The significance of HPA axis disturbance in panic disorder.

Agoraphobic and panic disorder patients underwent 1-mg Dexamethasone Suppression Tests (DST) before, during, and after an 8-week trial of diazepam, alprazolam, or placebo. Previously described, never-ill controls underwent similar testing. At baseline, 21 of 82 (25.6%) panic disorder and 5 of 38 (13.2%) controls were nonsuppressors. This difference grew more marked with multiple testing over a 2-month period; 18 of 44 (40.9%) panic disorder patients were nonsuppressors on at least 1 of 3 tests compared with only 5 of 35 (14.3%) controls (p = 0.006). DST results were related to severity, but not to the presence or absence, of depressive syndromes. Control for plasma dexamethasone levels left highly significant differences in postdexamethasone cortisol across diagnostic groups. Neither DST results nor plasma dexamethasone levels changed in concert with clinical change, and type of treatment had little differential effect on these measures. Nor did DST results predict subsequent course when active treatment was extended by 6 months. However, DST results during the initial 8 weeks of treatment were strongly related to relapse when medications were tapered, even though this occurred 6 months after the last DST.