Dr. Cameron: I appreciate the opportunity to speak today. Lyme disease is such an important topic that to have a panel like this precipitated by an Attorney General is great but it would be nice to have panels like this representing both perspectives at the major conferences. There was one at the IDSA once but there are so many associations where we need representations like Dr. Steere, Dr. Wormser and from ILADS really present the issues. And it would be great to really spell these issues out. I’m speaking as Dr. Daniel Cameron as a physician for patients who’ve been denied treatment based on the ideasof the IDSA guidelines. I’m speaking as President of the ILADS which is a multidisciplinary medical society dedicated to Lyme disease, and also I’m an epidemiologist so I’ve been writing considerably, I worked with a group to do the ILADS guidelines. We could spend considerable amount of time going line by line with the ILADS guidelines and saying, "Well what about this line? Is this process right?" But we’re focus today on really the IDSA guidelines in 2006.

Many of the thoughts that I’m talking about today are from the series of analytical reviews that have been published: The Consequences of Treatment Delay, Insufficient Evidence to Deny Antibiotics to Chronic Lyme Disease Patients, clinical trials, the four clinical trials that we’re talking about today, they validate the severity of the symptoms, and the generalizability issues of these trials.

So there’s a number statements in the IDSA guidelines that I have concerns with and others have concerns with including; that laboratory test are required to confirm the diagnosis and often it’s a two-tier with an ELISA and a western blot, prophylaxis where you take two doxycycline pills at the time of the tick, short term treatments of 14 to 21 days are fine. You know there’s always exceptions; treatment modality restrictions. And so there’s a number of statements beyond just this list.

So I’d like to go back to guidelines is that guidelines you know are not exactly what we do in our practice. When I go to my office on Monday and see patients there is something I always have to do with each individual patient. But the founder of the evidence based medicine, Dr. Sackett

tried to remind anybody when they get around and writing practice treatment guidelines to integrate the best research but also make sure that it includes clinical expertise and patient values. I’m going to look at that goal of Dr. Sackett.

First goal I want to do is talk about the evidence. And if we look at the trials, there’s 4 major trials in America funded by the National Institute on Health, that none of them made the sample size except for the Krupp study. The seropositive study by Klempner had hoped to get to 194 - 70. The seronegative part of Klempner hoped to get to 66 got to 45. Fallon had hoped to get to 45, ended up with 37. And Krupp had reached their goals. So whenever you have insignificant findings and you don’t reach the sample size there’s plenty of problems in trying to interpret that data and trying to say that antibiotic therapy has not been proven to be effective.

The other problem that happens in trials is that heterogeneity occurs. In those trials the Krupp study said that fatigue was better in the treatment group, 64% versus the placebo, 18. The Fallon had the same findings but when you look at a short form 36 for some reason there was no benefit and so you get 2 different conclusions and that’s hard to make strong, high quality, high energy guidelines and conclude that antibiotic therapy is not proven to be effective.

The next topic that happens is that - concern generalizability. The people that happened to enroll in the Klempner study, even though they try to get a broad representation. They were sick for 4.7 years; they had had 3 previous courses of treatments, so they were already in the treatment failure group. The Fallon study they were sick for an average of 9 years and there was a 2-year treatment delay. And when you look at the Fallon study 1% of the more than 3,000 people that they screened met the eligibility requirements, so therefore 99% never met those requirements. And so you can conclude that if you take too strict of an interpretation of those trials you might limit options for chronic Lyme disease patients. So again it’s difficult to conclude that antibiotics are not effective if you look at that evidence in those trials. And as I pointed out that the Fallon study those people, those 37 enrolling were sick an average of 2 years before they even got in the study. And that’s a particular type of patient that’s often more difficult and we don’t like those kinds of cases. We like them come in with the rash get treated, get better, get them to the office you know 1 follow-up, 2 follow-ups and then they’re done. So, this is the

particular type that’s very important. I had done a case control study of people with 5 bands on a western blot, and of those who had treatment delay, it was a 1.8 year delay and so this is a confounder that doesn’t seem to get mentioned in the IDSA guidelines and should be taken into account in clinical practice and it would be nice to include that.

