To the Editor.—
In a recent issue of JAMA, de Groote et al1 report that only 6 (16%) of 38 patients with gram-negative sepsis had detectable plasma levels of tumor necrosis factor. They suggest that this low frequency was due to the highly transient character of the rise in the level of tumor necrosis factor, although their data do not support this view. In fact, they present a table that shows persistent elevation of plasma tumor necrosis factor levels in 3 of the 6 patients rather than a transient rise. We performed a study similar to the one reported by de Groote et al and obtained comparable results that we interpret differently.
In 60 patients admitted to the medical care unit with infections of varying severity and all at risk of developing septic shock, we measured daily tumor necrosis factor levels over a period of 1 to