Progress In Adult Heart Valvular Disease Surgery Among Last Two Decades And The Change Of Risk Factors

【Objective】This project analysed the case data of adult Heart Valve Surgery retrospectivelyby using of a single-center database derived from Department of CardiothoracicSurgery, Changhai Hospital. The first purpose of our work is to observe the generalchange about adult cardiac valve surgery among last two decades. The second is toanalyze if there were any changes of risk factors about early postoperative prognosis.And the last, we try to establish a surgical risk prediction model of adult rheumaticheart disease.【Methods】1． Retrospective analysis of the adult heart valve surgery among last twodecades:(1).Patients in the database of the Department of Cardiothoracic Surgery,Changhai Hospital who is, or older than18years and underwent aortic and/or mitralheart valve surgery, both repair and replacement, from January1991to December2010were considered for inclusion. And99patients who had only pulmonary and/ortricuspid valve surgery were excluded to avoid computational problems associatedwith small mumbers. The final study sample comprised of6300patients withconcomitant Coronary Artery Bypass Graft (CABG) surgery, major vascular surgery,congenital heart disease surgery and other cardiac vascular surgery.(2). The clinical outcome considered was in-hospital mortality, defined as thepatient's status at discharge after the operation.(3). Patients were divided into four time span(1991to1995, from1996to2000,2001to2005,2006to2010) or year on year, according to the distributioncharacteristics of the data, in order to describe the basic trends. And than,[atientswere divided into the early surgery group (1991to2000) and late surgery group (2001 to2010) to determine whether there is a significant difference between the twogroups.2. The change of early postoperative risk factors about adult heart valvedisease:(1). The patients enrolled were as same as the first part. We select5postoperativeendpoints:1) hospital deaths: deaths in patients with valvular heart disease in thehospital for any reason;2) acute renal failure: acute renal failure after heart valveduring hospitalization;3) prolonged mechanical ventilation (PMV): postoperativeendotracheal intubation time>24hours;4) second surgery: Heart valve surgery againin the same hospital during the period line surgery of the heart and great vessels;5)prolonged postoperative ICU time: heart valve surgery, length of ICU stay>5days.(2). First, patients were divided into the early surgery group (1991-2000) and latesurgery group (2001-2010); second, screen the significant risk factors of two groupsby the mean of statistical anaylasis; third, compare and analyze whether there is anydiffenrnt or changer of risk factors among two groups.(3). The choice of clinical variables refers to the STS2009, the latest releasevalve disease risk-prediction model. In addition, we added some surgery informationsaccording to research purposes. And there were some modifications of Variable valuesbased on the specific situation of China's situation.3. Establishment of risk prediction model and risk score for in-hospitalmortality after adult rheumatic heart valve surgery:(1). The final study sample comprised of3889patients with adult rheumatic heartvalve surgery only. including aortic valve replacement, mitral valve repair, mitralvalve replacement, and aortic and mitral combination procedure. All patients weredivided into three subgroups according to the surgery site of left atrioventricular valve:mitral valve surgery group; aortic valve surgery group; and mitral and aortic valvesurgery group. Risk factors that were possibly associated with mortality were collected from the clinical variables in the database, and then the coding of candidatepredictor variables were determined.(2). The data was splited into development (60%) and validation (40%) data sets,and then the risk model was developed by using a logistic regression model accordingto the data in development data set.(3). Predicted mortality was compared to observed mortality for the developmentdata set, validation data set, and each subgroup. Model calibration was analysed byHosmer-Lemeshow goodness-of-fit statistic, and model discrimination was tested bycalculating the area under the receiver operating characteristic (ROC) curve.(4). Risk score was finally set up according to the coefficient βand rank ofvariables in logistic regression model.【Results】1. Retrospective analysis of the adult heart valve surgery among last twodecades:(1). A total of6300cases of adult heart valve surgery were included in the study,4309cases of late surgery group,1991cases of early surgery group. The overall trendof valvular heart disease for the number of patients is increasing year by year. Theproportion of female patients is more than male patients. The proportion of patientsolder than50years is increased year by year, from15.3percent in1991to55.3percent in2010. And patients who are older than70years were emerged since1998.The proportion of asymptomatic patients was increased year by year, and the averagetime from onset of symptoms to seeking treatment was shortend. Patients whosecardiac function in NYHA class III-IV level were reduced. The proportion ofpatients with preoperative hypertension, high cholesterol, diabetes, coronaryatherosclerotic heart disease, chronic lung disease and cerebrovascular disease wassignificantly increased. The proportion of patients with concomitant CABG surgerywas increased. (2). The composition of the cause of valvular disease is still rheumatic lesions,but the proportion of the degenerative valvular disease was growing. The ensuingchange is the reduction of the proportion of atrial fibrillation and double valvereplacement surgery, and the increasing of the proportion of mitral valve surgery andaortic valve replacement.