October 1994, Volume 44, Issue 10

Original Article

Post-Anaesthetic Pulmonary Oedema Following Upper Airway Obstruction

Mohammad Maroof  ( Department of Anaesthesiology and Intensive Care, King Fahad National Guard Hospital, Riyadh, Saudi Arabia. )
Rashid M. Khan  ( Department of Anaesthesiology and Intensive Care, King Fahad National Guard Hospital, Riyadh, Saudi Arabia. )
Bruce G. Ryley  ( Department of Anaesthesiology and Intensive Care, King Fahad National Guard Hospital, Riyadh, Saudi Arabia. )
Naveed Bad  ( Department of Anaesthesiology and Intensive Care, King Fahad National Guard Hospital, Riyadh, Saudi Arabia. )
Timothy Cooper  ( Department of Anaesthesiology and Intensive Care, King Fahad National Guard Hospital, Riyadh, Saudi Arabia. )


Twelve cases of post-anaesthetic pulmonary oedema (P0) secondary to upper airway obstruction (UAO) are reported. All were adult male patients undergoing uneventful elective surgical procedure under general anaesthesia. Post-anaesthetic laryngospasm was the single most important factor for the upper airway obstruction (UAO) in 5 (41.6%) patients. P0 secondary to partial UAO in drowsy patients was observed in 4 (33.3%) patients. UAO due to foreign body was responsible for P0 in two patients. A combination of negative intrathoracic pressure, hypoxia and associated hyperadrenergic state were the most likely causes of P0 in these patients with UAO. Early recognition, maintenance of patent airway and adequate oxygenation via face mask or endotracheal tube with mechanical ventilation resolved the syndrome within 6-36 horns in all of these patients. Invasive haemodynatnic monitoring or aggressive drug therapy were not applied in any of the patients. A heightened awareness among anaesthesiologists of the varied causes of post-anaesthetic UAO leading to P0 may help reduce the occurrence and facilitate early management of the potential complications (JPMA 44:244,1994.).


Since 1977 various case reports and larger series have described post-anaesthetic acute pulmonary oedema as a complication of upper airway obstruction (UAO). In majority of these patients post-extubation laryngospasm has been the cause of UAO1-9. However, we have closely examined all cases of pulmonary oedema (P0) during the immediate post-anaesthetic period from 1987 to 1993 at our institution and have observed a wide range of conditions responsible for producing the UAO. We present our experience of 12 cases of post-anaesthetic UA0 induced P0 in adults to highlight the need for early recognitionand appropriate management of this syndrome.

Case Reviews

Twelve ASA I and II adult male patients (26.2SD±8.3year, range 16-45 year; 68SD±15.4 kg. range 45-95 kg) undergoing general anaesthesiaforelective surgical procedures developed acute P0 in the post-anaesthetic period 1987 to 1993. Nine of these patients had undergone surgery of the head and neck region (Table).

