IMAGINE GOING THROUGH CHEMO OR CANCER SURGERY BECAUSE SOMEONE MISLABELED, MISREAD OR JUST MIXED UP YOUR X-RAYS
Nick Lombardo remembers the moment of his death sentence. As he lay in a Long Island hospital bed days after abdominal surgery in June 1977, he overheard a group of doctors come touring by his room.” …And this patient here is terminal,” noted one chart-wielding man M white for the benefit of his colleagues.
Lombardo called his own physicians, who confirmed it: He had inoperable colon cancer; in 90 days he would be dead. At age 50, Lombardo, a Queens, New York, real-estate appraiser, planned his own funeral and sold his investments to provide money for his wife and children. “I went home and lay on my living-room couch and waited to die,” he says.
Although the doctors held out no hope, they started chemotherapy anyway. Lombardo had diarrhea, vomited throughout his house, lost his hair and began abusing sleeping pills. “At under 100 pounds I got so weak from the chemo that my wife practically had to carry me around. She went through the tortures of hell,” he says. “My kids were going crazy. My son stopped going to school.”
Along this darkening path toward death lay one major obstacle: Nick Lombardo did not have cancer.
After 20 months and some 90 courses of chemotherapy, he went to New York City’s respected Memorial Sloan-Kettering Cancer Center to find out why he was still alive. Specialists there determined that he had Crohn’s disease an inflammation of the intestine that can sometimes resemble malignancy and removed the affected part. In 1984, Lombardo won $1.6 million in a lawsuit charging his original physicians with misdiagnosis.
For Lombardo, life has never been the same:
The chemotherapy damaged his immune system, making him susceptible to infections (he’s had five bouts with pneumonia since treatment stopped) and putting him at high risk of developing leukemia. “For 12 years after the chemo stopped,” he says, “I would get that metallic taste in my mouth the taste of the drugs.”
A positive cancer misdiagnosis that is, diagnosis of a malignancy in a person who does not have one is not an everyday occurrence. Doctors agree almost universally that negative misdiagnosis, or failure to detect an actual cancer, is far more common and problematic. Yet, as Lombardo’s case and others show, doctors can interpret benign diseases as malignancies, sometimes with hideous results.
There is scant research on the frequency of or reasons for positive misdiagnosis. One recent study of ovarian cancer, a malignancy now recognized as one of the most troublesome in this respect, found a 5 percent positive misdiagnosis rate. The positive error rate for other cancers is probably lower, but no one knows for sure.
“If there is any indictment to be made, it is that sometimes we still lack the knowledge to identify cancers when they are there,” says Saul Gusberg, M.D., past president of the American Cancer Society. “From everything I know, a positive misdiagnosis has got to be very rare. Let’s not give people false hope.”
However, pathologist Robert Anderson, M.D., contends that there is “a significant proportion of major-league discrepancies” in cancer diagnosis. In 1989, when he was at the University of New Mexico School of Medicine, Anderson published a catalog of autopsy studies done over the past 50 years and found that the positive misdiagnosis rates for three cancers he looked at have not changed since the 1930s; 10 percent for stomach cancer, 14.3 percent for lung cancer and nearly 23 percent for liver cancer.
Anderson’s error rates are probably inflated because his data come from the most difficult cases patients whose problems were puzzling enough for doctors to order an autopsy. Nevertheless, Anderson believes his study accurately plots the basic trend; zero improvement in positive diagnosis accuracy for many cancers. “Let’s not quibble about what the exact percentage is,” he says. To Anderson, the problem is that fancy technology may have hurt cancer diagnostics as much as it has helped. “Physicians have begun to assume that if a machine costs a million dollars, it must be right,” says Anderson. “They have begun to rely more on that than on a thorough exam, clear thinking and instinct. We are standing still because although we have all these wonderful tools, we have abandoned old fashioned skills.”
Supporting Anderson’s point, a 1988 investigation by physicians at Boston’s Brigham and Women’s Hospital noted that advanced technologies like computerized tomography, sonography and nuclear-medicine procedures, all used to pinpoint malignancies, can yield misleading results where, for example, an inflammation appears to be a tumor. Another study, published last year by Albert Wu, M.D., assistant professor of health policy, management and medicine at Johns Hopkins University, found that misdiagnoses (mostly negative ones, for ailments ranging from pneumonia and diabetes to tumors) can arise from workaday problems: A doctor did not have enough specialized knowledge to evaluate symptoms; was distracted with other problems; or was simply too overworked and tired to function intelligently.
