Proprionibacterium acnesInfection Complicating the Operative

CASE SERIES
Proprionibacterium acnes Infection Complicating the
Operative Treatment of Clavicle Fractures
ABSTRACT
BACKGROUND Management of failed surgical fixation of clavicle
fractures resulting in nonunion is a challenging entity and warrants
suspicion for indolent infection. Propionibacterium acnes (P. acnes)
and other Proprionibacterium species have been recognized as one
of the major sources of infection of the shoulder girdle. This study
reviews a series of patients diagnosed with P. acnes infection after
nonunion of a surgically treated clavicle fracture.
METHODS From 2002 to 2014, six patients were diagnosed with P.
acnes infection after a failed surgical fixation of a clavicle fracture
and were retrospectively analyzed. Clinical outcome was measured
using pre- and post-operative outcome scores.
RESULTS All patients had an initial diagnosis of clavicle fracture that
was treated with open reduction and internal fixation (ORIF). Four
patients subsequently developed chronic nonunion and wound cultures revealed P. acnes at the time of revision ORIF or irrigation and
debridement and were subsequently treated with intravenous antibiotics followed by a course of oral antibiotics. Two patients developed an acute infection after clavicle fixation within 3 months of the
initial fixation and went on to nonunion. All six patients ultimately
demonstrated union and had significantly improved physical function and reduction of pain.
CONCLUSION Failure of surgical fixation of clavicle fractures resulting in nonunion should raise suspicion for infection and cultures
should be obtained. P. acnes has been recognized as a source of infection in the shoulder girdle; when recognized and treated appropriately, excellent functional outcome and union can be achieved.
Clavicle fractures represent a common injury of the shoulder girdle.14
Various treatment options have been proposed, ranging from nonsurgical
treatment with a sling to surgical treatment using a variety of forms of internal fixation depending on location of the fracture. McKee et al.13 conducted
a meta-analysis analyzing the current evidence of operative versus nonop66
Arvind G. von Keudell, MD1
Sandra B. Nelson MD2
Jesse B. Jupiter, MD1
AUTHOR AFFILIATIONS
Massachusetts General Hospital, Department of Orthopaedic Surgery, Harvard
Medical School Boston, MA
1
Massachusetts General Hospital, Division
of Infectious Disease, Harvard Medical
School Boston 02114, MA
2
CORRESPONDING AUTHOR
Arvind G. von Keudell, MD
Hand and Upper Extremity Service
Massachusetts General Hospital
Department of Orthopaedic Surgery
Yawkey Center for Outpatient Care
55 Fruit Street, YAW-2-2C
Boston, Massachusetts 02114
Phone: 617-726-5100
[email protected]
One or more authors have potential
conflicts of interest related to this
work. The complete Disclosure of Potential Conflicts of Interest submitted
by the authors is available with the
online version of the article.
©2015 by The Orthopaedic Journal at
Harvard Medical School
THE ORTHOPAEDIC JOURNAL AT HARVARD MEDICAL SCHOOL
P. acnes Infection Complicating the Operative Treatment of Clavicle Fractures
erative treatment of displaced midshaft clavicular fractures. Operative treatment was found to result in earlier
return to function, a reduced rate of nonunion, and an
overall decreased complication rate when compared to
nonoperative treatment. The complication rate (including nonunion, symptomatic malunion, infection, hardware removal, neurologic symptoms and refracture)
was reported to occur in 29% in the operative group
versus 42% in the nonoperative group. Postoperative
pin site or wound infections occurred in 4% (9 of the
212 patients) treated operatively.