All patients underwent an uneventful general anaesthetic using nitrous oxide and oxygen with isoflurane or enflurane and intermittent fentanyl in appropriate dose. All patients were intubated and ventilation was controlled after relaxation with atracurium or vecuronium. At the end of the surgery, residual muscle paralysis was adequately reversed monitoring train-of-four ratio in each case. Patients were extubated after resumption of adequate spontaneous respiration. Patient # 1 to 6 (Table) were adequately breathing and saturating well (arterial oxygen saturation, Sa02>95%) prior to transfer from operation room (OR) to post-anaesthesia recovery room (PARR). However, patients # 1 to 4 were recorded to be drowsy on arrival in PARR. Patient # 1 was noted to have obstructed breathing while patient # 5 was observed to be apneic and cyanosed on arrival in PARR. Patient #6 developed laryngospasm on arrival in PARR. All these 6 patients gradually desaturated (Sa02<90%) and went on to develop signs and symptoms of acute P0 over the next 5-45 minutes despite appropriate airway management and oxygen (02) administration with face mask. Use of furosemide, morphine and appropriate respiratory manage­ment quickly resolved the pulmonary oedemain less than 24 hours in all patients. Patient # 7 (Table) underwent elective excision of a tongue lesion under general anaesthesia. A pharyngeal pack was placed by the attending anaesthesiologist. At the time of extubation the relieving anaesthesiologist, unaware of the phaiyngeal pack, extubated the patient. The patient soon developed obstructed breathing and oxygen desaturation. Laiyngoscopy within 5 minutes revealed the obstructing pack. Despite the relief of obstruction, patient developed bilateral chest crepitation 30 minutes later in the PARR. Patient responded well to conservative management without the need for intubation and positive pressure ventilation. Patient # 8 (Table) a 33 year old male with an oroniaxillary fistula underwent uneventful repair under gen­eral anaesthesia. Towards the end of the surgery, a size 18 Foley catheter was passed through his right nostril and the balloon inflated in the post-nasal space to provide tamponade to stop further bleeding. The free limb of the catheter was anchored to the right cheek with the aid of adhesive tape. Following uneventful extubation and spontaneous breathing for a few minutes, the patient was transferred to PARR. Within minutes of arrival, his oxygen saturation began to drop rapidly from the initial 100% despite oxygenation via bag mask. The patient became progressively cyanosed and bradycardic. Laryngoscopy revealed the balloon of the Foley catheter completely obstructing the larynx. Balloon was removed and the patient intubated. Florid PO was evidenced by the copious pink froth that issued from the endotracheal tube. He responded well to intra-venous morphine, diuretics and intermittent positive pressure ventilation (IPPV) with mild positive end expiratory pressure (PEEP). He was discharged from the ICU on the 2nd post-operative day. Patient #9 to 12 (Table) had uneventful intra-operative period. On extubation, each developed acute laiyngospasm. Suxamethonium 25-50 mg was administered to relieve the spasm after initial failure to ventilate them with bag mask. All these patients rapidly desatunted (Sa02<90%) and were re-intubated in the OR. Each of these 4 patients demonstrated florid pulnionaiy oedema in the form of copious pink froth issuing via the endotracheal tube. After initiation of therapy with fumsemide, morphine and IPPV they were transferred to the ICU for continuing management of acute P0. All were extubated within 24-36 hours and discharged from the ICU in less than 48 hours with complete resolution of acute P0. Increasing hypoxia, as demonstrated by the pulse oximetry, was theflrstsignofimpendingPO ineachcase. This was further confirmed by arterial blood gas analysis. The clinical diagnosis of the P0 varied from volume of fluid pouring up an endotraeheal tube to coughing blood stained sputum. All patients had crepitations on auscultation of the chest. Diagnosis of P0 was confirmed by chest x-ray in all patients. This was followed with serial chest x-rays which showed quick resolution. All patients had a final normal chest x-ray prior to discharge from the hospital.


