Near-Drowning in Cold Fresh Water: Current Treatment Regimen

NEAR-DROWNING IN COLD FRESH WATER: CURRENT TREATMENT REGIMEN

A.W. CONN, J.F. EDMONDS, AND G.A. BARKER

DROWNING is defined as death by suffocation recovery, .7-t9 partly due to the "diving under water I and annually accounts for approx- reflex". Is,z~ This is a neurogenic reflex, inde- imately 1,300 deaths in Canada, 7,000 in U.S.A., 2 pendent of baro-receptors and chemo-receptors, and 140,000 around the world. 3 A newer term, which shunts blood away from non-essential or- "near-drowning", is defined as a submersion vic- gans to the heart and brain and is associated with tim who arrives at an emergency facility and sur- marked bradycardia. This reflex is triggered by vives for 24 hours. 4 This definition excludes submersion of the face with apnoea and is most those who "drowned", but ignores the degree active in young children, in cold water (<20 ~ C) of subsequent recovery. Contrary to some re- and is potentiated by fear. After rescue, it is usu- ports, 5"6 most papers 4.7,a-to and personal obser- ally impossible to verify the presence of this vation suggested that, using routine therapy, ini- reflex, as both severe hypoxia and hypothermia tial improvement was too often followed by pro- can cause bradycardia. Nevertheless, even if gressive cerebral deterioration. Routine therapy cardiac arrest occurs after a prolonged period of included oxygen and artificial ventilation as indi- bradycardia due to the driving reflex, the prog- cated, rapid restoration of body temperature, cir- nosis is much more favourable. culatory Support and maintenance of urine vol- Immersion in very cold water (5~ C) 22 has been ume. In 1975, a five-year ,'etrospective review of shown to produce an incredibly rapid fall in body 30 near-drowning victims at The Hospital for tempe,-ature, 23 which can be accelerated by vig- Sick Children, Toronto, revealed permanent orous swimming. In small children the rate of brain damage in 30 per cent t~ (Table I). At that temperature fall is maximal because of a rela- time, many reports ~2-ts had noted that if the tively larger surface a,'ea and lack of insulation. 9 drowning incident happened in very cold water, If subme,-ged, the swallowing of cold water in complete recovery could occur despite prolonged large quantities and the aspiration of ice water (15 submersion (up to 40 minutes), t6 This paper dis- per cent of cases exceed 22 ml/kg 24) are addi- cusses those factors which affect the ultimate fate tional reasons for rapid core cooling. In our series of near-drowning victims in cold fresh water and of 26 cases ( 1975-1977) admitted to The Hospital recommends rational therapeutic measures to for Sick Children after various intervals of treat- minimize permanent brain damage. ment, routine rectal temperatures ranged between 29~ C and 39.5 ~ C but were below 35.5 ~ C in FACTORS AFFECTING SURVIVAL 46.1 per cent. In older children at approximately 35 ~ C alertness and at 33 ~ C consciousness are The survival and ,extent of recovery after a diminished, resulting in an early death fiom sub- submersion accident is determined by a number mersion and drowning* or a later death fi'om of factors. These include certain physiological progressive immersion hypothe,'mia, zs In either characteristics of the individual, the occurrence case the victim appears irretrievably dead and of immersion hypothermia, the submersion time, resuscitation may not be attempted, which is the initial resuscitation efforts and, later, the in- WRONG! The rapidly lethal effect of immersion tensive care management. hypothermia is partly counter-balanced by pro- It is generally accepted that, within narrow ducing simultaneously significant protection to margins, there is individual variation in tolerance the hypoxic brain. This is one of the few fringe to, and recovery from, severe hypoxia. Of greater benefits of the cold climate of central Canada. relevance is the age of the patient. The younger This protective effect is well documented in elec- the victim the greater the possibility for complete tive cardiovascular surgery :~176 where circulation can be discontinued for as long as 45 minutes A.W. Corm, M.D., F.R.C.P.