Annals Internal Medicine article-you can't rely on RCTs alone to determine drug safety

In the Feb.21,2006 issue of the Annals of Internal Medicine, Ioannidis et al have an excellent article on adverse efffects (AEs).(Ioannidis,et al "Adverse effects;The more you search, the more you find", Ann Int Med 144 (4) Feb 21, 2006)

"It is almost always inappropriate to make statments about no difference in adverse rates between groups of non-significant p values....rates of adverse events that are derived from single, modest-sized trials that are not statistically significant typically do not exclude with certainty the possibility of major,clinical important differences in harm between groups"

Randomized trials are often small and designed to test efficacy in patients in which the dosing is well controlled and often patients with co-morbidities are excluded. Only after many more patients are treated do side effects become apparent.

The latest two in what could be a very long list of medications for which serious safety issued arose after RCTs were done and FDA approval issued are Tegrin (gatifloxacin) and Ketex.

The issue with gatifloxacin drug is diabetes and now it seems to be contraindicated in patients with diabetes. Serious liver problems have been noted with Ketex.

Interestingly, gatifloxacin has been associated both with hypoglycemic reactions in diabetics on treatment and hyperglycemia in patients previously not known to be diabetic.One cannot but wonder what the mechanism(s)is/are.

A recent review of gatifloxacin-hyperglycemia cases found most were associated with decreased renal function. That the blood sugar effects are not limited to just gatifloxacin is suggested by another recent review that showed approximately equal numbers of "dysglycemias" with gatifloxacin and levofloxacin.Still another review study demonstrated a greater risk of blood sugar alterations with gatifloxacin.

Even though the RCT rests on the top perch in the party-line version of the heirarchy of the evidence used in evidence based medicine, we have to fall back on the lowly case report and observational studies to alert us to serious drug side effects. We have to realize RCTs may be best to determine efficacy not the risk of adverse effects.