We investigated psychologic influences on duodenal ulcer by examining
the effect of personality, stress, and mood, measured at diagnosis, on
subsequent ulcer healing. Stressful life events, psychopathology (ass
essed using the Minnesota Multiphasic Personality Inventory), anxiety,
depression, smoking, alcohol consumption, nonsteroidal antiinflammato
ry drug use, and serum pepsinogen I levels were determined immediately
after endoscopy showed duodenal ulcer craters in 70 patients with rec
ent onset of symptoms. Endoscopy was repeated following 6 weeks of ran
itidine therapy. Six ulcers (8.6%) persisted, and the duodenum remaine
d inflamed in an additional five cases, for a total of 16% with incomp
lete healing. The only baseline characteristic significantly associate
d with poor healing was anxiety (p = 0.03 for ulcer persistence, p = 0
.02 for incomplete healing). Being in the highest anxiety tertile was
associated with a more than fourfold elevation in the risk of incomple
te healing (p = 0.02). The association between anxiety and poor healin
g was not changed by modification of the anxiety score to eliminate ga
strointestinal symptom items or by adjustment for serum pepsinogen, se
x, or cigarette smoking. Anxiety inhibits the healing of duodenal ulce
rs treated with adequate antisecretory therapy.