Neuraxial homeostasis-guided labor analgesia to reduce neurologic symptom and enhance maternal satisfaction: a randomized clinical trial Authors: Jianwei Guo, Yan Cheng, Minmin Yi Published: 2026 ABSTRACT BackgroundNeuraxial analgesia is the gold standard for labor pain relief, yet procedure-related neurological symptoms may occur. Guided by the theory of neuraxial homeostasis, this study compared a sequential spinal-epidural analgesia (SSEA) technique with conventional combined spinal-epidural analgesia (CSEA) to determine whether preserving neuraxial stability was associated with a reduction in neurological symptoms and enhanced maternal satisfaction.MethodsIn this randomized trial, 740 parturients requesting labor analgesia were assigned to receive either SSEA (subarachnoid injection followed by epidural catheterization, n = 368) or conventional CSEA (needle-through-needle technique, n = 372). The primary outcome was analyzed in the intention-to-treat (ITT) population of all 740 randomized participants. For secondary outcomes assessed after delivery, 116 women who underwent cesarean section were not applicable for analyses requiring vaginal delivery, leaving 624 parturients who completed vaginal delivery (312 per group) for those specific analyses. The primary outcome was the incidence of intraprocedural neurological symptoms (radiating pain or involuntary muscle twitching during puncture). Secondary outcomes included post-procedural neurological symptoms, adverse effects, analgesic efficacy (Visual Analogue Scale [VAS] scores), analgesic quality, maternal/fetal outcomes, and maternal satisfaction (5-point Likert scale). An exploratory structural equation model (SEM) examined factors influencing satisfaction.ResultsBaseline characteristics were balanced. In the ITT analysis (n = 740), SSEA was associated with a significantly lower incidence of intraprocedural neurological symptoms (1.36% vs. 23.12%; risk difference [RD] − 0.218, 95% CI − 0.267 to − 0.169; P < 0.001). Among the 624 vaginal deliveries, SSEA also showed lower rates of neurological symptoms at 48 h (0.32% vs. 2.56%; RD − 0.022, − 0.041 to − 0.004; P = 0.019), lower back pain (1.28% vs. 6.41%; RD − 0.051, − 0.081 to − 0.021; P = 0.001), and persistent paresthesia at 1 week (0% vs. 5.13%; RD − 0.051, − 0.076 to − 0.026; P < 0.001). Overall maternal satisfaction was higher with SSEA (12.81 ± 1.46 vs. 11.26 ± 1.23; Cohen’s d = 1.15; P < 0.001), driven by greater satisfaction with the overall experience (d = 2.08) and willingness to recommend (d = 0.98), whereas satisfaction with pain relief did not differ (P = 0.139). VAS scores, adverse effects, and maternal/fetal outcomes were comparable between groups. In an exploratory SEM, intraprocedural neurological symptoms showed the strongest negative association with satisfaction (standardized β=–0.140, P < 0.001), followed by poor analgesic quality (β=–0.101, P = 0.011) and higher mean VAS (β=–0.092, P = 0.023).ConclusionIn this study, preserving neuraxial homeostasis by performing spinal puncture before epidural catheterization was associated with a lower incidence of observed neurological symptoms and higher maternal satisfaction scores compared with conventional CSEA, without compromising analgesic efficacy. While these findings suggest potential benefits of the SSEA technique, confirmation in blinded, multi-center trials is needed, and the assessment of satisfaction was limited by the use of a non-validated instrument.Trial registrationChiCTR, ChiCTR2500111828. Registered 30 November 2023 (prospectively registered); first participant enrolled 10 February 2024.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12967-026-08331-7. Introduction Neuraxial blockade is currently the universally acknowledged gold standard for labor analgesia [1, 2]. Not only does it effectively alleviate maternal pain, but it also provides rapid and satisfactory anesthesia should instrumental delivery or intrapartum cesarean section become necessary [3]. As healthcare models increasingly shift toward patient-centered care, the subjective experience and satisfaction of parturients have emerged as key indicators for assessing the quality of analgesia [4]. Yet clinical observations have revealed an intriguing pattern. During the performance of classic combined spinal-epidural analgesia (CSEA)—where the spinal needle is passed through the epidural needle to puncture the dura and arachnoid mater—approximately 25% of parturients experience neurological symptoms such as unilateral limb jerking or lower back muscle twitching. In some cases, these symptoms can progress to permanent sensory loss in areas like the lower back, abdomen, or lateral thigh. By contrast, when subarachnoid block is performed with a spinal needle first, followed by epidural catheterization, only about 1% of parturients report such symptoms. This striking disparity prompted us to investigate the underlying mechanism. In our previous work, we proposed the concept of “neuraxial homeostasis disruption“ [5] —the idea that the relatively stable state of intrathecal structures within the closed vertebral canal can be disturbed during CSEA. We hypothesized that when the epidural needle enters first, it may cause subtle movement of the dura and arachnoid, agitating the nerve roots floating in cerebrospinal fluid. Then, when the spinal needle is introduced, it might contact these oscillating nerve fibers, eliciting neurological symptoms. These symptoms, along with potential neurological sequelae, can profoundly impact maternal health [6], triggering psychological stress and anxiety, affecting emotional well-being, prolonging recovery, and ultimately diminishing overall satisfaction with labor analgesia. Current research on labor analgesia has largely focused on common complications such as hypotension, post-dural puncture headache (PDPH), epidural-related maternal fever (ERMF), breakthrough pain, inadequate analgesia, uterine atony, transient fetal heart rate abnormalities, urinary retention, and pruritus [7, 8]. What has received far less attention—particularly from a technical and procedural perspective—is the prevention and management of neurological complications. This gap in the literature makes it difficult to provide practical, evidence-based guidance for clinical practice. Furthermore, identifying the core clinical factors that influence maternal satisfaction, especially the intrinsic relationship between procedure-related neurological symptoms and satisfaction, is crucial for comprehensively improving the quality of analgesia. With this in mind, we designed the present study to test a specific hypothesis: that performing dural puncture before epidural catheterization (SSEA), thereby avoiding dural puncture with an epidural needle