Determinants of satisfaction with acute pain therapy during and after childbirth.

Background: An important task in obstetrics is to offer patients adequate pain management after vaginal delivery (VD) and after cesarean section (CS). The aim of the present analysis was to assess pain intensity at the day after childbirth, after 3 and 6 months in women after VD and after CS. Additionally, it was planned to identify determinants of maternal satisfaction with acute pain therapy following VD and CS. Methods: This secondary analysis of a single-center prospective cohort study included 210 parturients. The women gave birth in the University Hospital of Wuerzburg. They completed a survey before childbirth (P), on the first day (D1), 3 and 6 months after VD and CS. The survey included questions about the expected pain, the experienced pain, the birth, the pain afterwards and also psychological questionnaires. In addition, information was collected from the patient records. Women with planned CS received spinal anesthesia. A secondary CS was performed with an epidural, spinal or general anesthesia. Pain therapy on the ward was standardized. The primary outcomes were acute and chronic pain intensity at rest, during movement, determinants of satisfaction with pain therapy and the wish to have received more pain medication during and after VD and CS (D1). Results: A total of 175 women completed the survey at all time points and were available for the analysis (drop-out 16.8%). The results confirmed high pain levels at D1 after CS (median pain intensity during movement: 8 NRS points, interquartile range, IQR 6–9 points). After VD pain scores were much lower (median pain intensity during movement: 4 points, IQR 2–5 points). Of the mothers 7.4% reported pain at rest at all time points after VD or CS but pain intensity was low after 3 and 6 months (median pain intensity at rest: 2 points, IQR 0–3 points), 28% received extended-release opioids after CS and 33% of women (VD and CS) had an epidural. The most important influencing factor for lower satisfaction with pain therapy after CS or VD was inadequate pain relief. Women with VD who had an epidural, were more satisfied during delivery than women without but there was no difference at D1. Discussion: This study confirmed high acute pain levels following CS and an incidence of chronic pain of around 7% but pain intensity was low and one third received strong opioids after CS. Around 11% of women after primary CS (8% epidural, 3% combined spinal/epidural) and around 55% of women after secondary CS had an epidural, which could be used for pain therapy after birth. Women without an epidural or without opioids may not have had good pain management. The received pain relief appeared to be the most relevant predictor for satisfaction and the wish to have received more pain medication. Mothers having a VD with an epidural catheter were more satisfied with pain therapy during birth than those without. Therefore, a better multimodal pain management (including opioids, epidural and nonopioids) might provide better pain relief and might improve overall satisfaction with pain treatment following CS. Finally, the received pain relief might be a better patient-related outcome measure for satisfaction with pain treatment after childbirth than pain intensity alone.

