Archives of Gynecology and Obstetrics https://doi.org/10.1007/s00404-017-4605-4

MATERNAL-FETAL MEDICINE

Changes in birth‑related impact of neurobiological and psycho‑social factors

Sebastian Berlit1 · Stefanie Lis2 · Katharina Häfner3 · Nikolaus Kleindienst3 · Ulf Baumgärtner4 · Rolf‑Detlef Treede4 · Marc Sütterlin1 · Christian Schmahl3,5

Received: 9 October 2017 / Accepted: 21 November 2017 © Springer-Verlag GmbH Germany, part of Springer Nature 2017

Abstract Purpose To analyse post-partum short- and long-term pain sensitivity and the infuence of endogenous pain inhibition as well as distinct psycho-social factors on birth-related pain. Methods Pain sensitivity was assessed in 91 primiparous women at three times: 2–6 weeks before, one to 3 days as well as ten to 14 weeks after childbirth. Application of a pressure algometer in combination with a was utilised for measurement of pain sensitivity and assessment of conditioned pain modulation (CPM). Selected psycho-social factors (anxiety, social support, history of abuse, and of childbirth) were evaluated with standardised questionnaires and their efect on pain processing then analysed. Results Pressure pain threshold, cold pain threshold and cold pain tolerance increased signifcantly directly after birth (all p < 0.001). While cold pain parameters partly recovered on follow-up, pressure pain threshold remained increased above baseline (p < 0.001). These pain-modulating efects were not found for women with history of abuse. While CPM was not afected by birth, its extent correlated signifcantly (r = 0.367) with the drop in pain sensitivity following birth. Moreover, high trait anxiety predicted an attenuated reduction in pain sensitivity (r = 0.357), while there was no correlation with fear of childbirth, chronic pain and social support. Conclusion Pain sensitivity showed a decrease when comparing post-partum with prepartum values. The extent and direc- tion of CPM appear to be a trait variable that predicted post-partum hypalgesia without being changed itself. Post-partum hypalgesia was reduced in women with a history of abuse and high trait anxiety, which suggests that individual diferences in CPM afect childbirth experience.

Keywords Birth · Pain · Social support · Abuse · Anxiety · Fear of childbirth

Introduction can lead to pathological antepartum as well as post-partum efects. Birth pain considerably infuences fear of child- Giving birth is arguably one of the most emotionally intense birth, which in turn is known to have a predominant impact and relevant key events in women’s life. Besides positive on the continuously rising caesarean section rate in the connotations, the substantial pain experienced under labour Western world [1]. Moreover, excessive labour pain can be experienced as a traumatic event which may lead to pro- Sebastian Berlit, Stefanie Lis, Rolf-Detlef Treede, Marc Sütterlin longed psychological sequelae [2]. These aspects explain and Christian Schmahl contributed equally.

* Sebastian Berlit 3 Department of Psychosomatic Medicine and Psychotherapy, [email protected] Central Institute of Mental Health, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany 1 Department of Obstetrics and Gynaecology, University 4 Department of Neurophysiology, Centre of Biomedicine Medical Centre Mannheim, Heidelberg University, and Medical Technology Mannheim, Medical Faculty Theodor‑Kutzer‑Ufer 1‑3, 68167 Mannheim, Germany Mannheim, Heidelberg University, Mannheim, Germany 2 Institute of Psychiatric and Psychosomatic Psychotherapy, 5 Department of Psychiatry, Schulich School of Medicine Central Institute of Mental Health, Medical Faculty and Dentistry, Western University, London, ON, Canada Mannheim, Heidelberg University, Heidelberg, Germany