already in situ, would reduce neurological complications compared to the standard needle-through-needle CSEA approach [5]. Materials and methods Study participants This study was approved by the Ethics Committee of Shaanxi Provincial People’s Hospital (Approval No.: 2024-R175) and conducted in accordance with the Declaration of Helsinki [9]. Informed consent was obtained from all parturients and their families. The trial was prospectively registered with the Chinese Clinical Trial Registry (ChiCTR2500111828) on November 30, 2023—more than two months before the first participant was enrolled on February 10, 2024. The study protocol, finalized prior to registration, has remained unchanged throughout the study period and is available as supplementary material. A total of 740 parturients receiving labor analgesia at our hospital between February 2024 and September 2025 were enrolled and randomized to either the SSEA group (n = 368) or the CSEA group (n = 372). After randomization, 116 parturients were excluded due to conversion to cesarean section during labor (56 SSEA, 60 CSEA), leaving 624 parturients who completed vaginal delivery (SSEA: 312, CSEA: 312). A random allocation sequence was generated by a statistician not involved in patient enrollment, using a computer-generated random number table with a 1:1 allocation ratio. The group assignments (SSEA or CSEA) were placed in sequentially numbered, opaque, sealed envelopes by an independent research assistant. When a parturient was admitted for labor analgesia and met the inclusion criteria, the attending anesthesiologist opened the next envelope in sequence to reveal the group allocation. In this way, allocation was concealed from both the recruiting team and the parturient until the moment of the procedure. Inclusion, exclusion, and elimination criteria Inclusion Criteria: (1) Voluntary request for analgesia by the parturient; (2) Assessed by an obstetrician as suitable for vaginal delivery or trial of labor after cesarean [3]; (3) age 23–35 years, weight 60–90 kg, ASA physical status I or II. Exclusion Criteria: (1) Parturient refusal or unwillingness to sign informed consent; (2) Inability to cooperate with neuraxial puncture; (3) Contraindications to neuraxial blockade, such as coagulation disorders, infection or injury at the puncture site, uncorrected maternal hypovolemia or hypotension, increased intracranial pressure, severe spinal deformity; (4) Allergy to local anesthetics or opioids; (5) Neurological diseases or neuropathies assessed as contraindications to neuraxial analgesia, including patients with platelet function and/or coagulation disorders—for example, those on anticoagulant therapy (antiplatelet agents or anticoagulants) or with documented platelet dysfunction [3]. Criteria for Exclusion from the Vaginal-Delivery Subgroup Analysis: For secondary outcomes that could only be assessed in parturients who underwent vaginal delivery (e.g., analgesic quality during the second stage, maternal satisfaction with the birth experience), parturients who underwent intrapartum cesarean section were not applicable for that specific analysis. This does not apply to the primary outcome or to other intraprocedural adverse events, which were available for all randomized participants and were analyzed using intention-to-treat principles. Neuraxial labor analgesia technique Neuraxial labor analgesia was performed at the L3-4 interspace, identified using bony landmarks (iliac crests and lumbar spinous processes). All procedures were performed by a single experienced attending anesthesiologist with over 10 years of experience in obstetric anesthesia, who had performed more than 500 CSEA and 200 SSEA procedures prior to the study. To minimize bias, parturients were not informed of their group allocation. A separate trained research assistant, blinded to group assignment, conducted all follow-up assessments at 48 h and 1 week postpartum. This assistant was not present during the neuraxial procedure and was instructed to avoid asking about procedural details. With the patient in the left lateral decubitus position, and following standard monitoring (heart rate, blood pressure, ECG, pulse oximetry), the puncture site was determined. A midline approach was used, with the needle oriented perpendicular to the plane of the lower back [10]. SSEA Group: A standard spinal needle (0.5 × 113 mm) was first used for subarachnoid puncture. Two milliliters of 0.1% ropivacaine mixed with sufentanil (1 µg/ml) were injected intrathecally. The spinal needle was then withdrawn. Subsequently, a standard epidural needle (1.6 × 80 mm) was used for epidural puncture, and an epidural catheter (outer diameter 1.0 mm) was inserted approximately 4 cm into the epidural space. The catheter was connected to an epidural analgesia pump (containing 50 ml of 0.1% ropivacaine with 0.5 µg/ml sufentanil, basal rate 10 ml/h) 20 min after the intrathecal injection. This 20-minute interval was deliberately chosen to allow the intrathecal dose to achieve its peak effect and to confirm the absence of high spinal block before initiating the epidural infusion. No interim rescue boluses were permitted during this period. CSEA Group: A standard epidural needle was first used for epidural puncture. The spinal needle was then introduced through the epidural needle (needle-through-needle technique) for subarachnoid puncture. The same intrathecal medication was injected. After withdrawing the spinal needle, an epidural catheter was inserted approximately 4 cm. The catheter was connected to the epidural analgesia pump 20 min later. All parturients received continuous analgesia via the pump until the completion of the third stage of labor. Study outcomes Analysis Populations. The primary outcome (intraprocedural neurological symptoms) was analyzed in the intention-to-treat (ITT) population, which included all 740 randomized parturients. A per-protocol (PP) population of 624 parturients who completed vaginal delivery (312 per group) was used for the analysis of secondary outcomes that were specific to the labor, delivery, and postpartum periods (e.g., analgesic quality, satisfaction). A supplementary table (Supplementary Table S1) details which population was used for each outcome and the rationale for exclusions where applicable. The primary outcome was the incidence of intraprocedural neurological symptoms. We established a rigorous operational definition for this endpoint: any new-onset, transient sensation or movement disorder occurring during the neuraxial puncture procedure (from skin infiltration to epidural catheter fixation). This was specifically defined as either unilateral radiating pain—an electric shock-like sensation—in the lower back, buttock, or lower limb; or