INTRODUCTION
Labor pain is one of the most severe types of pain experienced by women during their lifetime. Pain perception varies widely depending on psychological, cultural, and physiological factors. Fear of labor pain contributes to the preference for cesarean section even in the absence of medical indications in some populations [1]. Therefore, effective prenatal counseling and appropriate analgesia play crucial roles in improving preference for vaginal delivery and maternal satisfaction.
Understanding the characteristics and mechanisms of labor pain is essential for developing a rational and effective analgesic strategy. Labor is divided into three phases. The first phase, from the onset of regular uterine contractions to full cervical dilatation, is characterized by visceral pain caused by uterine contractions and cervical dilatation. This pain is transmitted via the T10-L1 afferent nerves and is typically felt as a widespread discomfort in the lower abdomen and back. The second phase, ranging from full cervical dilatation to fetal delivery, is characterized by somatic pain due to tension and distension in the vagina, perineum, and pelvic floor. This pain is primarily transmitted through the pudendal nerves (S2-S4). The third phase, from fetal to placental delivery, involves mild visceral sensations associated with uterine contractions and placental separation [2,3].
These dynamic changes in pain characteristics and neurotransmitter pathways during labor have provided evidence supporting neuraxial labor analgesia (NLA) as the most effective pain management strategy, endorsed by the World Health Organization (WHO) and major professional societies as the gold standard for labor analgesia [4].
The efficacy and safety of NLA have significantly improved over the past two decades through extensive clinical experience and research, including technological advancements and pharmacological approaches. Collectively, these innovations have improved analgesic onset, block uniformity, sacral coverage, and maternal satisfaction, while minimizing motor blockade and drug exposure. Despite these advances, the clinical application of NLA remains influenced by persistent controversial and inconsistent interpretation of the literature. Long-standing concerns regarding delayed labor, increased operative delivery, higher oxytocin doses, maternal fever, and potential long-term effects on maternal mental health or child neurodevelopment continue to shape clinical practice and patient counseling.
Anesthesiologists are expected to not only provide technically proficient analgesia but also guide interdisciplinary teams and counsel patients based on clinical data that accurately reflect contemporary practice. By synthesizing the latest data on procedural approaches, pharmacological therapies, maternal and neonatal outcomes, and long-term safety, this review aims to support individualized and informed clinical decision-making. Furthermore, we aimed to provide mothers and their families with clear and balanced information on the benefits, potential risks, and expected clinical outcomes of NLA. Strengthening the shared understanding between healthcare providers and women undergoing labor will ultimately promote safer care, reduce anxiety related to labor pain management, and support a positive birth experience.
INITIATION TIME
The optimal timing of NLA initiation remains controversial. Traditionally, concerns have been raised about whether early NLA during the latent phase of labor (cervical dilation < 4 cm) may delay the progression of labor or increase the likelihood of a cesarean section [5]. A Cochrane review found no significant difference in cesarean section rates or maternal or fetal outcomes comparing early and delayed NLA (cervical dilation ≥ 4 cm), although early NLA increased the rate of oxytocin use [6]. Furthermore, very early NLA administration may slightly delay the onset of labor, and some investigators have suggested that the optimal timing for epidural analgesia (EA) is cervical dilation of ≥ 6 cm [7]. To date, conflicting results have been reported regarding the potential impact of NLA on obstetric outcomes, especially delayed labor [5-11]. Nevertheless, randomized controlled trials and meta-analyses have consistently shown that early NLA administration does not adversely affect maternal or neonatal outcomes.
Considering all these findings, the American College of Obstetricians and Gynecologists (ACOG) recommended in 2019 that maternal request alone is a valid reason to initiate analgesia unless there is a medical contraindication [12]. This interpretation supports the contemporary guidelines that prioritize patient-centered analgesia while avoiding unnecessary delays that may exacerbate pain, stress, and catecholamine-mediated physiological responses during labor.
