Negative pressure pulmonary edema after a tonsillectomy and adenoidectomy in a

Negative pressure pulmonary edema after
a tonsillectomy and adenoidectomy in a
pediatric patient: Case report and review
CAROLINE L. THOMAS, CRNA, MSN
St. Clair Shores, Michigan
TIMOTHY J. PALMER, CRNA, MS
PETER SHIPLEY, CRNA, MS
Royal Oak, Michigan
1998 Student Writing Contest Winner
Negative pressure pulmonary edema
(NPPE) continues to be reported as a
complication of upper airway obstructions
seen by anesthesia providers during
induction or emergence. The majority of
patients reported to have experienced NPPE
have been healthy, without underlying
pulmonary or cardiac disease. Factors
associated with the formation of NPPE
include young male patients and patients
with long periods of airway obstruction.
Overzealous intraoperative fluid
administration and preexisting heart and
lung disease also have been implicated as
predisposing factors. Negative pressure
pulmonary edema is the result of a marked
decrease in intrathoracic pressure caused by
ventilatory efforts against a closed glottis
resulting in a disruption of the normal
intravascular Starling mechanism,
ultimately leading to the transudation of
intravascular proteins and fluid into the
pulmonary interstitium. The onset of
NPPE is usually rapid, and without
prompt recognition and intervention, the
outcome can be fatal. A case of NPPE in a
pediatric patient after an otherwise
uncomplicated surgical procedure was
observed in our institution and is described
in this report.
Key words: Anesthesia, negative pressure
pulmonary edema, upper airway obstruction.
Case report
A 19-month-old, 11-kg boy was scheduled to
undergo a tonsillectomy and adenoidectomy. His
medical history included recurrent crouplike
infections, and he also had been recently discharged from a local area hospital after evaluation
to rule out meningitis. Symptoms at that time
included a stiff neck, headache, frequent crying
for 2 days, and an inability to ambulate without
assistance. Records obtained from that hospital
revealed negative cerebrospinal fluid culture
results. The patient’s surgical history included
bilateral myringotomies with tube placement and
bilateral inguinal hernia repairs without reported
problems. The patient was not taking any medications and had no known drug allergies; the
physical assessment was unremarkable. The preinduction heart rate was 162 beats per minute;
blood pressure, 98/52 mm Hg; and respiratory
rate, in the 30s.
An intravenous (IV) infusion of lactated
Ringer’s was started after performing a mask
induction of anesthesia with sevoflurane, 30%
oxygen, and 70% nitrous oxide. The trachea was
intubated easily with a 4.5 uncuffed endotracheal
tube (ETT), and a leak was noted at less than 20
cm of water. Anesthesia was maintained with
100% oxygen and isoflurane. The heart rate
remained between 160 and 170 beats per minute,
the systolic blood pressure between 96 and 120
mm Hg, and the diastolic blood pressure between
56 and 78 mm Hg throughout surgery. The
surgery proceeded without difficulty. Bupivacaine
0.5% with epinephrine 1:200,000 (exact amount
unclear) was injected into the tonsillar fossae by
AANA Journal/October 1999/Vol. 67, No. 5 425
the surgeon to adequately infiltrate the area, and
10 mg of IV meperidine was administered for postoperative pain relief. The patient received a total
of 350 mL of lactated Ringer’s during the 60minute procedure. His normal maintenance intravenous fluid (IVF) requirement per hour was calculated to be 42 mL, and the deficit from an 8hour overnight fast was estimated to be 336 mL.
The estimated blood loss was minimal (<50 mL).
The patient was extubated during deep anesthesia and immediately turned on the right side.
The patient began moving his extremities and
crying but was noted to have difficulty with phonation. Stridor and retractions ensued and the
arterial oxygen saturation, as measured by pulse
oximetry (SpO2), decreased into the 70% range.
Positive-pressure ventilation via bag and mask was
applied and the patient was assisted successfully
with spontaneous respirations. The SpO2
increased to 100% but could be maintained at
more than 90% only by using 1.0 FIO2 by mask. The
heart rate and blood pressure were only slightly
elevated. Hydrocortisone sodium succinate, 50
mg IV, was given to treat the laryngeal edema evidenced by stridor, and a laryngoscopy was performed. There was no evidence of obstruction due
to blood clots or ongoing bleeding. The vocal
cords could not be visualized, as the child was
moving about vigorously. Chest auscultation
demonstrated diminished breath sounds on the
left side. A chest radiograph performed in the
operating room revealed bilateral infiltrates
indicative of pulmonary edema; 5 mg of IV
furosemide was administered. Soon after, pink
frothy sputum was noted from the oropharynx.