involuntary, unilateral muscle twitching or jerking of the lower back or lower extremity. The symptom had to be temporally associated with needle advancement or injection, resolve spontaneously or upon cessation of the stimulus, and be clearly distinguishable from a voluntary pain response. The timing of symptom onset (e.g., during spinal needle advancement versus during drug injection) was also recorded. Intraprocedural neurological symptoms were recorded by the operating anesthesiologist. No independent observer or standardized intra-procedural checklist was used to confirm these events in real time, which may introduce measurement bias. All other outcomes were recorded by the blinded research assistant. Recorded secondary outcomes included: [11, 12] adverse effects of neuraxial analgesia (hypotension within 20 min, ERMF, pruritus within 48 h, nausea and vomiting (NAVO), urinary retention, chills, local anesthetic systemic toxicity, extensive block, lower back pain, PDPH), post-procedural neurological symptoms, and maternal satisfaction. Post-procedural neurological symptoms were assessed by the blinded research assistant at 48 h and 1 week postpartum. We defined “neurological symptoms at 48 hours” as any patient-reported new or persistent sensory abnormality (numbness, tingling, hypoesthesia) or motor weakness (difficulty moving a leg) in the lower extremities or back, not attributable to the residual effects of analgesia or the labor process itself. “Persistent paresthesia at 1 week” was defined as the presence of any of the above sensory symptoms that were still present at the 7-day follow-up. If a patient reported significant or persistent deficits, a formal neurological consultation was arranged, though this was not part of the routine outcome definition. Other outcomes included analgesic efficacy (VAS scores at 0, 1, 5, 10, 30, 60 min post-analgesia initiation, at complete cervical dilation, at fetal head delivery, and at placental delivery), analgesic quality assessment (asymmetric block, S2 blockage, Bromage scores, number of PCEA boluses, manual epidural bolus, epidural catheter adjustments), maternal outcomes (spontaneous vaginal delivery, instrumental delivery), and fetal outcomes (bradycardia, 1-/5-/10-minute Apgar scores). Patient satisfaction was assessed at 48 h postpartum by the blinded research assistant using a three-item questionnaire, with each item rated on a 5-point Likert scale (1 = very dissatisfied to 5 = very satisfied): (1) satisfaction with pain relief, (2) satisfaction with the overall birth experience, and (3) willingness to recommend the analgesic technique to others. We acknowledge that this was an ad-hoc questionnaire whose validity and reliability have not been formally established—a limitation we address later in the discussion. Sample size calculation The sample size was primarily calculated based on the incidence of intraprocedural neurological symptoms observed in a pilot study conducted in our department. This pilot study retrospectively reviewed 150 parturients (75 who received CSEA and 75 who received an early form of SSEA) and found an incidence of 24.0% (18/75) in the CSEA group compared to 1.3% (1/75) in the SSEA group. Assuming a more conservative incidence of 25% in the CSEA group and 1% in the SSEA group, a two-sided test with α = 0.05 and power (1-β) = 0.90 required 41 patients per group using PASS 15.0.5 software. To account for an anticipated cesarean section rate of approximately 15% and to ensure adequate power for the final analysis, we enrolled a larger sample of 740 parturients (368 SSEA, 372 CSEA). After excluding 116 who underwent cesarean section, this yielded 312 patients per group for the final analysis. Statistical analysis Statistical analyses were performed using SPSS 23.0 and AMOS 29.0. All primary analyses were conducted on both a per-protocol basis (including only parturients who completed vaginal delivery, n = 624) and an intention-to-treat basis (including all 740 randomized participants). For the ITT analysis, intraprocedural neurological symptoms were available for all participants, as these events occurred prior to any conversion to cesarean section. A sensitivity analysis was performed to compare the primary outcome between the ITT and per-protocol populations to assess the robustness of the findings. Normality of continuous data was assessed using the Kolmogorov-Smirnov test. Normally distributed continuous data—such as age, BMI, and VAS scores—are presented as mean ± standard deviation (X̄ ± SD) and were analyzed using the independent samples t-test. Non-normally distributed continuous or ordinal data are presented as median (interquartile range) and were analyzed using the Mann-Whitney U test. Categorical data, including the incidence of neurological symptoms and mode of delivery, are presented as n (%) and were analyzed using the Chi-square test or Fisher’s exact test, as appropriate. For all analyses, a two-sided P value < 0.05 was considered statistically significant. Given the number of secondary outcomes compared, these analyses should be considered exploratory. No formal adjustment for multiple comparisons was made; consequently, findings from secondary outcomes should be interpreted with caution, recognizing the increased risk of type I error. The primary outcome was the sole basis for the confirmatory hypothesis test. An exploratory structural equation model (SEM) was constructed using maximum likelihood estimation in AMOS 29.0 to examine the direct effects of key clinical variables on maternal satisfaction. The model was just-identified (saturated), resulting in zero degrees of freedom (χ²=0). As a result, fit indices (CFI, TLI, RMSEA, SRMR) are mathematically perfect (1.000 or 0.000) and cannot be interpreted as measures of model fit. Our focus in this analysis was on the significance and magnitude of the standardized path coefficients (β). Variables included in the model were intraprocedural neurological symptoms (binary), mean VAS pain score (continuous), a latent variable for “poor analgesic quality” (indicated by need for PCEA boluses, manual boluses, or catheter adjustments), and maternal and fetal outcomes. This analysis was purely exploratory and intended to generate hypotheses for future research. Standardized path coefficients were used to examine the mechanisms of variables affecting satisfaction. Results Baseline characteristics comparison As shown in Table 1, the two groups were well-balanced with respect to demographic and obstetric characteristics. No statistically significant differences were observed between the SSEA and CSEA groups in terms of age, height, weight, BMI, gestational age, parity, cervical dilation at placement, or rates of labor induction/augmentation (P > 0.05 for all