Every Patient Tells A Story Part 8

"You don't need it," he told her, ticking off the items that supported his diagnosis. It was early summer, when Lyme is most common. She lived in suburban Connecticut-not too far from the actual town of Lyme, where the disease was first identified. And she had a big, round rash typical of those seen in the early stages of Lyme disease.

He acknowledged that her symptoms weren't the cla.s.sic headache and stiff neck, but, still, she had the fever and body aches. The odds were overwhelming that this was Lyme, he told her. "Besides, this early in the disease, the Lyme test wouldn't tell us a thing." Then he was gone, off to the next room, the next patient, leaving Carol Ann with his scrawled prescription and a feeling of uncertainty.

Every spring and summer some version of this story is repeated tens of thousands of times in states of the Northeast, Midwest, and northern West Coast. Often, as in Carol Ann's case, the diagnosis will be made without a test, based on the patient's geography and symptoms, and cinched by the presence of the typical rash, known as erythema migrans. The diagnosis will be appropriate and reasonable, but not definitive. And in Lyme disease, that uncertainty has proved to be a particularly noxious ingredient.

Carol Ann took the antibiotics as prescribed. By the following weekend she felt almost back to her usual self. For most patients with Lyme disease, a single course of antibiotics is curative. But if Carol Ann had had a simple case of Lyme disease, I wouldn't be telling you this story. Instead, a few weeks later, Carol Ann developed pain and stiffness in her knees and hips. There was no swelling, no redness, just this strange reluctance in the joints of her lower body.

She went to her internist, who thought the symptoms were from Lyme disease. Untreated or inadequately treated, Lyme can attack the joints, causing pain, and usually swelling. He changed her to another antibiotic-doxycycline. She stayed on that for three more weeks but the stiffness continued. Her internist was baffled; he sent her to a rheumatologist. The rheumatologist wasn't sure what was going on either. So she went back to her internist. "He fobbed me off on his physician's a.s.sistant," Carol Ann said. "I was practically crying over the phone because of the pain. I told the PA I couldn't even sleep because the pain was so bad. She wasn't very sympathetic."

Carol Ann felt abandoned. Her doctor was a nice guy, she told me, but he clearly didn't know what was causing her pain or what to do about it. She decided to take matters into her own hands. She talked to friends; she rummaged the shelves of her local bookstore; she cruised the Internet. Everything seemed to point her back to her diagnosis of Lyme disease. She decided that she needed a Lyme specialist-someone who really understood the disease. So she set off to find one.

What Carol Ann didn't know-couldn't know-was that she was about to enter one of hottest controversies in medicine, a maelstrom of professional contention and confusion about Lyme disease, a controversy that would leave her in pain for two full years. Most patients believe that it is usually possible to determine what ails them with some kind of test-an X-ray, for example, or an MRI, or any of the hundreds and hundreds of blood and urine tests. If the test is positive, the patient may not be happy, but at least he believes he knows exactly what's wrong: a fractured wrist-bone, asthma, a tumor, a heart attack. If the test is negative, then the patient believes the result at least proves he doesn't doesn't have something, which can be a relief if what he thought he had was cancer or some other terrifying disease. Or it can be very frustrating-since often treatment and the possibility of cure depend on diagnosis. have something, which can be a relief if what he thought he had was cancer or some other terrifying disease. Or it can be very frustrating-since often treatment and the possibility of cure depend on diagnosis.

Doctors too put a great deal of faith in the power of diagnostic tests. And mostly for good reasons. Tremendous strides have been made in our ability to identify a disease using advanced technologies of one kind or another. While the patient's story and physical exam can often suggest a diagnosis, both doctors and patients like to see hard evidence-and that evidence usually comes in the form of results from some kind of diagnostic test.

But, as it turns out, tests and their results are not nearly as crisp and clear as many patients (and doctors) a.s.sume them to be. In fact, for all of their tremendous and invaluable power, the testing process can actually slow or sidetrack the diagnostic process in some cases-or derail it completely.

The complexities surrounding testing for Lyme disease have pitted doctors against doctors and led to a confusion bordering on chaos about the diagnosis of this common and highly treatable disease. The result has been a virtual epidemic of missed and mistaken diagnoses. Some patients end up suffering from undiagnosed acute Lyme disease. And hundreds-maybe thousands-of patients sick with other diseases are being "diagnosed" with a phantom illness and treated for a medical problem they don't have.

To Carol Ann it made sense that these aches and pains could be linked to her Lyme disease-after all, she'd been fine until then. Plus, her doctor had thought the first round of antibiotics ineffective. Why would the second round be any better? She finally located a Lyme specialist in nearby Wilton, Connecticut. He didn't take insurance-none of the Lyme specialists she called did-but his fee was reasonable and he was conveniently close.

By the time Carol Ann went to see the specialist, Dr. Matthew Davidson (not his real name), she was a wreck. Her body ached all the time. The joints didn't look injured-there was no swelling or redness-but they hurt so much that even sleeping was difficult. She was exhausted, her memory was shot, she couldn't concentrate, and normal daily frustrations often reduced her to tears.

Davidson is a general internist who has focused his practice on Lyme disease. A stocky man with thinning blond hair, he exuded a warmth and openness that impressed Carol Ann. She sat in his exam room and began to describe the symptoms that had taken over her life for the past year.

Davidson nodded his head as Carol Ann began to list her symptoms. Her illness was no mystery to him. Her presentation was cla.s.sic, he told her halfway through her story. What happened to her was common-far too common, in his opinion. She was right to seek his help. The antibiotics obviously hadn't worked; she hadn't been cured of her infection, and as a result she now had something he called "chronic Lyme disease."