INITIATION TECHNIQUES
The three major methods of analgesia currently used in obstetric practice are EA, combined spinal epidural (CSE), and dural puncture epidural (DPE) techniques [13]. Each technique differs in approach, analgesic onset, block characteristics, and maternal-fetal effects.
EA is the conventional and most widely performed analgesia technique. After identifying the epidural space, typically at the L3-L4 or L4-L5 interspace, using the loss-of-resistance method, a catheter is advanced 3-5 cm into the epidural space for intermittent or continuous infusion of dilute local anesthetic-opioid mixtures. EA provides reliable and adjustable analgesia throughout all stages of labor and can be easily extended to cesarean sections. The onset is relatively slow (15-20 min), and sacral coverage may be incomplete, sometimes requiring supplemental manual boluses [13]. Hemodynamic stability is generally well-preserved, and both maternal satisfaction and neonatal outcomes are favorable. Although a slight prolongation of the second stage and an increased rate of instrumental vaginal delivery have been reported, especially in conventional high-concentration drug use, high-quality meta-analyses have suggested that low-concentration EA does not increase cesarean section rates or adverse neonatal outcomes [14,15].
CSE analgesia combines the advantages of two techniques (epidural and spinal analgesia). After identifying the epidural space, a thin spinal needle (25-29 gauge [G]) is inserted through the epidural needle, and a small dose of local anesthetic and/or opioid is administered intrathecally before insertion of the epidural catheter. This method has a rapid onset within 2-5 min, effectively covers the sacral area, and achieves a high maternal satisfaction [16]. Despite concerns about delayed confirmation of epidural catheter function, the incidence of replaced catheters (CSE 1.49% vs. EA 3.18%) and the overall failure rate of CSE is lower than that of EA [17,18]. This is likely due to the need for a more centrally placed epidural needle for a successful spinal tap and the ability to confirm the epidural space indirectly by regurgitation of the spinal fluid. However, intrathecal medication administration can cause transient maternal hypotension, itching, or decrease the fetal heart rate (FHR) [19,20]. Large cohort and meta-analysis data have shown that CSE has no negative effects on cesarean section rates, length of labor, neonatal Apgar scores, or umbilical blood acidosis [19,21].
DPE is a modified form of CSE analgesia that involves puncturing the dura mater with a fine spinal needle, without intrathecal injection. This intentional micropuncture promotes communication between the intrathecal and epidural spaces, thereby improving the diffusion and homogeneity of epidural drugs. DPE avoids the itching and hemodynamic changes associated with intrathecal administration while offering a faster onset of action and superior sacral analgesia than EA alone [22-24]. DPE offers superior block quality and comparable obstetric and neonatal outcomes to conventional EA, establishing it as a viable alternative in modern labor analgesia practice [24,25]. DPE also reduces the need for catheter repositioning, likely for reasons similar to those for CSE. Intentional epidural puncture has been shown not to increase the incidence of post-dural puncture headache (PDPH) [26].
Comparative evidence suggests that differences in clinical outcomes for the EA, CSE, and DPE techniques are modest and largely mediated by onset kinetics, block density, and sacral coverage rather than by the intrinsic superiority of any single technique. CSE consistently provides the most rapid and reliable analgesia but is associated with a higher incidence of transient maternal and fetal side effects related to intrathecal drug administration. DPE offers an intermediate profile, improving block quality and symmetry while minimizing adverse intrathecal effects, whereas conventional EA remains a flexible and widely applicable option. Therefore, technique selection should be individualized to balance the need for rapid analgesia against maternal comorbidities, labor progression, and tolerance of potential side effects. Tables 1, 2 provide a brief comparison of these characteristics that will help clinicians select the most appropriate initiation technique.
DRUG CHOICES: LOCAL ANESTHETICS AND ADJUVANTS
With appropriate drug selection, concentration, and volume, all NLA methods can provide excellent pain relief with minimal impact on obstetric, maternal, and neonatal outcomes.