comparisons). This comparability at baseline strengthens our confidence that any subsequent differences in outcomes can be attributed to the analgesic techniques themselves rather than to underlying patient factors. Table 1Baseline characteristics between the two groups at randomization (n = 740)CharacteristicsSSEA (n = 312)CSEA (n = 312)tP-valueAge, yr32.43 ± 3.4532.71 ± 3.48-0.9940.321Height, cm161.61 ± 4.50161.62 ± 4.61-0.0090.993Weight, kg69.85 ± 8.8069.82 ± 9.010.0460.963BMI, kg.m− 226.73 ± 3.0626.72 ± 3.110.0640.949Gestational age, days275.80 ± 6.87274.87 ± 9.051.4400.150Nulliparous, n (%)201 (64.4%)208 (66.7%)0.3520.553Cervical dilation at placement, cm3.5 (2.5–4.5)3.5 (2.5–4.5)-0.2310.817Induction/Augmentation, n (%)112 (35.9%)118 (37.8%)0.2470.619*Data are presented for all 740 randomized parturients. Of these, 116 who converted to cesarean section during labor (56 SSEA, 60 CSEA) were not applicable for secondary outcomes requiring vaginal delivery; the remaining 624 parturients (SSEA: 312, CSEA: 312) completed vaginal delivery and were included in those specific analyses. There were no significant differences in baseline characteristics between groups among the randomized population. Values are presented as mean ± SD, median (IQR), or number (percentage). SSEA: Sequential Spinal-Epidural Analgesia. CSEA: Combined Spinal-Epidural Analgesia. BMI: Body Mass Index Table 1Baseline characteristics between the two groups at randomization (n = 740)CharacteristicsSSEA (n = 312)CSEA (n = 312)tP-valueAge, yr32.43 ± 3.4532.71 ± 3.48-0.9940.321Height, cm161.61 ± 4.50161.62 ± 4.61-0.0090.993Weight, kg69.85 ± 8.8069.82 ± 9.010.0460.963BMI, kg.m− 226.73 ± 3.0626.72 ± 3.110.0640.949Gestational age, days275.80 ± 6.87274.87 ± 9.051.4400.150Nulliparous, n (%)201 (64.4%)208 (66.7%)0.3520.553Cervical dilation at placement, cm3.5 (2.5–4.5)3.5 (2.5–4.5)-0.2310.817Induction/Augmentation, n (%)112 (35.9%)118 (37.8%)0.2470.619*Data are presented for all 740 randomized parturients. Of these, 116 who converted to cesarean section during labor (56 SSEA, 60 CSEA) were not applicable for secondary outcomes requiring vaginal delivery; the remaining 624 parturients (SSEA: 312, CSEA: 312) completed vaginal delivery and were included in those specific analyses. There were no significant differences in baseline characteristics between groups among the randomized population. Values are presented as mean ± SD, median (IQR), or number (percentage). SSEA: Sequential Spinal-Epidural Analgesia. CSEA: Combined Spinal-Epidural Analgesia. BMI: Body Mass Index Baseline characteristics between the two groups at randomization (n = 740) *Data are presented for all 740 randomized parturients. Of these, 116 who converted to cesarean section during labor (56 SSEA, 60 CSEA) were not applicable for secondary outcomes requiring vaginal delivery; the remaining 624 parturients (SSEA: 312, CSEA: 312) completed vaginal delivery and were included in those specific analyses. There were no significant differences in baseline characteristics between groups among the randomized population. Values are presented as mean ± SD, median (IQR), or number (percentage). SSEA: Sequential Spinal-Epidural Analgesia. CSEA: Combined Spinal-Epidural Analgesia. BMI: Body Mass Index Intervention outcomes and satisfaction analysis Analysis Populations. The primary outcome of intraprocedural neurological symptoms was analyzed in the ITT population of all 740 randomized participants, as this event occurred before any conversion to cesarean section and was recorded for all enrolled parturients. Of the 740 randomized parturients (SSEA: 368, CSEA: 372), 116 were not applicable for the vaginal-delivery subgroup analysis due to conversion to cesarean section during labor (56 SSEA, 60 CSEA), leaving a per-protocol (PP) population of 624 parturients who completed vaginal delivery (SSEA: 312, CSEA: 312). This PP population was used for the analysis of secondary outcomes specific to the vaginal delivery and postpartum periods. Follow-up at 48 h and 1 week was complete for all 624 parturients, with no losses to follow-up. As shown in Table 2, notable differences were observed between the groups for several key outcomes. The incidence of intraprocedural neurological symptoms was substantially lower in the SSEA group compared to the CSEA group (1.60% vs. 25.64%, P < 0.001). To assess the robustness of this finding, we performed an intention-to-treat analysis including all 740 randomized participants, for whom intraprocedural neurological symptom data were available prior to any conversion to cesarean section. The results remained consistent: the incidence of intraprocedural neurological symptoms was significantly lower in the SSEA group (5/368, 1.36%) compared to the CSEA group (86/372, 23.12%; risk difference [RD] − 0.218, 95% CI − 0.267 to − 0.169; P < 0.001), suggesting that the observed association was not biased by post-randomization exclusions. This difference of 24% points represents a number needed to treat of just 4.2, suggesting that for every 4.2 parturients treated with SSEA instead of CSEA, one fewer case of intraprocedural neurological symptoms was observed. Similarly, the SSEA group had significantly lower rates of neurological symptoms at 48 h (0.32% vs. 2.56%, P = 0.019), lower back pain (1.28% vs. 6.41%, P = 0.001), and persistent paresthesia at 1 week (0% vs. 5.13%, P < 0.001). Table 2Comparison of clinical outcomes between the two groupsOutcomesSSEA(n = 312)CSEA(n = 312)Test Statistic (χ²/t/Mann-Whitney)P-valueEffect Size (RD or Cohen’s d) [95% CI] Intraprocedural neurological symptom 5(1.60%) 80(25.64%) 76.612 0.001 RD: -0.240 [-0.291, -0.190]Adverse effects Hypotension within 20 min15(4.81%)12(3.85%)0.3480.555 ERMF7(2.24%)8(2.56%)0.0680.794 Pruritus within 48 h24(7.69%)31(6.73%)0.9770.323 NAVO9(2.88%)14(4.49%)1.1290.288 Urinary retention00// Chills3(0.96%)5(1.60%)0.5060.477 Toxicity00// Extensive block00// Neurological symptoms at 48 h 1(0.32%) 8(2.56%) 5.511 0.019 RD: -0.022 [-0.041, -0.004] Lower back pain 4(1.28%) 20(6.41%) 11.093 0.001 RD: -0.051 [-0.081, -0.021] PDPH4(1.28%)7(2.24%)0.8330.361 Persistent paresthesia at 1 week 0 16(5.13%) 14.464 0.001 RD: -0.051 [-0.076, -0.026]Analgesic efficacy VAS 0 min6.84 ± 1.056.76 ± 1.050.9530.341 VAS 1 min2.06 ± 0.791.99 ± 0.791.0180.309 VAS 5 min1.80 ± 0.751.89 ± 0.80-1.4430.150 VAS 10 min1.92 ± 0.802.04 ± 0.86-1.7930.073 VAS 30 min1.73 ± 0.761.84 ± 0.73-1.7140.087 VAS 60 min1.84 ± 0.831.89 ± 0.85-0.8080.420 VAS complete cervical dilation2.02 ± 0.822.08 ± 0.78-0.8980.369 VAS fetal head delivery2.15 ± 0.782.23 ± 0.81-1.2530.211 VAS placental delivery1.46 ± 0.501.50 ± 0.50-1.0410.298Analgesic quality Asymmetric blockage6(1.92%)4(1.28%)0.4070.524 S2 blockage262(83.97%)276(88.46%)2.9490.086 Bromage scores (any motor block)25(8.01%)24(7.69%)0.0220.882 PCEA boluses required25(8.01%)36(11.54%)2.1990.138 Manual epidural boluses required17(5.45%)12(3.85%)0.9040.342 Epidural catheter adjustments6(1.92%)7(2.24%)0.0790.779Maternal outcomes Spontaneous vaginal birth275(88.14%)264(84.62%)1.6480.199 Instrumental delivery37(11.86%)48(15.38%)//Fetal outcomes Fetal bradycardia00// Apgar score 1 min8.48 ± 0.508.52 ± 0.55-0.8380.402 Apgar score 5 min9.29 ± 0.589.23 ± 0.591.4410.150 Apgar score 10 min10.00 ± 0.0010.00 ± 0.00// Satisfaction 12.81 ± 1.46 11.26 ± 1.23 14.339 0.001 d: 1.15 [0.98, 1.32] Sat-Pain relief4.92 ± 0.274.88 ± 0.321.4830.139 Sat-Overall Experience 3.78 ± 0.91 1.91 ± 0.89 25.978 0.001 d: 2.08 [1.88, 2.28] Sat-Willingness to Recommend 4.11 ± 0.66 3.54 ± 0.50 12.256 0.001 d: 0.98 [0.82, 1.15]*Values are presented as mean ± SD or number (percentage). SSEA: Sequential Spinal-Epidural Analgesia, CSEA: Combined Spinal-Epidural Analgesia, ERMF: Epidural-Related Maternal Fever, NAVO: Nausea and Vomiting, PDPH: Post-Dural Puncture Headache, VAS: Visual Analogue Scale, PCEA: Patient-Controlled Epidural Analgesia, RD: Risk Difference, d: Cohen’s d Table 2Comparison of clinical outcomes between the two groupsOutcomesSSEA(n = 312)CSEA(n = 312)Test Statistic (χ²/t/Mann-Whitney)P-valueEffect Size (RD or Cohen’s d) [95% CI] Intraprocedural neurological symptom 5(1.60%) 80(25.64%) 76.612 0.001 RD: -0.240 [-0.291, -0.190]Adverse effects Hypotension within 20 min15(4.81%)12(3.85%)0.3480.555 ERMF7(2.24%)8(2.56%)0.0680.794 Pruritus within 48 h24(7.69%)31(6.73%)0.9770.323 NAVO9(2.88%)14(4.49%)1.1290.288 Urinary retention00// Chills3(0.96%)5(1.60%)0.5060.477 Toxicity00// Extensive block00// Neurological symptoms at 48 h 1(0.32%) 8(2.56%) 5.511 0.019 RD: -0.022 [-0.041, -0.004] Lower back pain 4(1.28%) 20(6.41%) 11.093 0.001 RD: -0.051 [-0.081, -0.021] PDPH4(1.28%)7(2.24%)0.8330.361 Persistent paresthesia at 1 week 0 16(5.13%) 14.464 0.001 RD: -0.051 [-0.076, -0.026]Analgesic efficacy VAS 0 min6.84 ± 1.056.76 ± 1.050.9530.341 VAS 1 min2.06 ± 0.791.99 ± 0.791.0180.309 VAS 5 min1.80 ± 0.751.89 ± 0.80-1.4430.150 VAS 10 min1.92 ± 0.802.04 ± 0.86-1.7930.073 VAS 30 min1.73 ± 0.761.84 ± 0.73-1.7140.087 VAS 60 min1.84 ± 0.831.89 ± 0.85-0.8080.420 VAS complete cervical dilation2.02 ± 0.822.08 ± 0.78-0.8980.369 VAS fetal head delivery2.15 ± 0.782.23 ± 0.81-1.2530.211 VAS placental delivery1.46 ± 0.501.50 ± 0.50-1.0410.298Analgesic quality Asymmetric blockage6(1.92%)4(1.28%)0.4070.524 S2 blockage262(83.97%)276(88.46%)2.9490.086 Bromage scores (any motor block)25(8.01%)24(7.69%)0.0220.882 PCEA boluses required25(8.01%)36(11.54%)2.1990.138 Manual epidural boluses required17(5.45%)12(3.85%)0.9040.342 Epidural catheter adjustments6(1.92%)7(2.24%)0.0790.779Maternal outcomes Spontaneous vaginal birth275(88.14%)264(84.62%)1.6480.199 Instrumental delivery37(11.86%)48(15.38%)//Fetal outcomes Fetal bradycardia00// Apgar score 1 min8.48 ± 0.508.52 ± 0.55-0.8380.402 Apgar score 5 min9.29 ± 0.589.23 ± 0.591.4410.150 Apgar score 10 min10.00 ± 0.0010.00 ± 0.00// Satisfaction 12.81 ± 1.46 11.26 ± 1.23 14.339 0.001 d: 1.15 [0.98, 1.32] Sat-Pain relief4.92 ± 0.274.88 ± 0.321.4830.139 Sat-Overall Experience 3.78 ± 0.91 1.91 ± 0.89 25.978 0.001 d: 2.08 [1.88, 2.28] Sat-Willingness to Recommend 4.11 ± 0.66 3.54 ± 0.50 12.256 0.001 d: 0.98 [0.82, 1.15]*Values are presented as mean ± SD or number (percentage). SSEA: Sequential Spinal-Epidural Analgesia, CSEA: Combined Spinal-Epidural Analgesia, ERMF: Epidural-Related Maternal Fever, NAVO: Nausea and Vomiting, PDPH: Post-Dural Puncture Headache, VAS: Visual Analogue Scale, PCEA: Patient-Controlled Epidural Analgesia, RD: Risk Difference, d: Cohen’s d Comparison of clinical outcomes between the two groups *Values are presented as mean ± SD or number (percentage). SSEA: Sequential Spinal-Epidural Analgesia, CSEA: Combined Spinal-Epidural Analgesia, ERMF: Epidural-Related Maternal Fever, NAVO: Nausea and Vomiting, PDPH: Post-Dural Puncture Headache, VAS: Visual Analogue Scale, PCEA: Patient-Controlled Epidural Analgesia, RD: Risk Difference, d: Cohen’s d Overall maternal satisfaction was significantly higher in the SSEA group (12.81 ± 1.46 vs. 11.26 ± 1.23, P < 0.001), corresponding to a large effect size (Cohen’s d = 1.15). When we examined the components of satisfaction, an interesting pattern emerged: the SSEA group scored significantly higher in “satisfaction with overall birth experience” (3.78 ± 0.91 vs. 1.91 ± 0.89, P < 0.001, d = 2.08) and “willingness to recommend” (4.11 ± 0.66 vs. 3.54 ± 0.50, P < 0.001, d = 0.98), yet no significant difference was found between groups in “satisfaction with pain relief” (4.92 ± 0.27 vs. 4.88 ± 0.32, P = 0.139). This dissociation suggests that the higher satisfaction in the SSEA group stems not from superior analgesia—since pain control was equivalent between techniques-but rather from factors related to procedural safety and the avoidance of alarming neurological symptoms. Notably, no statistically significant differences were observed between the two groups in other adverse effects (hypotension within 20 min, ERMF, pruritus within 48 h, NAVO, urinary retention, chills, local anesthetic systemic toxicity, extensive block, or PDPH), analgesic efficacy (VAS scores at all recorded time points), analgesic quality measures (asymmetric block, S2 blockage, Bromage scores, number of PCEA boluses required, manual epidural boluses, or epidural catheter adjustments), maternal outcomes (rates of spontaneous vaginal delivery vs. instrumental delivery), or fetal outcomes (bradycardia, 1-/5-/10-minute Apgar scores) (all P > 0.05). This pattern of findings—equivalent analgesia and common side effects but marked differences in neurological symptoms and satisfaction - points toward a specific mechanism related to neural protection rather than generalized improvement in all outcomes. Structural equation modeling analysis Because the model was saturated (df = 0), the fit indices are mathematically perfect (χ²=0.000, CFI = 1.000, TLI = 1.000, RMSEA = 0.000, SRMR = 0.000) and therefore do not provide meaningful information about model-data fit [13]. With this caveat in mind, we turn to the path coefficients, which offer insight into the relative strength of associations between clinical variables and maternal satisfaction. As shown in Table 3, after controlling for other variables, intraprocedural neurological symptoms showed the strongest negative association with satisfaction (β = -0.140, P < 0.001). This was followed by poor analgesic quality (β = -0.101, P = 0.011) and mean VAS scores (β = -0.092, P = 0.023). Fetal outcome demonstrated a marginally significant positive association (β = 0.078, P = 0.052), while maternal outcome showed no significant association (β = 