Cigarette smoking decreases the prolactin response to serotonergic stimulation in subgroups of alcoholics and controls

Background: The prolactin response to serotonergic stimulation has been use d as an index of central nervous system serotonin function. We evaluated th e prolactin response to d,l-fenfluramine to determine whether subtypes of a lcoholics differed in prolactin responsivity compared with nonalcoholics an d whether cigarette smoking affected prolactin response. Methods: One hundred ten healthy, abstinent men across four groups (control s [23% smokers]; alcoholics [72% smokers]; alcoholics with antisocial perso nality disorder [94% smokers]; nonalcoholic anti-socials [88% smokers]) rec eived d,l-fenfluramine (100 mg orally) in a randomized, double-blind, place bo-controlled study. Plasma prolactin levels were obtained at baseline and at half-hour intervals for 5 hr after fenfluramine/placebo administration. Plasma fenfluramine and norfenfluramine levels were obtained hourly. Results: Smokers had a blunted prolactin response to fenfluramine compared with nonsmokers without any alcoholism or antisocial personality effects. U sing a cutoff point of Delta peak prolactin < 10 ng/ml, more smokers (41/76 , 54%) had a dampened response to fenfluramine than did nonsmokers (7/34, 2 1%) [chi(2)(1) = 10.6,p < 0.003]. The percentage of low responders was grea test among smokers regardless of whether they were healthy controls, alcoho lics, or antisocial. Multiple regression revealed that three variables-(1) number of pack-years of smoking, (2) actual dosage of fenfluramine received , and (3) plasma norfenfluramine level obtained-explained 43% of the varian ce (R-2 = 0.43) in Delta prolactin area under the curve. Variables that inc luded alcoholism diagnostic status, antisocial personality diagnostic statu s, and impulsive aggressive personality, depressive, and suicidal traits fa iled to explain any additional unique variance. Conclusions: Cigarette smoking blunted the prolactin response to a pharmaco logical challenge with d,l-fenfluramine. Pharmacodynamic and pharmacokineti c factors related to smoking both appear to influence fenfluramine-induced prolactin secretion. Phenotypes of alcoholics did not differ in their prola ctin response to this serotonergic probe.