Davidson explained that often an initial course of antibiotics doesn't kill the bacteria that cause Lyme disease. The bacteria somehow manage to "hide" in the body, only to reappear and cause a host of symptoms that include joint pain, muscle pain, insomnia, and lack of concentration-all symptoms that Carol Ann had. Davidson said that her only hope was to take even more antibiotics. Maybe for a few more weeks, possibly for months, perhaps even for years, in the hope of finally eradicating the insidious bug and its symptoms. It could be a long process, he told her, but with his help she could defeat the infection and regain her health.

As Carol Ann left Davidson's office that day, she felt more optimistic than she had since her symptoms had started. That mood wouldn't last long.

The diagnosis that Carol Ann was given-chronic Lyme disease-is one that tens of thousands of patients have been given in the thirty-some years since Lyme disease was first identified. And a whole cadre of doctors like Davidson, who call themselves "Lyme literate," a.s.sert a special expertise in what they claim is a chronic and insidious infection. But despite the claims of these physicians-and the sometimes fervent belief of their patients-"chronic Lyme disease" is almost certainly a phantom illness. Contrary to the claims of doctors like Davidson, there is little evidence that the bacteria that cause Lyme disease can persist in the face of antibiotics, causing the symptoms attributed to "chronic Lyme disease." Furthermore, there is plenty of solid evidence that shows that long-term treatment with antibiotics will not not cure whatever it is that ails those diagnosed with this syndrome. cure whatever it is that ails those diagnosed with this syndrome.

Despite this evidence, thousands of patients continue to be prescribed months and sometimes years of multiple antibiotics in a desperate quest for relief. The danger of this diagnosis and treatment are twofold. First, it puts patients at risk of serious side effects from the powerful drugs that are used. Second, this erroneous diagnosis can postpone diagnosis and treatment of other diseases, leaving patients even worse off than when they started.

How can reasonable, well-meaning medical doctors such as Davidson continue to believe in this phantom and continue to prescribe treatments that don't work? The answer has to do, at least in part, with the difficulty of diagnosing this complex disease. But it also is closely linked to a very human discomfort with the uncertainty when faced with a patient in pain and in need of an answer.

The Discovery of Lyme Disease The discovery of Lyme disease is one of the great pieces of medical detective work of the twentieth century. In 1956, Polly Murray, a young artist and housewife in Ess.e.x, Connecticut, began to suffer an array of inexplicable health problems: fevers, rashes, joint pains, and fatigue. Her memory didn't seem as sharp as it once was. She felt unfocused and found it hard to concentrate on her artwork. She went to her doctor. He was baffled. So were the specialists she was sent to see. Several suggested that the symptoms were all in her head-a manifestation of some psychiatric illness.

By 1964, Polly, her husband, and four children had moved to the small town of Lyme, Connecticut, an affluent community wedged into the verdant countryside between the Connecticut River and Long Island Sound. By then everyone in the family was suffering from the same symptoms Polly had. Even the dog was afflicted. Visits to the doctor were frequent, relief was rare, frustration high.

Over time Polly discovered that other people in the area were experiencing the same constellation of symptoms. Together, sufferers in her town had racked up hundreds of doctor visits and seen dozens of specialists. No one had an answer. No one could explain what they had or why so many of them had it. Finally, in October 1975, Murray called the state health department to report the strange local epidemic.

The health department turned to Dr. Allen Steere, a Yale rheumatologist who had spent his first two years out of medical school working for the Epidemic Intelligence Service (EIS), the investigative arm of the nation's primary public health watchdog, the Centers for Disease Control (CDC) in Atlanta. Steere asked Murray to come to his office in New Haven and bring her notes. Unlike many of the doctors Polly had seen, Steere showed a profound interest in her story. He collected the names of other people she knew who were suffering. Steere called each family on her list. They, in turn, gave him additional names, and eventually he compiled a list of twelve adults and thirty-nine children who had the same collection of symptoms as Polly and her family.

Steere immediately noted that the individual cases resembled juvenile rheumatoid arthritis. But that was a relatively rare disorder. What, he wondered, could have caused the cl.u.s.tering of so many cases of this uncommon disease in such a small area? He set out to discover what, if anything, the fifty-one individuals had in common.

The outbreaks were seasonal, reaching a peak each summer and then again in the fall. Based on that, Steere quickly focused on the possibility that this was some kind of insect-borne disease. But few patients remembered being bitten. And those who did described the appearance of the bite differently. It took two years of hard work before Dr. Steere and his colleagues fingered a culprit. Steere remembers the day clearly: it was in the summer of 1977 when a young man walked into his office with a vial containing a tick he had found after a walk through the woods near his home. The hiker had never seen one like it. Neither had Steere. It turned out to be an immature Ixodes scapularis Ixodes scapularis tick, a tiny black-legged arachnid, new to the region. Local insect census takers had been tracking the tick's march across Connecticut. Comparing the location of Steere's mystery cases and the areas invaded by the tick produced a geographical match. tick, a tiny black-legged arachnid, new to the region. Local insect census takers had been tracking the tick's march across Connecticut. Comparing the location of Steere's mystery cases and the areas invaded by the tick produced a geographical match.

The final piece of the puzzle remained a mystery until 1981, when w.i.l.l.y Burgdorfer, an entomologist for the National Inst.i.tutes of Health, identified the corkscrew-shaped bacteria transmitted by the ticks that actually cause Lyme disease. It was a new bacterial species-and it was named in his honor: Borrelia burgdorferi Borrelia burgdorferi.

