Understanding maternal pain and psychological vulnerabilities associated with the development of sub-acute pain after childbirth.

This is a secondary analysis of a prospective cohort study involving data collected from a randomized clinical trial conducted at KK Women’s and Children’s Hospital between June 2017 and July 2021 [ 8 ]. The study was approved by SingHealth Centralized Institutional Review Board (Ref no. 2017/2090) on 25/03/2017 and registered on Clinicaltrials.gov ( NCT03167905 ) on 30/05/2017. The primary study examined the association between the use of labor epidural analgesia and postpartum depression at 6 to 10 weeks postpartum; and the SAPC outcomes were not relevant to the primary objective of the study. Written informed consent was obtained from all participants on the primary study, patients acknowledged that the collected data was also used for other secondary analyses. The analysis of this study on SAPC adheres to the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines [ 9 ]. The study cohort included women aged 21 years or older, having a singleton pregnancy with ≥ 36 gestational weeks, and classified as American Society of Anesthesiologists (ASA) physical status II. Those who had multiple pregnancies, non-cephalic fetal presentation, presence of obstetric and uncontrolled medical complications (e.g., pre-eclampsia, prolonged premature rupture of membrane (PROM) for ≥ 48 h, gestational diabetes mellitus (GDM) on insulin, hypertensive disorder of pregnancy on medication), or underwent elective cesarean delivery, were excluded from the study. As this study involved the use of validated questionnaires in English versions, participants who could not understand nor read English were also excluded. The primary exposure was measured before delivery (upon admission to the delivery suite when patients were not in distress or severe pain (numeric rating scale (NRS) pain score \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\le\:\:$$\end{document} 3), using the Central Sensitization Inventory (CSI). The CSI is a two-part psychometric instrument, with Part A comprising 25 items assessing symptoms of central sensitivity syndromes with total scoring from 0 to 100; and Part B consisting of ten items on diagnosis of pain-related conditions relevant to pain hypersensitivity [ 10 ]. The CSI has been validated in the context of chronic pain conditions including fibromyalgia and post-operative patients with endometriosis [ 11 , 12 ]. Participants were also administered the validated questionnaires listed below to assess their pre-delivery pain and psychological vulnerabilities: (i) Pain Catastrophizing Scale (PCS): A 13-item measure of maladaptive pain-related cognitions, with total scores ranging from 0 to 52 across three subscales: helplessness, magnification, and rumination [ 13 ]. PCS has demonstrated high internal consistency (Cronbach α = 0.87–0.93) [ 14 ]; (ii) Edinburgh Postnatal Depression Scale (EDPS): A 10-item self-report survey assessing antenatal and postnatal depression [ 15 , 16 ], and validated in pregnant women in Singapore with sensitivity and specificity of 82% and 86% respectively [ 17 ]; (iii) Fear-Avoidance Components Scale (FACS): A patient-report measure on fear-avoidance beliefs to pain, and demonstrates high internal consistency in both healthy volunteers and patients with chronic musculoskeletal pain disorders [ 18 ]; (iv) State Trait Anxiety Inventory (STAI): A measure of state (situational) and trait (general) anxiety, validated for use during pregnancy and peripartum period [ 19 , 20 ]; (v) Perceived Stressed Scale (PSS): A 10-item scale ranging from 0 to 40 and assessing perceived stress and stressful life situations [ 21 ]. PSS has been extensively validated in the pregnant population [ 22 ]; (vi) Angle Labor Pain Questionnaire (A-LPQ): A 22-item condition-specific psychometric instrument ranging from 0 to 220 and measuring childbirth pain across five subscales: The enormity of the pain, fear/ anxiety, uterine contraction pain, birthing pain, and back pain/ long haul [ 23 ]. It demonstrates excellent test-retest reliability and high internal consistency (Cronbach α = 0.94) [ 23 ]; (vii) EuroQol Five-Dimensional Three-Level (EQ-5D-3 L): The visual analog scale (VAS) score from EQ-5D-3 L ranging from 0 to 100 was collected to determine the health-related quality of life [ 24 ]. Lower scores have been associated with