“At the level of complexity and uncertainty that doctors, as human beings, work today, it’s inevitable that mistakes are made. There’s no way out of it,” says David Hilfiker, M.D., a Washington, D.C., family practitioner who once wrote a wrenching account for Harper’ s magazine of how he aborted a healthy fetus after tests misled him into thinking it was dead.
The Tunnel Vision Trap
Maurie Marlcman, M.D., vice chairman of medicine at Memorial Sloan-Kettering Cancer Center, says, “It is possible that anything that occupies space in the body can be mistaken for a tumor.” Parasitic infections or abscesses in the liver can mimic malignancy; inflammations like diverticulitis may be mistaken for colon cancer; chronic gastritis or ulcers can look like carcinoma. “Sometimes ‘cancers of the pancreas’ are actually alcohol-induced pancreatitis,” says Markman. “The symptoms are the same, and without proper treatment, so is the outcome: death.” Some physicians assume that patients have cancer because they have enlarged lymph nodes, says Joseph Bailes, M.D., a leading oncologist in McAllen, Texas, who chairs the clinical practice committee of the American Society of Clinical Oncology “a big mistake,” since these can be caused by infections or other nonmalignant conditions.
A diagnosis of cancer is shaky at best without a biopsy a piece of the suspicious tissue, examined under a microscope by a pathologist. These days, certain biopsy samples (most commonly from the breast) are also subjected to a battery of DNA and other tests to analyze the structure of the cells. (Nick Lombardo’s misdiagnosis apparently stemmed from the surgeon’s failure to take a biopsy a major break with normal procedure, said experts who testified in his case.)
But like other professions, pathology the study of how disease affects the body’s tissues is highly specialized, and pathologists can develop tunnel vision. “Some of us are better at spotting certain kinds of cancer than others,” says Steven Silverberg, M.D., chairman of the division of anatomic pathology at George Washington University Medical Center. “If you have a kidney tumor, you should go to a lung person. Even then, there is no one in this profession who has never made a mistake-including me.”
Ovarian Cancer, Minus the Ovaries
Among the most problematic types of growths are those arising on or around the ovaries. In a study published last year, gynecologic oncologist Larry McGowan, M.D., a professor of obstetrics and gynecology at George Washington University School of Medicine, reviewed the records of more than 300 women diagnosed with ovarian carcinoma. He found that 5 percent actually had benign growths, and 8 percent had malignancies originating in the nearby peritoneum or intestines begun to assume that-not the ovaries. As a result, he says, those with nonovarian cancer may have received the wrong treatment, while women without cancer received treatment that wasn’t needed at all.
McGowan and other specialists say that ovarian cancer is particularly difficult to diagnose because growths from other organs sometimes intrude on the ovary and can look just the same as ovarian cancer under the microscope. Moreover, McGowan found in his research that surgeons sometimes misidentified the location of the growth, leading the pathologist to make a wrong diagnosis, or that the attending physician did not pay enough attention to symptoms or medical history, which would have led to a correct diagnosis. The most startling demonstration of laxity: A few of the misdiagnosed women did not even have ovaries; they had been removed by previous surgery. These patients apparently meekly accepted the diagnosis ,under the assumption that something had been missed in the previous operation, said McGowan, and the mistake slipped through the cracks.
“You can’t just look at a biopsy under a microscope,” says McGowan. “People need to communicate with each other and really be sure they have all the facts.”
A misdiagnosis of ovarian cancer can have a frightening impact not only on the woman but on her female relatives as well, since the malignancy has a strong genetic component: Someone with a close relative who has had ovarian cancer is two to three times more likely to develop the disease. Because of the relatively high error rate of 13 percent in McGowan’s study (it may have been worse ten years ago, when fewer physicians knew of the potential for confusion), anyone who fears she might have inherited a predisposition to the disease may want to have a specialist double check an aunt’s, sister’s or mother’s medical records. Many hospitals keep biopsy slides and other records for years, says McGowan, and will usually release them to relatives.
The “Problem” with Early Detection
Some biopsy misdiagnoses, made under pressure while the patient is on the operating table, are soon reversedbut not quickly enough. Alexandra Sloan, a Washington-area dental hygienist, went for surgery ten years ago when her doctor found lesions in her genital tract. She and her husband discussed the possiblity of cancer. Despite the fact that they were hoping to have a baby, they decided that if a “frozen section”a preliminary biopsy examined during surgeryindicated malignancy, the surgeon would remove the uteri’s, ovaries and fallopian tubes on the spot; they were concerned about the risk that waiting for a “permanent section,” in which tissue is analyzed in greater detail over several days, could allow the cancer to “seed” itself elsewhere.
