Over the summer months, I completed the home study portion of the American Board of Internal Medicine (ABIM) recertification program. Its open book format consists of five Self-Evaluation Process (SEP) modules, each with 60 general medical questions and multiple-choice answers. ABIM encourages examinees completing these modules to consult with colleagues, perform Medline searches, and use textbooks and medical journals. Not surprisingly, I ended up using primarily online clinical references.

In answering ABIM's questions, I discovered great variability in both content and ease of navigation among seemingly reliable resources. To share what I discovered about these resources, I selected four clinically challenging questions whose answers for whatever reasons were not easily accessible on line. This forced me to use several Web sites and references before arriving at what I thought was the correct answer. In retrospect, this process reminds me of the insight gained when comparing cars under similar conditions during a road test.

Road Rules

Many primary care and specialty boards offer recertification programs that are clinically relevant and cover recent advances. This was certainly the case with ABIM's SEP modules -- so much so that using the Web here felt near-identical to looking up patient care questions in real time. As in clinical practice, my concern was not only the accuracy of information, but also the timeliness and efficiency of information retrieval.

Since I might need to defend my answers to ABIM, I only consulted online references that are well known and nearly universally respected. My primary resources were the electronic versions of general medicine textbooks such as the subscription-based Harrison's Online, Scientific American Medicine (SAM) Online¹, and a dozen or so of the 38 textbooks within MD Consult. Electronic textbooks differ from one another not only in breadth, depth, and timeliness of content and associated materials, but also in the ease of navigation using tools such as search engines, tables of contents, and indexes. These resources will be discussed in detail in Part 2.

A number of questions focused on recent advances or new clinical data, so I also performed Medline searches. I consulted PubMed, although there are a number of other free and subscription-based Medline interfaces available that may have provided somewhat different search results.

I also felt confident consulting a handful of Web sites housing unique electronic information such as the National Cancer Institute's Cancernet, GeneClinics, and sites sponsored by well-known medical organizations and agencies such as Centers for Disease Control and Prevention.

Occasionally I accessed several free online references such as eMedicine and Praxis MD. I also consulted several well respected, practically-oriented clinical journals whose databases are on line, principally the American Academy of Family Physicians' American Family Physician. Several colleagues gave positive reviews of the clinical database in the subscription-based UpToDate software. I was unable to gain access to its online database for this article as it was then in beta testing, but its review in included in Part 3 of this series.

ABIM warned me that while many questions had more than one possible answer, full credit was awarded only for the one best answer. Rarely, partial credit was awarded. I disagreed with a surprising number of their answers, including several presented below.

Prophylactic Treatment For Lyme Disease

One question asked about the management of an asymptomatic 17-year-old who was bitten by a tick two days earlier while camping in Connecticut. He promptly removed and disposed of the tick.

The two most frequently quoted studies,² both published in the mid-1990s and one of which is a meta-analysis, convincingly demonstrate no clinical benefit to the general population of prophylactic treatment with antibiotics. They found the overall risk for clinical infection with B. burgdorferi in patients with a known tick bite to be very low, even in areas where Lyme Disease is endemic. Therefore, the correct answer is observation and patient education.

However, other references, particularly those from the late 1990s, give cogent reasons to consider treating selected subgroups, and a few even recommend treating all patients who inquire, citing the documented popularity among physicians for prophylactic treatment.³ Even by conservative standards, prescribing a two week course of doxycycline to a young man from Connecticut with a recent tick bite and who is concerned about contracting Lyme Disease is well-supported by the literature, although it will not earn even partial credit from ABIM.

The most complete and up to date information was found in eMedicine's chapter on Tick-borne diseases, Lyme, which explains the reasoning behind observation and cites exceptions to the no-treatment rule, all in a succinct and authoritative manner. Kudos to Jonathan Edlow, MD, for writing such a well-referenced chapter on an important subject. MD Consult includes another winner: Mandell's Principles and Practices of Infectious Diseases (2000). This textbook provides a less succinct but otherwise excellent explanation of why treatment is not necessary yet often given anyway, especially for anxious patients or if follow-up is difficult.

Performing an MD Consult search in all book references for "lyme disease and treatment" brought up seven textbooks, two of which were consulted. Unfortunately, they provided less satisfying information. Conn's Current Therapy (2000) strongly argues against treatment, as "one severe life-threatening drug reaction would be expected for every 10 cases of Lyme disease theoretically prevented by a prophylactic course of amoxicillin." This is not surprising for a penicillin, but I doubt that datum holds up for doxycycline. A second text, Rosen's Emergency Medicine (1998), states that physicians tend to treat, but recommends patient education with serial serologic testing -- an unusual (and expensive) recommendation for emergency care.

Also less satisfying, SAM's reference on Lyme Disease goes into great detail why there is no need to treat patients with tick bites, citing clinical studies and offering no exceptions. Harrison's reference on Lyme Disease says there is no need to treat unless the patient is anxious or pregnant.

All other online references accessed were essentially useless. I was surprised that CDC's Lyme Disease provided just a superficial statement recommending treatment if risks outweigh advantages. A PubMed search for "lyme disease AND preventive treatment" retrieved 24 references, only three of which were useful clinically. Searching the American Family Physician for "lyme disease" came up with several superficial articles, sometimes with contradictory recommendations.

