Portal de Periódicos da CAPES

Hyperbaric Oxygen Treatment for Cerebral Air Embolism—Where Are the Data?

1991; Elsevier BV; Volume: 66; Issue: 6 Linguagem: Inglês

10.1016/s0025-6196(12)60525-4

1942-5546

A. Joseph Layon,

Traumatic Brain Injury and Neurovascular Disturbances

In this issue of the Proceedings (pages 565 to 571), Armon and colleagues present an intriguing case of a cerebral air embolism sustained during the replacement of a mitral valve. Approximately 100 to 200 ml of air passed to the aorta when the atrial vent was inserted. Postoperative seizures and coma resulted, as well as a left-sided paresis. Although funduscopic examination showed no streaming air bubbles, blanched and nonfilled retinal arteries were noted bilaterally; routine therapy (phenytoin and hyperventilation) was begun. Thirty hours after the event, hyperbaric oxygen therapy was initiated. Perhaps because of the hyperbaric treatment, the patient had only mild left hemispheric deficits at 53 days and minimal residua at 14 months after the embolism was sustained. The authors cite this case as evidence that hyperbaric oxygen is the treatment of choice for cerebral air embolism, even if begun late, and their discussion is persuasive. Several issues need to be addressed: 1. What prospective data show that hyperbaric oxygen is the therapy of choice for cerebral arterial gas embolism?2. Does the outcome differ between early and late therapy with hyperbaric oxygen?3. What is the course of cerebral arterial gas embolism with, as opposed to without, hyperbaric oxygen therapy?4. How is cerebral arterial gas embolism diagnosed? Can neuroimaging studies play a diagnostic role?5. What is the role of strict management of fluid and glucose in the patient with cerebral arterial gas embolism? Although it may not be possible to answer each of these questions rigorously, my attempt herein will be to shed light on the infrequently explored areas of hyperbaric oxygen practice. 1. What prospective data show that hyperbaric oxygen is the therapy of choice for cerebral arterial gas embolism? One of the many criticisms leveled at the use of hyperbaric oxygen therapy was that, by 1987, no prospective randomized trials had been conducted to prove its utility in comparison with standard therapy for air emboli.1Gabb G Robin ED Hyperbaric oxygen: a therapy in search of diseases.Chest. 1987; 92: 1074-1082Crossref PubMed Google Scholar Indeed, a review of the pertinent literature for the past 30 years yielded no prospective, randomized trials in humans or animals that would prove the beneficial effect of hyperbaric oxygen for cerebral arterial gas embolism. One series of reports2Leitch DR Greenbaum Jr, LJ Hallenbeck JM Cerebral arterial air embolism. I. Is there benefit in beginning HBO treatment at 6 bar?.Undersea Biomed Res. 1984; 11: 221-235PubMed Google Scholar, 3Leitch DR Greenbaum Jr, LJ Hallenbeck JM Cerebral arterial air embolism. II. Effect of pressure and time on cortical evoked potential recovery.Undersea Biomed Res. 1984; 11: 237-248PubMed Google Scholar, 4Leitch DR Greenbaum Jr, LJ Hallenbeck JM Cerebral arterial air embolism. III. Cerebral blood flow after decompression from various pressure treatments.Undersea Biomed Res. 1984; 11: 249-263PubMed Google Scholar, 5Leitch DR Greenbaum Jr, LJ Hallenbeck JM Cerebral arterial air embolism. IV. Failure to recover with treatment, and secondary deterioration.Undersea Biomed Res. 1984; 11: 265-274PubMed Google Scholar perhaps provides the only data on the subject that approach being definitive. In this series, dogs were anesthetized and were then subjected to instrumentation and to an intracarotid injection of air. Cortical somatosensory evoked potentials, physiologic variables (cardiac output, arterial blood gases, blood pressure, pulse, hematocrit, and cerebrospinal fluid pressure), cerebral perfusion pressure, brain water content, and cerebral blood flow were measured. The animals were treated with either US Navy Table 6 (Fig. 1) or Table 6A (Fig. 2). Although cortical somatosensory evoked potentials deteriorated after air embolization, no significant differences were noted between the treatment groups with regard to rapidity of recovery after compression. Furthermore, no significant differences were found between treatment groups in any factor measured (physiologic variables, cerebral perfusion pressure, brain water content, and cerebral blood flow) even though cortical somatosensory evoked potentials deteriorated significantly after compression in four of five dogs in the Table 6A group.Fig. 2US Navy Table 6A. Initial air and oxygen treatment table for cerebral arterial gas embolism. The first 30 minutes is spent breathing air at 165 ft. During the next 4 minutes, the depth is decreased to 60 ft; this change is followed by three 20-minute oxygen breathing cycles interrupted by three 5-minute air breaks. The depth is decreased to 30 ft during a 30-minute period while oxygen is breathed. Two 60-minute cycles of oxygen interrupted by a 15-minute air break follow. Ascent to the surface occurs during a 30-minute period while oxygen is breathed.