The SpO2 was maintained at more than 90% with
assistance throughout the event. The patient’s
condition began to improve, and he was transferred to the pediatric intensive care unit for
observation approximately 45 minutes after completion of the tonsillectomy and adenoidectomy
without evidence of respiratory distress. The SpO2
was 96% to 100% during administration of 2 L of
oxygen delivered via nasal cannula. The total IVF
administered was 550 mL of lactated Ringer’s.
The admission note to the pediatric intensive
care unit reported rales in the right lower lobe
without stridor or wheezes with symmetrical
expansion of the chest. Supplemental oxygen was
discontinued gradually 4 hours after admission
when further auscultation of the chest revealed
the resolution of rales. A chest radiograph performed the next morning demonstrated almost
complete resolution of the bilateral infiltrates.
426
AANA Journal/October 1999/Vol. 67, No. 5
The patient was discharged home without further
complications.
Discussion
The negative pressure pulmonary edema
(NPPE) resulted from a partial upper airway
obstruction of unknown cause. Since the patient
had a history of recurrent crouplike infections,
the cause of the upper airway obstruction may have
been preexisting laryngeal edema. Another
possibility was a partial laryngospasm, attributable to secretions pooling in the oropharynx.
Furthermore, vocal cord paralysis from the injection of bupivacaine into the tonsillar fossae may
have contributed to the subsequent upper airway
obstruction.1 Considering the anatomical location
of the nerves relative to the pharyngeal space, it is
apparent how closely the vagus and the hypoglossal
nerve travel in relation to the pharyngeal space
(Figure 1). Depending on the amount of solution
and the depth of the injection, the local anesthetic
could diffuse to the nerves or be inadvertently
injected directly into the sheath that contains the
nerves. In the pediatric population, the risk may be
greater than in adults because children’s shorter
necks decrease the distance between the nerves and
the pharyngeal space. This may have predisposed
the child in the present case to a unilateral recurrent laryngeal nerve block resulting in the stridor
and difficulty with phonation. In addition, the
hypoglossal nerve may have been blocked, causing a
soft tissue obstruction due to the loss of motor
function to the tongue and the upper pharyngeal
muscles.1 Patient positioning also may be a contributing factor that is associated with accidental
nerve blocks after injection of local anesthetics into
the lateral pharyngeal space. Placing the patient in
the lateral head-down position to prevent pooling
of secretions theoretically could promote cephalad
spread of the local anesthetic solution.1
Considering the rapid onset and recovery of
the upper airway obstruction, it is unlikely the
patient experienced an accidental nerve block. If
the local anesthetic reached the nerve by diffusion,
the onset more likely would have been gradual. If
bupivacaine had been injected directly into the
sheath, the block would have been prolonged.
Furthermore, the injection was relatively superficial; accidental nerve blocks are more common
with deeper injections, such as those performed
for glossopharyngeal nerve blocks.1 Although the
evidence does not support the expected clinical
manifestations, the possibility cannot be completely ruled out and deserves some consideration.
Figure 1.
Anatomical location of the nerves
relative to the pharyngeal space
Figure 2.
Physiologic state of the lungs at rest
(From Spence and Mason.2 Reproduced by permission from Wadsworth Publishing Company.)
(Artist: Jay Knipstein)
Pathophysiology of NPPE
In the normal physiologic state, intrapulmonary pressures are equal to that of the atmosphere. This, balanced with the intrapleural
pressure and collapsing force of the lung, keeps
the lungs expanded in the thoracic cage (Figure
2). According to Boyle’s law, as the volume of a
container increases, pressure decreases; thus,
expansion of the chest wall causes the intrapleural
pressure to decrease and become subatmospheric
with resultant inward airflow.2 At the level of the
respiratory membrane, the pulmonary interstitial
space between the capillary and the alveolus is
ordinarily relatively dry, consistent with the
Starling hypothesis. The major forces maintaining
this balance include the pulmonary capillary
osmotic force promoting fluid movement into the
capillary, a relatively weak hydrostatic pressure in
the capillary opposing the inward movement, and
a negative interstitial pressure working concurrently with the interstitial colloid osmotic pressure
drawing fluid into the interstitial space. Lymphatic
capillaries drain the interstitial space of any fluid
in excess of that managed by the Starling mechanism (Figure 3).3-5(p600-602)
Three major factors contribute to the transudation of fluid into the interstitial space after the
relief of an upper airway obstruction. As inspiration occurs against a closed epiglottis (Müeller
Figure 3.
