Ｏｂｊｅｃｔｉｖｅ　Patients with heart valve replacement may be prone to triple low state of propofol target controlled anesthesia, ie low BIS, low MAP and low target effect-site concentration(Ce) . The present study was to establish the threshold of triple low state of propofol anesthesia associated with 30-day mortality risk in patients undergoing valve replacement.　Methods　161 adult patients with rheumatic heart disease were scheduled for elective valve replacement surgery and anesthesied with propofol via an effect-site target. Patients were divided into survival group and mortality group after 30-day follow-up according to the different prognosis and the univariable and multivariable regression analysis of risk factors were established.　Results　The cumulative duration of BIS<45, MAP<65 mmHg（1 mmHg=0.133 kPa）, Ce<1.5 mg/L and triple low state in mortality group were respectively (278±95), (153±41), (125±40), (58±16) min, which were significantly longer compared with the survival group(163±53), (65±21), (63±20), (21±6) min, respectively)(P<0.05). In univariate regression model, smoking history, preoperative comorbidities, the cumulative duration of BIS<45, MAP<65 mmHg, Ce<1.5 mg/L and triple low state, average duration of surgery, anesthesia and cardiopulmonary bypass and postoperative ICU stay were risk factors of increased 30-day mortality(P<0.05). When we added these risk factors to a multivariate logistic regression analysis, cumulative duration of triple low state and postoperative ICU stay remained the significant predictor of 30-day mortality(P<0.05).　Conclusions　Triple low state of propofol anesthesia(BIS <45, MAP<65 mmHg and Ce<1.5 mg/L) and ICU stay were independent factors of 30-day mortality risk in patients with valve replacement surgery.