0.043, P = 0.281). Perhaps most tellingly, the model explained only 4.2% of the variance in maternal satisfaction (R² = 0.042)—a finding that underscores the complexity of satisfaction as a construct and suggests that factors beyond traditional clinical metrics play a dominant role in shaping the birth experience. Table 3Structural equation modeling analysis of factors influencing patient satisfaction (n = 624)PathUnstandardized coefficientS.E.C.R.P-valueStandardized coefficient (β)Neurologic symptom → Satisfaction –0.634 0.181 –3.493 <0.001 –0.140 VAS mean → Satisfaction –0.516 0.226 –2.280 0.023 –0.092 Poor analgesic quality → Satisfaction –0.345 0.135 –2.547 0.011 –0.101 Maternal outcome → Satisfaction0.1970.1831.0780.2810.043Fetal outcome → Satisfaction0.2940.1521.9420.0520.078* VAS: Visual Analogue Scale. Model fit: χ²/df = 0.000 (df = 0), CFI = 1.000, TLI = 1.000, RMSEA = 0.000, SRMR = 0.000; R²= 4.2%. Due to the saturated (just-identified) model, fit indices are mathematically perfect and do not indicate good fit; the low R² indicates that most variance in satisfaction remains unexplained Table 3Structural equation modeling analysis of factors influencing patient satisfaction (n = 624)PathUnstandardized coefficientS.E.C.R.P-valueStandardized coefficient (β)Neurologic symptom → Satisfaction –0.634 0.181 –3.493 <0.001 –0.140 VAS mean → Satisfaction –0.516 0.226 –2.280 0.023 –0.092 Poor analgesic quality → Satisfaction –0.345 0.135 –2.547 0.011 –0.101 Maternal outcome → Satisfaction0.1970.1831.0780.2810.043Fetal outcome → Satisfaction0.2940.1521.9420.0520.078* VAS: Visual Analogue Scale. Model fit: χ²/df = 0.000 (df = 0), CFI = 1.000, TLI = 1.000, RMSEA = 0.000, SRMR = 0.000; R²= 4.2%. Due to the saturated (just-identified) model, fit indices are mathematically perfect and do not indicate good fit; the low R² indicates that most variance in satisfaction remains unexplained Structural equation modeling analysis of factors influencing patient satisfaction (n = 624) * VAS: Visual Analogue Scale. Model fit: χ²/df = 0.000 (df = 0), CFI = 1.000, TLI = 1.000, RMSEA = 0.000, SRMR = 0.000; R²= 4.2%. Due to the saturated (just-identified) model, fit indices are mathematically perfect and do not indicate good fit; the low R² indicates that most variance in satisfaction remains unexplained Discussion Guided by the theory of “Neuraxial Homeostasis” and its disruption, this study represents—to our knowledge—the first large-scale randomized controlled trial to systematically compare the impact of SSEA versus CSEA on neurological complications during labor analgesia. The results show a consistent pattern: while the two techniques were equivalent in terms of analgesic efficacy, maternal-neonatal outcomes, and common adverse effects, the SSEA group demonstrated substantially lower incidences of intraprocedural neurological symptoms, neurological symptoms at 48 h, lower back pain, and persistent paresthesia at 1 week. These differences were accompanied by substantially higher overall maternal satisfaction in the SSEA group—a finding that takes on particular significance given that satisfaction with pain relief itself did not differ between groups. What explains this pattern? The core difference between the techniques lies not merely in the order of steps but in the physical context of dural puncture. In SSEA, a fine-gauge spinal needle penetrates the dura in an anatomically undisturbed environment. In CSEA, by contrast, dural puncture occurs with the spinal needle passing through a larger-bore epidural needle that has already been positioned - a configuration that may alter tissue dynamics, tent the dura, and reduce the distance to nerve roots, potentially increasing the risk of contact. The research flow diagram is presented in Fig. 1. Fig. 1Research flow diagram. SSEA, sequential spinal-epidural analgesia; CSEA, combined spinal-epidural analgesia. *P < 0.05, †P < 0.01. ‡P < 0.001 vs. CSEA group Research flow diagram. SSEA, sequential spinal-epidural analgesia; CSEA, combined spinal-epidural analgesia. *P < 0.05, †P < 0.01. ‡P < 0.001 vs. CSEA group In our previous work [5], we introduced the concept of “Neuraxial Homeostasis” to describe the relatively stable state of intrathecal structures within the closed vertebral canal. We hypothesized that during classic CSEA, the initial insertion of the epidural needle disrupts this homeostasis, causing subtle movement of the dura and arachnoid and agitating the nerve roots floating in cerebrospinal fluid. When the spinal needle is subsequently introduced, it may contact these oscillating nerve fibers, eliciting neurological symptoms. Our current findings are consistent with this theoretical framework. It is important to emphasize, however, that these mechanisms were not directly measured in our study and should be regarded as hypotheses generated from our findings—hypotheses that warrant direct investigation in future research. We hypothesize that the marked reduction in symptoms with SSEA stems from performing dural puncture with a fine-gauge needle before any epidural instrumentation, thereby minimizing disturbance to the neuraxial environment. The existing literature offers several plausible physiological explanations that align with our observations. Based on 3D-MR models, when an epidural needle is positioned prior to dural puncture, the dura mater may be displaced toward the subarachnoid space, reducing the distance to dorsal nerve rootlets [14]. This, combined with transient compression of the epidural space volume [15] and potential increases in subarachnoid pressure during maternal pushing [16–18], could increase the probability of nerve root contact with the advancing spinal needle. The sequential SSEA puncture may avoid this cascade by performing dural puncture with a fine-gauge needle before any epidural instrumentation. While these mechanisms provide a theoretical framework, we acknowledge that they were not directly measured in our study and represent hypotheses generated from our findings, warranting further investigation. Concepts from spinal cord injury research may also offer insight. Squair et al. [19] proposed that Spinal Cord Perfusion Pressure (SCPP) = mean arterial pressure (MAP) - CSF pressure (CSFP), noting that for every 1 mmHg increase in CSFP, the probability of improvement in ASIA motor scores decreases by 7%. A multicenter study on acute spinal cord injury [20] showed that lumbar CSF drainage of 5 ml could increase SCPP by 9 mmHg and improve nerve root blood perfusion by 12%. Although our patients did not have spinal cord injury, nerve root blood flow follows the same pressure gradient. During CSEA, increased CSFP leading to decreased SCPP could theoretically reduce blood flow to spinal nerves, and when combined with mechanical stimulation, might result in reversible nerve conduction block - manifesting as the observed paresthesia within 1 week or lower back pain. While these mechanisms are speculative in the context of healthy laboring women, and were not directly assessed in our study, they provide a coherent theoretical framework for understanding why symptoms persisted for over one week in 5.13% of the CSEA group but resolved within one week in all SSEA patients, and why 6.41% of the CSEA group reported residual back pain compared to only 1.28% in the SSEA group. The drug regimen used in this study for both groups (0.1% ropivacaine + sufentanil 1 µg/ml for intrathecal load and 0.5 µg/ml for PCEA) aligns with ASA obstetric analgesia guidelines [21, 22], overlapping with recommended concentrations. Consequently, the finding of equivalent analgesic quality and maternal-neonatal outcomes between groups is consistent with the guideline’s conclusion of “equivalent analgesia.