Forces causing fluid movement
(Modified from Guyton and Hall.4 Used by permission from WB
Saunders Company.)
maneuver), intrapleural pressures become subatmospheric, which in turn makes the interstitial
space markedly more negative, causing the transudation of intravascular proteins and fluids from
the pulmonary capillary into the interstitial
space.3,5(p600-602),6 In addition, hypoxia leads to pulmonary vascular constriction, which increases
AANA Journal/October 1999/Vol. 67, No. 5 427
pressure in the pulmonary capillaries. Concurrently, central nervous system-mediated αadrenergic stimulation leads to an increase in total
peripheral resistance.3,6 This systemic vasoconstriction leads to an increase in right ventricular
preload, pulmonary capillary volume increases,
and fluid leaks into the interstitial space. Left ventricular afterload also increases leading to
decreased left ventricular ejection (Figure 4).7–10 As
the interstitial space is flooded with fluid, compensatory mechanisms are overwhelmed, and
eventually the fluid penetrates and floods the
alveoli resulting in pulmonary edema. Symptoms
usually are seen only after the relief of the obstruction, due to the intermittent positive pressure
created with expiration against the closed glottis
(Valsalva maneuver). This positive pressure
opposes the negative pressure until the obstruction is removed.3,6,11
Clinical manifestations of NPPE
The onset of NPPE is usually immediate but
can occur up to several hours after an obstruction.12 Signs and symptoms of respiratory distress
are often present. Expectoration of pink frothy
sputum typically is seen as the alveoli are flooded
with fluid. The patient also may become hypoxic;
whether hypoxia occurs depends on the severity of
the condition.13 Auscultation of the chest reveals
rales and possibly wheezing as the peripheral
airways are compressed by excess fluid.11,13 The
chest radiograph frequently reveals diffuse interstitial and, depending on the severity of the condition, possibly alveolar infiltrates appearing as
“whited out” areas.11 Signs of sympathetic stress
stimulation, such as tachycardia, hypertension,
and diaphoresis, also will be apparent.14(p910)
Prevention of NPPE
Attempting to prevent or treat any upper
airway obstruction is the primary measure opposing
the development of NPPE. Some upper airway
obstructions common to the pediatric population in
the perioperative period are postextubation laryngeal edema and laryngospasm. Avoiding laryngeal
edema in the pediatric patient is crucial because a
small degree of edema in a pediatric patient may
lead to obstruction, whereas in an adult, edema may
result only in hoarseness. Measures to prevent laryngeal edema include ensuring a positive leak around
the ETT at less than 15 to 20 cm of water, delaying
surgery if crouplike symptoms are present, and
attempting to avoid excessive bucking on the ETT
on emergence from anesthesia.15(p1432)
428
AANA Journal/October 1999/Vol. 67, No. 5
Figure 4.
Pathophysiology of negative pressure
pulmonary edema
Inspiration against a closed glottis
Intrapulmonary
pressure becomes
subatomospheric
Markedly
negative
interstitial
pressure
Transudation
of fluid into
the interstitial
space
Hypoxia
α-adrenergic
stimulation
Increase in
total peripheral
resistance
Increase in
right ventricular
preload
Decrease in
left ventricular
ejection
Hypoxic
pulmonary
vasoconstriction
Increased
pressure in
the capillary
Fluid leaks
into the
interstitial
space
Contributes to
fluid in the
interstitial space
If signs and symptoms of croup develop
(such as inspiratory stridor, a barking cough,
hoarseness, or retractions) during the perioperative period, treatment must be implemented
promptly.16(p723) Cool humidified oxygen should be
administered. Racemic epinephrine (0.5 mL of
a 2.25% solution nebulized in 2.5 mL) may be
used for its vasoconstrictive properties. Steroids
may be administered (eg, dexamethasone, 0.1-0.5
mg/kg), although immediate relief may not
be seen due to the delayed onset of action. If
symptoms are severe or progress, reintubation
may be necessary with a smaller ETT.15(p1432),17
Laryngospasm is a common occurrence in
the pediatric patient undergoing a tonsillectomy
and is the most common cause of NPPE secondary
to an upper airway obstruction.6,10 The mechanism
involved in its onset is sensory stimulation of the
superior laryngeal nerves with resultant spasm of
the true vocal cords.18 It has been suggested that
prevention of laryngospasm on emergence may be
accomplished through extubation during deep
anesthesia when airway reflexes are absent or by
extubating in an awake state, thereby avoiding stimulation in a light depth of anesthesia when laryn-
gospasm is most common.17 Patients whose eyes are
deviated and pupils are dilated and who hold their
breath may be more susceptible to a laryngospasm.15(p1431) If extubating a patient who is awake,