(C), J.F. Edmonds, during profound hypothermia with complete re- M.B., F.R.C.P.(C), G.A. Barker, M.B., F.F.A.R.A.C.S., Intensive Care Unit, The Hospital for *This may be defined as submersion hypathermia Sick Children, Toronto and the Department of Anaes- where immersion hypothermia occurs with asphyxia of thesia, University of Toronto, Toronto. immediate onset. 259 Canad. Anaesth. Soc. J., vol. 25, no. 4. July 1978 260 CANADIAN ANAESTHETISTS' SOCIETY JOURNAL

TABLE I RESULTS (ALL CASKS)

Number Deaths CNS damage

1970-1974 30 1 (3.3~) 9 (30.07o) (routine) 1975-1976 li 0 2 08.17,) (Hyper R - 3 cases) 1977 15 4 (26.6~) 1 (6.6~)* (Hyper R - 7 cases) (Table 11) TOTALS (1970-1977) 56 5 (9~) 12 (21%

*Admitted five days post-drowning. covery. Since drowning is not elective (!) cooling obtained by Dr. Mathew Spence of Auckland, lags behind hypoxic effects but has still proven to N.Z., in adults with severe head injuries, 3-~ using be protective j3-j6 and should be continued after hypothermia and hyperventilation, among other rescue. In warm-water drowning with cardiac ar- measures, for as long as two to three weeks. It rest, immediate induction of hypothermia has was considered that victims of near-drowning in been recommended 9'2''31's2 and may play a simi- cold fresh water might have less severe brain lar but less successful role. injury and that, using a similar approach, our The duration of submersion is very difficult to results might improve. The prime aim of therapy determine after the event, because of the emo- became "cerebral salvage". At intervals during tional excitement existing at the time. Estima- the next three years, all known measures for this tions in our 26 cases (1975-1977) showed <5 purpose were gradually introduced and took ab- minutes, 8 cases; 5-10 minutes, 7 cases; >10 solute priority over any other therapeutic consid- minutes, 6 cases; and unknown, 5 cases. It is erations. The twin goals of therapy became the generally accepted thal 4 to 5 minutes under preservation of any surviving cerebral cells water at normothermla represents a maximum which might be temporarily non-functioning, and time before irreversible neurological damage oc- the prevention of post-asphyxial brain swelling curs. This time-period needs re-examination in and subsequent rise in intracranial pressure. Fol- view of treatment outlined later. However, there lowing three years of trial and error, the following is little doubt that full recovery is unlikely after measures can be recommended as providing op- prolonged submersions (>30 minutes) even in timal management in near-drowning victims. very cold water'. The need for immediate resuscitation in all METHOD OF TREATMENT cases and regardless of circumstances cannot be over-emphasized. Mouth-to-mouth breathing Time is short, and closed chest massage should be instituted in The odds are great, the water, if possible, and maintained as long as The margins small, necessary or until cerebral death supercedes. The stakes infinite ... The heart of a normothermic child, if untreated, Winston Churchill can be restarted as long as 45 minutes after sub- (paraphrased) mersion, but survival with an intact brain is most "Their Finest Hour" unlikely. If the resuscitation time of a hypoxic normothermic heart exceeds 15 to 20 minutes, Following rescue, resuscitation, and stabiliza- chances of complete cerebral recovery also di- tion of the circulation, certain clinical features minish. Even a few minutes delay in commencing appear which may affect brain recovery. These treatment may