It is reassuring that our results on NSAIDs and AD neuropathology1 were as surprising to Drs Mackenzie and Munoz as they were to us.
We do not feel that the number of cases was too small to reach some conclusions, although larger study numbers would have been preferable. The sample size was small because we analyzed brain tissue from longitudinally followed up, cognitively tested patients who, at autopsy, had either AD or no neuropathologic condition. The original sample was 340 individuals (170 with AD and 170 controls). Approximately 30% died during the collection period and 50% were autopsied. Three exclusion criteria were necessary: We excluded patients taking steroidal anti-inflammatory drugs to assess the effects of NSAIDs (all patients except one were taking NSAIDs for arthritis at dosages ≥1000 mg/d of Naprosyn [naproxen; Hoffman–La Roche Inc, Nutley, NJ] or an equivalent). We excluded patients and controls if they had any neuropathologic condition other than those associated with AD (particularly infarction, which increases inflammatory markers). We excluded patients with generalized sepsis, which potentiates cerebral inflammation.
Halliday G, Kril J. In reply. Arch Neurol. 2001;58(3):517-518. doi: