Factors Determining In-Hospital or Late Survival after Aortic Valve Replacement: Collection of Data and Statistical Analyses
Collection of Data and Statistical Analyses
Preliminary data on all patients was gathered retrospectively from patients’ hospital records and questioning patients. The condition of patients has been ascertained yearly since 1981. This was accomplished by examination of the patient by the cardiologist or cardiac surgeon (or both) and by telephone interview with the patient or with the referring physician (or both). No patient was lost to follow-up. The data have been computerized. canadian pharmacy
Univariate and multivariate semiparametric regression analyses* were used to determine the influence of 12 independent variables on in-hospital and late survival of patients. These were previous open-heart surgery, age at operation, preoperative NYHA functional classification, size of valvular prosthesis, status at time of operation (elective, urgent, or emergent), year of operation, concomitant procedure, valvular lesion (aortic stenosis, aortic regurgitation, or mixed lesion), cardiopulmonary bypass time, aortic cross-clamp time, surgeon, and valve-related events. Other statistical methods applied included time-related hazard function to show the instantaneous risk of death, and actuarial survival. A p value of 0.05 or less was considered statistically significant. (Statistical analysis was performed by one of the authors [S. A.]).
In-hospital mortality, defined as death within 30 days following operation, occurred in 16 (7 percent) of the 240 patients (70 percent CL = 5.0, 8.8). Six deaths (5 percent) occurred in the subgroup of 130 patients who had undergone isolated AVR (70 percent CL = 2.7, 7.4), compared to ten deaths (9 percent) in the subgroup of 110 patients who had concomitant procedures (70 percent CL = 6.2, 12.9) (not significant).