Correspondence
Hyperbaric Oxygen for Acute Carbon Monoxide Poisoning
N Engl J Med 2003; 348:557-560February 6, 2003
- Article
To the Editor:
The positive effect of hyperbaric-oxygen treatment on cognitive sequelae after carbon monoxide poisoning, reported by Weaver et al. (Oct. 3 issue),1 has important implications for patients and their physicians. Since facilities for the administration of hyperbaric oxygen are usually available only at specialized centers, a broader indication for the use of such facilities would lead to an increase in time-consuming and costly transportation of patients, which is not without risk. Therefore, it is important to indicate which subgroup of patients will benefit most from hyperbaric-oxygen treatment.
The positive effect of hyperbaric-oxygen treatment in conscious patients has been suggested by Thom et al. in a report on an unblinded study.2 However, in comatose patients on mechanical ventilation, a positive effect of hyperbaric oxygen treatment could never be demonstrated.3-5 In the study by Weaver et al., only 8 percent of the patients were intubated. Although the patients were stratified according to their age, whether they lost consciousness, and the interval between the end of exposure to carbon monoxide and the first chamber session, a subgroup analysis was not reported.
5 ReferencesAnne-Cornélie J.M. de Pont, M.D.
Evert de Jonge, M.D.
Margreeth B. Vroom, M.D.
Academic Medical Center Amsterdam, 1105 AZ Amsterdam, the Netherlands
a.c. depont@amc. uva. nl 1
Weaver LK ,Hopkins RO ,Chan KJ , et al. Hyperbaric oxygen for acute carbon monoxide poisoning. N Engl J Med 2002;347:1057-1067
Full Text | Web of Science | Medline2
Thom SR ,Taber RL ,Mendiguren II ,Clark JM ,Hardy KR ,Fisher AB . Delayed neuropsychologic sequelae after carbon monoxide poisoning: prevention by treatment with hyperbaric oxygen. Ann Emerg Med 1995;25:474-480
CrossRef | Web of Science | Medline3
Juurlink DN ,Stanbrook MB ,McGuigan MA . Hyperbaric oxygen for carbon monoxide poisoning. Cochrane Database Syst Rev 2000;2:CD002041-CD002041
Medline4
Raphael JC ,Elkharrat D ,Jars-Guincestre M-C , et al. Trial of normobaric and hyperbaric oxygen for acute carbon monoxide intoxication. Lancet 1989;2:414-419
CrossRef | Web of Science | Medline5
Scheinkestel CD ,Bailey M ,Myles PS , et al. Hyperbaric or normobaric oxygen for acute carbon monoxide poisoning: a randomised controlled clinical trial. Med J Aust 1999;170:203-210
Web of Science | Medline
To the Editor:
A larger controlled trial by Scheinkestel et al.1 showed no benefit of hyperbaric-oxygen therapy, and the patients in their study had more severe carbon monoxide poisoning than did the patients in the study by Weaver et al. Although Weaver et al. cited that study, I think it merits greater consideration. One major difference between the two studies is the duration of normobaric-oxygen therapy (three days in the study by Scheinkestel et al.). An important unanswered question is whether protracted normobaric-oxygen therapy provides the same benefits as intermittent hyperbaric-oxygen therapy; I know of no reason why this should not be the case. Such an approach would avert the need for expensive and dangerous transfers of patients to centers with facilities for the administration of hyperbaric oxygen.
1 ReferencesJames P. Finnerty, D.M.
Countess of Chester Hospital, Chester CH2 1UL, United Kingdom
finnerty@tiscali.co. uk 1
Scheinkestel CD ,Bailey M ,Myles PS , et al. Hyperbaric or normobaric oxygen for acute carbon monoxide poisoning: a randomised controlled clinical trial. Med J Aust 1999;170:203-210
Web of Science | Medline
To the Editor:
Weaver et al. recommend hyperbaric oxygenation for the treatment of acute carbon monoxide poisoning, regardless of its cause and severity. However, a larger randomized trial previously demonstrated that hyperbaric oxygenation was useless in patients with accidental, residential carbon monoxide poisoning who did not lose consciousness.1 This finding was confirmed in later studies.2 In addition, two hyperbaric-oxygenation sessions in comatose patients resulted in recovery rates at one month that were similar to the rates with one session.
It is noteworthy that in 31 percent of the patients in the study by Weaver et al., carbon monoxide intoxication was related to a suicide attempt. Obviously, in these cases, other gases were involved along with carbon monoxide. In addition, most such patients combine gas with drugs or alcohol intoxication, so there is no way to relate the symptoms specifically to carbon monoxide itself.