pregnancy complicated by medical conditions [ 25 ]. Pain Catastrophizing Scale (PCS): A 13-item measure of maladaptive pain-related cognitions, with total scores ranging from 0 to 52 across three subscales: helplessness, magnification, and rumination [ 13 ]. PCS has demonstrated high internal consistency (Cronbach α = 0.87–0.93) [ 14 ]; Edinburgh Postnatal Depression Scale (EDPS): A 10-item self-report survey assessing antenatal and postnatal depression [ 15 , 16 ], and validated in pregnant women in Singapore with sensitivity and specificity of 82% and 86% respectively [ 17 ]; Fear-Avoidance Components Scale (FACS): A patient-report measure on fear-avoidance beliefs to pain, and demonstrates high internal consistency in both healthy volunteers and patients with chronic musculoskeletal pain disorders [ 18 ]; State Trait Anxiety Inventory (STAI): A measure of state (situational) and trait (general) anxiety, validated for use during pregnancy and peripartum period [ 19 , 20 ]; Perceived Stressed Scale (PSS): A 10-item scale ranging from 0 to 40 and assessing perceived stress and stressful life situations [ 21 ]. PSS has been extensively validated in the pregnant population [ 22 ]; Angle Labor Pain Questionnaire (A-LPQ): A 22-item condition-specific psychometric instrument ranging from 0 to 220 and measuring childbirth pain across five subscales: The enormity of the pain, fear/ anxiety, uterine contraction pain, birthing pain, and back pain/ long haul [ 23 ]. It demonstrates excellent test-retest reliability and high internal consistency (Cronbach α = 0.94) [ 23 ]; EuroQol Five-Dimensional Three-Level (EQ-5D-3 L): The visual analog scale (VAS) score from EQ-5D-3 L ranging from 0 to 100 was collected to determine the health-related quality of life [ 24 ]. Lower scores have been associated with pregnancy complicated by medical conditions [ 25 ]. Demographic data, including age, race, marital status, occupation, etc. were collected accordingly. Additional data on labor pain management, obstetric and neonatal characteristics were also recorded. An online survey was conducted at 6 to 10 weeks postpartum to collect the pain outcome on SAPC. The primary objective was the association between pre-delivery CSI scores and status of SAPC at 6 to 10 weeks postpartum. The primary exposure of pre-delivery CSI score would be treated as continuous variable; whereas the primary outcome of SAPC was determined by an online survey at 6 to 10 weeks postpartum, and the presence of SAPC was defined as: answered “yes” to either having had ongoing pain at 4 weeks or more postpartum; or currently had pain relating to their delivery. All variables were presented and summarized based on their status of SAPC as follows: Continuous variables as mean (standard deviation (SD)), and categorical variables as frequency (proportion) respectively. Comparisons between with and without SAPC groups were performed either a two-sample independent t-test or a Mann-Whitney U-test (continuous variables, depending on the distribution); or Chi-square test (categorical variables). Univariate and multivariable logistic regression analyses were performed to determine the potential factors associated with risk of SAPC. The quantitative associations derived from the logistic regression models were expressed as odds ratio (OR) with corresponding 95% confidence interval (95% CI). Variables with p  < 0.10 in the univariate regression analysis were considered for the subsequent multivariable logistic regression analysis. The final multivariable model was developed using a stepwise variable selection method. Statistical significance was set at p  < 0.05 and all tests were two-tailed. All statistical analyses were performed using SAS version 9.4 software (SAS Institute, Cary, North Carolina, USA).