the patient should have spontaneous movement of
all extremities, spontaneous eye opening, and a
regular respiratory rate and rhythm.9
Careful suctioning of the oropharynx before
extubation is essential. This removes blood and
secretions, thus preventing them from falling on
the glottic structures, potentially resulting in stimulation of the superior laryngeal nerve.19 When
extubating the patient, positive pressure should be
administered while removing the ETT just as the
patient is about to exhale. This fills the lungs with
oxygen and acts as a reserve in case of a laryngospasm. Stretch receptors in the lungs also are
stimulated, which induces forceful exhalation as
the ETT is removed (ie, the Hering-Breuer reflex),
aiding in clearing blood or secretions that may be
in proximity to the laryngeal structures.2,15(p1431) The
administration of IV lidocaine, 2 mg/kg, 1 minute
before extubation also has been shown to decrease
the incidence and severity of laryngospasms.15
If laryngospasm occurs, treatment must be
instituted immediately. If possible, remove the
stimulus by suctioning secretions from the oropharynx. The administration of 100% oxygen with continuous positive pressure may be enough to force
the vocal cords open.19 If this is accomplished, continued positive inspiratory assistance via manual
insufflation concurrent with the patient’s spontaneous ventilatory efforts will effectively keep the
vocal cords open until the crisis is resolved. Pushing
the mandible forward also may be necessary as this
maneuver displaces the tongue, the hyoid bone,
and the epiglottic cartilage attached to it. This not
only widens the pharynx, but also opens the
entrance to the larynx.20 If the laryngospasm is not
relieved in approximately 30 seconds, succinylcholine should be administered.9 Some advocate a
subapneic dose of succinylcholine (0.1-0.2 mg/kg);
however, if the patient’s condition is deteriorating,
it may be efficacious to administer an intubating
dose (1-2 mg/kg), reintubate the patient, and
extubate once the patient is awake with stable pulmonary mechanics and optimal oxygen saturation
parameters.9,19
The judicious administration of IVF also will
decrease the risk of developing NPPE after the relief
of an upper airway obstruction.14(p910) Adherence to an
appropriately conceived IVF plan is prudent for all
patients undergoing surgery with or without
evidence of pulmonary or cardiac disease.
Treatment of NPPE
With overt evidence of pulmonary edema,
treatment is directed toward reversing hypoxia
and decreasing the fluid volume in the lungs.
Fortunately, in almost 50% of the patients, supplemental oxygen and maintenance of a patent
airway are usually all that is required to maintain
adequate oxygenation.3 If oxygenation does not
improve with the administration of high inspired
concentrations of oxygen, the patient must be
reintubated and positive pressure ventilation with
positive end-expiratory pressure initiated. Endexpiratory inflation of the alveoli is promoted by
positive end-expiratory pressure, thus maintaining
and increasing the patient’s functional residual
capacity. This facilitates an improved ventilation/perfusion ratio and decreases intrapulmonary shunting.6,11,18 The administration of diuretics (eg, furosemide, 1 mg/kg) to effectively remove
excess intrapulmonary fluid has been advocated
by some; however, this is controversial and may be
reserved for patients with marked hypervolemia.8,13,20,21 Even when mechanical ventilatory
support is required, NPPE usually will resolve
within 24 hours.6,8,22(p159) The patient described in
the present case report was weaned from supplemental oxygen within 4 hours, monitored in the
pediatric intensive care unit overnight, and sent
home without further sequelae.
Conclusion
Although NPPE usually resolves spontaneously with supportive treatment, the rapidity of onset
and severity of the disease ultimately may lead to
death if the disease is not recognized promptly and
immediate intervention instituted. Thus, it must be
emphasized that NPPE can occur in any patient,
particularly patients with potentially obstructive
processes of the upper airway. Vigilant anesthesia
care warrants recognition of the patients at risk for
NPPE, with the institution of measures during provision of anesthesia to prevent the occurrence of
upper airway obstructions.
REFERENCES
(1) Bean-Lijewski JD. Glossopharyngeal nerve block for pain relief
after pediatric tonsillectomy: Retrospective analysis and two cases of
life-threatening upper airway obstruction from an interrupted trial.
Anesth Analg. 1997;84:1232-1238.
(2) Spence AP, Mason EB. The respiratory system. In: Human Anatomy
and Physiology. 4th ed. St Paul, Minn: Wadsworth Publishing Company;
1992:718-747.
(3) Lang SA, Duncan PG, Shephard DAE, Ha HC. Pulmonary
oedema associated with airway obstruction. Can J Anaesth.
1990;37:210-218.
