The role of preoperative endocrine parameters in prediction of postoperative complication in adult patients undergoing elective heart surgery – Preliminary study

Abstract

Introduction The correct cost-benefit estimation and the more accurately perioperative risk evaluation is one of the more difficult task in the perioperative medicine. We wanted to study more than the obvious parameters, so we looked for less studied predictors, such as different peripherial hormone levels. Every hormone molecule acts a complex pathway, so we truely believe that the insufficient function or decreased serum level of these mediators could be heavy influence for the perioperative adverse events. Methods Our single center prospective, observational study approved by the IRB and registered at clinicaltrials.gov (NCT03736499). We enrolled 89 patients between 25.01.2019 and 30.04.2019 who underwent elective cardiac surgical procedure. In the preoperative period we measured different hormone levels in serum (TSH, free T3 and T4, prolactine and testosterone) and collected the demographic parameters. We corrigated the normal hormone levels for age and gender. We recorded the length of postoperative intensive care unit stay, length of mechanical ventilation, dose of different vasopressor and inotropic agents, fluid balance, blood transfusion and bleeding. The adverse outcomes were also collected. The primary outcome was the all-cause mortality. Secondary outcome were the length of ICU stay, length of mechanical ventilation and the frequency of different adverse events. We performed descriptive statistics, chi square test and Mann-Whitney U test using the IBM-SPSS 22.0 statistic software. Results Median age was 66.0 years (IQR: 59.0-73.0), 64.04% of the patients were male, median of BMI was 28.22 (IQR: 25.44-31.44). The median of EuroSCOREII was 1.94 (IQR: 1.17-3.05). The mean follow up time was 52.8±20.68 days. 8 patients (8.98%) died during the follow up time, the median length of ICU stay was 45 hours (IQR: 22.0-92.0), the median length of mechanical ventilation was 9 hours (IQR: 5.0-19.0). In univariable Cox regression serum free T3 was correlated with all cause mortality (OR: 0.266, CI 95%: 0.096-0.735, p = 0.011) and the abnormal level of free T3 was associated with higher infection rate (OR: 6.052, CI 95%: 1.271-28.832, p = 0.024). Using nonparametric tests we found association between abnormal serum prolactine level and postoperative arrhythmias (p = 0.017) and between abnormal serum testosterone level and prolonged length of ICU stay (p = 0.002). Discussion The evaluation of serum hormone levels could help to make more precise estimation of perioperative risk in adult patients waiting heart surgery. Further examination should be performed, to establish stronger correlation between fine hormonal dysregulation and postoperation complications.