be critical as far as recovery of the include evidence of hyperhydration and evidence brain is concerned. In the presence of severe of marked brain damage such as hyperventila- immersion or accidental hypot hermia, time limits tion, hyperpyrexia, hyper-excitability and hy- should be ignored until core temperature reaches per-rigidity. For discussion purposes and ease of 30 ~ C 33.J4 memory, the acronym H.Y.P.E.R. will be used, In 1975 a more aggressive approach was introduced in the intensive care management of near- Hyperhydration (H) drowning. This was based on )he good results During the drowning process, unknown quan- CONN, el C/].: NEAR-DROWNING 261 tities of fresh water may be both ingested and consumption, increased pulmonary shunting aspirated. After prolonged submersion, resusci- with hypoxaemia and hypocapnia. Blood oxygen tation requires large volumes of intravenous tensions on arrival in emergency departments fluids and sodium bicarbonate as well as other vary greatly, but in a high proportion of cases (77 volume expanders. Later, when the circulation per cent) ~~are Iow.t becomes stable, there is both a relative and abso- In all such cases, prophylactic nasotracheal lute overhydration in the presence of damaged intubation is done to prevent even minor airway capillaries. Fluid leaks into the tissues, particu- obstruction, which can be so disastrous to the larly into the lung and the brain. This state is post-hypoxic brain, lntubation permits I.P.P.V. analogous to that of the child after prolonged with P.E.E.P. and adequate suctioning to be per- cardiac by-pass surgery, who in the immediate formed. The presence of a small air leak around post-operative period has no clinical (except the tube prevents post-extubation complications. weight gain) or haemodynamic evidence (CVP, After global ischaemia, cerebral shunts or LA or CWP all normal) of excessive third space steals from stages of autoregulation cannot be fluid but who will later have a diuresis of several determined. Therefore, the Pacoz is arbitrarily litres without hypovolaemia! maintained at 4 kPa (30 mm Hg). This is a com- In the hr this process can easily be recog- promise between excessive hypocapnia with nized by falling oxygen tensions, with a "wet" possible cerebral vasoconstriction, and normo- lung on chest X-ray and it is treated with intermit- or hyper-capnia with excessive cerebral blood tent positive pressure ventilation (I.P.P.V.) using flow and raised intracranial pressure. Neverthe- increasing levels of both positive end-expiratory less, if an abrupt rise of intracranial pressure oc- pressure (P.E.E.P.) and inspired oxygen concen- curs, hyperventilation can be increased tem- trations. The use of I.P.P.V. and P.E.E.P. are porarily to deal with such emergencies. well recognized causes of increased production The inspired oxygen concentration is main- of anti-diuretic hormone and fluid retention. tained at 70 per cent or more and the risk of In the brain, this process of capillary leak and pulmonary toxicity is accepted. A high blood rising intracranial pressure is not recognizable in oxygen level theoretically favours oxygen diffu- early stages without the use of a Richmond sub- sion through the oedematous peri-capillary areas arachnoid screw or intra-ventricular drain ! When of the brain. P. E. E. P. of 0.67-1.33 kPa (5-10 mm it is recognized clinically the process may be too l-Ig) 37 provides adequate oxygenation in most advanced for emergency therapy. cases and avoids a possible rise of cerebral ven- In our experience this potential problem is best ous pressure (I.C.P.) and pulmonary micro- managed by prevention. Initial treatment con- atelectasis. sists of the immediate administration of furo- semide 1.0 mg/kg intravenously, repeated until Hyperpyrexia (P) adequate diuresis occurs. Later, fluid restriction Fever commonly follows near-drowning 2.3s to half-normal maintenance is instituted and con- and has exceptionally deleterious effects on a tinued until the presence of clinical or laboratory damaged brain. It is believed, on circumstantial evidence of dehydration is obtained. evidence, that damaged neurones which are not functioning may recover if oxygen requirements Hyperventilation (Y) are reduced and increases in intracranial pressure An estimated 10 percent of submersion victims at-e prevented. 