(From US Navy Diving Manual [NAVSEA 0994-LP-001-9010]. Vol 1: Air Diving. Revision 1. Washington, DC, US Government Printing Office, June 1985, pp 8–37.)View Large Image Figure ViewerDownload (PPT) The literature on humans is perhaps more problematic than the experimental animal work. In a retrospective review of 93 cases of air embolism from various causes, the mortality had been reduced from 93% to 33% by conventional emergency treatment; in that report, conventional therapy was defined as left lateral decubitus positioning, vasopressors, and oxygen under positive pressure.6Ericsson JA Gottlieb JD Sweet RB Closed-chest cardiac massage in the treatment of venous air embolism.N Engl J Med. 1964; 270: 1353-1354Crossref PubMed Scopus (76) Google Scholar Adding closed-chest cardiac massage to “conventional therapy” decreased mortality to approximately 28%. Because only seven study subjects received cardiac massage, however, the statistical validity is questionable. In another study, each of nine patients who had hemodialysis-associated air embolization was treated with conventional emergency therapy; this therapy resulted in only slight improvement of initial manifestations (predominantly in the cardiopulmonary and nervous systems).7Baskin SE Wozniak RF Hyperbaric oxygenation in the treatment of hemodialysis-associated air embolism.N Engl J Med. 1975; 293: 184-185Crossref PubMed Scopus (47) Google Scholar Compression to 165 feet seawater (fsw) resulted in dramatic improvement in seven patients within 10 minutes. The other two patients, who required a somewhat more extended treatment regimen, had no signs or symptoms at the conclusion of therapy. A retrospective study of 30 patients with air embolism treated with hyperbaric oxygen (US Navy Table 6) found that all but 6 patients (20%) recovered with minimal to no residua. Four of the six patients had severe residual neurologic findings; two patients died.8Hart GB Treatment of decompression illness and air embolism with hyperbaric oxygen.Aerospace Med. 1974; 45: 1190-1193PubMed Google Scholar This outcome represents a 7% mortality rate, in comparison with 33% for those who received conventional emergency therapy. A decrease in mortality from 93% with no therapeutic intervention, to 28 to 33% with conventional emergency treatment, to 7% with hyperbaric oxygen therapy seemingly provides a strong rationale for the use of hyperbaric oxygen in patients with cerebral air embolism. The comparison of data from different retrospective studies, however, especially when therapeutic interventions have changed over time, is questionable. Furthermore, I question (although no satisfactory answers are available) whether an iatrogenic air embolus caused by, for example, disconnection of a central venous line is pathophysiologically the same as a cerebral arterial gas embolism that results from rapid ascent from depth with breath holding. At minimum, a prospective, randomized animal study of factors similar to those previously evaluated must be conducted. 2. Does the outcome differ between early and late therapy with hyperbaric oxygen? Fully 30 hours after the occurrence of the embolism, Armon and colleagues opted to treat their patient with hyperbaric oxygen with US Navy Table 6A followed by US Navy Table 4 (Fig. 3). Would earlier treatment have resulted in a different or better outcome? So late after the initial event, would the patient have recovered even without the assistance of hyperbaric oxygen therapy? These questions have no definitive answers. Many investigators in this area agree that earlier, rather than later, treatment is most appropriate for an air embolus.3Leitch DR Greenbaum Jr, LJ Hallenbeck JM Cerebral arterial air embolism. II. Effect of pressure and time on cortical evoked potential recovery.Undersea Biomed Res. 1984; 11: 237-248PubMed Google Scholar, 9Rivera JC Decompression sickness among divers: an analysis of 935 cases.Milit Med. 1964; 129: 314-334PubMed Google Scholar, 10Davis JC Treatment of decompression sickness and arterial gas embolism.in: Bove AA Davis JC Diving Medicine. Second edition. WB Saunders Company, Philadelphia1990: 249-260Google Scholar The data supporting this statement, however, are