National Halothane Study

Halothane reached clinical practice in 1956 and was rapidly adopted because it was nonflammable, potent, and smoother to administer than the ether and cyclopropane it replaced. Within a few years, case reports began appearing of fatal hepatic necrosis in patients who had received it, some after a single exposure but more often after repeat anesthetics. Anesthesiologists disagreed over whether halothane caused the damage or whether hypotension, surgical trauma, or coincident disease explained it. No single institution had enough cases to answer the question, and there was no infrastructure for pooling data across hospitals.

The National Halothane Study was organized to provide that infrastructure. Under the leadership of John Bunker and with analytic direction from Lincoln Moses, a biostatistician at Stanford, a subcommittee of the National Academy of Sciences assembled postoperative records from 34 hospitals covering more than 800,000 anesthetics. It was among the largest retrospective outcome studies conducted in surgery or anesthesia to that point, and its summary appeared in JAMA in 1966.

The study found that massive hepatic necrosis was rare and, overall, no more common after halothane than after other anesthetics; most cases had other apparent causes. But among the few cases of acute hepatic necrosis with no aggravating cause, halothane was more likely to have been the anesthetic used, which is why the JAMA summary spoke of a possible association. To compare death rates fairly, the analysts adjusted for procedure, age, and preoperative physical status.

The result quieted the immediate alarm, but it did not close the case. In 1968 the Fulminant Hepatitis Group reported 150 cases from 73 centers: 24 percent of the patients had received halothane before the illness, and 77 percent of those had had more than one exposure. Granulomas and eosinophilia in some patients led to the proposal of a hypersensitivity mechanism, though fully recognizable hepatic necrosis after halothane remained very rare. Halothane continued in clinical use for decades, especially in pediatric anesthesia, but was eventually displaced by newer volatile agents with less hepatic metabolism.

The study's most lasting contribution came from its methods. After adjustment for procedure, age, and physical status, death rates still varied considerably among the 34 hospitals, a finding that helped start research on differences in hospital outcomes. In 1974 the Stanford Center for Health Care Research extended that kind of comparison to 1,224 hospitals.