39'4~ It is argued that if prophylac- show "dry" drowning. 36 The remainder aspirate tic elective hypothermia gives the brain excellent varying quantities of waler. Fresh-water aspira- protection, and simultaneous immersion hypo- tion is less deleterious to the lung than salt water, thermia gives good protection, then delayed but damage always results, especially when food therapeutic hypothermia may give some protec- or foreign material is present. Lung compliance tion if instituted immediately. Therefore, immer- falls and respiration increases. The unconscious sion hypothermia should be maintained and if the patient with brain damage also increases his res- unconscious patient is normothermic, then mod- piratory rate to as high as 60-80/minute, produc- erate hypothermia should be induced within two ing severe respiratory alkalosis. This stimulation hours of injury. 4~ Hypothermia reduces the arises fi'om an acidic cerebral spinal fluid secon- body's total oxygen requirements in a linear dary to brain damage and adds to the direct pul- manner 29 so that at 30~ C both total body and monary change. This process causes greatly in- cerebral oxygen requirements are ,-educed by creased work of breathing, increased oxygen half. Cerebral blood flow also falls and intracra- 262 CANADIAN ANAESTHETISTS' SOCIETY JOURNAL nial pressure falls 5.5 per cent/1 ~ C at 25~ C. 4z for 24 hours observation TM to detect subtle Cerebral metabolism is only 30 percent of normal cerebral or pulmonary changes at an early stage at 25~ C 4~ but there is a high incidence of spon- and to prevent delayed death. 22.ss,49,sl "Coning" taneous ventricular fibrillation. is unpredictable and can occur with little warning Using cooling blankets and relaxants, body and dire results. All immersion victims who are temperature should be rapidly reduced and main- unconscious should have the full therapy started tained at 30 - I ~ C. This level achieves a satisfac- and should be transferred to the intensive care tory reduction in both cerebral oxygen require- unit immediately. Fortunately for the patient ments and intracranial pressure and is above the reaching tertiary care facilities, there is usually a upper level for spontaneous ventricular fibrilla- one to two hour lag before intracranial pressure tion (approximately 28~ C). rises precipitously. After eight hours delay in Chlorpromazine 5-25 mg q8h is given in- treatment, little benefit can be expected. tramuscularly to promote vasodilation, to pre- Full treatment should be continued for 48 hours vent shivering and to potentiate other drugs. before discontinuing relaxants and barbiturate therapy. Fortunately recovery is slow and if the Hyper-excitabilily (E) intracranial pressure rises then therapy must be Acute brain damage is frequently associated re-instituted for another 48 hours, as cerebral with grunting, athetoid movements and "strain- compliance must be low. One of our cases in 1977 ing". Such movements are prone to raise intra- was fully awake after 48 hours of full treatment cranial pressure. Less commonly, grand mal sei- only to relapse and die a rapid neurological death. zures occur which are deleterious both directly Transient small rises in intracranial pressure can and indirectly, Barbiturates have been used re- be treated with mannitol 1.0 gm/kg intravenously cently in large doses to protect the brain. 