anecdotal rather than scientific. No clear distinction exists between “early” and “late” treatment. Davis10Davis JC Treatment of decompression sickness and arterial gas embolism.in: Bove AA Davis JC Diving Medicine. Second edition. WB Saunders Company, Philadelphia1990: 249-260Google Scholar noted that the classic description of early treatment for cerebral arterial gas embolism is therapy within minutes after the event. This definition apparently originated from the initial cases that were in military and commercial diving settings where the diagnosis was quickly entertained and treatment facilities were readily available. More recently, sport divers, generally without this type of support, have been stricken. Thus, “late” or “delayed” treatment may mean a hiatus of hours to days after the event. Regardless of the duration of the hiatus, most physicians would consider it appropriate to treat “late” presenters with air emboli aggressively in the hope of obtaining a beneficial response. In my personal (and anecdotal) experience, response—albeit incomplete—is the rule with aggressive therapy, even when initiation is delayed. Some clinicians advocate different treatment for early as opposed to late presenters with air emboli. Davis10Davis JC Treatment of decompression sickness and arterial gas embolism.in: Bove AA Davis JC Diving Medicine. Second edition. WB Saunders Company, Philadelphia1990: 249-260Google Scholar asserted that late presenters should be treated only to 60 fsw (US Navy Table 6) rather than the 165 fsw (US Navy Table 6A) to which early presenters are taken. Because Davis provided no rationale for his recommendation, the reasons for this difference are likely based on his extensive experience. At our center, my colleagues and I use Table 6A for both groups. Pertinent available scientific studies,11Bond JG Moon RE Morris DL Initial table treatment of decompression sickness and arterial gas embolism.Aviat Space Environ Med. 1990; 61: 738-743PubMed Google Scholar although lacking randomized treatment and having a strong selection bias, suggest that outcome is better with Table 6 than Table 6A or Table 6 with extensions. Until adequate data are available, however, we continue to use Table 6A for our patients with air emboli. Once again, well-designed human or animal studies are needed to clarify these issues. 3. What is the course of cerebral arterial gas embolism with, as opposed to without, hyperbaric oxygen therapy? The approximate mortality rate of 90% in patients with air emboli of various causes can be reduced to 28 to 33% by conventional emergency therapy (left lateral decubitus positioning, vasopressors, administration of oxygen by a mask-valve-bag device, and closed-chest cardiac massage).6Ericsson JA Gottlieb JD Sweet RB Closed-chest cardiac massage in the treatment of venous air embolism.N Engl J Med. 1964; 270: 1353-1354Crossref PubMed Scopus (76) Google Scholar The use of hyperbaric oxygen therapy has further reduced the mortality to less than 10%7Baskin SE Wozniak RF Hyperbaric oxygenation in the treatment of hemodialysis-associated air embolism.N Engl J Med. 1975; 293: 184-185Crossref PubMed Scopus (47) Google Scholar, 8Hart GB Treatment of decompression illness and air embolism with hyperbaric oxygen.Aerospace Med. 1974; 45: 1190-1193PubMed Google Scholar when either US Navy Table 6 or Table 6A is added. Are these data credible? In my opinion, the answer is an equivocal yes. Nonetheless, one is obliged to ask for well-designed randomized human or animal studies to clarify these issues. It seems generally unwise, except in attempting to define the problem, to compare series of patients treated between 1930 and 1950 with those treated in the 1960s and 1970s. 4. How is cerebral arterial gas embolism diagnosed? Can neuroimaging studies play a diagnostic role? In iatrogenic cases, air embolism is often diagnosed situationally—that is, a change in mental status and a disconnected central venous line are noted, or air in the cardiopulmonary bypass tubing is obvious.12Murphy BP Harford FJ Cramer FS Cerebral air embolism resulting from invasive medical procedures: treatment with hyperbaric oxygen.Ann Surg. 1985; 201: 242-245Crossref PubMed Scopus (131) Google Scholar For such patients, the differential diagnosis is usually limited and, if a hyperbaric chamber is available, treatment can be instituted rapidly. For divers, the differential diagnosis between neurologic symptoms caused by cerebral arterial gas embolism and those caused by type II decompression sickness usually relates to the rapidity of onset. Furthermore, the former disorder frequently manifests as cognitive dysfunction, whereas the latter often manifests as progressive motor or sensory loss (or both) of the limbs; however, weakness and paresthesias may also occur in some cases of cerebral arterial gas embolism. In virtually all cases, the symptoms of cerebral arterial gas embolism develop within minutes of ascent;13Pearson RR Goad RF Delayed cerebral edema complicating cerebral arterial gas embolism: case histories.Undersea Biomed Res. 1982; 9: 283-296PubMed Google Scholar type II decompression sickness is usually slower in onset14Dick APK Massey EW Neurologic presentation of decompression sickness and air embolism in sport divers.Neurology. 