The burgdorferi burgdorferi bacteria normally live in the blood of deer and various rodent species. As a larva, the bacteria normally live in the blood of deer and various rodent species. As a larva, the Ixodes Ixodes tick (commonly called a "deer tick") takes a meal of blood from its animal host and, if the animal harbors the bacteria, gets a dose of tick (commonly called a "deer tick") takes a meal of blood from its animal host and, if the animal harbors the bacteria, gets a dose of burgdorferi burgdorferi at the same time. The bacteria don't seem to bother the ticks. They just live quietly in the tick's gut. at the same time. The bacteria don't seem to bother the ticks. They just live quietly in the tick's gut.

The tiny arachnid has three life stages-larva, nymph, and adult. At each life stage the tick takes only one blood meal. Most cases of Lyme are transmitted in the tick's second stage. At that point the Ixodes Ixodes tick is dark and tiny-about the size of a pinhead-and the tick can easily latch on to a pa.s.sing human and find a quiet spot to attach itself and take its blood meal. In order to transmit the infection the tick needs to stay attached for at least twenty-four hours-which it can often do easily because it is so hard to see and the bite is usually painless. tick is dark and tiny-about the size of a pinhead-and the tick can easily latch on to a pa.s.sing human and find a quiet spot to attach itself and take its blood meal. In order to transmit the infection the tick needs to stay attached for at least twenty-four hours-which it can often do easily because it is so hard to see and the bite is usually painless.

The earliest and most common symptom of Lyme disease is an expanding circular rash, which usually appears around the site of the tick bite within a few weeks. The rash often presents in a "bull's-eye" pattern: a red ring circling a central clearing. However, some studies suggest that the most common presentation is a completely red, somewhat round patch that expands over the course of several days.

Left untreated, the burgdorferi burgdorferi bacteria migrate to other parts of the body and the immune system responds with inflammation, fever, muscle aches, and other symptoms as it tries to fight the infection. bacteria migrate to other parts of the body and the immune system responds with inflammation, fever, muscle aches, and other symptoms as it tries to fight the infection.

Until w.i.l.l.y Burgdorfer's discovery of the bacterial cause of Lyme disease, no test existed for it-for the simple reason that n.o.body knew what to look for. Even after the identification of the bacteria, testing for the disease remained difficult. Many types of bacterial infections can be diagnosed by culture-taking a sample (a throat swab, for example), rubbing it on a material that fosters bacterial growth, incubating the sample for a period of days, and then identifying the colonies of bacteria that form. But the Lyme bacteria don't grow well in culture.

Instead, doctors seeking to diagnose Lyme disease must depend on the body's response to the infection. To do this, doctors use two separate tests, neither of which is good enough to use alone but which, used together, can reliably identify those who have been exposed to the bacteria. It's an old strategy and while once commonplace, it has been replaced by better, more specific tests in many diseases. It's still used for a couple of diseases such as HIV and hepat.i.tis C-other diseases that cannot be easily grown in culture.

The first of the two tests is known as an ELISA (enzyme-linked immuno-sorbent a.s.say) and it detects antibodies to the invading bacteria or virus in a person's blood. Antibodies are part of the body's defense system and help kill invaders. This ELISA looks for antibodies to the Lyme bacteria. And it's a pretty good test but cannot distinguish between Borrelia burgdorferi Borrelia burgdorferi and many of its look-alike brothers, nephews, or distant cousins. Some types of normal flora can cause a positive reaction on the ELISA. and many of its look-alike brothers, nephews, or distant cousins. Some types of normal flora can cause a positive reaction on the ELISA.

That's why a second test is needed. If a person tests positive or equivocal on the ELISA, then a second test is conducted called the Western blot test. Again, this test is looking at antibodies, not actual bacteria. This test looks for antibodies not to the whole organism but to the basic building blocks of the Lyme bacteria-individual proteins. It's a complicated process because many types of bacteria use the same building blocks. So it's not enough to identify, for example, two or three of the proteins known to be part of burgdorferi burgdorferi bacteria. Those same proteins are also found in many other species. bacteria. Those same proteins are also found in many other species.

The CDC has determined a standard for interpreting results from the Western blot test series. The standards say that Lyme disease should be suspected only if a person's blood is found to have antibodies to five out of the ten proteins that are commonly tested for. If a patient tests positive on ELISA and and is positive on at least five out of the ten relevant Western blot tests, then it is very likely he has been exposed to Lyme disease. is positive on at least five out of the ten relevant Western blot tests, then it is very likely he has been exposed to Lyme disease.

If this were the end of the story, it wouldn't be so bad. The tests for Lyme disease would be indirect and require two steps, but the end result would be about as clear as we get in medicine. Unfortunately, it's more complicated than that.

First, it usually takes the body several weeks to develop enough antibodies against the bacteria to be measured by either of the two tests. In the earliest days of the infection, therefore, even though you may have the rash or other symptoms of early Lyme disease, neither of the two tests is likely to be positive. And if treatment is started early enough, the bacteria will be killed so quickly that the antibodies may never never get created. That means there's no way to definitely prove, years later, that a person did get created. That means there's no way to definitely prove, years later, that a person did not not have Lyme disease at some point in the past. But an even more important complication in testing for Lyme disease is that once the body does make antibodies, they stick around-for months or years-as protection against future infections. That means that the tests for Lyme will remain positive long after the bacteria that originally caused the symptoms are eradicated. It will look for all the world as though you still harbor the bacteria even though you don't, because the tests don't measure bacteria-they measure antibodies. have Lyme disease at some point in the past. But an even more important complication in testing for Lyme disease is that once the body does make antibodies, they stick around-for months or years-as protection against future infections. That means that the tests for Lyme will remain positive long after the bacteria that originally caused the symptoms are eradicated. It will look for all the world as though you still harbor the bacteria even though you don't, because the tests don't measure bacteria-they measure antibodies.