44.4s,~6 or increased ventilation. Surgical decompression Pentobarbitone has a shorter duration of action has not been necessary but has been very suc~ than phenobarbitone and is not dependent on cessful in our treatment of Reye's Syndrome. renal excretion. The summation of beneficial The only indication for discontinuing treatment metabolic effects of barbiturate treatment when at any time is the presence of "'cerebral death" combined with hypothermia has been reported 4~ (determined by flat electroencephalogram, "A" and refuted. 47 Steroids are ineffective in the waves, difficulty in maintaining temperature and treatment of the pulmonary lesion in fresh-water circulation, etc.). drowning48 but are known to prevent a rise in Full-time nursing and physician coverage is es- intracranial pressure *~'a9 and improve cerebral sential. Monitoring requirements are obviously compliance. extensive but available in any Intensive Care Currently recommended doses are: pentobar- Unit and will not be itemized. We have only lira- bitone 2-3 mg/kg/qlh intravenously (maintain ited experience so far with a Richmond Screw in blood level between 2.5 and 4.0 mg/100 ml), s~ near-drowning cases but have found its use in Methylprednisoline 1.0 mg/kg/day in 4 doses al similar situations invaluable. A subarachnoid (start treatment within 6 hours). screw (Bolt?) or an intraventricular drain for con- t~nuous monitoring of intraeranial pressure is Hyper-rigidity (R) mandatory. We have had no experience with con- Decerebrate rigidity is an obvious cause of se- tinuous E.E.G monitoring, which may be helpful. vere rise in intracranial pressure. More subtle causes include simple nursing procedures such as DISCUSSION lowering the head, or tracheal suctioning through the nasotracheal tube which may precipitate a This is a small series of cases containing a large rise in intracranial pressure of 30 minutes" dura- number of uncontrollable variables. After the tion! For these reasons, as well as the need to 1970-1974 period we began to introduce more control ventilation, complete muscle paralysis is aggressive measures in stages and therapy is still maintained continuously. For this purpose we evolving. In 1978, "HYPER" therapy was insti- use d-tubocurare 0.5-1.0 mg/kg/hr intravenously tuted on all unconscious drowned patients or p.r.n, to prevent movement or pancuronium regardless of possible benefit. It is believed that 0.1 mg/kg/hr intravenously or p.r.n, to prevent even in hopeless cases, a 24-hour period of ob- movement. servation and therapy allows both the physicians On arrival at hospital all submersion victims and parents to make adjustments and judge- who are conscious and alert should be admitted ments. In "lighter" levels of consciousness, full CONN, etal.: NEAR-DROWNING 263

TABLE II RESULTS (HYPER R)

Number Deaths CNS damage

1975 2 0 0 1976 I 0 1 1977 7 4* 0 TOTAL (1975-1977) I0 4 (40~) ] (10~) *One case, recovered completely and relapsed; One case, treated after 4 hours; One case, treated after 3 hours; One case, > 30 minutes submerges. therapy is also indicated because spontaneous The Hospital for Sick Children, Toronto, re- complete recovery is quite unpredictable. This vealed a mortality of 3.3 per cent and permanent may be considered therapeutic '~overkill ~ but a brain damage in 30 per cent. At the same time, decision to forgo full treatment without reliable excellent results were reported, using an aggres- guidelines can lead to tragedy. There were no sive treatment of closed-head injuries, following major complications of "HYPER" therapy ex- cold-water water drownings and in the use of cept for one child who made a complete recovery profound hypothermia in cardiovascular surgery. after 48 hours' treatment only to relapse abruptly In 1975 some specific therapeutic measures were and die a brain death. Other complications such instituted for near-drowning victims and have as pulmonary