Response to Letters Regarding Optimizing Therapy for Community-Acquired Pneumonia with the Goal of Rapid Resolution of Illness

To the Editor—As was pointed out by the letters by Solnick [1] and Leenders [2], and as was discussed in our commentary [3], there are several limitations of the study by Welte et al. [4] that potentially reduce the significance of their findings in their comparison of 2 treatment regimens. However, the primary emphasis of our commentary was not to address the relative efficacy of either of the 2 regimens evaluated in this study, but rather to discuss the importance of including a systematic evaluation of the speed of resolution of illness as an end point for future randomized, clinical trials. Although the time to symptom resolution was assessed in the study by Welte and colleagues by the use of a patient diary, this process needs to be better evaluated. The standard test-of-cure evaluation in most previously published trials to have compared antimicrobial regimens has usually been 7–10 days after end of therapy (and, therefore, usually 14–21 days after the initiation of therapy). This end point is often not sensitive enough to evaluate the efficacy of different treatment arms, because many infections, even when treated with marginal therapy, are improved by the end of this time period.

The trend in the treatment of most infectious diseases, especially with the availability of more-potent antimicrobial agents, is for shorter-course therapy [5, 6]. There are several potential advantages to short-course therapy in general and with regard to community-acquired pneumonia in particular. These include the potential to reduce the selection pressure for resistance and adverse events that may emerge with prolonged antibiotic exposure. Presently, there is a relative dearth of well-performed studies to evaluate the appropriate duration of therapy for most infections. Studies that include time to resolution of disease and eradication of pathogens can be useful in helping one determine the appropriate duration of therapy; however, to properly interpret the results regarding the time to resolution of disease in randomized, clinical trials, the trials need to be better designed than are the studies that have already been published.

As indicated in our commentary [3], we propose that future studies of pneumonia should not only compare monotherapy to combination therapy with regard to clinical or microbiological cure, but they should stratify patients on the basis of varying severities of illness and comorbidities. A validated symptom score to assess resolution of illness should be evaluated. Ideally, studies should be prospective and double-blind, and an aggressive attempt to identify pathogens and to determine the speed of eradication of pathogens is necessary to correlate with the clinical manifestations.

Acknowledgments

Potential conflicts of interest. T.M.F. and J.S.T. have received recent research funding from Abbott, Arpida, Bristol-Myers Squibb, Cubist, Enata, GlaxoSmithKline, Pfizer, and Wyeth; have served as consultants for Sanofi-Aventis, Bayer, GlaxoSmithKline, Ortho-McNeil, Merck, Oscient, Pfizer, and Wyeth; and have been on the speakers' bureaus for Abbott, Sanofi-Aventis, GlaxoSmithKline, Merck, Ortho McNeil, Oscient, Pfizer, Schering Plough, and Wyeth.