Thomas M. Bozzuto, D.O.

Introduction

Hyperbaric oxygen therapy (HBOT) has been alternately called “highly effective”¹ and “a therapy in search of diseases”.² Modern use of hyperbaric oxygen in clinical medicine began in 1965 with the work of Churchill­Davidson³ and Borema.⁴ Following its initial successful use in cardiac surgery, carbon monoxide poisoning, and gas gangrene, researchers were eager to treat a variety of other conditions in hyperbaric chambers, often without much scientific rationale. Because of this, hyperbaric oxygen therapy fell into disfavor until the 1970’s when several significant events took place. Medicare convened a panel of members in the Undersea Medical Society (UMS) to help establish guidelines in 1972. The same society convened a workshop in 1975 after which Davis and Hunt edited the first clinical textbook in hyperbaric medicine.⁵ An ad hoc committee of the UMS was convened in 1976 which led to the formation of the Hyperbaric Oxygen Committee that now publishes a report of accepted medical conditions every two years. Potential indications for HBOT are rigor­ously screened for data “at least as convincing as that for any other treatment modality for that disorder”.⁶

Studies into the physiologic effects of oxygen under pressure have elucidated much more information over the past twenty years concerning the mechanism of action of HBOT. Research in the effects of HBOT at the cellular level have provided enough data that the majority of rational physicians no longer consider HBOT as magic, voodoo, or merely a waste of time and money. In emergency medicine, there are two main indications for HBOT: Carbon Monox­ide, Cyanide or Hydrogen Sulfide Poisoning, and a constel­lation of symptoms categorized as Decompression Illness.

Carbon Monoxide Poisoning

Carbon monoxide is a leading cause of death by poison­ing in the United States.^7,8 Most common sources are automobile exhaust (accidental or purposeful), faulty heat­ers, and building fires. The pathophysiology of CO poison­ing can be found in any textbook of toxicology or emergency medicine and is beyond the scope of this article. A few significant points bear mentioning. Many factors affect the actual clinical presentation, such as the inhaled CO concen­tration, duration of exposure, rate and depth of breathing, heart rate, co-morbid illnesses and most importantly, the time between discovery of the patient after exposure and arrival at a hyperbaric chamber. Until recently, many text­

books listed severity of symptoms as they related to arterial carboxyhemoglobin levels. Recent data suggest that COHgb levels are merely an indication of exposure, and that arterial pH is a much more sensitive indication of severity of exposure.9, ¹⁰ Hy­perbaric oxygen causesCOdissocia­tion from hemoglo­bin to occur at a rate greaterthanachiev­able by breathing pure oxygen at sea level (Table 1).^11-13

There is evidence that CO is directly toxic to brain via a mechanism that is not related to hypoxia.¹⁴ The initial effect of CO on cerebral blood flow (CBF) is to induce a very large increase in flow, rather than to cause a hypotensive hypoperfusion.¹⁵ This increase in CBF is sufficient to main­tain oxygen delivery to the brain.¹⁶ The difference between CO exposure and dilutionally hypoxic rabbits was that brain function (measured as a cortical somatosensory re­sponse) was greatly inhibited in the former, but preserved in the latter.

Thom^17,18 has demonstrated a marked benefit with HBOT in CO poisoning that stands independent of the rhetoric in clinical literature regarding how to treat CO poisoned patients.^19,20 His studies show that CO causes a cascade of microvascular endothelial injury and that binding and sub­sequent activation of leukocytes is a central component of brain injury triggered by CO. The specific process by which hyperbaric oxygen inhibits this leukocyte-endothelial ad­hesion is related to a class of glycoproteins on the leukocyte surface called b₂-integrins. This oxygen related effect is quite discreet. Another interesting finding which supports this mechanism is that animals pre-treated before with HBOT prior to CO exposure received the same benefit as those treated after exposure to CO.

Clinical presentation varies greatly in patients poisoned with CO. The most common physical findings include tachycardia and tachypnea. There may be associated mild hypotension.²¹ Headache and nausea or vomiting are com­mon. Be suspicious of an entire family who goes to bed feeling well and all wake up sick at the same time. Inquire if there are pets in the house behaving strangely. The cherry-red color described in textbooks is rarely seen.

