Is epidural analgesia non‐inferior to intrathecal fentanyl as initiation for neuraxial analgesia in early non‐spontaneous labour?

Intrathecal fentanyl, using the combined spinal‐epidural (CSE) technique, provides rapid analgesia during early labour. Because of the technique's more complex and invasive nature, as its replacement we assessed the use of epidural analgesia in primiparous parturients with induced labour. The study was registered at www.clinicaltrials.gov (NCT04645823). The aim was to compare the efficacy, duration of analgesia and maternal satisfaction. The primary outcome was the difference in pain visual analogue scale (VAS) between the interventions at 20 min after the analgesia administration.

Conclusion: After 20 min, epidural analgesia by lidocaine and fentanyl was within the non-inferior threshold compared with intrathecal fentanyl in efficacy.The duration of action was not shorter than that of intrathecal fentanyl and maternal satisfaction was also similar.

K E Y W O R D S
epidural analgesia, fentanyl, labour analgesia, labour induction

Editorial Comment
The combined spinal and epidural technique requires a longer time to place than the standard epidural placement, even in skilled hands.The number of attempts needed for success for combined spinal-epidural is higher than that for a standard epidural catheter increasing the time the parturient must be positioned during her contractions.In this non-inferiority study the analgesic efficacy of epidural lidocaine with fentanyl was compared with intrathecal fentanyl in parturients undergoing induced labour.The authors concluded that adding fentanyl to lidocaine epidurally is non-inferior to intrathecal fentanyl in terms of intensity of the analgesia provided at 20 min after intervention.Epidural analgesia with fentanyl results in less pruritus, is less invasive, and requires less equipment.It may therefore be a favourable alternative for analgesic technique for parturients in non-spontaneous labour.

| INTRODUCTION
Initiation of labour analgesia by intrathecal fentanyl using the combined spinal-epidural (CSE) method has been shown to be an effective rapid-onset mode of analgesia in the early stage of labour. 1,2Additional benefits of using the CSE technique in early labour include a lack of motor block allowing ambulation and the possibility of limiting the total dose of opioids over the course of delivery, when the first dose is given intrathecally. 2,3An increasing number of deliveries are being induced, and these parturients often experience painful contractions early on during delivery. 4,5[8] Despite the rapid onset and expected high efficacy of analgesia, the CSE technique has some disadvantages.It involves an intentional perforation of the dura, thus making it more invasive compared with an epidural analgesia, and it also requires more time and skill and more expensive special needles. 2,9For these reasons, we have sought to replace the intrathecal fentanyl with an epidural analgesic in our hospital, where the CSE labour analgesia was used for over 2000 parturients in 2022.
Motivated by a prior study demonstrating the efficacy of epidural fentanyl, 10 we chose to study it further in primiparous parturients and found that the 100 μg dose of epidural fentanyl expected to be equipotent with a 20 μg intrathecal fentanyl dose resulted in median visual analogue scale (VAS) values of 53 mm compared with 10 mm after the intrathecal fentanyl. 11This difference was also evident in parturient satisfaction with analgesia with only 40% of the parturients in the epidural fentanyl group rating over 92/0-100 VAS values for satisfaction with analgesia as opposed to 70% in the intrathecal fentanyl group.However, the efficacy of intrathecal fentanyl is in some cases too high, resulting in totally pain free parturients who report annoying pruritus.Therefore, a continuation for the prior study was planned where the effect of epidural fentanyl would be improved by adding a rapidly acting local anaesthetic, lidocaine, into the epidural intervention.
Since the intrathecal fentanyl analgesia offers clinical use mainly in parturients undergoing non-spontaneous labour, we studied the effect of epidural analgesia containing both lidocaine and fentanyl and compared it with intrathecal fentanyl in this clinically relevant parturient cohort.The hypothesis in this non-inferiority study was that adding lidocaine to the epidural fentanyl would result in analgesia that was clinically comparable to intrathecal fentanyl.Given the prior experiences with pruritus, invasiveness, and material costs, it was considered sufficient if the epidural analgesia would result in pain VAS that was 20 mm above the level seen after intrathecal fentanyl assuming that the maternal satisfaction with the analgesia is not compromised.The initial inclusion criteria were: primiparous, ASA category 1 or 2, body mass index (BMI) <40 kg/m 2 , no contraindications for the neuraxial analgesia, age 18 years or older, sufficient knowledge of Finnish language, and non-spontaneous labour.Additional criteria before the randomisation step and the actual study intervention were: numeral rating scale value of 8-10/10 for pain during contraction, cervical dilatation of 5 cm or less, and no systemic opioid analgesics for 2 h before neuraxial analgesia.