In the past decade, Proprionibacterium acnes (P.
acnes) has been recognized as one of the major sources of infection after surgical interventions around the
shoulder girdle.17 This study reviews a series of patients
who were either from our institution or referred to
our institution and diagnosed with P. acnes infection
after either chronic nonunion or acute infection of a
surgically treated clavicle fracture. The primary goal
TABLE 1
Patient
1
2
3
Age
(yrs)
52
59
21
Gender
F
Type of Initial
Fracture
Distal third,
type 2
Initial
Treatment(s)
Location
and Type of
Nonunion
Coracoclavicular screw
Distal third
clavicle, atrophic
Mid-shaft
clavicle,
hypertrophic
This case series was approved by the institutional
review board. We retrospectively reviewed all patients
seen for infections after surgically repaired clavicle
fractures and treated at the Upper Extremity or Orthopaedic Trauma Department. Additionally, patients
were identified using the Infectious Disease Outpatient
Parenteral Antimicrobial Therapy registry. The retrospective review identified six patients who were treated
between January 2002 and April 2014 (Table 1). Clin-
Nonunion
Duration
(mos)
Secondary Surgeries
Final Fixation Method
Time to
Evidence of
Healing of
Nonunion
(wks)
3
2 (Hook plate/
removal of hook
plate)
Iliac crest bone graft,
dorsal lateral locking
compression plate &
anterior non locking
plate
6
8
3 (Two Revision
ORIF with allograft,
Revision ORIF with
autogenous graft)
Superior (3.5mm)
and anterior (2.7mm)
double plating with
intercalary metatarsal
allograft
8
4
2 (Removal of
hardware and Irrigation& Debridement and VAC
placement)
Superior and anterior
double non- locking
reconstruction plating
6
9-hole non-locking
pelvic reconstruction
plating with 5cm iliac
crest bone graft
4
M
Mid-shaft
fracture
ORIF (unkonwn
implant)
F
Mid-shaft
with
comminution
8-hole plate
with 6 3.5 mm
cortical screws
and 1 lag screw
Mid-shaft,
atrophic
ORIF 3.5mm
pelvic reconstruction plate
Mid-shaft,
atrophic with a
4.5 cm defect
10
3 (Revision ORIF
and Removal of
hardware and
irrigation and
debridement)
Mid-shaft,
atrophic
12
2 (ORIF revision
with intercalary
bone graft, unknown implant)
Iliac crest bone graft,
LCDC anterior plate
and 2.7mm superior
plate
12
3
2 (Irrigation and
Debridement,
Removal of hardware)
None
12
4
49
F
5
49
M
Mid- shaft
clavicle fracture with mild
comminution
ORIF (unknown
implant)
M
Distal third
calvicle fracture, Type 2
Hook plate and
screws
51
MATERIALS & METHODS
Patient demographic data and information on surgical treatments
Segmental
mid- shaft
fracture
6
is to report on (1) radiographic union and (2) clinical
outcome (Quick DASH, VAS and Range of motion) after the various treatment options applied for infection
control or eradication of infected clavicle nonunions.
Further goals are to (3) report on re- operations and (4)
complications of all treated patients. Three exemplary
cases will be discussed in further detail.
Distal third
Volume 16 June 2015
67
von Keudell et al.
ical outcome was measured by radiologic outcomes as
well as patient rated factors including Quick DASH
Score,3 Range of motion and Visual Pain Analogue.
Four patients presented with a diagnosis of a nonunion of a clavicle fracture following surgical treatment
and two patients developed acute infection following
primary fracture fixation at our institution.
Two patients developed a symptomatic postoperative infection within 3 months following surgical treatment while the other four patients developed positive
cultures at a later date obtained at the revision surgery
(mean 23.6 months, range 6-71 months). In all, the primary causative agent was P. acnes. Coagulase negative
staphylococcus was also found in two patients.
In two out of the four patients, P. acnes had been cultured during a second surgery prior to referral to our service. The organisms that grew on culture were initially
interpreted as contaminants, and prior to revision surgery
at our institution no antibiotic treatment was provided.