Three local anesthetics are clinically significant for labor analgesia: bupivacaine, levobupivacaine, and ropivacaine. The key physicochemical properties that determine the onset and duration of action include the ionization constant, lipophilicity, and the degree of protein binding [27]. As lipophilic drugs readily cross nerve cell membranes, their analgesic efficacy is closely related to their lipid solubility [13,27].
The primary goal of neuraxial drug selection is to achieve effective analgesia while minimizing motor blocks and adverse effects. The standard local anesthetic concentrations for EA have evolved from 0.25% bupivacaine to lower concentrations (0.1%) [15]. Low-concentration local anesthetics, when combined with adjuvant medications, have been shown to shorten the second stage of labor and reduce the rate of instrumental vaginal delivery by preserving motor function without compromising the quality of labor [15,28]. In recent years, ropivacaine and levobupivacaine have replaced racemic bupivacaine because of their reduced cardiotoxicity and tendency to induce motor blockade. These drugs are typically used at concentrations of 0.0625-0.1%, providing selective sensory blockade while minimizing motor disturbances [29].
Adding adjuvants to the epidural mixture further enhances analgesia and reduces the required local anesthetic dose, thereby reducing the risk of motor blockade and systemic local anesthetic toxicity. Commonly used adjuvants include lipophilic opioids (fentanyl and sufentanil), adrenergic agonists (clonidine and dexmedetomidine), neostigmine, and epinephrine.
Neuraxial opioids play a crucial role in labor analgesia by synergizing with local anesthetics to enhance analgesia, while reducing the minimum local analgesic concentration. Fentanyl is short-acting and reduces the minimum local analgesic concentration of bupivacaine by 31-72% depending on the dose used [30]. Sufentanil has a more rapid onset and shorter duration of action and is 4.5 times more potent than fentanyl, reducing the minimum local analgesic concentration of bupivacaine by up to 91% [31]. According to American Society of Anesthesiologists guidelines, adding fentanyl (1-3 µg/ml) or sufentanil (0.5-1 µg/ml) to epidural mixtures provides excellent analgesia while minimizing motor block. For intrathecal administration, a dose of 10-25 µg of fentanyl or 2-5 µg of sufentanil is recommended [32]. Furthermore, low- or ultra-low-concentration local anesthetic-opioid combinations (typically ropivacaine or levobupivacaine plus a lipophilic opioid) have been associated with increased rates of vaginal delivery, reduced need for instrumentation, and a shorter second stage of labor than conventional high-dose regimens [33,34].
α2-Adrenergic agonists exert their effects via spinal α2 receptors, producing dose-dependent analgesia and sedation. Intrathecal clonidine doses of 15-45 µg provide sustained analgesia but can cause hypotension [35]. Epidural doses of 30-150 µg are effective, but doses above 75 µg can cause sedation, hypotension, bradycardia, and FHR changes [13].
Epinephrine acts partially through α2 receptors and vasoconstriction. A systematic review showed that epinephrine prolongs intrathecal analgesia and motor blockade by approximately 60 min; however, its efficacy in EA during labor remains uncertain [36].
Overall, the current evidence supports a pharmacological strategy centered on low- and ultralow-concentration local anesthetic regimens combined with lipophilic opioids. This approach consistently optimizes the analgesic efficacy, maternal satisfaction, and obstetric outcomes by minimizing motor blockade through dose reduction and opioid synergy. Among commonly used medications, lipophilic opioids (fentanyl and sufentanil) are the only adjunctive agents approved by the U.S. Food and Drug Administration for epidural and intrathecal administration during obstetric anesthesia. However, other medications—including α2-adrenergic agonists, neostigmine, epinephrine, dexamethasone, midazolam, ketamine, and magnesium sulfate—are considered off-label for epidural or intrathecal administration, and their routine clinical use is not recommended until additional safety data become available [27]. The key pharmacological options, dosing ranges, and relative advantages and limitations are summarized in Table 1.
MAINTENANCE TECHNIQUES
Labor is a prolonged and dynamic process, and the choice of maintenance technique significantly influences the effectiveness and quality of NLA. Currently, two principal techniques are used: continuous epidural infusion (CEI) and programmed intermittent epidural bolus (PIEB), both of which are commonly supplemented with patient-controlled epidural analgesia (PCEA).