| MATERIALS AND METHODS
Of the 95 patients who gave their consent, 60 parturients proceeded to the randomisation and neuraxial analgesia step according to the initial plan.
An online tool was used to generate a random sequence of 60 number 1s and 2s in 1:1 ratio corresponding to the two treatment arms.The randomization sequence was generated by a person not involved in the study who also prepared the randomization envelopes.
The randomization key was stored on a flash drive along with a printout of the randomization list in a sealed envelope until the last parturient had delivered.The group allocation consisted of sealed non-transparent consecutively numbered envelopes.For the randomization step, the anaesthesiologist performing the intervention took the next available envelope and opened it in the labour room.All interventions were completed by senior anaesthesiologists working full-time in our hospital where CSE analgesia is used for over 2000 parturients per year.The parturient and the assessor remained blinded regarding the intervention.The analgesia provider who was not blinded regarding the procedure was advised not to disclose the provided analgesia type to anyone.
For both intervention groups, the back was washed with 70% coloured alcohol solution and the lumbar L2/L3 or L3/L4 interspace was identified in either sitting or lateral position.Lidocaine 1% was used for the local anaesthesia.For the intrathecal analgesia, a needlethrough-needle CSE technique (CSEcure, Smith Medical, USA) with a 27-gauge pencil point needle was used, and thereafter an 18-gauge epidural catheter was placed and secured to the back without a test dose.For the epidural group, the same needle set was used but the spinal needle was not inserted, and the epidural study solution was injected through the epidural catheter.The study drug solutions were prepared by the analgesia provider, which is a standard operating practice in our hospital.
For the intrathecal fentanyl intervention, a dose of 20 μg fentanyl was diluted to a volume of 2 mL with saline.For the epidural intervention, a volume of 8 mL of 1% lidocaine was mixed with a 100-μg dose of fentanyl to a total volume of 10 mL, which was given through the epidural catheter.In both interventions, the epidural catheter was then flushed with 2 mL of saline.
The analgesia providers informed the parturients that they may feel something at the time of either the spinal needle insertion or the epidural catheter insertion.The same information was given again either during the insertion of the epidural catheter for the CSE group or removal of the epidural needle in the epidural group to ensure blinding of the midwife who was requested to observe the foetal heart rate monitoring (cardiotocography, CTG) during the procedure.
An anaesthesiology resident who had not worked in our institution, but was familiar with both analgesia procedures, participated in the study and could not identify from the procedure or the analgesic effect which intervention she had received.
Following analgesia administration, the anaesthesiologist performing the analgesia confirmed on the case report form that the analgesia was provided as indicated in the randomization ticket (yes/no) and recorded the number of puncture attempts and the lumbar interspace used, along with the time when the analgesia was given, before exiting the room.
Thereafter, the assessor entered and asked the parturient to mark her subjective pain during the latest contraction on a 100 mm VAS line every 5 min for 30 min.At the end of the 30-min observation period, the parturient was asked to assess her satisfaction with the analgesia on a 100 mm VAS line (0 mm: complete dissatisfaction; 100 mm: complete satisfaction).At this time the parturient was also asked how her legs felt and if she could raise her upper leg.The parturient was also asked to grade pruritus on a five-step Likert scale (none-mild-moderate-severe-unbearable) every 5 min.
At the end of the 30-min study period, the parturients were offered ropivacaine 0.1% 20 mL epidurally according to the standard hospital practice.Those parturients who were satisfied with the analgesia were given identical instructions to inform the midwife at the latest when the NRS for pain reached 6/10 to receive their first topup dose.For the top-up dose, ropivacaine 20 mg and fentanyl 0.1 mg in a total volume of 20 mL was used from 45 min post-intervention onwards.The time when the parturient requested the additional dose was recorded as the effective time for the initial analgesia intervention.
The total number of subsequent epidural doses and epidural drug consumption for ropivacaine and fentanyl were recorded.
In the assessment of the stage and progression of labour, the time from the beginning of regular contractions until full cervical dilatation was considered as the first stage of labour and the time from full cervical dilatation until delivery as the second stage of labour.Cervical dilatation before analgesia intervention and the next measured cervical dilatation were recorded, along with the time of measurement, to assess the average rate of cervical dilatation at the time of intervention as previously described. 12e foetal heart rate monitoring was recorded with CTG before and after any neuraxial analgesia.The CTG recordings were classified according to the FIGO guidelines 13 as normal, suspicious, or pathological both before the intervention and during the 30-min period following the intervention, by an obstetrician (RJ) blinded to the intervention.Furthermore, possible interventions to improve foetal wellbeing, such as a change of maternal position, performed due to CTG abnormalities during the 30-min period were recorded.
The outcome of the labour was recorded as normal vaginal, instrumented vaginal or caesarean section (C-section).
The actions taken to induce or augment labour (such as artificial rupture of the membranes, dilatation of the cervix with a balloon catheter, use of oxytocin or oral misoprostol tablets) were retrospectively recorded from the patient files.
The material costs for each of the intervention arms consisted of the cost of the CSE-or epidural set, the prepacked needle-syringeswab-drape-set, syringes, needles, and the medicine costs according to the cost level for 2022.

