INTRAOPERATIVE DETERMINANTS OF PULMONARY COMPLICATIONS AFTER CARDIAC SURGERY: FOCUSING ON INTRAOPERATIVE TRANSFUSION AND TIME ON PUMP IMPLICATIONS ON SHORT AND LONG-TERM OUTCOMES

Abstract

Background and Introduction: Postoperative pulmonary complications remain a major contributor to morbidity in patients undergoing cardiopulmonary bypass grafting (CABG). In this study we aimed to investigate how intraoperative factors, focusing on intraoperative transfusion (iXFN) and use of cardiopulmonary bypass (CPB), affect the development PPCs. Furthermore, we aimed to explore how these factors and the development of PPCs influence survival after CABG. Methods: This investigation is a retrospective analysis of prospectively collected clinical registries collected at two community hospitals in Ohio, USA. All patients who underwent first-time or redo, isolated CABG (N = 6,151) between Jan 1, 1994 and Dec 31, 2005, were considered for this study, and compared for two factors. The two factors are 1) use of early or intra-operative blood transfusions and 2) the role of use (on pump versus off pump) and duration of CPB in case of on pump CABG [off pump (0 minutes); short (1–59 minutes); intermediate (60-119minutes) and long (≥120minutes)]. Long-term survival data were obtained from the United States Social Security Death Index. Univariate binary logistic regression was used to calculate unadjusted odds ratios (UORs) and 95% confidence intervals (CIs) for potential risk factors associated with pulmonary complications. Subsequently, multivariable logistic regression was applied to determine and quantify the risk-adjusted effects of independent predictors (AORs). Discrimination of logistic regression models was evaluated using the area under the receiver operating characteristic (ROC) curves. Then, the ROC curves were compared for each outcome. For the survival analysis, groups were compared by Cox proportional hazard models, Kaplan-Meier survival plots, and hazard functions to determine 1) pulmonary complications, 2) perioperative mortality and 3) intermediate and long term mortality at 1-year and 5-year follow up. Patient demographics, risk factor, operative and outcomes data were also compared for patients that did versus did not experience postoperative pulmonary complications. Results: Analysis of preoperative and operative risk factors showed a slightly higher risk profile in patients with intraoperative transfusion, while the risk profiles were generally comparable across different CPB groups. A total of 583 (9.5%) patients developed one or more pulmonary complication. We have found both factors, intraoperative blood transfusion and CPB use, to be associated with worse pulmonary outcomes; with higher pulmonary morbidity observed in longer CPB duration. The odds of developing any PPC was 2.7 higher (AORiXFN/NoiXFN = 2.7 [2.2-3.3]) in patients with intraoperative transfusion compared to those without intraoperative transfusion. As for CPB use/ duration, we noted higher odds of suffering any PPC in patients with longer CPB times compared to patients with zero CPB time (off pump); AOR1-59 mins/ Off Pump = 2.0 [1.1-3.8]), AOR60-119mins/ Off Pump = 2.5 [1.4-4.5], AOR>120 mins/ Off Pump = 3.9 [2.2-7.0]. In order to elucidate the combined effect of CPB and iXFN, a composite eight-category transfusion-CPB (2 x 4) variable based on intraoperative transfusion (Yes or No) and CPB use (off pump, 1-59 mins, 60-119 mins and > 120 min) was created. Compared to patients with no iFXN and without CPB use (the reference category), the odds of developing any PPC were consistently higher in the other categories, with greater values in patients with intraoperative transfusion and in those with longer CPB durations. Notably, the odds of developing any PPC was 10.8 times higher in the category of patients with iXFN and long CPB duration of > 120 mins, compared to the reference category of patients without iXFN and without CPB use (AOR iXFN&>120 mins/ NoiXFN&Off Pump = 10.8 [5.1-23]). Both factors, iXFN and CPB use/ duration, however, neither adversely affected perioperative mortality at 30-day follow up, nor intermediate and long-term 1-year and 5-year follow up. Only exception, however, is the 40% significant increase in 5-year mortality risk observed in patients who underwent surgery with long CPB time (> 120 mins) compared to those who underwent off pump surgery (AHR> 120mins/Off Pump= 1.4 [1.1-2.0]). Finally, the development of postoperative pulmonary complications was generally not associated with adverse mortality outcomes at 30-day, 1-year and 5-year follow up. Conclusion: In this study, we have identified a synergistic interaction between two widely studied intraoperative risk factors (intraoperative transfusion and cardiopulmonary bypass use/ duration), and found it to adversely affect the development of postoperative pulmonary complications in patients undergoing CABG. It is our hope that our findings will provide a knowledge base that lays the groundwork for future studies that aim to improve transfusion practices and further explore the role of CPB use/ duration.