Impact of Bedside Tracheostomy on Procalcitonin Kinetics in Critically Ill Patients.

Abstract

Critically ill patients are particularly vulnerable to invasive procedures and complications; however, tracheostomy is frequently performed in the intensive care unit (ICU). We analyzed the effects of tracheostomy on procalcitonin (PCT) kinetics and investigated whether PCT could reliably predict septic complications after tracheostomy. We retrospectively identified 134 patients who underwent bedside tracheostomy during their ICU stay at a Japanese university hospital from October 2010 to December 2015. We extracted PCT data from the day of the procedure (day 0) to postoperative day 2 and defined alert PCT as a PCT level ≥0.5 ng/mL, which had not decreased from the previous day. We divided patients into the following groups: nonevent, aseptic complication, and septic complication. Twelve (9.2%) patients developed acute aseptic complications, and 12 (9.2%) patients developed septic complications. In the nonevent group, the PCT value decreased continuously in the initial PCT ≥ 0.5 ng/mL subgroup (P < .001, P <.001 for trend). In contrast, significant changes were not observed in the initial PCT < 0.5 ng/mL subgroup. Significant differences and an upward trend in alert PCT incidence rate existed between the groups (P < .001, P < .001 for trend): nonevent group, 5.5%; aseptic complication group, 41.7%; and septic complication group, 66.7%. In a multivariate linear regression model, septic complications were independently associated with PCT change at postoperative days 1 and 2 (adjusted β = 3.58, P < .001; adjusted β = 9.84, P < .001, respectively). Procalcitonin predicted septic complications more accurately than C-reactive protein, with the area under the receiver operating characteristic curves of 0.8 versus 0.63 (P = .058) and 0.91 versus 0.69 (P = .036) at postoperative days 1 and 2, respectively. Our results demonstrated that PCT was not elevated after uncomplicated surgical tracheostomy in critically ill patients.