The Phantom Epidemic Carol Ann, of course, did not understand these layers of complexity about testing for Lyme disease when she consulted with Dr. Davidson. All she knew was that when she had left Davidson's office on her initial visit, he said he would send off a sample of her blood for a confirmatory test-which certainly seemed to Carol Ann like responsible medical practice. Two weeks later, the results came back: they were negative. This struck Carol Ann as odd, but didn't seem to bother Dr. Davidson at all. He told her that testing was notoriously unreliable in making this diagnosis. He sent off more blood and started her on an antibiotic despite the negative test.

Several weeks later Davidson told Carol Ann the results of the second test were positive. He did not tell her that he wasn't following the guidelines for interpreting the results set by the CDC. Rather than the standard of five out of ten antibodies that the CDC regards as evidence of Lyme disease, her test had been positive in only three-which Davidson interpreted as "positive."

Like his "Lyme literate" colleagues, Davidson justifies his more lenient testing standards as necessary for not missing anyone who may have the disease. But this is a poor argument. It's like saying that all patients who have a sore throat, runny nose, and a fever have the dreaded avian flu. You probably won't miss any cases of avian flu with such a vague set of criteria, but most of the time your diagnosis will be wrong. Instead most of the patients you identify this way will have other, far more common illnesses-a cold, maybe bronchitis, or maybe the usual flu.

But all this was invisible to Carol Ann, who took some comfort in the apparent confirmation of her illness. In any case, she was feeling a little better since starting the latest round of antibiotics. Her shoulder didn't hurt quite as much and she was sleeping better. But the medication was taking a toll on her stomach. She lost a few pounds because she felt nauseated-especially in the hours right after taking the medication. Davidson thought it was important to continue the antibiotic, especially in light of the "positive" Lyme test, and encouraged her to soldier on. The hope of a complete recovery and of being her old self again made Carol Ann determined to continue taking the medication despite how ill it made her feel.

At about the same time that Carol Ann was forging ahead with her prescribed regimen of antibiotics, forty-four-year-old Will Hammer was negotiating slippery November roads to see his own "Lyme literate" physician. He had been diagnosed with Lyme disease over a decade earlier but, when I spoke with him, he said he'd suffered from the disease for more than half his life. A tall man with short-cropped red hair and a quiet manner, he told me proudly that he hadn't missed a day of work because of his "chronic Lyme disease" in over five years. He attributed his success to Dr. Andrea Gaito, a rheumatologist and leader in the "Lyme literate" movement who had had him on daily antibiotics for nearly thirteen years.

Hammer said he first developed symptoms in high school following a camping trip. He never had the fever, muscle pain, and headache typical of Lyme, but felt tired and run-down. "Occasionally I'd feel not quite right and I'd wonder about it," he told me, "but it wasn't overwhelming." The symptoms worsened in his twenties. He would have periods of insomnia, body pains, and fatigue. He went to doctor after doctor. No one even had a name for the illness he suffered, much less a cure.

Eventually he heard about Lyme disease and wondered if that could be causing his symptoms. He'd been tested for it in the past and been told the test was negative, but now he was hearing that the test wasn't very reliable. That's when he ended up in Gaito's office, who quickly diagnosed him with "chronic Lyme disease" and started him on a six-week course of antibiotics.

The effect was immediate and life-altering, Hammer said. "The first time I was treated here I felt better than I had in my entire adult life. I felt great." But, he told me, that good feeling didn't last long.

A couple of months after completing his treatment, he started having trouble sleeping again. Then the fatigue and body aches reappeared. Then he began to have problems with his memory. "At first it was little things. Then one day, I was driving my son to soccer as I had every weekend for months and suddenly I couldn't remember where to go. I couldn't quite remember where I was." He pulled off the road into a parking lot. His heart was racing. Slowly and carefully, he figured out how to get where he needed to go. His young son, confused by his uncharacteristic behavior, asked what was the matter. Nothing was wrong, he rea.s.sured his son, but inwardly he wondered what the h.e.l.l was happening to him.

He went back to Gaito and started back on antibiotics. He's been on them with only a couple of short breaks ever since. At several points over the years since starting these medications, Hammer found himself as sick as he'd been to start with despite the daily antibiotics. He'd go back to his doctor-discouraged, frustrated, and depressed. Gaito would help him get through it, increasing the dose of the antibiotics or changing him to a new one and eventually he'd start to feel a bit better. When I last spoke to Hammer, he was going to Dr. Gaito every three to four months, but he worried that he might never be well enough to stop taking the antibiotics.

Both Carol Ann and Will tell versions of a common story in Lyme country: a diagnosis of Lyme disease, followed by antibiotic treatment, an initial improvement, and then a return of symptoms. This pattern emerged early in the history of the disease. Allen Steere noticed that while most of his patients got well after a two-to-four-week course of antibiotics, between 10 and 20 percent took months and occasionally even years to feel completely better. Like Will Hammer and Carol Ann DeVries, these patients often noticed a lingering fatigue and body aches. Some complained that they had trouble sleeping or problems with their memory. Still others would have recurrences of the joint pain and swelling that brought them to the rheumatologist in the first place. Steere dubbed this phenomenon PostLyme Disease syndrome.

In the early 1990s, Steere and researcher Nancy Shad.i.c.k set out to determine how common the syndrome really was. They recruited one hundred residents from Ipswich, Ma.s.sachusetts, an area that had been hard hit by Lyme disease. Half the patients had a doc.u.mented history of Lyme disease, which had been treated; the other half had never had the disease. Nearly one in three of the treated patients continued to have residual pain and other symptoms more than five years after getting Lyme disease. These symptoms were seen far more frequently in those with Lyme than in those who'd never had it. Other studies too have found that those with a history of Lyme disease report more joint pain, fatigue, and memory problems than those who have never had the disease.

