Pneumatic Retinopexy for the Repair of Retinal Detachments: A
Transcription
Pneumatic Retinopexy for the Repair of Retinal Detachments: A
SURVEY OF OPHTHALMOLOGY VOLUME 53 ! NUMBER 5 ! SEPTEMBER–OCTOBER 2008 DIAGNOSTIC AND SURGICAL TECHNIQUES MARCO ZARBIN AND DAVID CHU, EDITORS Pneumatic Retinopexy for the Repair of Retinal Detachments: A Comprehensive Review (1986--2007) Clement K. Chan, MD, FACS,1,2 Steven G. Lin, MD,3 Asha S.D. Nuthi, DO,3 and David M. Salib, MD3 1 Southern California Desert Retina Consultants,Palm Springs, California; 2Department of Ophthalmology, Loma Linda University, Loma Linda, California; and 3Southern California Desert Retina Consultants, Palm Springs, California, USA Abstract. Pneumatic retinopexy has become an important surgical technique in the modern era of retinal surgical management for retinal detachments. It is primarily indicated for uncomplicated retinal detachments with retinal breaks involving the superior 8 clock hours of the fundus, although more complex retinal detachments may be successfully managed with this technique on a selected basis. Qualified candidates must be willing to maintain a specific head posture for five or more days for optimal outcome with pneumatic retinopexy. Basic surgical steps of pneumatic retinopexy include retinopexy of retinal breaks with cryotherapy or laser, intraocular gas injection before or after retinopexy, and maintenance of proper head posture by the patient for the required time period after surgery. Phakic eyes fared better than nonphakic eyes for pneumatic retinopexy, with the single-operation successes of 71--84% for the former and 41--67% for the latter. Despite lower single-operation successes with pneumatic retinopexy in comparison to sclera buckling, the multicenter pneumatic retinopexy trial and other published reports have shown that the final anatomical and visual outcomes are not disadvantaged by the initial pneumatic retinopexy. An extensive discussion of complications associated with pneumatic retinopexy is presented. In addition, a key feature of this review is a comprehensive update in the outcome of pneumatic retinopexy in published reports from 1986 to the present in chronological order not available in the current literature. This comprehensive summary shows updated average surgical outcomes for the 4,138 eyes in the 21-year period to be similar to previous reports: single-operation successes (74.4%), final operation successes (96.1%), new retinal breaks (11.7%), and proliferative vitreoretinopathy (5.2%). (Surv Ophthalmol 53:443--478, 2008. ! 2008 Elsevier Inc. All rights reserved.) Key words. Complications of pneumatic retinopexy ! expansile gas ! gas injection ! pneumatic detachment of vitreous ! pneumatic retinopexy ! pneumoretinopexy ! surgical outcome of pneumatic retinopexy ! visual outcome of pneumatic retinopexy can be managed by pneumatic retinopexy.40 Its efficacy for repairing a retinal detachment depends on three cardinal steps: 1) induction of retinopexy around all retinal breaks with cryopexy or laser; 2) Pneumatic retinopexy (PR) is an office-based procedure designed for repairing selected retinal detachments. Brinton and Hilton estimated that at least 40% of rhegmatogenous retinal detachments 443 ! 2008 by Elsevier Inc. All rights reserved. 0039-6257/08/$--see front matter doi:10.1016/j.survophthal.2008.06.008 444 Surv Ophthalmol 53 (5) September--October 2008 intraocular gas injection; and 3) consistent postoperative head positioning for appropriate gas tamponade to achieve closure of the retinal breaks. Since its introduction in 1985,111 PR has become a well-accepted alternative surgical technique to scleral buckling and vitrectomy for uncomplicated rhegmatogenous retinal detachments with superior retinal breaks in the modern era of retinal surgical management. Proponents of PR have underscored the advantages of less tissue trauma, usual lack of major complications, no hospitalization, and lower expenses, associated with PR in comparison to scleral buckling (SBP) and vitrectomy.40,110,111,240 The purpose of this review is to present a comprehensive updated summary of the indication, anatomical and visual outcomes, and complications associated with PR reported in the world literature for the past two decades. The presentation of not only the conventional but also the controversial applications of PR is not an unqualified endorsement of the latter by the authors, but rather a synopsis of all aspects of PR found in the world literature in the past two decades. Historical Perspective Jules Gonin, the father of retinal detachment surgery, performed the first successful surgery to close retinal breaks with his ignipuncture operation in 1919. He drained subretinal fluid with a Graefe knife after making a radial incision through the sclera to the choroid, and cauterized the retina by the retinal tear with a red-hot cautery.22,91 The idea of injecting intravitreal gas for repairing a retinal detachment was first introduced by Ohm in 1911.189 Subsequently, Rosengren described in 1938 a technique of external diathermy through the fullthickness eye wall to induce adhesion of retinal breaks to the underlying retinal pigment epithelial (RPE) and choroidal layers, followed by subretinal fluid drainage and intraocular air injection for closing the associated breaks.206,207 He reported a success rate of 76% in his surgical series with his technique.206 With the introduction of the Custodis procedure in the early 1950s, scleral buckling rapidly became the dominant surgical technique.58,59 The idea of repairing a retinal detachment with an intraocular air or gas bubble faded from the surgical scene for the next 50 years. It was not until 1985 that George Hilton re-introduced the idea of intraocular gas tamponade and retinopexy in an office setting for managing select cases of retinal detachments.40,111 Contemporaneous with Hilton’s discovery, Dominguez independently reported a similar technique in repairing certain CHAN ET AL retinal detachments.62--65 In the ensuing years, multiple clinical studies have reported both the benefits and complications of PR, establishing its current important role in the repertoire of the retinal surgeon for managing retinal detachments.1-- 10,13--21,25,27,31--34,36--40,44--52,54,56,57,60,62--69,71--82,89,90, 95--100,102,104,105,107,110--112,114--117,120--124,127--129,131,132, 134--137,139,140,141,143--152,154,159,160,162--175,178--181,184--186, 190--195,197,200--203,208,209,212,214--218,220,225--228,231--235, 238--253,255--261,263--265,267,270,271,273--278 Previous surveys of the practice patterns showed its varying popularity among retinal surgeons, partially dependent on their prior experiences, geographic locations, and ages. Multiple surveys have shown an increasing trend for PR in certain regions of the United States (i.e., California and Florida), and its lower preference in the Midwest and the Northeast of the United States and certain parts of Europe (e.g., United Kingdom, Germany). Younger and more recently trained surgeons also favor PR more than their older colleagues trained in earlier years.4,14,23,146,227,265 Mechanisms and Basic Requirements All surgical techniques for successful repair of a retinal detachment (RD) require closure of all retinal breaks and resolution of subretinal fluid. PR accomplishes this goal by the tamponade of retinal breaks with an intraocular gas bubble and induction of chorioretinal adhesion with cryopexy or laser.40,110,111,240 The property of high surface tension associated with an intraocular gas bubble constitutes its main mechanism for closing retinal breaks and enhancement of subretinal fluid absorption. For instance, a gas bubble induces an interfacial tension of 70 dynes per cm on the retinal surface, whereas silicone oil creates a surface tension of only 21.3 dynes per cm.84 Buoyancy of an intraocular gas bubble plays a secondary role in re-attaching the detached retina. For an average eye, 0.3 ml of gas can cover 60 degrees of the retinal surface, whereas it takes 1.2 ml of intraocular gas to cover 90 degrees of the retinal surface.40,110,111,240 More gas is required to cover an equivalent arc of retinal surface for a highly myopic eye with a large globe. A lingering intraocular gas bubble with excessive expansion may cause unwanted vitreoretinal traction, resulting in new retinal breaks.40,110,111,240 Table 1 outlines the duration and extent of expansion of intraocular air and various gases. Choices of the agent for intraocular tamponade include air, sulfur hexafluoride (SF6), perfluoroethane (C2F4), and perfluoropropane (C3F8). The law of partial pressures defines the mechanism of 445 REVIEW: PNEUMATIC RETINOPEXY TABLE 1 Intravitreal Gas Duration and Expansion Gas Average Duration Largest Size Average Expansion Air 3 days SF6 12 days C3F8 38 days Immediate 36 hours 3 days No expansion Doubles Quadruples gas expansion in the eye. The intraocular expansion of an inert gas (SF6, C2F4, or C3F8) is related to its low water solubility and the net diffusion of nitrogen and oxygen molecules into the gas bubble (Fig. 1). SF6 doubles in size by 24 to 36 hours, and lasts from 6--12 days.40,110,111,240 Previous studies have shown that it takes 7--14 days for the formation of a retinopexy scar with optimal tensile strength following cryopexy or laser.197 Thus, the optimal intraocular gas is one that expands sufficiently to cover all retinal breaks, and lasts long enough for the retinopexy scar to develop optimal tensile strength before its rapid disappearance subsequently. Perfluoroethane has the characteristics that match these criteria the best, but it has not been approved by the FDA for clinical use. In contrast, both SF6 and C3F8 have been approved by the FDA for clinical use. The long duration of C3F8 is advantageous for selected cases that require prolonged gas tamponade. Because of its greater expansion, only a small volume of C3F8 is required. However, many surgeons consider the long duration of C3F8 to be excessive, causing unnecessary delay in the resumption of normal activities and traveling for the patient. By default, SF6 has become the gas of choice used most frequently by most surgeons for PR. The basic criteria required for PR include the following:40,110 1) presence of a single or multiple retinal breaks occupying one clock hour of the retinal arc; 2) location of all retinal breaks in the superior 8 clock hours of the globe; and 3) ability of the patient to maintain the proper head position for at least 16 hours a day for five days or more. Patients with a retinal detachment not satisfying these criteria are expected to have much poorer responses to PR than those that do. Patient Selection The following conditions associated with a retinal detachment constitute the ideal case scenarios for performing pneumatic retinopexy (class I or II evidence-based support [A to C]):40,110 LIMITED SUPERIOR RETINAL BREAKS The limitation of one or more small retinal breaks within one clock hour of the retinal arc in one or two superior quadrants without other retinal breaks is associated with a favorable prognosis for PR. SUPERIOR RETINAL DETACHMENT The limitation of the RD primarily to a superior quadrant is another highly favorable clinical situation for PR. PHAKIA Most clinical series on PR have shown substantially higher success rates for managing phakic RDs in comparison to pseudophakic and aphakic RDs.3,6,8,15,32,34,37--40,44,46,47,51,52,71,77,78,89,98,104,105,110-112,129,134,140,152,164,166--169,178,180,181,185,190,192--194,200, 201,214,216,217,220,225,234,235,240--243,248,249,251,252,257,259, 260,269,271,273,275--277 Fig. 1. Intraocular expansion of inert gas is related to its low water solubility and net diffusion of nitrogen and oxygen molecules into the gas bubble. The lower success rates associated with the latter two conditions are thought to be related to the presence of multiple small missed or new peripheral retinal breaks, frequently in conjunction with peripheral vitreoretinal traction. However, Hilton and others have reported favorable responses to PR for many nonphakic eyes. Thus nonphakia per se is not a contraindication for PR.40,110 Certain eyes with features of RD not consistent with the best case scenarios outlined above may still respond appropriately to PR, if they satisfy the basic criteria described in the preceding section. In addition to the basic criteria, the following 446 Surv Ophthalmol 53 (5) September--October 2008 considerations are crucial for proper patient selection for successful PR:40,110 1) Can the patient follow directions, 2) Does the patient have physical problems that may prevent compliance with positioning requirements, 3) Does the patient’s home situation allow consistent postoperative positioning, 4) Are there potential obstacles to vital postoperative visits and close monitoring, 5) Does the patient plan on air or mountain traveling shortly after surgery, 6) Is there risk of wound dehiscence related to recent eye surgery, and 7) Is there history of severe or endstage glaucoma. OTHER FAVORABLE CONDITIONS Brinton and Hilton have advocated pneumatic retinopexy to be particularly appropriate for the following clinical situations:40,110 LOCATION OF RETINAL BREAKS UNDER THE SUPERIOR RECTUS MUSCLE Placement of a segmental scleral buckle under the rectus muscle for vertical action may induce diplopia, a risk avoided by PR. GLAUCOMA FILTERING BLEB AND DRAINAGE DEVICE PR is associated with the theoretical advantage of averting the disturbance of conjunctival integrity critical for a functional bleb and any underlying drainage device for an eye that has undergone a glaucoma filtering procedure,24,40 although the peritomy may circumvent the filtering bleb and disturbance of the fibrous capsule surrounding a drainage device may be avoided in the case of SBP. There is also a lack of reports in the literature showing higher bleb failure rates after scleral buckling in comparison to PR. RECURRENT RETINAL DETACHMENT FOLLOWING SCLERAL BUCKLING A limited superior RD associated with new or recurrent superior retinal breaks after scleral buckling is often more easily managed with PR than buckle revision. Friberg and Eller reported successful repair of 36 of 40 eyes (90%) with a recurrent RD after SBP with laser pneumatic retinopexy, whereby laser is applied around new or missed retinal breaks with a binocular indirect ophthalmoscopic device after flattening of the detached retina with additional intraocular gas injection.80,82 OTHER MISCELLANEOUS CONDITIONS Besides the previous conditions, Hilton has pointed out the advantageous of PR over scleral CHAN ET AL buckling for treating an RD in the presence of the following clinical conditions:40,110 1) large irregular retinal breaks that are difficult to close with scleral buckling; 2) globe with thin sclera at risk for perforation during scleral buckling; 3) need for maintaining emmetropia and avoidance of anisometropia often associated with scleral buckling; 4) avoidance of strabismus that may be induced by SBP; 5) minimizing cosmetic defects (i.e. lid ptosis and enophthalmos); 6) need to cancel scleral buckling due to mishap from a retrobulbar or peribulbar anesthetic injection; 7) extensive conjunctival scarring that hinders the performance of a peritomy required for SBP; 8) patient with an illness or physical condition at increased risk for complications with a surgical procedure in the operating room; 9) patients undergoing anticoagulant therapy at risk for ocular bleeding during SBP or vitrectomy; 10) patients with a physical condition at risk for intravenous or general anesthesia associated with SBP or vitrectomy;11) delay in availability of the operating room; and 12) lack of equipment or limited financial resources for SBP or vitrectomy. Advantages, Benefits, and Disadvantages The major advantages and benefits of PR over SBP and vitrectomy consist of the performance of a relatively non-invasive and brief procedure in the office setting, bypassing numerous requirements and their associated expenses for surgery in the operating room.40,110,111,240 The disadvantage is greater attention to retinal break detection necessary for PR.40,110,111,240 A scleral buckle procedure is potentially ‘‘more forgiving’’ for missed retinal breaks and postoperative peripheral vitreoretinal traction that may be supported by an encircling buckle. Consistent and precise postoperative positioning is also more critical for PR to enhance adequate retinal break closure and subretinal fluid absorption. Contraindications The following clinical situations contraindicate PR for repairing a retinal detachment. 40,110,111,240 PRESENCE OF RETINAL BREAKS WITHIN THE LOWEST 4 CLOCK HOURS OF THE INFERIOR QUADRANTS Although a recent report showed the successful repair of certain retinal detachments with breaks along the inferior 4 clock hours (4 to 8 o’clock) for selected patients,49 PR usually fails to repair such 447 REVIEW: PNEUMATIC RETINOPEXY RDs. Most patients cannot tolerate the unnatural position required for proper gas tamponade of the retinal breaks found in the lowest 4 clock hours. retinal detachment, that is, substantial PVR, and diabetic retinopathy. HAZY MEDIA PRESENCE OF PROLIFERATIVE VITREORETINOPATHY GRADE C OR D An RD with substantial proliferative vitreoretinopathy (PVR) including considerable retinal traction is not a good candidate for PR. For optimal outcome, the basic criteria for PR require the absence of PVR grade C or D. However, successful outcome with PR has been reported for RD with limited grade C PVR (see following section on Expanded Use of Pneumatic Retinopexy for Special Clinical Conditions). NONCOMPLIANCE WITH REQUIRED HEAD POSITIONING The presence of retinal breaks outside of the 11 to 1 clock hours in the superior quadrants requires more extreme head tilt positioning following PR that may be difficult for patients with certain physical disabilities, neck and back problems, or mental incompetence. Ocular opacities that impede the identification of all retinal breaks may lower the success rate of pneumatic retinopexy. Certain aphakic and pseudophakic eyes with multiple small peripheral retinal breaks in the presence of a cloudy peripheral lens capsule are poor candidates for PR. However, obscuration of a portion of the retina does not necessarily prevent success with PR, as long as the retinal breaks are appropriately located and treated. Expanded Use Of Pneumatic Retinopexy For Special Clinical Conditions The following sections describe the unconventional application of PR for various special clinical conditions (class II or frequently only class III evidence-based support). Considerable controversy and differing opinions exist among retina surgeons regarding the appropriateness of PR for these conditions. Many surgeons continue to advocate scleral buckling and vitrectomy instead of PR for these conditions. SEVERE GLAUCOMA With the exception of advanced glaucoma, most eyes with concomitant RD and glaucoma can be safely managed with PR, as long as the intraocular pressure is closely monitored, and corrective measures are taken if necessary. EXTENSIVE LATTICE DEGENERATION AND VITREORETINAL TRACTION The proponents of PR advocate PR for treating RD associated with limited lattice degeneration. They point out that most eyes with mild or moderate lattice degeneration associated with an RD respond well to PR.40,241 Tornambe showed successful repair of RD with 3 or fewer clock hours of lattice degeneration with PR in his retrospective series of 302 eyes.241 In that series, the presence of limited lattice degeneration (less than 3 clock hours) did not lower the single operation success rate. However, there is a lack of published controlled prospective studies contrasting the efficacy and safety of PR for cases with and without lattice degeneration. In addition, most agree that extensive lattice degeneration may indicate increased risk for vitreoretinal traction and new retinal breaks that are best treated with scleral buckling or a vitrectomy. PR is also unable to relieve significant vitreoretinal and fibrous traction associated with a rhegmatogenous RETINAL BREAKS MORE THAN 1 CLOCK HOUR APART AND LIMITED PVR Although the basic criteria for pneumatic retinopexy require one or more retinal breaks to be within one clock hour in extent, some surgeons have reported the successful management of an RD with multiple retinal breaks spanning up to 3 clock hours with PR.249 The technique of alternate head tilt may be employed for proper tamponade of the multiple breaks in this situation (Fig. 2). An RD with limited and localized PVR (grade C or less) at a substantial distance from the retinal breaks may respond well to PR. Tornambe et al reported an overall 75% success rate of retinal reattachment in a series of 40 eyes with retinal lesions that exceeded the basic criteria for PR (i.e., multiple breaks in multiple quadrants, large retinal breaks up to 2.5 to 6 clock hours in size, presence of moderate PVR of not more than grade C2).249 For eyes with limited PVR, PR was most effective when the star folds were not in the area of the retinal breaks.249 New breaks developed in 12.5% of eyes. GIANT RETINAL TEARS OR DIALYSES In 1993, Irvine