| Outcome measures
The primary outcome measure was the difference in contraction pain at 20 min after the analgesia.The 20-min time point was chosen based on our prior study indicating that the maximum achievable analgesia with intrathecal opioids is reached by 15-20 min after the intervention. 11The secondary outcomes were the time until the next epidural dose and maternal satisfaction with the analgesia at 30 min after the neuraxial analgesia.

| Statistical analysis of the data
For the analysis of the absolute pain difference in outcomes, the pain VAS values for the intrathecal fentanyl (standard treatment) were deducted from the VAS values after epidural analgesia (the experimental treatment).A similar approach was used for the two secondary outcomes (duration and satisfaction).Thus, if intrathecal analgesia was more potent than epidural analgesia, positive values would be attained.The absolute differences were used to calculate the mean difference and the two-sided 95% confidence interval which was used to determine non-inferiority.For pain, the non-inferiority line was set at 20 mm above and the theoretical superiority line at 20 mm below the 0 mm no-difference line.No non-inferiority margin was set for the secondary outcomes for which the aim was to have the higher end of the 95% CI for difference to not fall below zero marking inferior performance compared with intrathecal fentanyl.Finally, the presence of the 95% CI within the required boundaries was verified by individual one-sided t tests (TOST) against the boundaries where P < .001was set as the significance level.
The group size calculation for non-inferiority setup with α set at 0.05 and β at 0.10 required a minimum of 27 parturients per group to detect the clinically significant 20 mm difference between the groups. 14We chose to recruit 30 parturients in each group to accommodate possible technical failures in some parturients.
All parturients who proceeded to the randomization step received their allocated treatment and the dataset is analysed for all 30 parturients per group as per protocol unless otherwise specified due to missing data for some datapoints.
All normally distributed data are expressed as mean (95% CI) and non-normally distributed data, (pain and satisfaction raw VAS scores), are expressed as median (inter-quartile range).Where appropriate, one-way ANOVA was used for normally distributed data and Mann-Whitney U-test was used for non-normally distributed data.Categorical data are expressed as N (%) and intergroup comparison was done by using the χ 2 test except for multiple ordered outcome option-data the Mantel-Haenszel test was used.All statistical analysis were performed with SPSS (version 25, IBM, Armonk, NY), and P values of <.05 were considered statistically significant for the main outcome while P values of <.025 were considered significant for the duration of analgesia and maternal satisfaction.
The study was planned to estimate the efficacy of epidural analgesia in reference to intrathecal fentanyl and, apart from pruritus and costs, no difference in adverse effect profile was expected.A P value of <.05 is considered significant in adverse effect comparisons.

