Failure to THRIVE: no cause for alarm

Abstract

EditordWe read with interest the retrospective study by Abe and colleagues 1 on the association between mode of anaesthesia for scheduled Caesarean delivery and maternal morbidity. As expressed here and in the accompanying editorial, 2 a randomised trial would generate the most rigorous data, but cannot ethically be performed. With a very large data set, Abe and colleagues 1 accounted for over 30 comorbidities with propensity-score matching. Whilst the editorial by Butwick and Palanisamy 2 addresses the possibility of residual confounding by indication, we would like to draw attention to another potential limitation in the methodology of the study: failure to analyse the data by the intended anaesthetic. The authors excluded cases requiring both neuraxial and general anaesthesia from their analysis, stating that these patients could not be accurately categorised. However, whilst in some cases both neuraxial and general anaesthesia might have been planned, these patients most likely received a neuraxial anaesthetic before conversion to general anaesthesia, and not the reverse. The authors acknowledge this when they state that conversions usually occur when 'neuraxial anaesthesia is insufficient or critical vital signs emerge'. Whilst the conversion group likely represents a small proportion of their data set (<233 of 95 232 scheduled Caesarean deliveries), our data 3 and the data of others 4 indicate that these cases include a disproportionate number of morbid outcomes compared with neuraxial anaesthesia alone. Given that the 10 046 cases with neuraxial anaesthesia contained only 76 cases of severe morbidity, it is possible that the inclusion of patients receiving both neuraxial and general anaesthesia significantly increased this number if analysed as a presumed intended neuraxial anaesthetic. This potential bias should be considered when interpreting the results of Abe and colleagues. 1 Declaration of interest None declared. ANE.0000000000003314. Access published on February 23 4. Guasch E, Gilsanz F. Treatment of postpartum hemorrhage with blood products in a tertiary hospital: outcomes and predictive factors associated with severe hemorrhage. Clin