P0 associated with UAO have been called negative pressure pulmonary oedemabecause it is largely related to the development of markedly negative intra-pleural pressure8,9. A number of conditions can produce UAO. Willms and Shure7 analyzed 26 cases of P0 due to UAO from the literature and observed that laryngospasm was the most common cause of UAO in 42.3% (11/26). 9 of these 11 patients (81.8%) had post-extubation laryngospasm. Our observations were nearly identical. UAO secondary to laryngospasm was observed in 41.6% (5/12) of our patients. Of these 80% (4/5) had post-extubation laryngospasm. Lorch and Sahn5 have identified 3 factors predisposing to UAO. These are: 1) Anatomically difficult intubation; 2) Nasal, Oral or Pharyngeal surgical site; 3) Obesity. One or more of these factors were present in 5 of our patients (41.6%). However, there are several case reports to demonstrate that this syndrome may also complicate UA0 in patients without additional identifiable risk factors3,7,10,12. The pathogenesis of P0 associated with UAO is multifactorial5. However, the principal factor leading to P0 following UAO appears to be the generation of markedly negative intra-thoracic pressure due to forceful inspiratory effort against a closed glottis. This results in a decrease in pulmonary interstitial pressure favouring transudation of oedema fluid from puhnonary capillaries13. It has been observed that markedly negative intra- thoracic pressure alone can explain the association of UAO and PO14,15. A decrease in intra-thoracic pressure may also lead to augmented venous return which theoretically increases pulmonary blood volume and pulmonary arterial pressure. This fact also may contribute to the hydrostatic transudation of fluid. In addition, increased venous return is associated with right ventricular distention and a decrease in left ventricular compliance as a result of the left ward shift of the intra-veniricular septum (ventricular inter-dependence) 16,17. This leads to an increase in the left ventricular end-diastolic pressure thereby further increasing the pulmonary vascular pressure and hydrostatic transudation of fluid. All our patients were young males and ASA I or II. It has been shown that these young healthy patients may be at increased risk for laiyngospasm-induced P0 because they can generate large negative intra-thoracic pressure18. In our series patients #2 to 5 (Table I)who were drowsy on arrival in the PARR, exhibited semi-obstructed airway possibly due to the approximation of the tongue to the posterior pharyngeal wall. Nevertheless they went on to develop clinical picture of P0 despite appropriate airway management. Nearly identical mechanism of UAO was observed by Warner et al19. We postulate that UAO need not be absolute like laryngospasm for the P0 to develop. Even mild to moderately obstructed airway can lead to negative intra­thoracic pressure, hypoxia and subsequent P0. Although negative intra-thoracic pressure is the primary pathological event in the development of P0 associated with UAO, hypoxia and hyperadienergic state both contribute to its development. Hypoxia can alter capillary integrity and pro­duce a hyperadrenergic state5,20. This hyperadrenergic state may be associated with a redistribution of blood from the periphery to the pulmonary circuit21 and to increased pulmo­nary vascular resistance22,23. Furthermore, hypoxia alters peri-capillary pulmonary vascular resistance ma non-uniform fashion24-26 leading to a generalized increase in pulmonary vascular pressure. Finally, hypoxia and metabolic acidosis may produce direct myocardial depression27 and may potenti­ate other factors known to enhance formation. Peri-operative administration of naloxone has been implicated intriggering PO28,29. However none of our patients received peri-anaes­thetic nareotic antagonists. In all our patients with post-exubation laiyngospasm (Case # 9-12) and in the patient with Foley balloon obstruc­tion, florid P0 was evidenced only after the relief of UAO by intubation. It is postulated that UAO creates more positive pressure during expiration which serves as a form of “auto-PEEP” to oppose transudation until the obstruction is re­moved5,7,30,31. In all these patients, the subsequent course of events was nearly identical to those previously reported6,18,32,34. Lang et al35 reviewed 77 cases of P0 associated with UA0 and observed that 85% required tracheal intubation for a short period, 50% needed mechanical ventilation and about 50% required continuous positive airway pressure or positive end-expiratory pressure. This was in contrast to 66.6% of our patients (8/12) who had to be intubated. All these patients were mechanically ventilated from 6 to 36 hours. Furosemide was administered to all our patients. Her­rick, Mahenderan and Penny36 reviewing 19 cases in literature of post- obstructive P0 found diuretic administration in 13 cases (68.4%). The use of diuretic has been questioned by some. in the light of normal pulmonary capillary wedge pressure (PCWP)37. However, it is recommended by others30. Invasive haemodynamic measurements were not per­formed in any of our patients as data from previous studies suggests a normal haemodynamic picture (central venous pressure, pulmonary artery pressure and PCWP) in these patients3,10,38,39. This is a characteristic finding into following UA0. Invasive haemodynamic monitoring is important in situations where the diagnosis is not clear especially in patients with uncertain iatrogenic volume overload and/or cardiogenic aetiologies. In few of our cases radiological differential diagnosis suggested aspiration pneumonia/PO despite a complete lack of history of regurgitation and aspiration. However, the management of aspirationpneumonia is identical to that of P0 associated with UAO unless an infectious complication ensues40. A rapid resolution of the radiological findings in less than 36 hours in all our cases pointed more towards P0 secondary to UAO than to aspirationpneumonitis. We have observed a variety of causes of UAO which all led to a similar syndrome of rapid onset of P0 followed by quick resolution with appropriate therapy. We should be vigilant therefore not only for post-anaesthetic laryngospasm but also for the partially obstructed airway as may be seen in the drowsy patients and rare iatrogenic causes of airway obstruction. Aggressive haemodynamic monitoring, me­chanical ventilation, or drug therapy are not mandatory. Maintenance of adequate oxygenation and a patent airway are the main stays of its treatment.


