To the Editor.
—Lehrer et al,1 in the August issue of the Archives, attempt to show a link between vertigo and insulin resistance or hypertriglyceridemia. Unfortunately, the study suffers in several basic respects.There is no description of the patients studied and no mention is made of any possible associated comorbid states, such as the use of other medications or ethanol. Criteria for inclusion in the study not only included vertigo, but also "disequilibrium," which was poorly defined.The authors found "abnormal" insulin response curves and "elevated" triglyceride levels in many of their patients, yet the prevalence of these unquantified findings in a representative sample of their total patient population is not addressed. Dietary recommendations were temporarily associated with subjective reports of improvement, but there was no control group of untreated patients and there was no attempt to document dietary compliance. Basic follow-up data, such as weight loss, improvement