The comparison of caesarean section bleeding between volatile and total intravenous anaesthesia in a Japanese nationwide database

Abstract

BACKGROUND Volatile anaesthesia may increase blood loss because of the uterine-relaxing effect of the volatile anaesthetics during caesarean section under general anaesthesia. OBJECTIVE This study compared the bleeding risk during caesarean section between volatile anaesthesia and total intravenous anaesthesia (TIVA) using a nationwide inpatient database in Japan. DESIGN Observational study. SETTING Nationwide inpatient database in Japan from April 2012 to March 2020. PATIENTS Women who underwent caesarean section under general anaesthesia. MAIN OUTCOME MEASURES Volume of blood loss. RESULTS We identified 26 585 women, including 19 320 in the volatile anaesthesia group (mean age = 32.9 ± 5.5 years) and 7265 in the TIVA group (mean age = 32.8 ± 5.5 years). The mean blood loss was 1113 ± 909 and 1136 ± 944 ml and the proportion of blood transfusion was 14.7 and 16.0% in the volatile and TIVA groups, respectively. With conventional regression analyses, volatile anaesthesia was associated with a slightly lower risk of bleeding: the adjusted mean difference for blood loss (95% CI) was −56.1 (−81.4 to −30.7). However, in the instrumental variable analysis, volatile anaesthesia was associated with a higher risk of bleeding: adjusted mean difference for blood loss (95% CI) was 154.3 (112.4 to 196.3) ml. CONCLUSION This large observational study with instrumental variable analyses suggested an increased bleeding risk associated with volatile anaesthesia, and the results were not identical to those in the conventional regression analyses.