97 Can pre-operative troponin levels predict post-operative mortality following non-cardiac surgery?

Introduction Despite advances in surgical and anaesthetic techniques, non-cardiac surgery still has a significant mortality. We hypothesised that pre-operative troponin levels might predict post-operative mortality. Methods Patients undergoing elective and urgent/emergency non-cardiac surgery excluding minor procedures were retrospectively assessed for known vascular disease (defined by diagnostic imaging or previous intervention rather than clinical assessment) and vascular risk factors including hypertension, treatment with lipid-modifying agents (irrespective of agent or dose) and chronic lung disease. Pre-operative high-sensitivity troponins and routine pre-operative bloods were recorded. Six- and twelve-month mortality data were collected; independent predictors of mortality and associations between pre-operative patient characteristics and pre-operative troponin were determined. Results 993 patients were assessed; 13% had an elevated pre-operative troponin with 3%u003e50 ng/L. 825 (83%) were elective patients; 8.6% had an elevated pre-operative troponin. Six-month mortality was 4.2% and 5.9% at twelve months. Elevated pre-operative troponin was associated with higher post-operative mortality; 2.5%, 12.5% and 25% for a troponin 50 ng/L respectively (figure?1). This trend was also evident at twelve months; 3.7%, 16.3% and 37% for the same troponin bands (figure 2). Lipid-modifying agents were independently associated with a lower rate of pre-operative troponin release (HR 0.446 (0.232 – 0.857) p=0.015). Impaired renal function (assessed as a continuous variable), emergency presentation and pre-operative troponin levels were independent predictors of six- and twelve-month mortality with emergency presentation the strongest predictor by hazard ratio (table 1). Age over 75 independently predicted twelve-month mortality only. Conclusion The role of pre-operative troponin monitoring and the prevalence of pre-operative troponin has not previously been established on an all-comer population. Pre-operative troponin level greater than 50 ng/L is an independent predictor of six and twelve-month mortality following non-cardiac surgery in an all-comers cohort although the mechanism of troponin release is not clear. The lower rate of troponin release associated with lipid-modifying medication has been seen in other studies (1). We hypothesise the known anti-inflammatory effects of statins may indicate a systemic inflammatory process responsible for the troponin release rather than unstable coronary disease. Further studies to assess this in the elective population may be useful to target pre-operative interventions. Reference Association between pre-operative statin use and major cardiovascular complications among patients undergoing non-cardiac surgery: the VISION study. ?Berwanger O, Le Manach Y, Suzumura EA, et al. Eur Heart J2016Jan 7;37(2):177?–?85