1. Jenkins, J.G. Pulmonary edema following laiyngospasm. Anesthesiology, 1984;60:611.
2. Melnick, B.M. Postlaryngospasm pulmonary edeman in adults Anesthesiology, 1984;60;516­-17.
3. Weissman, C., Damask MC. and Yang, J. Non cardiogenic pulmonary edema following laryngeal obstruction, Anesthesiology, 1984;60: 163-65.
4. Mc Gongle, M. and Kennedy, T.L. Laryngospasm induced pulmonary edema. Laiyngoscope, 1984;94: 1583-85.
5. Lorch, DO. and Sahn, S.A. Postextubation pulmonary edema following anesthesia indneedby upperaiiweyobstruction; are certain patients atincreasedrisk? Chest, 1986;90:802-5.
6. Frank, L.P. and Schreiber, D.C. Pulmonary edema following acute upper airway obstruction Anesthesiology, 1986;65: 106.
7. Willims, D. and Shure, D. Pulmonary edema due to upper airway obstruction in adults. Chest, 1988;94: 1090-92.
8. Stalcup, &A. and Mellin, RB. Mechanical forces producing pulmonary cedems in acute asthma. N. Engl. 3. Med. 1977;297:592- 6.
9. Newton-John, H. Pulmonary oedema in uppr airway obstruction. Lancet, 1977;2:510.
10. Cozanitia, DA, Leijala, M., Pesonen, S. et al. Acute pulmonary oedesna due to laryngeal spasm. Anaesthesia, 1984;37:1198-9.
11. Anderson, Kancir, C.B., Nielson, KD. Lsryngospsam-induced pulmonary oedcma. Acts Anaestheaiol. Scand, 1988;32:710-1.
12. t3lasser, S.A. and Bier, J.N. Delayed onset of laryng ospasm induced pulmonary edema in an adult outpatients Anesthesiology, 1985;62:370-1.
13. Galvis, AG., Stool, SE. and Bluestone, CD. Pulmonary edema following relief of acute uppersirwsyobstruction. Ann. Otol. Rhinol. Laryngol., 1980;89:124-8.
14. Smith-Erichsen, N. sndBo, G. Airwayclosure and fluid filtration in the lung. BL J.Anaeath., 1979;51 :475-9.
15. Lloyd, J.E., Nolop, KB., Parker, RE. et si. Effects ofinspiratoayresistance loadingoflung fluid balance inawake sheep. J. Appl. Physiol., 1986;60: 198-203.
16. Peters, J. Kindred, M.K. and Robotham, J.L. Transient analysis of cardiopulmonary interactionsl. Diastolic events. J. Apl. Physiol., 1988;64:1506-17.
17. Peters, J., Kindred, M.K. and Robotham, iL. Transient analysis of cardiopulmonary interactions!!. Systolicevents. J. AppL Physiol, 1988;64:1518-26,
18. Anderson, AR, Alfrey, D. and Lipscomb, AS. Acute pulmonary edema, and unusual complication following arthroscopy: A reportofthree cases. Arlhroscopy, 1990;6:235-7.
19. Warner, L.O., Msrtino, J.D., Davidson, P.J. et al. Negative pressure pulmonary oedema: a potential hazard of muscle relaxantain a wake infants. Can. 3. Anaesth., 1990;37:580-3.
20. Moss, G., Stanton, C. and Stein, A.A. The centrineurogenic etiology of acute respiratory diatresasyndrome. Universal,species Independent phenomenon. Am.J. Surg. 1973;126:37­-41.
21. Samoff, S.J., Berglund, E. and Samoff, L.C. Neurohemodynsmics ofpulmonary edema III Estimated changes in pulmonasy blood volurneac compsny in gaystemic vaso constriction and vasodilatalion. J. AppI. Physiol., 1953;5:367-74.