and Lahey reported the successful management with PR for four of five selected RDs 448 Surv Ophthalmol 53 (5) September--October 2008 CHAN ET AL tilt followed by cryopexy led to complete retinal reattachment of both eyes one day later. Additional cryopexy successfully treated a new break on inferior attached retina of left eye 6 months later.239 PNEUMATIC RETINOPEXY FOR RETINAL DETACHMENT ASSOCIATED WITH PROGRESSIVE RETINOSCHISIS Fig. 2. Alternate head tilt allows proper gas tamponade of multiple breaks more than 1 clock hour apart for expanded indication of pneumatic retinopexy. In 2000, Vrabec reported successful repair of a case of RD associated with a progressive retinoschisis.258 Three superior pinpoint inner-walled holes were first closed with PR after cryotherapy. After flattening of both RD and retinoschisis by the next day, laser was applied around two large temporal posterior outer-walled holes. Complete retinal attachment was maintained with a visual acuity recovery to 20/30, ten months later. She advocated consideration of PR for selected cases with combined retinoschisis and RD, instead of automatically choosing vitrectomy and scleral buckling for such cases. INFERIOR RETINAL BREAKS with a giant retinal tear within 180 degrees in extent and a mobile retinal flap.121 Melgren and Michels also published successful repair of RD with a giant retinal dialysis (3 or more clock hours in size) in the superior 8 clock hours of the fundus in four selected eyes with PR.173 RETINAL BREAKS ON BOTH ATTACHED AND DETACHED RETINA Some surgeons have also successfully employed PR for repairing retinal breaks present on both attached and detached retina,40,110,111,240 when proper steps are taken. First, laser or cryopexy should be applied on breaks and lattice degeneration on attached retina before injection of an intraocular gas bubble and treatment of the retinal breaks on detached retina. To prevent the migration of subretinal fluid under attached retinal breaks and lattice degeneration, the steamroller maneuver should be performed immediately after intraocular gas injection (see following discussion for details). BILATERAL RETINAL DETACHMENT In 1987, Tornambe reported the successful repair of bilateral retinal detachment with bilateral PR in a case report.239 Multiple superior large and small breaks and lattice degeneration were separated by more than 30 degrees in one eye with an RD that also had a macular hole. There were two large breaks separated by 30 degrees in the other eye. Sequential PR for the two eyes and alternate head Tornambe et al showed that RDs with retinal breaks in the inferior one-third of the fundus do not respond well to PR. Despite prior reports of poor responses with PR for RDs with inferior breaks, Chang et al presented successful outcome with ‘‘inverted’’ PR for RD with retinal breaks involving the inferior 4 clock hours (including 6 o’clock) in selected patients in 2003.49,79,165,212 In their series, highly motivated patients with inferior retinal breaks associated with an RD were required to assume an inverted head posture for successful closure of those breaks. They found closure of the inferior breaks after as little as 8 hours of appropriate gas tamponade. They achieved a single-operation success of 81.8%, and a final success of 100% of their cases. Through their experience with inverted PR, they showed the proof of concept that shorter duration than routine practice for gas tamponade is needed even for superior retinal breaks associated with PR. Further literature search also showed that the idea of utilizing PR to treat RD with inferior retinal breaks was first described by Friberg and Eller in their report of pneumatic repair of primary and secondary RD with a binocular indirect ophthalmoscopic laser device in 1988.81,82 On a practical basis, most patients are unable to tolerate and maintain the unnatural inverted head posture required for using PR to repair an RD with inferior retinal breaks between 4 and 8 o’clock. The use of PR for such cases remains to be a highly unorthodox and controversial practice. In 1989, Bagolini et al reported an unorthodox technique of temporary 449 REVIEW: PNEUMATIC RETINOPEXY surgical rotation of the extraocular muscles in order to shift retinal breaks located on the inferior fundus to the upper quadrants for superior gas tamponade during pneumatic retinopexy.17 RETINAL DETACHMENT ASSOCIATED WITH POSTERIOR RETINAL BREAKS, INCLUDING MACULAR HOLES, AND OPTIC NERVE PIT It is difficult for scleral buckling to repair an RD associated with posterior retinal breaks, including macular holes, and macular detachment associated with an optic nerve pit. Pars plana vitrectomy with or without membrane stripping and long-acting gas or silicone oil tamponade is the current standard therapy for repairing these conditions. Alternative unconventional techniques for their repair include macular buckling, exchange of liquefied vitreous fluid with intravitreal gas, and drainage of subretinal fluid followed by intraocular gas injection.29,30,130,159,203,237 Some surgeons have found PR to be an effective alternative for treating such cases under certain circumstances. For instance, Lincoff et al and Blodi et al proposed injecting intravitreal expansile gas followed by postoperative prone positioning for repairing selected cases of myopic RD associated with a macular hole.30,159 Kusuki and coworkers treated 12 eyes in 12 patients with macular-hole RD by injecting SF6 gas followed by head-prone positioning.141 Seven eyes (58.3%) were successful after one injection and three eyes after two injections, with a final success of 83.3%. Menchini et al reported a success rate of 77.8% for treating myopic retinal detachment with a macular hole with PR.175 They applied limited laser around the macular hole besides gas injection, followed by head-prone positioning for a number of days. Rashed and Sheta treated 50 eyes with retinal detachment associated with a macular hole.202 Ten eyes underwent PR with plain room air (group 1), 30 eyes underwent vitrectomy and fluid air exchange (group 2), and 10 underwent a vitrectomy followed by silicone oil tamponade due to PVR (group 3). Eight of 10 eyes in group 1, 21 of 30 eyes in group 2, and 8 of 10 eyes in group 3 achieved successful closure of the macular hole. The visual outcome was the best for eyes undergoing PR (group 1). In 1990, Focosi et al performed PR to treat the macular hole after repairing concomitant peripheral retinal breaks with cryopexy and scleral buckling in six eyes.74 They achieved successful retinal reattachment and macular hole closure in all six eyes. Lazzaroni achieved 50% and Santillo 58.6% success, whereas La Torre achieved 73.9% success for macular hole detachment in their series.143,147,209 In 1993, Zhai et al advocated intraocular gas injection followed by head prone posture for selected cases of posterior retinal perforation associated with trauma.278 Despite these reports, many surgeons consider the treatment of these conditions with PR or its variations a controversial practice. PNEUMATIC RETINOPEXY FOR RETINAL DETACHMENT ASSOCIATED WITH OTHER SPECIAL CLINICAL ENTITIES In 1993, Glacet-Bernard and Coscas described the successful employment of PR for repairing an RD complicating Vitelliform macular dystrophy.88 Mild grid laser was applied around the macular hole responsible for the RD after flattening of the RD with gas tamponade. In 1996, Haimovici et al reported a successful repair of a nasal RD associated with two small nasal flap tears as well as a supranasal choroidal melanoma with PR.102 Iodine-125 plaque radiation therapy was performed 3 months after PR. Six months later, the retina was completely attached and the visual acuity was improved to 20/70. There was absence of tumor regrowth, extension, and metastasis 11 months later. However, the patient’s survival status and the condition of the melanoma are not known at 5 years or beyond, the minimal time interval usually required for judging the efficacy and safety of a therapy involving a choroidal melanoma. In 1998, McAuliffe and Heinemann used PR to reattach a superior bullous RD associated with a single superior flap tear in an eye that has received a ganciclovir intraocular device for CMV retinitis with a visual recovery of 20/30.170 However, recurrent inferior retinal detachment associated with multiple inferior necrotic breaks developed 2 months later, requiring a vitrectomy and silicone oil tamponade. The visual outcome was 20/70 seven months later. In 1993, Greco and Ambrosino reported successful repair of recurrent RD associated with Marfan syndrome after failed SBP.92 Surgical Techniques IDENTIFYING BREAKS Prior to any surgical maneuvers, the surgeon must spend sufficient time to re-identify and confirm all retinal breaks associated with the retinal detachment, and devise his or her strategy for treating those breaks.40,110,111,240 Inadequate endeavors on this aspect of the surgery frequently lead to surgical failure. If any previously missed retinal breaks are now detected to be 2 or more clock hours from the originally identified superior breaks, alternate head tilt to allow appropriate gas tamponade for both sets of tears may be performed after retinopexy is 450 Surv Ophthalmol 53 (5) September--October 2008 applied (see previous section on Expanded Use of Pneumatic Retinopexy for Special Conditions), or PR may be abandoned for another procedure. For retinal tears located in the inferior quadrants, the surgeon must be prepared to treat the RD with another procedure, such as scleral buckling. The surgeon must also be prepared to treat all retinal breaks and lattice degeneration located on attached retina with laser or cryopexy first, prior to gas injection.40,110,111,240 ANESTHESIA For most patients, subconjunctival anesthesia supplemented by topical anesthesia is sufficient for the performance of PR. However, some patients need retrobulbar or peribulbar anesthesia for achieving adequate analgesia, particularly in cases requiring substantial cryotherapy and laser treatment for retinal breaks during PR.40,110,111,240 General anesthesia is almost never performed with PR, since it forfeits the major advantage of PR as an office-based procedure. In the rare instance of general anesthesia, the surgeon must remind the anesthesiologist to avoid nitrous oxide in order to prevent postoperative shrinkage of the intraocular gas bubble.40 RETINOPEXY The induction of appropriate adhesion of the retinal breaks to the underlying RPE and choroidal layers is a vital step for PR. This process is usually accomplished with either cryopexy or laser application. Although cryopexy can be performed in a single session, two separate sessions are usually required for laser. Experimental studies on the in vitro peeling force by Yoon and Marmor272 and on the in vivo adhesive force by Kita et al133 have shown that laser photocoagulation produces a bond that approaches normal adhesive tensile strength within 24 hours, possibly related to local effects such as fibrin formation. There is gradual increase in the adhesive strength that approximates twice normal by 2--3 weeks after photocoagulation.142 In vivo studies also showed an adhesive strength for diathermy with a time course similar to laser.133 For cryotherapy, however, there is weakening of the adhesion for the first week possibly due to local tissue inflammation or edema, before development of tensile strength similar to the other modalities during the second week.133 Thus, laser and cryotherapy result in a scar with a similar tensile strength over the long term. However, laser photocoagulation allows a more rapid bond formation. Cryopexy is required when there is substantial subretinal fluid separating the RPE from the retinal break, unless CHAN ET AL laser is applied during a second session after flattening of the retina with intraocular gas tamponade. Small peripheral retinal breaks are more easily treated with cryopexy, and are usually difficult to locate and may be obscured by multiple intraocular gas bubbles for laser treatment after gas injection. ‘‘Fish-egg’’ gas bubbles may be avoided by using the proper gas injection technique (injecting the gas with the needle in a vertical and downward orientation at the uppermost portion of the globe and toward the center of the eye). Once present, these annoying gas bubbles may be temporarily displaced by reversing the head tilt from the original position, or placing the head upside down to move the gas bubbles away from the superior fundus for laser application. Laser is more easily applied for treating retinal breaks posterior to the equator. There is also potentially less pigment cell release into the vitreous cavity with laser than cryotherapy for treating large retinal breaks and retinal breaks associated with a bullous RD.43,96 The best technique for transconjunctival cryopexy involves scleral indentation with the cryo-probe under the guidance of indirect ophthalmoscopy to induce RPE apposition to the underlying retinal break. Yellowish white discoloration of the fundus created by the freeze-thaw of the cryo-probe encompasses the choroidal, RPE, and sensory retinal layers around the margin of the retinal break. However, one can terminate the freeze as soon as a dull orange discoloration is visible at the level of the RPE and choroid alone without changes of the neurosensory retina, particularly in the presence of a bullous RD that impedes RPE and retinal apposition.40,110,111 Application of laser through shallow subretinal fluid with increased laser power while pressing the RPE against the retinal layer with forceful scleral indentation to obtain sufficient laser uptakes may inadvertently puncture the retina with new holes.40,110,111,240 In case of laser application after flattening of the retina with intraocular gas, it is best to prepare a detailed diagram with clear landmarks to pinpoint the locations of the breaks ahead of time, and their locations may be further obscured by multiple gas bubbles after gas injection. As mentioned previously, these gas bubbles may be avoided with the proper gas injection technique, or displaced with the appropriate head posture subsequently during laser application. Transpupillary laser indirect ophthalmoscope (LIO) is superior to slit-lamp delivery of laser in circumventing the interference of intraocular gas bubbles, particularly for reaching the retinal breaks on the peripheral fundus.80--82 The ideal laser application involves the creation of multiple rows of overlapping lesions with 451 REVIEW: PNEUMATIC RETINOPEXY moderate intensity that surround the entire retinal break. To prevent anterior migration of subretinal fluid, the surgeon must not neglect adequate laser application along the anterior margin of the retinal break. Care must be taken to avoid excessive laser power when treating through intraocular gas bubbles due to their insulation of heat generated by the laser. In fact, it is best to avoid performance of laser through the gas bubbles since the aberrant optical effects induced by the bubbles may cause inaccurate or harmful laser application. Laser instead of cryopexy must be used in case of new retinal breaks overlying a pre-existing scleral buckle.40,81,82,150 In recent years, transconjunctival diodepexy was introduced as an alternative to cryopexy or laser for inducing retinopexy.103 Similar to the latter modalities, this new technique is performed through indirect ophthalmoscopic guidance. It possesses certain positive qualities of both laser and cryopexy. Similar to laser, diodepexy creates immediate visible chorioretinal adhesions. Transconjunctival diodepexy may also be applied through a scleral buckle for retinal uptakes. In addition, it allows safe treatment through subretinal fluid as in the case of cryopexy. STERILE PREPPING Meticulous sterile prepping of the ocular surface and the surrounding adnexa is critical for prevention of postoperative infections. A common antiseptic for this task is povidone-iodine.229 The commercially available 5% povidone-iodine ophthalmic solution (Betadine 5% Sterile Ophthalmic Prep Solution; Alcon Laboratories, Fort Worth, TX) minimizes corneal toxicity in comparison to the regular 10% povidone-iodine solution. However, Tornambe and Hilton strongly favor 10% instead of 5% Betadine for sterilization during PR (Personal communication, Paul Tornambe, MD). The authors favor painting the lashes and margins of the upper and lower lids with sterile applicators soaked with the povidone-iodine solution after application of a few drops of the same on the globe, although this measure is not mandatory. Excessive antiseptic solution on the globe must be rinsed off with saline or balanced salt solution after a few minutes to prevent corneal toxicity and abrasion. Sterile gauzes are used to dry the sterilized lid margins before placement of a sterile lid speculum. Tornambe has emphasized the importance of using closed lid speculum to keep the eye lashes away from the surgical instruments. In the event of a history of prior allergic reactions to povidone-iodine, a topical antibiotic with broad-spectrum coverage may be substituted. CHOICE OF SULFUR HEXAFLUORIDE, PERFLUOROPROPANE, PERFLUOROETHANE, AND AIR AND GAS RETRIEVAL AND INJECTION The required size of the intraocular gas bubble depends on the extent of the retinal breaks. In general, a 1.0-ml bubble covering close to 3 clock hours of the retinal surface is usually adequate.40,109,110,240 Because SF6 expands to twice its size in 24 hours, an initial bubble size of 0.45 to 0.5 ml is sufficient.40,110,111,240 For intraocular air, a volume of at least 0.8 ml is required for adequate retinal coverage.40,110,111,24 Its more rapid absorption and lack of expansion may lead to inadequate duration of tamponade of the retinal breaks. Nevertheless, Carim, Kaplan, Sebag, and Waterhouse et al have presented in separate reports the successful repair of RD with the injection of room air for PR.44,45,128,214,260 A volume of 0.5--1.0 ml of filtered air was used by these investigators. The potential disadvantage of a larger volume of the required injected air is balanced by its quicker resolution, in comparison to gas. However, the quicker resolution of the injected air may not allow adequate tamponade of a large break long enough after its treatment with cryotherapy. Poliner and Tornambe noted that effective chorioretinal adhesion may take 7--10 days after cryotherapy, too long a duration for a dissipating air bubble; although breaks treated with laser may need a shorter duration of tamponade.197 Although only a small volume of perfluoropropane (C3F8) is required due to its expansion to four times its initial size, its greater half-life in comparison to SF6 and C2F4 may promote vitreoretinal traction for a longer duration than is desired and delay air or mountain traveling. In 1992, Bochow et al reported their results using C2F4 for pneumatic retinopexy in 17 eyes.31 C2F4 possesses certain theoretical advantages over SF6 and C3F8. A smaller volume of C2F4 may be used in comparison to SF6 due to its greater expansion. It also does not persist as long as C3F8, but lasts longer than SF6 until optimal tissue adhesion is achieved for the treated retinal tear. But as noted previously, C2F4 has not been approved by the FDA for clinical use in contrast to the other two gases. In 2000, Doi and coworkers reported thinning or disappearance of the outer plexiform layer in the superior retina in rabbit eyes that received C3F8, but not in rabbit eyes that received SF6 or air.61 In addition, immunohistochemical examination showed abnormal glutamine distribution of the superior retina in rabbit eyes injected with C3F8, SF6, or air.61 The clinical relevance of these findings to human eyes is unknown. For various reasons mentioned previously, SF6 possesses sufficiently favorable characteristics 452 Surv Ophthalmol 53 (5) September--October 2008 for PR. By default, it has become the gas of choice for most surgeons who perform PR (see preceding section on Mechanisms and Basic Requirements).40,110,111,240 A small amount of gas is delivered from a highpressure gas tank through a step-down device and via a Millipore filter into a tuberculin syringe. The step-down device in the form of a pressure-reducing valve or a low-pressure balloon or catheter allows the placement of a small volume of gas into the tuberculin syringe in a controlled fashion. The gas drawn through a Millipore filter into a syringe can be diluted due to infiltration with room air, if not used in a few minutes.118 In case the filtered gas is not injected immediately, a three-way stopcock with its valve in the closed position may be applied to the tip of the tuberculin syringe filled with gas, in order to preserve its purity for a period of time. Alternatively, Friedrichsen advocated the temporary storage of the filtered gas in the sterile interior of a Vacutainer (Becton Dickinson Inc., Franklin Lakes, NJ), originally intended for storing blood products obtained by venipuncture (personal communication, Eric Friedrichsen, MD). He has shown by mass spectrometry that the purity of the filtered gas stored in a Vacutainer remains undegraded for more than 24 hours. Multiple Vacutainers filled with the filtered gas can be safely transported during air traveling for mission service in the third world without the loss of the purity of the gas for a number of days (Personal communication, Herbert Cantrill, MD, Minneapolis, MN). GAS INJECTION TECHNIQUE The proper gas injection technique requires injecting the filtered gas bubble contained in a tuberculin syringe via a short 27- or 30-gauge needle at 3--4 mm posterior to the limbus into the vitreous cavity. 