Neuraxial opioids as analgesia in labour and postoperative after caesarean section and hysterectomy: A questionnaire survey in Sweden

Introduction

Intrathecal and epidural morphine improve postoperative analgesia after caesarean section (CS) and hysterectomy and intrathecal labour analgesia^1–3. In 1981, the Swedish Society of Anaesthetists conducted a nationwide survey of experience with intrathecal and extradural opiates⁴. They found intrathecal morphine was administered to only 90–150 patients and ventilatory depression requiring treatment with naloxone was needed in six of these patients. Since then, the use of intrathecal and epidural morphine has expanded, with a decrease in the doses used⁵. Still, the use of intrathecal and epidural morphine in these patient categories vary and the general use in Sweden at present is unknown. Spinal morphine may have undesirable side effects, such as nausea and vomiting, with the most feared side effect being respiratory depression, which is why extended postoperative monitoring is required⁶. Extended postoperative monitoring demands personal resources and is sometimes difficult to arrange, which may influence the use of spinal morphine.

The aim of the present questionnaire survey was to assess the use of intrathecal and epidural opioids in obstetric and gynaecological patients, factors that limit/holdback its use and monitoring routines implemented for patients treated with opioids.

Methods

This study was conducted in accordance with the principles outlined in the Declaration of Helsinki. Ethical committee approval was not applied for the present study, since the survey concerns only clinical practice and routines and not patient data in accordance to Swedish ethical board guidelines. The permission to carry out the research was obtained from the head of department of each hospital.

Results

Discussion

We found that all units used opioid supplementation and two thirds of responding units of Swedish obstetric anaesthesia used spinal/epidural morphine as adjunct to local anaesthesia for perioperative care of CS and hysterectomies. Spinal use was rather uniform with 100 µg for CS and up to 200 µg for hysterectomies of intrathecal morphine combined with a low dose of fentanyl or sufentanil. Epidural morphine was used to a lesser extent. Opioids were also commonly added to labour analgesia as sufentanil, but no unit uses morphine as an adjunct in labour analgesia. The common reason for withholding spinal/epidural opioid was the risk for respiratory depression, and thus the demands for post procedural monitoring.

There were some obvious differences in practice in the different units. Winther et al. found likewise inconsistencies in clinical guidelines for obstetric anaesthesia for CS⁷. The most effective method for providing pain relief during labour is epidural analgesia and there are good opportunities to get a beneficial epidural in Sweden⁸. In countries with limited resources, a single shot spinal anaesthesia may be a feasible option. Combination of low dose morphine, fentanyl and bupivacaine or morphine, sufentanil and bupivacaine are suggested to achieve effective analgesia with a prolonged effect⁹. However, we were previously unable to find a major advantage of the addition of morphine, when comparing 0, 50 or 100 μg morphine added to 1.25 mg bupivacaine and 5 μg sufentanil during established labour, as this did not show a significant increased duration of analgesia¹⁰. Morphine is not a preferred drug for labour epidural analgesia and questions about the addition of opioids other than morphine to labour epidurals were not included in this survey. Adding sufentanil or fentanyl is common practice, and may be seen more or less as the gold standard^11,12. Sufentanil is the most common opioid added in labour anaesthesia in Sweden, which was reported in a national obstetric anaesthesia meeting from a survey in 2009 (not published).

Spinal anaesthesia is the preferred analgesia in CS when a working epidural is not in place for conversion and top-up. Long acting spinal opioids as a component of the CS spinal anaesthesia has been proven superior as the post-caesarean analgesia over systemic counterparts, and made them a commonly used part of multimodal analgesic regimes¹³. In Sweden, the long acting opioid most commonly used in CS is morphine, and we found that 63% of units used intrathecal morphine as routine in CS spinal analgesia. Since larger units with higher numbers of CS more commonly use intrathecal morphine, the impact of the routine is even more pronounced. The units using intrathecal morphine as routine adjunct to local anaesthesia in CS, covered an estimated 73% of all CS annually performed in responding units. All responding units used intrathecal morphine doses of 100–125 µg. Palmer et al. found in a dose response study that morphine at 100 µg added to hyperbaric bupivacaine (12.75 mg) provided analgesia comparable to that provided by doses as high as 500 µg¹. The optimal effective dose for CS is not well defined. The balance between analgesia and side effects must be taken into account. The occurrence of pruritus, has been found to be dose related^1,6,14. For nausea and vomiting the relationship is not that clear but nausea and vomiting is seen in a higher frequency in higher doses than in lower dose groups¹⁴.

Intrathecal morphine is suggested to provide better pain control after CS than opioid-free epidural analgesia¹⁵. Intrathecal morphine has also been shown superior regarding postoperative pain relief, as compared to abdominal wall block (TAP)¹³. The more lipophilic diamorphine was the most commonly used opioid in a survey conducted in the UK in 2008⁵. Diamorphine has a more rapid onset of action than morphine because of a high lipophilicity (octanol-water coefficient =280)¹⁶. Diamorphine is metabolised to morphine with long-acting activity; however, diamorphine is not registered/available as pharmaceutical drug in Sweden¹⁶. Fentanyl and sufentanil are the available lipid-soluble opioids used, which have a quick onset for improved perioperative spinal and epidural analgesia¹⁷. A combination of one lipid-soluble opioid and the long-acting water-soluble morphine is added to hyperbaric bupivacaine, which is found to be commonly used in Sweden. It is an attractive combination providing a quick onset with improved perioperative quality and a long- acting postoperative analgesia¹⁶. Yet currently, the present study found that twelve units use solely fast opioid adjuncts to local anaesthesia and two of the units add morphine to local anaesthetics with no addition of fentanyl or sufentanil, but one of the units stated they plan to start adding fentanyl shortly.