oedema, haemolysis, anuria and subsequently been revised and extended. pneumonia were infrequent but will not be dis- The goal of treatment is to prevent a significant cussed as the brain remains the weakest link in rise in intracranial pressure while maintaining op- recovery from asphyxia! The risk that this ag- timal conditions for cerebral recovery. Therefore gressive treatment would leave many more pa- therapy is instituted immediately and continued tients in a permanent vegetative state appears to as long as necessary. Treatment is often facili- be groundless (Table II). All submersion victims tated by the presence of immersion hypothe,'mia. should be admitted 24 hours for study and obser- This regimen includes: (a) severe restriction of vation to prevent "secondary" drowning. hydration; (b) controlled ventilation to obtain high oxygen blood levels and mild hypocarbia CONCLUSIONS (Paco2 4 kPa [30 mm Hg]); (c) moderate hypothermia (30 ~ C); (d) control of hyper- Within obvious but uncertain limits, the excitability by large doses of barbiturate and therapeutic regimen recommended for near- steriods and (e) control of hyper-rigidity by relax- drowning in "cold" fresh-water victims should ants. Monitoring of all body systems is necessary improve their prognosis. Despite limited cases, a and the continuous monitoring of intracranial large number of uncontrollable variables, our re- pressure is mandatory. sults would appear to support this concept. Final The results during evolution of treatment proof awaits a larger series and the results (1975-1977) in 26 cases showed 15.4 per cent of animal experiments. Application of these mortality and brain damage in 7.7 per cent. In this "HYPER" principles can be justified but is less limited series, with numerous variables, these successful in brain injury from other causes, e.g., results suggest that this therapeutic trial should warm-water drowning, post-cardiac arrest, be continued. Reye's Syndrome, etc. Meanwhile, future near- In conclusion, statistical proof to justify these drowning victims must be treated as they arrive, measures is not yet available clinically or experi- and the principles of cerebral salvage must re- mentally. However, the magnitude of the prob- ceive prime consideration. lem, when associated with the minimal morbidity of treatment, justifies continuation of the pro- SUMMARY gram and possible extension to other types of brain injury. Near-drowning victims are admitted to tertiary care facilities within a few hours of submersion. RI~SUMf~ Following initial improvements, many patients undergo progressive cerebral deterioration. A re- Les survivants de noyade sont hospitalis~s trospective study (1970-1974) of 30 patients at darts des unit~s de soins tertiaires quelques 264 CANADIAN ANAESTHETISTS' SOCIETY JOURNAL heures apr~s I'accident. Apr~s une am61ioration REFERENCES initiale, beaucoup deces patients manifestent une 1, THE SHORTER OXFORD ENGLISH DICTIONARY, 3rd d6t&ioration c6r6brale progressive. Une analyse ed. 1964, Oxford University Press, Amen House, r6trospective portant sur trente patients hos- London E.C. 4. pitalis6s h I'H6pital Sick Children de Toronto 2. GIAMMONA,S.T. Drowning: pathophysiology and entre 1970 et 1974, a mis en 6vidence une mor- management. Current Problems in Paediatrics, talitd de 3.3 pour cent et des dommages c6rdbraux /(7): 1-33 (1971). 3. CLARKE, E.B. & NIGGEMANN, E.H. Near drown- permanents chez 30 pour cent des patients. Au ing. Heart and Lung4(6): 946-955 (1975). conrs de la m~me pdriode, I'on rapportait d'excel- 4. SCHUMAN, S.H., ROWE, J.R., GLAZIER, I'{.M., lents r6sultats avec ]'emploi d'une th6rapie ag- REDDING, J.S- The iceberg phenomenon of near ressive dans les cas de traumatismes crfiniens drowning. Society of Crit. Care Med. 4(2): 127 (I976). fermds, chez [es survivants de noyade en eau 5. ERIKSSON, H.F., GERDMAN, P., & THORSON, J. froide et en chirurgie cardiovasculaire sous Sequelae of accidental near-drowning in child- hypothermie profonde. En 1975, un protocole de hood, Scand. J. Soc. Med. / : 3 (1971). traitement ~tait &abli pour les survivants de 6. PEARN, J. Neurological and pscyhometric studies noyade, protocole qui a depuis 6t6 r6vis6 et in children surviving fresh-water immersion accidents. Lancet, January I (1977). modifi6 au besoin. 