If one looks at the evidence of this recommendation of 2 doxycycline pills, there’s a problem because nobody got studied longer than 6 weeks. They looked for objective evidence like a rash but they didn’t look for subjective symptoms which are important to patients. They didn’t look at psychiatric issues neurologic issues of this encephalopathy type. And when you have only one study and there was only about 8 rashes in one group – one group rash in the other group. It isn’t the type of study that’s really well designed that we can come up with strong recommendation to take 2 doxycycline pills. And so one should really put a lot of cautions on that type of recommendation.

Lastly, in terms of the evidence, is that there were a number of drugs and treatments that were on the not recommended list. Now that was based on level 3 evidence, which is expert opinion. The problem with level 3 is it – they didn’t get a chance on the guidelines to go to each line by line denial. It just said, "Well there’s no infection. There’s no persistent infection, let’s just deny them." Very few of them were worked out in detail and so whenever you start taking things off the list, off the table like prolonged treatment, combination treatment, you really should have a lot more of a developed topic on that subject before you start taking it away without giving clinical discretion, clinical options.

So as far as evidence goes is that going be important for this panel to not only look at what are the recommendations, but how are they graded. So I gave 2 examples. One example is that they took the antibiotic therapy for individuals with symptoms beyond 6 weeks and said, "We’ll give it an E1." E1 means they strongly feel against it. And it’s based on level 1. And for other antimicrobials, it’s also E3. That means they’re strongly against it but this is clinical judgment which is the weakest.

So if you look at how the IDSA scored it is that if you look at level 1 that says, "Evidence from at least one properly randomized, controlled study." Now there are a lot of things in medicine where you have thousands of people who get enrolled in order to make some big decisions, major decisions, landmark studies. So when the studies are all under a hundred they’re not really as big as you’d like and also when you have limitations heterogeneity, confounders of being 2 year delayed. If you have 4.7 year delays, 9 year delays, there’s all kinds of concerns that that may not be a properly randomized study. It’s a study. We learn something, but it’s more of exploratory level rather than a definitive landmark stuff. And so I would hardly ever grade anything 1. Also, one study of 2 doxycycline pills, one could say is that properly labeled? Could we call that a 1? So, one could go back to that scoring system and say, "Well perhaps I should call that a B2." You know, moderate in favor of treatment some people use C which is they can’t make an opinion but still level 2. Also for other antimicrobials, I would suggest moderately in favor. Of course each drug has its own scoring if you wanted to really get specific. And there’s moderate evidence, there’s more case reports, case series those types of things, for some of those. Now not everything on that list but each one has each own characteristic. Should you use Bicillin? Years ago they did a double-blind randomized study of Bicillin for Lyme in 25 years ago and Bicillin worked better than placebo.

If one looks at the second goal of Dr. Sackett, it’s clinical expertise. So even though there’s a statement that chronic Lyme is nothing more than aches and pains of daily living. If you look at those 4 trials, the Klempner, Krupp and Fallon, there’s 22 separate validated instruments that are used in many diseases: Short-Form 36; fatigue severity scale, fibromyalgia scale cognitive scales and are all measures of quality of life, fatigue, role function, pain, psychopathology. They’re severe not just from a normal population but they’re as bad as many of the major chronic diseases that are in America.

So even Klempner wrote that "The deficits in physical health as measured by short form 36 were equivalent to those observed in patients with congestive heart failure or osteoarthritis." And Fallonin his paper wrote, "Compared with published samples, reports of pain were similar to those of postsurgery patients, fatigue was similar to that of patients with multiple sclerosis and

limitations of physical functioning were comparable with those of patients with congestive heart failure."

So if you look at one of the measures, Short-Form ,36 the goal in life is to be at least average which is 50. Zero is the low, 100 is the high. So in the Fallon and Klempner the physical component of health was 33 and 37, quite severe. They are so severe that if you look at diabetes for Short-Form 36 its 42, cancer 41, and depression 45, and so it shows how bad they are from a quality of life which is of major marker in America of where we really want to be. We want to get all these patients up to a 50 who have chronic Lyme. We don’t want to have a chronic disease population.

And if we look at the mental component, again we want it to be 50. Fallon had 39 on average Klempner had 43 and the only one that scored the lowest was depression but that makes some sense because mental health and depression, they kind of make sense in an intuitive sense. But look at the Fallon study, we’re almost there and he was recruiting people with neurologic cognitive problems.