(3). As matter of surgery, due to technological advances, cardiopulmonary bypasstime, aortic cross-clamping time and assisted circulation time was significantlyshorter than before (P <0.05). Application of the ratio of biological valve in the mitraland aortic valve replacement surgery. The effectiveness of reopretion is significantlyimproved.(4). There were significant reduction of early postoperative mortality (P<0.05),early postoperative acute renal failure rate (P <0.05), postoperative mechanicalventilation time (P<0.05), rate of second surgery (P <0.05) and postoperative ICUtime (P<0.05) among the last decade than those in the first decade.2. The change of early postoperative risk factors about adult heart valvedisease:(1). There are1991cases in the early surgery group Total, mean age42.92±10.78yearsm,965male cases (48.5%) and1026female(51.5%) cases. Death in113cases (5.7%), postoperative acute renal failure in58cases (2.8%), delayed extubationin642cases (32.2%), the second operation in84cases (4.2%), prolonged ICU time963cases (48.4%). And4309cases of late surgery group consisted of patients withan average age of48.85±12.33years old,1988male cases (46.1%) and2321femalecases (53.9%). Death in183cases (4.2%), postoperative acute renal failure in93cases (2.2%), delayed extubation in678cases (15.7%), the second surgery in135cases (3.1%), prolonged ICU time745cases (17.3%).(2). Previous valve surgery is the independent risk factor of hospital mortality(2.47,95%CI1.31-4.66), postoperative delayed extubation (2.43,95%CI1.53-3.84)and prolonged ICU time(3.06,95%CI1.76-5.31) of early surgery group, but which did not constitute risk factors of late surgery group. Surgical status is the independentrisk factor of early death (2.27,95%CI1.08-4.79) of early surgery group and ofprolonged ICU time(2.23,95%CI1.21-4.09) of late surgery group. The effect of allthis risk factors talk before is more significant to the patients in early surgery group.(3). The procedure of concomitant CABG (11.76,95%CI3.48-39.76) is the mostimportant independent risk factor of early death in early surgery group. Although it isthe risk factors of early death in late surgery group patients (1.95,95%CI1.13-3.37),but the effect is not so significant.(4). In late surgery group, diabetes, hypertension and Previous cerebral embolismhistory are the risk factors of prolonged mechanical ventilation(1.23,95%CI1.00-1.50), prolonged ICU time(1.28,95%CI1.05-1.56), acute renal failure (1.82,95CI,1.02-3.24) and early death (2.09,95%CI1.13-3.87), respectively. All of theabove risk factors is meaningful for the prognosis of patients in late surgery group.(5). Risk factors of age, cardiac NYHA functional class, left ventricular ejectionfraction (EF), CPB time, and concomitant other cardiac surgery are meaningful forboth groups and there is no significant difference.3. Establishment of risk prediction model and risk score for in-hospitalmortality after adult rheumatic heart valve surgery:(1). We established a risk prediction model and found seven risk factors: heartfunction in NYHAfunctional class≥II grade (3.36,2.42-4.67), preoperative creatinine>110mmol/L (2.69(1.51-4.79), history of previous chest pain (2.33,1.07-5.11),surgical status (2.32,0.94-5.73), previous history of hypertension (2.24,1.19-4.23),preoperative critical state (2.14,1.27-3.60) and age>50years (1.57,1.18-2.09).(2). Our risk model showed good calibration calibration and discriminative powerfor the development data set,validation data set, and three subgroup in whichHosmer–Lemeshow test's P value were greater than0.05and the area under the ROCcurve were greater than0.70.(3). Scoring methods: age51-60years:1point, age61-70:2points, age>70 years:3points; history of hypertension:1point; creatinine>110umol/L:4points;NYHA class stageⅡ:2points, NYHA class stage III:4points; NYHA class stage IV:6points; history of previous chest pain:1point; preoperative critical condition:2points;urgent surgery:2points: emergency surgery:4points.【Conclusion】1. Patients in early surgery group had higher rate of combined symptoms,longer time of symptoms and poorer heart function classification. While patients inlate surgery group had older average age, more incidence of hypertention, diabetes,high cholesterol, smoking history and concomitant coronary heart disease. The majorcause of heart valve disease is still rheumatic lesions in our country, but theproportion of degenerative valve disease is gradually increased. In recent years, theresults of heart valve surgery is better and better than before, as cardiopulmonarybypass time and aortic cross-clamping time was significantly shortened, the earlypostoperative mortality, delayed extubation rate, the rate of secondary surgery andICU time was significantly decreased.Currently, the main problems are: the average age of patients is increasing, theproportion of elderly patients also showed an increasing trend. There are moredifficult of treatment and diagnosis for this kind of patients who still had a higherincidence of postoperative mortality and complications.2. the risk factors for adult heart valve surgery for early death andcomplications are in a dynamically changing, in:1) risk factors meaningful to patientsin early surgery group but no longer significant for patients in late surgery group:previous history of heart valve surgery, surgical status (elective, urgent, emergency);2), the degree of influence weakened: concomitant coronary artery bypass grafting(CABG);3) meaningless for patients in early surgery group but significant forpatients in late surgery group: diabetes, hypertension, cerebral infarction, history,creatinine>110mmol/L;4), meaningful for both surgery groups with no significant change: age, cardiac NYHA functional class, left ventricular ejection fraction (EF),CPB time, concomitant other cardiac surgery.3. We have created a new risk prediction model and risk score, which canaccurately predicts outcomes in patients undergoing heart valve surgery for our center.Furthermore, our risk model can also enable benchmarking and comparisons betweenmulticenters in a meaningful way in the furture.