Initially there was concern that these symptoms represented an ongoing infection, which persisted despite a full course of antibiotics. Patients themselves said that it felt felt like an infection, and so Steere, Shad.i.c.k, and many other doctors responded initially by treating these symptoms with a second or third course of antibiotics. like an infection, and so Steere, Shad.i.c.k, and many other doctors responded initially by treating these symptoms with a second or third course of antibiotics.

But it soon became obvious to Steere and others that while many of those with these persistent symptoms got better after multiple courses of antibiotics, so did people who only got the one round of antibiotics-those prescribed at the time of their original diagnosis. Ultimately, most people got better and it wasn't at all clear whether the repeated courses of antibiotics made any difference at all, once the disease had been treated initially.

To better understand what was going on, researchers in the field did what researchers do-they set up experiments to study in a very careful, controlled way whether antibiotics really helped people with PostLyme Disease syndrome recover faster.

The New England Journal of Medicine New England Journal of Medicine published the first experimental results in 2001. Researchers at Tufts Medical Center in Boston and YaleNew Haven Hospital looked at 129 patients who had doc.u.mented proof of a previous infection with Lyme disease and persistent symptoms even after being given the recommended regimen of antibiotics. Most had some degree of musculoskeletal pain. Half were treated with ninety days of antibiotics and half were treated with an identical-looking placebo. Neither doctor nor patient knew who was getting which. Each partic.i.p.ant was evaluated for physical and psychological health before, during, and after treatment with the drug or placebo. published the first experimental results in 2001. Researchers at Tufts Medical Center in Boston and YaleNew Haven Hospital looked at 129 patients who had doc.u.mented proof of a previous infection with Lyme disease and persistent symptoms even after being given the recommended regimen of antibiotics. Most had some degree of musculoskeletal pain. Half were treated with ninety days of antibiotics and half were treated with an identical-looking placebo. Neither doctor nor patient knew who was getting which. Each partic.i.p.ant was evaluated for physical and psychological health before, during, and after treatment with the drug or placebo.

Just over 40 percent of the patients treated with the antibiotics felt better after the first month. And almost that many felt better, overall, after the full course of antibiotics and three months later. What about those on the placebo? The response was almost identical: 35 percent of the patients getting a completely inert substance felt better after the first month and 40 percent felt better by the end of the study. The antibiotics had made no difference at all.

Two other rigorous trials investigated the same issue. One showed a small improvement in symptoms of fatigue in those getting the antibiotics, but nothing else. The third study was done by psychologist Brian Fallon, a "Lyme literate" researcher and the head of a research center at Columbia University that is funded in part by a "chronic Lyme disease" advocacy group. But even Fallon failed to find any significant difference between the group on antibiotics and the group getting the placebo. Moreover, in each of these studies a significant number of partic.i.p.ants had complications stemming from antibiotic treatment. Many experienced the kinds of adverse reactions felt by Carol Ann, and sometimes the complications were serious enough that the study partic.i.p.ants had to be hospitalized.

These kinds of consistent results from three separate, rigorous experiments, published in high-quality, peer-reviewed journals, would normally be considered a slam dunk in the medical world. The results clearly showed that antibiotics were not not helpful for patients with persistent symptoms after being treated for Lyme disease, which strongly suggested that whatever was wrong with these people, it was helpful for patients with persistent symptoms after being treated for Lyme disease, which strongly suggested that whatever was wrong with these people, it was not not because they were harboring some "super because they were harboring some "super burgdorferi burgdorferi" in their bodies. Leading medical groups involved in the study of Lyme disease published guidelines recommending against the use of multiple courses of antibiotics to treat the persistent symptoms. Normally, this would settle the matter and medical science would move on to a new mystery.

But that's not what happened here. Instead a small but vocal group of doctors and patients refused to accept these results, refused even to accept the designation of PostLyme Disease syndrome. They clung, instead, to "chronic Lyme disease" and insisted that these symptoms did did reflect an ongoing infection that warranted continuing treatment with antibiotics. They countered the randomized controlled trials with research of their own, which often showed improvement in patients given antibiotics. But none of these studies compared the antibiotics against a placebo. The randomized controlled trials showed that while patients getting antibiotics did improve, so did those getting the salt.w.a.ter placebo. Studies done without the placebo had no way of telling whether the antibiotics were really effective or if the improvement was due to something in the normal ebbs and flows of any human condition. reflect an ongoing infection that warranted continuing treatment with antibiotics. They countered the randomized controlled trials with research of their own, which often showed improvement in patients given antibiotics. But none of these studies compared the antibiotics against a placebo. The randomized controlled trials showed that while patients getting antibiotics did improve, so did those getting the salt.w.a.ter placebo. Studies done without the placebo had no way of telling whether the antibiotics were really effective or if the improvement was due to something in the normal ebbs and flows of any human condition.

Advocates of the "chronic Lyme disease" model of understanding also seized on the complexities of Lyme tests and argued that because of their limitations, they should simply be disregarded. According to the International Lyme and a.s.sociated Diseases Society (ILADS), a group founded in 1999 to promote the dissemination of information on "chronic Lyme disease," diagnosis should be based only only on the patient's symptoms. They don't trust either physical exams or the two tests for Lyme disease. on the patient's symptoms. They don't trust either physical exams or the two tests for Lyme disease.

"Available data suggest that objective evidence alone is inadequate to make treatment decisions," say the ILADS guidelines, "because a significant number of Chronic Lyme Disease cases may occur in symptomatic patients without objective features on examination or confirmatory laboratory testing."