COHgb levels above 15% increase the risk for myocar­dial infarction and a 9% COHgb level lowers the ventricular fibrillation threshold.⁷ The reported incidence of delayed neuropsychiatric syndrome (cognitive dysfunction, per-

Site	# of patients	Loss of consciousness	Deaths		Neurologic Sequelae
Copenhagen (100%O2 at sea level)	79	76	23	(30%)	11 (14.5%)
Seattle (HBOT)	115	74	11	(14.9%)	2 (2.7%)

sonality changes, aphasia, apraxia, apathy, disorientation, hallucinations, gait disturbances, mood changes, violence, verbal aggressiveness and impulsiveness) ranges from 3 to 40%.^22-24 Of equal importance and interest is the possibility of long-term cardiac sequelae and myocardial dysfunction as a result of CO poisoning.²⁵

Methylene chloride is a component of paint thinner. It is converted to CO by the cytochrome P-450 after it enters the liver. Endogenously created CO derived from methylene chloride has a much longer half life than exogenously inhaled CO, but the symptoms are similar. CO poisoning should be suspected in persons who exhibit toxicity after using paint thinning products in a closed environment. Another increasingly more common source of CO is pro­pane-powered motors such as those found in warehouse fork lifts and industrial floor buffers. Ideally, these propane engines should burn cleanly if properly maintained. How­ever, a poorly maintained propane engine will emit nearly as much CO as an internal combustion (gasoline) engine.

Performing double-blind studies in a hyperbaric chamber is difficult. A sham treatment must include pressurization to avoid patients knowing that they were not under pressure. This requires sophisticated gas mixing systems beyond the capability of most clinical hyperbaric facilities. In addition, most hyperbaric physicians have enough respect for the clinical data available that they consider it unethical not to treat appropriate patients with HBOT. The advantages of HBOT over surface oxygen in treating acute CO poisoning can be seen by comparing clinical reports from Copenhagen and Seattle.^26,27 The results of these studies are summarized in Table 2. Except for the HBOT, treatment of patients in each study was similar.

The decision for the emergency physician is: When should I refer a patient for hyperbaric oxygen therapy? There is no concrete consensus even among hyperbaric physicians. Epidemiologic studies suggest that prognosis is poorer for patients who have underlying cardiovascular disease, are more than 60 years old, or have suffered any interval of unconsciousness due to CO.21,27

We use the following criteria to determine a patient’s candidacy for HBOT:

·    history of an interval of unconsciousness;

·      an objective neurologic deficit or altered mental status;

·      ischemic EKG changes or chest pain in a patient ex­posed to CO;

·    pregnant patient with COHgb > 15% (fetal hemoglobin binds CO more ‘tightly’ than maternal hemoglobin);

·    recurrent symptoms within three weeks of original treat­ment;

·    patient with COHgb > 25-30%;

·      symptoms in less severe poisoning that do not resolve after four hours of 100% O₂ via non-rebreather mask in the emergency department.

Cyanide Poisoning

Carbon monoxide and cyanide poisoning frequently oc­cur simultaneously in victims of smoke inhalation. ^28-33 In combination, these two agents exhibit synergistic toxic­ity.34, ³⁵ The difficulty in cyanide detection is that blood levels are more difficult to obtain in the emergency depart­ment, and levels are ordered with much less frequency than are COHgb levels. While CO binds reversibly to the ferrous (Fe²⁺) iron in the cytochrome oxidase a3 system, cyanide irreversibly binds to the ferric (Fe³⁺) iron at the cytochrome a oxidase level. Therefore, while HBOT is a primary therapy utilized in the dissociation of CO from hemoglobin, it is supportive to the use of a cyanide antidote kit in situations of cyanide poisoning. The rationale for utilizing HBOT in cyanide poisoning is several fold: it mitigates the hypoxia induced by cyanide by supersaturating the plasma with oxygen until the sodium nitrite in the antidote kit converts cyanhemoglobin to methemoglobin and it lessens the hy­poxic effect of methemoglobin itself.^36-43 One must be cautious in administering HBOT simultaneously with a cyanide antidote kit because the methemoglobin level may be directly lowered by hyperoxia (at least 4 ATA), possibly reducing the efficacy of antidotal therapy.⁴⁴ In reality, 100% oxygen is never administered at pressures greater than 3 ATA because of the significant increase in CNS oxygen toxicity. HBOT is recommended as an adjunct to the treatment of combined CO poisoning complicated by cyanide poisoning.⁶