CEI—Initially, labor epidural doses were maintained with provider-administered boluses. With the introduction of automated epidural pumps, CEI has become the standard maintenance technique, providing stable analgesia through a basal infusion of 8-12 ml/h into the epidural space. Although this technique can provide stable analgesia, it results in a higher total anesthetic consumption and a higher incidence of motor blockade than intermittent bolus techniques [13].
PCEA, first described in 1988, allows patients to self-administer additional doses in response to breakthrough pain [37]. Better pain relief and reduced provider callbacks for additional top-ups can be achieved when background CEI is added to PCEA compared with PCEA alone, although at the expense of increased overall local anesthetic consumption [38]. When CEI is combined with PCEA, additional boluses (4-8 ml) are administered at a lockout time of 15-20 min.
PIEB was proposed as a more optimal maintenance technique compared with CEI in 2004 [39]. With PIEB, an epidural pump provides automated boluses administered at predetermined intervals (e.g., every 45 or 60 min). Rapid bolus administration promotes wider epidural spread and more effective dermatomal coverage than a slow infusion of the same hourly volume [40,41]. Consistent evidence shows that PIEB provides superior analgesia, a more uniform sensory block, reduced local anesthetic requirements, fewer rescue boluses, and higher maternal satisfaction compared with CEI [42,43]. Several studies have investigated the optimal PIEB settings for NLA; the optimal interval was found to be 40 min (90% effective interval), and the ED90 bolus dose was approximately 11 ml when administered at 40-min intervals [44,45].
The maintenance strategy plays a pivotal role in determining the quality and consistency of NLA. Accumulating evidence supports PIEB with or without PCEA as the most physiologically favorable approach, providing superior dermatomal spread, reduced local anesthetic consumption, and higher maternal satisfaction than CEI [38]. However, the benefits of PIEB are influenced by pump availability and patient tolerance to bolus dosing. CEI, while less optimal in terms of drug efficiency, remains an appropriate and pragmatic alternative for selected patients. These findings underscore that the maintenance technique, rather than the initiation method alone, is a key determinant of clinical success.
CLINICAL OUTCOMES
When assessing the quality of labor analgesia, clinicians should consider the method selected for initiating neuraxial technique, maintenance strategy, and analgesic medications discussed thus far, and assess outcomes at three levels: obstetrics, maternal, and offspring outcomes.
1. Obstetric outcomes are generally assessed in three domains: duration of labor (first and second stages), mode of delivery (spontaneous, instrumental vaginal, or cesarean), and the need for labor augmentation with oxytocin.
(1) Duration of labor: With modern low- or ultralow-concentration neuraxial regimens, none of the commonly used techniques consistently prolong the first stage of labor. The second stage may be modestly prolonged when a denser motor block develops; however, this effect can be mitigated by lowering local anesthetic concentrations and avoiding excessive background infusion rates [28,29]. Because CSE and DPE often provide more reliable sacral coverage, some studies have reported fewer supplemental boluses and a neutral or slightly shorter second stage than older high-concentration epidural regimens [19].
(2) Modes of delivery: Modern diluted neuraxial techniques do not increase cesarean delivery rates [28]. The likelihood of instrumental vaginal delivery correlates more strongly with the extent of motor blockade than with the neuraxial technique. Strategies that minimize motor block, such as using low-concentration local anesthetics and reducing unnecessary continuous basal infusions, are associated with lower instrumental delivery rates [3,28]. Overall, CSE and DPE perform similarly to EA with comparable or slightly improved maternal satisfaction and obstetric outcomes.
(3) Labor augmentation: An association between NLA and increased oxytocin doses has been reported [46]. Previous observational studies have suggested that the increased oxytocin use among women receiving EA, likely reflected confounding by prolonged labor or differing obstetric practices [9]. More recent analyses have demonstrated no independent association between EA and the need for augmentation; however, a large retrospective study found that although total oxytocin doses were higher overall, the proportion of women requiring augmentation during vaginal delivery was lower in those receiving EA [46].