1985; 35: 667-671Crossref PubMed Google Scholar (from minutes to hours) but may manifest almost as rapidly as air embolism. Although in some cases distinguishing between these two conditions will not be possible, treatment with US Navy Table 6 or Table 6A may be used for both. Several reports15Adkisson GH Macleod MA Hodgson M Sykes JJW Smith F Strack C Torok Z Pearson RR Cerebral perfusion deficits in dysbaric illness.Lancet. 1989; 2: 119-122Abstract PubMed Scopus (40) Google Scholar, 16Warren Jr, LP Djang WT Moon RE Camporesi EM Sallee DS Anthony DC Massey EW Burger PC Heinz ER Neuroimaging of scuba diving injuries to the CNS.AJR Am J Roentgenol. 1988; 151: 1003-1008Crossref PubMed Scopus (69) Google Scholar, 17Hodgson M Beran RG Shirtley G The role of computed tomography in the assessment of neurologic sequelae of decompression sickness.Arch Neurol. 1988; 45: 1033-1035Crossref PubMed Scopus (17) Google Scholar, 18Levin HS Goldstein FC Norcross K Amparo EG Guinto Jr, FC Mader JT Neurobehavioral and magnetic resonance imaging findings in two cases of decompression sickness.Aviat Space Environ Med. 1989; 60: 1204-1210PubMed Google Scholar suggest that, although computed tomographic scanning of the head is of little use, analysis by magnetic resonance imaging, single photon emission computed tomography, or stable xenon-enhanced computed tomographic scanning of regional blood flow may be helpful. The experience at our center with both computed tomographic and magnetic resonance imaging scanning has been less than encouraging. Thus, no pathognomonic sign or high-technology study allows the definitive diagnosis of cerebral arterial gas embolism. Nevertheless, the rapidity of onset, the type of symptoms, and, perhaps, selected imaging studies will assist in the diagnosis in most cases. Treatment should not be appreciably delayed to obtain these test results. 5. What is the role of strict management of fluid and glucose in the patient with cerebral arterial gas embolism? To date, the role of strict management of fluid and glucose in patients with cerebral air emboli has not been adequately addressed. Because the pathophysiologic features of air emboli include cerebral edema and regional abnormalities in blood flow, in addition to giving corticosteroids, I initiate treatment with one-half to three-quarters maintenance fluid therapy using a solution such as 0.9% saline (osmolarity, approximately 308 mosmol/kg) rather than a more dilute fluid. Furthermore, on the basis of animal work conducted at the Mayo Clinic19Lanier WL Stangland KJ Scheithauer BW Milde JH Michenfelder JD The effects of dextrose infusion and head position on neurologic outcome after complete cerebral ischemia in primates: examination of a model.Anesthesiology. 1987; 66: 39-48Crossref PubMed Scopus (169) Google Scholar as well as by others elsewhere,20Hoffman WE Braucher E Pelligrino DA Thomas C Albrecht RF Miletich DJ Brain lactate and neurologic outcome following incomplete ischemia in fasted, nonfasted, and glucose-loaded rats.Anesthesiology. 1990; 72: 1045-1050Crossref PubMed Scopus (51) Google Scholar I aggressively control the serum glucose to maintain a concentration of 100 to 150 mg/dl. These variations in treatment are not endorsed by all clinicians; they are within the realm of therapy based on clinical judgment and anecdotal experience. Whether such modifications are of use should be determined by a prospective, randomized trial. Armon and colleagues describe a reasonable course of hyperbaric oxygen therapy initiated in a patient 30 hours after she suffered a cerebral air embolism. The patient's recovery after treatment suggests that it was beneficial. The rationale, however, for using hyperbaric oxygen therapy for a substantial number of medical problems is not based on prospective and randomized scientific studies. Such data must be obtained, first in animals (although which model is best suited—cat, dog, monkey, baboon, or piglet—needs some thought and discussion) and then, once focused, in humans.