Zusammenfassung: Hintergrund: Eine zentrale Aufgabe in der Geburtshilfe ist es, Patientinnen nach Spontanpartus (SP) und insbesondere nach Kaiserschnitt (KS) eine adäquate peripartale Schmerztherapie anzubieten. Ziel der vorliegenden Studie war es, Schmerzintensitäten nach einer Geburt zu erfassen sowie Einflussfaktoren auf die Zufriedenheit mit der Schmerztherapie während und nach der Geburt zu identifizieren. Methodik: Die vorliegende Sekundäranalyse einer prospektiven Längsschnittstudie umfasste 210 Patientinnen. Diese füllten pränatal, einen Tag, 3 sowie 6 Monate nach Geburt einen Fragebogen aus. Als primäre Endpunkte fungierten die Schmerzintensitäten in Ruhe und bei Belastung (Median [Interquartile Range, IQR]), die Zufriedenheit mit der Schmerztherapie sowie der Wunsch nach mehr Schmerztherapie während der Geburt und am ersten postpartalen Tag. Ergebnisse: Einen Tag nach einem KS zeigten sich sehr hohe Schmerzintensitäten (mediane Schmerzintensität bei Belastung: 8 NRS-Punkte [6–9]). 28 % der Frauen erhielten nach einem KS retardierte Opioide. 7,4 % aller Mütter gaben Schmerzen zu allen Zeitpunkten an, wobei die Schmerzintensität nach 3 bzw. 6 Monaten niedrig war (mediane Intensität in Ruhe: 2 NRS-Punkte [0–3]) Als wichtigster Einflussfaktor für die Zufriedenheit mit der Schmerztherapie konnte in der Kohorte eine stärkere Schmerzlinderung identifiziert werden. Diskussion: In der vorliegenden Studie bestätigten sich hohe Schmerzintensitäten nach KS im Gegensatz zu niedrigeren Schmerzintensitäten nach SP sowie eine Inzidenz von 7 % mit chronischen postoperativen Schmerzen – aber von geringer Intensität. Eine ungenügende Schmerzlinderung war der wichtigste Einflussfaktor für eine geringere Zufriedenheit und einen stärkeren Wunsch nach mehr Schmerztherapie. Eine suffiziente multimodale Schmerzbehandlung (retardierte Opioide, Regionalanästhesieverfahren, nichtmedikamentöse Verfahren) mit Fokus auf eine adäquate Schmerzlinderung könnte einen Beitrag zu Verbesserung der Zufriedenheit leisten.

Keywords: Satisfaction; Postoperative pain; Delivery; Epidural analgesia; Obstetrics; Zufriedenheit; Postoperativer Schmerz; Geburt; Epiduralanalgesie; Geburtshilfe

Supplementary Information The online version of this article (https://doi.org/10.1007/s00101-023-01260-w) contains additional tables.

Acute pain treatment following childbirth remains a relevant healthcare problem. A large database analysis revealed that cesarean section (CS) was in the top five of the most painful surgeries [[6]]. An insufficient pain treatment [[12]] and a higher risk to suffer from severe acute pain [[14]] might be possible explanations. Furthermore, insufficiently treated acute pain following childbirth might become chronic [[8], [21]] and postdischarge opioid use after CS might be higher than expected [[13]].

Although discussions about handling pain treatment after VD and CS have existed for a long time, there are still current attempts to define the most relevant patient-related outcomes (PROs) and how to measure them (PROMs) [[16]]. Pain intensity might not be the most important PRO [[11]]. Currently, there is no definition of a core outcome set (COS) for acute pain following childbirth, but satisfaction with pain treatment might be a relevant PRO [[9]]. Within an international acute pain registry (PAIN OUT), Schwenkglenks et al. found out that satisfaction with pain treatment is strongly correlated with no desire to have received more pain treatment and the experience of an initial higher pain relief [[19]]. Additionally, a single-center trial examined predictors for satisfaction with pain treatment following thoracoscopic surgery and found interesting results [[20]]; participation concerning treatment decisions, information, pain intensity at the time of survey, lowest pain score in the last 24 h and pain-related sleeping interference were significant predictors for satisfaction; however, specific prenatal and postnatal pain, as well as psychological data with a focus on satisfaction with pain therapy and the wish to have received more pain treatment in women following childbirth are currently lacking.

The aims of this study were first to identify determinants of maternal satisfaction and the wish to have received more pain treatment during as well as the first day after VD or CS and second to investigate acute and chronic pain outcomes.

The present study is a secondary analysis of a prospective cohort trial investigating predictors for acute and chronic pain and depressive symptoms after childbirth with a special focus on maternal attachment styles [[18]].

Study approval for the prospective cohort trial was given by the institutional review board of the University Hospital of Wuerzburg (registration number: 144/15) from 15 October 2015. The study was registered in the German Clinical Trials Registry (DRKS00010433). All included women gave a written informed consent for participation in the trial.

All pregnant women visiting a special consultation for delivery preparation in the outpatient clinic of the department of obstetrics in the University Hospital Wuerzburg were asked to participate.

