Hypervigilance as predictor of postoperative acute pain: its predictive potency compared with experimental pain sensitivity, cortisol reactivity, and affective state.

OBJECTIVES Pain hypervigilance--a strong attentional bias toward pain--is thought to accompany chronic pain and modulate pain management. Its usefulness as predisposing factor for the development and maintenance of pain has been discussed. The aim of our study was to demonstrate the predictive power of hypervigilance for the development of acute postoperative pain. METHODS Fifty-four young male patients were assessed 1 day before surgery (correction of chest malformation) on a range of psychologic predictors. These predictors included the assessment of hypervigilance (questionnaires as the Pain Catastrophizing Scale, Pain Anxiety Symptom Scale, the Pain Vigilance and Awareness Questionnaire, and the dot-probe task) and affective state, experimental pain sensitivity, and cortisol reactivity. Acute postoperative pain was assessed by ratings of pain intensity 1 week postsurgery and through the amount of analgesics [patient-controlled epidural analgesia (PCEA)] requested during the first days after surgery. RESULTS Pain intensity was significantly explained (17% explained variance) by hypervigilance, whereas PCEA performance was not (10%). Adding all other predictors led to a significant increase of explained variance (35%) for pain ratings and a nonsignificant increase (19%) for PCEA. A more parsimonious solution with only heat pain threshold added led to a significant increase in explained variance (30%) for pain intensity. Hypervigilance was only moderately correlated with the other predictors. DISCUSSION Hypervigilance proved to be a powerful predictor of subjective acute postoperative pain, but was less useful with regard to the amount of requested analgesics. The overlap with other psychologic predictors (affective state, experimental pain sensitivity, and cortisol reactivity) is sufficiently small to consider hypervigilance a promising supplement in psychologic predictor research.