“ [21] Where our findings diverge from the existing literature is in the domain of neurological complications. Previous research has largely focused on common complications like PDPH, hypotension, and ERMF [7, 8], with immediate neurological insult primarily reported in case reports and lacking large-scale controlled data. The ASA guidelines list neurologic injury as a rare event (< 0.1%), but these statistics often rely on discharge diagnoses or insurance databases that do not capture immediate procedural radiating pain. By refining the observation window, we found a 25.64% incidence of immediate neurological symptoms in the CSEA group, suggesting that guidelines may substantially underestimate the incidence of reversible neural stimulation. The fact that 5.13% of these parturients had sensory disturbances lasting over a week further suggests that transient neural stimulation can sometimes lead to more persistent symptoms, particularly affecting sensory nerves. Perhaps our most clinically significant finding concerns the determinants of maternal satisfaction. Previous studies have often attributed high satisfaction with neuraxial analgesia primarily to analgesic strength [23–25]. In this study, however, VAS scores at all time points showed no significant differences between groups, both groups achieved mean VAS scores below 3 after analgesia initiation, indicating mild pain and adequate analgesic efficacy for both techniques. Yet overall maternal satisfaction was substantially higher in the SSEA group. When we examined the three satisfaction components, an intriguing dissociation emerged: satisfaction with pain relief showed no intergroup difference, suggesting that satisfaction is not solely related to whether pain is present but is closely tied to the perception of safety. Our exploratory structural equation model should be interpreted with caution. The model explains only 4.2% of the variance in satisfaction, and the satisfaction instrument used has not been formally validated. These findings identify associations that may inform hypothesis generation, but do not establish causality. With this caveat in mind, our exploratory SEM provided quantitative support for this interpretation. Among the clinical factors examined, intraprocedural neurological symptoms (β = -0.140, P < 0.001) had a stronger negative association with satisfaction than the pain experience itself (β = -0.092, P = 0.023) or poor analgesic quality (β = -0.101, P = 0.011). It must be noted, however, that the satisfaction instrument used was ad hoc and has not been formally validated; this finding should therefore be viewed as hypothesis-generating rather than confirmatory. This finding, which should be interpreted with caution due to the use of a non-validated satisfaction instrument and the exploratory nature of the SEM, may suggest that parturients’ concern for neurological health could potentially influence satisfaction in a manner that, within the limits of our model, showed a stronger association than pain scores.Labor pain, after all, is an anticipated and expected event for which parturients are psychologically prepared. Even if VAS scores are relatively high, if pain is perceived as a normal part of labor, its impact on overall satisfaction may be relatively contained. Neurological symptoms, limb jerking, paresthesia, unexpected radiating pain, are qualitatively different. They represent unanticipated complications that can provoke strong psychological distress, undermine perceived safety, and trigger anxiety about long-term sequelae. The threat to long-term functional safety naturally carries more weight than temporary pain perception. Thus, the higher satisfaction scores in the SSEA group, measured with a non-validated ad hoc instrument, point to a potential association that warrants investigation. The higher satisfaction appears to stem not from superior analgesia but from a procedural experience associated with fewer neurological symptoms. By preserving neuraxial homeostasis through sequential puncture, SSEA avoids the nerve fiber oscillation and potential contact associated with the epidural needle-guided approach in CSEA. This finding is consistent with, but does not directly prove, the “Neuraxial Homeostasis” hypothesis, suggesting that maintaining the stability of the neuraxial environment and avoiding perturbation of neural structures may be clinically as important as analgesic efficacy alone. These findings must be interpreted with extreme caution. The exploratory structural equation model was just-identified (saturated) and used a non-validated satisfaction measure; it explained only 4.2% of the variance in satisfaction scores. The results do not establish definitive predictors but rather identify exploratory associations that generate hypotheses for future, more robust research. Several aspects of our exploratory SEM merit careful interpretation. The model was just-identified (saturated), resulting in a Chi-square of 0 and degrees of freedom of 0. Consequently, fit indices (CFI = 1.000, RMSEA = 0.000, SRMR = 0.000) achieved perfect values by mathematical necessity, not indicating superior model-data fit, and should not be interpreted for goodness-of-fit [13, 26]. The value of the model lies instead in the significance and stability of the parameter estimates. We found that intraprocedural neurological symptoms, mean VAS scores, and analgesic quality were significantly negatively associated with patient satisfaction, suggesting the clinical importance of optimizing neuraxial techniques to maintain homeostasis. Equally important, however, is what the model does not show. These clinical indicators explained only 4.2% of the variance in maternal satisfaction, a finding that strongly suggests satisfaction with the delivery experience is a