Intra‐operative bradycardia in gynaecological surgery — a reply

Abstract

We have read the valuable comments of Dr Buggy about our paper in which we compared the incidence and severity of intra-operative bradycardia in patients undergoing elective total abdominal hysterectomy (Anaesthesia 1997; 52: 1221–4) and would like to clarify a few points. Dr Buggy's observation that prospective power of study was not calculated is correct, but there have been other recent studies on similar subjects with fewer patients where the prospective power of study was not calculated, or at least not mentioned [1]. Moreover Dr Buggy used simple bradycardia to calculate retrospectively the power of the study which was an insignificant finding in our study and therefore will not affect clinical practice. Regarding the statistical analysis of severe and symptomatic bradycardia, Dr Buggy's observation is correct that Chi-square test is not applicable and either Yates' correction or Fischer's exact test should be used. Because of the small number of patients we also performed the statistical analysis using Fischer's exact test on the Epi-Info-6 computer statistical package which gave a p value of < 0.05 (print-outs available on request). We accept our mistake for not mentioning this detail of statistical analysis in our paper, but this does not affect the clinical validity of this study. We do not agree with the point raised by Dr Buggy regarding the cut-off value of heart rate for labelling bradycardia under anaesthesia, as the definition we used was taken from a standard anaesthesia textbook and such observations cannot be made on the basis of institutional practice [2]. Furthermore, the rest of the anaesthetic technique clearly stated that the drugs used for the maintenance of anaesthesia (enflurane and pethidine) are known to cause tachycardia in the doses we used [3, 4]. Thus the occurrence of bradycardia in the presence of drugs that tend to increase heart rate is clinically a significant finding. In conclusion our observations are supported by our data. However, the argument would become stronger if the number of patients was larger.