Depression before and after cardiac surgery: do all patients respond the same?

OBJECTIVE: To characterize the prevalence, incidence, and risk factors for depression before and after cardiac surgery.
METHODS: Patients awaiting nonemergent surgery (N = 436), completed the Patient Health Questionnaire-9 for depression, as well as the International Physical Activity Questionnaire short version preoperatively (Q1, n = 436) and at hospital discharge (Q2, n = 374). At baseline patients were categorized depression "naïve," "at risk," or "depressed." At each interval patients were identified as "not depressed" (Patient Health Questionnaire-9 score 0-3), "possibly depressed" (score 4-9), or "depressed" (score >9). Multivariate logistic regression analysis identified independent baseline and perioperative variables associated with depression.
RESULTS: Depression prevalence at Q1 and Q2 was 23.6% and 37.7%, respectively (P < .001). The incidence of new depression at discharge was 29.2%. Multivariate logistic regression identified independent variables associated with depression: preoperative left ventricular ejection fraction 30% to 49% (Q1: odds ratio [OR], 1.81; 95% confidence interval [CI], 1.02-3.21; P = .042), left ventricular ejection fraction < 30% (Q1: OR, 2.81; 95% CI, 1.13-6.96; P = .026), physical inactivity (Q1: OR, 2.03; 95% CI; 1.26-3.28; P = .002), baseline at-risk group (Q2: OR, 2.16; 95% CI, 1.28-3.67; P = .004), baseline depressed group (Q2: OR, 7.46; 95% CI, 4.06-13.69; P < .0001), hospital length of stay >7 days (Q2: OR, 1.62; 95% CI, 1.03-2.55; P = .039).
CONCLUSIONS: Depression is prevalent in one-third of cardiac surgery patients at time of discharge. It is not associated with operative or postoperative risk factors, with the exception of prolonged hospital stay >7 days. Preoperative depression or being at risk for depression, is associated with the highest risk for postoperative depression.