The same guidelines a.s.sert that when the two-step testing for Lyme is used as recommended by the CDC, up to 90 percent of cases are missed. But this is misleading at best. The "research" offered in the guidelines to support this "fact" consists of a single unpublished study. And if you use the test on everyone with the common symptoms of fatigue, insomnia, and muscle pains, 90 percent of them probably will have a negative test because they don't really have Lyme disease.

In fact, when used as recommended, the tests accurately diagnose Lyme correctly more than 90 percent of the time.

Instead of testing or the physical examination, ILADS and "Lyme literate" doctors suggest that diagnosis be made on symptoms alone. The problem here is that the symptoms these doctors have chosen to rely on are hopelessly broad and over-inclusive. They include fatigue, sore throat, muscle pain, joint pain, insomnia, chest pain, abdominal pain, dizziness, nausea, poor concentration, headaches, irritability, depression, back pain. These are some of the most common symptoms of patients presenting to a primary care office. As an internist who sees patients regularly, most of the patients I see complain of these symptoms-and they certainly don't all have Lyme disease. Because I have an urban practice, I see only a few cases of Lyme disease a year.

And yet this is the diagnostic strategy that allows doctors like Davidson and Gaito to maintain that patients like Carol Ann or Will have "chronic Lyme disease" and that they should be continually treated with antibiotics because of a lingering infection with the Lyme bacteria.

But there is probably an additional factor at work in the stubborn refusal of the "Lyme literate" doctors and patients to accept evidence that most other doctors find compelling. It's an unavoidable aspect of all medical practice and is particularly related to the limitations of diagnostic testing-limitations that are particularly acute with Lyme disease. I'm talking about a discomfort with uncertainty, with ambiguity, with not knowing. And the doctors most uncomfortable in this way are the ones most likely to seize upon a diagnosis or diagnostic label and distort their own thinking in an attempt to prove to themselves and their patients that they know what's going on.

The fact is that, more often than doctors like to admit, they cannot find a cause for a patient's symptoms. This was powerfully ill.u.s.trated in a 1998 study of two hundred patients diagnosed with Lyme disease. It turned out that more than half of the patients had no no evidence of ever being infected with Lyme disease. As we've seen, it could be that some of these patients were treated for Lyme so quickly that they never developed the antibodies that would prove their infection. Maybe. But that surely could not account for such a huge fraction of these patients who tested negative for Lyme. evidence of ever being infected with Lyme disease. As we've seen, it could be that some of these patients were treated for Lyme so quickly that they never developed the antibodies that would prove their infection. Maybe. But that surely could not account for such a huge fraction of these patients who tested negative for Lyme.

Only 44 of the 200 patients (20 percent) were found to have active Lyme disease-with symptoms, physical exam findings, and blood tests consistent with Lyme disease. Another 40 (19 percent) were found to have PostLyme Disease syndrome-with a clear history of Lyme infection, adequate treatment, and persistent symptoms. The other 116 patients in the study-60 percent of those enrolled-had no evidence of either current or past Lyme disease, even though all had been diagnosed with the disease. Clearly these results show that Lyme disease is being seriously over-diagnosed. But the results also shed light on the issue of uncertainty in medicine.

If the 116 patients who thought they had Lyme didn't didn't, then what did did they have? A lot of things, it turns out. Fourteen had rheumatoid arthritis. Fifteen had osteoarthritis. Eight were diagnosed with another infection, and another dozen had some kind of neurological disorder such as multiple sclerosis or amyotrophic lateral sclerosis (ALS), often referred to as Lou Gehrig's disease. Several were diagnosed with depression. they have? A lot of things, it turns out. Fourteen had rheumatoid arthritis. Fifteen had osteoarthritis. Eight were diagnosed with another infection, and another dozen had some kind of neurological disorder such as multiple sclerosis or amyotrophic lateral sclerosis (ALS), often referred to as Lou Gehrig's disease. Several were diagnosed with depression.

These diagnoses covered about half of the 116-and clearly demonstrate how the phantom diagnosis of "chronic Lyme disease" is concealing a great many real diseases that patients should be getting help with. But the other half of the partic.i.p.ants are also interesting. These people were obviously suffering from real ailments-real symptoms-but neither the study doctors nor their own physicians could determine a cause. These patients had what doctors call "medically unexplained symptoms."

n.o.body likes not knowing, but doctors, perhaps, find this state of being even more intolerable because it utterly thwarts their ability to alleviate suffering, which is often the fundamental motivation for their entire career. But a doctor's discomfort in the face of the unexplained can lead them astray. Rather than accepting a patient's symptoms as real, but unexplained, physicians all too often either dismiss the symptoms as unreal ("all in your head") or, alternatively, read too much into scanty evidence in an effort to banish uncertainty with a crisp diagnosis. Neither response serves patients.

We can see the results of both types of responses in Will Hammer's experience.

"My doctors would let me list my symptoms and they'd send me off for a bunch of tests and when they all came back normal they were like 'What are you still doing here? We can't find anything wrong with you so it must be all in your head.' But these pains I have, this fatigue, this confusion, it's not in my head. It's in my body. I needed someone to help me with what was happening to me. That's when I found Dr. Gaito."

By dismissing his symptoms, Will's doctors, in a very real sense, abandoned him. He didn't have a disease they knew about and so he wasn't really sick. This left him vulnerable to a wide range of other pract.i.tioners-both traditional and "alternative"-who offer a sympathetic ear, a ready explanation for symptoms, and a confident plan of treatment. These are precisely what Dr. Gaito provided. She listened to Will, validated his feelings, and offered him a compelling narrative of his symptoms-a seemingly solid and substantial diagnosis.