40,110,240 To reduce the development of small intraocular ‘‘fish-egg’’ gas bubbles, the surgeon should inject the gas at an upper position of the globe, point the needle perpendicular to the sclera and downward toward the center of the eye. After inserting half of the needle to penetrate the vitreous humor, the needle should be partially withdrawn until only 1 mm of it is still in the eye. Brisk injection of gas through the shallowly inserted needle allows continued gas placement inside the same single enlarging gas bubble (Fig. 3).40,110,240 The surgeon should also inject the gas away from a large retinal break, in order to reduce the risk of subretinal gas migration. ‘‘Fish-egg’’ bubbles have greater tendency to enter subretinal space through a retinal tear. Eyes with retinal tears larger than 1 clock hour or more rigid retinal tears are at greater CHAN ET AL Fig. 3. Injecting gas at upper position of globe with the needle oriented perpendicular to the sclera and directed toward the center of the globe reduces the tendency for fish-egg formation. risk for subretinal gas entry. Most surgeons perform the gas injection with the patient lying in a supine position. For some eyes, the authors found certain advantages in performing the gas injection with the patient in the upright position. For instance, the upright position during gas injection reduces the tendency for gas migration and vitreous prolapse into the anterior chamber when encountering elevated intraocular pressure, although this technique should be avoided in the presence of a large retinal tear due to the risk of the formation of ‘‘fishegg’’ gas bubbles. An alternative is to perform preinjection paracentesis for eyes with substantial rise in intraocular pressure. STEAMROLLER MANEUVER Yeo et al first described extension of subretinal fluid under detached retina to an area of attached retina following intraocular gas injection associated with PR.274 The steamroller maneuver is designed to prevent or minimize such an occurrence, particularly in case of potential macular involvement with the migrating subretinal fluid.40,110,240 Other situations that indicate this maneuver include: 40,110,240 1) avoidance of iatrogenic detachment of attached retinal breaks; 2) obstruction of appropriate viewing of critical posterior retinal structures by a bullous RD; 3) difficult treatment of retinal breaks due to excessive subretinal fluid under a bullous RD; 4) fish mouthing or enlargement of a retinal tear caused by excessive subretinal fluid, complicating its treat- REVIEW: PNEUMATIC RETINOPEXY ment; 5) the need to move subretinal fluid from the macula in the presence of delayed subretinal fluid absorption; and 6) the requirement to flatten a retinal tear against the RPE in the presence of multiple small bubbles to prevent their subretinal migration. Fig. 4 illustrates details of the steamroller technique and clarifies its mechanisms. The patient’s immediate assumption of the head prone position after gas bubble injection constitutes the first step of this maneuver, promptly placing the injected gas on the macula and minimizing subretinal fluid collection under the posterior retina. As the patient’s head is slowly elevated over the next 10 to 15 minutes, the subretinal fluid is gradually pushed by the gas bubbles from the posterior to the superior fundus, away from the macula, and out through the superior retinal breaks into the vitreous cavity. During this process, the retina is gradually flattened from the posterior to the superior direction. When the patient’s head finally reaches the upright position, the gas bubbles are positioned against the uppermost fundus, re-apposing the retinal breaks to the underlying RPE, thus simplifying their subsequent treatment with cryotherapy or laser. One theoretical risk of the steamroller technique is the potential release of pigment cells from the subretinal space into the vitreous cavity.96,101 Thus the performance of this maneuver after extensive cryotherapy of retinal breaks runs the risk of excessive collection of RPE cells in the vitreous cavity and subretinal fluid, capable of creating unwanted periretinal fibrocellular proliferation and even proliferative vitreoretinopathy (PVR).43,96,101 Tornambe et al Fig. 4. Steamroller maneuver consists of gradual elevation of head from the prone position after gas injection. 453 advocated the judicious use of the steam roller maneuver, and the application of cryopexy or laser after retinal flattening to decrease the chance of excessive pigment cell release into the vitreous cavity.240,247,248 They noted that the steamroller maneuver was utilized in 18% of eyes in the Pneumatic Retinopexy Clinical Trial. For all those eyes, cryopexy was applied after the steamroller maneuver. None of those eyes developed PVR or visually significant macular pucker.248 In a prospective study of PR involving 40 eyes with primary RD,271 Yanyali et al reported in 2007 no statistically significant differences in efficacy and safety for 21 eyes treated with steamroller technique versus 19 eyes treated with the basic technique. Only one eye (5%) in each group developed PVR. THE ROLE AND TIMING OF PARACENTESIS AND POSTOPERATIVE INTRAOCULAR PRESSURE MANAGEMENT Due to its lower tendency for underestimating the intraocular pressure of a gas-filled eye, applanation tonometry (including the Tono-Pen [Medtronic, Inc, Minneapolis, MN]) is preferred for the monitoring of the intraocular pressure after gas injection.153,196 Schiötz tonometry will provide falsely low IOP readings.12,182 Mild to moderate intraocular pressure rise after intraocular gas injection is well tolerated by most eyes without special treatment, as long as perfusion of the central retinal artery is not compromised at the end of surgery.40,240 However, corrective measures for elevated intraocular pressure should be considered for eyes with significant glaucoma or those at risk for deficient arterial perfusion after surgery. Simple treatment may include application of topical and oral hypotensive agents.40,240 However, additional measures may be taken if necessary. Many surgeons avoid the performance of a paracentesis during PR due to reports in the literature suggesting increased risk of endophthalmitis associated with paracentesis in general.11,108,125 However, there is a lack of reports specifically associating paracentesis during PR with increased incidence of endophhalmitis or other complications. Complications such as hyphema, iris prolapse, and corneal endothelial and lens damages may be routinely avoided when proper paracentesis technique is utilized. The authors favor the performance of paracentesis prior to gas injection for the following reasons: 1) reduced tendency for vitreous displacement into the anterior chamber due to sudden decompression of the globe; 2) decreased propensity of reflux of gas from the vitreous cavity to the subconjunctival space; and 3) lower chance of 454 Surv Ophthalmol 53 (5) September--October 2008 corneal or scleral wound dehiscence. A paracentesis may also be required after gas injection, if excessive intraocular pressure is not lowered, and central retinal artery pulsation persists beyond a few minutes. The usual method of paracentesis involves passive egress of aqueous humor through a 25- or 27- gauge needle connected to a tuberculin syringe without its plunger via the limbus. Pars plana paracentesis is performed on a less frequent basis due to greater risk of vitreous incarceration into the syringe. To decrease the latter complication, the plunger of the syringe is left in place for pars plana paracentesis. The needle is directed through the pupil, around an intraocular implant, or through the iris into the anterior chamber with the pars plana approach. The following conditions may indicate pars plana instead of limbal paracentesis: 1) shallow anterior chamber; 2) pre-existing vitreous herniation into the anterior chamber; 3) pre-existing corneal endothelial decompensation; and 3) presence of an iris-fixated or anterior chamber implant at risk of contact with the corneal endothelium.40,110,111,240 The Josephberg 1-port vitrectomy biopsy cutter (Visitec portable vitrector; Becton Dickinson, Inc, Franklin Lakes, NJ) allows safe and effective vitreous removal to make room for a sizable volume of intraocular gas injection with the patient sitting in an upright position at the slit lamp in the office setting. This device can be considered if a large volume of gas injection is required. The Intrector (Insight Instruments Inc., Stuart, FL) is the latest 23-gauge version of this device. Due to the risk of harmful vitreoretinal traction, a vitreous tap via the pars plana is rarely necessary, and reserved for inadequate lowering of excessive intraocular pressure following aqueous paracentesis alone in eyes with liquefied vitreous. Sufficient liquid vitreous is usually present in a highly myopic or syneretic eye for such a maneuver to be safely performed.40,110,111,240 The sequential injection of two smaller bubbles to achieve a larger intraocular gas volume may avoid acute intraocular rise and the need of a paracentesis. FREQUENCY AND TIMING OF POSTOPERATIVE VISITS AND POSTOPERATIVE CARE After applying topical antibiotic-steroid and cycloplegic medications, the operated eye is patched. Drawing an arrow on the outer surface of the eye patch corresponding to the axis of the retinal breaks is a useful means of reminding the patient of the appropriate head tilt for the required postoperative period.40,110 The patient is encouraged to check the proper orientation of the arrow pointing at the CHAN ET AL ceiling by looking into a mirror from time to time. An alternative is the application of the Tornambe leveler(Escalon Medical Corp., New Berlin, WI), a reliable device that reminds the patient of the proper head posture after surgery. Previous reports advocated maintaining the correct head posture for at least 5 days.40,110,111,240 Three days of proper head positioning may be sufficient for some eyes.110 However, we ask our patients to maintain the proper head positioning for at least a week to achieve the best outcome, considering the published data of 7 to 14 days for formation of a scar with the optimal tensile strength after retinopexy.133,142,197,272 The authors usually allow the patient to take a break from the required head tilt for 15 minutes every hour during this period. However, the patient is reminded not to assume a supine position that may lead to any contact between the intraocular gas bubbles and the lens and shallowing of the anterior chamber.40,110,111 Close monitoring of the operated eye after surgery is required to ensure optimal improvement and minimal complications. The authors favor examining the patient the day after surgery, 1 week later, 2--3 weeks later, and every month thereafter for the next few months. More frequent examinations may be required for some cases. In case of laser therapy for induction of retinopexy, it should be applied as soon as the retinal breaks are flattened by the gas bubbles. This is usually performed 24--48 hours after gas injection. Most of the subretinal fluid absorbs within 72 hours after closure of the retinal breaks.40,110,111,240 Surgical Outcome ANATOMICAL AND VISUAL OUTCOMES In the retrospective review of their first 100 cases of PR performed by multiple clinical centers, Hilton et al reported an initial 91% reattachment rate, and a 6-month single-operation success of 84%.112 Fourteen (14) of the 16 failures were successfully repaired with subsequent SBP, resulting in a final 98% rate of retinal reattachment with multiple operations. For eyes with pre-operative macular detachment, 65% achieved 20/20 to 20/50 visual outcome. For eyes with extramacular detachment, 86% accomplished stable or improved vision after surgery. They emphasized that the advantages offered by PR include reduced tissue trauma, fewer complications, no hospitalization, and lower expenses. In a subsequent report of more complex RD with retinal characteristics exceeding the basic criteria for PR (multiple breaks in multiple quadrants, breaks exceeding 1 clock hour, and PVR C1 or C2), Tornambe et al reported single-operation successes of 75%.249 Phakic eyes fared better than REVIEW: PNEUMATIC RETINOPEXY aphakic or pseudophakic eyes. Eighteen percent of phakic eyes failed, whereas 38% of aphakic and pseudophakic eyes failed. Eight eyes that failed were successfully reattached with additional surgery (five with SBP and three with vitrectomy). Vision was improved in 75%, unchanged in 17%, and worse in 8% after surgery. Visual improvement was achieved in 58% of macula-on RD, and 79% of macula-off RD. Whereas 23% of eyes had a VA of 20/40 or better before surgery, 48% of eyes achieved a final VA of 20/40 or better after surgery. Despite their favorable results, others reported less successful outcome. In a series of 51 consecutive eyes, Chen et al reported a single-operation success of 63%, and the results were much better for phakic eyes (74%) than nonphakic eyes (41%).51 For their eyes with macula-on RD, the vision was stabilized or improved in 71%, and decreased in 4%. For their macula-off RD, a VA of 20/20 to 20/50 was achieved in 52 % of eyes. To overcome confounding variables inherent in a retrospective study, Tornambe and Hilton organized the Retinal Detachment Study Group to prospectively compare the results of PR with scleral buckling.243 SBP was performed for 95 eyes, and PR was performed for 103 eyes. They reported singleoperation successes of 82% for SBP eyes versus 73% for pneumatic retinopexy eyes. The difference was not statistically significant. With reoperations, 98% of SBP eyes, and 99% of PR eyes achieved retinal reattachment. Phakic eyes did better than nonphakic eyes for both groups. Single-operation successes were found in 90% of phakic eyes that underwent SBP, and 75% of phakic eyes that underwent PR, whereas 73% of nonphakic eyes after SBP and 67% of nonphakic eyes after PR achieved the same. Regarding the visual outcome, those eyes with a duration of macular detachment of 14 days or less, postoperative VA of 20/50 or better at 6 months was achieved in 80% of the PR eyes, but only 56% of the SBP eyes (p 5 0.01). For eyes with macular detachment of 7 days or less, post-operative VA of 20/50 or better at six months was attained in 83% of the PR eyes and 62% of the SBP eyes (p 5 0.01).243 For eyes with a duration of macular detachment of 3 days or less, postoperative VA of 20/50 or better at 6 months was achieved in 91% of PR eyes and 64% of SBP eyes (p 5 0.05). Despite no statistical differences in preoperative VA, restoration of VA was more rapid after PR. And despite lower single-operation successes with PR, the final anatomic and visual outcomes of PR eyes were not disadvantaged by the initial PR for this study. In a consecutive series of 101 eyes with simple primary RD that underwent PR, Ambler et al reported 77% single-operation successes.8 Their phakic eyes and 455 those nonphakic eyes with an intact posterior capsule had an 84% success rate in comparison to a 56% success rate for their nonphakic eyes with an opened capsule. Their analysis of eyes that initially failed with PR showed that the final visual outcome was not adversely affected by PR. Retinal reattachment was eventually accomplished with multiple procedures for all 23 eyes that experienced initial failure after PR. The two-year study of the multicenter PR clinical trial showed continued visual improvement over the long term, so that 94% of eyes with macula-on RD, and 90% of eyes with macula-off RD achieved VA of 20/50 or better after 2 years; thus, implying the importance of more than 1 year of follow-up.248 In their summary of 1,274 eyes in 26 published reports on PR performed between 1986 and 1989, Hilton et al found singleoperation successes for PR to be ranging from 53% to 100%, with the average at 80%.40,114,240 There was one study involving only 4four eyes that reported a 100% success rate for PR in that series. They also reported final success rates ranging from 92--100% with the average at 98% after PR in their summary.40,114,240 In 1995, Grizzard et al performed retrospective analyses on 107 consecutive eyes that underwent PR in their practice together with a literature review of 25 statistical series on PR with primary attention to failures.97,98 They performed univariate and multivariate analyses and calculated odds ratios for risk factors using logistic regression. They reported 69% single successes, and 98% final successes. Missed breaks were found in 14.9% eyes, reopened breaks in 11.2% of eyes, and unclosed breaks in 4.6% of eyes. Risk factors correlating with failure included male sex (adjusted odds ratio [OR] 5 2.65), eyes with preoperative visual acuity worse than 20/50 (adjusted OR 5 1.21), eyes with four quadrants of RD or total RD (adjusted OR 5 2.03), aphakic or pseudophakic eyes (adjusted OR 5 1.91), and eyes with additional pathologic findings (adjusted OR 5 3.14). In their series, poor visual outcome was associated with initial visual acuity of less than 20/50 (adjusted OR 5 15.7), and eyes with four quadrants of RD or total RD (adjusted OR 5 5.01). In 2007, Kulkarni et al140 also found a greater tendency for males than females to develop recurrent RD after PR (32% versus 15.3%, p 5 0.07). Both Grizzard and Kulkarni advocated better patient education in proper head tilt positioning for patients undergoing PR that might correct the discrepancy in success rates between males and females.98,140 With proper instruction for strict posture compliance, even teenagers with RD may respond well to PR, as reported by Chen and Hwang in 2007.52 In 1997, Tornambe retrospectively analyzed the characteristics and results of his series of 456 Surv Ophthalmol 53 (5) September--October 2008 302 eyes that underwent PR. 241 He found a singleoperation success of 68% for all cases, and final success of 95% after additional surgery. When cases fitting the PR multicenter clinical trial (PRCT) criteria were selected and staged supplemental 360-degree retinopexy was performed, the single operation success rate was raised to 85%.241 However, detractors of PR have criticized the low initial success rate of only 55% for eyes undergoing focal retinopexy during PR for that series in comparison to other series, and they also pointed out the lack of a prospective study to confirm the efficacy of 360degree retinopexy for PR. In that series, 86% of eyes reattached with single PR procedure achieved visual acuity (VA) $20/40, whereas only 52% of eyes that underwent multiple procedures achieved the same. For macula-on RD, 85% achieved VA $20/40, but only 66% macula-off RD achieved the same. For eyes with macular detachment within 7 days, 73% recovered VA $20/40. In a subsequent retrospective study in 1998, Han and coworkers reported less favorable outcome for PR in their comparison of 50 eyes that underwent SBP with 50 eyes that underwent PR.104,105 Singleoperation successes were significantly higher for SBP eyes (84%) in comparison to PR eyes (62%), (p # 0.01). They found reoperation rates to be significantly higher for PR eyes (38%) in comparison to SBP eyes (14%), (p # 0.01). They used multivariate analysis with logistical regression to show that PR was the sole factor predictive of retinal detachment after a single procedure. However, they reported final retinal reattachment rate to be similar for both groups (98%), consistent with previous studies. They also found the final VA and rates of PVR to be similar for phakic eyes that underwent PR in comparison to eyes that underwent SBP. However, nonphakic eyes that underwent PR had a higher rate of single-operation failure, a higher rate of PVR (p # 0.05), and worse visual outcome in comparison to phakic eyes that underwent PR and phakic eyes that underwent SBP (p # 0.01). Tornambe criticized Han for lumping macula-on and macula-off RD together in his visual analysis, despite more maculaon eyes in the SBP group (48%) in contrast to the PR group (36%), possibly biasing the latter group in the visual analysis. He also questioned the familiarity of the surgeons to PR in Han’s series, because many of them contributed only a few cases each to the series.250 Han replied that the regression analysis utilized for the visual analysis had already adjusted for the baseline visual acuity, macular status, and other factors related to vision. Subsequent stratification of their eyes with RD according to macular status and pre-operative vision as suggested by Tornambe showed no significant differences in the CHAN ET AL visual outcome between SBP and PR eyes.104 Han also contended that surgical experience did not influence the outcome in their