Hydrogen Sulfide Poisoning

Hydrogen sulfide (H₂S) is a highly toxic, inflammable, colorless gas, readily recognized by its characteristic “rot­

Treatment

There is no universally accepted treatment protocol for the treatment of CO, cyanide or hydrogen sulfide poisoning with HBOT. Because the T_½ of CO at 3 ATA is 23 minutes, a common treatment protocol is to subject the patient to two or three half-lives of CO at 3 ATA on 100% oxygen with five minute air breaks in between the oxygen-breathing periods. These “air breaks” have been found to lessen the likelihood of oxygen toxicity. Patients are instructed to follow-up within 24 hours for reassessment. Retreatment is based on recurrence of significant symptoms (neurologic deterioration, headaches, confusion, nausea, irritability or personality change) if any. A syndrome of chronic carbon monoxide poisoning has been recognized in which patients are exposed to long-term low levels of CO. These patients are symptomatic with relatively low (normal) levels of COHgb because CO is lipophilic and tissue levels will be elevated. On occasion, repeat ABG’s six hours after initial HBOT will reveal COHgb elevations again without addi­tional exposure, suggesting long term exposure. These patients will require repeated HBOT.

Decompression Illness

Decompression illness (or DCI) is a general term used to describe a broad spectrum of signs and symptoms of inert gas (N₂) problems or dysbaric injuries related to SCUBA diving. An arterial gas embolism (AGE) is characterized by gas bubbles in the arterial system generally caused by air passing through the walls of the alveoli into the blood­stream. AGE can result after breathing compressed gas followed by voluntary breath-holding (such as during a rapid ascent); or it can result from a pathologic condition which traps air in the lungs while ascending to the surface. Only a 4-ft ascent during breath holding is sufficient to bring about enough of a pressure increase to cause AGE. Symptoms of AGE are usually immediate in onset and generally involve changes in level of consciousness, pa­ralysis or other cerebral symptoms. Decompression sick­ness (DCS) is a syndrome caused by bubbles of inert gas (N₂) formed in the tissues and blood stream after SCUBA diving. DCS usually results from a deep dive or prolonged exposure to breathing compressed gas at depths greater than 20 feet/6.1 meters. Symptoms may be confined to the musculoskeletal system and consist of joint or muscle pain, or may involve the central nervous system with symptoms of numbness, tingling and other complaints. Type I DCS refers to pain involving the joints or muscles, or skin bends, or fatigue without other symptoms. Type II DCS includes neurological and cardiorespiratory symptoms. It bears mentioning that breath-holding diving (where no com­pressed gas is involved) does not expose a diver to DCI risk.

Florida has the dubious honor of having both the highest annual number of cases of DCI (42.9%) and the highest number of fatalities from diving accidents in the United

States (21.2%).⁴⁶ For this reason, it behooves emergency physicians in Florida to become familiar with the symp­toms, signs and treatment of divers with DCI. Divers are generally divided into two groups: those who have very little knowledge of recognition of DCI, and those who think they have a great deal of knowledge (sometimes inappro­priate) about the risk profiles and likelihood of DCI. A common fallacy includes not being able to get DCI if diving within the tables (there is a 1-5% inherent risk of DCI when following tables, and many other factors contribute to the likelihood of a DCI incident than merely the dive profile).

Factors that contribute to the likelihood of DCI are listed in Table 3. A good rule of thumb is that a diver who exhibits symptoms after an ex­posure to compressed air breathing has DCI until proven otherwise. Certainly, based on the medical history of the diver, and other factors, every under-water oc­currence may not be DCI (See Tables 4 and 5). The differentiation of Type I, Type II DCS, or AGE is not so im­portant in the emer­gency department as is therecognitionofaDCI that requires recom­pression therapy.