The following inclusion criteria were used: age ≥ 18 years, able to speak and understand the German language, planned delivery at the University Hospital Wuerzburg, delivery in or after the 34th week of gestation.

The cohort study included surveys at four different time points: prenatal in the last trimester (pregnant/P), first day (D1), 3 (3M) and 6 months (6M) after VD or CS. The survey included different psychological questionnaires, the Edinburgh Postnatal Depression Scale (EPDS, P, D1, 3M, 6M), the Adult Attachment Scale (AAS, P), the Experiences in Close Relationships (ECR-RD8, P), the Pain Catastrophizing Scale (PCS, D1) and the Anxiety Sensitivity Index (ASI‑3, P), a self-designed postoperative pain outcome form (mean pain intensity at rest, NRS from 0 no pain to 10 maximum pain and during movement, NRS from 0 no pain to 10 maximum pain), mean proportion of time suffering from severe pain within the last 24 h (NRS from 0% never in severe pain to 100% always in severe pain), satisfaction with pain treatment (NRS from 0 very dissatisfied to 10 maximum satisfaction), the degree of received pain relief (NRS from 0 no pain relief to 10 maximum pain relief), wish to have received more pain medication (yes/no), use of pain medication (yes/no, D1, 3M, 6M) and a questionnaire focusing on pregnancy and delivery-related issues according to the time point. Generally, only women receiving pain medication during or after delivery were asked about satisfaction with pain treatment, their wish to have received more analgesia and the degree of received pain relief. The first evaluation of data showed only a small proportion of patients who selected low scores on the satisfaction scale. Therefore, the cohort was divided into four groups (group 1: 0–4 points, group 2: 5–6 points, group 3: 7–8 points, group 4: 9–10 points) [[20]]. Additionally, on the first day after VD or CS data concerning the delivery were collected (e.g., mode of delivery, use of epidural or other analgesic techniques during VD or CS and postoperative pain medication). Pain therapy during and after VD and CS was standardized. Women undergoing a planned CS received a spinal regional analgesia up to 12 ml isobaric 0.1% bupivacaine (1 mg/ml) with sufentanil (1 µg/ml) as already described in another publication [[7]]. If a secondary CS was performed, women with an epidural catheter received either an additional 15–20 ml ropivacaine 0.75% or were treated with spinal regional anesthesia or general anesthesia. Women undergoing a VD usually received a weak opioid first (meptazinol), had the possibility to use nitrous oxide (rarely used) and/or were offered an epidural catheter as patient-controlled analgesia (ropivacaine 0.13%, sufentanil 0.5 µg/ml; continuous rate: 5 ml/h; bolus: 5 ml; lockout interval: 15 min). After a VD all women received a nonopioid analgesic (e.g., diclofenac or ibuprofen plus paracetamol) on demand, whereas all women after CS received a fixed combination of paracetamol combined with ibuprofen or diclofenac. If pain intensities at rest assessed by these women were above 3 points (measured on a NRS), 10 mg extended-release oxycodone could be administered up to twice daily. Postnatal pain management after discharge was not standardized.

Primary endpoints were pain at rest and during movement, satisfaction with pain therapy and the wish to have received more pain medication during childbirth and on the first day after VD or CS. Secondary endpoints were effectiveness of pain control during and after VD or CS using several relevant PROs.

As described in the primary analysis of these data the sample size calculation was based on Döring [[3]]. The response rate of women filling out all questionnaires at all time points was estimated to be 36%. To obtain 64 complete datasets 200 women were included and 10 more participants were included to fill out an extra form in order to test the feasibility of the questionnaires.