7. SEGARRA, F. & REDDING, R.A. Modern concepts L'objectif du traitement est de pr6venir une about drowning. C.M.A. I10: 1057-1062(1974). augmentation de la pression intra-cr~.nienne tout 8. PLUECKHAHN, V.D. Death by Drowning? Geelong en maintenant des conditions optimales pour la 1959-1974. Med. Journal of Australia 2:904-906 (I975). r6cup6ration c6r6brale. Ceci implique: (a) une 9. CRAIG, A.B., CRANES, S.A,, KWUDSON, R.D., & hydratation r6duite au minimum; (b) une ventila- MODELL, J.H. Drowning's deadly toll - can more tion contr616e pour obtenir une Po2 dlev6e avec be saved? The Physician and Sports Medicine hypercapn6e moddr6e (Paco2 4 kPa (30 mm Hg)); Magazine, pp. 28-41 (August 1977). (c) une hypothermie moddrde (30~ C); (d) le con- 10. PETERSEN, B. Morbidity of childhood near- drowning. Pediatrics 59: 364-370(1977). trfle de l'hyper-excitabilit6 cdr6brale au moyen I I. CONN, A.W. The role of hypothermia in neat'- de doses elev6es de barbituriques et de st6ro'ides; drowning. Proceedings of "Cold Water Sym- (e) et enfin le contr61e des 6tats d'hyperrigidit6 au posium", Royal LifeSaving Society of Canada, pp. moyen de curarisants. Une monitoring efficace 33-35 (May 8, 1976). 12. MODELL, J.H. The pathophysiology and treatment de toutes les grandes functions, y compris le of drowning. Chas. C. Thomas, Springfield, Illinois monitoring constant de la pression intra- (1971). crfinienne est essentiel. 13. DOMINGUEZDE VILLOTA, E., BARAT, G., PERAL, Nuns avons observ6 une incidence de 7.7 P., JUFFIE, A., FERNANDEZ, J.M., & AVELLO, F. pour cent de dommages c6r6braux permanents Recovery from profound hypothermia with cardiac arrest after immersion. Brit. Med. J. 4:394 (1973). avec une mortalit6 fi 15.4 pour cent chez 26 pa- 14. KVITTINGEN, T.D. & NAESS, A. Recovery from tients ainsi trait6s entre 1975 et 1977. MHme si drowning in fresh water. Brit. Med. g. 1:1315 notre s6rie est petite et pr6sente de nombreuses (1963). variables, de tels r6sultats justifiem, croyons- 15. HUNT, P.K. Effect and treatment of diving reflex. nous, la poursuite de I'dvaluation de cette ap- C.M.A. Ill: 1330-1331 (1974). 16. SIEBKE, H., ROD, T., BREIVIK, H., & LIND, B. proache thdrapeutique. Survival after 40 minutes submersion without En conclusion, m6me si la preuve statistique de cerebral sequelae. Lancet I, 1275-1277 (1975). la validit6 clinique et exp6rimentale de ces me- 17. PEARN, J., NIXON, J., & WILKEV, I. Freshwater sures n'est pas encore disponible, le maintien drowning and near-drowning accidents involving children: a 5-year total population study. Med. J. of d'un tel protocole et peut-6tre son application h Australia 2: 942-946 (1976). d'autres cas de traumatismes cdrdbraux, nous 18. STRONG, M.J. & KEATS, A.S. Induced hypother- semblent justifi6s, surtout si l'on considbre la mia following cerebral anoxia. Anesthesiology 28: gravit6 de la pathologie en cause et la morbidit6 920-923 (1967). 19. ORLOWSKI,J.P. Prognostic factorsindrowningand minime li6e au traitement lui-m6me. the post-submersion syndrome. Crit. Care Med. 6(2): 94 (1978). ACKNOWLEDGEMENTS 20. Editorial: Drowning and the diving reflex. C.M.A.J. 108: 1209(1973). The authors wish to thank Dr. George Montes 21. NATURAL LIFE PRESERVERS. Time Magazine (Medicine), pp. 45-46 (Aug. 22, 1977). and the other Residents, Nurses and Technicians 22. GOLDEN, F. ST C. & RIVERS, J.F. Thoughts on of the Intensive Care Unit who have worked so immediate care: the immersion incident, Anaes- hard in the interests of these patients. thesia 30:364-373 (1975). CONN, el al.: NEAR-DROWNING 265