Out of the 881 parturients enrolled from the primary study, 880 completed the pre-delivery surveys, with one patient withdrawn from the study as she changed her mind for the study participation before the questionnaire administration (Fig.  1 ). A subsequent 816 completed the online survey at 6 to 10 weeks postpartum, with 99 (12.1%) developed SAPC. There were no significant differences in the demographic characteristics between the groups with and without SAPC, except race and history of having diarrhea during menstrual cycles (Table  1 ). Fig. 1 Study flow diagram. SAPC subacute pain after childbirth Study flow diagram. SAPC subacute pain after childbirth Table 1 Demographic characteristics of patients Variables No SAPC ( n  = 717) Developed SAPC ( n  = 99) P Age (years), mean (SD) 31.0 (4.0) 30.7 (3.9) 0.473 Race, n (%) 0.055  Chinese  Malay  Indian  Others 400 (56.2) 39 (5.5) 190 (26.7) 83 (11.7) 41 (41.8) 6 (6.1) 37 (37.8) 14 (14.3) 0.010 0.979 0.030 0.558 Marital status, n (%) 1.000  Single/ divorced/ separated/ others  Married 18 (2.5) 691 (97.5) 2 (2.0) 96 (98.0) - - Housing type, n (%) 0.334  Public  Private 561 (89.2) 68 (10.8) 82 (93.2) 6 (6.8) - - Housing status, n (%) 0.925  Renting  Owned 66 (11.7) 497 (88.3) 10 (12.8) 68 (87.2) - - Number of biological children, n (%) 0.153  0  1  2  3 442 (63.9) 164 (23.7) 62 (9.0) 24 (3.5) 72 (73.5) 14 (14.3) 10 (10.2) 2 (2.0) 0.080 0.050 0.831 0.661 Age at first menstrual period, mean (SD) 12.8 (2.4) 12.7 (2.5) 0.737 Mood changes during menstrual period, n (%) 0.960  Always/ sometimes  Rarely/ never 486 (71.8) 191 (28.2) 69 (72.6) 26 (27.4) - - Cramps during menstrual period, n (%) 0.248  Always/ sometimes  Rarely/ never 448 (66.1) 230 (33.9) 69 (72.6) 26 (27.4) - - Nausea during menstrual period, n (%) 0.501  Always/ sometimes  Rarely/ never 126 (18.6) 551 (81.4) 21 (22.1) 74 (77.9) - - Vomiting during menstrual period, n (%) 0.982  Always/ sometimes  Rarely/ never 83 (12.3) 594 (87.7) 11 (11.6) 84 (88.4) - - Diarrhea during menstrual period, n (%) 0.013  Always/ sometimes  Rarely/ never 196 (28.9) 482 (71.1) 40 (42.1) 55 (57.9) - - Headache during menstrual period, n (%) 0.401  Always/ sometimes  Rarely/ never 299 (44.3) 376 (55.7) 47 (49.5) 48 (50.5) - - SAPC subacute pain after childbirth, SD standard deviation Demographic characteristics of patients Chinese Malay Indian Others 400 (56.2) 39 (5.5) 190 (26.7) 83 (11.7) 41 (41.8) 6 (6.1) 37 (37.8) 14 (14.3) 0.010 0.979 0.030 0.558 Single/ divorced/ separated/ others Married 18 (2.5) 691 (97.5) 2 (2.0) 96 (98.0) - - Public Private 561 (89.2) 68 (10.8) 82 (93.2) 6 (6.8) - - Renting Owned 66 (11.7) 497 (88.3) 10 (12.8) 68 (87.2) - - 0 1 2 3 442 (63.9) 164 (23.7) 62 (9.0) 24 (3.5) 72 (73.5) 14 (14.3) 10 (10.2) 2 (2.0) 0.080 0.050 0.831 0.661 Always/ sometimes Rarely/ never 486 (71.8) 191 (28.2) 69 (72.6) 26 (27.4) - - Always/ sometimes Rarely/ never 448 (66.1) 230 (33.9) 69 (72.6) 26 (27.4) - - Always/ sometimes Rarely/ never 126 (18.6) 551 (81.4) 21 (22.1) 74 (77.9) - - Always/ sometimes Rarely/ never 83 (12.3) 594 (87.7) 11 (11.6) 84 (88.4) - - Always/ sometimes Rarely/ never 196 (28.9) 482 (71.1) 40 (42.1) 55 (57.9) - - Always/ sometimes Rarely/ never 299 (44.3) 376 (55.7) 47 (49.5) 48 (50.5) - - SAPC subacute pain after childbirth, SD standard deviation The pre-delivery pain and psychological vulnerabilities of patients are summarized in Table  2 . The primary exposure of CSI score was significantly different between the groups with and without SAPC at 6 to 10 weeks postpartum ( p  = 0.003). In addition, the two groups also showed significant differences in FACS Activity subscale ( p  = 0.007), EQ-5D-3 L VAS ( p  = 0.003), all A-LPQ parameters excluding the birthing pain subscale ( p  < 0.05). By examining the obstetrics, analgesic, and neonatal characteristics, the number of pain relief requests ( p  = 0.006) and administrations ( p  = 0.005), mode of labor onset ( p  = 0.003), mode of delivery ( p  = 0.004), fetal presentation ( p  = 0.019), volume of blood loss ( p  = 0.005), infant height ( p  = 0.004) and weight ( p  = 0.015), were significantly different in groups without and without SAPC (Table  3 ). Table 2 Psychological and pain vulnerabilities of patients Variables No SAPC ( n  = 717) Developed SAPC ( n  = 99) P CSI total score 52.8 (12.3) 57.0 (13.2) 0.003 A-LPQ  Total score  Uterine contraction pain  Fear/ anxiety  Back pain/ long haul  Birthing pain  The enormity of the pain 126.7 (48.8) 25.5 (10.3) 22.7 (10.8) 29.9 (14.1) 21.5 (13.3) 27.1 (15.3) 139.7 (48.8) 27.7 (9.8) 25.2 (10.5) 33.7 (14.6) 22.7 (14.0) 31.0 (16.2) 0.017 0.043 0.032 0.018 0.431 0.027 EPDS total score 7.5 (4.2) 8.2 (5.1) 0.193 EQ-5D-3 L VAS 75.0 (15.9) 68.8 (19.2) 0.003 FACS  Total score  Pain  Activity  Victimization 37.4 (18.1) 13.0 (8.2) 21.0 (9.6) 3.4 (3.1) 41.3 (18.4) 14.1 (8.4) 23.9 (9.9) 3.7 (3.2) 0.055 0.224 0.007 0.389 PCS  Total score  Helplessness  Magnification  Rumination 18.5 (11.2) 7.5 (5.3) 4.0 (2.8) 7.0 (4.4) 20.2 (12.1) 8.0 (5.7) 4.5 (2.9) 7.8 (4.4) 0.190 0.415 0.105 0.095 PSS total score 16.1 (6.0) 16.6 (6.5) 0.470 STAI  Total score  State anxiety  Trait anxiety 77.8 (19.1) 39.4 (11.4) 38.4 (9.2) 79.0 (18.7) 39.9 (10.6) 39.2 (9.9) 0.553 0.667 0.451 The values are expressed in mean (SD) A-LPQ Angle Labor Pain Questionnaire, CSI Central Sensitization Inventory, EPDS Edinburgh Postnatal Depression Scale, EQ-5D-3 L EuroQol Five-Dimensional-Three-Level, FACS Fear-Avoidance Components Scale, PCS Pain Catastrophizing Scale, PSS Perceived Stress Scale, SAPC subacute pain after childbirth, SD standard deviation, STAI State Trait Anxiety Scale, VAS visual analog scale Psychological and pain vulnerabilities of patients Total score Uterine contraction pain Fear/ anxiety Back pain/ long haul Birthing pain The enormity of the pain 126.7 (48.8) 25.5 (10.3) 22.7 (10.8) 29.9 (14.1) 21.5 (13.3) 27.1 (15.3) 139.7 (48.8) 27.7 (9.8) 25.2 (10.5) 33.7 (14.6) 22.7 (14.0) 31.0 (16.2) 0.017 0.043 0.032 0.018 0.431 0.027 Total score Pain Activity Victimization 37.4 (18.1) 13.0 (8.2) 21.0 (9.6) 3.4 (3.1) 41.3 (18.4) 14.1 (8.4) 23.9 (9.9) 3.7 (3.2) 0.055 0.224 0.007 0.389 Total score Helplessness Magnification Rumination 18.5 (11.2) 7.5 (5.3) 4.0 (2.8) 7.0 (4.4) 20.2 (12.1) 8.0 (5.7) 4.5 (2.9) 7.8 (4.4) 0.190 0.415 0.105 0.095 Total score State anxiety Trait anxiety 77.8 (19.1) 39.4 (11.4) 38.4 (9.2) 79.0 (18.7) 39.9 (10.6) 39.2 (9.9) 0.553 0.667 0.451 The values are expressed in mean (SD) A-LPQ Angle Labor Pain Questionnaire, CSI Central Sensitization Inventory, EPDS Edinburgh Postnatal Depression Scale, EQ-5D-3 L EuroQol Five-Dimensional-Three-Level, FACS Fear-Avoidance Components Scale, PCS Pain Catastrophizing Scale, PSS Perceived Stress Scale, SAPC subacute pain after childbirth, SD standard deviation, STAI State Trait Anxiety Scale, VAS visual analog scale Table 3 Obstetrics, analgesic, and neonatal characteristics of patients Variables No SAPC ( n  = 717) Developed SAPC ( n  = 99) P Gravida, mean (SD) 1.9 (1.2) 1.8 (1.2) 0.435 Termination of pregnancy, n (%) 0.510  0  1  2 65 (75.6) 18 (20.9) 3 (3.5) 5 (62.5) 3 (37.5) 0 0.698 0.527 1.000 Analgesia received, n (%) 0.159  Epidural  Entonox  Pethidine  Remifentanil 565 (82.5) 453 (66.1) 114 (16.6) 6 (0.9) 88 (92.6) 64 (67.4) 27 (28.4) 2 (2.1) 0.018 0.902 0.008 0.568 Number of pain relief requests, mean (SD) 1.7 (0.9) 2.0 (1.0) 0.006 Number of pain relief administrations, mean (SD) 1.7 (0.9) 2.0 (1.0) 0.005 Mode of labor onset, n (%) 0.003  Spontaneous  ARM without oxytocic  Oxytocic without ARM  ARM and oxytocic  Spontaneous and oxytocic  Prostin induction  Emergency cesarean  Others 17 (2.4) 14 (2.0) 13 (1.8) 70 (9.8) 5 (0.7) 329 (45.9) 68 (9.5) 289 (40.3) 3 (3.0) 1 (1.0) 2 (2.0) 20 (20.2) 0 25 (25.3) 9 (9.1) 56 (56.6) 0.959 0.798 1.000 0.003 0.884 < 0.001 1.000 0.003 Blood loss during delivery (mL), mean (SD) 268.1 (119.7) 346.6 (268.8) 0.005 Duration of second stage of labor (mins), mean (SD) 64.0 (60.7) 67.9 (54.9) 0.581 Third-degree tear, n (%) 9 (1.3) 5 (5.1) 0.285 Shoulder dystocia, n (%) 6 (0.8) 2 (2.0) 0.157 Mode of delivery, n (%) 0.004  NVD  Instrumental  Cesarean section 481 (67.1) 80 (11.2) 156 (21.8) 50 (50.5) 19 (19.2) 30 (30.3) 0.002 0.033 0.076 Position of head, n (%) 0.019  Occiput anterior  Occiput posterior  Others 541 (75.8) 15 (2.1) 158 (22.1) 63 (63.6) 5 (5.1) 31 (31.3) 0.014 0.153 0.057 Infant height (cm), mean (SD) 49.1 (1.9) 49.7 (1.9) 0.004 Infant weight (kg), mean (SD) 3.2 (0.4) 3.3 (0.4) 0.015 Head circumference, mean (SD) 33.9 (1.3) 33.9 (1.2) 1.000 APGAR at 1 min, mean (SD) 8.9 (0.7) 8.9 (0.3) 1.000 APGAR at 5 min, mean (SD) 9.0 (0.4) 9.0 (0.1) 1.000 Infant resuscitation, n (%) 274 (38.5) 47 (47.5) 0.109 ARM artificial rupture of membranes, NVD normal vaginal delivery, SAPC subacute pain after childbirth, SD standard deviation Obstetrics, analgesic, and neonatal characteristics of patients 0 1 2 65 (75.6) 18 (20.9) 3 (3.5) 5 (62.5) 3 (37.5) 0 0.698 0.527 1.000 Epidural Entonox Pethidine Remifentanil 565 (82.5) 453 (66.1) 114 (16.6) 6 (0.9) 88 (92.6) 64 (67.4) 27 (28.4) 2 (2.1) 0.018 0.902 0.008 0.568 Spontaneous ARM without oxytocic Oxytocic without ARM ARM and oxytocic Spontaneous and oxytocic Prostin induction Emergency cesarean Others 17 (2.4) 14 (2.0) 13 (1.8) 70 (9.8) 5 (0.7) 329 (45.9) 68 (9.5) 289 (40.3) 3 (3.0) 1 (1.0) 2 (2.0) 20 (20.2) 0 25 (25.3) 9 (9.1) 56 (56.6) 0.959 0.798 1.000 0.003 0.884 < 0.001 1.000 0.003 NVD Instrumental Cesarean section 481 (67.1) 80 (11.2) 156 (21.8) 50 (50.5) 19 (19.2) 30 (30.3) 0.002 0.033 0.076 Occiput anterior Occiput posterior Others 541 (75.8) 15 (2.1) 158 (22.1) 63 (63.6) 5 (5.1) 31 (31.3) 0.014 0.153 0.057 ARM artificial rupture of membranes, NVD normal vaginal delivery, SAPC subacute pain after childbirth, SD standard deviation The univariate factors with p  < 0.10 were considered for the development of the multivariable model (Table  4 ). Seven independent association factors for SAPC multivariable model included: Having higher CSI score (adjusted odds ratio (aOR) 1.03, 95%CI 1.01–1.04, p  = 0.006), increased number of pain relief administrations (aOR 1.55, 95%CI 1.23–1.95, p  < 0.001), use of artificial rupture of membrane (ARM) and oxytocic for labor induction (aOR 2.72, 95%CI 1.51–4.91, p  < 0.001), greater volume of blood loss during delivery (every 10-mL increase; aOR 1.02, 95%CI 1.01–1.03, p  = 0.002), having had third-degree tear during delivery (aOR 4.12, 95%CI 1.28–13.27, p  = 0.018) and greater infant weight (aOR 1.15, 95%CI 1.02–1.30, p  = 0.028) were independently associated with greater risk of SAPC. The use of prostin for labor induction was protective against the risk of SAPC (aOR 0.56, 95%CI 0.34–0.93, p  = 0.025) (Table  5 ). The area under the curve (AUC) of the multivariable model was 0.73 (95% CI 0.67–0.78). Table 4 Univariate logistic regression analysis for SAPC Variable Unadjusted OR (95% CI) P Age 0.98 (0.93, 1.03) 0.415 Race (Reference: Chinese)  Indian  Malay  Others 1.90 (1.18, 3.06) 1.50 (0.60, 3.76) 1.65 (0.86, 3.16) 0.008 a 0.386 0.134 Marital status (Single/ divorced/ separated/ others versus married) 0.80 (0.19, 3.50) 0.767 Housing type (private versus public) 0.60 (0.25, 1.44) 0.254 Housing status (renting versus owned) 1.11 (0.54, 2.26) 0.779 Number of biological children 0.82 (0.61, 1.10) 0.192 Age at first menstrual period 0.98 (0.89, 1.08) 0.681 During menstrual period, rarely/ never versus always/ sometimes having:  Mood changes  Cramps  Nausea  Vomiting  Diarrhea  Headache 0.96 (0.59, 1.55) 0.73 (0.46, 1.18) 0.81 (0.48, 1.36) 1.07 (0.55, 2.08) 0.56 (0.36, 0.87) 0.81 (0.53, 1.25) 0.864 0.205 0.417 0.849 0.010 a 0.343 A-LPQ  Total score  Uterine contraction pain  Fear/ anxiety  Back pain/ long haul  Birthing pain  The enormity of the pain 1.006 (1.001, 1.01) 1.02 (1.00, 1.05) 1.02 (1.00, 1.05) 1.02 (1.00, 1.04) 1.01 (0.99, 1.02) 1.02 (1.00, 1.03) 0.017 a 0.053 a 0.031 a 0.014 a 0.405 0.023 a CSI total score 1.03 (1.01, 1.04) 0.002 a EPDS total score 1.04 (0.99, 1.09) 0.153 EQ-5D-3 L VAS 0.98 (0.97, 0.99) < 0.001 a FACS  Total score  Pain  Activity  Victimization 1.01 (1.00, 1.02) 1.02 (0.99, 1.04) 1.03 (1.01, 1.05) 1.03 (0.96, 1.10) 0.046 a 0.240 0.007 a 0.396 PCS  Total score  Helplessness  Magnification  Rumination 1.01 (1.00, 1.03) 1.02 (0.98, 1.06) 1.06 (0.99, 1.15) 1.04 (0.99, 1.09) 0.164 0.419 0.105 0.108 PSS total score 1.01 (0.98, 1.05) 0.450 STAI  Total score  State anxiety  Trait anxiety 1.00 (0.99, 1.01) 1.00 (0.99, 1.02) 1.01 (0.99, 1.03) 0.552 0.683 0.471 Gravida 0.97 (0.81, 1.16) 0.719 Termination of pregnancy 1.38 (0.39, 4.89) 0.618 Analgesia received  Epidural  Entonox  Pethidine  Remifentanil 2.67 (1.21, 5.91) 1.06 (0.67, 1.67) 1.99 (1.22, 3.24) 2.43 (0.48, 12.24) 0.015 a 0.811 0.006 a 0.280 Number of pain relief requests 1.47 (1.18, 1.83) < 0.001 a Number of pain relief administrations 1.47 (1.18, 1.84) < 0.001 a Mode of labor onset: Spontaneous 1.29 (0.37, 4.47) 0.691 Mode of labor onset: ARM without oxytocic 0.51 (0.07, 3.94) 0.521 Mode of labor onset: Oxytocic without ARM 1.12 (0.25, 5.02) 0.886 Mode of labor onset: ARM and oxytocic 2.34 (1.31, 4.05) 0.002 a Mode of labor onset: Prostin induction 0.40 (0.25, 0.64) < 0.001 a Mode of labor onset: Emergency cesarean 0.95 (0.46, 1.98) 0.900 Mode of labor onset: Others 1.93 (1.26, 2.95) 0.002 a Blood loss during delivery 1.03 (1.01, 1.04) b < 0.001 a Duration of second stage of labor 1.001 (0.997, 1.005) 0.602 Third-degree tear 4.18 (1.37, 12.73) 0.012 a Shoulder dystocia 2.44 (0.49, 12.28) 0.278 Mode of delivery (Reference: NVD) 0.005+  Instrumental  Cesarean section 2.29 (1.28, 4.08) 1.85 (1.14, 3.01) 0.005 a 0.013 a Position of head (Reference: Occiput anterior) 0.022+  Occiput posterior  Others 2.86 (1.01, 8.14) 1.69 (1.06, 2.68) 0.049 a 0.028 a Infant height 1.17 (1.05, 1.31) 0.005 a Infant weight 1.001 (1.00, 1.001) 0.020 a Head circumference 1.02 (0.86, 1.21) 0.788 APGAR at 1 min 1.16 (0.74, 1.83) 0.516 