Minimal Effective Weight-based Dosing of Ondansetron to Reduce Hypotension in Cesarean Section Under Spinal Anesthesia

A Randomized Controlled Superiority Trial

Maliwan Oofuvong; Thitikan Kunapaisal; Orarat Karnjanawanichkul; Nussara Dilokrattanaphijit; Jaranya Leeratiwong


BMC Anesthesiol. 2018;18(105) 

In This Article


In this study, we found that a weight-based dose of prophylactic ondansetron did not influence the incidence of maternal hypotension. The results of this study were consistent with two other studies[8,13] but were in contrast with others.[6,7,9,10] Two recent systematic reviews[14,15] showed ondansetron reduced the incidence of hypotension (relative risk ratio [RR] = 0.63, 95% confidence interval [CI]: 0.45–0.88 and RR = 0.67, 95% CI: 0.54–0.83, respectively) and bradycardia (RR = 0.31, 95% CI: 0.19–0.50 and RR = 0.49, 95% CI: 0.28–0.87, respectively) after spinal anesthesia in cesarean section even though a low quality of evidence was reported.[15] However, the minimal effective weight-based dosing of ondansetron in parturients was not determined.[9,10,14,15]

In our study, ondansetron had little effect on hemodynamic stability after subarachnoid block in parturients. Factors that impact the level of anesthesia and analgesia given to the patient might also affect the patient's arterial pressure as well as the way to measure arterial pressure. There are some possible reasons for the difference in the incidence of hypotension compared to other reports. First, we did not monitor invasive arterial blood pressure or cardiac output. Real time arterial pressure monitoring may be better at detecting actual variations of blood pressure than noninvasive blood pressure monitoring although we measured blood pressure every minute during the first 15 min after spinal anesthesia. A noninvasive blood pressure measurement might not be sufficiently sensitive to notice hemodynamic changes to promptly treat hypotension in parturients, especially in preeclamptic patients.[16] Langesaeter et al.[17] and Rosseland et al.[18] showed that continuous invasive arterial pressure monitoring during cesarean section under subarachnoid block provided hemodynamic stability in parturients, especially when vasopressor was given. Second, vasodilatation from subarachnoid block might have a stronger influence on blood pressure than the Bezold–Jarisch reflex in our study. From the results, 0.05 mg of ondansetron caused more episodes of hypotension before delivery compared to the control group (P = 0.02), especially at 4 and 9 min after spinal anesthesia (Figure 2) which led to failure to prevent the Bezold–Jarisch reflex. Therefore, ondansetron may have little effect on preventing hypotension in these cases since we administered a vasopressor to treat vasodilatation immediately after hypotension occurred causing no differences in the incidence of hypotension among the three groups. We recorded blood pressure after a subarachnoid block was performed every minute for 15 min. If hypotension occurred, ephedrine or norepinephrine was administered instantly until the MAP reached the lower limit of baseline level. The peak onset of ephedrine or norepinephrine is approximately 1–2 min[19] and repeated doses were given as needed. Third, the weight-based dosing of 0.1 mg/kg of ondansetron might not be adequate to maximize the effect to prevent the Bezold–Jarisch reflex. A high dose of ondansetron (12 mg) or 0.15 mg/kg of ondansetron was reported to increase MAP in patients who had a cesarean section under spinal anesthesia compared to the control group.[9,20] In our study, the mean dose of 3 to 7.9 mg of ondansetron was given which might not be adequate to prevent the Bezold–Jarisch reflex even though some studies found a lower dose of ondansetron (4 or 6 mg) could significantly prevent hypotension in the same setting.[10,20] However, we were concerned that the dose of 0.15 mg/kg of ondansetron might be harmful to the fetus in terms of possible umbilical arterial vasoconstriction due to ondansetron acting on the 5-HT1B receptor.[21] Fourth, our analysis was based on an intention-to-treat principle which included repeat administration of subarachnoid blocks following failure of the first spinal anesthesia. A repeat dose of subarachnoid block can affect the level of anesthesia and analgesia, although there were no significant differences in the number of blocks given between the three groups (P = 0.87).

The definition of hypotension used in this study was a decrease in the MAP by 30% from the baseline level which was quite similar to many studies.[7,8,22,23] MAP was used instead of SBP to define hypotension because MAP provides a physiologically more appropriate measurement of hypotension than SBP.[24] The overall incidence of hypotension among women undergoing cesarean section following spinal anesthesia in our study (80%) was lower than reported in a previous study done in Songklanagarind Hospital (85%).[2] One-third of each intervention group received metoclopramide as premedication. Even though a case report showed 10 mg of metoclopramide could cause severe hypotension,[25] the baseline blood pressure before spinal anesthesia was not different among the three groups (Figure 2).

Ondansetron did not cause any adverse events in our study. The parturients experienced neither QT abnormalities nor burning sensation during the injection and the newborns did not experience bradycardia or hypothermia 24 h postoperatively. The incidence of urticaria or arrhythmia was not higher in the two ondansetron groups. Although the incidence of vomiting was lower in the high dose ondansetron (0.1 mg/kg) group, the incidence was not statistically different (P = 0.09). However, postoperative metoclopramide requirement was significantly lower in the high dose ondansetron compared to the normal saline group (P = 0.01). Therefore, the effect of ondansetron to prevent or treat nausea and vomiting was achieved as expected. There were no significant differences in terms of hypotension or tachypnea in newborns between the three groups. The incidence of tachypnea was slightly higher among the newborns of women who received the low dose of ondansetron compared to the control group (7% vs 2.8%). Tachypnea resolved within 2–3 h after oxygen therapy was administered. Although ondansetron can transfer across the placenta to newborn babies,[26] adverse events related to ondansetron in newborns have not been reported.

Strengths and Limitations

The strengths of this study were reduced selection bias due to the randomized treatment allocation and the triple-blind design (patients, investigators, and statistician). We used the intention-to-treat principle in the analysis which helped decrease the bias and ensure a balance of patients between the three groups. There are two weaknesses of this study. First, several resident anesthetists performed the spinal block which might have affected the level of anesthesia and incidence of hypotension in the study participants. However, these residents had a high level of experience (> 20 spinal blocks) and a fixed rate of subarachnoid injection (0.2 mL/s) was used in all participants. Second, we did not monitor invasive arterial blood pressure because invasive monitoring is not routinely used in healthy parturients in our setting and it is also costly. However, the accuracy of the study is high and reproducibility of the results for a similar setting (a tertiary care hospital) should be achievable even though the study was conducted in a single hospital.