Under the microscope, Sloan’s frozen section appeared to be positive for ovarian cancer, and the surgeon cut out her reproductive organs, sterilizing her forever at the age of 29. A couple of days later, says Sloan, the hospital’s head pathologist visited her bedside with “good news”: She wouldn’t have to have chemotherapy she didn’t have cancer.
In the ensuing lawsuit, it never did become clear whether the revised diagnosis a benign growth that probably did not require radical surgery was made from permanent section or from a review of the “frozen” section. But Sloan, who settled out of court, knows this: “If I knew what I know now about the incidence of misdiagnosis, I never would have gone with a frozen section. I’m much more frightened by the mistake than I was by the cancer. Disease you can fight. With this, I feel permanently violated.”
A soon-to-be-published study done at the University of Southern California Medical Center confirms the pitfalls of on-the-table diagnosis of ovarian cancer. Among women whose frozen sections showed low-malignant potentiar ovarian growths–borderline tumors that may or may not be truly cancerous-10 percent turned out benign on the permanent section, and 18 percent were full-blown malignancies. “The lesson is that you should get as much information as possible before doing an irreversible operation, especially with borderline tumors,” says C. Paul Morrow, M.D., director of gynecologic oncology at USC and one of the authors of the study.
Because early-detection tests such as mammograms have recently come into wide usage, doctors are seeing more and more borderline growths. The irony is that experts sometimes disagree about whether such incipient growths are malignant and require drastic treatment, or are harmless and should be left alone. In a study at the Yale University School of Medicine published last year, a researcher sent slides of what he considered borderline breast lesions to five top pathologists; in not a single case did all five agree on a diagnosis, with evaluations of the same tumors ranging from fully benign to fully malignant.
Janet Osuch, M.D., an associate professor of surgery at Michigan State University, says, “We are finding more of these tiny lesions, but we don ‘t know what to call them. Are they cancer? Or are they hyperplasia (a benign overgrowth of cells)?”
Charles Smart, M.D., who recently retired as the chief of the National Cancer Institute’s Early Detection Branch says similar issues are coming up with growths in the cervix, stomach, prostate, colon, mouth and lungs. “It used to be that when they told you that you had cancer, it was so far gone there was no question,” he says. “Now that we see things so much earlier, there are a lot of questions about what is really cancer and what is not.”
A boom in new cancer-screening procedures has meant that questions about the nature of a growth often arise before a biopsy is ever done. “Everybody wants an easy screening test” that will reveal whether cancer is present, but “some screening tests can put you through a lot of terror for nothing,” says Joseph Bailes. One example: the ultrasound tests meant to spot early prostate and gynecological cancers. In one study at the University of Kentucky, researchers who screened 1,600 women for ovarian cancer found 28 tumors. Only three were malignant, and eight were possibly premalignant. A blood test for the protein PSA, thought to indicate increased risk of prostate cancer, suggests possible malignancy in 16 percent of men who prove not to have prostate cancer on further testing. In one study, reported by the NCI, another protein test, called ELISA, when used for bladder cancer detected 100 percent of real cancers, but had a false-positive rate of 4 percent. (False negatives are a much more serious problem with some tests; Pap smears, with a miss rate of at least 10 to 15 percent, are the most notorious.)
One of the most popular screening tests and, in some cases, one of the most questionable is an assay for CA-125, a substance produced by ovarian tumors. The problem is that CA-125 levels can also be elevated by pregnancy, endometriosis, benign growths, cardiovascular disease, smoking, non-ovarian malignancies or, in 1 percent of patients, nothing at all. All told, the likelihood of false positive results is so high that many physicians see the test as useful only for monitoring patients who are at risk because of a past bout of ovarian cancer.
Robert Higgins, M.D., associate director of gynecologic oncology at Carolinas Medical Center in Charlotte, North Caroline,says, “Many women hear about CA-125 on TV, and they pressure their doctors for it because they’re worried about ovarian cancer. The doctors know the problems with the test, but they go along. And if it’s positive, patients may end up going through a surgical biopsy that wasn’t needed. Sure, they find out in the end there’s no cancerbut by that time they’ve been cut open.”