Recent evidence has consistently demonstrated that modern NLA using dilute local anesthetic regimens does not adversely affect labor duration, mode of delivery, or the need for operative intervention. In particular, the 2018 WHO Intrapartum Care Guidelines advise against prophylactic oxytocin administration to prevent labor delay in women receiving NLA, emphasizing that transient changes in cervical dilation following epidural placement do not justify routine augmentation and that oxytocin should be reserved for clearly defined clinical indications [4].
Taken together, the obstetric outcomes of NLA are primarily driven by the density of the motor block [3]. Concerns regarding prolonged labor or increased instrumental delivery largely originate from historical practices associated with dense motor blockade using high-dose epidural regimens. Although analgesic techniques and dosing strategies have evolved substantially, such concerns persist in some obstetric settings and may continue to influence clinical decision-making. These perceptions can result in delayed or restricted access to effective analgesia, exposing laboring women to unnecessary pain, and in some cases, contributing to the avoidance of vaginal delivery. In this context, anesthesiologists should play a critical role in reducing obstetric concerns by applying evidence-based analgesic strategies and facilitating interdisciplinary communication.
2. Maternal outcomes include neuraxial procedure-related complications, analgesic efficacy and side effects, epidural-related maternal fever, and other long-term consequences.
(1) Procedure-related complications
- Unintended dural puncture (UDP) occurs in 0.5-1.5% of labor epidurals. In a 10-year observational study of 7,718 procedures, the incidence of UDP was 1.25% (95% confidence interval [CI] 1.01-1.53) [47]. Patients who underwent UDP experienced complications including total spinal analgesia (1.0%), dilation block (3%), hypotension (4.1%), abnormal FHR (2%), inadequate analgesia (14.4%), and general anesthesia after failed NLA (33.3% of emergency cesarean sections) [47]. Overall, 82.5% of women experienced prolonged hospital stays. If UDP occurs, healthcare providers should immediately inform patients and advise them of the available options for labor analgesia: using a non-neuraxial analgesic (systemic opioids, inhalation analgesia, etc.), repositioning the epidural catheter (usually in the upper intervertebral space), or inserting a spinal catheter for patients who are particularly difficult to re-insert. If an epidural catheter is reinserted after UDP, the initial analgesia should be administered slowly in divided doses to manage the expected analgesia. For spinal catheters, 0.1% levobupivacaine (2.5 mg) and sufentanil (3 µg) are administered at a 6 cm cervical dilatation, followed by 0.1% levobupivacaine (2.5 mg). However, there is no established agreed-upon protocol for managing spinal catheters during labor or cesarean section anesthesia. Patients experiencing UDP should be evaluated daily for headache and neurological symptoms.
- PDPH: Although PDPH is rare after uncomplicated NLA (0.44%), more than half of women develop PDPH after UDP with a 16- to 18G epidural needle [13]. The use of pencil-point 27-29G needles during CSE or DPE does not increase the risk of PDPH [13]. Conservative treatment includes analgesics, hydration, supine rest, and prophylactic anticoagulation therapy. According to the 2019 guidelines of the Association of Obstetricians and Gynecologic Anesthesiologists (AOGA), an epidural blood patch (EBP) is considered the gold standard for patients with persistent or severe headaches that do not respond to conservative treatment, with 50-80% complete or partial relief [48-50]. If symptoms are not completely relieved, a second EBP may be needed in addition to evaluating other potential causes of headache. Numerous adjunctive treatments have been investigated, including medications, such as caffeine, theophylline, adrenocorticotropic hormones, steroids, and triptans, as well as invasive procedures, such as acupuncture, occipital or sphenopalatine nerve blocks, epidural morphine, and epidural or spinal fluid injections. However, more evidence is needed to make definitive recommendations and further research is needed to understand their effectiveness.