Data were analyzed separately for women having a VD or CS. Demographic data (e.g., age) were calculated as mean (± standard deviation, SD). Outcome data rated on scales (e.g., NRS pain scales) were assessed as median (interquartile range, IQR). Within the group of women having a VD or a CS the results were compared within the four groups with different satisfaction levels (see above) [[20]] and different wishes to have received more pain medication (yes/no) with the nonparametric Kruskal-Wallis test or Wilcoxon-Mann-Whitney test. Statistical significance was determined using α ≤ 0.05. Categorial variables (e.g., number of patients treated with opioids, number of patients having an epidural) were analyzed with the χ2-test or Fisher's exact test. All statistical analyses were carried out with SPSS 22.0 (IBM Corp., Armonk, NY, USA).

A total of 222 pregnant women were screened for inclusion and asked for their consent. We finally included 210 women, whereas 1 parturient withdrew consent after inclusion and the data were deleted from the database. The overall drop-out rate for the whole study period was 16.8% and data of 175 women were finally available for the analysis for all time points (Fig. 1).

Graph: Fig. 1Flow diagram of included and excluded patients. T1 before delivery, T2 first day, T3 3 months and T4 6 months after delivery

The prenatal demographic data are based on 209 included women and summarized in Table 1. The mean (±SD) age was 31.7 (±4.4) years, 7.7% suffered from prenatal chronic pain and 23.6% reported other chronic diseases (such as thyroid diseases 10%, allergies 3.3%). Additionally, 16.7% complained about psychological problems during the current pregnancy, whereas only 2.9% received psychotherapy and/or medications.

Table 1 Demographic data of all included women

The postnatal demographic data are based on 199 included participants at D1. The mean week of pregnancy for the delivery was 40.1 (±1.4), 135 parturients (68.2%) had a VD and 63 (31.8%) had primary or secondary CS. No emergency CS was performed and 39 deliveries were medically initiated. Within the group of VD 37.8% received an epidural catheter, whereas 45.2% were treated with a weak intravenous opioid. Of the parturients 5 needed general anesthesia for postnatal curettage due to retained placenta, 36 parturients received regional anesthesia within the group of mothers having a primary CS, while 1 parturient was treated with general anesthesia. One child died after delivery due to several medical problems and the data of the mother were therefore excluded.

Women with VD had much higher maximum pain scores during delivery (10 NRS points, IQR 9–10 points) than women with CS (5 NRS points (1–9)). On the first day women with CS suffered from more pain at rest (4 NRS points (3–7)) and during movement (8 NRS points (6–9)) than women with VD (pain at rest: NRS 2 (1–3); pain during movement: 4 NRS points (2–5)) (Fig. 2). After 3 and 6 months following VD or CS around 18% (3M) and 15% (6M) of women suffered from pain at rest, respectively, but only 7.4% of women had pain at all 3 time points. The average level of pain at rest was low (after 3 months: 2 NRS points (0–3); after 6 months: 1 NRS point (0–4)) (supplemental tables 1 and 2).

Graph: Fig. 2Postoperative pain intensities (NRS) at rest and during movement in women after vaginal delivery or cesarean section. D1 first day after delivery

Results are presented in the supplemental tables 3–6. Analyzing the most important factors which influence satisfaction with pain medication, the degree of pain relief was significantly lower in women with lower satisfaction levels during CS (p < 0.001) and 1 day after (p = 0.01). Additionally, pain at rest (p = 0.023) as well as helplessness (p = 0.01) and rumination (p = 0.049) on the PCS were significantly higher in women with lower satisfaction levels at D1.

The wish to have received more pain medication was also significantly different in women with different degrees of pain relief during childbirth (p = 0.005) and during the first day after CS (p < 0.001). Women reporting high pain intensity during CS (p = 0.025), high pain scores after giving birth (pain at rest and during physical activity p < 0.001) and a high proportion of time with severe pain at D1 (p < 0.001), had a higher wish for more pain medication. The number of women treated with opioids was not significantly different in the groups with different satisfaction levels (p = 0.938) or in women with the wish to have received more pain medication (p = 0.767).