series, because the anatomic success rate was similar for the four surgeons who performed 88% of the PR cases versus the five surgeons who performed the remaining procedures.104,105,250 Table 2 summarizes the surgical outcomes of published reports of PR involving nine or more eyes for the repair of RD found in the world literature from 1986 to 2006.3,5,6,8,15,19,21,25,27,31,32,34,37--40,44,46, 47,49,51,52,62--66,68,69,71--75,77--82,89,95,97,98,100,104,105,111, 112,121,128,129,132,134,137,140,141,143,144,148,150,152,159,162, 164,166,167,168,169,172,175,178--181,185,191--194,200,202,209,214, 216,220,225,234,235,241--243,248,249,251,252,257,259,261,271,273, 275--278. This series includes 4,138 eyes among 81 reports from the world literature. The average single operation success was 74.4%, and the average final surgical success was 96.1%. There was an average of 11.7% of missed or new breaks, and 5.2% of proliferative vitreoretinopathy. Table 3 summarizes the visual outcomes of the same series found in the same 81 reports throughout this 21-year period, which varied from surgeon to surgeon. Although such a retrospective review summarizing a large number of clinical series consisting of dissimilar numbers of patients performed by multiple surgeons of diverse experience with PR in vastly different settings likely introduces multiple confounding variables (e.g., inconsistent and incongruous reports of visual outcome) that are difficult to control, it provides an accurate collective overview of PR on the world stage over the past two decades. Nevertheless, caution should be exercised when interpreting the results of many of these reports due their uncontrolled nature and small numbers of study subjects. The places of origin of these 81 reports include the following: West Coast USA (18), Midwest USA (9), East Coast USA (3), Western Europe (30), Eastern Europe (8), Asia (10), South America (1), Africa (1), and Middle East (1). COST ANALYSIS FOR PNEUMATIC RETINOPEXY VERSUS SCLERAL BUCKLING Hilton et al and Tornambe et al have repeatedly emphasized the cost-effectiveness of PR in comparison to scleral buckling and vitrectomy for repairing RD in multiple reports.40,110,240,241,243,248 In his retrospective review of 302 cases, Tornambe calculated that the average total cost of PR and subsequent reoperations for the entire series was 59% in comparison to SBP as the first procedure and subsequent reoperations. When the same comparison was made for selected cases of PR fitting the PR clinical trial criteria that underwent staged Surgical Outcome (1986 to 2007) Single Operation Success (SOS) Author 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Dominguez (1986) Hilton (1986) Dominquez(1987) Gnad(1987) Hilton (1987) Lemmen (1987) Menchini (1987) Van Effenterre (1987) Algvere (1988) Bovey(1988) Chen (1988) Lowe (1988) McAllister (1988) Menchini (1988) Mortensen (1988) Tornambe(1988) Tornambe(1988) Binder (1989) Brinton(1989) Chan (1989) Friberg(1989) Kelly (1989) Kusuki(1989)* Lemmen (1989) Liggett(1989) Miyakawa(1989) Dos Santo Motta (1989) Packo(1989) Poliner(1989) Rashed (1989) Sato (1989) Sato (1989)* Termote(1989) Terubayashi (1989) Vygantas (1989) Zakka (1989) Zegarra(1989) Skoog (1989) Location 4 Madrid, Spain Oakland, California1 Madrid, Spain4 Vienna, Austria4 Oakland, California1 Cologne, Germany4 Udine, Italy4 Paris, France4 Stockholm, Sweden4 Lausanne, Switzerland4 Portland, Oregon1 San Francisco, California1 Cleveland, Ohio2 Udine, Italy4 Odense, Denmark4 Poway, California1 Poway, California1 Vienna, Austria4 Oakland, California1 Loma Linda, California1 Pittsburq, Pennsylvania2 Sacramento, California1 Nagasaki, Japan6 Cologne, Germany4 Los Angeles, California1 Tokai, Japan6 Rio de Janeiro, Brazil7 Chicago, Illinois2 Poway, California1 Cairo, Eqypt8 Niigata, Japan6 Niigata, Japan6 Brussels, Belgium4 Kyoto, Japan6 Des Plaines, Illinois2 Los Angeles, California1 Cleveland, Ohio2 Linköping, Sweden4 Finals Success New Breaks No. Eyes No. Eyes % No. Eyes % No. Eyes 31 20 43 27 100 62 36 60 58 27 51 55 56 9 12 34 40 37 70 38 14 268 12 54 30 23 21 33 39 10 30 12 20 31 12 20 56 52 29 18 40 24 84 53 31 51 49 18 32 45 40 7 10 28 30 28 57 30 12 217 7 27 16 17 16 27 24 8 19 7 17 25 8 16 40 41 93.55 90.00 93.02 88.89 84.00 85.48 86.11 85.00 84.48 66.67 62.75 81.82 71.43 77.78 83.33 82.35 75.00 75.68 81.43 78.95 85.71 80.97 58.33 50.00 53.33 73.91 76.19 81.82 61.54 80.00 63.33 58.33 85.00 80.65 66.67 80.00 71.43 78.85 30 20 42 27 98 56 N/A 59 55 27 N/A 54 56 N/A 11 33 38 37 68 38 14 265 10 50 N/A 23 21 33 37 10 29 11 19 31 11 19 56 47 96.77 100.00 97.67 100.00 98.00 90.32’ N/A 98.33 94.83 100.00 N/A 98.18 l00.00 N/A 91.67 97.06 95.00 l00.00 97.14 l00.00 100.00 98.88 83.33 92.59 N/A 100.00 100.00 100.00 94.87 100.00 96.67 91.67 95.00 100.00 91.67 95.00 100.00 90.38 N/A 2 2 1 7 11 0 4 7 8 11 6 11 N/A N/A 4 5 5 11 5 N/A N/A 0 12 2 2 N/A 9 3 0 2 0 3 4 N/A 2 9 N/A % N/A 10.00 4.65 3.70 7.00 17.74 0.00 6.67 12.07 29.63 21.57 10.91 19.64 N/A N/A 11.76 12.50 13.51 15.71 13.16 N/A N/A 0.00 22.22 6.67 8.70 N/A 27.27 7.69 0.00 6.67 0.00 15.00 12.90 N/A 10.00 16.07 N/A PVR No. Eyes % N/A 1 1 0 3 3 0 1 6 2 5 3 1 N/A 1 3 4 3 3 0 0 N/A 0 3 1 2 N/A 0 6 0 1 0 1 2 N/A 1 1 4 N/A 5.00 2.33 0.00 3.00 4.84 0.00 1.67 10.34 7.41 9.80 5.45 1.79 N/A 8.33 8.82 10.00 8.11 4.29 0.00 0.00 N/A 0.00 5.56 3.33 8.70 N/A 0.00 15.38 0.00 3.33 0.00 5.00 6.45 N/A 5.00 1.79 7.69 457 (Continued) REVIEW: PNEUMATIC RETINOPEXY TABLE 2 458 TABLE 2 Continued Author Tornambe (1989) Ambler (1990) Berrod (1990) La Torre (1990)* Morin (1990) Pecold (1990) Santillo (1990)* Balestrazzi (1991) Emi (1991) Foscosi(1991) Pecold (1991) Tornambe (1991) Wedrich(1991) Algvere(1992) Bochow(1992) Pavlovic(1992) Freyler(1993) Sebag(1993) Trillo (1993) Grizzard (1995) Carstocea(1995) Gribomont(1996) Mulvihill (1996) Kishimoto(1997) Tornambe (1997) Sharma (1997) Fiser(1998) Han (1998) Lisle (1998) Assi (1999) Snady-McCoy(1999) Abecia (2000) Eter (2000) Friberg (2000) Kovacevic (2001) Kleinmann (2002) Böhm (2003) Chanq (2003) Yospaiboon (2005) Zaidi (2006) 1 Poway, California Cleveland, Ohio2 Nancy, France4 Florence, Italy4 Marseille, France4 Poznan, Poland5 Rome, Italy4 Roma, Italy4 Sakai, Japan6 Rome, Italy4 Poznan, Poland5 Poway, California1 Vienna, Austria4 Stockholm, Sweden4 St. Louis, Missouri2 Belgrade, Serbia5 Vienna, Austria4 Huntington Beach, California1 Genoa, Italy4 Tampa, Florida3 Bucarest, Romania5 Brussels, Belgium4 Dublin, Ireland4 Osaka, Japan6 Poway, California1 Madras, India6 Prague, Czekoslovakia5 Milwaukee, Wisconsin2 Odense, Denmark4 London, England4 Providence, Rhode Island3 Zaragoza, Spain4 Bonn, Germany4 Pittsburq, Pennsylvania2 Rijeka, Croatia5 Rehovot, Israel9 Bratislava, Slovakia5 Los Angeles, California1 Khon Kaen, Thailand6 San Francisco, California1 No. Eyes 103 101 56 23 29 35 29 9 54 37 40 93 24 51 17 46 89 45 55 107 16 63 10 67 302 36 13 50 36 31 32 219 78 40 28 44 30 11 116 61 No. Eyes 75 78 37 17 26 29 17 6 38 30 32 N/A 17 44 12 27 65 38 46 74 7 49 7 30 205 25 11 31 30 19 19 179 51 36 22 33 23 9 76 33 New Breaks PVR % No. Eyes % No. Eyes % No. Eyes % 72.82 77.23 66.07 73.91 89.66 82.86 58.62 66.67 70.37 81.08 80.00 N/A 70.83 86.27 70.59 58.70 73.03 84.44 83.64 69.16 43.75 77.78 70.00 44.78 67.88 69.44 84.62 62.00 83.33 61.29 59.38 81.74 65.38 90.00 78.57 75.00 76.67 81.82 65.52% 54.10 102 101 N/A N/A N/A 32 N/A 9 53 37 35 93 24 50 17 45 89 45 53 105 7 63 9 62 286 33 13 49 35 26 32 215 77 40 N/A 42 28 11 107 61 99.03 100.00 N/A N/A N/A 91.43 N/A 100.00 98.15 100.00 87.50 100.00 100.00 98.04 100.00 97.83 100.00 100.00 96.36 98.13 43.75 100.00 90.00 92.54 94.70 91.67 100.00 98.00 97.22 83.87 100.00 98.17 98.72 100.00 N/A 95.45 93.33 100.00 92.24 100.00 24 11 10 0 N/A 0 N/A 1 2 5 2 N/A 1 3 3 N/A 24 4 9 16 N/A 11 1 N/A 99 0 1 N/A 5 7 N/A 18 4 5 3 2 N/A 0 N/A 13 23.30 10.89 17.86 0.00 N/A 0.00 N/A 11.11 3.70 13.51 5.00 N/A 4.17 5.88 17.65 N/A 26.97 8.89 16.36 14.95 N/A 17.46 10.00 N/A 32.78 0.00 7.69 N/A 13.89 22.58 N/A 8.22 5.13 12.50 10.71 4.55 N/A 0.00 N/A 21.31% 3 1 9 0 2 4 2 0 0 1 0 N/A 2 1 2 N/A 4 1 2 5 0 8 2 N/A 29 1 1 6 1 4 0 8 4 0 3 6 N/A 1 N/A 0 2.91 0.99 16.07 0.00 6.90 11.43 6.90 0.00 0.00 2.70 0.00 N/A 8.33 1.96 11.76 N/A 4.49 2.22 3.64 4.67 0.00 12.70 20.00 N/A 9.60 2.78 7.69 12.00 2.78 12.90 0.00 3.65 5.13 0.00 10.71 13.64 N/A 9.09 N/A 0.00 CHAN ET AL 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 Location Finals Success Surv Ophthalmol 53 (5) September--October 2008 Single Operation Success (SOS) 459 5.2 360-degree retinopexy, the average cost was as little as 28% of cases undergoing SBP first.241 EVIDENCE-BASED ANALYSIS OF PNEUMATIC RETINOPEXY VERSUS SCLERAL BUCKLING 11.7 Sharma and Robbins used utility analysis methodology to analyze the quality of life associated with PR versus SBP, in order to compare the costeffectiveness of the two procedures.218 Using the Markov model and a Monte Carlo simulation, they found PR confers an improvement of 1.86 qualityadjusted life months (QALMs), a 7.4% increase in comparison to SBP during a 3-year period. Thus, their calculations based on combining all VA outcomes and clinical complications from the PR multicenter clinical trial into a single analysis determined that PR is the preferred treatment method over SBP. 96.1 19 148 40 3728 51.8 100.00 98.67 100.00 2 26 4 464 7.4 10.50 17.30 21.00 0 5 2 182 2.6 0.00 3.30 5.00 REVIEW: PNEUMATIC RETINOPEXY Complications and Their Management 74.4 Minimum of 9 eyes per series for inclusion; N/A5not available. *Series involving macular holes 1 Western US(18) 2 Midwest US (9) 3 East Coast US (3) 4 Western Europe (30) 5 Eastern Europe (8) 6 Asia (10) 7 South America (1) 8 Africa(1) 9 Middle East (1). 79 80 81 Chen (2007) Kulkarni (2007) Yanyali (2007) Chang-hua, Taiwan6 New Brunswick, New Jersey3 Istanbul, Turkey5 Total Average 19 150 40 4138 16 115 25 3010 37.6 84.21 76.67 70.00 NEW OR MISSED BREAKS AND SURGICAL FAILURE In their survey of 26 reports with 1,274 patients that underwent pneumatic retinopexy, Brinton and Hilton reported a 13% incidence of new or missed retinal breaks.40,242,243 Some investigators have attempted to differentiate between ‘‘new’’ from ‘‘missed’’ retinal breaks.40,76,114,195,241 According to their definition, retinal breaks found in area of previous RD during repeat surgery not detected during the initial PR constitute ‘‘missed’’ breaks. However, breaks located in an area of new retinal detachment distant from the original breaks are considered ‘‘new’’ breaks. In a retrospective review of 13 cases, Poliner et al reported two cases of new breaks associated with extension of the RD to previously uninvolved quadrants within two days after PR.195 They ascribed the early new retinal breaks and RD to traction on condensed vitreous distal to the sites of the original breaks following intraocular gas injection. In 1988, Chen et al reported new breaks occurring in 11 of 31 eyes (21%) after PR.51 In 1989, Tornambe et al also reported 23% new or missed breaks for the PR eyes in contrast to 13% of this complication for SBP eyes in their multicenter PR study.243 Tornambe’s subsequent study of 302 eyes reported 14% missed breaks and 19% new breaks.241 These rates are in contrast to the historical rates of 1.8% to 20.8% after non-pneumatic retinopexy techniques for repairing retinal breaks and retinal detachment in previous reports.26,35,55,57,83,94,97, 157,158,199,204,219,222,266 New or missed breaks may develop in any quadrants. However, 76% are located in the superior 8 clock hours of the retina, and 52% are found within 3 clock hours of the original 460 TABLE 3 Visual Outcome (1986 to 2007) Authors 1 2 Dominquez (1986) Hilton (1986) No. Eyes 51 55 35% $ 20/50 54.8% $ 20/50 71% $ Pre-op N/A McAllister (1988) Menchini (1988) Morternsen (1988) 56 9 12 69% $ 20/50 N/A N/A 25% $ Pre-op N/A N/A 16 17 18 19 20 Tornambe (1988) Tornambe (1988) Binder (1989) Brinton (1989) Chan (1989) 34 40 37 70 38 N/A 79% $ Pre-op N/A N/A N/A N/A 58% $ Pre-op N/A N/A N/A 21 22 23 Friberg (1989) Kelly (1989) Kusuki (1989)* 14 268 12 N/A N/A N/A N/A N/A N/A 24 Lemmen (1989) 54 25 26 27 Liggett (1989) Miyakawa (1989) Motta (1989) 30 23 21 28.0% Post-op VA $ 20/50: 92% O Pre-op N/A 80% $ 20/50 N/A 87.1% $ 20/50; 41.9% $ Pre-OP; 29% O Pre-op N/A 61.5% $ 20/50 N/A 28 29 Packo (1989) Poliner (1989) 33 39 N/A N/A N/A N/A 11 12 Chen (1988) Lowe (1988) 13 14 15 43 27 100 62 36 60 58 27 Other Categorization N/A N/A N/A N/A N/A N/A N/A N/A N/A Post-op VA O Pre-op VA 74.1 %; Post-op VA 5 Pre-op VA 14.8 %; Post-op VA ! Pre-op VA 11.1 % N/A 83.6% O Pre-op VA; 10.9% 5 Pre-op; 5.5% # Pre-op N/A Post-op VA O Pre-op VA 5 100% Post-op O Pre-op VA 5 83.3%; Post-op VA $ 20/40 5 66.7%; Post-op VA $ 20/120 5 16.7% N/A N/A N/A N/A (SOS) Median VA 5 20/30; (Re-op) Median VA 5 20/70 N/A N/A 66.67% O Pre-op VA; 16:67% 5 Pre-op VA; 16.67% ! Pre-op N/A N/A 65.2% $ 20/50 for entire series All SOS eyes final VA $ Pre-op; 5 Re-op eyes final VA: 2 $ Pre-op; 3 ! Pre-op N/A N/A CHAN ET AL N/A N/A RD, 65% $ 20/50 N/A N/A 60% $ 20/40 74%, VA O 20/50 N/A Dominguez (1987) Gnad (1987) Hilton (1987) Lemmen (1987) Menchini (1987) Van Effenterre (1987) Algvere (1988) Bovey (1988) N/A 50% $ 20/70; 50% $ 20/200 Macula-on N/A 50% Post-op VA unchanged; 20% Post-op VA improved 2 or 3 lines N/A N/A RD, 41% $ Pre-op VA N/A N/A 20/100 # 60% # 20/40 N/A N/A 3 4 5 6 7 8 9 10 31 20 Macula-off Surv Ophthalmol 53 (5) September--October 2008 Visual Outcome Rashed (1989) Sato (1989) Sato (1989)* Termote (1989) Terubayashi (1989) Vygantas (1989) Zakka (1989) Zegarra (1989) Skoog (1989) Tornambe (1989) 10 30 12 20 31 12 20 56 52 103 40 41 42 43 44 45 46 47 48 49 50 51 52 53 Ambler (1990) Berrod (1990) La Torre (1990)* Morin (1990) Pecold (1990) Santillo (1990)* Balestrazzi (1991) Emi (1991) Foscosi(1991) Pecold (1991) Tornambe (1991) Wedrich Algvere(1992) Bochow (1992) 101 56 23 29 35 29 9 54 37 40 93 24 51 17 54 55 56 Pavlovic (1992) Freyler (1993) Sebag (1993) 46 89 45 57 58 Trillo (1993) Grizzard (1995) 55 107 59 60 61 Carstocea (1995) Gribomont (1996) Mulvihill (1996) 62 63 64 Kishimoto (1997) Tornambe (1997) Sharma (1997) 16 63 10 N/A 302 36 N/A N/A N/A 80% $ 20/50 N/A N/A N/A N/A N/A (#14days)80%$ 20/50; (#3 days) 91% $ 20/50 66% O 20/50 N/A N/A N/A N/A N/A N/A 72.7% $ 20/50 N/A N/A (#14 days) 89% $ 20/50 28.5%: 20/50-20/100; 43% $ 20/40 65% $ 20/40 87.5% VA OPre-op; VA 2 or O lines N/A N/A N/A Macular hole series; Final VA 20/60 to 20/240 N/A N/A N/A N/A N/A N/A N/A 97% O Pre-op VA N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 66.7% O 20/40; 88.9% O Pre-op VA 46% O 20/22 N/A N/A N/A N/A N/A N/A N/A N/A 85 % O 20/50 N/A N/A N/A N/A N/A N/A 87% O 20/50 N/A N/A 94% $ 20/40 90%: 20/25-20/20 N/A 77.8% 5 Pre-op VA; 22.2% O Pre-op VA N/A N/A N/A Post-op O Pre-op VA N/A Pre-op 5 Post-op VA N/A N/A N/A N/A N/A 91.5% $ Pre-op VA N/A N/A 66% $ 20/40 N/A N/A 85% $ 20/40 N/A REVIEW: PNEUMATIC RETINOPEXY 30 31 32 33 34 35 36 37 38 39 Post-op VA $ 20/50 5 56.5% Post-op VA $ 20/40 5 57.3% Post-op VA O Pre-op 2 lines or more 66.7%; Post-op VA 5 Pre-op VA 31.1%; Post-op VA ! Pre-op VA 2.2% N/A 73.9% $ 20/30; VA # 20/400 for 9.45% SOS, 27.3% Re-op (P #0.05); age: negative correlation to VA (r 5 "0.37, P # 0.001); þ correlation for Pre-op & Post-op VA (r 5 0.5, P #0.001). N/A 95%$ Pre-op VA: 92% $ 20/50 80% Post-op VA O Pre-op VA; 20% Post-op VA ! Pre-op VA N/A N/A Post-op VA $ 20/60 5 47.2%;Successful eyes: Post-op VA $ Pre-op 5 66.7% 461 (Continued) 462 TABLE 3 Continued Visual Outcome No. Eyes 65 Fiser (1998) 13 66 67 68 69 70 71 Han (1998) Lisle (1998) Assi (1999) Snady-McCoy (1999) Abecia (2000) Eter (2000) 72 73 74 75 76 Friberg (2000) Kovacevic (2001) Kleinmann (2002) Böhm (2003) Chang(2003) 40 28 44 30 11 77 Yospaiboon(2005) 78 Macula-off Macula-on Single eye 5 20/67 91.7% (11/12 eyes) $ 20/40 47% $ 20/50 65% $ 20/50 45 % $ 20/40 90% $ 20/50 33.3% $ 20/40 (SOS) Mean VA 5 20/35; (Re-op) Mean VA 5 20/56 N/A N/A N/A 66.7% $ 20/50 N/A 76% $ 20/50 23.1 % $ 20/50 100% $ Pre-op N/A 56.2% $ 20/40 (SOS) Mean VA 5 20/29; (Re-op) Mean VA 5 20/31 N/A N/A N/A 80% $ 20/50 N/A 116 N/A N/A Zaidi (2006) 61 N/A N/A 79 Chen(2007) 16 84.6% $ 20/40 80 Kulkarni(2007) 150 77.6% $ 20/50 83.3%5 20/70 16.7% 5 20/70 83% $ 20/50 81 Yanyali(2007) N/A N/A 50 36 31 32 219 78 40 Other Categorization Post-op Post-op Post-op Post-op N/A N/A N/A Post-op N/A N/A VA VA VA VA $ 20/29 5 76.9%; O Pre-op VA 5 61.5%; 5 Pre-op VA 5 30.8 %; ! Pre-op VA 5 7.7% VA $ 20/40 5 81.3 % N/A N/A Post-op VA O 20/60 5 61.9% N/A Post-op O Pre-op 100%; Post-op mean VA 5 3 lines O Pre-op 84.1% with improved VA; 26.1%$ 20/60; 54.2% $ 20/60 to 20/200 Mean final VA for SOS eyes 5 20/25; Mean final VA for Re-op eves 5 20/100 65.4%: pre-op O post-op 31.6%: pre-op 5 post-op Macula off: 88.4% $ 20/50 for SOS eyes Macula on: 87.8% $ 20/50 for SOS eyes Basic group: Final VA: 20/80 Steamroller group: 20/63 Surv Ophthalmol 53 (5) September--October 2008 Authors N/A Not available; RD 5 retinal detachment; SOS 5 single-operation success. series involving macular holes. * CHAN ET AL 463 REVIEW: PNEUMATIC RETINOPEXY causative break(s).117 In their multicenter prospective study, Tornambe et al attributed the majority of new breaks to initial incomplete posterior vitreous detachment, because 59% of new breaks develop during the first postoperative month.243 New or missed retinal breaks invariably induce reaccumulation of subretinal fluid. If not managed appropriately, they result in surgical failure. New breaks located in the superior quadrants and not widely separated (more than 1 or 2 clock hours apart) associated with limited subretinal fluid, may be successfully treated with additional cryotherapy or laser and gas injection. Otherwise, scleral buckling or vitrectomy is frequently required to repair the surgical failure. In his report of 302 eyes over a 12-year period, Tornambe found that risk factors associated with increased tendency for surgical failure include: aphakia and pseudophakia, multiple retinal breaks, and detachment involving more than 50% of the retina.241 He advocated circumferential (360-degree) laser retinopexy along the vitreous base and ora serrata to decrease the incidence of missed or new breaks and enhance the rate of surgical success.241 He reported 85% single operation success for PR after 360-degree retinopexy, in contrast to 55% after focal retinopexy alone. Critics of PR have cited Lincoff’s study on cynomulgus monkeys’ eyes that showed the stretching and tearing of vitreous lamellae by an expanding intraocular gas bubble, leading to preretinal membrane formation.155 Lincoff et al also found reduction of concentration of vitreous hyaluronic acid following intravitreal injection of an expansile gas bubble in rabbit eyes. They reported a decrease of 18.6% of hyaluronic acid concentration after a 90% vitreous displacement by the expansile gas.155 However, proponents of PR have emphasized that such ultrastructural and biochemical changes of the vitreous are only of theoretical concern and have not resulted in adverse clinical manifestations associated with PR. They also indicated that although PR may be associated with a higher incidence of new breaks, the final anatomical and visual outcomes of the operated eyes are not adversely affected, and that the frequency of PVR is equal to SBP (see subsequent section on PVR).6,105,114,115,131,240,241,243,248 Rarely, a giant retinal tear may form after pneumatic retinopexy.270 CATARACT PROGRESSION Cataract development or progression is uncommon after PR.114,115,131,135,136,184,243,267 In their multicenter randomized trial comparing PR with encircling scleral buckling, the RD Study Group reported cataract development for only one eye after PR, due to anterior lens capsular touch during paracentesis.243 In their report of 2-year follow up of the multicenter PR trial, Tornambe et al indicated cataract progression in 4% of eyes after PR, in contrast to 18% of eyes after SBP.248 Tornambe also reported in his 12-year analysis of 302 cases the overall incidence of cataracts to be within 1% after PR.241 In their prospective study published in 1994, Mougharbel et al confirmed the general lack of cataract progression after PR.184 By using slit-lamp examination and Scheimpflug photography, Koch et al detected small but insignificant degradation of lens transparency in eyes that underwent PR in comparison to non-operated eyes after 2 months of follow-up.135 After 6 months, the difference became statistically significant for the anterior lens cortex.136 However, subsequent study by the same group of investigators showed no correlation between PR and cataract formation or progression in 33 eyes that were followed prospectively for 2 years after PR.184 This complication can be minimized by careful placement of the paracentesis needle away from the lens and the avoidance of contact between the intraocular gas and the lens by proper head positioning after PR. SUBCONJUNCTIVAL GAS Subconjunctival gas is an infrequent occurrence due to either incomplete penetration of the sclera by the needle for gas injection or outward leakage of gas through the needle track for the gas injection.114,131,267 Hilton et al reported this complication in only three eyes in a review of 1,274 cases of PR.114 This complication