DCI cannot occur unlessthereissufficient volume of inert gas dis­solved in the tissues, so that when the ambient pressure maintaining it in solution is suffi­ciently reduced, the gas leaves solution and  forms bubbles. There are both mechanical effects and physiologic effects of bubble formation. Among the me­chanical (bubble-related) effects is assumed to be the occlu­sion of arterioles. Most of the evidence for the presence of in vivo gas bubbles in the blood stream following DCI is based on their detection by Doppler flowmeters. Many limitations have been attributed to this technique.⁴⁷ How­ever, the overwhelming evidence is that during decompres­sion, gas bubbles are first detected on the venous side of the circulation and that arterial bubbles are rarely observed except in cases of severe DCI.^48-50

The ability of bubbles to distort tissue and obstruct blood flow would be injurious even if the fluid surrounding the gas phase were inert. The blood, however, is a highly reactive fluid, and the effects of a bubble are amplified by the activation of systems usually quiescent during normal vascular flow. The interface between the gas phase and blood is a physical-chemical discontinuity, and its mainte­nance is associated with enormous electrochemical forces. These forces cause the denaturation of proteins, accumula­tion of globules of free fat, and expose active sites on enzymes in the blood that activate coagulation and comple­ment systems. All these mechanisms extend any mechani­cal blockage of the circulation with progressive clotting and further damage tissue by a reduction of blood flow, the formation of edema, toxic oxygen species (free radicals) and by the attraction of leukocytes to the area. The physi­ologic role of HBOT in the amelioration of these effects of DCS are similar to the effects shown in the prevention of neutrophil-endothilial adhesions by the inhibition of b₂­integrins in CO poisoning.

Type I DCS is characterized by aching pain in a limb that occurs 20 minutes to several hours after surfacing from a compressed air dive. There are usually no physical signs associated with Type I DCS. Initially the deep, aching pain is characterized as dull and vaguely localized. When the pain is well localized, it is often described as being adjacent to rather than within the joint. In short dives on compressed air, the upper limbs are affected two to three times more often than the lower limbs, with the shoulder being the most common site.⁵¹ The sternoclavicular joint has never been reported to be involved in a DCI. When more than one site is involved, they are not usually symmetrically distributed. Fatigue is a frequent sequel to exposure to pressure, even if the workload is light. It is usually transient, therefore often ignored. Occasionally, the exhaustion is sufficient to pro­voke comment, and then is often a harbinger or accompa­niment of more serious signs of DCS.⁵²

DCS Type II may occur alone or in combination with musculoskeletal pain. One series estimated that about 30% of Type II DCS was accompanied by pain.⁵³ There are three manifestations of Type II DCS. Pulmonary DCS (chokes) is

relatively rare and generally occurs with rapid emergency ascents. The onset is usually heralded by a sensation of substernal discomfort that commences within minutes after reaching the surface. Pneumothorax, pneumomediastinum or subcutaneous emphysema may be found. It may be accompanied by a cough and deep inspiration may provoke paroxysms of coughing. The breathing pattern becomes shallow and rapid, cyanosis develops, as well as signs of right heart failure. At this stage the patient is in cardiovas­cular shock and immediate recompression is required as well as fluid resuscitation and pressors. The incidence of Neurological Type II DCS is common in sport divers. It is actually more common (62.7%) than the milder DCS I (25.3%) or the more severe AGE (12.0%).⁴⁶ The central nervous system is the particular target organ of DCS. Brain dysfunction is often manifested by confusion, drowsiness, fatigue or indifference. The pain and paraplegia of spinal cord decompression occurs earlier and more often attracts the most attention. The anatomy of the spinal cord, which protects it so well from minor trauma, renders it uniquely susceptible to DCS. The mechanical structure of the verte­bral column shields the cord from most injury. The redun­dant collateral arterial supply assures that the cord will be nourished directly from the aorta even when less favored tissues are allowed to become ischemic. The venous drain­age of the cord, slowed and made pendular by respiratory pressure changes, makes it uniquely vulnerable to venous infarction. This vulnerability causes a unique spinal cord DCS different from any other neurological syndrome. Spi­nal DCS presents a clinical picture of diffuse multilevel cord disease. Vestibular DCS is relatively common and presents as dizziness, nausea, nystagmus vomiting and, occasionally, hearing loss and tinnitus. The incidence of these symptoms varies from 13 to 72 percent of Type II DCS. Differentiating this syndrome from otic barotrauma is important. The symptoms of vestibular DCS occur during ascent, those of otic barotrauma occur during descent.