One also wonders about the clinical relevance of the cognitive tests used by Weaver et al. It is very likely that the suicidal persons had previously abnormal results of cognitive tests. Yet base-line information on abnormal cognitive tests is not provided. Given the heterogeneity of the population and the rather small sample, one cannot rule out an imbalance between the treatment groups with respect to abnormal results of cognitive tests, just as there was an imbalance with respect to cerebellar signs.
2 ReferencesJean Claude Raphael, M.D.
Djillali Annane, M.D., Ph.D.
Raymond Poincaré Hospital, 92380 Garches, France
jean-claude.raphael@rpc. ap-hop-paris. fr Sylvie Chevret, M.D., Ph.D.
St. Louis Hospital, 75475 Paris, France1
Raphael JC ,Elkharrat D ,Jars-Guincestre M-C , et al. Trial of normobaric and hyperbaric oxygen for acute carbon monoxide intoxication. Lancet 1989;2:414-419
CrossRef | Web of Science | Medline2
Annane D ,Chevret S ,Jars-Guincestre C , et al. Prognostic factors in unintentional mild carbon monoxide poisoning. Intensive Care Med 2001;27:1776-1781
CrossRef | Web of Science | Medline
To the Editor:
We disagree with the claim by Weaver et al. that hyperbaric-oxygen therapy improves the outcome of carbon monoxide poisoning, for several reasons. First, the “battery” of six neuropsychological tests used by Weaver et al. was limited, since they failed to include measure of delayed memory, which is common after anoxia. Second, testing was undertaken at six weeks “to identify patients in whom delayed cognitive sequelae developed.” However, since no base-line measure was obtained immediately after the poisoning episode, any judgment about the development of impairment is impossible, nor can there be any assessment of differential recovery of function in the first six weeks (which might reasonably be expected). Third, there was no control group of subjects without poisoning, so regression to the mean cannot be ruled out as a factor in the differences noted.1 Finally, the study did not evaluate activities of daily living or other measures of functional performance. Thus, the results of the study are unconvincing, both for methodologic reasons and because of the failure to demonstrate “real life” benefits of hyperbaric-oxygen treatment.
1 ReferencesGeoffrey K. Isbister, M.B., B.S.
University of Newcastle, Newcastle 2298, Australia
gsbite@bigpond.com Patricia McGettigan, M.D.
Newcastle Mater Misericordiae Hospital, Newcastle 2298, AustraliaIrina Harris, Ph.D.
Macquarie University, Sydney 2109, Australia1
Kim H . Regression of lateral asymmetry scores toward the mean. Cortex 1994;30:331-341
Web of Science | Medline
To the Editor:
Weaver et al. hypothesize that one of the mechanisms by which hyperbaric-oxygen therapy reduces cognitive sequelae in patients with acute carbon monoxide poisoning could be preservation of mitochondrial ATP production, since their patients benefited from the therapy even in the presence of nearly normal levels of carboxyhemoglobin. In support of this mechanism in humans, we report data from six patients with acute carbon monoxide intoxication (mean [±SD] carboxyhemoglobin level, 22±6 percent). Peripheral-blood lymphocytes were obtained from the patients before and 3 and 10 days after they received hyperbaric-oxygen therapy. We measured oxygen consumption and the activity of complexes III and IV of the mitochondrial respiratory chain (both of which contain cytochromes, a target for carbon monoxide),1 using standard procedures.2
There was a marked inhibition of enzyme activity before the administration of hyperbaric oxygen, which was still clearly decreased 3 days after treatment (and in some patients even 10 days after treatment), despite already normal carboxyhemoglobin levels (Figure 1Figure 1
Mean (+SE) Enzyme Activity of Complex III and Complex IV of the Mitochondrial Respiratory Chain and Oxygen Consumption by Intact Lymphocytes before and after the Addition of Pyruvate (Complex I Substrate) and Succinate (Complex II Substrate) in 6 Patients Treated with Hyperbaric Oxygen and 30 Controls.). The decreases in the activity of complexes III and IV were accompanied by a reduction in the maximal rate of mitochondrial oxygen consumption, both in the absence and in the presence of substrates.These findings and our previous data in humans3 also argue for subcellular mechanisms in hyperbaric-oxygen therapy, distinct from elevated carboxyhemoglobin levels. However, the slow recovery of mitochondrial-respiratory-chain function suggests that other therapeutic approaches, in addition to hyperbaric-oxygen therapy, may be useful in some mitochondrial-respiratory-chain deficiencies.4
4 ReferencesFrancesc Cardellach, M.D.
Òscar Miró, M.D.
Jordi Casademont, M.D.
