Safety of post-operative epidural analgesia in the paediatric population: A retrospective analysis.

BACKGROUND AND AIMS: Epidural infusion analgesia (EIA) is among the common procedures performed in children to provide analgesia. However, the administration of epidural is not without complications. Limited studies are available regarding the safety of EIA in children with no studies from the Indian subcontinent. The aim of this study was to analyse all the complications that occured during administration and maintenance of EIA in paediatric patients.
METHODS: All children undergoing elective or emergency surgeries under general anaesthesia and given concomitant epidural analgesia for post-operative pain management were included. Data were collected by reviewing patient medical records, anaesthesia registers and post-operative intensive care unit charts. Statistical averages were drawn to assess the complication rates.
RESULTS: Seventy children received epidural analgesia during the span of study, of them five were neonates and fifteen were infants. No major complications that were life-threatening or leading to permanent disability were documented. Two children (2.85%) had blood tap during procedure. Eleven children (15%) had peri-catheter leaks and 14 children (20%) had catheter dislodgements.
CONCLUSION: EIA seems to be a relatively safe method of providing analgesia. It is associated with the occurrence of complications which are at best temporary.

INTRODUCTION

Provision of safe and effective post-operative analgesia is a challenging task, especially in children. Despite being an effective method of providing post-operative pain relief,[1] epidural analgesia remains a frequently underutilised modality in children. The big question faced by paediatric anaesthesiologists is, ‘how safe are epidurals in children?’ Risk versus benefit information derived from adult epidural literature is often extrapolated to paediatric patients while obtaining consent from parents. Limited studies are available regarding the safety of epidural infusion analgesia (EIA) in children with no studies from the Indian subcontinent.[12345] Hence, we undertook a retrospective analysis of all complications that occurred during administration, maintenance and discontinuation of post-operative EIA of our hospital.

METHODS

All children undergoing elective or emergency surgeries under general anaesthesia and given concomitant epidural analgesia over 2 years were included in our study. There were no lower age limits imposed. The following age groups were decided upon: Neonates, infants (>1 month and <1 year of age), children 1–5 years of age and above 5 years of age. Neonates were categorised separately to determine whether neonatal EIA presented any particular challenges or risks.

Data were collected by reviewing patient medical records, anaesthesia registers and post-operative intensive care unit (ICU) charts. Statistical averages were drawn.

Standard protocols of pre-anaesthetic check-up, investigations and technique were followed. The lateral decubitus position with midline approach and 19-gauge Touhy needle and 22-gauge epidural catheter were used in most of the children (<3 years of age or weight <15 kg).

After identification of the epidural space, test dose of lignocaine with adrenaline (5 μg/ml) was given at a dose of 0.1 ml/kg. Bupivacaine 0.125% at 0.3 ml/kg was given intraoperatively to provide analgesia, half of the dose was repeated every 90 min. Postoperatively, continuous infusion of 0.0625% bupivacaine with fentanyl 2 µg/ml at 0.3 ml/kg was given to provide analgesia along with rectal paracetamol suppository, at 20 mg/kg. Fentanyl was not used in preterm babies. Maximum dose of bupivacaine per hour while under continuous infusion was limited to <0.2 mg/kg/h in neonates and infants and <0.4 mg/kg/h in older children. Post-operative pain was assessed by physiological monitoring of vitals in neonates and infants, and by Wong–Baker Scale and level of discomfort expressed by patients/parents in older children.

It was found that epidural catheter was kept in situ for 3–5 days depending on the requirement. Daily inspection of catheter site and assessment of epidural catheter function were done during the period.

Complications during administration, maintenance and removal of epidural were audited as follows: At the time of insertion - failure to identify the epidural space, blood tap, wet tap, epidural haematoma, during maintenance -nerve injuries, drug errors in the form of excess of local anaesthetics leading to central nervous or cardiac toxicity or in case of opioids causing respiratory depression, peri-catheter leak (identified by mild soakage of dressing applied at the site of insertion), catheter migration, infections and during removal -catheter breakage.

RESULTS

A total of 700 paediatric and neonatal surgeries were performed during the span of which 70 children (10%) received concomitant epidural analgesia. 49 (70%) children were male and the remaining 21 (30%) were female. The age distribution of children who received EIA is shown in Table 1.

Table 1

Age distribution of the children receiving epidural

Of these 70 epidurals administered, 16 were for abdominal surgeries, 52 were for urological surgeries and 2 were for thoracic surgeries. All surgeries were major surgeries with an average duration of time ranging about 3–5 h.

34 were low thoracic epidurals inserted at the level of T9-T11 and 36 were inserted at lumbar level L1-L5 level. After identification of the space, 4–5 cm of the catheter was left in situ. 57% of children received EIA during the 1st year and 42% received during the 2nd year of the study. The percentage remained fairly constant every year.

Majority of complications documented were during the maintenance of EIA. The total complications that were encountered are mentioned in Table 2. 14 (20%) children had catheter disconnections categorised separately and not as complications.

Table 2

List of all complications encountered in the audit

DISCUSSION

Epidural analgesia is an effective method of providing pain relief to children in the post-operative period, has widespread use and offers more benefits over conventional methods. The main factor which has limited its usage is difficulty in assessing the safety of the procedure. The process of obtaining informed consent from parents has been hampered by the lack of data on the incidence and severity of complications associated with the technique and also the possible consequences of these.

Studies have shown that administration of epidurals is not without complications. Adult studies looking at spinal and epidural techniques suggest an incidence of permanent neurological injury of 2 to 7: 10,000 epidurals and of transient neurological injury of 1 to 8: 10,000.[67] Local anaesthetic toxicity was described with an incidence of 1:10,000 patients. However, risk factors reported in adults need not be observed in the paediatric population and extrapolations may not be correct.