"Dr. Gaito believes that this is chronic Lyme disease," he told me. "She's been treating me for it, and while I'm not all better, I shudder to think what I'd be like if I hadn't been taking antibiotics all these years. No, I think I'll probably end up taking antibiotics for the rest of my life and if that's what it takes, I'm willing."

Escape from the Phantom Will's story is the result of a tangled web of factors, at the heart of which lie the unusually complicated and uncertain diagnostic tests used-or not not used-for patients with Lyme disease. But whereas Dr. Gaito and Will remain solidly wedded to their belief in the phantom "chronic Lyme disease," Carol Ann eventually escaped this psychological and medical dead end. used-for patients with Lyme disease. But whereas Dr. Gaito and Will remain solidly wedded to their belief in the phantom "chronic Lyme disease," Carol Ann eventually escaped this psychological and medical dead end.

For weeks Carol Ann took the medicine prescribed by Dr. Davidson. Nausea continued to plague her, but she persisted. Then her symptoms, which at first seemed to be getting better, slowly returned. Davidson changed her to a different dose of the antibiotic and when that didn't help, to yet another one. With each change, Carol Ann would start to feel a little better, but it never lasted. After months of this Carol Ann returned to Davidson's office-frustrated and sick. All of her symptoms had come back and the new medicine was making her feel just as nauseated as the old one had. At that point she'd been on antibiotics for over six months and she was-if anything-worse off than she had been when she first came to see him.

"I'm only sixty and I feel like an old woman," she told him. "What is wrong with me? If this is chronic Lyme disease, why am I getting worse?" It's an interesting question, he told her. If this were a persistent infection, he would expect her to get better. So-and he paused-maybe this isn't chronic Lyme disease after all. Perhaps this is something else. He encouraged her to return to her primary care physician. Perhaps he could help her. Davidson only took care of patients with Lyme disease. He had done all he could.

Discouraged and depressed, she agreed. Her internist referred her to a new rheumatologist and finally, nearly two years after her symptoms started, Carol Ann walked into the office of Dr. Linda Bockenstedt at Yale School of Medicine. As she sat in the dreary waiting room, Carol Ann wondered if she'd made a mistake. This looked more like a clinic than a regular doctor's office. There were nearly two dozen patients waiting for any one of a whole string of doctors whose names she'd seen on the door as she came in. Finally she was taken to a small, brightly lit exam room. There were no pictures on the walls, no personal items on the desk. It was as cool and impersonal as a room in a chain hotel.

That chill left the room the moment Bockenstedt entered. She was tall with light hair and warm brown eyes. After introducing herself, she sat on a metal chair and, looking Carol Ann in the eye, she asked her why she had come. And then she listened-without interrupting-as Carol Ann went through her entire story. She described how she was initially diagnosed with Lyme disease, and the crazy symptoms that followed. She recounted the long unsuccessful treatment with antibiotics and the terrible toll it had taken on her stomach and her body. Now during the day she could barely move her arms, her shoulders hurt, and at night her hips and knees throbbed, making sleep almost impossible. She was tired; she could barely concentrate. Her memory was shot. She felt irritable and lost her temper frequently. Bockenstedt took notes as Carol Ann spoke, and when she was done asked a few more questions to help her pa.r.s.e the puzzling symptoms.

Bockenstedt then examined Carol Ann carefully, paying particular attention to her hurting joints. Her neck and shoulders were tender to the touch and too stiff to move fully. Her hands, the joints in which are most frequently involved in rheumatoid arthritis and lupus, were supple and without pain. During the examination Carol Ann's hips and knees were pain-free and had a normal range of motion but she complained that they hurt at night and that in the morning they were so stiff it was hard to get out of bed. The rest of the exam was unremarkable.

By the end of the visit, Bockenstedt focused on three possible diagnoses. First-and most concerning-was a disease not of the joints but of the blood vessels, a disease known as giant cell arteritis. This disease is most common in women over fifty and attacks large blood vessels in the body. Untreated, it can cause blindness and strokes. The most common symptoms are fatigue, weight loss, and body aches-which Carol Ann had-as well as headaches and jaw pain-which she didn't have. Still, it wasn't a disease that Bockenstedt wanted to risk missing.

Another possibility, and the one Bockenstedt thought most likely, was a common but poorly understood disease of the muscles and joints known as polymyalgia rheumatica, abbreviated as PMR. This disease frequently causes stiffness of the neck, shoulders, and hip joints as well as fatigue and sometimes fevers. One of the most interesting aspects of PMR is that it appears full blown out of the blue. Patients often say that they feel like they came down with a flulike illness that never went away.

Rheumatoid arthritis was the third possibility-her symptoms weren't cla.s.sic, but if not treated, this type of arthritis can cause permanent injury to the joints.

Bockenstedt explained her thinking to Carol Ann and sent her to the lab to look for evidence of any of these joint diseases and to test again for Lyme disease. She also ordered X-rays of her shoulders, which would reveal evidence of rheumatoid arthritis joint damage if it existed.

Two weeks later, Carol Ann was again sitting in Bockenstedt's exam room. Bockenstedt didn't waste any time: she was very confident that Carol Ann had polymyalgia rheumatica. The X-rays ruled out rheumatoid arthritis and the blood work showed no signs of a bacterial infection-by the Lyme bacteria or any other type of bacteria.