In this cohort women who experienced more pain relief were more satisfied at both time points (during VD, p < 0.001, 1 day after VD, p = 0.019) (supplemental tables 7–10). Additionally, the proportion of time with worst pain during VD was significantly longer in patients with lower satisfaction levels during birth (p = 0.026). In women wishing to have received more pain medication the median degree of pain relief during birth (p = 0.018) and on the first day after (p = 0.034) was lower and worst pain was higher during VD (p = 0.007). At 1 day after VD women wishing to have received more pain medication reported a higher pain expectation before (p = 0.01), reported a higher median pain during movement (p = 0.002), showed a lower pain relief (p = 0.034) and a higher amount of time spent with worst pain within the last 24 h (p = 0.047). Out of 59 women who consumed pain medication, 10.2% requested more pain treatment (p = 0.007). Women who had no medication after VD had no wish to receive them. Furthermore, we investigated the influence of epidural anesthesia in women having a VD on the primary endpoints and 82% of women with an epidural catheter rated the satisfaction during VD in the upper half. Only 30.2% did so without an epidural (p < 0.001) but there were no differences at D1.

This survey is a secondary analysis of a prospective longitudinal cohort study and included 210 parturients. It investigated acute and chronic postoperative pain intensity and possible factors which might influence poor maternal satisfaction levels with pain management and the wish to have received more pain medication. Data were analyzed during VD or CS, on the first day, 3 and 6 months after childbirth. Women reported much higher pain scores at rest and during movement after a CS than a VD. Lower received pain relief was reported in women with lower satisfaction levels and with a wish to have received more pain medication during childbirth and at the first day after.

Within the last years several observational trials investigated the prevalence of acute and chronic pain after childbirth [[5], [8], [21]]; however, the results were not conclusive. The present data confirmed high pain intensities on the first day after CS and lower pain intensities after VD. Additionally, it was confirmed that 7% of all parturients suffered from chronic pain after VD or CS [[21]]. A new finding is that the reported intensities and other relevant pain outcomes were low and therefore these outcomes might be not clinically relevant; however, these results might be influenced by the fact that only 28% of women received extended-release opioids after CS. Furthermore, 55% of all women with secondary CS had an epidural catheter, which could be used for pain therapy after surgery; in contrast only 11% of all women with primary CS (8% epidural, 3% combined spinal and epidural) had an epidural catheter for pain treatment. Therefore, pain management after CS probably was not adequate in these women, either because women were reluctant to use effective analgesia or because the barrier of the physicians to give additional potent opioids was too high. The results are confirmed by another observational study by Marcus and Gerbershagen et al. concerning low opioid prescription after CS. In their survey 63% of women did not receive an opioid and also showed high pain intensities after CS [[12]]. Therefore, it might be speculated that a more liberal opioid prescription or the increased use of catheter-based regional analgesia and the use of neuraxial hydrophilic opioids might have changed the results and might improve pain treatment following CS. The data clearly confirmed that women using an epidural during birth showed high satisfaction scores with pain treatment during VD and 82% were satisfied. In contrast, only 30.2% of women giving birth without an epidural reported higher satisfactions levels. Adequate relief of pain could be the key factor for this outcome [[2]].

More specifically, other data showed that opioids given via a patient-controlled analgesia pump or a self-administered medication scheme might be even more effective in women after VD or CS, because these women less often reported the wish to receive more analgesics and were more satisfied compared to the group with a fixed scheme [[4]].

