Background:
Worldwide, 200 million adults undergo non-cardiac surgery annually, and completing pre-operative assessments is a major role of today’s hospitalists. In previous years, decisions regarding anti-platelet agents were less evidence-based and often fell to provider preference. With the release of POISE-2 in 2014, there is now stronger evidence to support holding aspirin before elective surgery partly based upon the increased rates of major bleeding in those continued on anti-platelet agents before and after the surgical period. The quantitative measure of major bleeding and the impact on clinical practice was not well described. Our aim with this study was to identify the potential benefit to resource utilization, including blood transfusions, which may occur as a result from practice changes in accordance with the POISE-2 trial findings.
Methods:
We conducted a retrospective review of the medical records of 739 consecutive patients who underwent elective hip or knee arthroplasty at both an 805-bed tertiary teaching hospital and its affiliated 191-bed community hospital from January to December 2010. Data collected included basic demographics, ongoing use of anti-platelet agents pre-operatively, and the indications for the use of these agents. Outcomes were assessed by pre- and post-operative hemoglobin measurements, the need for transfusion of packed red blood cells (pRBC), the quantity of units transfused, and the occurrence of myocardial infarction (MI), stroke, venous thromboembolism (VTE), or death.
Results:
Of the 739 patients, 233 (31.5%) were taking aspirin prior to surgery. Of these, 20 patients (8.6%) had no interruption of aspirin use [continuation group], while 213 (91.4%) were instructed to stop taking aspirin [cessation group]. When comparing these two groups, the continuation group had significantly higher occurrence of major bleeding (13/20 [65%] vs 56/213 [26%], p < 0.0001). In regards to bleeding and blood bank resources, there was no statistical difference in the absolute drop in hemoglobin (-3.61 + 1.53 continuation group vs -3.39 + 1.46 cessation group, p = 0.48), nor the number of patients requiring transfusion (2/20 [10%] vs 41/213 [19.2%], p = 0.34), nor the number of units of pRBCs administered to those who required transfusion (3.00 per patient continuation group vs 2.58 per patient cessation group, p = 0.38). As for other outcomes, there was no statistical difference in the rates of MI, stroke, VTE, or death between the two groups.
Conclusions:
In accordance with the results of the POISE-2 trial, our study demonstrated that continuation of aspirin in the peri-operative period resulted in an increased risk of significant bleeding, and had no appreciable difference in negative outcomes of MI, stroke, VTE or death. However, the degree of bleeding and the utilization rate and quantity of blood products was similar regardless of aspirin use. While the resources used to address less life-threatening bleeding was not captured by this study, there appears to be no difference from the perspective of blood bank utilization in the management of elective surgery patients. Evaluation of other populations and other elective surgeries is needed to draw more definitive conclusions.