Antibiotic therapy after surgical debridement and
revision consisted of six to eight weeks of an intravenous antibiotic regimen including cefazolin, penicillin,
ceftriaxone, clindamycin, vancomycin or daptomycin
for an average of 7 weeks (range 6-11 weeks). This was
followed by an oral regimen with one agent (clindamycin, amoxicillin, or cefadroxil) or a combination of
TABLE 2
one of these drugs with rifampin. Oral antibiotics were
continued for an average of 21 weeks (range 2 and 6
months; one individual was placed on life long antibiotic suppression). The parenteral antibiotics were tailored towards the sensitivities of the bacterium when
available (Table 2).
Case 1
A 52-year-old woman and avid triathlete sustained a closed, displaced distal clavicle fracture of
her non-dominant limb. She initially underwent coraco-clavicular screw fixation with a 4.5 mm cannulated screw. Three weeks postoperatively, the screw was
noted to have migrated superiorly. She returned to
the operating room, where the fracture was internally
fixed with a hook plate and screws. Two months later, radiographs suggested the fracture had united and
the plate was removed due to hardware irritation. Follow-up radiographs at eight weeks following plate removal demonstrated a nonunion, and the patient was
referred for reconstruction. A third operation was performed four months following hook plate removal using a dorsolateral plate with four locking screws in the
distal piece in addition to a 2.4mm anterior plate with
five screws along with 3 cm autogenous iliac crest bone
Culture data and antibiotic therapy
Patient
Time to
Growth
1
7 days
P. acnes in anaerobic culture, one
colony of Coagulase-negative
Staphylococcus
Stapylococcus resistant only to
Penicillin
IV Cefazolin and oral Rifampin (6
weeks)
Doxycycline (8 weeks)
Rifampin (4 weeks)
2
5 days
P. acnes in anaerobic culture
Sensitivities were not obtained
Penicillin (6 weeks)
Amoxicillin (lifelong suppression)
Rifampin (1 year)
3
11 days
P. acnes in anaerobic culture
Sensitivities were not obtained
IV Ceftriaxone and oral Rifampin
(6 weeks)
4
9 days
P. acnes in anaerobic culture
Unknown
Clindamycin (unknown)
Clindamycin (unknown)
5
8 days
P. acnes in anaerobic culture
Sensitivities were not obtained
IV Vancomycin (3 weeks), IV Daptomycin (8 weeks)
Clindamycin (8 weeks)
6
8 days
P. acnes in anaerobic culture, one
colony of Coagulase-negative
Staphylococcus
Staphylococcus resistant to
Penicllin and Clindamycin
IV Cefazolin (6 weeks)
Cefadroxil (6 months)
68
Culture
Sensitivities
Initial Antibiotic Treatment
(Duration)
THE ORTHOPAEDIC JOURNAL AT HARVARD MEDICAL SCHOOL
Suppressive Antibiotic
Treatment
Amoxicillin and rifampin
(4 months)
P. acnes Infection Complicating the Operative Treatment of Clavicle Fractures
FIGURE 1
Initial trauma CT scan 3D reconstruction
A
B
C
D
E
F
(A) Evidence of failure of fixation with a 4.5mm coraco-clavicular screw (B). Subsequent fixation with a hook plate and
screws (C) and radiographs taken after removal of hardware demonstrating nonunion (D). Final clavicle radiographs (E/F)
demonstrating healing of nonunion status post superior & anterior plating and screw fixation.
graft. Cultures revealed one colony of Coagulase-negative staphylococcus and moderate P. acnes.
She was treated with intravenous cefazolin in combination with oral rifampin for a total of six weeks, followed by eight weeks of oral doxycycline.
At thirty-four months status post revision surgery,
she reported 0/10 pain on VAS, and an improvement
from 66 to 9 points on the short Disability Arm, Shoulder and Hand Score (DASH). Physical examination
demonstrated full range of motion of her shoulder. The
patient was able to return to her competitive athletic
lifestyle (Figure 1).
Volume 16 June 2015
69
von Keudell et al.