Our review, though limited has allowed us to quantify at least to some extent the magnitude of risks associated with post-operative epidural analgesia in paediatric surgical patients.

Llewellyn and Moriarty calculated an incidence rate of serious complications as 1 in 2000 in a large national audit of 10,663 children who received epidural infusion.[1] Flandin-Bléty and Barrier identified an incidence of complications of approximately 40:10,000, with five severe neurological complications and three deaths.[2] Giaufré et al. suggested an incidence of 15: 10,000 in single-shot techniques, including two incidents of post-dural puncture headache, three intravenous (IV) local anaesthetic overdoses, one incident of arrhythmia and two transient neuropathies, together with more minor technical problems.[3] These figures from large paediatric studies demonstrate that serious incidents which have the potential to cause severe or long-term harm fortunately are rare.[4]

Similarly, serious complications such as epidural haematoma, nerve injury and local anaesthesia toxicity or infections, which can lead to mortality or permanent disability, were not documented in our study. Other smaller audits such as ours and Kotzé et al. reported no major complications.[5] Our study recorded two blood taps (2.85%) while performing the procedure. In both of these patients, epidural catheter placement was successful through the next higher space with no other consequences. Kotzé et al. reported 2/46 (4.34%) cases of a single blood tap, followed by an uneventful second attempt.[5] Complications such as blood tap, dural puncture and failure to identify correct space to a major extent can be reduced by experienced anaesthetist's supervision, utmost precautions and are less in high volume centres.

Our study included 7.14% of neonates and 21.4% of infants. However, we did not have any complications specifically occurring in neonates and infants. Ecoffey et al. reported that complications are 4 times more frequent in children aged <6 months than in children aged more than 6 months, despite the fact that the younger patients are probably managed by the most experienced paediatric anaesthesiologists with maximum precautions.[4] Our results may be attributed to the smaller sample size of our study, lower doses of the drugs used, avoidance of fentanyl in preterm babies and monitoring of all babies in exclusive tertiary level paediatric ICU, whereas in the larger audits, the patients were from various centres with wide geographical spread, broad range of specialities and levels of expertise.

One child (1.42%) in the group had a transient episode of bradycardia while under continuous infusion of 0.125% bupivacaine. Immediately, epidural infusion was terminated, IV atropine 20 μg/kg was given and level of sensory and motor blockade was assessed. After reconfirming that there was no excess dosage (0.4 mg/kg) or undue high level of block, the infusion was again restarted and no further episodes were noted. Likewise, other studies also reported 0.048% incidence of single electrocardiogram changes or transient arrhythmias in their audit, which, however, required no active treatment.[4]

Our study recorded the need for rescue analgesia in four children (5.71%). When they complained of pain, the catheter was rechecked for position and adjusted, if required. As there was still no change in pain, IV fentanyl bolus 1 μg/kg followed by 0.5 μg/kg/h was instituted. Other smaller studies also reported a higher incidence of moderate–severe pain (18%) and severe pain (2%) in their patients.[5] This stark contrast could be because their study included only patients of empyema undergoing thoracotomy, whereas our study comprised of non-infected paediatric urological surgeries predominantly.

Peri-catheter leak is a unique complication noted in our study seen in 11 (15.71%) children, which has not been documented in other reviews. Minimal soakage of the epidural site dressing with infused solution was noted in these patients, but it did not result in need for rescue analgesia nor result in local site infections. The most plausible explanation for the peri-catheter leak may be the discrepancy between the size of needle (19-gauge) and size of catheter (22-gauge), which was a manufacturing limitation.

Reports suggest catheter dislodgement rate of 4.7%,[5] while Dolin et al. suggested a rate of 5.7%.[8] Our study did not have any catheter dislodgements; nevertheless, the most common problem encountered in our study was catheter disconnections which was seen in 14 (20%) patients. These disconnections were at the junction between epidural catheter and connector or between filter and high-pressure tubing from infusion pump. These disconnections can be graded as minor complications as they did not harm the patient but required frequent anaesthetist reviews and increased the nursing burden.

It appears that both peri-catheter leaks and catheter disconnections may be avoided by the availability of age-specific epidural catheters at least for the neonates and infants.

Although there are concerns regarding the safety of administration of epidural while children are anaesthetised, none of our children had any post-operative neurological deficits. Hence, we concur with the conclusion of other previous studies that the complication rates are low despite most of the epidurals being performed under general anaesthesia or heavy sedation.[4] Studies even recommend that due to the significant differences between children and adults with respect to self-control and the ability to communicate effectively, general anaesthesia or heavy sedation should not be considered an absolute contraindication to regional anaesthesia in children.[9]

Our study did not record any local site infections or epidural abscesses. Darchy et al. concluded that majority of local infections and catheter colonisations in critically ill patients are caused by skin flora.[10] An audit also showed epidurals were not associated with infection even in presence of empyema.[5] The authors suggest that the most important controllable factor in the prevention of catheter-related epidural abscesses is the sterility of the site and that equipment and meticulous aseptic technique can certainly make a difference.[1112]

A retrospective study such as ours is not without its limitations. The sample size is too small to generate accurate data that can be extrapolated to whole population. The study particularly reflects the practice of the institution and cannot be generalised. The main limitation was absence of any major complications which would require multicentre enrolments to get accurate incidence value.

CONCLUSION

EIA seems to be a safe and effective method of providing analgesia to children. However, all children should be monitored in tertiary level dedicated paediatric ICU to improve safety profile.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.