is usually associated with insignificant clinical effects. A small amount of subconjunctival gas usually resolves without clinical consequences. A large amount of subconjunctival gas can be easily released from the subconjunctival space with a needle or a small conjunctival incision. To minimize this complication, the surgeon should ensure proper insertion of the needle for gas injection into the vitreous cavity, and immediately place a cotton applicator on the needle track during withdrawal of the injection needle.114,131,267 Rotating the head immediately after withdrawal of the needle so that the injection site is no longer at the highest position may also move the intraocular gas bubble away from the injection site and reduce the tendency for the leakage of the injected gas from the injection site into the subconjunctival space. VITREOUS LOSS Friberg et al reported a small bead of formed vitreous on the globe surface corresponding to the external opening of the needle track following pneumatic retinopexy in three cases.82,114 Factors 464 Surv Ophthalmol 53 (5) September--October 2008 that may lead to subconjunctival gas may also lead to this minor complication. Thus the prevention of this complication can be accomplished by taking the same precautions as stated in the preceding paragraph. SUBRETINAL GAS ‘‘Fish-egg’’ gas bubbles may migrate through retinal breaks into the subretinal space, particularly in the presence of large breaks or breaks with rigid or scrolled edges.40,114,131,137,171,241,243,267 All six cases reported by McDonald et al involved ‘‘fishegg’’ bubbles.171 Hilton indicated that any eyes with large retinal breaks larger than 1 clock hour in size are associated with an increased risk of subretinal gas after PR.40,114,131 An expanding subretinal gas bubble may reopen a retinal break. Injection of the gas bubble at an uppermost location of the globe and in a brisk fashion reduces the chance of forming ‘‘fish-egg’’ bubbles (see the section Gas Injection Technique).40,114,131,137,240,267 In the presence of a large retinal break (more than 1 clock hour), the patient’s head should be tilted so that the injection site is uppermost and away from the break. Thus, any resultant intraocular ‘‘fish-egg’’ bubbles would not rise toward the retinal break. For an eye with intravitreal fish-egg bubbles in the presence of a large horizontal tear, the patient’s head should be positioned in such a way that the bubbles migrate away from the tear. The surgeon should wait for the ‘‘fish-egg’’ bubbles to coalesce (usually taking 24 hours) before allowing the gas to contact the retinal break. This maneuver does not work so well when a large retinal tear is located at the superior fundus. In that situation, the steam-roller technique is used to flatten the large retinal tear against the RPE first, before allowing gas contact with the tear. Although the flicking of the globe with moderate firmness has been advocated to be a useful technique for breaking up ‘‘fish-egg’’ bubbles, the authors have found this maneuver to be frequently ineffective. Upon the migration of ‘‘fish-egg’’ bubbles under the retina, certain correctives measures may be taken. Placing the patient in a supine position and tilting his or her head so that the break is at the most superior position, followed by gentle scleral depression on the globe may release the subretinal gas into the vitreous cavity. Three of McDonald’s cases were managed successfully with this method.171 Corrective measures should be taken immediately, because it is virtually impossible to displace the trapped gas from the subretinal space once the ‘‘fishegg’’ bubbles coalesce and expand in the subretinal space. Under certain circumstances, persistent small subretinal gas bubbles can be managed by scleral CHAN ET AL buckling alone. In 1988, Lewen and Eifig reported successful repair of a retinal detachment with a small subretinal gas bubble under a superior retinal break at 12 o’clock with an encircling 240 band augmented with a superior 279 circumferential segmental buckling element and intravitreal air, after subretinal fluid drainage.151 Finally, a vitrectomy with internal drainage is the most direct and effective way of eliminating the trapped subretinal gas. This approach is particularly appropriate in the case of a pseudophakic or aphakic eye without the risk of post-vitrectomy cataract progression. Three of McDonald’s cases required a vitrectomy, and one case underwent scleral buckling for management of this complication.171 GAS ENTRAPMENT IN PRE-HYALOID SPACE Occasionally, the injected gas is trapped in the shallow pre-hyaloid or sub-pars plana space. This potential space (Canal of Petit) is bordered centrally and anteriorly by Wieger’s hyaloideocapsular ligament and the lateral and posterior lens surface, laterally by the anterior pars plana epithelium, the pars plicata of the ciliary body, and the lens zonules, and posteriorly by the anterior vitreous hyaloid face (Fig. 5). 40,110,114,115,131,231,240,264,267 When viewed through the lens, the trapped bubble appears as a partial ring behind the peripheral edge of the lens, otherwise known as the ‘‘sausage sign’’ or ‘‘donut sign.’’40,231,264,267 No specific treatment is required for a small amount of trapped gas, which should dissipate rapidly.114 Having the patient assume a head prone position may allow a large and expanding trapped bubble to break through the anterior hyaloid face into the vitreous cavity and rise toward the retinal breaks. Maintaining the facedown posture overnight almost always resolves this problem. Under certain circumstances, the trapped bubble is removed by a method described by Hilton.40,110,114,115,131,231,240,264,267 This maneuver involves the insertion of a short 27-gauge needle connected to a plungerless tuberculin syringe filled with some balanced salt solution or saline through the eye wall into the trapped bubble. The patient’s head should be tilted so that the needle penetration site is uppermost. Bubbling of gas through the fluid in the syringe indicates the successful elimination of the trapped gas.40,110,114,115,131,231,240,264,267 To reduce the chance of repeating the same complication, re-injection of gas into the vitreous after release of trapped pre-hyaloid gas should be performed at an area away from the original site of complication.114 To prevent this complication, the surgeon must verify the penetration of the vitreous cortex with the inserted needle before gas injection during pneumatic retinopexy.231,233,264 465 REVIEW: PNEUMATIC RETINOPEXY minimize such a complication (see preceding paragraphs on Gas Injection Technique and Intraocular Pressure Management). ENDOPHTHALMITIS Fig. 5. Inadvertent trapping of gas in the pre-hyaloid space (Canal of Petit) may result in the ‘‘donut sign’’ or ‘‘sausage sign.’’ Positioning the needle deeper into the vitreous cavity before gas injection may also reduce the chance of this complication, although deeper injection may result in ‘‘fish-egg’’ bubbles that may migrate through a large retinal tear into the subretinal space (see preceding section). ANTERIOR CHAMBER GAS ENTRAPMENT AFTER PNEUMATIC RETINOPEXY This complication is an extension of pre-hyaloidal gas entrapment. In 2001, Taher and Haimovici reported a case of intravitreal gas migration into the anterior chamber of a phakic eye that underwent pneumatic retinopexy.233 They theorized that the gas was inadvertently injected into the potential space between the lens and the anterior hyaloid face (Canal of Petit) as described by Steinmetz et al (see previous paragraph).231,233 They further conjectured that a forceful injection of a relatively large volume of gas (0.6 ml) into this small space created a pressure gradient that led to zonular dehiscence and entrapment of the gas in the anterior chamber. Taher et al recommended anterior chamber paracentesis to evacuate the trapped gas in order to avoid excessive intraocular pressure rise and gas-induced lenticular opacities.233 The same measures that reduce the tendency for entrapment of pre-hyaloid gas also decrease the likelihood of this complication (see preceding paragraph). In addition, keeping the patient in an upright position during gas injection may also Similar to scleral buckling for repairing RD, the incidence of endophthalmitis following PR is very low.67,114,115,131,243,267 The rates of endophthalmitis among these series varied. In their multicenter randomized study of PR versus scleral buckling, Tornambe et al reported only a single case of staphylococcal endophthalmitis for the 103 eyes that underwent PR.243 They attributed improper sterilizing techniques (including diluted Betadine) to have contributed to the infectious process for that case. In 1987, Eckhardt reported a case of endophthalmitis due to Staphylococcus epidermidis following PR.67 He attributed the cause of the infection to entry of the organism through the scleral injection site for that case. Meticulous attention to sterile surgical techniques and postoperative topical antibiotics are measures that reduce the likelihood of this complication. MACULAR HOLE AND OTHER NEW POSTERIOR BREAKS Macular hole development has been reported to occur after retinal detachment repairs by various surgical procedures.16,40,107,161,183,208,221,243,254 Surgical procedures associated with the formation of a macular hole include not only pneumatic retinopexy, but also SBP, vitrectomy, and even after YAG laser for posterior capsulotomy.16,28,41,85,107,161,183,208,221, 243,254 In 1988, Runge et al and Avins et al presented case reports of macular hole formation following pneumatic retinopexy.16,208 Vitreomacular traction during the perioperative period is thought to play a role in its pathogenesis.16,208 Those eyes that developed a macular hole after PR were found to have vitreomacular attachment before surgery and vitreomacular detachment after surgery induced by the expanding intraocular gas. It has been speculated that the fellow eye of patients that previously developed a macular hole may have greater tendency for macular hole formation after pneumatic retinopexy.16,267 The most likely mechanism for macular hole formation in those eyes is thought to be the tracking of an expanding gas bubble between the retina and the posterior vitreous hyaloid face toward the macula.16,208 A secondary mechanism involves the stress induced on a fragile macula by the shifting of a large amount of subretinal fluid under it due to the expanding gas bubble.208 Runge et al advocated careful biomicroscopic evaluation of the 466 Surv Ophthalmol 53 (5) September--October 2008 vitreomacular relationship followed by careful avoidance of injecting gas under the posterior hyaloid face to prevent the development of a macular hole related to the first mechanism.208 In case of significant vitreomacular traction, they suggested the technique of liquid vitreous--gas exchange described by Blankenship et al for myopic eyes with a retinal detachment associated with a macular hole.29,208 To reduce the possibility of the second mechanism, the steam roller maneuver can be attempted.208 In his series of 302 eyes, Tornambe reported a 1% incidence of macular hole formation after PR.241 Conventional surgical techniques for repairing idiopathic macular holes can be utilized with good successes for treating macular holes related to pneumatic retinopexy.53,107,130 Hejny et al reported favorable outcome with vitrectomy for managing a macular hole that developed after pneumatic retinopexy.107 In 1988, Freeman and coworkers reported three cases of new posteriorly located retinal breaks other than a macular hole following PR.76 Similar to Runge et al, they believed the pathogenesis of this occurrence to be related to preoperative incomplete posterior vitreous detachment (PVD) and the intraoperative or postoperative tracking of gas under the posterior hyaloid face. They reported an incidence of 4--6% of this complication after PR. All three of their cases were successfully treated with additional surgery, resulting in good visual outcome. In 2002, Sharma reported a case of posterior radial extension of the nasal margin of a retinal break within 3 weeks after PR using C3F8, likely due to retrograde vitreous dissection under the posterior hyaloid face by the expanding gas bubble.215 CHAN ET AL UVEITIS AND VITREOUS HAZE Uveitis after PR is usually mild and of no lasting clinical relevance.114,131,232,257 There has been no report of severe uveitis associated with PR. Hilton and Tornambe reported one case of mild uveitis following PR that cleared spontaneously within 3 weeks.114,115,131,243,267 Van Effenterre et al reported six eyes that developed vitreous haze for 3 to 8 days after PR.257 Judicious application of cryotherapy or laser should reduce the occurrence of postoperative uveitis. CYSTOID MACULAR EDEMA The occurrence of cystoid macular edema (CME) is uncommon after PR.117,120,137,255,267 Treatment for CME associated with PR is similar to treatment for CME associated with SBP. Those eyes with uveitis and increased vitreous pigmentary dusting may have greater tendency for postoperative CME. Avoidance of excessive retinopexy during PR may reduce this complication. Topical steroidal and nonsteroidal anti-inflammatory medications and periocular steroidal medical therapy are usually effective for treating this complication. In 2007, Tunc et al reported an 11% incidence of CME following PR, in contrast to 29% after scleral buckling.255 They noted a significant correlation of pseudophakia with CME for the PR eyes (27%, p 5 0.02). The lower incidence of CME associated with pneumatic retinopexy may explain the more frequent and rapid visual recovery of pneumatic retinopexy in comparison to scleral buckling.176,241,267 INTRAOCULAR HEMORRHAGE Intraocular hemorrhage due to PR is rare.114, Temporary hyphema developed in one eye following PR in the multicenter PR study.243 One eye also developed vitreous hemorrhage that resolved spontaneously with a final visual acuity of 20/ 20. Hilton and Tornambe advocated intraocular gas injections away from the anterior ciliary vessels along the vertical and horizontal meridians, in order to reduce the chance of ocular surface and vitreous hemorrhage.40,114,115,131,243,267 In the multicenter PR trial, peripheral subretinal hemorrhage developed in one eye, possibly due to the fracturing of intraocular vessels by the release of the cryoprobe before its appropriate thawing.243 The final visual outcome was still excellent (20/40). There has been no report of submacular hemorrhage after PR. Careful handling of the cryoprobe is recommended to reduce this potential complication. The cryoprobe should not be removed from the eye until appropriate thawing of the ice ball on the cryoprobe and the surrounding surface of the globe is 115,243,267 INCARCERATION OF IRIS AND VITREOUS The sudden release of aqueous during paracentesis in the presence of marked intraocular pressure elevation after gas injection may lead to iris and vitreous incarceration at the paracentesis wound and anterior vitreous prolapse, particularly if the patient is in the supine position.114,115,131,243,267 This complication may be prevented by one or more of the following measures: 40,114,115,131,267 1) performing paracentesis via the limbus before gas injection; 2) creating a small paracentesis wound and sealing the wound with a cotton applicator upon the release of the paracentesis needle; 3) performing the paracentesis via the pars plana instead of limbus to remove either aqueous or vitreous; or 4) keeping the patient in an upright position during the paracentesis (see preceding paragraphs on Gas Injection Technique and Intraocular Pressure Management). 467 REVIEW: PNEUMATIC RETINOPEXY evident.114,115,131,267 The rare occurrence of subretinal hemorrhage associated with PR is in contrast to a 3.0--4.5% rate of the same complication associated with SBP.55,110,114,116 VITREOUS AND SUBRETINAL PIGMENT RELEASE The release of pigment granules into the vitreous cavity and subretinal space occurs on a routine basis during the process of a rhegmatogenous RD.109,131 In fact, pigment ‘‘fall-out’’ under the retina and ‘‘tobacco dusting’’ of pigments in the vitreous cavity may be encountered after both SBP and PR for the repair of RD.109,131 The amount of pigment release can be influenced by the extent of cryotherapy and laser during surgery.43,109,131 Although pigment release is frequently a benign phenomenon, too much pigment release may increase the risk of PVR.43,96 Thus excessive cryotherapy or laser should be avoided during PR. As discussed in the preceding section, the steamroller technique may enhance pigment migration into the vitreous cavity, and therefore should be applied judiciously (please also see the subsequent section on proliferative vitreoretinopathy).39,40,110,247 EPIMACULAR FIBROSIS AND MACULAR PUCKER The incidence of epimacular fibrosis (EMF) after PR is relatively low.105,112,114,131,240--243 The multicenter PR study reported a 4% incidence of asymptomatic EMF (mild macular pucker) for the 103 eyes that underwent PR.243 In contrast, asymptomatic EMF developed in 3% of eyes and symptomatic EMF (severe macular pucker) developed in 2% of eyes that underwent scleral buckling in that study. Similar to the case of PVR, avoidance of excessive cryotherapy and laser treatment may minimize the tendency for this complication. In 1989, Sato et al analyzed 42 eyes with RD that underwent PR with and without cryopexy.211 They found that of the 20 eyes that underwent cryopexy, 5 eyes (25%) developed a macular pucker, whereas none of the non-cryopexy eyes developed a macular pucker (p ! 0.05). They proposed the potential cause of this difference to be the excessive release of RPE cells into the subretinal space and vitreous cavity.211 However, Saran and Brucker found no statistical difference in the frequency of epimacular membranes in eyes with retinal breaks treated with cryopexy versus laser.210 The management of macular pucker after PR is similar to the treatment of this condition associated with other clinical situations, with usually good outcome. MACULAR FOLDS Macular folds may rarely occur when a superiorly positioned intraocular gas bubble displaces large amount of subretinal fluid associated with a bullous retinal detachment inferiorly, resulting in retinal redundancy involving the posterior fundus.267 The steamroller maneuver is most effective in avoiding this complication (See preceding paragraph on this technique under the section of Surgical Techniques).267 PROLIFERATIVE VITREORETINOPATHY The reported rates of PVR after PR varied depending on the clinical series.39,40,105,110,114, 115,131,240,242,243,267 The multicenter PR trial reported a 3% incidence of PVR for the PR eyes in comparison to a 5% incidence of the same for the SBP eyes.243 There was no statistical difference in the incidence between the two groups. In their summary of 1,274 eyes in 26 published reports of PR, Hilton et al cited an average rate of 4% for PVR after PR.114 However, Chen et al reported a 9.8% incidence of PVR in a prospective series of 51 eyes that underwent PR.51 In his review of 302 eyes that underwent PR over 12 years, Tornambe also noted a 9.6% incidence of PVR associated with PR.241 He found no difference in the incidence of PVR between those eyes that underwent cryotherapy versus laser in his series. He also found a lack of influence in the size and the duration of the intraocular gas bubble in the incidence of PVR.131,241 Similar to PVR associated with other surgical techniques for repairing RD, eyes with PVR after PR can be managed successfully with vitrectomy and membrane stripping with or without SBP, and gas or silicone oil tamponade. Campochiaro et al and Griffiths and Richardson speculated excessive release of subretinal pigments into the vitreous cavity to be the most likely cause for PVR after PR.43,96 Thus the avoidance of excessive cryotherapy or laser treatment, and the application of retinopexy after flattening of the retina with gas in case of large amount of preoperative subretinal fluid are recommended.39,40,110,112,114 CHOROIDAL DETACHMENT Choroidal detachment is uncommon after PR.114,131,240,242,243 Tornambe et al reported an incidence of 3% of choroidal detachment for the PR eyes in comparison to an incidence of 17% for the scleral buckling eyes in the multicenter PR study.243 The difference in the incidence of choroidal detachment between the two groups was statistically significant (p 5 0.001)114,243 However, none of those cases resulted in any clinically 468 Surv Ophthalmol 53 (5) September--October 2008 significant adverse outcome (e.g., angle closure).114,131,243 In his review of 302 cases performed over 12 years, Tornambe reported an overall 2% incidence of choroidal detachment after PR.241 Mild choroidal detachment after PR usually resolves spontaneously without any clinical consequences, as in all of such cases in the multicenter PR study.114,131,243 This complication may be reduced by avoiding hypotony due to too much aqueous release during paracentesis and excessive cryotherapy.114,131 In 2000, Baker and Hainsworth reported a case of suprachoroidal gas associated with pneumatic