Patients don’t always find out about misleading test results or other errors. In Albert Wu’s study, doctors anonymously admitted that while they usually discussed serious mistakes with other physicians, less than a quarter told their patients. (A third of the errors involved diagnoses; others had to do with treatment or procedure.)
One case of silence involved Diane Weiner, a 30-year old resident of Long Island who went into a local hospital with chest congestion so severe that a bronchoscopy and a lung biopsy were performed. Somewhere along the line, her biopsy apparently was mixed up with a cancerous sample from another woman’s uterus. Based on that slide, Weiner was diagnosed with lung cancer and sent to Memorial Sloan-Kettering, where a surgeon took out a third of her right lung. After Memorial doctors combed through slice after slice of Weiner’s lung-184 tissue samples in all without finding cancer, they contacted the original hospital to find out what was going on. At some point, it became clear to physicians that Weiner had been misdiagnosed. But Weiner was left out of the information loop: In the months after the operation, she bought a cemetery plot, expecting to use it soon.
Memorial “just engaged in a massive cover up,” claims Richard Frank, her lawyer. “They punched her ticket on the Cancer Express; and when they found out she wasn’t supposed to be riding, they didn’t tell her.” About two months later Weiner’s personal doctor found out about the apparent mix-up and wrote a letter to physicians at both institutions demanding that they tell her the truthor else he would. Shortly after that, a Memorial surgeon contacted Weiner and told her what had happened from his point of view.
She sued various parties and was awarded $2.8 million. (Memorial doctors, who didn’t return our phone calls, claimed in court that endometrial tissue could resemble lung tissue, and that Weiner’s lung was so inflamed that she needed the operation anyway. Frank claims it was just a bad case of bronchitis.) Weiner still lives on Long Island; Frank said she preferred not be be interviewed. “She was just a little person in a big system,” he says. “Everyone was getting information except her.”
Avoiding Disaster on Your Own
Is there anything a patient can do to minimize the hopefully small chance of becoming a Diane Weiner, or of risking even a short-lived cancer scare? Honest errors aside, the answer is yes.
- Discuss with your doctor whether any cancer screening tests are advisable for you and, if so, which are the most reliable. Unless you have symptoms or belong to a risk group for a certain kind of cancer, some tests are not worth the risk of false positives. In general, doctors recommend routine screening for cancers of the breast, cervix, colon, slcin and testicles. Screening tests for cancers of the head, neck, ovaries, stomach and prostate are more apt to mislead. In some cases combining tests may increase their reliability. Blood-flow cytometry, which measures the amount of blood going through a tumor picked up by ultrasound, may help distinguish between cancerous and benign tissue (most malignancies have more blood flow).
- If cancer is suspected, find a pathologist and a surgeon who specialize in that kind of tumor. The continual flow of new information makes it impossible for a generalist to keep up. For example, different tumors are sometimes upgraded or downgraded. Certain growths that were labeled malignant two or three years ago are now considered benign, and vice versa.
- Never accept a cancer diagnosis without a biopsy. Request, if possible, that the pathologist observe crucial parts of the operation, since the exact location of a tumor and the condition of surrounding tissues can be an important aid to the diagnosis.
- An opinion from a second pathologist is a good idea and should definitely be pursued if the lesion is described as borderline, or if the doctor tells you that it is an unusual kind of cancer for the part of the body in which it appears. Consult a physician who specializes in that kind of tumor.
Finally, don’t wait for your doctor to ask all the questions; volunteer as much information as possible, including symptoms, medications you are taking and other medical history. Don’t be afraid to volunteer details you may think of as trivial; small physical signs may lead the diagnosis in a different direction, and some medicines can cause side effects that might throw off test results or otherwise be misinterpreted as signs of disease. “You have to approach any illness with the awareness that doctors do not have magic vision,” says David Hilfiker. “Be specific not only about your symptoms but about the level of concern you feel. Listen to yourself; people often know whether they are seriously ill or not.”
In the end, of course, you will either believe what the doctors tell you or not. “Let’s face it, it’s hard to participate in your own diagnosisit’s not like treatment, where you get to make the basic decisions,” points out Albert Wu. In the unlikely event there’s a slipup, he says, you’re not always going to find out. But you will greatly increase you chances if you “ask a lot of questions and learn to think aloud with your doctor. It will keep him thinking, too. In general, just be skeptical.”
Copyright 1992, Kevin Krajick. Distributed by the New York Times Special Features