Pulse
In Defense of a Dying Art
Autopsies yield important information for medicine and public health. With fewer hospital autopsies being done, what clues about disease and death are physicians missing?
Throughout his 27-year career, internist Eugene Ollila, M.D., often has sat down with the families of patients who have died and gently suggested pursuing an autopsy. More and more, he’s encountering resistance to the idea. Some families contend that the patient, having undergone tests and procedures before dying, has endured enough. Others believe that if the patient recently had surgery, the physician has all the information he or she needs. To continue pressing for a postmortem exam, “we have to have an awfully good reason,” says Ollila, who practices at Allina Medical Clinic in downtown Minneapolis.
Such negotiations have not always been necessary. Half a century ago, all patients who died in the hospital received an autopsy. Since the 1970s, however, the rates of autopsies conducted in hospitals have been in a steady decline. For John Jones, M.D., a general pathologist at Abbott Northwestern Hospital, the change in recent decades has been more than a little noticeable. When he started practicing in 1980, pathologists at the hospital conducted about 250 autopsies a year. The numbers have “declined about 80 percent,” he says. “We won’t get to 50 [adult autopsies] this year.” Even the training requirements for pathology residents have changed. Residents used to be responsible for performing 200 autopsies during training; now, they’re required to do 50 to be board-eligible, and they have to rotate among hospitals in the area to attain those numbers.
Death and Disincentives
Pathologist Eric Pfeifer, M.D., of Mayo Clinic has observed a similar downward trend. He estimates that families decide to pursue an autopsy “about 20 percent of the time.” Both he and Jones concur that the reasons for the decline appear to stretch far beyond families’ reluctance. The rise in the use of diagnostic imaging techniques such as CT and MRI has made some physicians wonder whether autopsies are necessary at all. “There may be a sense with clinicians that [with these technologies] … fewer [patients] are dying without a correct clinical diagnosis,” Jones says. Fear of medical malpractice may be another reason why physicians are skittish about pursuing the procedure often referred to as “the final consultation.”
There are clear financial disincentives to doing autopsies, too. Including the pathologist’s fee, autopsies cost about $1,500 and insurance does not pay for them. Most hospitals absorb the cost of the procedure when patients die in-house. When patients die at home or in hospice, family members who want an autopsy have to pay for it themselves. “Sadly, there is anecdotal evidence,” acknowledges Pfeifer, “that some pathologists may charge a prohibitive price for a postmortem exam.”
Further, for hospital pathologists, who have time-sensitive obligations to check biopsies and tissue samples to determine diagnoses for living patients, the task of performing autopsies can seem like an added chore. Autopsies arise unexpectedly, and the two-hour investigation generally falls to the end of the day, when physicians otherwise would be heading home. In addition, says Lindsey Thomas, M.D., medical examiner for the Minnesota Regional Medical Examiner’s office in Hastings, a lot of pathologists just don’t like doing them—“What we do is messy, it smells,” she states—and haven’t rallied in resistance to the decline that’s taking place.
What Gets Missed
Even so, many pathologists and clinicians agree that there’s much at stake as the numbers dwindle. For one, autopsies offer “important quality assurance” for physicians, says Jones. “If we do find a major unexpected finding that might have resulted in a change in therapy or improved survival with [a different] premortem diagnosis, we refer [the information] back to the appropriate medical clinic.” In addition, internal medicine residents at Abbott Northwestern periodically review autopsied cases and correlate the findings with the patient’s diagnosis and care. “So it’s an important teaching tool … not just for pathology residents,” he points out.
In fact, studies have sought to determine whether autopsies yield important information. One German study published in Medicine in 1996 found from randomly sampled autopsies culled over four decades that modern diagnostic technologies such as CT, ultrasound, and MRI miss potentially life-saving diagnoses in about 10 percent of patients. The most commonly missed diagnoses included pulmonary emboli, myocardial infarctions, infections, and malignancies.
Autopsies also yield important public health information. They can detect infectious processes of new diseases such as SARS, point out consumer-product malfunctions that are putting lives in danger, and define trends in new, dangerous street drugs. As records, they’re invaluable tools. In a study published in Archives of Internal Medicine in February of this year, a Mayo Clinic epidemiologist reviewed a sampling of autopsy records from Olmsted County and found early signs of atherosclerosis: Plaque-clogged arteries were present in 83 percent of people ages 16 to 64 who had died of causes other than heart disease.
Even unusual cases make a strong argument for autopsies. Thomas and her colleague, Susan Roe, M.D., often turn up surprising results in the forensic cases they see. “Those are the ones that get you fired up,” Thomas says. (In Minnesota, each medical examiner’s or coroner’s office is a county agency, and forensic autopsies, which investigate sudden unexpected natural deaths as well as accidental deaths, suicides, and homicides are paid for by tax dollars, explains Andrew Baker, M.D., chief medical examiner of Hennepin County. In contrast to hospital autopsies, the number of forensic autopsies is not in decline. “Our numbers are as good as ever,” he says, “if not rising.”) One case that a partner of Thomas and Roe encountered was that of a 45-year-old man who had high blood pressure, high cholesterol, and a history of smoking. For two weeks, he’d complained of shortness of breath and coughing, and his physician had diagnosed a reactive airway disease and prescribed inhalers. Six days later, the man died at home. The autopsy revealed that he’d had an intimal sarcoma of a pulmonary artery, an extremely rare tumor.
Although some autopsies reveal diagnoses so rare a pathologist might never encounter them again, others can yield life-saving information for living family members. “Not infrequently, we find something that has hereditary implications,” Thomas says. The autopsy findings of a patient who died in a car accident included arrhythmogenic right ventricular cardiomyopathy, a genetic condition that can cause sudden death. Thomas and Roe took the information back to the family to let them know members could be checked for it.
According to Baker, the medical examiner’s office is “the tip of the spear” in a public health disaster, the place where the first sign of an outbreak or the presence of a biological weapon might be discovered. “I’ve never seen a case of anthrax or botulism,” he acknowledges, “but I would hope that we would at least recognize the case is weird enough that we’d collect the appropriate specimens and get a hold of public health right away.” (He adds, “Those are the kinds of cases that keep you up at night. Being a worrywart is kind of an occupational hazard in forensic pathology.”)
Closure and Comfort
More often, both hospital and forensic pathologists provide closure for the bereaved. Baker says, “Families have very logical and natural questions, like, ‘When my dad had whatever killed him, did it cause him to suffer? If he’d gone to the doctor, would he still be alive today?’” Some of those questions, he acknowledges, can’t be answered. “The ones you can answer, you answer them forthrightly and honestly. And in many ways that provides a lot of comfort to the family.”
Sometimes, providing closure means telling a difficult truth. A sudden death thought to have been caused by heart disease may be revealed to have been the result of a drug overdose. “It’s appropriate that you convey that information to [families] in an understanding way,” Baker says. “You don’t really know if they’re going to be surprised by it or not. But at the end of the day, you owe them the truth. That’s why we do what we do. We get paid to call ’em like we see ’em.”—Kate Ledger