Praxis MD's physician reference contains no real information on Lyme Disease, much less on prophylaxis. The site does contain a patient reference on Lyme Disease but it does not address the issue of prophylaxis (although it gives instructions on how to properly remove a tick). The site also links to LymeNet, whose May 2000's Diagnostic Tricks and Treatment Guidelines by Joseph Burrascano Jr., MD, recommends treatment for high risk depending on tick bite. But elsewhere on LymeNet, the information advocates that patients request prophylactic treatment from their physicians after a tick bite, citing reasons not consistent with mainstream medical thinking.

Emergency Treatment For vWD

A patient with continued bleeding after otherwise unremarkable nasal surgery is found to have a family history and lab tests suggestive of the most common inherited coagulopathy disorder, von Willebrand's disease (vWD). The question asks what treatment should be given after drawing confirmatory blood tests.

Text references differ with respect to recommended treatment. Some advocate desmopressin (DDAVP), though the strongest references suggest cryoprecipitate despite its viral risks, citing the possibility of DDAVP-induced thrombosis should the patient have type 2B, or its ineffectiveness in treating types 2A and type 3. Since this patient's post-op oozing was the first known sign of a coagulopathy, this patient has not been previously diagnosed or typed.

ABIM's correct answer is DDAVP. They do not give partial credit for fresh frozen plasma (FFP), which was cited as another option. (Although the only reasonable choice to avoid DDAVP-induced thrombosis, one hematologist told me that, unlike cyroprecipitate, FFP does not correct the coagulopathy associated with vWD.)

Textbooks do not always coincide with real world practice, and DDAVP is a well-known treatment for episodic bleeding in vWD patients, about 75% of whom have type 1. I have not yet found a colleague familiar with DDAVP-induced thrombosis in type 2B, though most emergency physicians I know use FFP to treat bleeding due to an undiagnosed coagulopathy.

Harrison's Online provided the most succinct and easily accessible information on vWD treatment, including a recommendation to avoid DDAVP in patients with an unknown type of vWD such as this case. MD Consult textbooks also provided accurate information. Its hematology text, Wintrobe's Clinical Hematology (1999) is less succinct but confirms Harrison's information. Searching all of MD Consult's textbooks for "von Willebrand" and "treatment" yielded three results, two of which -- Cecil's Textbook of Medicine and Conn's Current Therapy -- provided similar, accurate information.

No other online reference was helpful. Using the table of contents to access section 5/chapter XIII within SAM Online took me to the subsection on vWD, which did not mention treatment when the diagnosis was in question. EMedicine mentions vWD within its Acute Anemia chapter, and recommends cryoprecipitate for known vWD patients; this is the only treatment mentioned. I did not find any physician references using the Praxis MD search engine.

Air hunger?

Does a patient with borderline severe emphysema (pO₂ of 60mmHg on room air both at rest and with exercise) require supplemental oxygen when on a commercial flight, no special accommodations, or is flying not recommended due to lung disease?

My only online success with this question came by searching "lung AND air travel" on Harrison's Online. This led directly to the correct answer, within the Treatment subsection of Chronic Bronchitis. "Even with modern pressurization, cabin altitudes may reach the equivalent of 8000 ft, at which the pO₂ may fall by 25 mmHg below that found at sea level. Hence O₂ supplementation during prolonged flights should be considered for patients with sea level pO₂ values in the mid-70s."

Virtually all other search terms resulted in no hits or with unrelated material, on Harrison's and on nearly all other online references. I was even unable to search out appropriate information in any textbook within MD Consult. When I broadened my search to include all information within MD Consult's entire site, I found two pertinent articles in the Clinics of North America series, but both gave incorrect (and contradictory) information.

Diagnosing Cat-Scratch Disease

A middle-aged woman with symptoms suggestive of cat-scratch disease requires diagnosis. The correct test, as explained in Mandell's text under MD Consult, is to look for high concentration of antibodies to B. henselae on serologic tests two weeks after presentation.

The weakest information was provided by eMedicine's chapter on cat-scratch disease, which cites outdated diagnostic criteria and mentions serologic testing only in passing.

Harrison's subsection on diagnosis of cat-scratch disease and SAM Online section 7, chapter XVIII, subsection on cat-scratch fever both mention serologic testing, though without Mandell's satisfying explanation.

There was essentially nothing on any aspect of cat-scratch disease in Praxis MD, or in American Family Physician.

Disclosure: Dr. Lazoff wrote three chapters for eMedicine and submitted an article to Praxis MD.
 
Footnotes
1. SAM Online is temporarily available free of charge to the public when accessed through WebMD.
2. Shapiro ED, et al. A controlled trial of antimicrobial prophylaxis for Lyme disease after deer-tick bites. N Engl J Med 1992;327:1769-73
Warshafsky S, et al. Efficacy of antibiotic prophylaxis for prevention of Lyme disease: A meta-analysis. J Gen Intern Med 1996;11:329
3. Fix AD, Strickland GT, Grant J. Tick bites and Lyme disease in an endemic setting: problematic use of serologic testing and prophylactic antibiotic therapy. JAMA 1998; 279:206-10

continue on to Part 2: Selected References

 
 
Related resources on MCToday:

• Frequently Updated Medical References
• Online Medical Services