retinopexy.18 Jabaly-Habib et al also reported two similar cases of unintentional gas injections into the suprachoroidal space during pneumatic retinopexy in 2003.123 Prolonged and severe pain shortly after suprachoroidal gas injection is a typical manifestation of this complication.123 Re-tapping the same injection site with needle connected to a plunger-less syringe may allow immediate removal of gas trapped in the suprachoroidal space. This rare complication may be prevented by meticulous techniques, especially under the guidance of indirect ophthalmoscopy to ensure the location of the needle tip in the vitreous cavity during gas injection. MYOPIC SHIFT Tornambe et al noted a 68% incidence of myopic shift of 1 diopter or more after SBP in comparison to 3% of the same after PR in the multicenter PR trial.242,243 This high rate associated with SBP was largely related to the utilization of an encircling element for buckling cases in that trial.114,243 The rate is expected to be much lower with segmental instead of encircling scleral buckling techniques.224 The etiology of the myopic shift in the few PR cases was unknown, but possibly due to increased nuclear sclerosis of the lens after surgery.114 INTRAOCULAR PRESSURE RISE AND GLAUCOMA Although Abe et al reported an intraocular pressure (IOP) elevation of as much as 180 mm Hg during PR,1 such extreme IOP rise usually lasts only for a brief interval even without any corrective measures.56 As stated in the preceding section Intraocular Pressure Management, mild to moderate IOP rise is well tolerated by the average eye with no or minimal corrective measures after PR. Hilton et al stated that with the exception of eyes with advanced or endstage glaucoma, temporary IOP elevation during or after PR usually does not lead to any retinal or optic nerve damages.40,240 In the multicenter PR trial, one patient developed the rare complication of malignant glaucoma after disobeying instructions by sleeping on his back with his face CHAN ET AL up. Apparently this patient had a PC IOL and zonular dehiscence in his eye that underwent PR.114,131,243 Paracentesis followed by face-down positioning led to resolution of the glaucoma for that case. The final visual acuity of the affected eye was 20/20. Hilton et al recommended careful monitoring of the central retinal artery after intraocular gas injection. In the absence of central retinal arterial reperfusion after 10 minutes of observation (i.e., a lack of arterial pulsation), he advised the performance of paracentesis to immediately lower the IOP.40,110,240 Most importantly, the surgeon must ensure maintenance of light perception or better vision after the gas injection for the surgical eye. The authors tend to perform paracentesis during PR on a frequent basis, because it is associated with minimal morbidities when performed properly (see preceding section on Intraocular Pressure Management). ISCHEMIC OPTIC NEUROPATHY AND RETINAL VASCULAR OCCLUSION This rare complication was reported in 1 of 302 eyes (0.3%) after PR by Tornambe.241 Although not reported in the literature, branch and central retinal artery or vein occlusion can theoretically occur with PR.1,114 Use of subconjunctival anesthesia, careful administration of the peribulbar or retrobulbar anesthesia, and avoidance of excessive and prolonged intraocular pressure elevation may reduce the chance of such occurrences. DELAYED SUBRETINAL FLUID ABSORPTION Tornambe reported 15 of 302 eyes (5%) that developed delayed subretinal fluid absorption after PR.241 There are two types of residual subretinal fluid accumulation after PR. The first type is similar to delayed subretinal fluid absorption after RD repair by SBP. Chen et al reported substantial residual subretinal fluid involving the dependent portion of the RD that took longer than 6 months to resolve in three eyes after PR.51 They attributed this complication to choroidal vascular insufficiency, similar to the same phenomenon found in certain cases of RD after nondrainage SBP.51,158,205 The typical characteristics of this type of residual subretinal fluid consist of shifting of subretinal fluid to a dependent position and its slow absorption over a prolonged period. The second type of residual subretinal fluid after PR is much more subtle in appearance. In 1989, Chan and Wessels first reported this type of delayed subretinal fluid absorption (DSRFA) in 8 of 38 consecutive eyes with a retinal detachment that were treated with pneumatic retinopexy.47 The primary characteristic 469 REVIEW: PNEUMATIC RETINOPEXY of DSRFA consists of a small loculated pocket of shallow subretinal fluid not associated with any retinal breaks. They found heavy cryotherapy and subretinal pigment precipitates to correlate with DSRFA. DSRFA may have a negative visual impact when the macula is involved, because the resolution of the residual subretinal fluid is usually prolonged, taking many months. In contrast to the typical residual subretinal fluid associated with nondrainage scleral buckling, the loculated pocket of DSRFA related to pneumatic retinopexy is subtle in appearance and does not shift. Other investigators also reported similar cases of DSRFA.9,33,60 In 1990, Ambler and associates described three cases of DSRFA with good visual outcome.9 Böker et al reported 6 of 60 eyes with DSRFA after PR in 1991.33 The time for complete absorption of these circumscribed subretinal bullae ranged from 8 to 52 weeks (mean 22.7 weeks). None of their cases developed impaired vision as a result of the DSRFA. In 2001, Desatnik et al reported seven phakic eyes with DSRFA, with macular involvement in four eyes.60 The time interval for complete fluid absorption ranged from 10 to 26 months in their report. Consistent with Chan’s report, they found the rate of resolution of DSRFA not to be influenced by any postoperative maneuvers or additional surgical procedures. Complete retinal reattachment eventually occurs for most eyes on a spontaneous basis. Recent OCT scanning of the loculated pocket of submacular fluid after RD repair suggests its content to include photoreceptor byproducts.213 In the absence of worsening of the submacular fluid, it is best to wait for its resolution instead of reoperation. surgery associated wounds.127,253 with fresh corneal limbal MUSCULOSKELETAL COMPLICATIONS Due to the rigorous requirement of a consistent head tilt for a prolonged period, PR may lead to various musculoskeletal and neurological injuries, especially for elderly patients already afflicted with various arthritic and neuromuscular ailments. The cervical spine and neck muscles are particularly vulnerable to potential injuries on account of the head tilt frequently required during PR. In addition, lower back problems associated with prolonged turning of the body, and ulnar nerve injury due to excessive pressure on the elbow corresponding to the arm that supports the head during the head tilt may also occur. Such complications are usually minor and of a temporary nature. In their analysis of intraoperative and postoperative complications associated with PR, Hilton et al reported only two cases of postoperative neck problems.114,131 Nevertheless, meticulous attention should be given for proper support of the patient’s body to avoid musculoskeletal injuries related to PR. For instance, appropriate padding should be provided for the elbow sustaining the pressure. The patient should also be advised to avoid prolonged head and body tilt in mid air without any support. Instead, a pillow, cushion, a table or desk with the correct height, or a reclining chair or bed, may supply the required support and prevent unwanted musculoskeletal and neurological injuries. If necessary, analgesics and muscle relaxants may be prescribed to relieve musculoskeletal discomfort associated with PR.114 HYPOTONY MACULOPATHY In 1999, Aslanides et al reported an unusual case of hypotony maculopathy associated with progressive visual deficit 3 months after pneumatic retinopexy.13 The cause of the hypotony (3--5 mm of Hg) was a subclinical dehiscence of the cataract wound on the scleral limbus revealed only by ultrasonic biomicroscopy and indentation gonioscopy. Subsequent surgical repair of the dehiscence with interrupted 10-0 nylon sutures reversed the hypotony and improved the visual acuity to 20/50. CORNEAL WOUND DEHISCENCE In 2000, Jun et al reported two cases of corneal wound dehiscence during PR due to acute intraocular pressure rise after intraocular gas injection causing leakage through self-sealing, clear corneal incisions.127 To avoid this complication, they advised the performance of paracentesis prior to gas injection for eyes with RD following recent ocular Contrast of Pneumatic Retinopexy with Scleral Buckling, Pars Plana Vitrectomy, and Other Techniques SCLERAL BUCKLING, TEMPORARY BALLOON, PARS PLANA VITRECTOMY Despite the favorable reports of PR, some surgeons continue to favor SBP over PR for repairing simple RD.188 Others favor the temporary balloon buckling technique over PR.20,138,139,154-157,166 There have been multiple criticisms on the multicenter PR trial. For example, Valone noted that despite the simple RD, the single-operation success for the SBP eyes in the multicenter trial was only 82%, suggesting variations in outcome among participating surgeons.166,256 He also pointed out that SBP eyes had a longer duration of macular detachment than PR eyes, possibly influencing the visual outcome. Lincoff et al claimed that despite the validity of the multiple center PR trial statistics, 470 Surv Ophthalmol 53 (5) September--October 2008 the differences might be due to potentially deleterious effects of encircling buckles, SRF drainage, and air or gas injection in the SBP eyes, making their visual prognosis worse.138,154,155 They reported 93% success of 466 eyes with the former procedure, a much higher single-operation success than PR. However, Hilton et al pointed out that the balloon buckling group in Kreissig’s series had a low incidence of pseudophakia and macular detachment, predisposing them to a more favorable outcome. They also noted that Kreissig compared only the balloon buckling to the initial experience of PR with a lower success rate. In addition, Hilton and Tornambe emphasized that complications encountered with relative frequency with SBP but rarely or never occur with PR include the following:113,116,119,177,223,236,262 1. Inadvertent scleral perforation (5%) 2. Subretinal fluid drainage complications119 : i) subretinal hemorrhage (3.0 to 4.5 %), ii) retinal incarceration (2.2 to 3.0%), iv) retinal breaks (0.54 to 4.0%), v) vitreous loss (0.36 to 3.0%) 3. Extraocular muscular problems inducing diplopia (20 to 50%):177,223,248 In 1989, Smiddy et al reported extraocular muscle imbalance after scleral buckling, occurring primarily associated with encircling procedures.223 4. Refractive and corneal topographic changes after scleral buckling:219,224,262 Smiddy et al reported induction of myopic shift with majority of encircling buckling, but radial buckling infrequently results in significant postoperative refractive changes.224 In 1999, Weinberger et al assessed corneal topography by videokeratography for 46 patients that underwent PR (11 eyes), SBP (25 eyes), or vitrectomy (10 eyes).262 They found negligible corneal topographical changes for eyes that underwent PR and vitrectomy, but significant changes for those that underwent SBP. In their comments on Green et al’s report of favorable outcome with temporary balloon for repairing selected RD in 159 eyes, Tornambe and Hilton pointed out the lack of efficacy of temporary balloon for macular breaks, posterior breaks, giant breaks, giant dialysis, recurrent detachment adjacent to a scleral buckle, RD associated with an optic pit, and breaks in areas of scarred conjunctiva.246 Each of these conditions has been successfully treated with pneumatic retinopexy under selected circumstances. Tornambe also pointed out that pneumatic retinopexy has been reported to successfully treat selected complex RD involving multiple breaks in multiple quadrants involving more than 30 CHAN ET AL degrees that would usually require encircling instead of radial buckling if scleral buckling were chosen to repair such detachments.244 For such RD, PR but not encircling buckling would avoid a myopic shift. Primary pars plana vitrectomy for repairing RD is associated with higher incidence of cataract formation or progression, intraocular lens subluxation, iris capture, and flat anterior chamber.42,70,86,87,106,230 DRAINAGE, AIR INJECTION, CRYOTHERAPY, AND EXPLANT (D-ACE) AND SIMILAR TECHNIQUES Contemporaneous with the development of PR in the late 1980s, multiple surgeons in the United Kingdom advocated an alternative surgical technique combining certain features of both PR and scleral buckling.163,268 They coined the eponym ‘‘D-ACE’’ to depict the key steps of this procedure: (1) Drainage of subretinal fluid; (2) injection of intraocular Air; (3) Cryotherapy for retinal break(s); and (4) Explant placement. This procedure is advantageous for more accurate and prompt localization and support of retinal breaks in comparison to PR for a bullous RD. The proponents of this procedure reported a single-operation success rate of 85%, a figure higher than PR. The final success rates of 97--98% are comparable to PR. D-ACE provides the advantages of more accurate support of multiple and large breaks at or anterior to the equator while avoiding excessive cryotherapy for a bullous RD in comparison to PR. However, disadvantages include potential complications relating to subretinal fluid drainage and buckling not commonly associated with PR, that is, retinal incarceration or perforation, choroidal and subretinal hemorrhage, and extraocular muscle imbalance and diplopia. Purohit et al achieved a 95% single-operation success in 58 consecutive patients, using a technique similar to D-ACE that they referred to as ‘‘pneumatic buckle’’.198 They reported unchanged or improved final visual acuity in 88% eyes with macula-on RD, and final visual acuity of 20/20 to 20/50 in 67% eyes with macula-off RD. Gündüz and Günalp reported 90% single-operation success in reattaching 30 bullous RD by PR immediately following drainage of subretinal fluid.99 One eye developed D3 PVR after additional SBP, and refused further surgery. Thus, final success was 96.7%. Final VA was maintained or improved in 96.7% of treated eyes. Terubayashi et al reported 80% success in 31 eyes undergoing a modified PR technique with the following steps:235 1) lowering of the intraocular pressure; 2) cryopexy of retinal breaks; 3) drainage of subretinal fluid; 4) intra- 471 REVIEW: PNEUMATIC RETINOPEXY vitreal gas injection; and 5) 2 hours of prone positioning followed by bed rest. FLUID--GAS EXCHANGE FOR MACULAR-HOLE DETACHMENT A number of surgeons have reported the exchange of intraocular gas for liquid vitreous fluid for retinal detachment associated with a macular hole.2,93,203 Gas--fluid exchange is an appealing technique for repairing a macular hole--induced RD associated with high myopia, either during the initial procedure or afterwards. For instance, Blankenship et al advocated exchange of posterior vitreous fluid with a gas bubble,29 and Ripandelli et al proposed intraocular gas injection at the same time as subretinal fluid drainage for such cases.203 The proponents of this technique pointed out that the greater gas volume attained with this technique in contrast to a small volume of gas with the pure PR technique (see previous section on Expanded Use of Pneumatic Retinopexy for Special Clinical Conditions) allows more effective closure of the macular holes frequently associated with atrophic or staphylomatous posterior fundi in highly myopic eyes. In 1995, Chan reported the induction of posterior vitreous detachment followed by gas tamponade with a small expansile gas bubble for high-risk impending macular holes and small stage 2 macular holes.48 In recent years, Jorge et al and Mori et al also reported the utility of this technique for small macular holes.126,179 OCT confirmed the closure of the macular holes for the latter series. Mori reported that two important factors associated with the success of this technique include: 1) diameter of the macular hole of less than 200 microns, and 2) preoperative visual acuity of better than 20/40.179 SUPPLEMENTAL GAS INJECTION FOR FAILED PNEUMATIC RETINOPEXY, OR PRIMARY OR FAILED BUCKLING CASES Multiple surgeons also advocated intraocular gas injection for failed pneumatic retinopexy, during primary buckling, or to salvage failed scleral buckling cases with limited vitreoretinal traction due to new or missed retinal breaks.81,150,187,216,269 Weinberger et al reported 100% single-operation success for late-onset recurrent retinal detachment (6 months to 3 years after the original surgery).263 Eight of the 12 cases had primary PR and 3 had primary SBP. Final visual acuity remained unchanged in 10 eyes and improved in 2 eyes. Sharma et al reported 69.4% single-operation success for failed buckling cases with this technique.216 Multivariate analysis identified two risk factors for failure: location of retinal break either on the posterior slope or posterior to buckle (p 5 0.01), and extent of retinal detachment greater than two quadrants (p 5 0.02). They pointed out that because 100% final success was attained in all eyes that underwent further reoperations, supplementary PR did not compromise subsequent surgical outcome, a conclusion similar to primary PR. In 2007, McGimsey et al reported the successful treatment with gas tamponade followed by laser for recurrent rhegmatogenous RD without a detectable retinal hole after scleral buckling or vitrectomy in six eyes in six patients172 PNEUMATIC RETINOPEXY FOLLOWED BY SCLERAL BUCKLING FOR GIANT RETINAL TEAR DETACHMENT Ando et al performed PR with SF6 gas followed by cryotherapy and encircling scleral buckling to six of nine eyes. Two eyes required additional injection of SF6 gas, and one eye that developed postoperative grade C PVR was successfully repaired with a subsequent vitrectomy.10 A final VA of 20/30 was achieved in eight eyes. Conclusion Pneumatic retinopexy has become an important surgical technique in the modern era of retinal surgical management for RD. It is primarily indicated for uncomplicated RD with limited retinal breaks involving the superior 8 clock hours of the fundus, although more complex RD may be successfully managed with this technique on a selective basis.40,110,240,248 Qualified candidates must be willing to maintain a specific head posture for 7 days or more for optimal outcome with PR. Basic surgical steps of PR include retinopexy of retinal breaks with cryotherapy or laser, intraocular gas injection before or after retinopexy, and maintenance of proper head posture by the patient for the required time period after surgery.40,110,240 Proponents of PR have emphasized its advantages of minimal tissue trauma, no hospitalization, and reduced expenses, in comparison to scleral buckling and vitrectomy.8,40, 110,111,240,241,243,248 Critics of PR have pointed out its lower single-operation success rates, so that both the surgeon and patient need to be prepared for multiple procedures with PR on a more frequent basis.104,105,188 Published series of single-operation success rates of PR ranged from 53% to 100% for PR.114 Phakic eyes fared better than nonphakic eyes, with the single-operation successes of 71--84% for the former and 41--67% for the latter. Despite lower single-operation successes with PR in comparison to SBP, the multicenter PR trial and other published 472 Surv Ophthalmol 53 (5) September--October 2008 reports have shown that the final anatomical and visual outcomes are not disadvantaged by the initial PR.240,241,243 An excellent final average anatomical success rate of 98% was noted in published reports.114 In addition, the multicenter PR trial showed that cataracts, choroidal detachment, and myopic shift developed significantly less frequently with PR than SBP, although missed or new retinal breaks occurred more often after PR than SBP.243 Regarding the visual outcome, the multicenter PR trial has shown that those eyes with a duration of macular detachment of 14 days or less as well as 3 days or less, achieve a postoperative VA of 20/50 or better at 6 months more frequently after PR than SBP (p 5 0.01 and p 5 0.05, respectively).243 Recovery of VA was also more rapid after PR for many cases. Published reports of PR from 1988 to 1998 showed 35--80% of eyes with pre-operative macular detachment and 76-100% of eyes with pre-operative macular attachment achieve a postoperative VA of 20/50 or better with 6 or more months of follow up after surgery. The 2-year follow-up study of the multicenter PR clinical trial showed that vision continued to improve over the long term, so that 90% of PR eyes with macular-off RD achieved VA of 20/50 or better at 2 years. Thus, that study showed the importance of more than one year of follow-up.248 In summary, despite its limitations and shortcomings, multiple clinical studies have clearly shown important and favorable features of PR, and established its continued important role in the armamentarium of the retinal surgeon for managing an RD. The retinal surgeon should carefully consider all preoperative ocular features unique for each individual eye with an RD, modify his or her surgical techniques accordingly, and strive for the best possible anatomic and visual outcome for each patient. The authors have no financial and proprietary interests in any commercial products mentioned in this review. The authors thank Ms. Katherine Kreuter, Mr. Sandro Cinelli, Elizabeth Szambulan, MD, Doriana Cosgrove, MD, Paul Inae, MD, Ivan Fišer, MD, Victor Victorov, MD, and Ms Kelly Schultz for their expertise in foreign language translation. We also thank Paul Tornambe, MD, for his thoughtful review and helpful suggestions for the manuscript before and after its submission. References 1. Abe T, Nakajima A, Nakamura H, et al: Intraocular pressure during pneumatic retinopexy. Ophthal Surg Lasers 29:391--6, 1998 CHAN ET AL 2. Abe T, Takaji T, Maeda S, et al: The relationship between retinal tear area and prognosis following pneumatic retinopexy [Japanese]. Folia Ophthalmol Jpn 43:1220--5, 1992 3. Abecia E, Pinilla I, Olivan JM, et al: Anatomic results and complications in a long-term follow-up of pneumatic retinopexy cases. Retina 20:156--61, 2000 4. Ai E, Gardner TW: Current patterns of intraocular gas use in North America. Arch Ophthalmol 111:331--2, 1993 5. Algvere P, Gjötterberg M, Olivestedt G, Fituri S: Results of pneumatic retinopexy with air. Act Ophthalmologica 70: 632--6, 1992 6. Algvere P, Hallnas K, Palmquist BM: Success and complications of pneumatic retinopexy. Am J Ophthalmol 106: 400--4, 1988 7. Ambler JS, Meyers SM, Zegarra H: Delayed reabsorption of subretinal fluid after pneumatic retinopexy. [Correspondence]. Ophthalmol 97:695, [Chan CK, Wessels IF: Reply. 