to the editor: The article, “Diagnosing Night Sweats,”¹ was based on a MEDLINE literature review from 1966 to July 2001. As a result, the authors' literature search missed our study² of the epidemiology of night sweats in primary care that was published in 2002. There also has been at least one other relevant epidemiologic study³ that was not mentioned in their article.¹

Based on our published and unpublished results, it appears that the classic definition of night sweats “does not describe the majority of patients who may complain of the symptom.”² Night sweats are commonly experienced by primary care patients who, for unclear reasons, rarely mention them to their primary care physician even when they become severe and bothersome. The most common causes of night sweats have yet to be elucidated and, therefore, the appropriate evaluation of patients with this symptom is unclear. Given the level of our collective ignorance, many of the recommendations made by the authors¹ seem reasonable, however.

in reply: We would like to thank Dr. Mold for his comments and appreciate the study¹ conducted by him and his colleagues on the prevalence of night sweats in primary care patients. When we conducted the research for our article and revised it, that study was not yet published. An informal survey of medical students, house officers, and senior physicians (including infectious disease specialists, oncologists, and generalists) found that tuberculosis is the first disease to come to mind when a patient reports night sweats.² The same pool of physicians and students reported that they rarely encountered night sweats as a symptom among patients with tuberculosis. In our original manuscript,³ we cited the study⁴ by Lea and Aber that discussed the descriptive epidemiology of night sweats, but that section was subsequently edited out of the final article. This study⁴ reported that out of 174 randomly selected university hospital patients, 72 reported night sweats within three months before admission. None of these patients had tuberculosis. These observations that many possible common causes of night sweats were not known served as the initial basis for our article.