Case 2
A 59-year-old male sustained a closed midshaft fracture of his right dominant clavicle from a fall. Surgery
was performed at an outside institution with a plate
and screws, which failed to heal. Two subsequent surgical attempts to achieve union with plate and screws
and allograft bone graft followed.
A fourth attempt at our institution included autogenous iliac crest graft plus plate and screw fixation,
which also proved to be unsuccessful. Intraoperative
cultures demonstrated P. acnes, which was thought to
be a contaminant due to multiple prior surgeries. The
patient was referred to the senior author. Physical examination revealed a painful shoulder girdle motion
with his pain scale on the VAS being 10/10 and DASH
score of 95 points. The erythrocyte sedimentation rate
was 16 mm/hr and C-reactive Protein 10.4 mg/L. Radiographs revealed an obvious non-union in addition
to loose internal fixation.
Surgery was performed using a dual plate fixation
including a 6-hole 2.7mm plate on the dorsal surface and 9- hole 3.5 mm on the anterior surface. A
metatarsal allograft was used to bridge the defect at
the nonunion site following debridement. Intraoperative cultures again revealed P. acnes. Parenteral
penicillin was given for total of 8 weeks followed by
oral amoxicillin and rifampin, which was anticipated
to be lifelong. At three months post-op, radiographs
demonstrated that his nonunion to have healed. At
43 months following his last revision surgery, his
VAS score was 3/10, and his DASH improved from
95 to 61 points with full shoulder range of motion
(Figure 2).
Case 3
A 21-year old college soccer player sustained a
displaced midshaft clavicle fracture in her non-dominant extremity while playing soccer and underwent
surgical fixation with a 3.5mm plate. Six weeks postoperatively, wound dehiscence occurred. Radiographs did not show any evidence of bony healing.
She was taken to the operating room for irrigation
and debridement of the surgical site. The plate was
left in place. Cultures were obtained and grew P.
acnes. Intravenous ceftriaxone in addition to rifam70
pin was administered for six weeks followed by oral
amoxicillin with rifampin. Four months status post
irrigation and debridement, the plate was removed
and revision internal fixation with double plating using a 5-hole 3.5mm anterior plate and 8-hole 3.5mm
superior plate was performed. Cancellous bone grafting from the iliac crest was used to fill the 2cm gap
from the apparent nonunion site. Cultures that were
obtained at the revision operation remained without
growth. Antibiotics were discontinued when bony
union was evident on CT scan four months after the
revision procedure. At five months following her revision surgery, the patient report 0/10 on VAS with a
DASH score of 2 and returned to playing soccer.
RESULTS
After eradicating the infectious organism, all six
patients demonstrated healing of the infected nonunion on follow-up radiographs.
All six patients improved in regards to their
Quick DASH Score, functional motion, and pain
scale when the treatment was completed. The DASH
Score improved from 54±32 (n=5) to 22±72 (n=5)
points at final follow-up (25±19 months, range 343months). The range of motion (n=5) of the affected shoulder postoperatively demonstrated on
average abduction of 160±35 degrees, forward flexion of 153±46 degrees, external rotation of 75± 26
degrees and VAS pain scale decreased from 4.8±3.1
to 1±1.7 (Table 3).
Overall, two patients underwent four procedures,
three patients underwent three operations and
one patient underwent two operations to achieve
healing of their infected non-unions. All patients
demonstrated union at final follow-up. Evidence of
bony healing following revision surgery was apparent in less than 3 months in all patients (8± 3 weeks,
range 4-12 weeks).
Complications included an initial cortical screw
backing out slightly in one patient, which remained
mildly symptomatic but did not interfere with functional activities. Two patients had allergic reactions to
the antibiotic therapy. Upon change of the antibiotic,
the symptoms subsided.
