Post-spinal hypotension remains a common and clinically-important problem during caesarean section, and accurate pre-operative prediction of this complication might enhance clinical management. We conducted a prospective, single-centre, observational study of heart rate variability in 102 patients undergoing elective caesarean section in a South African regional hospital. We performed Holter recording for ≥ 5 min in the hour preceding spinal anaesthesia. The low-frequency/high-frequency ratio component of heart rate variability was compared, using a logistic regression model, with baseline heart rate and body mass index (BMI) as a predictor of hypotension (defined as systolic arterial pressure < 90 mmHg) occurring from the time of spinal insertion until 15 min after delivery of the baby. We also assessed clinically relevant cut-point estimations for low-frequency/high-frequency ratio. Low-frequency/high-frequency ratio predicted hypotension (p = 0.046; OR 1.478, 95%CI 1.008-1.014), with an optimal cut-point estimation of 2.0; this threshold predicted hypotension better than previously determined thresholds (p = 0.003; c-statistic 0.645). Baseline heart rate (p = 0.20; OR 1.022, 95%CI 0.988-1.057) and BMI (p = 0.60; OR 1.017, 95%CI 0.954-1.085) did not predict hypotension. Heart rate variability analysis is a potentially useful clinical tool for the prediction of hypotension. Future studies should consider a low-frequency/high-frequency ratio threshold of 2.0 for prospective validation., (© 2017 The Association of Anaesthetists of Great Britain and Ireland.)