PFA-100© closure times and postoperative blood loss after total knee arthroplasty in patients taking NSAIDs

Abstract

OBJECTIVES: PFA-100© is useful for monitoring various congenital and acquired platelet dysfunctions, and the effect of aspirin and non-steroidal anti-inflammatory drugs (NSAIDs). However the lack of clinical outcome correlation remains the weakness of this monitor. We previously showed arthroplasty patients taking NSAIDs preoperatively have a wide range of PFA-100©closure times(Ng KFJ et al., 2008) but the different closure times does not seem to correlate with intraoperative blood loss(Ng KFJ et al., 2007). This study aims to correlate different preoperative PFA-100© closure times in these patients with postoperative blood loss as measured by drain output, and thus provide a more comprehensive evaluation of its clinical relevance in surgical patients. METHODS: With IRB approval, we recruited patients taking NSAID and undergoing total knee arthroplasty. All patients were allowed to continue the NSAID up to the time of surgery. No LMWH was used. Preoperative blood was taken for PFA-100© ADP and epinephrine closure times (ADPCT and EPICT) measurement immediately before surgery. In some patients, the PFA-100©measurement was repeated with the blood sample haemodiluted by 20% with saline in vitro. Non-closures were assigned ADPCT or EPICT values of 300 s. In addition, preoperative platelet count, prothrombin time (PT), activated partial thromboplastin time (aPTT) and fibrinogen concentrations (FIB) were measured. Surgery was performed by the same group of surgeons in a standardised manner under general anaesthesia. No deliberate hypotension was performed and body temperature was maintained. Before skin closure one drain under low suction was placed. Postoperative output in the drain and surgeon assessment of ease of haemostasis were the outcome measures. Linear correlation was sought between ADPCT and EPICT and these outcome measures. P < 0.05 was considered statistically significant. RESULTS: 30 patients were recruited (age, 66.3 ± 8.6; BMI, 27.4 ± 5.3; M/F, 7/23), representing 51 knee arthroplasties. Mean ADPCT was 120 ± 60 s and EPICT was 199 ± 79.6 s. There was significant correlation between EPICT, but not ADPCT and drain output in the ward (r = 0.30; P = 0.03). The correlation becomes stronger in the haemodiluted sample (r = 0.42; P = 0.01). ADPCT also become correlated with drain output in the haemodiluted sample (r = 0.41; P = 0.02). There was no correlation between platelet count, PT, aPTT, FIB or surgeon assessment and drain output. CONCLUSIONS: Although previously found not to predict intraoperative blood loss, the EPICT of the PFA-100© predicts postoperative drain output in patients taking NSAIDs and undergoing total knee arthrosplasty. The sensitivity of the prediction seems to improve with in vitro haemodilution of the sample before measurement, which probably reflects the clinical postoperative condition better. The lack of proven clinical relevance of the PFA-100© has been a major limitation of this monitor. PFA-100©closure times has never been correlated with blood loss during surgery, except in open heart surgery3. This study provides the first evidence that the PFA-100© can actually predict bleeding perioperatively in general surgical patients. It is a useful monitor of platelet function in patients taking NSAIDs presenting for non-cardiac surgery. Conventional coagulation tests are not useful.