- Neurological complications: Most postpartum neurological symptoms are due to obstetric or positional causes rather than labor analgesia, and severe neurological impairment due to NLA is rare [51]. In patients with neurological impairment, it is crucial to differentiate between symptoms requiring immediate neurological or neurosurgical evaluation; however, NLA can interfere with this differentiation [52]. The American Society of Anesthesiologists and Association of Women’s Health, Obstetrics, and Neonatal Nurses (AWHONN) guidelines support the assessment of motor and sensory deprivation after NLA [53,54]. Postpartum neurological monitoring should include hourly assessments until the motor block resolves and consultation with an anesthesiologist for evaluation if the motor block persists 4 h after the last medication [53,54]. Tools for assessing motor function include the Bromage scale and the straight leg-raising test. Early detection and intervention are essential to prevent long-term damage. Table 3 summarizes the incidence, risk factors, and warning signs of neurological complications.
- Ultrasound-guided lumbar EA: Surface landmarks are inaccurate in 30-60% of pregnant women [55]. Ultrasound guidance also improves the accuracy of intervertebral space identification, which in turn allows optimal puncture site selection and prediction of epidural depth. The correlation between the ultrasound-estimated depth and the actual Tuohy needle depth at the loss of resistance has been reported to be excellent [56]. Compared with landmark-based techniques, ultrasound-guided lumbar neuraxial procedures are associated with lower technical failure rates (relative risk [RR], 0.5), fewer traumatic insertions (RR, 0.3), and fewer needle passes and redirections [56,57]. Improved midline identification and lower lumbar positioning are associated with improved sacral coverage and maternal satisfaction [58].
Ultrasound is particularly useful in patients with obesity, scoliosis, prior spinal surgery, or difficulty in determining the epidural position [59,60]. Specifically, ultrasound mapping performed before resident procedures has been associated with higher initial success rates and fewer needle redirections and catheter repositions, suggesting that this procedure has particular educational value [61]. However, the benefits appear to be limited in patients with readily palpable landmarks or when the procedure is performed by expert anesthesiologists [62].
(2) Analgesic efficacy and adverse effects: Recent studies have examined whether the insertion level of epidural catheters influences the quality of NLA. Traditionally, the L3-4 or L4-5 interspace has been the most practical and widely used site [13]. A randomized controlled trial demonstrated that neuraxial procedures at L2-3 resulted in a faster onset of analgesia and higher rates of pain-free uterine contractions, although sensory and motor block characteristics and adverse events did not differ significantly from neuraxial procedures at L3-4 [63]. However, landmark-guided techniques frequently result in catheter placement at one level higher than intended [55], and targeting the L2-3 interspace without ultrasound guidance may increase the risk of nerve root contact. Additionally, when UDP occurs, subsequent attempts often require placement at a higher level, further complicating the use of L2-3 in routine practice.
Although L5-S1 insertion has been proposed to improve sacral coverage, the available evidence does not demonstrate a clinically meaningful advantage over traditional epidural placement. In one study, S2 blockade after 10 ml of 0.125% bupivacaine was similar between L5-S1 and standard techniques (91% vs. 81%, P = 0.24) [64].
Overall, improvements in sacral coverage and block uniformity were driven by the initiation technique and maintenance regimens rather than by the catheter insertion level. Starting with CSE or DPE, PIEB, with or without PCEA is performed using low- and high-dose local anesthetics and opioids. Across all techniques, the order of analgesic initiation and block uniformity generally is as follows: CSE > DPE > EA [32]. Local anesthetic and opioid consumption are lowest when block distribution is optimized, and maintenance therapy using PIEB with or without PCEA is used rather than continuous infusion [13].