| The parturients
All randomized parturients received their allocated treatment and there were no dropouts or change over between the treatment arms.
A flowchart of the patient flow is shown in Figure 1 and the parturient demographics in Table 1.The groups were homogenous in terms of the groups' baseline characteristics and stage of labour at the time of neuraxial analgesia.The time from the onset of regular contractions and the cervical dilatation state were similar in the majority of parturients, with cervical dilatation of 3-5 cm before neuraxial analgesia.In the epidural arm the catheter insertion was technically successful on the first attempt in 26 (87%) of the parturients, whereas the adequate approach to the epidural space to facilitate clear cerebrospinal fluid flow for the CSE intervention took more than one attempt (range 2-5) in 12 (40%) cases (P = .02).

F I G U R E 1
The CONSORT flowchart for the patient flow in the study.2. After epidural analgesia, the VAS values were 3.3 (À0.06 to 6.66) mm higher than following intrathecal fentanyl.Thus, epidural analgesia appears to result in slightly higher pain VAS scores than intrathecal fentanyl at 20 min, but the efficacy is still well within the clinical non-inferiority area.A graphical presentation of the assessment is shown in Figure 3.

| Duration of analgesia
Six parturients (20%) in both treatment groups requested additional analgesia within 15 min after the initial 30-min observation period.
These parturients are identified in Figures 4 and 5  Note: For continuous data mean (95% CI) are shown, while for the counts, N (%) are shown.

| Maternal satisfaction with analgesia at 30 min
The groups were identical in terms of reported satisfaction with the analgesia score (Figure 4).The median satisfaction with analgesia VAS was 92.Note: Duration of analgesia marks the time in min until the parturient requested supplemental epidural analgesia.For neuraxial continuation, the number of doses is shown for the parturients that completed the first stage of labour and reached full cervical dilatation, with the total amount of ropivacaine and fentanyl given, which is also referenced to the duration of the first stage of labour (shown as drug consumption/h).Labour analgesia costs are shown for the initial study intervention and the total neuraxial analgesia costs for all parturients in the group.VAS values are expressed as median (IQR) and analysed by Mann-Whitney U test.All other continuous data expressed as mean (95% CI) and analysed by one-way ANOVA.Data for pruritus and foetal heart rate pattern analysed by Mantel-Haenszel test.

| Motor block
All parturients could move in their bed during the 30-min observation period normally as well as support themselves on all fours, if needed due to CTG changes.At the end of the 30-min observation period they were asked to abduct their lower limb and asked for subjective assessment of their lower limbs.While two parturients in the epidural group reported subjective heaviness of the lower limbs, all parturients could perform the leg-raising test.

| Pruritus
Pruritus was experienced in both treatment arms, with a markedly higher incidence and severity in the intrathecal fentanyl group.The distribution of pruritus on the five-step Likert-scale is shown in Table 2.

| Progress of labour and analgesia
Twenty parturients in the intrathecal fentanyl group and 18 in the epidural analgesia group completed the first stage of labour and reached full cervical dilatation after the analgesia.The time spans of the deliveries are shown in Table 3.The cervical dilatation levels and the time from the measurement before and after the analgesia intervention are shown in Figure 5, along with the mean cervical progress during the neuraxial analgesia initiation.There were no differences between the groups in the time until full cervical dilatation or the time until delivery (Table 3).
The analgesia was continued subsequently by midwifeadministered boluses, and there were no differences in the total number of boluses given throughout delivery or in the total amount of fentanyl given, even when the different fentanyl amounts during the initial dose are accounted for.The epidural analgesic consumption is shown in Table 2.