APGAR at 5 min 1.36 (0.49, 3.76) 0.551 Infant resuscitation 1.45 (0.95, 2.20) 0.088 a A-LPQ Angle Labor Pain Questionnaire, ARM artificial rupture of membranes, CI confidence interval, CSI Central Sensitization Inventory, EPDS Edinburgh Postnatal Depression Scale, EQ-5D-3 L EuroQol Five-Dimensional-Three-Level, FACS Fear-Avoidance Components Scale, NVD normal vaginal delivery, OR odds ratio, PCS Pain Catastrophizing Scale, PSS Perceived Stress Scale, SAPC subacute pain after childbirth, SD standard deviation, STAI State Trait Anxiety Scale, VAS visual analog scale a Variables with p  < 0.10 in the univariate regression analysis were considered for the subsequent multivariable logistic regression analysis b Blood loss during delivery is reported per 10‑mL increase + refers to type 3 p – value Univariate logistic regression analysis for SAPC Indian Malay Others 1.90 (1.18, 3.06) 1.50 (0.60, 3.76) 1.65 (0.86, 3.16) 0.008 a 0.386 0.134 Mood changes Cramps Nausea Vomiting Diarrhea Headache 0.96 (0.59, 1.55) 0.73 (0.46, 1.18) 0.81 (0.48, 1.36) 1.07 (0.55, 2.08) 0.56 (0.36, 0.87) 0.81 (0.53, 1.25) 0.864 0.205 0.417 0.849 0.010 a 0.343 Total score Uterine contraction pain Fear/ anxiety Back pain/ long haul Birthing pain The enormity of the pain 1.006 (1.001, 1.01) 1.02 (1.00, 1.05) 1.02 (1.00, 1.05) 1.02 (1.00, 1.04) 1.01 (0.99, 1.02) 1.02 (1.00, 1.03) 0.017 a 0.053 a 0.031 a 0.014 a 0.405 0.023 a Total score Pain Activity Victimization 1.01 (1.00, 1.02) 1.02 (0.99, 1.04) 1.03 (1.01, 1.05) 1.03 (0.96, 1.10) 0.046 a 0.240 0.007 a 0.396 Total score Helplessness Magnification Rumination 1.01 (1.00, 1.03) 1.02 (0.98, 1.06) 1.06 (0.99, 1.15) 1.04 (0.99, 1.09) 0.164 0.419 0.105 0.108 Total score State anxiety Trait anxiety 1.00 (0.99, 1.01) 1.00 (0.99, 1.02) 1.01 (0.99, 1.03) 0.552 0.683 0.471 Epidural Entonox Pethidine Remifentanil 2.67 (1.21, 5.91) 1.06 (0.67, 1.67) 1.99 (1.22, 3.24) 2.43 (0.48, 12.24) 0.015 a 0.811 0.006 a 0.280 Instrumental Cesarean section 2.29 (1.28, 4.08) 1.85 (1.14, 3.01) 0.005 a 0.013 a Occiput posterior Others 2.86 (1.01, 8.14) 1.69 (1.06, 2.68) 0.049 a 0.028 a A-LPQ Angle Labor Pain Questionnaire, ARM artificial rupture of membranes, CI confidence interval, CSI Central Sensitization Inventory, EPDS Edinburgh Postnatal Depression Scale, EQ-5D-3 L EuroQol Five-Dimensional-Three-Level, FACS Fear-Avoidance Components Scale, NVD normal vaginal delivery, OR odds ratio, PCS Pain Catastrophizing Scale, PSS Perceived Stress Scale, SAPC subacute pain after childbirth, SD standard deviation, STAI State Trait Anxiety Scale, VAS visual analog scale a Variables with p  < 0.10 in the univariate regression analysis were considered for the subsequent multivariable logistic regression analysis b Blood loss during delivery is reported per 10‑mL increase + refers to type 3 p – value Table 5 Multivariable logistic regression model on associated factors with SAPC Variable Adjusted OR (95% CI) P CSI total score 1.03 (1.01, 1.04) 0.006 Number of pain relief administrations 1.55 (1.23, 1.95) < 0.001 Mode of labor onset: ARM and oxytocic 2.72 (1.51, 4.91) < 0.001 Mode of labor onset: Prostin induction 0.56 (0.34, 0.93) 0.025 Blood loss during delivery 1.02 (1.01, 1.03) a 0.002 Third-degree tear 4.12 (1.28, 13.27) 0.018 Infant weight 1.15 (1.02, 1.30) 0.028 ARM artificial rupture of membranes, CI confidence interval, CSI Central Sensitization Inventory, OR odds ratio a Blood loss during delivery is reported per 10‑mL increase Multivariable logistic regression model on associated factors with SAPC ARM artificial rupture of membranes, CI confidence interval, CSI Central Sensitization Inventory, OR odds ratio a Blood loss during delivery is reported per 10‑mL increase

In this study, we identified having greater central sensitization was an independent risk factor of SAPC, along with other analgesic and obstetric factors that suggest the potential involvement of inadequate management on labor pain and physical trauma. Given the model’s moderate predictive performance in our findings, future studies should aim to refine and externally validate predictive models incorporating these factors, with the goal of improving their discriminative performance and clinical utility in identifying patients at increased risk of SAPC. These efforts may then contribute to the formulation of personalized care plans that include monitoring and early interventions to enhance postpartum pain management.