University of Barcelona, 08036 Barcelona, Spain
fcardell@clinic.ub. es 1
Brown SD ,Piantadosi CA . Reversal of carbon monoxide-cytochrome c oxidase binding by hyperbaric oxygen in vivo. Adv Exp Med Biol 1989;248:747-754
Medline2
Miro O ,Alonso JR ,Jarreta D , et al. Smoking disturbs mitochondrial respiratory chain function and enhances lipid peroxidation on human circulating lymphocytes. Carcinogenesis 1999;20:1331-1336
CrossRef | Web of Science | Medline3
Miro O ,Casademont J ,Barrientos A ,Urbano-Marquez A ,Cardellach F . Mitochondrial cytochrome c oxidase inhibition during acute carbon monoxide poisoning. Pharmacol Toxicol 1998;82:199-202
CrossRef | Medline4
Shoffner JM ,Wallace DC . Oxidative phosphorylation diseases and mitochondrial DNA mutations: diagnosis and treatment. Annu Rev Nutr 1994;14:535-568
CrossRef | Web of Science | Medline
Author/Editor Response
Dr. de Pont and colleagues ask for subgroup data. Our study was underpowered for post hoc subgroup analyses, and subgroup data could therefore be misleading.1
Dr. Finnerty incorrectly suggests that our results differ from those in a previous report2 because our patients were less severely poisoned. In fact, 80 percent of our patients (121 of 152) (vs. 73 percent2) met criteria used for severe poisoning. Dr. Finnerty also proposes that three days of normobaric-oxygen therapy should be equivalent to intermittent hyperbaric-oxygen therapy, yet the rate of cognitive sequelae was 25 percent with hyperbaric-oxygen therapy in our study, whereas it was 70 percent with normobaric-oxygen therapy in the study by Scheinkestel et al.2 Finally, our 132 medical transfers were carried out without adverse events.
It is not easy to compare our results with previous findings,3 because our study was different with respect to blinding, the definition of sequelae, neuropsychological testing, the patient population, and the dose and frequency of hyperbaric-oxygen treatment. Apparently overlooking our inclusion and exclusion criteria, Raphael et al. incorrectly assert that we recommend hyperbaric-oxygen therapy regardless of the cause or severity of carbon monoxide poisoning.3 They claim that the 31 percent rate of attempted suicide in our study confounds the results, because they expect cognitive function to be abnormal before carbon monoxide poisoning in patients who attempt suicide. If they are correct, the higher proportion of suicide attempts in the hyperbaric-oxygen group (36 percent, as compared with 26 percent in the normobaric-oxygen group)3 should have resulted in an increased rate of cognitive sequelae in the hyperbaric-oxygen group. However, we found a reduced rate of cognitive sequelae in the hyperbaric-oxygen group. We reanalyzed our data and found a lower, though statistically insignificant, rate of cognitive sequelae among patients who had attempted suicide (28 percent [13 of 47 patients]) than among those who had been accidentally poisoned (39 percent [41 of 105]) — the opposite of their expectation.
Dr. Isbister and colleagues correctly identify the absence of delayed-memory tests in our study. However, the tests we used detect impairments due to hypoxia. Since we found impairments with less sensitive cognitive measures, our results may underestimate cognitive dysfunction. Isbister et al. claim that base-line cognitive measures are necessary, and they appear to overlook the power of between-group comparisons in randomized clinical trials.1 Furthermore, on the basis of comparisons of our patients with normal matched control subjects,4 “regression to the mean” did not bias our results. Scores for activities of daily living and other functional scores, as well as “real life” benefits, are reported and discussed in our article.
We thank Dr. Cardellach and colleagues for sharing their data that support the concept of mitochondrial dysfunction due to carbon monoxide poisoning.
4 ReferencesLindell K. Weaver, M.D.
Ramona O. Hopkins, Ph.D.
Alan H. Morris, M.D.
LDS Hospital, Salt Lake City, UT 84143
lweaver@ihc.com 1
Spilker B. Guide to clinical trials. New York: Raven Press, 1996.
2
Scheinkestel CD ,Bailey M ,Myles PS , et al. Hyperbaric or normobaric oxygen for acute carbon monoxide poisoning: a randomised controlled clinical trial. Med J Aust 1999;170:203-210
Web of Science | Medline3
Raphael JC ,Elkharrat D ,Jars-Guincestre MC , et al. Trial of normobaric and hyperbaric oxygen for acute carbon monoxide intoxication. Lancet 1989;2:414-419
CrossRef | Web of Science | Medline4
Hopkins RO ,Weaver LK ,Churchill S . Neuropsychological performance in patients with carbon monoxide (CO) poisoning and demographically matched normal control subjects. Undersea Hyper Med 1999;26:Suppl:52-52 abstract.
- Citing Articles (2)