Rapid ascent (without breath holding) may result in inert gas coming out of solution (plasma) in a rapid fashion and traverse the arterial supply into the cerebral circulation causing an arterial gas embolism (AGE). A sudden change in sensorium (during the ascent or shortly after surfacing) is the most common symptom and ranges from disorienta­tion to coma. Focal neurological deficits such as hemiplegia or monoplegia may occur depending on the location of the lesions. Liebermaster’s sign (presence of pallor or mottling of the tongue) may be present.

Skin bends is a benign condition which commences with itching, which may be intense. The pruritic areas, usually limited to the trunk, are first reddened by vasodilation in the dermis, then vascular stasis results in a characteristic mot­tling of the skin. Although recompression will promptly relieve the itching, it will resolve over a period of days if left alone. Treatment is more for comfort than to avoid possible complications. The main risk of non-treatment is the possi­bility of missing more severe symptoms of DCI that would require HBOT.

There are other “dysbaric injuries” that, although they occur while under pressure, do not require HBOT. These include pneumomediastinum, barosinusitis or barotitis and caloric-induced vertigo (which must be differentiated from vestibular DCS).

Treatment

The emergency department treatment in most cases of DCS or AGE is fairly straightforward. One hundred per­cent oxygen should be administered by mask or endotra­cheal tube. The reasoning is that one is not only trying to deliver oxygen to relatively ischemic, hypoxic tissues, but also trying to eliminate as much inert gas as possible. Patients with DCI can become severely dehydrated. Immer­sion in water that is colder than body temperature inhibits antidiuretic hormone and results in a diuresis that can often be significant. Intravenous fluid hydration is preferred even if the patient is totally conscious because providing enough oral hydration at the same time as maintaining adequate oxygenation is difficult. Patients may need up to 8 liters of IV fluids before urine output is obtained. It warrants men­tioning that the symptoms of DCI may resolve on supple­mental oxygen during transport or during emergency department treatment. This does not eliminate the need for hyperbaric therapy. There is almost always a recurrence of symptoms, and the delay to HBO increases the likelihood of residual symptoms after treatment. Diagnosis of DCI is made by history and physical examination (Table 6). No laboratory findings will assist in the diagnosis. Chest x-rays are helpful to diagnose a pneumothorax that will need decompression prior to HBOT. Waiting for laboratory results, CT scans, or other diagnostic studies will serve to delay HBOT without adding any additional value to the diagnosis.

Despite the basic requirements of oxygen and IV fluids being considered elementary and essential in the treatment of DCI, statistics show that only 33% of DCI cases receive oxygen before transport to HBOT, and only 25% receive IV fluids.⁴⁶ In cases of neurologic involvement, IV corticoster­oids have been suggested.⁵⁴

Folklore states that a Trendelenberg, left lateral decubitus position should be utilized for patients with AGE. The theory for lateral decubitus is to trap any bubbles in the right atrium thereby lessening circulating air, and to protect the airway in case of vomiting. This may have some merit. The Trendelenberg rationale takes advantage of gravity and bubble buoyancy to minimize cerebral embolization. Buoy­ancy of bubbles has no effect on arterial distribution⁵⁵, and prolongedhead-downpositioncanpotentiatecerebraledema in the injured brain.⁵⁶ For these reasons, Trendelenberg is not recommended. If the nearest hyperbaric facility is too far for surface transportation, air evacuation must be used. It is important that the patient not be exposed to decreased barometric pressure at altitude, so helicopter transportation is frequently utilized, and flight altitude must be kept to no greater than 800-1000 feet above ground level (AGL).

