The purpose of our investigation was not to question the safety of the technique but rather to investigate the respiratory depressive effects of neuraxial morphine in women after cesarean delivery. Opioid analgesics are well known to produce dose-dependent respiratory depression affecting both the hypoxemic and hypercapnic respiratory drives.1 The increase in transcutaneous carbon dioxide we observed after intrathecal opioid administration is consistent with the respiratory response to decreased minute ventilation after administration of opioids.1 Our findings are also consistent with those of Dalchow et al, who found an incidence of respiratory depression of 17.8% using transcutaneous carbon dioxide monitoring with a lower equivalent dose of intrathecal morphine than the current study.2
We agree with Drs Carvalho and Weiniger that clinically important respiratory depression is generally accompanied by bradypnea, hypoxemia, and sedation. Nonetheless, respiratory depression can also be defined as a blunted ventilator response to hypercapnia (monitored via capnography) and is well studied after opioid administration in the postoperative setting.3–6 We used a standard threshold for hypercapnia (transcutaneous CO2 > 50 mm Hg), but we agree that this degree of respiratory depression did not predict adverse respiratory events, supporting the idea that the threshold for hypercapnia and the length of hypercapnia that leads to adverse respiratory events require further study. It is likely that monitoring of transcutaneous CO2 has high specificity but low sensitivity for detection of respiratory compromise.
The hypercapnic events we observed were episodic but were likely not artifacts of monitoring. We agree that little is known regarding the normal physiologic recovery of ventilatory drive and effects on carbon dioxide levels in the immediate postpartum period. However, the sustained levels of transcutaneous CO2 (median duration of the longest event per patient was 25.6 minutes) and the higher starting baseline CO2 levels (median 38 mm Hg) in those who reached the hypercapnia threshold compared to those who did not argue against artifact. Although none of the patients who had hypercapnia had a respiratory event that required intervention, a subset of patients in our study had hypoxemia in association with hypercapnia. Seventy-six percent of women required additional opioid analgesia, but we do not know the degree to which these supplemental opioids contributed to our findings. It is common clinical practice to administer systemic opioids for breakthrough pain, as multimodal analgesia with intrathecal morphine and systemic nonsteroidal anti-inflammatory drugs may not completely mitigate postoperative pain. We acknowledge that systemic opioids may have contributed to the effect we observed, but this reflects common clinical practice, and therefore, we felt it was important to continue our usual postoperative analgesic regimen and measure carbon dioxide levels in our patients.
Our study should not be interpreted as a call to discontinue the use of intrathecal morphine. Instead, our results suggest that further study of the respiratory effects of intrathecal morphine is indicated, as well as study of the sensitivity and specificity of respiratory monitoring methods, particularly at different morphine doses and for high-risk patients who were not included in our study.