Somewhat ironically, there is no test for polymyalgia rheumatica. Instead, tests are used to rule out other possible candidates, and a diagnosis is based on these tests and the patient's constellation of symptoms. Bockenstedt explained why she felt the evidence for PMR was compelling. Carol Ann had a cla.s.sic presentation of the disease. To begin with, women are four times more likely to get PMR than men. Carol Ann was older than fifty-the age group at highest risk of the disease (one of every two hundred women over fifty get PMR). Her symptoms came on suddenly and felt like an infection. Her pain was located primarily in the large supporting joints of the body-the shoulder and neck, hips, and knees. The spine and the smaller joints of the arms and feet are typically spared. And, of course, the X-rays and blood tests showed no signs of other rheumatologic disease or infection.

Carol Ann took in all of the information without speaking. If this was really true, then Dr. Davidson had been wrong and she had endured those months of nausea needlessly. She wasn't quite ready to trust this new diagnosis, even though she trusted Bockenstedt. She had also trusted Davidson. Plus she had read on the Internet that prednisone-the medication that Bockenstedt was suggesting-could worsen a hidden infection, if she had one.

"So you really don't think I have chronic Lyme disease?" Carol Ann asked.

Bockenstedt paused.

This was dangerous territory. Bockenstedt knew from bitter firsthand experience that the "Lyme literate"-whether physicians or patients-could be ferocious in their attacks on doctors who were skeptical about this disease. She had trained at Yale and watched in 2000 as the "literate" protested outside Allen Steere's lab. They shouted and carried signs accusing the onetime hero of Lyme disease of being a killer and a monster. He had been heckled and had even gotten death threats. Why? Because he agreed with the data. He stated publicly that no evidence supported the administration of repeated courses of antibiotics following an initial treatment for Lyme disease. And just last year when the Infectious Diseases Society of America had come out against the treatment of Lyme with months of antibiotics, ILADS charged the members with being in the pockets of the insurance companies who didn't care about the patients and simply didn't want to pay. (Bockenstedt's vivid memories of the ad hominem attacks even made her reluctant to partic.i.p.ate in my writing of this book, though in the end her commitment to educating the public won out.) Bockenstedt therefore weighed her words to Carol Ann very carefully.

There was no evidence that Carol Ann had Lyme disease right now, she told her. None of the tests she had were positive by CDC criteria. In reviewing DeVries's records, Bockenstedt noted that in the first two tests, done by Dr. Davidson, a couple of bands in the Western blot had been positive but not the five recommended by the CDC. And in the tests Bockenstedt had run, none of the bands had "lit up." Carol Ann's symptoms and the weight of the tests all convincingly pointed to polymyalgia rheumatica as the source of her ongoing symptoms.

"No," she told Carol Ann. "I don't think you have chronic Lyme disease. You've been through a lot, I know. But I think you can trust this diagnosis."

Any doubts Carol Ann might have harbored vanished within days of starting the treatment. The prednisone worked quickly and soon her joints were pain-free. After two years of insomnia, she finally slept through the night. Her flulike feelings disappeared. She could think, she could concentrate, she could remember. She felt like a new woman.

That was four years ago. Carol Ann stayed on prednisone for just over a year, tapering her dose slowly at the end as Bockenstedt recommended to let her body adjust. Since then she's had one flare-up of the old symptoms, but a week or so of prednisone tamed the pain and eased the stiffness.

So, did Carol Ann ever ever have Lyme disease? Probably not, Bockenstedt told me, but it's impossible to know for sure. Certainly by the time she'd come to Bockenstedt's office she had no evidence of the disease. Her painful joints weren't swollen-as they usually are in a Lyme-related arthritis. And none of the test results that Carol Ann had gotten for Lyme reached the level of confidence required by the CDC guidelines. It could be that Carol Ann was one of those in whom the Lyme bacteria were killed before her antibody defenses formed. Or maybe the rash that the ER doctor based his diagnosis on was a single hive, left over from her attack earlier in the week. Bockenstedt strongly suspects that, in fact, Carol Ann had suffered from PMR all along, but she is careful to acknowledge she can't know for sure. Which is exactly how real diagnosis of real disease often works. have Lyme disease? Probably not, Bockenstedt told me, but it's impossible to know for sure. Certainly by the time she'd come to Bockenstedt's office she had no evidence of the disease. Her painful joints weren't swollen-as they usually are in a Lyme-related arthritis. And none of the test results that Carol Ann had gotten for Lyme reached the level of confidence required by the CDC guidelines. It could be that Carol Ann was one of those in whom the Lyme bacteria were killed before her antibody defenses formed. Or maybe the rash that the ER doctor based his diagnosis on was a single hive, left over from her attack earlier in the week. Bockenstedt strongly suspects that, in fact, Carol Ann had suffered from PMR all along, but she is careful to acknowledge she can't know for sure. Which is exactly how real diagnosis of real disease often works.

We have tools that are essential in the making of a diagnosis. You have the history. You have the physical exam. You have tests. Ultimately you have treatment. All pieces of the puzzle, all clues leading to the final diagnosis. Says Bockenstedt, "If I had treated Carol Ann with steroids and she hadn't gotten better, I'd have to wonder-is this really what she has?" But the steroids did work-nearly perfectly. And so while Carol Ann had a good story for Lyme disease, her physical exam wasn't consistent with that disease, her testing wasn't consistent with that disease, and the treatment didn't help her. "Seeing all this, I don't understand how anyone could argue that Lyme disease is what was causing her all that pain," Dr. Bockenstedt concludes.

Testing has changed how medicine is practiced. Doctors can now be far more certain of a given diagnosis with the help of tests than ever before in the long history of medicine. But tests don't make a diagnosis-thinking does. Better tests certainly make for better thinking and Lyme disease sure could use a better test. Until there is one, and possibly long afterward, concerns about missed diagnoses of Lyme and late diagnoses of Lyme will continue to be a source of confusion and contention for doctors and patients alike.