Volume 10, No. 2 • Summer 1995

In My Experience:Non-Cardiogenic Pulmonary Edema After Difficult Intubation

James Kie-Chul Ohn, M.D.

Non-cardiogenic pulmonary edema associated with acute airway obstruction in an adult age group appears to be very rare. (1) Non-cardiogenic pulmonary edema has been described in pediatric age groups who had croup and epiglottitis (2) laryngospasm at the end of anesthesia, (3) and succinylcholine before induction of anesthesia.

We observed acute non-cardiogenic pulmonary edema on awakening from anesthesia in an adult whose trachea was intubated with difficulty in the beginning of anesthesia.

Case Report

A 57-year-old man was scheduled for right colon resection. He was found to have a cecal mass with some hemorrhage. The patient was 6 feet tall and weighed 200 pounds. His past medical history was unremarkable. He had no previous surgery or anesthesia. Physical examination did not reveal any abnormalities. Electrolytes were within normal limits. Hemoglobin was 14.5 gm/dl and hematocrit was 44. Chest x-ray was normal. EKG showed nonspecific ST-T changes. The patient was premedicated with 50 mg of mephedrine, 75 mg of pentobarbital and 0.4 mg of atropine one hour before the anesthesia started. The anesthesia was induced with 400 sodium thiopental, and 100 mg of succinylcholine was given to facilitate intubation.

However, the larynx was not able to be visualized because of its extreme anterior location. The patient was ventilated with 100% 02 and 2% of enflurane was administered. After a few trials with Macintosh and Miller’s blade, it was decided to bring spontaneous breathing back to perform bend nasal intubation. However, severe laryngospasm was encountered and succinylcholine had to be given to break the spasm. This was attempted twice. Luckily, the third anesthesiologist was able to intubate the trachea by making a hockey stick shape with the stylet inserted to the endotracheal tube. The procedure took an hour using 600 mg of sodium thiopental and 260 mg of succinylcholine. Blood pressures were 160-170/90-110 and pulse rates were 100130/min. Mask ventilation was moderately difficult.

A few minutes after intubation, mild cyanosis in finger tips was observed and rales were heard on both lung fields. Very little white secretions were auctioned and lungs became clear as the anesthesia was deepened. Cyanosis in the finger tips disappeared. One thousand ccs of lactated Ringer’s with 5% dextrose solution was given during that time. He had been NPO since midnight and anesthesia was started at noon. The anesthesia was maintained with 1% of enflurane/N20/02 and fentanyl. Pancuronium was used for muscle relaxation. He was mechanically ventilated with tidal volume 800 ml and rate 10/min. Inspiratory pressure was 28 cmH20. Blood pressure was maintained 10012060-80 and pulse rate was 70-100/min. Total fluid given was 2,300 ml and estimated blood loss was 500 ml. The operation lasted three and a half hours from the beginning of anesthesia. Pancuronium was reversed with 10 mg of pyridostigmine and 1.0 mg of atropine and spontaneous breathing was resumed.

The patient was brought to recovery room and put on T-piece. Upon awakening, he was bucking on the endotracheal tube and bringing up pinkish frothy material through the tube. He was cyanotic, but awake and following verbal command. Blood pressure was 170/90, pulse rate 96/min. Coarse rales were heard on both lung fields. ABC’s were pH 7.32, P02 48, PCO2 46 on T-piece with 10/min. 100% O@. Chest x-ray showed widespread alveolar densities in a butterfly distribution in both perihilar regions. The heart size was normal. EKG did not reveal any evidence of ischemia or infarction. Furosemide and morphine sulfate were given and he was mechanically ventilated. In 24 hours, pulmonary edema was completely resolved and he was transferred to a regular floor from ICU. Serial EKG and cardiac enzymes did not show any abnormal changes.


Inspiratory efforts against closed glottis, hypoxia, difficult intubation under light anesthesia and awakening from anesthesia might have contributed to leakage of fluid to alveoli. All of these are interrelated to cause imbalance in Starling forces. The forces tending to move fluid outward are mean capillary pressure, negative interstitial pressure and interstitial fluid colloid osmotic pressure. Plasma colloid osmotic pressure holds fluid inward. Net effect results in slightly more outward forces than inward. It is balanced by fluid return to the circulation through the lymphatics. Difficult intubation under light anesthesia stimulates the sympathetic nervous system. Release of catecholamines causes peripheral vascular constriction which shifts systemic blood to pulmonary circulation. Emergence from anesthesia may increase catecholamine release because of pain, hypothermia and shivering. Shift of blood volume to the pulmonary circulation causes an increase in capillary pressure which favors outward movement of fluid. Neurogenic pulmonary edema is often explained by this mechanism. (5) Hypoxia in alveoli is known to cause hypoxic pulmonary vasoconstriction which exerts favorably to minimize ventilation perfusion mismatches. Pulmonary arteriolar constriction increases pulmonary capillary pressure.

Systemic hypoxemia increases catecholamine release which may contribute to systemic vasoconstriction and shift of blood to pulmonary circulation. Thus, both alveolar hypoxia and hypoxemia can raise pulmonary capillary pressure and enforce outward fluid movement in Starling equilibrium.

