BACKGROUNDAlthough the results of surgical treatment in cardiac valve disease continue to improve, the postoperative mortality rate and the rate of complications in patients with advanced valvular heart disease (AVHD) are still very high. We did this retrospective study to summarize the surgical experience of heart valve replacement for patients with AVHD and discuss effective ways to improve the surgical outcome.

METHODSFrom January 1994 to October 2003, surgical procedures of heart valve replacement were performed on 227 (136 men and 91 women) patients with AVHD in our Department of Cardiothoracic Surgery. The clinical data of all patients were collected and analysed. Patients' age ranged from 10 years to 77 years. In preoperative cardiac function grading, 157 cases were NYHA III and 70 cases NYHA IV. Fifty-one patients had had cardiac operations. The ultrasonic cardiac graphs showed that 145 patients suffered from moderate or severe pulmonary hypertension and 73 had combined giant left ventricle. Mitral valve replacement was performed in 32 cases, aortic valve replacement in 90, tricuspid valve replacement in 1, combined mitral and aortic replacement in 103 and combined mitral and tricuspid replacement in 1. Nineteen patients also received surgical corrections for other minor abnormalities during the operations. A logistic model was established to evaluate the influence of perioperative factors on the mortality rate.

RESULTSThe operative mortality rate was 13.2% (30/227). The main causes of death included multiple organ dysfunction syndrome (MODS), low cardiac output syndrome and ventricular fibrillation. From the results of the binary noncounterpart multivariate logistic regression, the following statistically significant factors were found to influence the operative mortality rate: redo operation, age >/= 55 years, preoperative NYHA cardiac function grading, extracorporeal circulation time >/= 120 minutes and postoperative usage of GIK (glucose, insulin and potassium) solution. All factors were risk ones except postoperative application of GIK. The Hosmer-Lemeshow goodness of fit coefficient of this model was 0.976.

CONCLUSIONSThe risk factors associated with postoperative mortality rate in the patients with AVHD were redo operation, age >/= 55 years, preoperative NYHA cardiac function grading and extracorporeal circulation time >/= 120 minutes. Postoperative usage of GIK acted as a kind of metabolic therapy and will improve the recovery for patients with AVHD. Active perioperative management and care will play a very important role in reducing the operative risk and improving the short term outcome of surgical treatment for the patients with AVHD.