695--6], 1990 8. Ambler JS, Meyers SM, Zegarra H, Paranandi L: Reoperations and visual results after failed pneumatic retinopexy. Ophthalmol 97:786--90, 1990 9. Ambler JS, Zegarra H, Meyers SM: Chronic macular detachment following pneumatic retinopexy. Retina 10: 125--30, 1990 10. Ando F, Hirose H, Nagasaka T, et al: Treatment of retinal detachment with giant tear by pneumatic retinopexy. Eur J Ophthalmol 3:201--6, 1993 11. Apt L: Purulent anterior segment endophthalmitis following paracentesis. Ophthalmic Surg 17:526, 1986 12. Aronowitz JD, Brubaker RF: Effect of intraocular gas on intraocular pressure. Arch Ophthalmol 94:1191--6, 1976 13. Aslanides IM, Pavlin CJ, Giavedoni LR: Hypotonic maculopathy following pneumatic retinopexy: a UBM study. Br J Ophthalmol 83:1088, 1999 14. Assi AC, Charteris DG, Gregor ZJ: Practice patterns of pneumatic retinopexy in the United Kingdom. Br J Ophthalmol 85:244, 2001 15. Assi AC, Charteris DG, Pearson RV, et al: Pneumatic retinopexy in the treatment of primary rhegmatogenous retinal detachment. Eye 13:725--8, 1999 16. Avins LR, Krummenacher TR: Macular holes after pneumatic retinopexy. Arch Ophthalmol 106:724--5, 1988 17. Bagolini B, Oliva G, Focosi F, et al: Pneumoretinopexy with eyeball rotation: A new method to treat tears and breaks located in the lower quadrants. [Italian]. Ann Ottalmol Clin Ocul 115:633--9, 1989 18. Baker SR, Hainsworth DP: Suprachoroidal gas as a complication of pneumatic retinopexy. Retina 20:224--5, 2000 19. Balestrazzi E, Spades L, Marullo M, et al: The pneumoretinopexy in the treatment of some special cases of separation in the retina extensive rhegmatogenous detachment to at least three quadrants [Italian]. Clin Ocul Patol Ocul 12:111--5, 1991 20. Barr CC: Pneumatic retinopexy and balloon buckles as alternatives to conventional scleral buckling surgery. Sem Ophthalmol 10:2--8, 1995 21. Baltatzis S, Djacos MC, Micha M: The treatment of retinal detachment by means of internal tamponade by gas/ pneumatic retinopexy. [French]. J Fr Ophthalmol 12:441--4, 1989 22. Benson WD: History, in Benson WB (ed): Retinal Detachment, Diagnosis and Management, chapter 4. Philadelphia, PA, J.B. Lippincott Company, ed 3 1988, pp. 57--8 23. Benson WE, Chan P, Sharma S, et al: Current popularity of pneumatic retinopexy. Retina 19:238--41, 1999 24. Benz MS, Scott IU, Flynn HW, Gedde SJ: Retinal detachment in patients with a preexisting glaucoma drainage device—anatomic, visual acuity, and intraocular pressure outcomes. Retina 22:283--7, 2002 25. Berrod JP, Bazard MC, Bodart E, et al: Pneumatic retinopexy by SF6 or C3F8: results and complications concerning 56 patients. [French]. Bull Soc Opht France 90:117--20, 1990 473 REVIEW: PNEUMATIC RETINOPEXY 26. Binder S: Repair of retinal detachments with temporary balloon buckling. Retina 6:210, 1988 27. Binder S, Freyler H, Velikay M, et al: Comparative results of pneumatic retinopexy, Lincoff balloon and permanent buckle operation in retinal-reattachment surgery. Spektrum Augenheilkd 3:193--6, 1989 28. Blacharski PA, Newsome DA: Bilateral macular holes after Nd:YAG laser posterior capsulotomy. Am J Ophthalmol 105:417--8, 1988 29. Blankenship GW, Ibanez-Langlois S: Treatment of myopic macular hole and detachment: Intravitreal gas exchange. Ophthalmol 94:333--6, 1987 30. Blodi CF, Folk JC: Treatment of macular hole retinal detachment with intravitreal gas. Am J Ophthalmol 98:811, 1984 31. Bochow TW, Olk RJ, Hershey JM: Pneumatic retinopexy perfluoroethane (C2F6) in the treatment of Rhegmatogenous retinal detachment. Arch Ophthalmol 110:1723--4, 1992 32. Böhm P, Záhorcová M: Pneumatic retinopexy-method of choice in treatment of retinal detachment [Slovak]. Cesk Slov Oftalmol 59:154--9, 2003 33. Böker T, Koch F, Spitznas M: Delayed resorption of subretinal fluid after pneumatic retinopexy [German]. Fortschr Ophthalmol 88:354--7, 1991 34. Böker T, Schmitt C, Mougharbel M: Results and prognostic factors in pneumatic retinopexy [German]. Ger J Ophthalmol 3:73--8, 1994 35. Boniuk I, Okun W, Johnston GP, Arribas N: Xenon photocoagulation vs cryotherapy in the prevention of retinal detachment. Mod Probl Ophthalmol 12:81--102, 1974 36. Borgioli M, Cona F, Lambertucci D, et al: Pneumatic retinopexy and argon laser photocoagulation in some selected cases of rhegmatogenous retinal detachment. [Italian]. Ann Ottalmol Clin Ocul 114:259--65, 1988 37. Bovey EH, Bucher PJM, Gonvers M, Gailloud Cl: Treatment of retinal detachment with cryocoagulation and intravitreal gas injection. [French]. Klin Mbl Augenheilk 192:468--70, 1988 38. Brinton DA, Hilton GF: Clinical experience: Oakland, California, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, A Clinical Symposium. Des Plaines, IL, Greenwood, 1989, pp. 130--4 39. Brinton DA, Hilton GF: Pneumatic retinopexy. Ophthalmol Clin North Am 7:1--12, 1994 40. Brinton DA, Hilton GF: Pneumatic retinopexy and alternative retinal detachment techniques, in Ryan SJ, Wilkinson CP (eds): Retina, vol. 3. St. Louis, MO, Mosby Inc, ed 3 2001, pp. 2047--62 41. Brown GC: Macular hole following rhegmatogenous retinal detachment repair. Arch Ophthalmol 106:765--6, 1988 42. Campo RV, Sipperley JO, Sneed SR, et al: Pars plana vitrectomy without scleral buckle for pseudophakic retinal detachments. Ophthalmol 106:1811--6, 1999 43. Campochiaro PA, Kaden IH, Vidaurri-Leal J, Glaser BM: Cryotherapy enhances intravitreal dispersion of viable retinal pigment epithelial cells. Arch Ophthalmol 103: 434, 1985 44. Carim MM: Air injection pneumatic retinopexy, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, a Clinical Symposium. Chicago, IL, Greenwood Publishing, 1989, pp. 53--6 45. Carim MM: Scleral buckling after failed air injection pneumatic retinopexy. [letter]. Arch Ophthalmol 108:11, [Poliner LS, Tornambe PE: Reply. 11--12], 1990 46. Carstocea B, Filip M, Selaru D, et al: Retinopexia pneumatic!a cu SF6. [Portuguese]. Oftalmologia 39:62--5, 1995 47. Chan CK, Wessels IF: Delayed subretinal fluid absorption after pneumatic retinopexy. Ophthalmol 96:1691--700, 1989 48. Chan CK, Wessels IF, Friedrichsen EJ: Treatment of idiopathic macular holes by induced posterior vitreous detachment. Ophthalmology 102:757--67, 1995 49. Chang TS, Pelzek CD, Nguyen RL, et al: Inverted pneumatic retinopexy, a method of treating retinal de- 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. tachments associated with inferior retinal breaks. Ophthalmology 110:589--94, 2003 Chawla HB, Coleiro JA: Retinal detachment treated with intravitreal air: an evaluation of 241 cases. Br J Ophthalmol 61:588--92, 1977 Chen JC, Robertson JE, Coonan P, et al: Results and complications of pneumatic retinopexy. Ophthalmology 95:601--8, 1988 Chen SN, Hwang JF: Treatment of rhegmatogenous retinal detachment in teenagers by pneumatic retinopexy technique. Am J Ophthalmol 143:217--21, 2007 Chignell Ah, Billington B: The treatment of macular holes by pars plana vitrectomy and internal air/SF6 exchange. Graefe’s Arch Clin Exp Ophthalmol 224:67--8, 1986 Chignell AH: Pneumoretinopexy: a personal view. Br J Ophthalmol 74:117, 1990 Chignell AH, Fison LG, Davies EWG, et al: Failure in retinal detachment surgery. Br J Ophthalmol 57:525, 1973 Coden DJ, Freeman WR, Weinreb RN: Intraocular pressure response after pneumatic retinopexy. Ophthal Surg 19: 667--9, 1988 Committee on Ophthalmic Procedure Assessment: The repair of rhegmatogenous retinal detachments. Ophthalmol 103:1313--24, 1996 Custodis E: Bedeutet die plombenaufnahung auf die sklera einen fortschritt im der operativen behändlung der netzhautablosung. [German]. Ber Dtsch Ophthalmol Ges 58:102--5, 1953 Custodis E: Scleral buckling without excision and with polyviol implant, in Schepens CL (ed): : Importance of the Vitreous Body in Retinal Surgery with Special Emphasis on Reoperations. St Louis, Mosby Inc, 1960, p. p 175 Desatnik H, Alhalel A, Treister G, Moisseiev J: Management of persistent loculated subretinal fluid after pneumatic retinopexy. Br J Ophthalmol 85:189--92, 2001 Doi M, Ning M, Semba R, et al: Histopathologic abnormalities in rabbit retina after intravitreous injection of expansive gases and air. Retina 20:506--13, 2000 Dominguez DA: Cirugı́a precoz y ambulatoria del blationinopexy de retina. Arch Soc Esp Oftalmol 48:47-54, 1985 Dominguez DA, Boyd BF, Gordon S: Repeated insufflation of expansile gas. [letter]. Highlights Ophthalmol 4:1--14, 1986 Dominguez DA, Fonseca A, Gomez-Montaña J: Cirugı́a ambulatoria inicial en determinados desprendimientos de retina. Stud Ophthalmol 4:11--8, 1985 Dominiguez DA, Fonseca A, Gomez-Montana J. Gas tamponade for ambulatory treatment of retinal detachment. In: Proceedings of the XXV International Congress of Ophthalmology, Rome, May 4--10, 1986. Amsterdam, Kugler and Ghedini, 1987, pp 2038--45 Dos Santos Motta MM: An evaluation of pneumatic retinopexy with sulphur hexafluoride in rhegmatogenous retinal detachments. [Portuguese]. Rev Bras Oftalmol 48: 3--12, 1989 Eckardt C: Staphylococcus epidermidis endophthalmitis after pneumatic retinopexy. Am J Ophthalmol 103:720--1, 1987 Emi K: Pneumatic retinopexy, long-term follow-up results. [Japanese]. Folia Ophthalmol Jpn 42:1143--7, 1991 Emi K, Danjo S: Long-term follow-up of pneumatic retinopexy, in Khoo CY (ed): : New Frontiers in Ophthalmology. New York, Elsevier Science Publishers BV, 1991, pp. 1104--6 Escoffery RF, Olk RJ, Grand MG, et al: Vitrectomy without scleral buckling for primary rhegmatogenous retinal detachment. Am J Ophthalmol 99:275--81, 1985 Eter N, Böker T, Spitznas M: Long-term results of pneumatic retinopexy. Graefe’s Arch Clin Exp Ophthalmol 238:677--81, 2000 Fišer I, Kocur R, Handlová R: Pneumatic retinopexy: one year experience. [Czech]. Cesk Slov Oftalmol 54:30--5, 1998 474 Surv Ophthalmol 53 (5) September--October 2008 73. Focosi F, Oliva G, Tamburrelli C, Manganelli C: Echodriven pneumatic retinopexy in superior retinal detachment. [Italian]. Ann Ottalmol Clin Ocul 114:569--75, 1988 74. Focosi F, Tamburelli C, Molle F, Fava S: Treatment of macular hole and peripheral retinal tears with scleral buckling and intraoperative pneumoretinopexy. [Italian]. Ann Ottalmol Clin Ocul 116:675--81, 1990 75. Focosi F, Manganelli C, Fava S, et al: Complication prevention in pneumatic retinopexy. [Italian]. Ann Ottalmol Clin Ocul 117:537--45, 1991 76. Freeman WR, Lipson BK, Morgan CM, Liggett PE: New posteriorly located retinal breaks after pneumatic retinopexy. Ophthalmol 95:14--8, 1988 77. Freyler H, Radax U: Pneumatic retinopexy-a study of 7 years experience [German]. Klin Monatsbl Augenheilkd 202:212--7, 1993 78. Friberg TR: Clinical experience: Pittsburgh, Pennsylvania, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, A Clinical Symposium. Des Plaines, IL, Greenwood, 1989, pp. 117--23 79. Friberg TR: Inverted pneumatic retinopexy [letter]. Ophthalmol 110:2261, 2003; Chang TS: Reply, pp 2261--2262 80. Friberg TR: Laser photocoagulation using binocular indirect ophthalmoscopic laser delivery system. Ophthal Surg Lasers 26:549--59, 1995 81. Friberg TR, Eller AW: Laser pneumatic retinopexy for repair of recurrent retinal detachment after failed scleral buckle-Ten years experience. Ophthalmic Surg Lasers 32: 13--8, 2001 82. Friberg TR, Eller AW: Pneumatic repair of primary and secondary retinal detachment using a binocular indirect ophthalmoscope laser delivery system. Ophthalmol 95: 877--83, 1988 83. Gailloud C, Dufour R, Geinoz, et al: Failures in retinal surgery. Mod Probl Ophthalmol 12:2--14, 1974 84. Gallemore RP, McCuen BW: Silicone oil in vitreoretinal surgery, in Ryan SJ, Wilkinson CP (eds): Retina, Vol. 3. St Louis, Mosby Inc, ed 3 2001, pp. 2195--220 85. Garcia-Arumi J, Palau MM, Espax AB, et al: Reopening of 2 macular holes after neodymium: YAG capsulotomy. J Cataract Refract Surg 32:363--6, 2006 86. Gastaud P, Rouhette H, Negre F, et al: Place of ‘‘exploratory’’ vitrectomy of retinal detachment without proliferative vitreoretinopathy. [French]. J Fr Ophtalmol 23:482--7, 2000 87. Girard P, Karpouzas I: Vitrectomy in the treatment of uncomplicated retinal detachment [French]. J Fr Ophthalmol 18:88--93, 1995 88. Glacet-Bernard A, Coscas G: Full-thickness macular hole and retinal detachment complicating Best’s disease. Eur J Ophthalmol 3:53--4, 1993 89. Gnad HD, Skorpik CH, Paroussis P, et al: G3F8-Gas als innere tamponade in der netzhautchirugie-eine ‘‘pneumatische’’ blation-operation ohne eröffnung der bindehaut. [German]. Fortschr Ophthalmol 84:462--4, 1987 90. Goldberg RE, Boyer DS: Sequential retinal breaks following a spontaneous initial retinal break. Ophthalmol 88:10--2, 1980 91. Gonin J: Guérisons opératoires des décollements retiniens. Rev Gen Ophthalmol 37:295, 1923 92. Greco GM, Ambrosino L: Treatment of retinal detachment in Marfan syndrome. Ann Ophthalmol 25:72--6, 1993 93. Greco GM, Bonavolonta G: Treatment of retinal detachments due to macular holes. Retina 7:177--9, 1987 94. Green SN, Yarian DL, Masciulli L, Leff SR: Office repair of retinal detachment using a Lincoff temporary balloon buckle. Ophthalmol 103:1804--10, 1996 95. Gribomont AC, Rolland F: Efficacy of pneumatic retinopexy: 7 year retrospective study. Bull Soc belge Ophtalmol 263:131--5, 1996 96. Griffiths PG, Richardson J: Causes of proliferative retinopathy following pneumatic retinopexy [letter]. Arch Ophthalmol 108:1515, Campochiaro PA, Conway BP, Hackett SF: Reply, p 1515, 1990 CHAN ET AL 97. Grizzard WS, Hilton GF, Hammer ME, et al: : A multivariate analysis of anatomic success of retinal detachment treated with scleral buckling. Graefes Arch Clin Exp Ophthal 232: 1--7, 1994 98. Grizzard WS, Hilton GF, Hammer ME, et al: Pneumatic retinopexy failures: cause, prevention, timing, and management. Ophthalmol 102:929--36, 1995 99. Gündüz K, Günalp I: Pneumatic retinopexy with drainage of subretinal fluid. Intl Ophthalmol 18:143--7, 1994 100. Gunnarson G, Byhr E, Leyon H: Pneumatic retinopexy. Acta Ophthalmol 68:125--8, 1990 101. Hackett SF, Conway BP, Campochiaro PA: Subretinal fluid stimulation of retinal pigment epithelial cell migration and proliferation is dependent on certain features of the detachment or its treatment. Arch Ophthalmol 107:391--4, 1989 102. Haimovici R, Mukai S, Schachat AP, et al: Rhegmatogenous retinal detachment in eyes with uveal melanoma. Retina 16:488--96, 1996 103. Haller JA, Walsh J, Smiddy W, et al: Multicenter trial of transscleral diode laser retinopexy in retinal detachment surgery. Trans Am Ophthalmol Soc 95:1--30, 1997 104. Han DP, Hartz A, Guse CE: Comparison of pneumatic retinopexy and scleral buckling in the management of primary rhegmatogenous retinal detachment. [letter]. Am J Ophthalmol 127:742--3, 1998 105. Han DP, Mohsin NC, Guse CE, et al: Comparison of pneumatic retinopexy and scleral buckling in the management of primary rhegmatogenous retinal detachment. Am J Ophthalmol 126:658--68, 1998 106. Harkin KN, Lavin MJ, Leaver PK: Primary vitrectomy for rhegmatogenous retinal detachment. Graefes Arch Clin Exp Ophthalmol 231:344--6, 1993 107. Hejny C, Han DP: Vitrectomy for macular hole after pneumatic retinopexy. Retina 17:356--7, 1997 108. Helbig H, Noske W, Kleineidam M, et al: Bacterial endophthalmitis after anterior chamber paracentesis. Br J Ophthalmol 79:866, 1995 109. Hilton GF : Subretinal pigment migration. Arch Ophthalmol 91:445--50, 1974 110. Hilton GF, Das T, Majji AB, Jalali S: Pneumatic retinopexyprinciples and practice. Indian J Ophthalmol 4:131--43, 1996 111. Hilton GF, Grizzard WS: Pneumatic retinopexy: a two-step outpatient operation without conjunctival incision. Ophthalmol 93:626--41, 1986 112. Hilton GF, Kelly NE, Salzano TC, et al: Pneumatic retinopexy. A collaborative report of the first 100 cases. Ophthalmol 94:307--14, 1987 113. Hilton GF, McLean EB, Norton EWD: Retinal Detachment. 4th ed. San Francisco, American Academy of Ophthalmology, 4th ed. 1981. pp 77--98 114. Hilton GF, Tornambe PE: The Retinal Detachment Study Group: Pneumatic retinopexy. An analysis of intraoperative and postoperative complications. Retina 11:285--94, 1991 115. Hilton GF, Tornambe PE, Brinton DA, et al: The complications of pneumatic retinopexy. Trans Am Ophthalmol Soc 88:191--207, 1990 116. Hilton GF, Tornambe PE, Grizzard WS: Results of a temporary balloon buckle in the treatment of 500 retinal detachments and a comparison with pneumatic retinopexy. [letter]. Am J Ophthalmol 107:612, 1989 117. Holz ER, Mieler WF: View 3: The case of pneumatic retinopexy. Br J Ophthalmol 87:787--9, 2003 118. Humayun MS, Yeo JH, Koski WS, et al: The rate of sulfur hexafluoride escape from a plastic syringe. Arch Ophthalmol 107:853--4, 1989 119. Humphrey WT, Schepens CI, Elzeneiny IL, et al: The release of subretinal fluid and its complications, in Pruett RC, Regan CDJ (eds): Retina Congress. New York, Appleton-Century-Crafts, 1972, pp. 383--90 120. Hutter JC, Luu HMD, Schroeder LW: A biological model of tamponade gases following pneumatic retinopexy. Current Eye Research 25:197--206, 2002 475 REVIEW: PNEUMATIC RETINOPEXY 121. Irvine AR, Lahey JM: Pneumatic retinopexy for giant retinal tears. Ophthalmol 101:524--8, 1994 122. Jarrett W: Questions and Answers: Pneumatic retinopexy. Arch Ophthalmol 108:657, 1990 123. Jabally-Habib HY, Fineberg EM, Tornambe PE, et al: Prolonged pain following unintentional injection of gas into the suprachoroidal space during pneumatic retinopexy 23:722--3, 2003 124. Johnston PB, Maguire CJF, Logan WC: Failed retinal surgery caused by intraocular gas injection. Br J Ophthalmol 72:322--5, 1988 125. Joondeph BC, Joondeph HC: Purulent anterior segment endophthalmitis following paracentesis. Ophthal Surg 17: 91--3, 1986 126. Jorge R, Costa RA, Cardillo JA, et al: Optical coherence tomography evaluation of idiopathic macular hole treatment by gas-assisted posterior vitreous detachment. Am J Ophthalmol 142:869--71, 2006 127. Jun AS, Pieramici DJ, Bridges WZ: Clear corneal cataract wound dehiscence during pneumatic retinopexy. Arch Ophthalmol 118:847--8, 2000 128. Kaplan S: Room air pneumatic retinopexy. Trans Pa Acad Ophthalmol Otolaryngol 41:827--30, 1989 129. Kelly NE: Clinical experience: Sacramento, California, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, A Clinical Symposium. Des Plaines, IL, Greenwood, 1989, pp. 151--4 130. Kelly NE, Wendel RT: Vitreous surgery for idiopathic macular holes. Results of a pilot study. Arch Ophthalmol 109:654--9, 1991 131. Kim RY, D’Amico DJ: Postoperative complications of pneumatic retinopexy. Int Ophthalmol Clin 40:165--73, 2000 132. Kishimoto N, Imaizumi M, Nishimura T, et al: Surgical outcomes in rhegmatogenous retinal detachment. [Japanese]. Jpn J Clin Ophthalmol 51:1805--10, 1997 133. Kita M, Negi A, Kawano SI, Honda Y: Photothermal cryogenic, and diathermic effects on retinal adhesive force in vivo. Retina 11:441--4, 1991 134. Kleinmann G, Rechtman E, Pollack A, et al: Pneumatic retinopexy-results in eyes with classic vs relative indications. Arch Ophthalmol 120:1455--9, 2002 135. Koch F, Hockwin O, Böker Th, Mougharbel M: Linsenveränderungen nach pneumatischer retinopexy: Vorläufige ergebnisse der linear densitometrischen bildanalyse von scheimpflug-photographien zwei monate nach operation. Klin Mbl Augenh 198:515--7, 1991 136. Koch F, Spitznas M, Böker T, Mougharbel M, et al: Entwicklung von linsentrübungen in einem zeitraum von 6 monaten nach pneumatischer retinopexy. Fortschr Ophthalmol 88:216--8, 1991 137. Kovacevic D, Bedenicki I: Complications of pneumatic retinopexy. Coll. Antropol 25(Suppl):97--9, 2001 