F I G U R E 3
The absolute pain VAS differences at 20 min after epidural analgesia (100 μg) and lidocaine 80 mg against intrathecal fentanyl (20 μg).The hatched vertical lines indicate the equivalence line (0 mm) and the margin of non-inferiority (+20 mm).The mean difference, and the 95% confidence interval for the difference is shown.
The maternal satisfaction and pain during contraction measured with 0-100 mm visual analogue scale (VAS) for each parturient 30 min after either (A) epidural analgesia (80 mg lidocaine and 100 μg fentanyl) or (B) intrathecal fentanyl 20 μg.The solid dots indicate parturients who requested supplemental analgesia at 30 min after the intervention and were subsequently satisfied with the analgesia provided, whereas the solid square symbol indicates a parturient who required a new epidural catheter.
analgesic effect.The differences between these interventions come from the side effect profile, with intrathecal fentanyl causing significantly more pruritus and epidural analgesia having the potential for a motor block.Intrathecal fentanyl also results in higher supply-related costs and requires a more complicated and invasive installation.
Our previous study addressing the possibility to replace the intrathecal fentanyl with an epidural fentanyl dose showed median pain VAS of 10 mm for the intrathecal fentanyl and 53 mm for the epidural fentanyl. 11Some of the parturients in the prior study reported lower satisfaction values due to pruritus after intrathecal opioids and expressed that they would be more satisfied with a lower analgesic effect if the pruritus was less severe.An unpublished post hoc analysis of the parturient data from the previous study revealed that the optimal cut-off-line at 30 min for pain VAS was <31 mm which would show the best predictive value to reach parturient satisfaction of >92 mm on a 0-100 mm VAS line corresponding to "excellent" satisfaction. 15Therefore, the clinical non-inferiority margin was set at 20 mm above the pain recorded in the intrathecal fentanyl group marking approximately half of the pain VAS difference between a fentanyl-only epidural and the intrathecal fentanyl at 20 μg.
To verify the effect size difference between groups, the analgesia-VAS values representing reduction of the VAS from the baseline values were also compared and showed 10.0 mm lower analgesic effect from epidural analgesia compared with intrathecal fentanyl (95% CI for the difference 5.8-14.2mm, TOST P < .001for the boundaries).However, labour pain may not stay constant since many of the parturients in the study cohort proceeded to the labour ward for the studied pain relief either soon after the initiation of the oxytocin infusion or to enable increasing its infusion rate.Additionally, and contrary to many other conditions with acute pain, labour pain is expected to increase as the labour progresses.Thus, although the use of raw VAS values has been questioned, we believe they can be used for comparison of the effects in this kind of setting. 16e 20-μg intrathecal fentanyl dose has been shown to cause an almost complete analgesia within 10 min in a primiparous parturient population, and in a dose-finding study no benefit was shown in increasing the dose from 20 μg. 1 The median VAS values at 20 min after the intrathecal fentanyl were comparable between our prior study and the current study, 10 and 13 mm with 75% and 80% of the parturients reaching VAS <30 mm, respectively.
The addition of lidocaine (80 mg) into the epidural analgesia regimen offered a significantly higher efficacy compared with an intervention consisting of fentanyl alone.The results are in line with the previously reported efficacy of epidural fentanyl (100 μg) with a lidocaine test dose (45 mg) and epidural lidocaine (150 mg) alone. 10,17,18rthermore, both the onset (Figure 2) and time of action for epidural analgesia appear to be like those achieved with intrathecal fentanyl (Table 2).
Subjective parturient satisfaction with the provided analgesia has been shown to be a more important outcome than the actual pain scores. 15In our study, the parturients' pain rating of more than 30 mm at 30 min after the analgesia was associated with lower satisfaction than a pain VAS of under 30 mm.Furthermore, pain ratings greater than 30 mm associated with a rapid parturient request for more epidural pain relief (Figure 4).On the other hand, complete analgesia does not necessarily mean the highest possible level of parturient satisfaction as side effects such as pruritus or actions needed for potential CTG changes appear to decrease the satisfaction scores.
Mild pruritus was reported by 14% of the parturients in the epidural group.In contrast, in the intrathecal fentanyl group, 13% of the parturients reported intense or unbearable pruritus, which was also a reason for lower satisfaction scores in three parturients.Thus, rapidly acting epidural analgesia could be a better choice of analgesia even if its analgesic potential was slightly lower than that of intrathecal opioid.
