Dr. Roberts responds: Thanks to Dr. Peterson for his insightful comments. It probably should be no surprise that I might not always agree with Drs. Bukata and Hoffman, but I do not interpret their comments to mean that you should never ever treat a sore throat. I would offer that treatment does, in fact, provide some benefit on symptoms (albeit minimal, but even a 12-hour faster cure is welcomed by the patient).
Maybe I missed something, but treating a sore throat will lessen the chance for rheumatic fever by all the authorities that I have read. Glomerulonephritis is, of course, not affected. I do not think anybody really knows if a peritonsillar abscess will be thwarted, but it seems logical if treatment is begun soon enough. A peritonsillar abscess killed George Washington, and some of my patients have access to a similar medical safety net as did the erstwhile president, so I'll take any help I can get from antibiotics. I would be interested in seeing good data to the contrary.
To answer Dr. Peterson's questions: An antibody titer takes about two days to come back in our hospital, and I think that test is clearly a waste of time and money, so I never send them. In a previous column I wrote on diagnosing a sore throat, I specifically and emphatically eschew any diagnostic testing, especially a culture or strep screen. I also argue against overkill with expensive antibiotics. I believe we agree on these issues. I would treat in the ED as soon as I examined the patient, almost always with IM benzathine penicillin. I am all for evidence-based medicine, and I think my overall approach is largely evidenced-based, flavored with a modicum of common sense and practicality.
Maybe it's not a good argument on my part, but try telling a mother that her child has a strep throat and to tough it out, and then try getting them back in school in 24 hours with that logic. And, by the way, those arguments on preventing rheumatic fever wouldn't work with a plaintiff's attorney from Philadelphia. I will concentrate my antimicrobial educational efforts on trying to keep my residents from prescribing a $100 course of antibiotics in viral bronchitis and viral syndromes in kids where they fantasize a red eardrum. I would intuit that most physicians readily hand out prescriptions for bronchitis, a condition where the data are clear that they do not help at all for anything.
I'll take the criticism on overprescribing penicillin for some severe sore throats, and it took me four months of columns to explain that approach. A question: Can any physician honestly say that if his 20-year-old son had a temperature of 103°F, pus covering his tonsils, nodes the size of a Philadelphia cockroach, and couldn't swallow his saliva without wincing that you would tell him that you had nothing to offer, the condition is benign, and suggest some Tylenol. I suspect not. If you did, and your wife is like mine, she would certainly opt for a second opinion in a few nanoseconds.
Dr. Hoffman responds: It's with some trepidation that I disagree even a little with the always-erudite and extremely well-read Dr. Roberts, but I must admit that I was a little surprised when I read his column in EMN. I too inferred that he was recommending antibiotics for many more sore throats than just the killer case described in his response to Dr. Peterson, and I'm delighted to learn that he apparently agrees that antibiotics should be limited to the few really bad sore throats that also have the multiple associated clinical characteristics that are generally predictive of group A strep. (I'm also delighted to see Dr. Roberts state plainly that he too opposes diagnostic testing [in run-of-the-mill cases], and especially discourages the use of inaccurate and temporally useless [but nevertheless expensive!] throat cultures.)
I just recently tried representing emergency medicine, along with my colleague Richelle Cooper, on a CDC-organized panel charged with addressing the overuse of antibiotics in common upper respiratory infections. We were amazed by how much heat was generated by the topic of pharyngitis. Although our panel (and reviewers from various outside bodies such as ACEP, ACP, AAFM, and IDSA) have long since agreed about recommendations regarding URIs, otitis, bronchitis, and sinusitis, we're still arguing about pharyngitis. That's because some of us think you could make a strong case for not using antibiotics at all (in the absence of special circumstances, such as a patient with an post-surgical heart valve, an immunosuppressed child, or a documented epidemic of strep, etc), or limiting them only to patients with a very high probability of strep (three or even four out of the four Centor criteria mentioned in Jim's article), while others (some infectious disease specialists, primarily), want to do tests (typically antigen screens) on most everybody, as the basis for treatment decisions.
While cognizant of Dr. Roberts' hypothetical mother and plaintiff's attorney, I nevertheless would point out that emergency physicians face tensions created by patient desires and demands and concerns about being blamed if something goes wrong all the time, and yet we're still obligated to provide the best medicine that we can, even when it might be a bit unpopular. In the large majority of cases of pharyngitis, I personally go along with Dr. Peterson and don't prescribe antibiotics. I don't think they make much difference in duration of symptoms (especially if you provide good supportive care), and the best evidence available today does not suggest they affect the development of peritonsillar abscess. Rheumatic fever is the big issue because preventing a single case is truly important, so I'd be quick to change my mind if there were even a whiff of ARF in my community. Because there isn't, however, and hasn't been in a long time, it's difficult to make a case for using millions of doses of antibiotics to prevent a disease that no longer seems to be around (in most American non-immigrant communities).
The flip side is that overuse of antibiotics is a real problem, and even worse, it engenders a vicious cycle of even greater demands for antibiotics in the future. Dr. Roberts is right that antibiotic treatment of URIs and colds is an even bigger issue, but that doesn't mean we should be complacent about misuse in cases of pharyngitis.
Reading Dr. Peterson's letter and Dr. Roberts' response, I suspect we all agree about most of the big picture items, though we may quibble a bit on some of the details, which means for the usual case of pharyngitis: don't do throat cultures; only consider antibiotics if there is a special circumstance or the patient presents very early in the course (less than one or at most two days), and has at least three or all four of the clinical findings that suggest a greater than 50 percent risk of strep; and don't go to war over it, but try your best to convince an insistent mom (or dad or patient) not only why her child doesn't need antibiotics, but why they're a bad idea, as well as how she can get excellent results with symptomatic care alone.
Dr. Bukata responds: A young adult has been waiting to be seen for an hour in the ED with a very painful exudative tonsilitis, generalized myalgia, a fever of 103°F, and no URI symptoms. Here's what I do: I don't order tests unless the presentation is atypical, and I think it is mononucleois, in which case a CBC is done first, and, if showing a lymphocytic pattern, then a monospot). I then prescribe 10 days of 1 gm of penicillin V BID (likely costs the same as 7 days).
Why do I do what I do? Tests for strep are a waste of time; there are too many false positives (carriers) and false negatives. Penicillin will reduce some symptoms and the duration of the illness, the earlier treated the better. I have no delusions that penicillin makes a big difference, and I tell the patients so. Perhaps it will decrease the length of the illness by half a day or a day (no small amount if you've ever had one of these infections).
I do believe in the germ theory of disease. If you kill the bugs, family members are less likely to get the disease, and the patient feels better somewhat sooner. This is by no means a miracle cure, but what I feel is a reasonable approach, all in all, to a most perplexing disorder.