This study represents a secondary analysis of observational data and provides important insights into factors associated with the development of SAPC, offering a foundation for future hypothesis‑driven and prospective investigations by associating central sensitization, labor pain experiences, and physical trauma with SAPC. Specifically, we showed that greater pre-delivery central sensitization, increased number of pain relief administrations, use of ARM and oxytocic for labor induction, greater volume of blood loss during delivery, having had third-degree tear and greater infant weight were independently associated with increased risk of SAPC. Conversely, the use of prostin for labor induction was protective against the development of SAPC. Our primary exposure of pre-delivery central sensitization emerged as an independent predictor of SAPC in our multivariable logistic regression model, which is in line with previous studies emphasizing the role of central sensitization in chronic pain syndromes such as fibromyalgia and complex regional pain syndrome [ 6 , 11 ]. These findings suggest that altered central pain processing may contribute to prolonged postpartum pain even in the absence of ongoing tissue damage. Antenatal screening using validated tools such as the CSI may therefore help identify patients at increased risk of SAPC and inform future targeted strategies aimed at modulating central sensitization to support postpartum pain management [ 26 ]. Apart from central sensitization, other pain and psychological vulnerabilities including labor pain, fear-avoidance, and lower status on general health were associated with SAPC in univariate analysis but were not retained in the multivariable model, possibly due to residual confounding or interactions not captured in this study. Previous studies nonetheless support the relevance of these factors in postpartum pain trajectories. For instance, greater labor pain has been linked to increased acute postpartum pain, while fear-avoidance during pregnancy has been associated with higher genito-pelvic pain intensity at three months postpartum [ 27 , 28 ]. Although indicators related to physical trauma, such as uterine contraction pain, back pain, and fear‑avoidance of physical activity during labor, were also significant in univariable analyses, these experiences may be transient and resolve within the early postpartum period, limiting their contribution to the development of SAPC [ 29 ]. Additionally, elevated EPDS scores have been more consistently associated with poorer perceived health status and subsequent mental health outcomes such as postpartum depression, rather than SAPC [ 30 , 31 ]. The significant association between a higher number of pain relief administrations and SAPC aligns with the findings from Maeda et al., who reported that women who received neuraxial labor analgesia experienced greater pain and higher analgesic requirements compared with those with unmedicated childbirth [ 32 ]. Given that inadequately controlled labor pain has been implicated in the development of persistent postpartum pain [ 33 ], our findings suggest that such inadequate pain control may be associated with longer-term postpartum pain outcomes, potentially mediated by nociceptor sensitization and inflammatory processes leading to secondary hyperalgesia [ 3 ]. The independent association between obstetric factors (greater volume of blood loss during delivery, third degree tear, and greater infant weight) and SAPC highlights the role of physical trauma and its downstream effects on pain perception and development. Traumatic birth events, including third-degree perineal tears, are well-established risk factors for acute and chronic postpartum pain and may contribute to increased blood loss, poorer recovery, and adverse pain outcomes [ 27 , 34 , 35 ]. Greater infant weight is also associated with more complicated deliveries [ 36 , 37 ], however perineal trauma is multifactorial with increased risk mediated through delivery complications such as shoulder dystocia or operative vaginal delivery. This underscores the potential importance of identifying individuals at increased risk of obstetric complications to support timely postpartum pain management. The protective effect of Prostin E 2 (Dinoprostone) for labor induction is a noteworthy and novel finding in the context of SAPC. Prostaglandin E 2 (PGE 2 ), including Dinoprostone, have been used to promote cervical ripening prior to induction of labor (IOL) and stimulations of uterine contractions [ 38 ]. Oxytocin, on the other hand, is widely used for IOL and augmentation; however, it does not have a significant role in cervical ripening and is typically administered following prostaglandin-mediated cervical ripening, often in conjunction with amniotomy [ 39 ]. Socha et al. previously reported that patients who received Dinoprostone required more oxytocin administration for IOL and had a longer time to delivery compared with those who received Misoprostol, a prostaglandin E1 (PGE 1 ) analog [ 39 ]. Interestingly, those who had Misoprostol required more labor analgesia than those receiving Dinoprostone, despite similar rates of uterine hyperstimulation (tachysystole). Similarly, Kunt et al. previously showed that the time from IOL to active labor onset and delivery was significantly shorter with oxytocin compared with PGE 2 despite comparable maternal and neonatal outcomes [ 40 ]. As Dinoprostone is the standard cervical ripening agent in our institution, comparative analyses with Misoprostol were not feasible; and differences in induction protocols that may influence labor trajectory and pain experience were beyond the scope of this study. On the other hand, artificial rupture of membrane (ARM) may shorten labor duration, although it is unclear whether this accelerated course contributes to intensified labor pain due to increased uterine compression and physical stress over a shorter timeframe [ 41 ]. The above suggests that PGE 2 may be associated with a more gradual labor progression, potentially offering advantages for pain experience and maternal outcomes. Further investigation is needed to clarify the mechanisms underlying these differential effects in postpartum pain. The results of this study give rise to several important clinical implications. The independent association between pre delivery central sensitization and subsequent SAPC suggests that routine antenatal screening using a validated instrument, such as the CSI, may facilitate earlier identification of at-risk individuals [ 10 ]. This in turn could enable more targeted postpartum surveillance, including timely follow up appointments and proactive pain management