Fatigue decreased     27 percent with Provigil,     but by 43 percent     with the inactive     placebo.

After a wait of more than a dozen years, the U.S. government study of modafinil (Provigil) for treatment of post-polio fatigue has been published. But wait! There's more. The University of Alberta beat the Americans and published their own study of modafinil and post-polio fatigue. Does Provigil reduce post-polio fatigue, you eagerly ask? Well, that depends on whether you read the actual study or listen to the author's interview about the study. Let's start in Canada, where cooler heads prevail. 

Researchers at the University of Alberta gave 14 polio survivors reporting moderate to severe fatigue 200 mg of Provigil or placebo twice a day for five weeks. Half took Provigil first and then the inactive placebo; the other half took placebo first. Anxiety and dry mouth were reported by 60 percent of the subjects on Provigil during the first days of treatment, so they likely knew they were on the drug. Subjects completed fatigue and sleepiness scales, plus memory and reaction time tests, before the study and at weekly intervals.

Fatigue decreased 27 percent with Provigil but by 43 percent with the inactive placebo, not exactly a ringing endorsement. However, neither this difference, nor smaller differences on the other test scores, were statistically significantly different between Provigil and placebo. The authors note that Provigil is thought to affect neurons that are responsible for wakefulness, not those in the brain stem's reticular activating system that "bore the brunt" of the damage done by the poliovirus. The authors conclude that the damaged brain activating neurons responsible for post-polio fatigue are not affected by the drug, which is why "Provigil was not effective in alleviating the symptoms of fatigue" in polio survivors. 

In the U.S. study, 33 polio survivors who had had paralytic polio and reported at least moderate fatigue received 200 mg of Provigil or placebo twice a day for six weeks, with a two-week "wash out" between drug and placebo. Subjects completed three fatigue severity scales and the Short-Form Health Survey before and after the study. There were no statistically significant differences between Provigil or placebo on any of these measures, with the American authors concluding that "Provigil was not superior to placebo in reducing self-reported fatigue or improving quality of life" in polio survivors, stating "our results validate" the Canadian study.

But, then, the U.S. authors strangely wrote that Provigil "appears ineffective in most polio survivors with PPS. We say most because, despite the overall negative results of our study, a considerable fraction (22) of the 33 subjects reported benefit." Provigil had no effect on 11 standardized tests of fatigue, health and well-being, yet 22 subjects "reported benefit"? That's like saying subjects taking a new blood pressure-lowering drug "reported benefit" even though their actual measured blood pressure didn't decrease.            

One reason 22 of the subjects "reported benefit" may have been that 13 of them reportedly guessed correctly that they were taking Provigil. What's more, 41 percent reported benefit when taking placebo, an unusually large placebo effect. The authors grudgingly admit, "The magnitude of a placebo effect on the outcome measures used in our study was not considered during the design, representing a possible shortcoming." That "possible shortcoming" invalidates 60 percent of the so-called "reported benefit." 

However, in a wire-service article headlined, "Provigil may have limited effect on some patients with post-polio fatigue," the study's lead author states that "despite the overall negative results of our study" an "after-study analysis (revealed) that nearly half of the patients in our trial improved substantially." If that were true, why wasn't the substantial improvement published in the journal article, instead of the "overall negative results"? The author also stated that a "significant proportion of polio survivors in the community may actually benefit from a low daily dose (100 mg) of Provigil." If the study data show that 400 mg had no significant effect on fatigue, why put people on only 100 mg? The author concluded, "the negative results of our study should be considered with caution. Several factors could be involved, including a high placebo effect or the faulty selection of patients."  

Indeed, consider with caution any study that has a "high placebo effect" and "overall negative results" where the author presents the results of a positive "after-study analysis" in the media, instead of publishing them in a medical journal. 

Dr. Richard Bruno is director of The Post-Polio Institute at Englewood Hospital and Medical Center. E-mail him at postpolioinfo@aol.com

---

Send a letter to the editor to comment on this story

Order this issue for $ 5.00

Search Magazine Archives:

Website Services: Degnan, Co.