Apart from pain intensities, satisfaction with pain treatment and more specifically the wish to have received more pain medication might better reflect the quality of the pain management. This analysis revealed that women receiving CS and reporting low satisfaction levels or expressed a wish to have received more pain medication reported low degrees of pain relief during birth and on the first day after. The same holds true for women undergoing a VD. This result is confirmed by Schwenkglenks et al., who described the same result in patients undergoing surgery [[19]]. They reported that relief of pain even in patients reporting high pain intensities improved satisfaction. Clivatti et al. described comparable results in their survey of parturients treated with epidural analgesia [[1]]. Despite an epidural catheter 45% of analyzed women still suffered from severe pain; however, 92% were satisfied with pain treatment and 91% would choose the same method again. This might be explained by the fact that mothers expect pain after childbirth and are prepared for the pain. In contrast, if pain relief is insufficient, this might not be expected and therefore might be responsible for lower satisfaction rates [[19]] as also seen in our analysis. This highlights that pain relief received during and after birth might be a better PRO than pain intensity, which should be evaluated in focus groups in the future. Interestingly, several studies showed that the amount of pain relief and the possible different degree of pain thresholds during or after birth might be predicted, e.g., by prenatal quantitative sensory [[17]] testing or psychological tests (e.g., pain catastrophizing scale, pain expectation) [[15]]. In this survey women undergoing CS, who were more afraid of giving birth, more often requested pain medication during birth than women who were less afraid (p = 0.046). One explanation could be a lower pain threshold as discussed above. Additionally, it was shown that higher helplessness and rumination scores were observed in women undergoing CS reporting lower satisfaction levels. The data clearly showed that this is relevant, because it might be related to lower satisfaction levels and a higher wish to have received more pain medication. Therefore, it may be helpful if these psychological factors could be routinely assessed before childbirth. Based on these findings, an individual plan including sufficient pharmacological and nonpharmacological (e.g., psychological interventions, distraction methods) pain treatment options could be offered in order to improve overall satisfaction with pain treatment.

Apart from that, we recently showed that significantly more women reporting higher EPDS scores in the prenatal screening had a primary and secondary CS compared to those having a VD [[18]]; therefore, prenatal screening might identify women at risk for having an unplanned secondary CS, which is possibly associated with higher pain intensity and lower satisfaction level as shown above.

There are several possible limitations of this study. First, the results might be influenced by selection bias because we only performed a single-center cohort study. Second, we did not evaluate side effects and the dose of pain medication, because this was not the primary aim of the initial study. Satisfaction in this analysis was not influenced by consumption of opioids but it might be speculated that side effects or fear of side effects for mother and the neonate might have influenced this result. Third, we cannot exclude a possible recall bias regarding data during birth, but the period between VD or CS and survey at D1 was kept as short as possible (less than 24 h). Finally, this was a longitudinal prospective study and not a randomized controlled trial with blinding, which might have influenced our findings regarding epidurals and analgesic drugs.

This study showed high acute pain levels in women following CS and an insufficient analgesia after CS. The prevalence of chronic pain after VD or CS in all parturients was approximately 7% but pain intensity for chronic pain was low and may not be clinically relevant. Parturients who reported low satisfaction levels with pain treatment or wished to have received more pain medication during and after CS or VD expressed poor perceived pain relief. Therefore, this PRO might better reflect satisfaction with pain treatment after birth and should routinely be assessed. Finally, a better multimodal pain management (more liberal use of opioids for a fixed time and fixed application of sufficient doses of nonopioids) and maybe further nonpharmacological (e.g., psychological interventions, distraction methods) support might provide better pain relief and might improve overall satisfaction with pain treatment following CS.

K. Schnabel, A.-M. Drusenbaum, P. Kranke, P. Meybohm, A. Wöckel and A. Schnabel confirm that there is no conflict of interest.

All procedures performed in studies involving human participants or on human tissue were in accordance with the ethical standards of the institutional and/or national research committee and with the 1975 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Graph: Tables S1–S10. Further pain trajectories and satisfaction scores.

• COS

• Core outcome set

• CS

• Cesarean section

• D1

• First day after vaginal delivery and cesarean section

• IQR

• Interquartile range

• M

• Median

• 3M

• 3 months after vaginal delivery and cesarean section

• 6M

• 6 months after vaginal delivery and cesarean section

• NRS

• Numeric rating scale

• P

• Pregnant

• PCS

• Pain catastrophizing scale
• PRO (M)
• Patient-related outcome (measures)

• VD

• Vaginal delivery