Now there’s a second statement in the IDSA guidelines that chronic Lyme may not exist or it doesn’t exist. Essentially is that the fact they were able to enroll people in the clinical trials sponsored by the NIH, supports that they are out there. But also the two best studies, two best cohort studies; a population based study in Massachusetts 34% were sick, 6 - 2.2 years later - great study. Generally, we can look at population-based as a representative. Also there was a consecutive case series in Westchester, where 62% were sick, and these types of numbers show that they are out there. Now you can disagree on whether you should treat or not but we have a public health problem if that many people are sick.

Another statement when it comes to clinical expertise is well should we throw anything out of the 40 states that they don’t have Lyme because there’s not as much Lyme there’s not as many ticks out in those other 40 states. The problem there is that there’s a growing number of ticks spreading through the country. Birds are spreading them to Canada. Those numbers that we’re using are fairly old. It’s under reporting so that the denominator of how many we really have in

the Midwest is really not known. So to use the Bayes theorem or use the predictive pretest probability is really a little bit shaky. The other thing is that type of statement doesn’t take into a fact that a doctor with clinical expertise can tell somebody with aches and somebody who’s sick. So if you find somebody who’s that sick with that much pain, that bad a quality of life, you’re going to consider Lyme and you’re going to rule out a lot of other conditions, and follow them up. And so you can’t just take them – just a random number without taking into account that the doctor actually uses their own judgment.

One more statement for clinical expertise is that the guidelines say that once having objective evidence of Lyme is that, "Must be as conditioned Sine qua non." Now what that means is that if you look at the dictionary, Websters dictionary is absolutely indispensable or essential. There’s not anything in life besides death and taxes that are indispensable and absolute. So when you get into this kind of statement, you know it’s a nice proposal to have chronic Lyme meet that definition but not everybody is going to have rash. In fact, if they don’t have a rash, sometimes they get lost in the system. Sometimes they took 2 doxycycline and didn’t get a positive test, didn’t get a two-tier test and all of sudden they don’t have that object evidence and so it’s premature to be taking this type of rigid statements. I’d suggest a little bit more flexibility. Koopman, who’s from Stanford, wrote in a written submission a paper that said we should have open minded, humble methodologic, rigorous way to look at the patients rather than jump too quickly to this of a statement.

Now the third section, the last section that I want to talk about is Dr. Sackett’s goal of patient values. Now they do say in the guidelines that they consider the risk of antimicrobial therapy, complications with I.V. catheters, inconvenience of prolonged therapies, the potential use of antibiotics on resistant organisms, and the economic cost of treatment. And those are all good and all guidelines try to include those, but there’s some risks that were not included, which is no mention that 34% or 62% were sick on long term in those studies. No mention of how long the Klempner and Fallon people were sick you know 4.7 to 9 years. That would have been nice to know when you put the risk and benefit together. Also no mention of severity of symptoms; it was dismissed, no mention that that there is this frustrating problem with people enrolling in the trials. Also cost of being sick wasn’t considered. Because this paper came out in 2006 and it

wouldn’t have been available for consideration by the IDSA but this was in a Maryland database put out by Emerging Infectious Disease and it’s out of– it was a co-author at the CDC and they were able to show that people who had chronic Lyme and they call it clinically defined in their database, they weren’t really getting most of them - so it was $16,199 per year over 16,000 per year per person in the database that had a Lyme disease diagnosis. Now it wasn’t the doctor treating Lyme, it was mostly and in fact 95% was indirect medical cost, non-medical costs and productivity loss. So the authors wrote that if you look how many cases are reported to the CDC which is over 23,000 cases that equates to $203 million for the society, I mean for the American society alone. And if you figure that 10 times more cases are actually around that ever make it to the CDC 10 times 200 million turns out to be a $2 billion annual cost which was something that’s you know it’s a public health problem when you have $2 billion cost over here you have resist organisms over here balancing, those two is still worth throwing into the guidelines and considering it.

