All right, we can see you stifling that yawn, and we know what youre thinking: What on earth is there to say about treating panic disorder other than to use SSRIs or benzos?

Well, were up for the challenge! There have been some new approvals over the past couple of years, as well as data on some off-label treatments that you may want to try on some of your more treatment-resistant patients.

Its true that SSRIs remain the mainstay of panic treatment, with Prozac (fluoxetine), Paxil (paroxetine), and Zoloft sertraline) all officially indicated for this condition. Recently, Effexor XR (venlafaxine) won approval for panic as well, based on the results of two placebo-controlled trials that lasted 12 months each. These were fixed-dose studies, meaning that patients were assigned to several specific doses of Effexor XR (75 mg, 150 mg, and 225 mg). All three doses beat placebo, which is reassuring for those who prefer not to risk the possibility of Effexor-induced hypertension when using higher doses. (Summaries of this data are available on the Wyeth website

Lexapro (escitalopram), which has become a best-selling SSRI based on excellent marketing by Forest and a possible advantage in terms of side effects, is indicated for GAD as well as depression, so youd assume that winning the panic disorder indication would be a slam dunk. However, the FDA recently issued Forest two consecutive non-approvable letters for the panic disorder indication. According to the Forest website, the FDA was not impressed by some of the research methods used in its placebo-controlled trials. Whether Lexapro is really not effective for panic is unclear, but this news does tend to temper our enthusiasm for the youngest SSRI on the block.

To treat panic disorder, start at half the usual dose of an SSRI to minimize initial jitteriness. Adding benzos at the outset is very common clinically, and over the past few years some good studies have been published bolstering this practice (Arch Gen Psychiatry 2001; 58:681-686, J Psychopharm 2003; 17:276-82). Both studies involved adding Klonopin (clonazepam) to an SSRI and compared this to adding placebo. Using Klonopin quickens response dramatically, but after four weeks there is no difference in response rates. In both studies, patients had little problem gradually tapering off Klonopin after this short-term treatment.

Aside from SSRIs, SNRIs, benzos, and CBT (cognitive behavioral therapy), what else can we offer our patients with panic disorder? Here is a laundry list of things to try, some of them with more robust research evidence than others:

Wellbutrin (bupropion). This is a blessedly low-side-effect drug that has been unfairly maligned as ineffective or anxiety. While Wellbutrin can be overstimulating for the first several days, it definitely works over time for anxiety. One series of studies found no difference between Zoloft and Wellbutrin for the anxiety accompanying depression (J Clin Psychiatry 2001; 62:776-781), and an open-label study of Wellbutrin SR in 20 patients with panic disorder found it to be effective (Psychopharm Bull 2003; 37:66-72). It strikes us as unlikely that well ever see a large controlled trial of Wellbutrin for panic disorder, because all formulations except Wellbutrin XL are available generically, reducing the financial incentives for drug-makers to fund the necessary research.

Zyprexa (olanzapine). Two patients with panic disorder, both on Paxil, improved within days of starting Zyprexa 5 mg QD as augmentation (J Clin Psychopharm 2003; 23:100-101).

Abilify (aripiprazole). In a retrospective chart review study, a majority of patients with a variety of anxiety disorders responded to the addition of Abilify 15-30 mg QD to their SSRI (Int Clin Psychopharmacol; 2005 20:9-11).

Tricyclics. While its generally accepted that tricyclics work as well as SSRIs for panic disorder (J Clin Psych 2004;65 [suppl 5]: 24-28), most psychiatrists are loathe to start anyone on them, because of lack of experience and the fear of side effects. Recently, researchers analyzed side effects specific to imipramine over one year of maintenance treatment, and found that it did indeed produce sustained dry mouth, sweating, tachycardia, and significant weight gain (J Clin Psychopharm 2002; 22:155-61).

Beta-blockers. Many psychiatrists are accustomed to prescribing beta-blockers such as propranolol and atenolol to treat situation-specific social phobia like stage-fright, or to alleviate lithium-induced tremor. In one study, the beta-blocker pindolol was compared with placebo as augmentation of Prozac treatment in 25 patients with treatment-resistant panic disorder. Pindolol outperformed placebo robustly. The dosage of pindolol used was 2.5 mg TID (roughly equivalent to propranolol 20 mg TID), and it was well tolerated in all patients (J Clin Psychopharm 2000; 20:556-559). However, using beta-blockers as monotherapy for panic has yielded mixed results (see, for example, J Clin Psychopharm 1989; 9:22-7).

Buspirone. Unfortunately, buspirone, which is as effective as any medication for generalized anxiety disorder (GAD), doesnt work for panic disorder (Acta Psychiatr Scand 1993; 88:1-11), although one small case series found it helpful as an adjunct to benzodiazepines, which might be a nice way of avoiding the benzo dosage creep that occurs in some patients (Am J Psychiatry 1989; 146:914- 916).

Gabitril (tiagabine). Gabitril (a Cephalon product) has been knocking on the door of the antianxiety market for several years now but hasnt yet won approval for anything beyond adjunctive treatment of epilepsy. Published placebo-controlled studies for GAD have been unimpressive (J Clin Psychiatry 2005; 66:1401-1408), showing no separation from placebo on the primary measure. Nonetheless, open trials have been intriguing, particularly those that used Gabitril as adjunctive treatment for patients with anxiety disorders who were unresponsive to the initial agent. In one study, for example, 13 of 17 patients achieved a response with addon Gabitril (mean dose 13 mg QD), and 10 patients achieved remission (Ann Clin Psychiatry 2005; 17:167-172). The main side effects to watch for are dizziness, sedation, jitteriness, and tremor. See the Gabitril medication fact sheet on our web site ( for more information.

Neurontin (gabapentin). One lonely placebo-controlled trial showed no drug/placebo difference on the Panic and Agoraphobia Scale in 103 patients with panic disorder (J Clin Psychopharm 2000; 20:467-471). Nonetheless, many clinicians are convinced that Neurontin can be helpful for refractory anxiety in select patients.

Lyrica (Pregabalin). Lyrica appears to have a more promising future in psychiatry than either Gabitril or its cousin Neurontin. Three placebo-controlled studies using Lyrica for GAD have been published, all of which were positive (J Clin Psychopharm 2003; 23:240-249, J Clin Psychopharm 2004; 24:141-149, Arch Gen Psychiatry 2005; 62:1022- 1030). In fact, Lyrica compared favorably to both Xanax (alprazolam) and Ativan (lorazepam) in these studies. The best dose to shoot for appears to be 200 mg TID. Side effects are similar to those with Gabitril, namely, dizziness and sedation. It does appear to cause weight gain of about 2 kg over four weeks. While it hasnt received FDA approval for GAD (it is currently approved for treatment of neuropathic pain), it did receive the green light from Europes Committee for Medicinal Products for Human Use (CHMP), meaning that it will likely receive approval from the European Commission (Europes FDA) within the next few months. Were not aware of any good studies of Lyrica for panic disorder, but the impressive GAD data bodes well for this condition.

TCR VERDICT: Panic disorder: Think outside the SSRI/benzo box