THE ORTHOPAEDIC JOURNAL AT HARVARD MEDICAL SCHOOL
P. acnes Infection Complicating the Operative Treatment of Clavicle Fractures
FIGURE 2
Case 2
A
B
C
D
E
F
(A/B) Initial presentation demonstrating hardware breakage and nonunion of mid shaft clavicle fracture. (C/D) Radiographs
demonstrating revision fixation of the clavicle fracture and nonunion. (E/F) Imaging of the re-revision fixation with double
plating and screw fixation demonstrating healed nonunion.
TABLE 3
Final outcome scores
Patient
Final Follow-up (mos)
Preoperative VAS (0-10)
Preoperative DASH
Postoperative VAS
Postoperative DASH
Final ROM
1
2
37
4
66
43
10
96
0
9
Full
3
61
Full
3
5
5
64
0
2
Full
4
3
3
9
n/a
n/a
n/a
5
36
2
36
0
18
Full
6
14
n/a
n/a
2
20
Full
Volume 16 June 2015
71
von Keudell et al.
DISCUSSION
Infection after surgery about the shoulder girdle
is an uncommon condition but can have devastating
impact on the shoulder function.17 P. acnes was once
thought to be a contaminant but now is recognized
as one of the most common and important pathogens
causing infection after surgical procedures about the
shoulder girdle.10, 16, 20 With the increase in operative
treatment of clavicle fractures, there is a potential for
an increase in surgical complications such as infections.
Duncan et al.6 were the first to report on P. acnes infection after surgical treatment of the clavicle. In their
series of six patients, three patients were found to have
P. acnes inoculation. All three patients were noted to
have pain even after eradication of the infection. Liu
et al.11 reported on seven patients with clavicle fracture who developed postoperative infection, although
none were caused by P. acnes.
P. acnes is a gram-positive anaerobic bacillus that is
found on the skin, especially in sebum excreting sebaceous follicles. It was first noted to be the source of infection in shoulder surgery in 1999,19 with an increasing incidence reported in subsequent studies.2, 7 The
explanation for why the shoulder appears to be the
most commonly affected area after surgical intervention by this bacterium remains not fully understood.
The shoulder area has the highest density of sebaceous
glands and most lipophilic skin, which may be a predisposing factor. P. acnes is notorious for having slow
growth in a culture medium (requiring a minimum of
6 days) and can be difficult to detect.
Lutz et al. found that it took an average of 11 days
for this bacteria to grow from specimens associated
with arthroplasty and osteosynthesis infections. In
their study, early infections resulted in a slightly shorter duration until cultures were positive in comparison
to late infections. (8.4 and 13.5 days, respectively)12
It should be noted that ESR and CRP concentration
are insensitive indicators for Propionibacterium osteosynthetic shoulder infection. It is rare for P. acnes
to present itself with a sinus tract communicating
with the osteosynthetic material and similarly rare
to demonstrate visible purulence at the implant site.
However, it is prudent to have a high suspicion for
infection, in particular P. acnes, in chronic nonunion
72
after surgically reconstructed clavicle fractures. It is
advisable to obtain at least three samples of anaerobic
cultures, which should be held for at least 14 days to
increase the diagnostic value of detecting the pathogen.1, 4, 9, 18 Asseray et al.1 determined different criteria
to improve the probability of diagnosing infection.
They found that a probability of >90% was achieved if
at least two positive deep cultures were present with
one of the following criteria: intraoperative signs of
infection, the presence of an orthopaedic implant or
revision surgery. If only one positive deep culture was
present, the diagnosis of P. acnes required three of the
following criteria: local signs of infections, inflammatory biological markers, orthopaedic device and revision surgery to achieve >90% probability of an actual
P. acnes infection.
Osmon et al.15 published guidelines on the treatment of prosthetic joint infection and recommended
parenteral penicillin or ceftriaxone for P. acnes infection. The optimal choice of antibiotic medication,
however, has not been established due to the paucity
of the data for osteosynthetic infections with P. acnes.