There are several standard treatment tables utilized in the treatment of DCI. The most common are the United States Navy Treatment Tables. The table chosen depends on the severity of symptoms, and the length of treatment also depends on the rapidity with which symptoms resolve

under pressure. If there are residual symptoms after the initial recompression treatment, the patient is treated daily until symptoms are totally resolved, or there are three consecutive treatments without objective improvement. Likelihood of total resolution is multifactorial, but depends on symptom severity and length of time from symptom onset until commencement of recompression therapy. De­spite appropriate recompression therapy, many divers will exhibit residual symptoms. Prompt treatment with HBOT is directly correlated with success of treatment as measured by its ability to reduce or totally resolve symptoms (see Figure 1). Approximately 18% of divers treated for DCI will have some permanent sequelae after treatment is com­pleted. Divers who are treated with HBOT require observa­tion for the following 24 hours to watch for recurrence of symptoms. They can be discharged from the hyperbaric department after 30 minutes if there is a reliable adult at home to observe them. We recommend that they remain within 30 minutes travel distance of the chamber in case there is a need for repeat treatment. Patients are to return for follow-up and repeat examination 24 hours after their initial treatment. Recommendations for returning to work or fly­ing after diving depend on the dive profile, the degree of resolution of symptoms after treatment, and the treatment table used. The Divers Alert Network maintains a 24-hour emergency hotline to assist medical personnel in diagnosis and treatment options and referral to the nearest hyperbaric facility. Their number is 919-684-8111.

Side-Effects

Side-effects during HBOT are rare. The most common is barotitis, which occurs more frequently in patients being treated for conditions other than those related to diving. Barosinusitis may also occur. These side effects are easily treated with decongestants and analgesics. Oxygen toxicity seizures are the most serious consequences of HBOT. There are many factors which influence a patient’s individual

Current research into other uses of HBOT include a multi-center study on HBOT and thrombolytics in the treatment of acute myocardial infarction (HOT Study), and another on the utility of HBOT in early stroke, traumatic brain injury, and spinal cord trauma.

There are approximately 28 hyperbaric chambers in Florida. Baptist Medical Center maintains the only treat­ment facility in Northeast Florida, with other nearby cham­bers being Tallahassee, Gainesville and Orlando.

REFERENCES

1. Neubauer RA, Walker M. Hyperbaric Oxygen Therapy. Avery Publishing Group, NY, 1998:ix.

2. Gabb G, Robin ED. Hyperbaric Oxygen: a therapy in search of diseases. Chest. 1987; 92:1074-82.

3. Churchill-Davidson I, Sanger C, Thomlinson. High-pressure oxygen and radiotherapy. Lancet. 1955;1:1091-95.

4. Boerema I, Kroll JA, Meijne NG, Lokin E, et al. High atmospheric pressure as an aid to cardiac surgery. Arch Chir Neerl. 1956;8:193-211.

5. Davis JC, Hunt TK. Hyperbaric Oxygen Therapy. Undersea Medical Society, Bethesda, 1977.

6. Hyperbaric Oxygen Committee. Hyperbaric Oxygen Therapy: A Committee Report. Undersea and Hyperbaric Medical Society, Bethesda, 1996.

7. Dolan MC. Carbon Monoxide Poisoning. Can Med Assoc J. 1985;133:392.

8. U.S. Public Health Service: Vital Statistics of the United States, Washington, DC, Government Printing Office, 1976.

9. Myers RAM. Do arterial blood gasses have value in prognosis and treatment decisions in carbon monoxide poisoning? Crit Care Med. 1989:1720:139-142.

10. Myers RAM, Messier LD, Jones DW, Cowley RA. New direction in the research and treatment decisions in carbon monoxide poisoning. Am J Emerg Med. 1983;2:226.

