Comparison of Perioperative High-Sensitive Troponin T and Troponin I Assays in Cardiac Surgery

Background Troponin measurements are among the standard parameters for monitoring perioperative myocardial ischaemia after cardiosurgical procedures. As high-sensitive assays continue to replace older analytic parameters with lower sensitivity, this study aimed to compare perioperative profiles of a high-sensitive troponin T assay (hsTnT, Roche Diagnostics, Mannheim, Germany) with a troponin I assay (sTnI, Siemens Healthcare Diagnostics, Eschborn, Germany). Methods A total of 287 consecutive patients undergoing a typical spectrum of cardiac procedures from August 2017 to March 2018 monitored with the hsTnT assay were compared with a propensity-matched collective analysed with the sTnI assay. For side-by side comparison, the peak troponin (Tmax) values were scaled to a z-score distribution before comparison. Results Despite absolute postoperative hsTnT and sTnI values differing by an order of magnitude, parameters could be scaled to a common distribution with kernel density curves overlapping 92%. Both parameters showed equal behaviour in subgroup analyses regarding relevant perioperative factors, such as type of procedure, cross-clamping time, and type of cardioplegic solution. However, there were some differences regarding pre-existing renal impairment between both parameters. In both groups, renal failure patients with chronic kidney disease stages IV or V as well as patients on haemodialysis exhibited a marked Tmax increase of .100% compared with normal kidney function (hsTnT, +121%; 2,383.5 vs 1,078.8 ng/L; p=0.0006; and sTnI, +149%; 27.3 ng/mL vs 11.0 ng/mL; p=0.009). However, in patients with moderately impaired renal function, those in the hsTnT group, but not in the sTnI cohort, showed significantly increased Tmax values (CKD stages II or III, 1,233.5 ng/L [+14%] and 1,314.1 ng/L [+22%] vs 1,078.8 ng/ L; p=0.01 and p=0.03). In these patients, the postoperative interval until Tmax was reached was also significantly increased (14.4 and 19.0 hrs vs 12.4 hrs for chronic kidney disease stages II and III; p=0.0038 and p,0.001), indicating a higher rate of accumulation in the hsTnT parameter. Conclusion In the context of cardiac surgery, this study found that both parameters behaved in a similar manner under most relevant circumstances. Despite significant difference in the absolute serum concentration, hsTnT and sTnI can be scaled to virtually identical distributions. However, renal impairment did affect both parameters differently with troponin T but not troponin I, showing evidence of accumulation in moderately impaired renal disease.