The other thing and I’m not going to go into the details, it might be touched on later is there’s different responses to treatment. There are some people who do well no matter how long they’ve been sick and some don’t. There’s co-infections, you know the most famous is that Babesia makes Lyme disease more difficult to treat. There’s co-morbidities - people have diabetes, people have depression, people have multiple sclerosis and Lyme on top. Or they might have multiple sclerosis and it turned to be Lyme but there’s strain vulnerability that Lorraine just talked about there’s quality of life tradeoffs and there’s a degree of functional impairment that I talked about. So patients need a chance to self govern. They need to know the facts, therapeutic choices, benefits and risk andit’s our responsibility as a doctor at least give more information and you’d to have in the guidelines in the 2006 guidelines if it’s revised so that the doctor at least can know what’s out there.

So what I would look at as we go through any of the sentences, the recommendations, the conclusions is that I’d like to go through a checklist. Was the research evidence looked at and throw in the limitations if there are some? Was clinical expertise allowed? Clinical expertise was allowed in the guidelines only in the disclaimer section. It was not integrated into the paper, it wasn’t integrated when they interpret test. It wasn’t integrated in treatment. In the disclaimer

section discounts the value of me as a doctor and other doctors out there making those clinical decisions.

Dr. Baker: Dr. Cameron, I’m going to have you sum up, you’ve gone past your time.

Dr. Cameron: Okay. And the last one is to make sure patient values are included. The following is a list of contested recommendations and flaws and that’s included for you to read rather than me discuss it. And I appreciate the opportunity to use my full 20 minutes, so much.

Dr. Carol Baker: Thank you very much for your testimony. I’ll begin with a question about your practice. I know that you’re an Internist. It’s a two-part question with respect to Lyme disease do you take care of acute and acute disseminated patients as well as persistent infections and then part two of question is do you do internal, general internal medicine in your patient practice?

Dr. Cameron: I’ve been in primary care since 1987 and so I have a general practice and you know had a lot of HMOs at the beginning and that’s droped some. I have a Medicare practice with all of the things and bumps that happened in Medicare. In Lyme we see people that day or the next day of when they call. So we have a broad-base of early, very rewarding patients that do great. So I’m talking today on those who either don’t get treated on a timely basis or when I have them come back in 30 days, aren’t doing as well I would like. So I have a very broad practice.

Dr. Carol Baker: Thank you very much. Now are there questions from the panel, Dr. Medoff will begin.

Dr. Medoff: I’ll ask you a similar question which I did before. Is it your feeling that the guidelines should be just junked or do you want just a few maybes added?

Dr. Cameron: Well I feel that there’s a lot of work that’s gone into those guidelines but I would recommend that even if any statements are made that each time, you know like they’re not sick –

or that it doesn’t exist, that full information be provided that they you know give treatment options when it’s just based on clinical expertise.

Dr. Medoff: So you’re saying a few maybes?

Dr. Cameron: I think there’s quite a few if you go line by line. I only gave a couple of examples. There’s quite a few different places where if you use that Sackett, of make sure the evidence is reflected, patient values, and expertise that you’ll find using that criteria that there’s a lot of changes that should take place to make that a better document.

Dr. Medoff: Do you find yourself using any of the guidelines in your practice?

Dr: Cameron: I might use some of the guidelines for early Lyme. For co-infections, there’s a lot of work done in describing Ehrlichia, Babesia, and so I still read those guidelines and get to know them.

Dr. Medoff: You find them useful.

Dr. Cameron: I don’t find that they’re useful other than because mainly because they don’t give the balance you know, of the argument, the balance of the evidence. I already know the other evidence because I’ve been working with it so much but I’m just concerned with those who don’t have the time in a busy practice to get those limitations in there, get the patients perspective thrown in there and that’s why I get nervous when I read a document that’s so limited.

Dr. Carol Baker: Other questions. Dr. Sanders?

Dr. Sanders: I’d like to thank you for the submission from ILADS. It was extremely well organized and very helpful. So thank you for not only for your talk today but the material that you submitted. And I would hate to fight against you in an editorial board in a journal, very precise arguments. Going to your sample size argument on the 4 studies, what’s your understanding as to why those studies had trouble meeting sample size? What was the problem?

Dr. Cameron: I think that part of it was the eligibility requirement because with the Fallon trial, you had to have 5 bands on a western blot out of 10 which is a certain type of test. You had to have previous IV already so that you have already had to have been a treatment failure. You also have to be willing to be in a placebo study. You know, known cognitive problems, poor short form 36 quality of life, pain scales a mess, psychopathology is a mess and you’re saying "hey here’s a study you get the treatment, you get the placebo with a poor quality of life and being followed for 6 months without any treatment?" If you’re in the placebo arm, it gets challenging to have someone sign on the dotted line. It is very hard to mimic actual practice with a placebo study. I did my own clinical trial and found trying to capture these issues was very difficult even though I’m an epidemiologist. The logistical issues are monumental. And so when I published my paper, I could only write it as an exploratory paper and just list off all kinds of flaws, limitations, challenges. With the review articles I did the same thing, it is hard to do a study. I can tell you it’s hard to do a study and those studies reflect the complexity at the NIH of trying to solve these problems. For now, you know expert judgment and patient values are what we’re stuck with. Fortunately doctors still try to do that it’s just that the IDSA guidelines doesn’t give enough flexibility.

Dr. Carol Baker: Dr. Charini?

Dr. Charini: Yes, thank you Cameron. Just to follow up on Dr. Medoff’s question. What would you see then as the ultimate satisfactory outcome of this discussion in this panel? Would it be a rewriting of the guidelines, or for example, listening to what you’re saying, it sounds like there are certain key issues that could be incorporated into the guidelines? For example the reanalysis of the grading of the evidence that was done and taking into account the balancing of patient value issues, that you mentioned cost of treatment and all that. If the guidelines were amended to include those issues and perhaps some expansion on the discussion of, and controversy surrounding chronic Lyme diseases, would that be something that you would consider an improvement on the guidelines or as Dr. Medoff said should they be just completely re-written?

Dr. Cameron: I think if you took those items, that checklist, and applied it to present guidelines, it could be amended in a number of locations. When Dr. Wormser and his colleagues wrote options, the main options were with regard to a tick bite. So if he carefully followed that template of what to do, should you treat with2 doxycycline or should you wait and should you watchful wait, you know those types of things, what are the benefits so there on a major tasks you should take that template and redo it so that the really controversial subjects like treating beyond 30 days or treating the chronic symptoms, taking that giving the options would be helpful. Giving more information that I have presented in the slides and what we’ll talk about later in the day will help a doctor be informed. Now that doesn’t mean that IDSA has to only give one opinion and change it on everything but they should at least give the doctor who’s reading it that option. And they should get clinical judgment out of the disclaimer and build it somehow into some of these arguments.

Dr. Carol Baker: I think Dr. Sanders has a follow-up question.

Dr. Sanders: Yes, yes thank you. Going back to the sample size question and then to follow that up on the suggested evidence suggested changes to the evidence grade. I understand you’re reasoning from going from a 1 to a 2 questioning whether or not the randomized control trials had met that level of evidence. But going from an E to a B on the recommendation. Are you suggesting that the, that you could look at the results of those trials and come up with a completely different recommendation than the IDSA the guidelines came from even varying the level of evidence?

Dr. Cameron: I think if you, 2 is reasonable based on the evidence but there’s going to be different viewpoints on suggestion of treatment; a C means. It’s really not as clear a statement as supporting treatment or not supporting treatment. So that if you know but certainly a clear A absolutely treatment or a clear E of absolutely not is hard to support based on the trials so that if let’s say if you change it like to a C or just say that we recommend a D against and there’s another group of doctors who say a B based on their interpretation, that gives more information to the reader and I think that’s why it’s going be a fine tuning on a few of those.

Dr. Sanders: I guess my confusion is specifically that if the IDSA guidelines, based on those clinical trials, had made recommendations against treatment calling the E level recommendation.

Dr. Cameron: E and then, right E and against it level 1 E and I would say that it’s got to be level 2 and at least a C which is a draw. You know because I have – I feel that there’s some value in some of those studies. I’m an advocate that there’s some value and needs some exploratory studies and the case series and my own practice to support without the level 1, but that’s, I’m giving that perspective. The panel will have to decide what’s they’re read on whether they want to call it a B, C, or D.

Dr. Carol Baker: Thank you very much Dr. Cameron.