While sleeping pills have been around for a long time, the modern age of hypnotics began December 16, 1992, when the FDA approved Ambien (zolpidem). Sonata (zaleplon), a late bloomer, received approval in 1999.

Both Ambien and Sonata are officially classified as "non-benzodiazepines," although they do their pharmaco- slumbering magic on the same site as the benzodiazepines (BZs), namely by binding to the GABAA/ benzodiazepine/chloride channel receptor complex. However, the non-BZs bind more selectively to the Type 1 subtype of this complex, which is likely why these meds are more specifically helpful for insomnia, and not particularly helpful for either anxiety or muscle relaxation.

Let's look at how Ambien and Sonata differ from each other, and also take a peek at the soon-to-be new kid on the block, Estorra (eszopiclone).

Ambien, with a half-life of 2.5 hours, is effective both at putting patients to sleep and at increasing their total sleep time. Sonata, on the other hand, with a tiny half-life of 1 hour, gets patients to sleep but does not increase total sleep time. Ambien's superior versatility comes at a price: it causes more next-day impairment than Sonata (J Clin Psychopharm 2002; 22:576-583).

Over the years, Ambien (10 mg QHS) has been used most by patients who have both difficulty falling asleep and frequent awakenings throughout the night. Sonata (10 mg QHS) has carved out a niche for two situations: 1. Patients who can't fall asleep easily but who sleep well once they drop off, and 2. Patients who fall asleep easily but wake up in the middle of the night and need something that will allow them to wake up reasonably refreshed in the morning. Its lack of next day hangover apparently caught NASA’s eye, as it has become the official sleeping pill used by the astronauts. Rock on, Sonata.

Estorra (eszopiclone) is zeroing in on FDA approval, which it should receive toward the end of 2004. It's another non-BZ, has a longer half-life of 5 hours, and probably works similarly to Ambien. Its claim to fame will be data showing efficacy and safety over a full year of treatment (see Formulary 2003; 38:583-593 for a review) allowing us to prescribe chronic insomniacs lots of refills without any twinges of guilt.

Are non-BZs really less "addictive" than BZs? Probably, but hard data on this issue is sparser than the drug-makers would like to admit. There was a recently published postmarketing study in which researchers surveyed graduates of three addiction centers in the UK and asked them about the street values and potential to make you high of a variety of sleep meds, including BZs, non-BZs, antidepressants, and antihistamines. The BZs had the highest abuse liability, followed by the ADs and non-BZs (which were judged as equally abusable). Benadryl was singularly unrecreational, according to these experts on getting high (Addiction, 2004; 99:165), though Dr. Mick’s anecdote (below) suggests otherwise.

Of course, all these new hypnotics are expensive, and those of us inclined toward health care frugality find that meds like trazodone, Benadryl, and generic benzos work just fine for insomnia, thanks very much. Well, be prepared for an onslaught of "infotisements" from the hired guns and hired journals.

The most recent example comes from the Journal of Clinical Psychiatry, whose cover article in June 2004 was entitled "The Use of Trazodone as a Hypnotic: A Critical Review." The first author is a consultant to King Pharmaceuticals (Sonata), and the second author has consulted for Sanofi (Ambien). Given these credentials, you might predict that the article slams trazodone, a drug that has the pharmacokinetic property of competitively displacing the costly non-BZs from many psychiatrists' prescription pads. Your prediction would be accurate.

The gist of this review is that since there is little hard data on the efficacy or safety of trazodone as a hypnotic, we shouldn't be prescribing it so much. Particularly since it can cause priapism (they cite a reported 12% rate (!)--we suspect that "priapism" was confused by subjects in this study with nocturnal erections), orthostatic hypotension, and cardiac arrhythmias (three single cases reported 20 years ago in patients with significant preexisting cardiac disease).

As with most well-written industry-serving articles, nothing said is frankly inaccurate, with the main intellectual sins being those of omission and inappropriate emphasis. For example, the authors conclude that "Although trazodone benefits some aspects of sleep in patients with major depressive disorder, there is little or no evidence from systematic studies that it aids sleep in insomnia patients." Left out is the important fact that the lack of systematic studies showing trazodone's efficacy is hardly the drug's fault, but rather a reflection of an economic reality. Since trazodone is generic, no company in its right mind will spend the millions of dollars required to conduct "systematic" studies for FDA approval, especially when it is already widely used for insomnia. In fact, dozens of small scale, non-industry supported studies have already shown trazodone to be an effective and safe medication for a variety of psychiatric conditions that present with insomnia, including depression, dysthymia, alcohol withdrawal, and SSRI-induced insomnia.

A final word to the wise (and surely, the weary): Beware the Mallinckrodt reps peddling Restoril (temazepam) 7.5 mg, an "Effective Low-Dose Choice." Temazepam has been a great low cost generic sleeping pill for years, and with no active metabolites, it's considered a good choice among the BZs for insomnia in the elderly. Here's the hitch: only the 15 mg and 30 mg forms are generic; the FDA granted patent exclusivity for the 7.5 mg capsule in 1991. The sales pitch is that 7.5 mg works as well as 15 mg, but with less “concern” for side effects. They chose that word carefully--they can hardly say "fewer" side effects when the major company-funded study (Curr Med Res Opin 2004; 20:441-449) showed no difference in side effects between the two doses!

So. Difference in effectiveness or side effects? None. Cost? Restoril 7.5 mg, about $100 per month, temazepam 15 mg, $10 per month. Lesson learned about the pharmaceutical industry? Priceless.

TCR VERDICT
Non-BZs: Good, but are they worth their price?