LAUSR

Background Procalcitonin (PCT) is efficient in reducing antibiotic usage without increasing complications for its sensitivity and specificity in detecting bacterial infection. However, its role in guiding antibiotic-spectrum escalation has not been studied. This study was performed to validate the role of PCT in indicating antibiotic spectrum escalation when pathogen results are unknown for ICU patients of suspected bacterial infections. Methods This was a single-center retrospective study including patients who were admitted to Peking Union Medical College Hospital from January 2014 to June 2018 for suspected bacterial infections. Patients were divided into “escalation” or “non-escalation” groups according to the change of employed antibiotic spectrum before and after the occurrence of “PCT alert”. The main study endpoint was the length of ICU stay (LIS), and LIS longer than 7 days was defined as “prolonged-ICU-stay (PIS)” while LIS equal to or shorter than 7 days was defined as “non-prolonged-ICU-stay(nPIS)”. Demographics, clinical characteristics, and infection characteristics were compared between patients in the “nPIS” and “PIS” groups. Multivariable logistic regression was used to evaluate independent risk factors for PIS. Results Totally, 1109 patients were included, and 654 in the PIS group, other 455 in the nPIS group. Respiratory infection was the main cause in both groups. Patients were older in PIS group than in nPIS group(PIS vs. nPIS: 58.99 ± 16.30 vs. 56.12 ± 15.93 years, P=0.002). The baseline Sequential Organ Failure Assessment (SOFA) score was 11.16 ± 7.33 and 9.73 ± 3.70 in PIS and nPIS groups. Fewer patients received antibiotic escalation in face of “PCT alert” in PIS group (PIS vs. nPIS: 27.68 vs.35.38%, P=0.014). In the multivariable logistic regression model, older age, higher heart rate, not undergoing surgery, higher baseline SOFA score, and not escalating antibiotics in face of “PCT alert” were associated with a prolonged ICU stay. The odds ratio of antibiotic escalation for PIS was 0.582 (95% CI: 0.365, 0.926, P=0.022). Conclusions Using PCT to guide antibiotic escalation when pathogen evidence is unavailable could be associated with a shorter length of ICU stay for ICU patients of suspected bacterial infection.