Abstract
Purpose: Peripheral intravenous catheter (PIVC)-administered noradrenaline offers faster treatment for septic shock but risks complications like phlebitis. We aimed to investigate the relationship between the total noradrenaline dose administered via PIVCs and the development of phlebitis by considering the influence of noradrenaline as a time-dependent covariate. Methods: A post hoc analysis was conducted on prospective multicenter cohort data from 23 intensive care units in Japan. The total noradrenaline dose was included as a time-dependent variable in a multilevel Cox regression model, and smoothing splines assessed nonlinear relationships. The primary endpoint was phlebitis. Directed acyclic graphs were used to define confounding factors for the analysis. Results: The analysis included 3410 PIVCs from 1351 patients, with noradrenaline administered to 70 patients (5.2%) with 91 PIVCs (2.6%). The median dwell time and interquartile range of PIVCs was 46.2 h (21.3-82.9). No significant association was observed between the total noradrenaline dose and the occurrence of phlebitis through analysis using the multilevel Cox regression model with time-dependent covariate, which assumed the linear relationship between phlebitis occurrence and the total noradrenaline dose (hazard ratio 1.06, 95% confidence interval [CI] 0.93-1.20). Spline curve analysis suggested a nonlinear relationship between the total noradrenaline dose and phlebitis, and the risk of phlebitis increased when the total administered dose of noradrenaline exceeded 6 mg as the lower limit of the 95% CI exceeded the significant threshold of 1.0. Sensitivity analyses, including additional potential risk factors, showed consistent results compared with those of the primary analysis. Conclusions: Administering noradrenaline within a total dose not exceeding 6 mg reduces the risk of phlebitis, potentially allowing safer administration through PIVCs. Trial Registration: UMIN Clinical Trials Registry (UMIN-CTR): UMIN000028019.