(4) Guyton AC, Hall JE. Pulmonary circulation; pulmonary edema;
pleural fluid. In: Textbook of Medical Physiology. 9th ed. Philadelphia,
AANA Journal/October 1999/Vol. 67, No. 5 429
Pa: WB Saunders Co; 1996:491-499.
(5) Benumof JL. Respiratory physiology and respiratory function
during anesthesia. In: Miller RD, ed. Anesthesia. 4th ed. New York,
NY: Churchill Livingstone; 1994:577-620.
(6) DeVane GG. Acute postobstructive pulmonary edema. CRNA
Clin Forum Nurse Anesthetists. 1995;6:110-113.
(7) Baker MD, Ruddy RM. Pulmonary emergencies. In: Fleisher
GR, Ludwig S, eds. Textbook of Pediatric Emergency Medicine. 3rd ed.
Baltimore, Md: Williams & Wilkins; 1993:874-887.
(8) Halow KD, Ford EG. Pulmonary edema following post-operative
laryngospasm: A case report and review of the literature. Am Surg.
1993;59:443-447.
(9) Lee KWT, Downes JJ. Pulmonary edema secondary to laryngospasm in children. Anesthesiology. 1983;59:347-349.
(10) Willms D, Shure D. Pulmonary edema due to upper airway
obstruction in adults. Chest. 1988;94:1090-1092.
(11) Westenkirchner DF, Eigen H. Noninfectious disorders of the
respiratory tract. In: Hilman BC, ed. Pediatric Respiratory Disease:
Diagnosis and Treatment. Philadelphia, Pa: WB Saunders Co;
1993:505-507.
(12) Segal E, Perel A. Anesthetic considerations, special problems,
approaches, and problem solving: Pulmonary edema. In: Kirby RR,
Gravenstein N, eds. Clinical Anesthesia Practice. Philadelphia, Pa: WB
Saunders Co; 1994:969-981.
(13) Wilson GW, Bircher NG. Acute pulmonary edema developing
after laryngospasm: Report of a case. J Oral Maxillofac Surg.
1995;53:211-214.
(14) Rojo J, Iacopelli M. Respiratory Anesthesia. In: Nagelhout JJ,
Zaglaniczny KL. Nurse Anesthesia. Philadelphia, Pa: WB Saunders
Company; 1997:882-954.
(15) Stone DJ, Gal TJ. Airway management. In: Miller RD, ed.
Anaesthesia. 4th ed. New York, NY: Churchill Livingstone; 1994:1403-
430
AANA Journal/October 1999/Vol. 67, No. 5
1435.
(16) Chipas A. Airway management. In: Nagelhout JJ, Zaglanczny
KL. Nurse Anesthesia. Philadelphia, Pa: WB Saunders Company.
1997:708-725.
(17) Brett CM, Maurice SZ, France NK. Eyes, ears, nose, throat, and
dental surgery. In: Gregory GA, ed. Pediatric Anesthesia. 3rd ed.
1994:657-692.
(18) Ingrams D, Burton M, Goodwin A, Graham J. Acute pulmonary
oedema complicating laryngospasm. J Laryngol Otol. 1997;3:482-484.
(19) Murray-Calderon P, Connolly MA. Laryngospasm and noncardiogenic pulmonary edema. J Perianesth Nurs. 1997;12(2):89-94.
(20) Bainton CR. Management of the airway. In: Benumof JL,
Saidman LJ. eds. Anesthesia and Perioperative Complications. St Louis,
Mo: Mosby-YearBook; 1992:3-15.
(21) Soldano SL, Place LB, Edwards FH, Cohen AJ. Post-extubation
non-cardiogenic pulmonary edema. Mil Med. 1993;158:278-280.
(22) Hall SM. Respiratory anatomy and physiology. In: Nagelhout JJ,
Zaglaniczny KL. Nurse Anesthesia. 1997:150-165.
AUTHORS
Caroline L. Thomas, CRNA, MSN, is currently a staff nurse anesthetist at St. John Hospital, Detroit, Mich. At the time this article was
written, she was a second-year student of the Oakland UniversityBeaumont Graduate Program of Nurse Anesthesia, Royal Oak, Mich.
Timothy J. Palmer, CRNA, MS, is a staff anesthetist at William
Beaumont Hospital, Royal Oak, Mich. He is also a clinical and didactic
instructor in the Oakland University-Beaumont Graduate Program of
Nurse Anesthesia.
Peter Shipley, CRNA, MS, is a staff anesthetist at William
Beaumont Hospital, Royal Oak, Mich, and a clinical instructor in the
Oakland University-Beaumont Graduate Program of Nurse Anesthesia.