National Halothane Study: a Study of the Possible Association Between Halothane Anesthesia and Postoperative Hepatic Necrosis; Report. Edited by John P. Bunker [and Others] (1969)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

PART VI. FORMAL RECOMMENDATIONS John P. Bunker Stanford University, School of Medicine Palo Alto, California Although this report might make it appear as though there were two separate studies (he- patic necrosis and surgical death rates), it should be emphasized that the Subcommittee, ably supported by its consultants, worked as a single remarkably harmonious team, and that all major decisions in the design and implementation of the Study were made by the full Subcommittee. The final action of the Subcommittee, taken early in 1966, was to formulate a series of rec- ommendations, which it was hoped might be of special interest and importance to the practice and science of medicine. Four recommendations were agreed on, of which three appear below. The fourth called for the implementation of studies of the marked variability in reported death rate from one institution to another. This recommendation was withdrawn from the sum- mary report (1) because of the difficulties in ap- praising and interpreting the observed differ- ences; but the possibility was left open that fur- ther analyses could lead in this full report to a new recommendation for studying institutional variation. Further analyses were carried out (Chapter IV-6), and their implications are dis- cussed at length in Chapter IV-8, where the orig- inal recommendation is tentatively reintroduced. The personal recommendations of Lincoln E. Moses and Frederick Mosteller, that new con- sideration be given to building teams of medical research workers for carrying out large cooper- ative studies and that new mechanisms be de- veloped for evaluating, particularly through large-scale cooperative trials, new therapies that have passed the developmental stage, also appear in Chapter IV- 8. Their proposals are closely related to the Subcommittee's Recom- mendation 2. Attention should also be called to Chapter ffi-5, in which two members of the Subcommittee suggest how the results of the Study might be applied to the practice .of anesthesia. This they do with some hesitation, for each physician should, and will, draw his own conclusions and apply them to his practice of medicine as he sees fit. Here, then, are the formal recommendations of the Subcommittee on the National Halothane Study: 1. We recommend that consideration be given to the initiation of limited randomized studies of death rates associated with anesthetic agents. The present Study has provided baseline data on death rates and frequency of various pro- cedures, and similar data, which should be par- ticularly valuable in planning such studies. This Study has left unexplained the relatively high death rate of cyclopropane and the observed but possibly misleading low death rate of ether. Although we can trust the indications that ether, nitrous oxide-barbiturate, and halothane had lower death rates, we are not able to say whether they lead to lower death rates, rather than merely being in association with them, possibly through bias due to selection. Such trials should not be undertaken unless, when compared with other uses of medical resources, it is thought worth while and feasible to realize a reduction in mortality of the order of one in 200 or unless firm baselines for death rates are in themselves regarded as highly valuable. If such objectives are to be sought, it would be advisable to choose operations for the study that have the following characteristics: (1) two or more anesthetics are regarded as equally suitable for the operation; (2) the death rate for the operation is appreci- able, say, at least 2 percent; (3) the operation is one that is frequently performed; and (4) necropsy rates can be anticipated to be sufficiently high if necropsies are needed to ensure success of the study. 2. We recommend the establishment of a cooperating group of institutions to serve as a panel-laboratory for the acquisition of trust- worthy information on new drugs (not merelyC T - anesthetics) as they come into useT In the history of medicine, it is doubtful whether any drug was ever more extensively studied both before and after its introduction than halothane. Yet, after halothane had been given to patients perhaps 10 million times, it was impossible to give firm, reliable answers to many basic questions about its effects. Two such questions were: "How does the death rate after operations under halothane anesthesia com- pare with death rates when other anesthetics are used?" "Does halothane induce significantly more hepatic dysfunction than other widely used anes- thetics?" The National Halothane Study attempted to answer these questions by using existing rec- ords. Although 856,500 operations were brought under scrutiny, the answers given are predictably and regrettably short of those desired. For ex- ample, the important questions of nonfatal he- patic injury was not taken up by the Study. The 417

limitations of knowledge on halothane are cer- tainly not peculiar to it. Limitations at least equally compelling apply to nearly any drug in- troduced in the past. Had halothane been ad- ministered a few scores of thousands of times in the context of an experimental information- gathering system, similar in kind to a cooperative randomized clinical trial, reliable information might have been acquired for over-all death rates, and possibly for nonfatal hepatic injury as well. 3. We recommend consideration of the es- tablishment of a registry for the collection of clinical, laboratory, and pathologic*findings in case of hepatic necrosis. Massive hepatic necrosis is rare, but usu- ally fatal. In some patients it follows what ap- pears to be typical viral hepatitis. Massive he- patic necrosis may also follow some major surgical procedures, shock, congestive heart failure, and the use of large amounts of pressor drugs. But in some patients the cause of hepatic necrosis is not so apparent. A number of the recently introduced drugs, such as iproniazid phosphate and zoxazolamine, are thought to be perhaps occasionally responsible; similar sus- picions concerning halothane formed the basis for the present Study. The National Halothane Study has not en- tirely ruled out a rare relationship between halothane and massive hepatic necrosis. It will be important to know, as further data accumu- late, whether this association will continue, in- crease, or disappear. New, possibly hepatotoxic, drugs will continue to be introduced and, because of its infrequency, any associated massive ne- crosis may go unnoticed unless looked for with care. The proposed registry would provide the mechanism for collecting such information. In designing such a registry, it must be rec- ognized that for many, if not most, purposes effective interpretation of the data requires knowledge of the size and composition of the population from which the registered cases arise. Some registries have no provision for obtaining such "denominator" data and are hampered in carrying out their mission. Possibly such a registry should be developed in relation to the kind of panel mentioned in Recommendation 2, so that the needed background information would be readily available, or in association with an exist- ing registry that has access to information about its population. In establishing this registry, the most care- ful consideration must be given to the many in- herent limitations and pitfalls. These include: (1) the historical, nonexperimental nature of the Study; (2) the very low incidence of the variable of interest; and (3) the loss of data by nonre- sponse, such as missing laboratory data and failure to obtain necropsy. In addition, other, possibly serious, difficulties will undoubtedly become apparent only as experience with such registries develops. It is apparent that, unless the greatest efforts are made to identify and overcome these problems, neither this nor any other registry can achieve its goal. On the con- trary, it will likely generate misleading or er- roneous information. Finally, a decision to establish a project of the magnitude of such a registry should be made in the light of the total needs of the public health and the availability of medical resources. REFERENCE 1. Subcommittee on the National Halothane Study of the Committee on Anesthesia, National Academy of Sciences.National Research Council. Summary of the National Halothane Study: Possible association between halo- thane anesthesia and postoperative hepatic necrosis. J.A.M.A. 197:775-788, 1966. U.S. eOVHHHEHT PRINTING OFFICE i Ifil O —334-553 418

I » .y. ,M i

as ' t::: • --• >--!•i' i.:::;>. • l*l« - •lilt* s.:"iC T - rww '":'Z ^fwc '?S5 gj H BS& ;;!.>. .»:.«» •a.»''• s:s:s