11. End E, Long CW. Oxygen under pressure in carbon monoxide poisoning. J Ind Hyg Toxicol. 1942;24:302-6.

12. Pace N, Strajman E, Walker EL. Acceleration of carbon monoxide elimination in man by high pressure oxygen. Science. 1950;111:652-4.

13. Britten JS, Myers RAM. Effects of hyperbaric treatment on carbon monoxide elimination in humans. Undersea Biomed Res. 1985; 12:431-8.

14. Haldane JBS. Carbon monoxide as a tissue poison. Biochem J. 1957; 21:1068-1075.

15. Meyer-Witting M, Helps SC, Gorman DF. Acute CO exposure and cerebral blood flow in rabbits. Anaesth Intens Care. 1991;19:373-7.

16. Thom SR. Antagonism of carbon monoxide-mediated brain lipid peroxidation by hyperbaric oxygen. Toxicol Appl Pharmacol. 1990; 105:340-4.

17. Thom SR. Functional inhibition of leukocyte ?₂-integrins by hyperbaric oxygen in carbon monoxide-mediated brain injury in rats. Toxicol Appl Pharmacol. 1993;123:248-256.

18. Olson KR, Seger D. Hyperbaric oxygen for carbon monoxide poisoning: Does it really work? Ann Emerg Med. 1995; 25:535-7.

19. Seger D. The science (or lack thereof) in the treatment of carbon monoxide poisoning. Am J Emerg Med. 1993;11:616-8.

20. Whorton MD. Carbon monoxide intoxication: A review of 14 patients. J Am Coll Emerg Physicians. 1976;5:505.

21. Choi IS. Delayed neurologic sequelae in carbon monoxide intoxication. J Toxicol Clin Toxicol. 1982;19:297.

22. Smith JS, Brandon S. Morbidity from acute carbon monoxide poisoning at three year follow-up. Br Med J. 1973;1:318.

23. Youngberg JT, Myers RAM, Use of hyperbaric oxygen therapy in carbon monoxide, cyanide and sulfide intoxication. Hyperbaric Oxygen Therapy: A Critical Review. Camporesi EM and Barker AC (Eds.) Undersea and Hyperbaric Medical Society, Bethesda, 1991:23-53.

24. Hadley M. Coal-gas poisoning and cardiac sequelae. Br Heart J. 1952;14:534-6.

25. Krantz T, Thisted P, Strom J, Sorrenson MB. Acute carbon monoxide poisoning. Acta Anaesthesiol Scan. 1988;32:278-282.

26. Norkool DM, Kirkpatrick JN. Treatment of acute carbon monoxide poisoning with hyperbaric oxygen: A review of 115 cases. Ann Emerg Med. 1985;14:1168-1171.

27. Min SK. A brain syndrome associated with delayed neuropsychiatric sequelae following acute carbon monoxide intoxication. Acta Psychiatr Scand. 1986;73:80-6.

28. Birky MM, Clarke FB. Inhalation of toxic products from fires. Bull NY Acad Med 1981;57:997-1013.

29. Clark CJ, Campbell D, Reid WH. Blood carboxyhemoglobin and cyanide levels in fire survivors. Lancet. 1981:1:1332-5.

30. Mohler SR. Air crash survival: Injuries and evacuation toxic hazards. Aviat Space Environ Med. 1975;46:86-8.

31. Terrill JB, Montogomery RR, Reinhardt CF. Toxic gasses from fires. Science. 1978; 200:1343-7.

32. Symington IS. Cyanide exposure in fires. Lancet. 1978;2:91-2.

33. Hart GB, Strauss MB, Lennon PA, Whitcraft DD. Treatment of smoke inhalation by hyperbaric oxygen. J Emerg Med. 1985;3:211-5.

34. Norris JC, Moore SJ, Hume AS. Synergistic lethality induced by the combination of carbon monoxide and cyanide. Toxicology. 1986;40:121­9.

35. Barillo DJ, Goode R, Rush BF Jr, Lin RL, et al. Lack of correlation between carboxyhemoglobin and cyanide in smoke inhalation injury. Curr Surg. 1986;43:421-3.

36. Ivanov KP. The effect of elevated oxygen pressure on animals poisoned with potassium cyanide. Pharmacol Toxicology. 1959;22:476-9.