Maternal physiological adverse events vary depending on the neuraxial technique and intrathecal administration. CSE is consistently associated with the highest incidence of transient hypotension and pruritus, which are primarily attributable to intrathecal administration. A meta-analysis by Grangier et al. [65] found that CSE opioid administration significantly increased the incidence of nausea and vomiting (RR 1.31), pruritus (RR 4.26), and fetal bradycardia (RR 2.38) compared with EA alone, as well as a higher incidence of maternal hypotension (RR 1.54). These hemodynamic effects reflect rapid sympathectomy and a rapid decrease in circulating catecholamines following intrathecal administration. Standard precautions, such as left lateral tilt, concurrent fluid administration, and vasopressor support, are effective in minimizing this risk. Pruritus is primarily observed in CSE and is directly associated with intrathecal opioids. Nausea and vomiting are typically multifactorial and are more related to systemic opioid exposure or transient hypotension than to neuraxial drugs alone. In contrast, DPE is associated with stable maternal hemodynamics and minimal pruritus, a side-effect profile similar to that of EA [32,65]. Consequently, CSE provides the fastest and most potent analgesia, but it should be noted that it may carry a greater burden of transient maternal side effects.
(3) Epidural-related maternal fever: Epidural-related maternal fever is a non-infectious intrapartum hyperthermia (typically ≥ 38.0°C) occurring in women receiving epidural labor analgesia [66]. Observational studies have reported an incidence of approximately 15-25% [66,67]. The factor most consistently associated with epidural-related maternal fever is the duration of epidural catheterization, independent of the neuraxial technique, drug regimen, or technical characteristics [13].
The mechanisms underlying epidural-related maternal fever remain unclear. Proposed hypotheses include altered hypothalamic thermoregulation, cytokine-mediated inflammatory activation, and decreased heat production due to epidural-induced sympathetic blockade [66,67]. Management of epidural-related maternal fever focuses on minimizing prolonged catheter stay and ongoing surveillance for infectious causes such as chorioamnionitis. Glucocorticoids appear to be effective in reducing fever; however, empirical antibiotics do not prevent epidural-related maternal fever [67].
Maternal fever of any cause can negatively impact fetal outcomes (decreased Apgar scores, increased neonatal sepsis evaluations, and increased neonatal intensive care unit [NICU] admission rates). Early communication with neonatal providers is essential to prevent unnecessary antibiotic exposure or NICU admissions due to suspected sepsis. However, unlike infectious fever, it remains unclear whether epidural-related maternal fever causes independent short- or long-term harm. Current evidence suggests that neonatal sequelae are secondary to conservative treatment (e.g., empiric antibiotics) rather than direct thermal injury or the spread of inflammation [68,69].
(4) Long-term outcomes—Postpartum depression (PPD): According to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), PPD is a major depressive disorder that occurs during pregnancy or within 4 weeks of delivery [70]. Its prevalence is reported to be as high as 20%, and if left untreated, PPD can persist for months or years. Although the most well-known risk factor for PPD is a history of mood disorders, pain and depression are universally and bidirectionally associated.
NLA has been proposed as a potential protective factor against PPD because it effectively reduces pain during labor. However, the evidence for this remains inconsistent. A meta-analysis of 11 studies (n = 85,928) found no association between NLA and PPD (adjusted odds ratio [OR] 1.03, 95% CI 0.77-1.37) [71]. Li et al. [70] reported a trend toward a lower incidence of PPD in women receiving NLA within 1 and 4 weeks postpartum. However, this finding was not confirmed in a subsequent meta-analysis of nine studies (n = 4,442), which found no significant difference (OR 1.02, 95% CI 0.62-1.66) [72]. To date, the observational nature of most studies, the heterogeneity of PPD assessment tools, and residual confounding factors make it impossible to establish a causal relationship between NLA and the risk of PPD.
Given the strong association between pain and depressive symptoms, optimizing labor analgesia is clinically justified, but there is insufficient evidence to recommend administering or discontinuing NLA solely for the purpose of preventing PPD. Future well-designed prospective studies incorporating predefined covariates and standardized PPD follow-up periods are required to confirm these relationships.
3. Offspring (fetal and neonatal) outcomes
Evaluation of offspring outcomes commonly includes FHR abnormalities, Apgar scores, umbilical cord acid-base status, NICU admission, early feeding/breastfeeding, and long-term neurodevelopment.