22. Samoff S.J. and Samoff L.C. Neurohemodynsmic of pulmonsay edema II. The role of sympathetic pathways in the. elevation of pulmonary and systemic vascular pressure following theintacisternal injection of fibrin. Circulation, 1952;6:51-62.
23, Theodore, J., Robin, ED. Speculations onneurogenic pulmonary edema. Am. REv. Respir. Dis., 1976;113-405-11.
24. Vlswanathan, R., Subrasnsnian, S. and Radha, TO. Effects of hypoxia onregional perfusion by scanning. Respiration, 1979;37: 142-7.
25. Schoene, RB. Pulmonary edema at high altitude: a review, pathophysiology and update. Chin. Chest. Med., 1985;6:491-507.
26. Kabayaahi, T., Kajama, S., Kubo, K. et al. Clinical features of patients with high altitude pulmonary edema in Japan. Chest, 1987;92:814-21.
27. Mitchell, R.A. Wildenthsl, K. and Johnson, R.L. The effects of acid-base disturbance on cardiovascular and pulmonary fuction. Kidney Int., 1972;1 :375-89.
28. Taff, RH. Pulmonary edema following naloxone administration in a patient without heart disease. Anesthesiology, 1983;59:576-7.
29. Flacke, J.W,. Flscke, WE. and Williams, 0. Acute pulmonary edema following naloxone reversal of high-dose morphine anesthesia. Anesthesiology, 1977;47:376­
30. Kanten, R.K. and Watchko, J.F. Pulmonary edema associated with upper airway obatmction. Am. 3. Dis. Child., 1984;138:356-8.
31. Sofer, S., Bar-ZIv, 3. and Scharf, SM. Pulmonary edema following relief of upper airway obatmction. Cheat, 1984;86:401-3.
32. Jackson, F.N., Rowland, V. and Corasan, G. Laryngoapasm induced pulmonary edema. Cheat, 1980;78:819-211.
33. Lee, K.W.T. and Downes, J.J. Pulmonary edema aecondsry to laryngoapaam in children. Anesthesiology, 1983;59:347-9.
34. Oswalt, CE., Gates, GA., Holstrom, M.G. Pulmonary edema as a complication of acute airway obstruction. JAMA, 1977;238: 1833-5.
35. Lang, S.A., Duncan, P. G.,Shephard, D.A.E. et al. Pulmonary oedema associated with airway obstruction. Can. J. Anaeath., 1990;37:210-8.
36. Herrick, IA., Mahendran, B. and Penny F.J. Postobatructive pulmonary edema following anesthesia. J. Clin. Anesth., 1990;2:116-20.
37. Barin, E. S., Stevenson, I.E adn Donnelly, G.L. Pulmonary oedema following acute upper airway obstruction. Aneath. Intena. Care, 1986;1] 4:54-7.
38. River, M., Hadlock, F.B. and O’Mear, ME. Pulmonary edema secondary to acute epiglottitia. Am. J. Roentg., 1979,132:991-92.
39. Steadling, JR. and Bolton, P. Upper airway obstruction as cause of pulmonary oedema. Lancet, 1982;1:1352-4.
40. Miller, R.D. Anesthesia. 2nd ed., New Tork, Churchill Livingatone, 1986,pp.2043-5.
41. Atkinson, R.S., Ruahman, GB. and Alfred Lee,i. Preanasthetic assessment In a Synposia of Anaesthesia. 10th ed. Bristol. lOP Publishing Limited, 1987;P.109.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: