This page contains medical essays for physicians and other health care providers.

return to Home Page


Nomenclature of Pressure Disorders

Risk Factors for Pulmonary Barotrauma

Fatal Pulmonary Barotrauma

Patent Foramen Ovale







This section contains two essays on categorizing the diving disorders. They were published in the journal Undersea and Hyperbaric Medicine 24: 1-4, 1997. They are provided with permission of the Undersea and Hyperbaric Medical Society and Dr. Richard Moon who wrote the second essay which expresses the opposing point of view on nomenclature. Return to table of contents


Nomenclature of Pressure Disorders

Alfred A. Bove, M.D. Ph.D. Philadelphia, Pa.


Recently several physicians have proposed a new method for describing diseases related to diving. In particular, the descriptions are related to disorders resulting from decompression. The proposed nomenclature is Decompression Illness (DCI). This term is proposed to encompass disorders previously known as Decompression Sickness (DCS), and Arterial Gas Embolism (AGE). The reason for the proposed change is to simplify the description of the disorders. Francis and Smith (1) have provided a list of descriptions which include the location, the clinical symptoms, and the time-related nature of the disorder, but not the underlying pathophysiology. They suggest that the etiology of the decompression disorders is impossible to define, the two illnesses are difficult to separate from one another, and both require the same therapy. Under these circumstances, there is no need to differentiate the pathophysiologic mechanisms of the disease, and symptomatic description would be adequate to propose a plan of therapy.

In a recent workshop Dutka (2) proposed that the Term "Decompression Injury" be used to encompass the two pathophysiologic disorders described by Decompression Sickness (DCS) and Arterial Gas Embolism (AGE). This proposal includes a symptomatic description-based classification which replaces the two previously described disorders. Dutka does not suggest that the two disorders be considered the same clinical entity, but suggests that confusion over pathophysiology is avoided by purely descriptive nomenclature. In reviewing the International Classification of Diseases 9th edition (ICD-9) one finds code 993 identifying "Effects of Pressure." Dutka would use the term "Decompression Injury" for the code 993 . The category "Effects of Pressure" includes barotrauma and decompression sickness. The term "Pressure Related Illness" could be used to describe this code, but since it contains barotrauma not related to decompression, using "Decompression Illness or "Decompression Injury" would be inappropriate to describe barotrauma of descent. Air embolism is coded separately as 958.0.

Francis and Smith (1) suggest that the nomenclature delete the terms Decompression Sickness and Arterial Gas Embolism and replace both with the term Decompression Illness accompanied by a description of the symptoms and clinical manifestations. An example would be "Abrupt Paresthetic DCI" instead of type II decompression sickness. An analogous example would be "Abrupt, Painful Chest Illness" to describe a manifestation of coronary artery disease, for which the treatment could be immediate catheterization and angioplasty of the culprit coronary artery, immediate infusion of a thrombolytic agent, a sublingual nitroglycerin tablet, or possibly, emergency repair of a dissecting aortic aneurysm. In this case, the need to understand that the pain originated from an occluded coronary artery would be supplanted by an algorithm describing a course of action for acute pain in the chest.

One could develop an entire practice of medicine based on symptomatic manifestations of disease ("Evolving, Painful Gastrointestinal Illness", instead of appendicitis). A comprehensive list of symptoms would be created with a stated treatment associated with each symptom complex. This approach would allow a clerk to record symptoms, look up the treatment, and send the patient to the appropriate treatment team. Although the list of symptom complexes would greatly exceed the list of pathophysiologic disorders, these long lists could be stored in a computer, the clerk could search on key words (pain, nervousness, tingling, hot, cold, etc.) to send the patient to the appropriate therapy. This would eliminate the need for physicians and nurses in the triage process (i.e. the emergency room), and increase efficiency since the patient could go right to the treatment procedure based on the clerk's computer search of symptom complexes. Besides reduced emergency medical costs and shorter hospital stays (diagnostic workups would be greatly shortened), insurers would find reduced health care costs in their capitated population because of the added efficiency and increased mortality with consequent reduction of hospital costs. An economic analysis would show this to be a highly successful nomenclature system, and it would likely be permanently incorporated into the health care system through legislative initiatives.

But what is wrong with this picture? We in medicine have been taught from the first day of medical school to understand the nature of disease processes. This is ingrained into all physicians, and is not a random choice of training methods. This process occurred because human beings for many millennia have made efforts to understand the processes of nature. This has occurred constantly in biology, mathematics, physics, and the social sciences. Gallileo proposed that the planets revolved around the sun. This observation in astronomy reduced thousands of observations about the sun, moon and planets to a single rule from which all the previous rules could be derived. James Maxwell found three equations which described the actions of electric and magnetic fields. These three simple equations explained hundreds of earlier observations, allowed prediction of yet unobserved phenomenon in electricity and magnetism, and provided a basis for research that brought us the disciplines of electronics and radio transmission. The consequence of understanding is simplification. Lack of understanding results in highly complex schemes for categorizing observations. The power and simplicity of Maxwell's equations, Gallileo's heliocentric hypothesis, the role of DNA in genetics, Einstein's description of Relativity are a tribute to human intellectual curiosity. The descriptive nomenclature "DCI" brings an increase in complexity, not a simplification. It no longer advances understanding of pathophysiologic processes which would simplify our concept of these diseases. The descriptive process will create an ever expanding list of descriptions. One which the clerk can use to send everyone to a hyperbaric chamber who has pain, tingling, or weakness. The nomenclature suggests a fixation of knowledge about diving disorders in time (circa 1994). It does not allow the next generation of intelligent diving physicians to propose new concepts of pathophysiology which we cannot presently conceive, and implies that we now know all there is to know. Why should we express such arrogance, and disallow new, even radical theories to be proposed and tested. The clerk reading the symptom list is not going to propose these ideas, the chamber operator, told to provide a certain pressure and a certain time is not. We give up a cherished responsibility to advance understanding if we deny the understanding of pathophysiology, no matter how incomplete our current knowledge.

I propose that we embrace the pathophysiology, and teach physicians and other care givers involved with diving medicine the proper rules to classify these disorders based on understood mechanisms. These mechanism can be divided into mechanical or direct pressure effects (barotrauma in its widest sense), and indirect or dissolved gas effects. These mechanisms lead to the nomenclature for all of the barotraumatic injuries (lungs, ears, sinuses, etc.), and the consequences, i.e. Arterial Gas Embolism, pneumothorax, round window rupture, etc.; for the disorders which involve supersaturation of gases in tissues, i.e. Decompression Sickness; and for osmotic and other dissolved gas effects, i.e. High Pressure Nervous Syndrome.

In seeking a categorization scheme for decompression sickness, the classification first described by Golding et al (3) still makes sense. They described a systemic form of DCS which involves the central nervous system, the lungs, the circulation (serious, type II), and a non-systemic (peripheral) form which involved the skin, bones and joints (minor, type I). Admittedly, we are still at a loss to understand the pathophysiology of limb bends, but this should drive us to more research, not to a complex symptom based description. Even if the complex symptom scheme works well (outcome assessment is still an issue), we cannot be satisfied without an understanding of the pathologic mechanisms; otherwise knowledge will be frozen at this point in time. Similar arguments hold for serious DCS. The concern that some patients with limb bends may have more serious manifestations is not an excuse for damning the nomenclature. The diagnosis in that case is not type I, but type II. If the physician is in error, education is needed not a name change.

The classic nomenclature still allows for combined DCS and AGE, and still allows the clinician to state that the exact nature of the disorder is unclear, but a treatment plan will be proposed on the information available. It also allows the patient to have peripheral manifestations in the presence of systemic symptoms. This is no different that what we do every day in clinical medicine. The simplicity of this diagnostic scheme allows the physician to account for a wide spectrum of symptoms which can still represent DCS or AGE or both, and avoids the need to create a complex unique description for every patient. If a data base of signs and symptoms is wanted in these disorders, then keep a data base of signs and symptoms, but do not call them a diagnosis. Use them to arrive at a diagnosis.


1. Francis TJR, Smith DH: Describing Decompression Illness. 42nd UHMS workshop. Undersea and Hyperbaric Medical Society, Kensington, MD, 1991

2. Dutka AJ: Clinical Findings in Decompression Illness. A proposed Terminology. In Moon RE and Sheffield PJ: Treatment of Decompression Illness. Undersea and Hyperbaric Medical Society, Kensington, MD, 1996. pp 1-9.

3. Golding FC, Griffiths P, Hempleman HV, Paton WDM, Walder DN: Decompression Sickness during construction of the Dartford Tunnel. Brit. J. Indus. Med. 17: 167-180, 1960

Return to table of contents

Classification of the decompression disorders:

time to accept reality

Richard E. Moon, M.D. Durham, North Carolina

The constellation of signs and symptoms associated with a reduction in ambient pressure was first observed in compressed air workers in the 19th century and was called "compressed air illness" or "caisson disease." Similar signs and symptoms were subsequently observed in divers breathing compressed air and were referred to as "diver's paralysis" or "diver's palsy." Early in this century the elucidation of the pathophysiology of this disorder led to the terms "decompression sickness" (DCS) for in situ bubble formation caused by inert gas supersaturation, and "arterial gas embolism" (AGE) for intravascular gas due to pulmonary over-pressurization.

In 1960, in reference to compressed air workers, Golding classified decompression sickness as "type I," referring to cases exhibiting only pain, and "type II," in which there were symptoms other than pain, or abnormal physical signs, including neurological manifestations (1). This sub-classification (type I DCS, type II DCS, and arterial gas embolism) has been used by the US Navy (USN) as a guide to diagnosis and treatment. According to the USN algorithm, type I DCS is to be treated under certain conditions with USN Table 5, whereas type II requires USN Table 6. The recommended treatment for AGE was USN Table 6A. For the majority of cases this treatment algorithm has been efficacious, and it has been widely accepted. Epidemiologic data pertinent to divers have traditionally not been recorded in sufficient detail to allow a more specific classification. However, as long as the majority of cases were associated with commercial or military diving there was no need to do so. Prediction of outcome and triaging were unnecessary, since chambers were available on site, and in the vast majority of cases in which recompression treatment was administered shortly after the onset of symptoms there was complete resolution. As a descriptive and epidemiologic tool, however, the "type I/type II" scheme has been lacking. The type II classification encompasses a huge spectrum of disease, ranging from paresthesias to quadriplegia. Furthermore, a universally accepted definition of the two types of DCS does not exist. Whereas type I in the Golding scheme includes only pain, the USN definition encompasses skin and Iymphatic manifestations (2). Type II has also some times been extended to include type I symptoms that occur during decompression. Both types are mutually exclusive in the Golding classification while the USN Diving Manual refers to types I and II symptoms and allows both types to coexist. Because of such inconsistencies it is impossible to methodically compare published series of diving accidents, as authors often do not define the manner in which they use the classification. In recent years, the usefulness of this traditional classification scheme has declined, due to several factors. First has been the growth of civilian recreational diving and the associated accidents, where often 24 h or more delay before recompression therapy is typical, and the outcome after treatment is less than uniformly successful. Second is the recognition of the difficulty of accurately classifying cases of decompression-associated symptoms. The traditional classification scheme is frequently applied inconsistently and incorrectly, undoubtedly in part due to the difficulty of ascertaining the exact cause of a diving accident (DCS vs. AGE). The diver may not remember, and the buddy has often not observed the breathing pattern during ascent. The clinical picture may be similarly unrevealing: gas embolism superimposed upon tissues with a significant inert gas load may result in a clinical presentation that resembles DCS rather than AGE. Third, the distinction between DCS and AGE has been further blurred by the realization that arterial bubbles probably contribute to the pathophysiology of decompression sickness in settings other than pulmonary barotrauma. Finally, the previously close relationship between DCS classification and choice of treatment table has now become less distinct. Many civilian diving physicians now treat all types of DCS with USN Table 6, irrespective of their classification, and the US Navy Diving Manual now recommends that recompression treatment of both DCS and AGE should begin at 60 fsw (18 msw). Even in USN practice, the only remaining therapeutic use for the old classification is to decide whether USN Table 5 can be utilized (i.e., to classify decompression sickness into "type I" and "non-type I"). If the old classification is to be discarded, with what should it be replaced? To reexamine the issue of diagnostic terminology, a workshop was held in 1990 in which there was a consensus in favor of abolishing the classification based on etiology (AGE vs. DCS) (3). The workshop participants recognized the difficulty of determining accurately the pathophysiology of decompression accidents and accepted the all-encompassing term decompression illness (DCI). While no new concept is engendered by this simple semantic change, it permits reference to gas bubble disease without requiring any insight into pathophysiology, as is implied when "decompression sickness" or "arterial gas embolism" are used. It was further proposed that it would be more appropriate to describe DCI descriptively, according to onset, evolution, and some estimation (e.g., depth-time profile) of inert gas load. The various reasons for classification of any disease include the need to predict prognosis and susceptibility to treatment and to design treatment algorithms. Identification of subsets that are particularly amenable or resistant to treatment is essential for the design of clinical trials. In the literature on neurologic bends it is impossible to identify such subsets because such a wide spectrum of disease is lumped into the category type II DCS. However, recent studies have demonstrated the feasibility of identifying such subsets by sub-classifying neurologic bends according to severity (4,5). It is traditional in medicine to attempt to classify diseases according to causation. However, in the absence of a unified knowledge of pathophysiology, many classification schemes in current use in other medical disciplines incorporate symptoms or signs, for example schizophrenia, lymphoma, migraine, and leprosy. The availability of a specific diagnostic procedure which can differentiate subsets (the usual requirement for classification of individuals cases by etiology) is missing for such entities, as well as, for the present, the decompression disorders. Irrespective of whether the data structure suggested by Francis and Smith (3) is the best one, to allude to either DCS or AGE, "decompression illness" is unquestionably a useful shorthand term requiring no insight into the pathogenesis of a particular case of gas bubble disease. Indeed, it has been widely accepted as such. The term "DCI" has been embraced by organizations whose task it is to collect epidemiologic data as well as by scientists and clinicians. A literature search at the time of writing reveals 35 indexed publications using the term "decompression illness." The development of large databases containing detailed information about symptomatology now makes available the tools necessary to examine the possibility that a new classification system, for example, frequent clustering of symptoms, or even better, response to treatment and long-term outcome, might be more clinically useful than the present one. Just such an approach was used to develop a classification for the muscular dystrophies (6). Whether decompression disorders are amenable to similar analysis is an unanswered question, but even to attempt it would require more detail than is available in the present classification.

Accepting the term "DCI" does not imply that the old terminology "DCS" and "AGE" should be eliminated. These terms are unambiguous and perfectly appropriate for denoting the pathophysiologic concepts for which they were defined. However, when classifying an individual patient the inherent uncertainty is best reflected by the term decompression illness.


1. Golding F, Griffiths P, Hempleman HV, Paton WDM, Walder DN. Decompression sickness during construction of the Dartford Tunnel. Br J Ind Med 1960; 17:167-180.

2. Navy Department. US Navy Diving Manual, vol. 1 revision 3. Air diving. NAVSEA 0994-LP-001-9110. Flagstaff, AZ: Best Publishing, 1993.

3. Francis TJR, Smith DJ, editors. Describing decompression illness. Kensington, MD: Undersea and Hyperbaric Medical Society,1991.

4. Kelleher PC, Pethybridge RJ, Francis TJR. Outcome of neurologicaL decompression illness: development of a manifestation-based model. Aviat Space Environ Med 1996; 67:654 65B.

5. Ball R. Effect of severity, time to recompression with oxygen, and retreatment on outcome in forty-nine cases of spinal cord decompression sickness. Undersea Hyperbaric Med 1993; 20:133-145.

6. Walton JN, Nattrass FJ. On the classification, natural history and treatment of the myopathies. Brain 1954; 77:169-231.

Return to table of contents



The comments below are from an editorial which was published in Chest (1997;112:576-578); it refers to the paper by Tetzlaff et al on Risk factors for Pulmonary Barotrauma (Chest 1997;112:654-659). This paper is a retrospective review of cases of pulmonary barotrauma (PBT). The data suggest that evidence for risk factors for PBT can be found in many cases of "undeserved" pulmonary barotrauma and air embolism:

Pulmonary Barotrauma in Divers.

Can prospective pulmonary function testing identify those at risk?

Alfred A. Bove, MD, PhD

In the 1930's a number of deaths were observed during submarine escape training in the U.S. Navy. Studies by Behnke (1), Polack and Adams (2), and Shilling (3) showed that the mechanism was arterial air embolism caused by pulmonary barotrauma. They reported the features of this injury as separate from the already understood problem of decompression sickness. Subsequently, fatalities were identified in which no evidence of pulmonary disease was present, but pulmonary barotrauma occurred, resulting in air embolism. In some cases, obstruction of a single bronchus caused over expansion of a lung segment. From this point, diving and submarine training was prohibited for subjects with a history of any disease that caused airway obstruction, and asthma was included. Later studies (4,5,6) provided further insight into mechanisms and prevention of pulmonary barotrauma. Asymptomatic anatomic abnormalities of the lung (7) or airways reactivity (8) were thought to contribute to the risk of barotrauma. Until recently, military and commercial diving prohibited candidates from diving who had a history of asthma. The concern was for bronchospasm, air trapping, pulmonary over pressure, arterial air embolism, and serious embolic complications from diving. Diving authorities thought that a significant reduction in injury and death would be achieved by disallowing anyone with asthma from diving. This practice in undersea medicine continued until the 50's when the sport diving community began to expand rapidly, and the strict medical criteria which were imposed upon military, and commercial divers were relaxed. Recent surveys of the sport diving population (9,10,11) indicated that 6-8% of the sport diving population had asthma or a history of asthma, and indicated that screening the sport diving community was unsuccessful in prohibiting the asthmatic population from diving. At the same time, data from diving accident reports maintained by the Diver's Alert Network at Duke University identified a small, non significant increase in risk of a diving accident in divers with a history of asthma. (12). The accident statistics in the diver's alert network were interesting in that the risk was in asthmatics who were actively wheezing at the time they were diving. Studies from England pointed out that many asthmatic divers were diving successfully without injury or fatality (13). These observations and requests from the sport diving community to develop rational guidelines for asthma and diving led to several workshops which pointed out that many asthmatics were diving safely, that a history of reactive airway disease was inadequate to prohibit sport diving, and measures of pulmonary function could be used to discriminate the diver at risk from the individual with a history of reactive airway disease who would not be at risk. The conclusion of a workshop in 1995 (14) suggested that pulmonary function with flow-volume curves demonstrating that mid- expiratory flow was within 80% of expected normal would allow safe diving without concern for pulmonary overpressure and barotrauma.

The paper by Tetzloff et al, provides important clinical observations that support a role for prospective pulmonary function testing as a means of separating individuals with a history of reactive airway disease who can dive safely from those who might be at risk for pulmonary barotrauma. Although their study is retrospective, they were able to identify changes in the flow-volume curve, specifically, reduced mid expiratory flow at 25% of vital capacity in pulmonary barotrauma patients compared to patients who had decompression sickness, but not pulmonary barotrauma. FEV1 was similar in both groups.

The workshop on asthma and diving held in 1995 (14) suggested similar criteria for selection of safe divers among an asthmatic population. The current study is the first since the workshop to confirm that an abnormal flow volume curve is an important criteria for selecting divers at risk from the asthmatic population. Expiratory flow at the late portions of the flow volume curve appear to be more valuable than peak expiratory flow, FEV1 or flow at 75% of vital capacity. The fact that the obstruction to flow occurs at small lung volumes is also significant in that simple bedside spirometry measuring FEV1 will not detect the changes.

The authors also performed computed tomography of the chest and found 13 abnormalities among the 15 patients with pulmonary barotrauma. Although only 4 cases in the control group had computed tomography of the chest, none had an abnormal finding.

Points on the flow-volume curve (peak, MEF 75, MEF 50, and MEF 25) can identify diving candidates who would be at risk for pulmonary barotrauma. An additional study of the chest by computed tomography would add anatomic information, particularly regarding lung cysts and blebs which might not be identified by a functional test. In Tetzloff's population 5 of the barotrauma patients had lung cysts.

The finding that FEV1 and vital capacity were normal in patients with barotrauma is important to physicians evaluating patients for diving as these tests which are simpler to perform are not indicative of risk. The flow-volume curve with measurements near the end of expiration provides more information.

Based on the assumption that exercise induced asthma can occur during diving, current recommendations for assessing safety in diving candidates with a history of asthma is to perform flow-volume curves before and immediately after exercise. Heavy exercise in diving often occurs when divers are swimming on the surface. Thus exercise-induced asthma is not likely to contribute to pulmonary barotrauma, but may contribute to drowning when exercise on the surface causes severe dyspnea and panic.

As understanding of reactive airway disease improves, and the relationship between exercise, reactive airway disease and functional testing is clarified, a more precise screening capacity will be available for the practicing physician. The fact that Tetzloff et al were able to identify risk factors in 93% of the patients who suffered pulmonary barotrauma after diving indicates that we can develop the necessary screening to eliminate individuals who are at risk.

As usual, a careful history is important. Identifying lung cysts or blebs on computed tomography and reduction of the MEF 25 flow to below 80% of normal, all may contribute to the screening of high risk individuals. Larger studies are needed to confirm this combination of screening techniques, and attention must be paid to the cost of screening particularly if screening involves computed tomographic studies. On the other hand, if flow-volume curves can identify high risk individuals who have a history of asthma and wish to dive, the costs of pulmonary screening is an excellent investment.


1. Behnke AR : Analysis of accidents occurring in training with the submarine "lung". U.S. Naval Med Bull, 30: 177-184, 1932.

2. Polak B and Adams H: Traumatic Air Embolism in Submarine Escape Training. U.S. Naval Med Bull, 30:165-177, 1932.

3. Shilling CW: Expiratory force as related to submarine escape training. U.S. Naval Medical Bull. 31:1-7, 1933.

4. Schaeffer KE, Nulty WP, Carey C, et al: Mechanisms in development of interstitial emphysema and air embolism on decompression from depth. J Appl Physiol, 13:15-29, 1958.

5. Liebow AA, Stark JE, Vogel J, et al: Intrapulmonary air trapping in submarine escape casualties. U.S. Armed Forces Med J, 10:265-289, 1959.

6. Malhotra, MC, Wright CAM: Arterial air embolism during decompression and its prevention. Proc R Soc Med B, 154:418-427, 1960.

7. Mellem H, Emhjellem S., Hurgen O: Pulmonary barotrauma and arterial gas embolism caused by an emphysematous bulla in a scuba diver. Aviat Space Environ Med, 61: 559-562, 1990.

8. Wagner PD, Dantzker DR, Iacovoni VE, Tomlin WC, West JB: Ventilation-perfusion inequality in asymptomatic asthma. Ann Rev Respir Dis, 118:511-524, 1978.

9. Neuman TS, Powers AT, Osborne DE: The prevalence of asthma, diabetes and epilepsy in a population of divers. Undersea Biomed Res 988; 15(Suppl):62-63.