(1) Immediate neonatal outcome: Transient FHR changes (typically brief bradycardias or decelerations) occur more frequently after CSE and are associated with the rapid onset of analgesia. These symptoms typically resolve with conservative treatment and are rarely clinically significant. In contrast, EA and DPE have low and similar incidences of clinically significant FHR abnormalities, respectively. Modern low-concentration NLA therapy does not increase the risk of neonatal acidosis, and multiple studies have reported no consistent differences in Apgar scores or cord blood gas values between the EA, CSE, and DPE techniques. The NICU admission rates were not associated with maternal or neonatal etiologies.
(2) Neurodevelopmental outcomes: A 2020 retrospective cohort analysis (147,895 children) reported a 37% increased risk of autism spectrum disorder (ASD) after maternal NLA exposure [73]. However, the study has significant methodological limitations, including incomplete adjustment for socioeconomic, obstetric, and familial factors and reliance on data from a single healthcare system. Nonetheless, the findings generated public concern and prompted several professional societies in obstetrics, gynecology, and anesthesiology to issue official reassurance statements.
Since then, several large registry-based studies across diverse populations have reexamined the association between NLA and neurodevelopment [74-76]. Although small positive associations (hazard ratio 1.08-1.14) were significantly attenuated or eliminated in sibling-matched analyses, this suggests that residual familial confounding likely explains the underlying signal [74].
Using Scottish National Data (n = 435,000), Kearns et al. [77] analyzed data from the Scottish National Health Service (n = 435,000) and reported a small reduction in the overall neurodevelopmental delay in children exposed to maternal NLA after adjusting for perinatal and family variables. (RR 0.96, 95% CI 0.93-0.98). Similarly, Ren et al. [76] using a Danish national cohort reported an 11% increased hazard of ASD and a 12% increased hazard of specific developmental disorders in unadjusted models, but these associations disappeared in within-sibling analyses, again implicating shared familial factors rather than exposure itself.
Although animal experimental studies have shown local anesthetic-induced neuronal apoptosis in immature spinal cord tissue [78-80], the clinical relevance of these findings remains uncertain given the limited placental transmission of epidural medications and the significant neurodevelopmental differences between animal models and human fetuses.
Collectively, recent epidemiological evidence suggests little or no causal association between NLA and adverse neurodevelopmental outcomes. The relationship between ASD and NLA likely reflects unmeasured confounders rather than the true effects of neuraxial techniques.
CONCLUSION
Modern NLA has evolved into a flexible evidence-based approach that can be tailored to individual maternal and obstetric circumstances. The selection of initiation and maintenance techniques should be guided by labor progression, maternal comorbidities, and tolerance of potential side effects rather than by outdated concerns regarding labor prolongation or increased cesarean delivery. Technological advances, particularly the incorporation of ultrasound guidance, have further improved the safety and success of neuraxial procedures and reduced the incidence of technical complications. Establishing structured management strategies to address complications is equally important.
In addition, maternal fever associated with EA must be carefully differentiated from infectious etiologies through appropriate treatment pathways and multidisciplinary collaboration among anesthesiologists, obstetricians, and neonatologists. Accordingly, a consolidated clinical algorithm integrating ultrasound-guided neuraxial placement, management of technical complications, and differentiation of epidural-related maternal fever from infectious causes has been proposed and is illustrated in Fig. 1.
Ultimately, the safe and effective application of NLA depends not only on technique selection, but also on anesthesiologist-led stewardship, continuous clinical assessment, and clear communication with patients and the multidisciplinary care team. In this context, the present review aimed to bridge persistent clinical controversies and daily practice, thereby supporting confident, individualized analgesic care throughout labor.
There is increasing interest in the long-term maternal and neonatal outcomes associated with NLA. Future research should prioritize the use of standardized outcome measures and well-designed prospective studies with extended follow-up periods. In parallel, continued refinement of individualized analgesic management algorithms that account for maternal characteristics, labor dynamics, and institutional resources is essential to further optimize clinical practice.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