A surprisingly high number of parturients (20%) in both treatment arms requested additional analgesia within 45 min of the initial analgesic intervention.The initial poor efficacy of neuraxial analgesia can be attributed to one of three reasons: technical failure (misplacement), rapid progression of labour, or dystocia causing abnormal contraction pain.All 60 epidural catheters placed in the study were used to give 1-7 top-up boluses, and all but one catheter in the epidural group provided sufficient analgesia at least after the first epidural top-up dose.In the 11 cases where an epidural top-up was successfully used within 45 min of the initial analgesic intervention, three epidural analgesia cases and one intrathecal fentanyl case appeared to be progressing rapidly (with cervical dilatation rate of 1.6-5.0cm/h), while two epidural and five intrathecal fentanyl cases showed protracted labour and a low cervical progression rate (Figure 5).Despite differences in the suspected aetiology for the poor response to the initial dose, these parturients were satisfied with the analgesia after the first epidural top-up dose.Thus, it remains possible that these initial failures could have been avoided by either addition of local anaesthetics in the intrathecal analgesia regimen or epidural administration of the higher volume-low concentration ropivacaine with the fentanyl as the initiation of the neuraxial analgesia.
The CSE-based initiation of neuraxial analgesia has been subject to numerous studies and meta-analyses. 2,19,20The rapidity of onset of analgesia has been most firmly established, while other possible outcomes are not as clear. 20It should also be noted that most of the literature around CSE based analgesia involves a co-administration of epidural or intrathecal local anaesthetic.The intrathecal fentanyl intervention is expected to cause a marked sacral analgesia and, because of the cephalic spreading of fentanyl in the cerebrospinal fluid, it is likely to cause analgesia in the thoracic segments innervating the uterus, thus alleviating contraction pain during the first phase of labour.Interestingly, 5/6 initially insufficiently effective labour epidural analgesics could also be fixed by injecting dilute ropivacaine into the epidural catheter.Extrapolating from data using stronger solutions used for lumbar epidural for the C-section anaesthesia, the initial lidocaine dose (80 mg) corresponds to about 30 mg of ropivacaine. 21nce, insufficient local anaesthetic dose is unlikely to be the reason for the poor efficacy of the initial dose.Instead, a more likely reason is the insufficient cephalic spread of the epidural drug dose from the lumbar epidural injection site after the limited (10 mL) volume of the epidural intervention used in this study.In this context, it would be beneficial to compare the efficacy of intrathecal fentanyl with a higher volume of dilute epidural analgesia.
Using a setup with patient-controlled-epidural analgesia after initial analgesia by intrathecal opioid-CSE with or without intrathecal local anaesthetic, the CSE technique has been shown to associate with lower consumption of epidural analgesic mixture, resulting in a lower maternal opioid load during delivery. 3Since the interventions in our study provide practically equal analgesia in terms of duration, the opioid-sparing effect of the initial intrathecal fentanyl dose would be more pronounced with a shorter duration of labour and hence, with fewer epidural top-up doses.However, when analgesia is started in the early phase of the delivery, more boluses are given and the relative difference in the total ropivacaine and fentanyl consumption becomes diluted over time.Our study population received similar amounts of midwife-administered epidural boluses during the deliveries that were of equal duration, and therefore showed no significant differences in the total fentanyl consumption or consumption rate per hour (Table 2).
Another presumed benefit of intrathecal opioid based analgesia is the lack of a motor block.In our study, motor block was not assessed until at the end of the 30-min observation period.The epidural intervention was known not to induce any clinically significant motor block due to institutional practise of using the same epidural intervention for the second phase of labour after repeated 20 mg ropivacaine boluses often given over a period of several hours.Since a motor block has not caused clinically detected problems even after prior loading of the epidural space with long-acting local anaesthetics, it remains unlikely that the first dose would cause it.It remains possible that signs of a motor block would have been found with a more precise testing method, such as the modified Bromage scale, which showed unaltered motor function in 76% of the parturients after an epidural lidocaine dose of 45 mg in a volume of 10 mL. 18garding the costs, the difference between the treatment groups arises from the price difference between the supplies for the initial intervention, with the relatively expensive CSE needle set accounting for most of it.