23. KEATINGE, W. The concept of hypothermia. Pro- 37. MODELL.J .H., CALDERWOOD, HaW. , Ru,z, B.C.. ceedings of"Cold Water Symposium", The Royal DOWNS, J .B., d~. CHAPMAN, R., JR. Effects of ven- Life Saving Society Canada, Toronto (May 8, tilatory patterns on arterial oxygenation after 19761. near-drowning in sea-water. Anesthesiology 40(4): 24. MODELL, J.H. & DAVID, J.H. Electrolyte changes 376-384 (19741. in human drowning victims. Anesthesiology 30(4): 38. SEGARRA, F. (~. REDO, NO, R.A. Modern concepts 414-420 (1969). about drowning, C.M.A. 110: 1057-1062 (19741. 25. KEAT, NG, W.R. Survival in cold water, Blackwell 39 F, SK, et aL Influence of duration of circulatory a,- Scientific Publications, p. 5 (19691. rest at 20~ C on cerebral changes, Anaesthesia and 26. CONN, A.W. & MUSTARD, W.T. Hypothermia for lmensive Care 4:126 (1976). cardiac surgery in paediatric surgery. Chicago 40. BLACK, P.R., VAN DEVAWTER, S., &COHN, L.H. Year Book Publishers (1969). Effects of hypothermia on systemic and organ sys- 27. WAGNER, H.R. & SUBRAMAN,AN, S. Deep tem metabolism and function, J. or Surg. Research, hypothermia in infant cardiac surgery. Paediatrics 20:49-63 (1976). 61(31:479-483 (19781. 41, BATZDORF, U. The Management ofcerebI~-d edema 28. HAKA-IKSE, K., BLACKWOOD, M.J.A., & STEW- in paediatric practice. Paediatrics 58(11:78-87 ARD, D.J. Psychomotor development of infants (19761. and children after profound hypothermia during 42, ROSOMOFF, H.L. & G,LBERT, R. Brain volume and surgery for congenital heart disease. Develop. C.S.F. pressure during hypothermia. Am. J. Med. Ch[Id. Neurol. 20: 62-70(19781. Physiology 183 : 19 (1957). 29. VIRTUE, R.W. Hypothermic anaesthesia. Charles 43. MOWAT, D. Hypothermia, in clinical surgery. Eds. C. Thomas, Springfield, Ill. (19551. Rob, C. & Smith, R. Butterworth and Co. (Pub- 30. STEWARD, D.J., SLOAN, I.A., & JOHNSTON, A.E. lishers) Ltd., pp. 146-168 (1964). Anaesthetic management of infants undergoing 44. SM,TH, A.L. Barbiturate protection in ce,'eb]~d profound hypothermia for surgical correction of hypoxia. Anesthesiology 47:28;5-293 (1977). congenital heart defects. C.A.S.J. 21: 15-22 45. MARSH, M.L.. MARSHALL, L.F.. & SHAPIRO. (1974). H.M. Neurosurgical intensive care. Anesthesi- 31. REDOING, J.S., YAKAIT,S, R.W., & HERSCHEL- ology 47:149-163 ( 19771. KING, C. Fh'oblems in the management of drowning 46. SHAP,RO, H.M. Anaesthesia, intensive care and victims. Maryland State Medical Journal 19:58 the neurosurgical patient. A.S.A. Refresher Lec- O970). ture Series (1975). 32. OHLSSON, K. & BECKMAN, M. Drowning- reflec- 47. H.;/.GERDAL, M., WELSH, F.A., KEYKNAH, M., & tions based on two cases. Acta. ChiT. Scand. 128: HARP, J .R. The protective effects of a combination 327-339 ( 19641. of hypothermia and barbiturates in cerebral 33. Fallis, J.C. Accidental hypothermia (personat hypoxia. Crit. Care Med. 6(2): 110-111 (1978). communication). 48, CALDERWOOD, H.W., MODELL, J.H., & Ru,z, 34. BRISTOW, G., SM,TH, R., LEE, J., AUTY, A., & B.C. The ineffectiveness of steroid therapy for TWEED, W.A. Resuscitation from cardiopulmon- treatment of fresh-water near-drowning, Anes- ary arrest during accidental hypothermia due to the siology 43(6): 642-650 (1975). exhaustion and expost,re. C.M.A.J. 117:247-249 49. FISHMAN, R.A. Brain edema. N.E.J.M. 293: (1977). 706-711 (19751. 35. MATHEW SPENCE, Auckland, N.Z. (personal 50- MARSHALL, L.F., SHAPIRO, H.M., RAUSCHER, R., communication). & KAUFMAN, N.M. Pentobarbital therapy For 36. MODELL, JEROME. Pulmonary insults, drowning- intracranial hypertension in metabolic coma. Crit. aspiration pneumonitis. A.S.A. Refiesher Course Care Med. 6(I): I-5 (19751. Lecture # 108:1-6 (October 12-16, 1974). 51. BARR, A.M. & TAYLOR, M. A case of drowning. Anaesthesia 31(5): 651-657 (19761.