When a functional labour epidural is in place, it’s recommended to convert this to a full epidural anaesthesia if an emergency CS becomes necessary¹⁸. Twelve of the 32 units add morphine to the epidural local anaesthesia, the most common dose is 2 mg. Singh et al. found epidural morphine at 1.5 mg provided none-inferior post caesarean analgesia and caused fewer adverse effects compared with 3 mg epidural morphine¹⁹. Still the majority of responding units chose not to add morphine in epidural analgesia for CS postoperative analgesia. The addition of sufentanil (15 units) or fentanyl (9 units) was more common, together covering 74% of CS performed annually in responding clinics. Malhotra et al. found no support for adding 75 µg fentanyl in top-up epidural for CS regarding time to onset nor quality of analgesia in women already receiving epidural fentanyl during labour²⁰. Yet, the addition of epidural opioids has been proven to enhance postoperative analgesia after CS with earlier onset^21,22.

The addition of morphine intrathecal or epidural as adjunct for intra as well as postoperative pain relief is also regarded by many as the ‘gold-standard’, due to its positive analgesic interaction with local anaesthetics and prolonged duration of action⁶. Spinal anaesthesia with added morphine was used in 63% of responding units for hysterectomy perioperative analgesia. The most common dose used was 200 µg, which is the well in line with the optimal dose found in a study of the three doses (100 µg, 200 µg and 300 µg) versus placebo in abdominal hysterectomy². Intrathecal morphine use in hysterectomy varied regarding type of surgical technique: five units chose to use intrathecal morphine in abdominal, vaginal and laparoscopic hysterectomies, while seven units use intrathecal morphine in abdominal and vaginal, but not in laparoscopic, hysterectomy.

All responding units had guidelines regarding postoperative monitoring after neuroaxial opioids. The routine of dividing the postoperative monitoring between a more monitoring intensive postoperative ward for the first 2–6 hours and a regular surgical ward for the following hours, up to 12 hours, according to the guidelines of SFAI, was common, and was found to be already routine in a European survey in 1996 by Rawal et al²³. In eight units, CS were initially monitored for 2–6 hours in the obstetrical ward. Two units using intrathecal morphine for CS monitored postoperatively only in a regular maternity ward for the whole period. Seven of the eleven units that chose not to use spinal/epidural morphine describe a risk of respiratory depression and difficulties to monitor postoperatively as the main reason for withholding its use. In these units, which do not use morphine because of risk for respiratory depression and difficulties in postoperative monitoring, approximately 15% of CS reported are performed.

Respiratory depression is the most serious side effect associated with neuroaxial morphine; however, its occurrence is rare⁶. In 2010, the subject “All patients receiving neuraxial morphine should be monitored with continuous pulse oximetry” was debated during Controversies in Obstetric Anesthesia²⁴. It was commented that a requirement for pulse oximetry might decrease the use of neuroaxial morphine as post caesarean analgesia and risk more pain in patients with more parenteral morphine consumed with the risk of increased risk of respiratory depression⁶. We have in our institution some 15 years’ experience of approximately 2000 annual CS performed, the majority in spinal anaesthesia using a combination of fentanyl (10 µg), morphine (100 µg) and hyperbaric bupivacaine (10–12 mg). Our routine is an initial 2–3 postoperative hours monitoring in a postoperative ward, and the following hours up to 12 hours in the maternity regular ward with midwife hourly supervision of level of consciousness and in case of sedated (sleeping) patient monitoring of respiratory rate. In the case of an unstable patient, morbid obesity or known sleep apnoea, the patient stays in the postoperative department for a longer time, determined on an individual basis. In line with responding units in this survey, we have no known experience from severe respiratory depression.

The return rate of the questionnaire was 32/47 units (68%), a figure in line with previous questionnaire survey around anaesthesia practice in conjunction to obstetrics^25,26. Taking into account the size of the different units, the response routines cover approximately 82% out of 111,364 annual deliveries, as seen in The Swedish Medical Birth Register (2013), and 83% of CS are performed in responding units. All units performing more than 400 annual CS returned the questionnaire. Our main interest was to find the routine use of morphine in CS, since we learned from anaesthesia meetings with colleges that this routine, regarded as “gold standard” was not in use in several units, due to fear of respiratory depression⁶. We included questions regarding spinal/epidural addition of fentanyl and sufentanil as well, but did not in detail ask for monitoring after these adjuvants when morphine was not included. We asked for numbers of performed CS and hysterectomies, including morphine, for each unit. However, those numbers were not possible to extract from the questionnaire because of few answers. When the majority of anaesthesia units are reporting anaesthesia to the Swedish PeriOperative Register (SPOR), hopefully in the coming years there will be better opportunities to answer questions about common routines and outcomes.

Conclusion

The use of neuraxial opioids is widely spread in Sweden; however somewhat varying regimes exists, where some units chose to use only lipophilic opioids with a rapid onset and few use only long-acting water-soluble morphine. However, a majority use a combination together with local anaesthesia. Still in some units the use of morphine is restricted because of the concern for respiratory depression and difficulties in monitoring.

Data availability

Dataset 1: Raw data for IT and EDA opioid survey. IT, intrathecal; EDA, epidural; MO, morphine. doi, 10.5256/f1000research.10705.d151418²⁷