138. Kressig I: The treatment of primary retinal detachment. Minimal extraocular or intraocular? [German]. Ophthalmologe 99:474--84, 2002 139. Kressig I, Failer J, Lincoff H, Ferrari F: Results of a temporary balloon buckle in the treatment of 500 retinal detachments and a comparison with pneumatic retinopexy. Am J Ophthalmol 107:381--9, 1989 140. Kulkarni KM, Roth DB, Prenner JL: Current visual and anatomic outcomes of pneumatic retinopexy. Retina 27: 1065--70, 2007 141. Kusuki Y, Yamanaka H, Mizoguchi T, et al: Treatment of retinal detachment with macular hole by pneumatic retinopexy. [Japanese]. Folia Ophthalmol Jpn 40:1650--3, 1989 142. Kwon OW, Kim SY: Changes in adhesive force between the retina and the retinal pigment epithelium by laser photocoagulation in rabbits. Yonsei Med J 36:243--50, 1995 143. La Torre A, Crociani L, Santoro P, et al: Treatment of macular hole retinal detachment with pneumatic retinopexy. [Italian]. Ann Ottalmol Clin Ocul 116:1223--8, 1990 144. La Torre A, Crociani L, Nocentini, et al: The treatment of retina detachment by means of pneumoretinopexy in the 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. day-hospital. [Italian]. Ann Ottalmol Clin Ocul 118:61--4, 1992 Langmann G: Guidelines for surgery in the treatment of rhegmatogenous detachments (buckle, pneumatic retinopexy, endosurgery). [German]. Klin Mantsbl Augenheilkd 221:160--74, 2004 Laqua H, Honnicke K: Is buckling surgery still used in practice? [German]. Ophthalmologe 98:881--5, 2001 Lazzaroni F, Morinelli G, Scorolli L, et al: Our experience in the surgical treatment of the retinal detachments caused by hole in high myopia. [Italian]. Ann Ottalmol Clin Ocul 123:237--44, 1997 Lemmen K, Heimann K: Preliminary-results of pneumatic retinopexy. [German]. Klin Mbl Augenheilk 191:496, 1987 Lemmen K, Heimann K: Problems of pneumatic retinopexy. [German]. Klin Mbl Augenheilk 195:150--3, 1989 Levkovitch-Verbin H, Treister G, Moisseiv J: Pneumatic retinopexy as supplemental therapy for persistent retinal detachment after scleral buckling operation. Acta Ophthalmol Scand 76:353--5, 1998 Lewen RM, Eifrig DE: Scleral buckling of failed pneumatic retinopexy. Arch Ophthalmol 106:18, 1988 Liggett PE: Clinical experience: Doheney Eye Institute, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, A Clinical Symposium. Des Plaines, IL, Greenwood, 1989, pp. 155--6 Lim JI, Blair NP, Higginbotham EJ, Farber MD, et al: Assessment of intraocular pressure in vitrectomized gascontaining eyes. A clinical and manometric comparison of the Tono-Pen to the pneumotonometer. Arch Ophthalmol 108:684--8, 1990 Lincoff H: Discussion of pneumatic retinopexy. A multicenter randomized controlled clinical trial comparing pneumatic retinopexy with scleral buckling. Ophthalmol 96:784, 1989 Lincoff H, Horowitz J, Kreissig I, et al: Morphological effects of gas compression on the cortical vitreous. Arch Ophthalmol 104:1212--5, 1986 Lincoff H, Kreissig I: Changing patterns in the surgery for retinal detachment: 1929 to 2000. [German]. Klin Mbl Augenheilk 216:352--9, 2000 Lincoff H, Kreissig I: Results with a temporary balloon buckle for the repair of retinal detachment. Am J Ophthalmol 92:245--51, 1981 Lincoff H, Kreissig I: The treatment of retinal detachment without drainage of subretinal fluid (modifications of the Custodis procedure: part VI). Trans Am Acad Ophthal Otolaryng 76:1221--33, 1972 Lincoff H, Kreisig I, Brodie S, et al: Expanding gas bubbles for the repair of tears in the posterior pole. Int Ophthalmol 5:98, 1982 Lincoff H, Kreissig I, Goldbaum M: Reasons for failure in non-drainage operations. Mod Probl Ophthalmol 12:40--8, 1975 Lipham WJ, Smiddy WE: Idiopathic macular hole following vitrectomy: implications for pathogenesis. Ophthalmic Surg Lasers 28:633--9, 1997 Lisle C, Mortensen KK, Sjølie AK: Pneumatic retinopexy, a long term follow-up study. Acta Ophthalmol Scand 76: 486--90, 1998 Little BC, Inglesby DV, Wong D, Chignell AH: Results and complications of conventional repair of bullous retinal detachment using posterior segment air injection. Eye 4: 222--5, 1990 Lowe MA, McDonald HR, Campo RV, et al: Pneumatic retinopexy. Surgical results. Arch Ophthal 106:1672--6, 1988 Mansour AM: Inverted pneumatic retinopexy [letter]. Ophthalmol 111:1435, 2004; Equi R, Chang TS: Reply, pp 1435--1436 Marcus DM, D’Amico DJ, Mukai S: Pneumatic retinopexy versus scleral buckling for repair of primary rhegmatogenous retinal detachment. Int Ophthalmol Clin 34:97--108, 1994 476 Surv Ophthalmol 53 (5) September--October 2008 167. Martins M, Motta S: An evaluation of pneumatic retinopexy with sulphur hexafluoride in rhegmatogenous retinal detachments. [Portuguese]. Rev Bras Oftalmol 48:3--12, 1989 168. McAllister IL, Meyers SM, Zegarra H, et al: Comparison of pneumatic retinopexy with alternative surgical techniques. Ophthalmol 95:877--83, 1988 169. McAllister IL, Zegarra H, Meyers SM, et al: Treatment of retinal detachments with multiple breaks by pneumatic retinopexy. Arch Ophthalmol 105:913--6, 1987 170. McAuliffe PF, Heinemann M: The use of pneumatic retinopexy to delay surgical repair of a retinal detachment associated with the ganciclovir intraocular device. Ophthal Surg Laser 29:244--6, 1998 171. McDonald HR, Abrams GW, Irvine AR, et al: The management of subretinal gas following attempted pneumatic retinal reattachment. Ophthalmol 94:319--26, 1987 172. McGimpsey SJ, Cooke CA, Best RM: Pneumatic retinopexy. [letter]. Ophthalmology 114:401--2, 2007 173. Melgren SE, Michels M: Pneumatic retinopexy for the treatment of giant retinal dialyses. Am J Ophthalmol 118: 762--5, 1994 174. Menabuoni L, Paoletti Perini R, Parducci F, et al: Pneumatic retinopexy considerations on the surgical methodology. [Italian]. Atti Fon G Ronchi 45:833--6, 1990 175. Menchini U, Scialdone A, Visconti C, Brancato R: Pneumoretinopexy in the treatment of retinal detachment with macular hole. Int Ophthalmol 12:213--5, 1988 176. Meredith TA, Reeser TH, Topping TM, Aaberg TM: Cystoid macular edema after retinal detachment surgery. Ophthalmol 87:1090--5, 1980 177. Mets MB, Wendell ME, Gieser RG: Ocular deviation after retinal detachment surgery. Am J Ophthalmol 99:667--72, 1985 178. Miyakawa N, Ando F: Pneumatic retinopexy. 3. Indications. [Japanese]. Folia Ophthalmol Jpn 40:2644--9, 1989 179. Mori K, Saito S, Gehlbach PL, Yoneya S. Treatment of Stage 2 macular hole by intravitreous injection of expansile gas and induction of posterior vitreous detachment. Ophthalmol 114:127--33, 2007 180. Morin B, Said T, Cohen H, et al: Pneumatic retinopexy: our experience of 29 cases. [French]. Bull Soc Opht France 90: 543--8, 1990 181. Mortensen KK, Sjolie AK: Retinal detachment treated by pneumatic retinopexy. Acta Ophthalmol 66:187--9, 1988 182. Moses RA: Schiotz tonometry with an air bubble in the eye. Am J Ophthalmol 62:281--2, 1966 183. Moshfeghi AA, Salam GA, Deramo VA, et al: Management of macular holes that develop after retinal detachment repair. Am J Ophthalmol 136:895--9, 2003 184. Mougharbel M, Koch FH, Böker T, et al: No cataract two years after pneumatic retinopexy. Ophthalmol 101:1191--4, 1994 185. Mulvihill A, Fulcher T, Datta V, et al: Pneumatic retinopexy versus scleral buckling: a randomized controlled trial. Ir J Med Sci 165:274--7, 1996 186. Nerome K, Tomoyose H: Surgical results in rhegmatogenous retinal detachment. [Japanese]. Folia Ophthalmol Jpn 45:684--9, 1994 187. O’Connor P: Intravitreous air injection and the custodies procedure. Ophthalmic Surg 7:86--9, 1976 188. O’Malley P, Swearingen K: Scleral buckle with diathermy for simple retinal detachments: 100 pneumatic retinopexy eligible eyes. Ophthalmol 99:269--77, 1992 189. Ohm J: Über die behändlung der netzhautablosung durch operative entleerung der subretinalen flussigkeit und einspritzung von luft in den glasköper. [German]. Graefes Arch Clin Exp Ophthalmol 79:442--50, 1911 190. Pallan LA, Eller AW, Friberg TR: Pneumatic retinopexy: an overview. Semin Ophthalmol 6:27--35, 1991 191. Pavlović S, Latinović S, Tomić Z: Treatment of retinal detachment by pneumatic retinopexy. [Croatian]. Med Pregl 45:395--7, 1992 CHAN ET AL 192. Pecold K, Richard G: Indications and limits of temporary balloon-buckle and pneumatic retinopexy. [German]. Spektrum Augenhelkd 4:48--50, 1990 193. Pecold K, Richard G: Principles of qualification of retinal detachment for pneumoretinopexy. [Polish]. Klin Oczna 93:208--10, 1991 194. Poliner LS, Grand MG: Clinical experience: St. Louis, Missouri, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, A Clinical Symposium. Des Plaines, IL, Greenwood, 1989, pp. 127--9 195. Poliner LS, Grand MG, Schoch LH, et al: New retinal detachment after pneumatic retinopexy. Ophthalmol 94: 3158, 1987 196. Poliner LS, Schoch LH: Intraocular pressure assessment in gas-filled eyes following vitrectomy. Arch Ophthalmol 105: 200--2, 1987 197. Poliner LS, Tornambe PE: Failed retinal detachment repair after intravitreal air injection. Arch Ophthalmol 107:487--8, 1989 198. Purohit SS, Matthews GP, Zakov ZN: Pneumatic buckle for the repair of primary rhegmatogenous retinal detachment. Ophthal Surg Lasers 30:509--12, 1999 199. Rachal WF, Burton TC: Changing concepts of failures after retinal detachment surgery. Arch Ophthalmol 97:480--3, 1979 200. Raichand M, Peyman G: Pneumatic retinopexy, chapter 8: The retina and vitreous, in Ernest JT, Deutsch TA (eds): Year Book of Ophthalmology. 1987, pp 173--6 201. Rapkin JS: Pneumatic retinopexy: Experience in southwest Oklahoma. J Okla State Med Assoc 84:55--9, 1991 202. Rashed O, Sheta S: Evaluation of the functional results after different techniques for treatment of retinal detachment due to macular holes. Graefe’s Arch Clin Exp Ophthalmol 227:508--12, 1989 203. Ripandelli G, Vincenzo P, Friberg TR, et al: Retinal detachment associated with macular hole in high myopia. Ophthalmol 111:726--31, 2004 204. Robertson DM, Norton EWD: Cause of failure in prophylactic treatment of retinal breaks. Mod Probl Ophthalmol 12:74--80, 1974 205. Robertson DM: Delayed absorption of subretinal fluid after scleral buckling procedures. Am J Ophthalmol 87:57--64, 1979 206. Rosengren B: 300 cases operated upon for retinal detachment: methods and results. Acta Ophthalmol 30:117-22, 1952 207. Rosengren B: Über die behandlung der netzhautablosung mittelst diathermie und luftinjektionen in den glasköper [German]. Acta Ophthalmol 16:573--9, 1938 208. Runge PE, Wyhinny GJ: Macular hole secondary to pneumatic retinopexy. Arch Ophthalmol 106:586--7, 1988 209. Santillo C, Maucione V, Rossi S, Garzione F: Pneumatic retinopexy: Pros and cons [Italian]. Ann Ottalmol Clin Ocul 116:219--34, 1990 210. Saran BR, Brucker AJ: Macular epiretinal membrane formation and treated retinal breaks. Am J Ophthalmol 120:480--5, 1995 211. Sato K, Ando N, Kim T, et al: Postoperative complication of pneumatic retinopexy-macular pucker [Japanese]. Folia Ophthalmol Jpn 40:530--4, 1989 212. Schechter RJ: Inverted pneumatic retinopexy [letter]. Ophthalmol 110:2260--2261, 2003; Chang TS: Reply, p 2261 213. Schocket LS, Witkin AJ, Fujimoto JG, et al: Ultrahighresolution optical coherence tomography in patients with decreased visual acuity after retinal detachment repair. Ophthalmol 113:666--72, 2006 214. Sebag J, Tang M: Pneumatic retinopexy using only air. Retina 13:8--12, 1993 215. Sharma T: Posterior extension of retinal break following pneumatic retinopexy. Ann Ophthalmol 34:47--8, 2002 216. Sharma T, Badrinath SS, Mukesh BN, et al: A multivariate analysis of anatomic success of recurrent retinal detachment treated with pneumatic retinopexy. Ophthalmol 104: 2014--7, 1997 477 REVIEW: PNEUMATIC RETINOPEXY 217. Sharma T, Gopal L: Recent developments in vitreoretinal surgery. J Indian Med Assoc 98(754--8):760--2, 2000 218. Sharma S, Robbins R: The value component of evidencebased medicine-Decision analysis and pneumatic retinopexy versus scleral buckling surgery. Evid-Based Eye Care 2: 243--9, 2001 219. Shiotani Y, Emi K, Gomi F, Motokura M: Effect of scleral buckle on postoperative visual acuity [Japanese]. Folia Ophthalmol Jpn 45:523--6, 1994 220. Skoog K, Textorius O, Wrigstad A: Pneumatic retinopexy in 50 patients. Acta Ophthalmologica 67:130--5, 1989 221. Smiddy WE: Atypical presentations of macular holes. Arch Ophthalmol 111:626--31, 1993 222. Smiddy WE, Flynn HW, Nicholson DH, et al: Results and complications in treated retinal breaks. Am J Ophthalmol 112:623--31, 1991 223. Smiddy WE, Loupe DN, Michels RG, et al: Extraocular muscle imbalance after scleral buckling surgery. Ophthalmol 96:1485--90, 1989 224. Smiddy WE, Loupe DN, Michels RG, et al: Refractive changes after scleral buckling surgery. Arch Ophthalmol 107:1469--71, 1989 225. Snady-McCoy L: In office retinal detachment surgery: pneumatic retinopexy, a Rhode Island experience. Med Health RI 82:46--8, 1999 226. Snead M: Pneumatic retinopexy: new breaks for old? [editorial]. Eye 13:709--10, 1999 227. Snyder WB, Bloome MA, Birch DG: Pneumatic retinopexy versus scleral buckle: preferences of Vitreous Society members. Retina 12:43--5, 1992, 1990 228. Souma N, Nakauchi K, Kubota Y, et al: Pneumatic retinopexy for bullous rhegmatogenous retinal detachment with small breaks [Japanese]. Folia Ophthalmol Jpn 50:721--4, 1999 229. Speaker MG, Meinkoff JA: Prophylaxis of endophthalmitis with topical povidone-iodine. Ophthalmol 98:1769--75, 1991 230. Speicher MA, Fu AD, Marlin JP, et al: Primary vitrectomy alone for repair of retinal detachments following cataract surgery. Retina 20:459--64, 2000 231. Steinmetz RI, Kreiger AE, Sidikaro Y: Pre-vitreous space gas sequestration during pneumatic retinopexy. Am J Ophthalmol 107:191--2, 1989 232. Subramanian ML, Topping TM: Controversies in the management of primary retinal detachments [review]. Int Ophthalmol Clin 44:103--14, 2004 233. Taher RM, Haimovici R: Anterior chamber gas entrapment after phakic pneumatic retinopexy. Retina 21:681--2, 2001 234. Termote H: Pneumatic retinopexy: analysis of the 20 first cases [French]. Bull Soc Belge Ophtalmol 231:107--16, 1989 235. Terubayashi H, Ban Y, Yokoi N, Akagi Y: Pneumatic retinopexy for retinal detachment. 2. Indications and complications [Japanese]. Folia Ophthalmol Jpn 40:2033-8, 1989 236. Theodossiadis G, Nikalakis S, Apostolopoulas M: Immediate postoperative muscular disturbance in retinal detachment surgery. Mod Probl Ophthalmol 20:367--72, 1979 237. Theodossiadis GP, Theodossiadis PG: The macular buckling procedure in the treatment of retinal detachment in highly myopic eyes with macular hole and posterior staphyloma: mean follow-up of 15 years. Retina 25:285--9, 2005 238. Thomas EL: Pneumatic retinopexy. West J Med 148:695, 1988 239. Tornambe PE: Bilateral retinal detachment repaired with bilateral pneumatic retinopexy. Arch Ophthalmol 105: 1489, 1987 240. Tornambe PE: Pneumatic retinopexy. Surv Ophthalmol 32: 270--81, 1988 241. Tornambe PE: Pneumatic retinopexy: The evolution of case selection and surgical technique, a twelve-year study of 302 eyes. Trans Am Ophth Soc 95:551--78, 1997 242. Tornambe PE: Patient selection and preoperative evaluation; pneumatic retinopexy: a multicenter randomized 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 253. 254. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. controlled trial comparing pneumatic retinopexy and scleral buckling, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, A Clinical Symposium. Des Plaines, IL, Greenwood, 1989, pp. 18--25, 215--24 Tornambe PE, Hilton GF: The Retinal Detachment Study Group. Pneumatic retinopexy, a multicenter randomized controlled clinical trial comparing pneumatic retinopexy with scleral buckling. Ophthalmol 96:772--84, 1989 Tornambe PE, Hilton GF: Pneumatic retinopexy and buckling [letter]. Arch Ophthalmol 108:318--9, Smiddy WE, Michels RG: Reply, p 319, 1990 Tornambe PE, Hilton GF: Scleral buckling versus pneumatic retinopexy [letter]. Ophthalmol 99:1642--3, O’Malley P, Swearingen K: Reply, p 1643, 1992 Tornambe PE, Hilton GF: Office repair of retinal detachment [letter]. Ophthalmol 104:1059--60, Green SN, Yarian DL, Maschiulli L, Leff SR: Reply, p 1060--1, 1997 Tornambe PE, Hilton GF: The steamroller maneuver and proliferative vitreoretinopathy [letter]. Arch Ophthalmol 110:15, 1992 Tornambe PE, Hilton GF, Brinton DA, et al: Pneumatic retinopexy, a two-year follow-up study of the multicenter clinical trial comparing pneumatic retinopexy with scleral buckling. Ophthalmol 98:1115--23, 1991 Tornambe PE, Hilton GF, Kelly NF, et al: Expanded indications for pneumatic retinopexy. Ophthalmol 95: 597--600, 1988 Tornambe PE, Poliner LS, Hilton GF, Grizzard WS: Comparison of pneumatic retinopexy and scleral buckling in the management of primary rhegmatogenous retinal detachment [letter]. Am J Ophthalmol 127:741--2 Han DP, Hartz A, Guse CE: Reply, 742--3, 1999 Trillo M, Facino M, Mosci C, et al: Pneumatic retinopexy and episcleral surgery. Comparison between the two techniques [Italian]. Ann Ottalmol Clin Ocul 19:425--9, 1993 Trillo M, Facino M, Terrile R, et al: Treatment of uncomplicated cases of rhegmatogenous retinal detachment with an expanding gas bubble. Ophthalmologica 207: 140--3, 1993 Tripathi A: Factors in the prevention of wound dehiscence during pneumatic retinopexy [letter]. Arch Ophthalmol 119:621, Pieramici DJ: Reply, p 621, 2001 Tsujikawa M, Saito Y, Lewis JM, Tano Y: Secondary vitrectomy for the treatment of macular holes occurring after vitrectomy. Ophthalmic Surg Lasers 28:336--7, 1997 Tunc M, Lahey JM, Kearney JJ, et al: Cystoid macular edema following pneumatic retinopexy vs scleral buckling. Eye 21:831--4, 2007 Valone J: Pneumatic retinopexy, revisited [letter]. Ophthalmol 96:1811, Tornambe PE, Hilton GF: Reply, 1812, 1989 Van Effenterre G, Abi-Richard J, Vachet JM, Khairallah M: Exclusive use of gas injection in the treatment of certain retinal detachments [French]. Ophtalmologie 2:209--12, 1987 Vrabec TR : Pneumatic retinopexy in a progressive rhegmatogenous retinoschisis retinal detachment. Arch Ophthalmol 118:720--1, 2000 Vygantas CM: Clinical experience : pneumatic retinopexy with C4F8 (octafluorocyclobutane), in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, A Clinical Symposium. Des Plaines, IL, Greenwood, 1989, pp. 135--41 Waterhouse WJ, Bradshaw DJ, O’Boyle JE: Pneumatic retinopexy with air. Retina 13:270--1, 1993 Wedrich A, Datlinger P, Velikay M, et al: Comparison of the functional results following pneumatic retinopexy versus scleral buckle in retinal detachment surgery [German]. Spektrum Augenheilkd 5:191--6, 1991 Weinberger D, Lichter H, Loya N, et al: Corneal topographic changes after retinal and vitreous surgery. Ophthalmol 106:1521--4, 1999 Weinberger D, Stiebel-Kalish H, Crystal O, et al: Pneumatic retinopexy for late-onset recurrent retinal detachment. Ophthalmic Surg Lasers 29:43--7, 1998 Wendel RT: Anterior hyaloidal gas—The ‘‘Donut Sign’’, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, 478 265. 266. 267. 268. 269. 270. 271. 272. Surv Ophthalmol 53 (5) September--October 2008 A Clinical Symposium. Des Plaines, IL, Greenwood Publishing, 1989, pp. 94--101 Wild MR, Ruby AJ, Rosenshein J: Pneumatic retinopexy: a survey of current practice patterns among the Vitreous Society members. Ophthal Surg Lasers 31:76--81, 2000 Wilkinson CP, Bradford RH: Complications of draining subretinal fluid. Retina 4:1--4, 1984 Wirostko WJ, Han DP, Perkins S: Complications of pneumatic retinopexy. Curr Opin Ophthalmol 11:195--200, 2000 Wong D, Chignell AH, Inglesby DV, et al: The treatment of bullous rhegmatogenous retinal detachment. Graefe’s Arch Clin Exp Ophthalmol 230:218--20, 1992 Worsley DR, Grey RH: Supplemental gas tamponade after conventional scleral buckling surgery—a simple alternative to surgical revision. Br J Ophthalmol 75:535--7, 1991 Yam JC, Liu DT, Lee VY, et al: Giant retinal tear after pneumatic retinopexy. Acta Ophthalmol Scand, 2007 Yanyali A, Horozoglu F, Bayrak YI, et al: Steamroller versus basic technique in pneumatic retinopexy for primary rhegmatogenous retinal detachment. Retina 27:74--82, 2007 Yoon YH, Marmor MF: Rapid enhancement of retinal adhesion by laser photocoagulation. Ophthalmology 95: 1385--8, 1988 CHAN ET AL 273. Yospaiboon Y, Sanguansak T, Ratanapakorn T, Sinawat S: Surgical outcomes in rhegmatogenous retinal detachment in Srinagarind hospital. J Med Assoc Thai 88(Suppl 9):69--76, 2005 274. Yeo JH, Vidaurri-Leal J, Glaser BM: Extension of retinal detachments as a complication of pneumatic retinopexy. Arch Ophthalmol 104:1161--3, 1986 275. Zaidi AA, Alvarado R, Irvine A: Pneumatic retinopexy: success rate and complications. Br J Ophthalmol 90:427--8, 2006 276. Zakka K: Clinical experience: utilizing 15% C3F8, Los Angeles, California, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, A Clinical Symposium. Des Plaines, IL, Greenwood, 1989, pp. 161--4 277. Zegarra H, Meyers SM: McAllister I: Clinical experience: Cleveland, Ohio, in Tornambe PE, Grizzard WS (eds): Pneumatic Retinopexy, A Clinical Symposium. Des Plaines, IL, Greenwood, 1989, pp. 142--7 278. Zhai SQ: Postoperative posture of patients with retinal perforation treated with pneumatic retinopexy [Chinese]. Zhonghua Hu Li Za Zhi 28:343--5, 1993 Reprint address: Clement K. Chan, MD, PO Box 2467, Palm Springs, CA 92263, USA.