Undue Pressure

Inspiration through a closed glottis may increase transpulmonary intrathoracic pressure gradient. Transmission of this pressure to peribronchial interstitial space becomes additive to outward fluid movement from the capillary. In other words, negative interstitial fluid pressure which normally exists is greatly enlarged. (3) A large intrathoracic subatmospheric pressure affects left ventricular function. (6) Increase in afterload, decrease in left ventricular volume and ejection fraction, and decrease in velocity of contraction were observed during inspiration against closed glottis. (6) These changes favor a rise in pulmonary blood volume and pulmonary capillary pressure which, in turn, leads to the transudation of the fluid from the capillary into the alveolus.

Endothelial Injury

Simple imbalance in Starling forces does not explain all facts of non-cardiogenic pulmonary edema. Physical capillary endothelial cell injury has been suggested as a mechanism of fluid leakage. (7) As vascular distending pressure is increased, the tangential stress developed in the walls of microvascular endothelial cells ultimately pulls open the intercellular junctions. It may explain higher edema fluid to serum protein ratio in noncardiogenic pulmonary edema than cardiogenic pulmonary edema. (8) Although the documented cases in which pulmonary vascular pressures were measured are rare, (9) transient rise of pulmonary capillary pressure during episodes of catecholamine release may cause this type of physical injury to endothelial cells. Normal or lower than normal pulmonary vascular pressures can be expected after the actual incident.(1) The injury done by this mechanism appears to be reversible by supportive therapy in a relatively short period of time (1,3,4) Intramuscular succinylcholine administered to unanesthetized infants was associated with non-cardiogenic pulmonary edema. (2) Succinylcholine to our patient had a doubtful effect on pulmonary edema. Frightening due to paralysis while being awake might have increased catecholamine release, Repeated laryngoscopies to intubate the trachea in the patient under inadequate anesthesia might have increased catecholamine release which could lead to systemic vasoconstriction and shift of blood to pulmonary capillaries. Laryngospasm occurred twice on resuming spontaneous breathing to attempt blind nasal intubation. It resulted in hypoxia and excessive transpulmonary intrathoracic pressure gradient: both increased pulmonary capillary pressure and the latter causes more negative interstitial pressure. By the time endotracheal intubation was accomplished, stretched pores between capillary endothelial cells had perhaps been formed and leakage of fluid into alveoli already started. This may explain initial cyanosis and rales in the chest. Controlled ventilation and deepening of anesthesia by enflurane must have halted progression.

Fluid Leak

Upon awakening from anesthesia in the recovery room, negative pressure in interstitial space created by spontaneous breathing and catecholamine release by pain and shivering set up favorable conditions to move capillary fluid through already physically injured membrane into alveolus. This could explain the event in the recovery room.

Non-cardiogenic pulmonary edema associated with difficult intubation in an adult is rare, but it can be reasonably explained by the mechanism involved in neurogenic and pulmonary edema occurring in acute airway obstruction in the pediatric age group. Awareness of this complication by anesthesiologists should lead to prompt diagnosis and treatment. High edema fluid to serum protein ratio (0.84) (7) and normal or lower pulmonary capillary wedge pressure should confirm the diagnosis. (1) Deep anesthesia and maintenance of good airway are thought to be the best ways to prevent this complication when difficult intubation is encountered. The treatment consists of mechanical ventilation with PEEP or CPAP and oxygen. Diuretics may or may not be needed. Measurement of the pulmonary capillary wedge pressure should best guide safe administration of diuretics and fluid.

Dr. Ohn is Chief of Anesthesia at Landmark Medical Center, North Smithfield, RI.


1. Gardaz JP, Foster A, Suter PM. Edeme pulmonaire aigu fuminant sans defaillance cardioque en fin d’anesthsie. Cana Anaesth Soc J 1979; 26: 34-37.

2. Cook DR, Westman HR, Rosenfeld L, Hemershot RJ. Pulmonary edema in infants: possible association with intramuscular succinylcholine. Anesthesia and Analgesia 1981; 6: 220-223.

3. Lee Kwt, Downer JJ. Pulmonary edema secondary to laryngospasm in children. Anesthesiology 1983; 59:347349.

4. Travis KW, Todres ID, Shannon DC. Pulmonary edema associated with croup and epiglottitis. Pediatrics 1977; 59: 695-698.

5. Robin ED, Theodore J. Pathogenesis of neurogenic pulmonary edema. Lancet 1975; 2:749-751.

6. Buda AJ, Pinsky MR, Ingels NB. Effects of intrathoracic pressure on left ventricular performance. N Engl J Med 1979; 301: 453459.

7. Staub NC. The pathogenesis of pulmonary edema. Prog Cardio Vac Dis 1980,23:53.

8. Sprung CL et al. The spectrum of pulmonary edema: Differentiation of cardiogenic, intermediate, and non-cardiogenic forms of pulmonary edema. Am Rev Respir Dis 1981; 124:718.

9. Wary Np, Nicotra MB. Pathogenesis of neurogenic pulmonary edema. Am Rev Respir Dis 1978,118: 783-786.