strategies [ 1 , 3 ]. The observed association between a higher number of pain relief administrations and SAPC may reflect inadequately controlled labor pain in these patients. This finding underscores the importance of optimizing intrapartum analgesia, not only to improve immediate comfort but also to promote better long-term pain outcomes. Furthermore, the divergent associations between induction methods (ARM with oxytocics versus Prostin) highlight the need for further evaluation of how specific labor induction protocols may influence postpartum pain trajectories. Collectively, these findings support a more individualized approach to perinatal care, integrating pre delivery psychological vulnerability, intrapartum analgesic management, and obstetric factors to enable risk stratification and guide timely postpartum interventions. There are several limitations in this study. Firstly, this study was carried out in a maternity facility that primarily treats an English-speaking Asian population. Kim et al. previously demonstrated in a systematic review that Asians may exhibit different pain perception by having a lower pain threshold compared to other ethnics such as non-Hispanic whites [ 42 ]. As a result, the generalizability of these findings to broader populations may be limited. The predominant population recruited in this study was of Chinese ethnicity, and this cultural bias may further influence the pain perception and management practices; thereby limiting the results’ applicability to other demographic groups [ 43 ]. Similarly, the study findings may not be fully applicable to populations in low-resource settings or with limited access to obstetric care including anesthesia services and pain management. Secondly, although multiple pain-related and psychological vulnerability measures were assessed, only pre-delivery central sensitization and number of pain relief administrations was retained as independent factors in the final multivariable SAPC model following stepwise selection. Other pain and psychological variables were not selected, likely reflecting overlap among constructs and potential multicollinearity that limited their independent contributions once modeled jointly. It is important to note that this does not imply that their potential clinical relevance is diminished, since unmeasured biological or physiological factors (e.g., hormonal changes, genetic factors, or inflammatory processes) may also play an important role in the development of SAPC but were not assessed in this study. Taken together, central sensitization may represent a broader, integrative marker of pain‑related psychological vulnerability within this cohort. Thirdly, given the evaluation of multiple psychological, obstetric, and neonatal variables, the possibility of type I error due to multiple comparisons should be considered, and statistically significant findings should be interpreted with appropriate caution [ 44 ]. In addition, the analytic strategy of univariable screening followed by multivariable regression modelling represents a conventional and widely used approach in observational clinical research; yet the use of stepwise variable selection may increase the risk of model instability and overfitting, potentially limiting the generalizability of the findings to other populations or clinical settings [ 45 , 46 ]. Similarly, overfitting cannot be excluded due to the number of candidate variables relative to the number of SAPC events, thus future studies with external validation are warranted to confirm the robustness and generalizability of our findings. As this was a secondary analysis of observational data, residual confounding cannot be excluded, and causal relationships between exposures and SAPC cannot be inferred. In particular, increased analgesic use may reflect greater underlying pain severity rather than a causal effect of analgesia on SAPC [ 47 ]. Lastly, we relied on self-reported measures that were assessed at different time points throughout the first stage of labor, which might potentially introduce variations in the physical and emotional states. As SAPC was also assessed using self-reported measures, the findings may be subject to recall bias. Moreover, the study setting may have triggered a Hawthorne effect, where patients altered their questionnaire responses due to awareness of being observed during the study period [ 48 ]. Nonetheless, the integration of psychological, obstetric, and analgesic factors within our study framework provides clinically meaningful insights into the risk of SAPC.

Postpartum pain resulting from the labor and delivery process has multifactorial etiology, including tissue trauma, surgical interventions, and inflammatory processes [ 1 ]. It may be categorized as acute, sub-acute, and chronic pain based on its duration; and is associated with adverse maternal outcomes including postpartum depression and chronic pain [ 2 ]. Current literature has largely concentrated on acute pain immediately following childbirth or chronic pain that manifests later, leaving the sub-acute phase relatively understudied [ 2 , 3 ]. Sub-acute pain after childbirth (SAPC) refers to postpartum pain lasting for ≥4 weeks and is a common yet frequently overlooked health issue that impacts a significant number of postpartum women [ 4 ]. With an incidence of approximately 8% in Singapore, SAPC has the potential to progress into chronic pain that further impairs daily functions [ 4 ]. Recent evidence demonstrates that psychological (depression, anxiety, stress), obstetric (blood loss, fetal position), and analgesic factors (inadequate labor analgesia) are associated with the development of SAPC [ 2 , 4 , 5 ]. Central sensitization is characterized by amplified response of the central nervous system to non-noxious stimuli and has been extensively studied in chronic pain conditions such as fibromyalgia [ 6 ]. Previous study in parturients having undergone cesarean delivery demonstrated a positive association between central sensitization and high risk of SAPC [ 7 ], yet its role in laboring women remains poorly understood. In addition, no comprehensive model currently integrates pre-delivery psychological and other patient factors to predict the risk of SAPC. This represents a significant unmet clinical need whereby early identification with preventive intervention is required to mitigate the postpartum acute-to-chronic pain transition. For the primary outcome, we investigated the association between central sensitization and the risk of SAPC at 6 to 10 weeks postpartum. We also evaluated other pain and psychological vulnerabilities, obstetric factors, analgesic choices, and their association with SAPC.