Despite being carefully planned, blinding remains a limitation in our study set-up.Blinding the analgesia provider by incorporating either intrathecal or epidural placebo injection would have turned the epidural intervention into a "dural-puncture epidural," while the intrathecal intervention could have been boosted by "epidural volume expansion." 19Also, the midwife remained in the labour room during the analgesia provision and data collection.While we cannot be certain about the blinding of the midwifes in our setup, their role in the study was limited to providing the subsequent epidural boluses and thus they participated only in the measurement of the duration of analgesic effect.All other data were collected directly by a blinded assessor from the blinded parturient.
The CSE technique involves the installation of a standard epidural catheter and the intrathecal injection, which requires a longer time than the standard epidural installation, irrespective of provider experience.The maternal perception of the neuraxial analgesic procedure was not assessed in the study but the longer the time that the parturient is required to stay positioned while having contractions could be an important factor from her viewpoint.The number of attempts needed for successful completion of the allocated intervention shows that even in experienced hands the CSE analgesia is more difficult to establish than normal epidural analgesia, being three times more likely to require more than one attempt (OR 3.0 [1.1-8.3]).
Although no differences were noted in the progression of labour, our small sample size does not allow conclusions regarding this or estimation of the rarer adverse events such as initial failure of neuraxial analgesia and factors associated with it.It has been proposed that the CSE technique may increase the likelihood of installing a properly functioning epidural catheter, and in line with this, in our cohort the only misplaced epidural catheter was seen in the epidural analgesia group. 9 conclusion, coadministration of fentanyl and lidocaine appears to provide analgesia that is non-inferior to intrathecal fentanyl in terms of intensity of the analgesia provided at 20 min after intervention during the early stage of induced labour.In comparison to intrathecal fentanyl, epidural analgesia results in less pruritus, easier and less invasive administration, and markedly lower costs.
This study was authorised by the Finnish Medical Agency (FIMEA) and local hospital authorities following a permission of the hospital Ethical board (Helsinki University Hospital Ethical Board I, decision HUS/3483/2020, 3 March 2021).Ninety-five informed parturients initially consented to the study.The parturients were admitted to the hospital between March 2021 and May 2023 for the purpose of induction of labour and recruitment was carried out at the induction ward before contractions started.
Figure 2A and the proportion of parturients reaching analgesia level . The mean time until the first epidural top-up dose was 82 (69-95) min in the intrathecal fentanyl group and 91 (75-106) min in the epidural analgesia group (P = .360).All 60 epidural catheters placed during the study were used to provide subsequent analgesia.One epidural catheter was replaced after both the initial epidural study intervention and the subsequent rescue bolus failed to provide satisfactory analgesia, whereas all other catheters provided adequate analgesia at least after the first top-up dose.The mean difference in duration was 9.0 (5.5-12.4)min longer after epidural analgesia.The one-sided t test P value against zero is <.001, but since the study was not planned to show superiority, this result should be considered to indicate non-inferiority in duration compared with intrathecal fentanyl.F I G U R E 2 (A) The onset time (0-30 min) for neuraxial analgesia following either intrathecal fentanyl (20 μg; open bars and symbols) or epidural analgesia (lidocaine 80 mg and fentanyl 100 μg; grey bars and symbols) measured by a 0-100 mm visual analogue scale (VAS) for the maximum pain during the latest contraction.The bars show the middle two quartiles while the whiskers show the overall range, excluding outliers, for each measurement point.At 5 min: N = 22 and 20 observations for intrathecal fentanyl and epidural analgesia, respectively.All other timepoints: 30 parturients per treatment arm.(B) The proportion of parturients remaining at pain VAS >30 mm during contraction at the given time point (the dashed line: intrathecal fentanyl, solid line: epidural analgesia).The P-value is shown for overall comparison between the treatment groups (Mantel-Cox Log Rank test).T A B L E 1 The parturient baseline characteristics.
Before the interventions, the CTG recordings were mostly rated as normal, and no signs of foetal hypoxia were detected.Cardiotocography recordings were available for assessment in 29/30 parturients in both treatment arms.Over the 30-min period following the interventions, the CTG recordings were recorded as unaffected in 21 (70%) and 19 (63%) of the parturients in the intrathecal fentanyl and epidural analgesia groups, respectively.The distribution of CTG classifications across the categories is shown in Table2.All suspicious or pathological tracings were reversible and required at maximum a change of maternal position and initiation of invasive foetal heart rate monitoring by a scalp electrode.