Thus, a few authors have used a variety of antibiotics such clindamycin, vancomycin, cephalosporins or
linezolid. Rifampin appears to offer synergy in the
eradication of P. acnes biofilms,2, 5, 7 though the role
of this antibiotic in treatment is not clear yet due to
the paucity of the available data, and drug interactions
complicate its use. Rifampin should only be used in
combination with another antibiotic, as it is thought
to aid in antibiotic penetration into the bone architecture. Due to the limited number of patients in this series we are not able to give general recommendations
on the antibiotic treatment therapy.
Currently there is no treatment algorithm for managing P. acnes infection of clavicular nonunion. Controlled clinical research is necessary to identify the optimal antimicrobial approach to these infections. Two
general approaches appear feasible for chronic nonunions (>3 months post-op), which can be amenable
to either single-stage revision plating versus two-stage
revision including irrigation and debridement and revision fixation.8 This decision should be made on an
individualized basis. Based upon the senior author’s
experience, we suggest a staged procedure if there
is suspicion for chronic infection or massive bone
loss in comparison to chronic infection with mild to
THE ORTHOPAEDIC JOURNAL AT HARVARD MEDICAL SCHOOL
P. acnes Infection Complicating the Operative Treatment of Clavicle Fractures
moderate bone loss. Double plating with intercalary bone graft either from the iliac crest or allograft
may be required to provide the stability to counteract the inherent rotational forces across the clavicle.
Parenteral antibiotics should be started as soon as the
pathogen is identified. Acute infection (<3 months
post-op) should be treated either with local irrigation
and debridement with retaining hardware until union
is observed with subsequent removal of hardware or
revision fixation with plate and screw fixation. It is
advisable to keep three samples of cultures obtained
during a revision procedure for at least 2 weeks if
there is suspicion for P. acnes infection.
culture for diagnosis of Propionibacterium acnes prosthetic joint
CONCLUSION
7. Herrera MF, Bauer G, Reynolds F, Wilk RM, Bigliani LU, Levine
This is the first series of P. acnes infection after clavicle fracture ORIF. It is difficult to diagnose and when
unrecognized can result in significant long-term impairment of the patient. It is necessary to have high
suspicion of infection in clavicle nonunion after surgical fixation. No clinical guidelines currently exist.
Single stage or two stage procedures in combination
with long-term parenteral antibiotic therapy may lead
to union and good outcomes.
bow Surg. 2002 Nov-Dec;11(6):605-8.
infection. J Clin Microbiol. 2011 Jul;49(7):2490-5. doi: 10.1128/
JCM.00450-11. Epub 2011 May 4.
5. Crane JK, Hohman DW, Nodzo SR, Duquin TR. Antimicrobial
susceptibility of Propionibacterium acnes isolates from shoulder
surgery. Antimicrob Agents Chemother. 2013 Jul;57(7):3424-6.
doi: 10.1128/AAC.00463-13. Epub 2013 Apr 29.
6. Duncan SF, Sperling JW, Steinmann S. Infection after clavicle
fractures. Clin Orthop Relat Res. 2005 Oct;439:74-8.
WN. Infection after mini-open rotator cuff repair. J Shoulder El-
REFERENCES
8. Jain AK, Sinha S. Infected nonunion of the long bones. Clin Orthop Relat Res. 2005 Feb;(431):57-65.
9. Kowalski TJ, Berbari EF, Huddleston PM, Steckelberg JM, Osmon DR. Propionibacterium acnes vertebral osteomyelitis: seek
and ye shall find? Clin Orthop Relat Res. 2007 Aug;461:25-30.
10. Levy PY, Fenollar F, Stein A, Borrione F, Cohen E, Lebail B,
Raoult D. Propionibacterium acnes postoperative shoulder ar-
1. Asseray N, Papin C, Touchais S, Bemer P, Lambert C, Boutoille
thritis: an emerging clinical entity. Clin Infect Dis. 2008 Jun
D, Tequi B, Gouin F, Raffi F, Passuti N, Potel G. Improving diag-
15;46(12):1884-6. doi: 10.1086/588477.
nostic criteria for Propionibacterium acnes osteomyelitis: a retrospective analysis. Scand J Infect Dis. 2010 Jul;42(6-7):421-5. doi:
10.3109/00365540903527330.