complex, multi-dimensional phenomenon shaped by factors far beyond traditional biomedical metrics. In our own clinical experience, we have observed that factors such as the quality of communication with healthcare providers, the presence of supportive companions, the sense of involvement in decision-making, and alignment between expectations and actual experience can profoundly influence how parturients evaluate their birth experience. Future research must incorporate these psychosocial factors—maternal expectations, quality of provider communication, environmental support in the delivery room, and the parturient’s sense of involvement and control—to fully understand the determinants of satisfaction and to develop interventions that address the full spectrum of influences on the birth experience. These findings must be interpreted with extreme caution. The exploratory structural equation model was just-identified (saturated) and used a non-validated satisfaction measure; it explained only 4.2% of the variance in satisfaction scores. The results do not establish definitive predictors but rather identify exploratory associations that generate hypotheses for future, more robust research. To our knowledge, this study provides preliminary quantitative evidence suggesting that, among the clinical factors examined, intraprocedural neurological symptoms may show the strongest negative association with satisfaction. This provides key clinical evidence for the theory of “Neuraxial Homeostasis,” indicating that maintaining neuraxial environmental stability and avoiding neural disturbance may be a crucial mechanism for optimizing the analgesic experience. The modified sequential SSEA puncture technique, by performing fine-needle subarachnoid block prior to epidural catheterization, significantly reduces the incidence of procedure-induced neurological symptoms. Our findings suggest that the primary pathway through which this technique improves maternal satisfaction is not by providing stronger analgesia, since pain control was equivalent, but by offering a procedural experience associated with fewer reported intraprocedural neurological symptoms, thereby mitigating the strongest negative factor associated with satisfaction at its source. These results have important implications for how we evaluate and optimize labor analgesia. They suggest that the traditional paradigm, which focuses predominantly on analgesic outcomes, should shift toward a new paradigm that equally emphasizes procedural safety and analgesic outcome. An ideal analgesic protocol must provide effective pain relief while minimizing the additional risks introduced by the medical intervention itself. From a practical standpoint, SSEA appears to achieve this balance, offering equivalent analgesia and was associated with a lower incidence of observed intraprocedural neurological symptoms. This study has several limitations that warrant consideration. The primary outcome was recorded by the unblinded operator without independent confirmation or use of a standardized checklist, increasing the risk of measurement bias. As such, findings related to intraprocedural neurological symptoms should be interpreted with caution. While the single-operator design eliminated inter-operator variability and strengthened internal validity, it may limit the generalizability of our findings to other practitioners or routine clinical settings with varying levels of expertise. The lack of blinding of the operating anesthesiologist was unavoidable due to the nature of the procedure, though we attempted to mitigate bias by blinding participants, outcome assessors, and data analysts. The analysis of multiple secondary outcomes without adjustment for multiplicity increases the risk of false-positive findings; these results should be viewed as exploratory and hypothesis-generating, requiring confirmation in future studies. 116 parturients (15.7%) were excluded after randomization due to conversion to cesarean section during labor (56 in the SSEA group, 60 in the CSEA group). While this rate is consistent with typical obstetric populations and was anticipated in our sample size calculation, the exclusion of these patients may limit the generalizability of our findings to parturients who require cesarean delivery. The large effect size for “overall birth experience” satisfaction, measured with a non-validated instrument, may be influenced by recall bias and the strong negative emotional impact of unexpected intraprocedural symptoms, which could disproportionately affect this global measure compared to the more specific rating of pain relief. The eligibility criteria restricted the sample to relatively low-risk parturients, limiting generalizability to older, higher BMI, or medically complex populations. The proposed mechanisms for our findings, while theoretically grounded, were not directly measured and remain hypothetical. Finally, it is important to reiterate that the satisfaction instrument used was non-validated and the exploratory SEM explained only 4.2% of the variance in satisfaction; thus, satisfaction-related findings should be interpreted as hypothesis-generating rather than confirmatory. Looking forward, several directions for future research emerge from this work. Direct measurement of CSF pressure dynamics during both techniques would help validate the proposed mechanisms. Multi-center studies involving multiple practitioners would establish generalizability. Inclusion of a broader range of parturients, including those with higher BMI, medical comorbidities, and advanced maternal age, would extend applicability. Most importantly, future studies should incorporate comprehensive assessment of psychosocial factors to better understand the determinants of satisfaction and to develop interventions that address the full spectrum of influences on the birth experience. In conclusion, guided by the theory of Neuraxial Homeostasis, optimizing the puncture sequence to reduce neurological complications appears to be an effective strategy for improving the quality of neuraxial labor analgesia and achieving higher maternal satisfaction. This study provides clear operational optimization directions and a theoretical basis for clinical practice, while highlighting the need for future research to confirm these findings in broader populations and to directly investigate the underlying mechanisms. For the practicing anesthesiologist, our findings suggest that a seemingly minor modification in technique, performing spinal puncture before epidural catheterization, can yield substantial benefits in reducing reported intraprocedural neurological symptoms and improving satisfaction without compromising analgesic efficacy. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1