37. Skene WG, Norman JN, Smith. Effect of hyperbaric oxygen in cyanide poisoning. In: Brown IW, Cox B (eds) Proceedings of the Third International Congress on Hyperbaric Medicine, Washington DC: National Academy of Sciences-National Research Council, 1966:705­710.

38. Takano T, Miyazaki Y, Nashimoto I, Kobayashi K. Effect of hyperbaric oxygen on cyanide intoxication: In situ changes in intracellular oxidation reduction. Undersea Biomed Res. 1980; 7:191-7.

39. Cope C. The importance of oxygen in the treatment of cyanide poisoning. JAMA. 1961;175:1061-4.

40. Isom GE, Way JL. Effect of oxygen on cyanide intoxication: VI. Reactivation of cyanide inhibited glucose metabolism. J Pharmacol Exp Ther. 1974;189:235-243.

41. Burrows GE, Way JL. Cyanide intoxication in sheep. Therapeutic value of oxygen or cobalt. Am J Vet Res. 1977;38:223-227.

42. Way JL, Gibbon SL, Sheehy M. The effect of oxygen on cyanide intoxication. I. Prophylactic protection. J Pharmacol Exp Ther. 1966; 153:381-5.

43. Sheehy M, Way JL. The effect of oxygen on cyanide intoxication. III. Mithridate. J Pharmacol Exp Ther. 1968; 161:163-8.

44. Goldstein GM, Doull J. Treatment of nitrite-induced methemoglobinemia
with hyperbaric oxygen. Proc Soc Exp Biol Med. 1971;138:137-9.

45. Jain KK. Textbook of Hyperbaric Medicine. Hogrefe & Huber Publishers, NY, 1990, 165.

46. Divers Alert Network. Report on Decompression Illness and Diving Fatalities: The DAN Annual Review of Recreational Scuba Diving Injuries and Fatalities Based on 1996 Data. Durham, NC. 1998.

47. Francis TJR, Dutka AJ, Hallenbeck JM. Pathophysiology of Decompression Sickness. In Bove AA, Davis JC (eds). Diving Medicine, WB Saunders Co., Philadelphia, 1990. 172.

48. Gardette B. Correlation between decompression sickness and circulating bubbles in 232 divers. Undersea Biomed Res. 1979; 6:99-107.

49. Nashimoto I, Gotoh Y. Relationship between precordial doppler ultrasound records and decompression sickness. In Shilling CW, Beckett MW (eds). Underwater Physiology VI. Bethesda. FAESB, 1978;497­501.

50. Powell MR, Johansen DC. Ultrasound monitoring and decompression sickness. In Shilling CW, Beckett MW (eds). Underwater Physiology VI. Bethesda. FAESB, 1978;503-510.

51. Francis TJR, Dutka AJ, Hallenbeck JM. Pathophysiology of Decompression Sickness. In Bove AA, Davis JC (eds). Diving Medicine, WB Saunders Co., Philadelphia, 1990. 177.

52. Kidd DJ, Elliott DH. Clinical manifestations and treatment of decompression sickness in divers. In Bennett PB, Elliott DH (eds). The Physiology and Medicine of Diving and Compressed Air Work. Baillere Tindall and Cassell, London, 1969. 464-490.

53. Hallenbeck JM, Bove AA, Elliott DH. Decompression sickness studies. In Lambertsen CJ (ed). Underwater Physiology V. Bethesda, FASEB, 1976, 273-286.

54. Bove AA. The basis for drug therapy in decompression sickness. Undersea Biomed Res.1982;9:91-111.

55. Butler BD, Laine, GA, Lieman BC, Warters D, et al. Effect of Trendelenberg position on the distribution of arterial emboli in dogs. Ann Thoracic Surg. 1988;45:198-202.

56. Dutka AJ. Therapy for dysbaric central nervous system ischemia: Adjuncts to recompression. In Bennett PB, Moon RE. Diving Accident Management. Undersea and Hyperbaric Medical Society, Bethesda, 1990, 222-234.