11. Liu PC, Hsieh CH, Chen JC, Lu CC, Chuo CY, Chien SH. Infection after surgical reconstruction of a clavicle fracture using a
reconstruction plate: a report of seven cases. Kaohsiung J Med Sci.
2. Athwal GS, Sperling JW, Rispoli DM, Cofield RH. Deep infec-
2008 Jan;24(1):45-9. doi: 10.1016/S1607-551X(08)70073-1.
tion after rotator cuff repair. J Shoulder Elbow Surg. 2007 MayJun;16(3):306-11. Epub 2007 Feb 22.
12. Lutz MF, Berthelot P, Fresard A, Cazorla C, Carricajo A, Vautrin AC, Fessy MH, Lucht F. Arthroplastic and osteosynthetic
3. Beaton DE, Wright JG, Katz JN. Development of the Quick-
infections due to Propionibacterium acnes: a retrospective study
DASH: comparison of three item-reduction approaches. J Bone
of 52 cases, 1995-2002. Eur J Clin Microbiol Infect Dis. 2005
Joint Surg Am. 2005 May;87(5):1038-46.
Nov;24(11):739-44.
4.Butler-Wu SM, Burns EM, Pottinger PS, Magaret AS, Rakeman
13. McKee RC, Whelan DB, Schemitsch EH, McKee MD. Opera-
JL, Matsen FA, 3rd, Cookson BT. Optimization of periprosthetic
tive versus nonoperative care of displaced midshaft clavicular frac-
Volume 16 June 2015
73
von Keudell et al.
tures: a meta-analysis of randomized clinical trials. J Bone Joint
17. Saltzman MD, Marecek GS, Edwards SL, Kalainov DM. In-
Surg Am. 2012 Apr 18;94(8):675-84. doi: 10.2106/JBJS.J.01364.
fection after shoulder surgery. J Am Acad Orthop Surg. 2011
Apr;19(4):208-18.
14. Nordqvist A, Petersson C. The incidence of fractures of the
clavicle. Clin Orthop Relat Res. 1994 Mar;(300):127-32.
18. Schafer P, Fink B, Sandow D, Margull A, Berger I, Frommelt L. Prolonged bacterial culture to identify late periprosthet-
15. Osmon DR, Berbari EF, Berendt AR, Lew D, Zimmerli W,
ic joint infection: a promising strategy. Clin Infect Dis. 2008 Dec
Steckelberg JM, Rao N, Hanssen A, Wilson WR; Infectious Dis-
1;47(11):1403-9. doi: 10.1086/592973.
eases Society of America. Diagnosis and management of prosthetic joint infection: clinical practice guidelines by the Infectious Dis-
19. Settecerri JJ, Pitner MA, Rock MG, Hanssen AD, Cofield RH.
eases Society of America. Clin Infect Dis. 2013 Jan;56(1):e1-e25.
Infection after rotator cuff repair. J Shoulder Elbow Surg. 1999 Jan-
doi: 10.1093/cid/cis803. Epub 2012 Dec 6.
Feb;8(1):1-5.
16. Patel A, Calfee RP, Plante M, Fischer SA, Green A. Propion-
20. Sperling JW, Kozak TK, Hanssen AD, Cofield RH. Infec-
ibacterium acnes colonization of the human shoulder. J Shoul-
tion after shoulder arthroplasty. Clin Orthop Relat Res. 2001
der Elbow Surg. 2009 Nov-Dec;18(6):897-902. doi: 10.1016/j.
Jan;(382):206-16.
jse.2009.01.023. Epub 2009 Apr 11.
74
THE ORTHOPAEDIC JOURNAL AT HARVARD MEDICAL SCHOOL