10. Bove AA, Neuman T, Kelsen S, et al: Observation on asthma in the recreational diving population. (Abstract). Undersea Biomedical Research 1992;19(Suppl.):18.

11. Corson KS, Moon RE, Nealen ML, et al: A survey of diving asthmatics. (Abstract). Undersea Biomed Research 1992;19 (Suppl.):18-19.

12. Corson KS, Dovenbarger JA, Moon RE, et al: Risk assessment of asthma for decompression illness. (Abstract). Undersea Biomed Research 1991;18 (Suppl.):16-17.

13. Farrell PJ, Glanville P: Diving practices of Scuba Divers with Asthma. Br Med J, 300:609-610, 1990.

14. Elliott DE: Are asthmatics fit to Dive? Undersea and Hyperbaric Medical Society, Kensington, MD, 1996.

Return to table of contents






The paper described in the reference and abstract below explains some of the difficulties in resuscitating victims of severe air embolism who have a cardiac arrest. The injection of large amounts of air into the circulation causes a "vapor-lock" of the heart chambers and great vessels and prevents blood from flowing through the heart.


Fatal Pulmonary Barotrauma Due to Obstruction of the Central Circulation with Air. Neuman TS, Jacoby I, Bove AA.. J. Emergency Medicine 16:413-417, 1998


Cardiac arrest in cases of barotraumatic arterial gas embolism (AGE) is usually ascribed to reflex dysrhythmias secondary to brainstem embolization or secondary to coronary embolization. Several case reports suggest that obstruction of the central circulation (i.e. the heart, pulmonary arteries, aorta, and arteries of the head and neck) may play a role in the pathogenesis of sudden death in victims of pulmonary barotrauma. We report three consecutive cases of fatal AGE in patients in whom chest roentgenograms demonstrated confluent air lucencies filling the central vascular bed, the heart, and great vessels. In none of the victims was there evidence by history or at autopsy that the intravascular gas was iatrogenically introduced. Total occlusion of the central vascular bed with air is a mechansim of death in some victims of AGE, and resuscitation efforts for such patients should take this possibility into consideration.

Return to table of contents






The article recently published in Undersea and Hyperbaric Medicine provides some insight into the effect of a PFO on risk of DCS.

Bove, AA: Risk of Decompression Sickness with Patent Foramen Ovale. Undersea and Hyperbaric Research Sept, 1998


Several reports have described populations of divers with decompression sickness (DCS) who have a patent foramen ovale (PFO). The presence of a PFO is known to occur in about 30% of the normal population, hence, 30% of divers are likely to have a PFO. Although observations have been made on the presence of a PFO in divers with and without DCS, the risk of developing DCS when a diver has a PFO has not been determined. In this study, we used Logistic Regression and Bayes' theorem to calculate the risk of DCS from data of three studies which reported on echocardiographic analysis of PFO in a diving population, some of whom developed decompression sickness. Overall incidence of decompression sickness was obtained from the sport diving population, from the U.S. Navy diving population, and from a commercial population. The analysis indicates that the presence of a PFO produces a 2.5 times increase in the odds ratio for developing serious (type II) DCS in all three types of divers. Since the incidence of type II DCS in these three populations averages 2.28/10,000 dives, the risk of developing DCS in the presence of a PFO remains small, and does not warrant routine screening of sport, military or commercial divers by echocardiography.

Return to table of contents

Table 1 from the article

Incidence of DCS in several diving populations. Table 1 from the article shows the incidence of decompression sickness in several diving populations. References are from the article bibliography listed below.

