Fast Fourier transforms as prophecy: predicting hypotension during spinal anesthesia.

HYPOTENSION during spinal anesthesia for cesarean delivery has been of concern since the 1960s. Along with multiple strategies aimed at treating or preventing hypotension, some investigators have attempted to identify patients more likely to have hypotension in the hopes of targeting treatment. In this issue of ANESTHESIOLOGY, Hanss et al. report that measurement of heart rate variability (HRV), an assessment of sympathetic and parasympathetic balance, can identify those women at risk for spinal-induced hypotension. Heart rate variability investigations started in obstetrics, with the observation that changes in fetal HRV precede changes in actual heart rate in cases of intrauterine asphyxia. Power spectral analysis of HRV uses fast Fourier transforms to display power (variance) by frequency and reflects autonomic control of the cardiovascular system. Standards for performance and analysis of HRV have been published. Using spectral analysis of 2to 10-min electrocardiographic recordings, two main power “components” of variability can be identified: low frequency (LF; 0.04–0.15 Hz) and high frequency (HF; 0.15–0.4 Hz). The major contributor to HF variability is vagal efferent activity. LF variability is a result of parasympathetic and sympathetic outflow. The ratio of LF to HF power is an indication of the balance of sympathetic to parasympathetic influences. A larger LF/HF ratio is interpreted as reflecting higher sympathetic versus parasympathetic activity. HRV has clinical relevance: Decreased variability is associated with heart failure, mortality after myocardial infarction, and is an early sign of diabetic neuropathy. HRV is altered by pregnancy, preclampsia, and regional anesthesia or analgesia. The current report from Hanss et al. consists of two separate, closely related studies: The investigators first retrospectively determined a threshold LF/HF ratio related to the risk of development of hypotension and then prospectively confirmed its validity. HRV was assessed in 41 women at three separate times before elective cesarean delivery: the day before surgery, and the day of surgery before and after intravenous hydration. Based on the systolic blood pressure response to spinal anesthesia, responses were classified as mild (no systolic blood pressure 100 mmHg), moderate (lowest systolic blood pressure 80–100 mmHg), or severe (systolic blood pressure 80 mmHg or requiring more than 1 ml of the vasopressor mixture) hypotension. LF/HF ratios on the day before surgery were not significantly different between groups. However, on the day of surgery, before hydration, the patients “destined” for development of moderate or severe hypotension had significantly higher LF/HF ratios (median 2.8 for moderate, 2.7 for severe) than those who went on to have mild hypotension (median 1.2). After hydration, moderate-hypotension patients decreased LF/HF ratios to mild levels, whereas severe patients were unchanged. Based on these results, the authors prospectively studied 19 patients to examine the hypothesis that a LF/HF ratio of 2.5 or greater on the day of surgery would predict hypotension. This hypothesis was confirmed; patients with high LF/HF ratio had significantly more hypotension than the patients with low LF/HF ratios. Is it physiologically plausible that HRV parameters can predict hypotension during cesarean delivery? Are the criteria defined by Hanss et al. optimal? Can this type of technology be adapted to the routine clinical environment? Will it be? Should it be? Heart rate variability as a predictor of hypotension seems physiologically plausible. A recent report by Chamchad et al., using a retrospective protocol similar to the first part of the current report by Hanss et al., used a nonlinear mathematical method of analysis but also suggested that HRV predicts hypotension during spinal anesthesia. The underlying assumption that preexisting higher sympathetic activity indicates a higher risk of hypotension during anesthesia is consistent with classic teaching in anesthesiology. Other measurements that may reflect sympathetic activity, including systemic vascular resistance index, a “supine stress test,” and baseline heart rate, have been reported to correlate with the risk of hypotension. However, recent studies suggest that hypotension during spinal anesthesia is significantly less likely in preeclamptic patients than in healthy pregnant women, despite higher sympathetic tone and LF/HF ratios in preeclampsia. Have Hanss et al. determined and defined the correct criteria and threshold for prediction of hypotension? This seems unlikely to have resulted from this one study. Further work replicating, refining, and/or refuting these qualitative and quantitative findings will no doubt be necessary. It should also be noted that the HRV criteria This Editorial View accompanies the following article: Hanss R, Bein B, Ledowski T, Lehmkuhl M, Ohnesorge H, Scherkl W, Steinfath M, Scholz J, Tonner PH: Heart rate variability predicts severe hypotension after spinal anesthesia for elective cesarean delivery. ANESTHESIOLOGY 2005; 102:1086–93.