 SOURCE (reference)


SPORT (11,12)



Total Dives*

Total DCS*

Type II DCS*

Incidence DCSÝ

Incidence DCS IIÝ





















* Values are number of events. Ý incidence per 10,000 dives, DCS II -DCS Type II

Return to table of contents


1. Hagen PT, Scholz DG, Edwards WD: Incidence and Size of Patent Foramen Ovale During the First 10 Decades of Life: an Autopsy Study of 965 Normal Hearts. Mayo Clin Proc 1984; 59: 17-20

2. Lynch JJ, Schuchard GH, Gross CM, Wann LS: Prevalence of Right-to-left Atrial Shunting in a Healthy Population: Detection by Valsalva Maneuver Contrast Echocardiography. Am. J. Cardiol 1984; 53: 1478-1480

3. Lechat PH, Mas JL, Lascault G, et al. Prevalence of Patent Foramen Ovale in Patients with Stroke. New Engl. J. Med. 1988; 318: 1148-1152

4. Hanna JP, Sun JP, Furlan AJ, Stewart WJ, Sila CA, Tan M: Patent Foramen Ovale and Brain infarct. Echocardiographic Predictors, Recurrence and Prevention. Stroke 1994; 25: 782-786

5. Labovitz AJ, Camp A, Castello R, Martin TJ, Ofili EO, Rickmeyer N, Vaughan M, Gomez CR: Usefulness of Transesophageal Echocardiography in Unexplained Cerebral Ischemia. Am. J. Cardiol. 1993; 72: 1448-1452

6. McNeil BJ, Keeler E, Adlestein SJ: Primer on Certain Elements of Medical Decision Making. New Engl. J. Med 1975; 293: 211-215

7. Kleinbaum, DG: Logistic Regression Springer-Verlag, New York, 1994

8. Wilmshurst PT, Byrne JC, Webb-Peploe MM: Relation between Intraatrial Shunts and Decompression Sickness in Divers. Lancet 1989; ii: 1302-1305

9. Moon RE, Kisslo JA, Massey EW, Fawcett TA, Theil DR: Patent Foramen Ovale (PFO) and Decompression Illness. Undersea Biomedical Res 1991; 13(Suppl): 15

10. Cross SJ. Evans SA, Thomson LF, Lee HS, Jennings KP, Shields TG. Safety of Subaqua Diving with a Patent Foramen Ovale. BMJ 1992; 304: 481-482

11. Gilliam BC: Evaluation of decompression sickness Incidence in Multi-day Repetetive Diving for 77,680 Sport Dives. In Lang MA, Vann RD (eds): Repetitive Diving Workshop. American Academy of Underwater Sciences, Costa Mesa, Ca, 1990: 15-25

12. Report on Diving accidents & Fatalities. Divers Alert Network, Durham, NC, 1996

13. Howsare CR, Jackson RL, Rocca AF, Morrison IJ. U.S. Navy Decompression Illness and Fatalities, 1990-1995. Undersea and Hyperbaric Medicine 1997; 24 (Suppl): 77

14. Imbert JP, Fructus X, Montbarbon S: Short and Repetitive Decompressions in Air Diving Procedures: the Commercial Diving Experience. In Lang MA, Vann RD (eds): Repetetive Diving Workshop. American Academy of Underwater Sciences, Costa Mesa, Ca, 1990: 63-72

15. Moon RE, Camporesi EM, Kisslo JA: Patent Foramen Ovale and Decompression Sickness in Divers. Lancet 1989; i: 513-514

16. Hallenbeck JM, Bove AA, and Elliott DH: Mechanisms underlying spinal cord damage in decompression sickness. Neurology 25:308-3l6, l975.

17. Sykes JJW, Yaffee LJ: Light an Electron Microscopic alterations in spinal cord myelin sheaths after decompression sickness. Undersea Biomed Res. 12: 251-258, 1985

18. Lever MJ, Miller KW, Paton WDM, Smith EB: Experiments on the genesis of bubbles as a result of rapid decompression. J. Physiol. Lond. 184: 964-969, 1966

19. Malhrota MC Wright CAM: Arterial Air Embolism during decompression and its prevention. Proc. R. Soc Med B 154: 418-427, 1960

20. Vann RD: Mechanisms and risk of decompression sickness. In Bove AA: Bove and Davis' Diving Medicine. W.B. Saunders, Phila, 1997: 146-158

Return to table of contents