Vol.:(0123456789)1 3 Archives of Gynecology and Obstetrics the importance of research on labour pain, which has been of painful stimuli such as a or the cold pressor addressed by various authors [3–10]. test (CPT) [4, 5, 12, 20]. Furthermore, most investigations Besides the disadvantageous psychological efects, birth in this context focused on particular labour pain infuencing pain is assumed to sustainably change physiological patterns factors [3]. of pain perception, i.e. to result in a reduced pain sensitiv- As labour pain can be afected by multiple physiologi- ity, which even persists after birth [4, 7]. Studies identifed cal, social and psychological determinants, we designed both biological and psycho-social factors as determinants this prospective investigation to assess short- and long-term of this change: alterations of pain processing after birth and the infuence of One of the biological mechanisms is endogenous pain multiple factors, which previously have been shown to have inhibition. First described in men severely wounded in battle an impact on birth pain. Incorporating the potentially bias- many studies have confrmed the existence of a descending ing neurobiological mechanism of CPM, we focused on the pain modulatory circuit that diminishes pain [11]. The status impact of the psycho-social determinants fear of childbirth, of this system can be evaluated by activation of pathways anxiety, social support and history of abuse on birth-related in the nociceptive system and quantitative sensory testing pain perception [3, 4, 6–9]. To our knowledge, this is the (QST) [12]: conditioned pain modulation (CPM) can be frst investigation with a comprehensive analysis of altera- used to measure the descending inhibition of nociceptive tions in pain processing after birth over time, which focuses signals indicated by the attenuation of the perceived inten- on both the efects of endogenous pain inhibition by means sity of a painful test stimulus after application of a painful of pressure pain sensitivity in the context of a CPM para- conditioning stimulus [13, 14]. It thus provides an objective digm and psycho-social factors. In line with the literature, measurement tool that allows to test whether and to which we hypothesised that giving birth will result in a reduced extent this mechanism contributes to both short- and long- pain sensitivity, indicated by increases in pain thresholds term reduction in post-partum pain sensitivity. to standardised painful stimuli. We expected that women’s Furthermore, some authors have confrmed that inter- endogenous pain inhibition before birth predicts the severity individual diferences in pain processing depend on a variety of change in pain perception after birth, with a stronger con- of psycho-social factors and this holds true for birth-related ditioned pain modulation being related to a stronger decrease pain processing: In a prospective investigation using a cold in pain sensitivity. Moreover, we hypothesised that psycho- pressure test to measure pain sensitivity, Saito et al. found social factors will afect birth-related alterations. Based on that patients with fear of childbirth tolerated pain signif- previous studies, we expected that both general anxiety and cantly shorter than did women without fear both during in fear of childbirth would attenuate the reduction in pain sen- the pregnancy and in the post-partum period [15]. Beyond sitivity after birth. fear specifcally related to giving birth, inter-individual dif- ferences in anxiety, as an individual’s stable personality fea- ture, seem to contribute to a variability of pain perception Methods changes during birth: Various studies revealed that increased anxiety is linked to augmented labour pain [3, 16–18]. More- Subjects over, the experience of social support and psycho-social factors such as a history of adverse childhood trauma may This investigation was approved by the Medical Ethics Com- infuence pain sensitivity after birth: a history of a traumatic mittee II of the University Medical Centre Mannheim, Hei- sexual experience increased pain sensitivity, leading to a delberg University, Mannheim, Germany (2010-347 N-MA) more traumatic birth experience in general [9, 19]. and conducted according to the Declaration of Helsinki as In sum, both psychological as well as biological factors well as the Declaration of the World Medical Association. are supposed to afect pain processing in the context of giv- After obtaining written informed consent, a total of 131 ing birth. However, most of the investigations so far focussed pregnant women registering for planned vaginal birth at the on pain processing only during or shortly after delivery using Department of Obstetrics of the University Medical Cen- solely single measurements. Studies capturing antepartum tre Mannheim were included in this prospective study. At pain processing and its changes following birth in longi- the time of recruitment, women were not in labour neither tudinal studies with repeated measurements over weeks or did they experience contractions nor did they have a his- months are sparse [4]. Thus, it is still unclear whether alter- tory of psychiatric illness. In order to be included in this ations of pain perception indeed persist over an extended investigation, patients had to be of legal age, right-handed period of time or not. Moreover, most studies focused on and primipara. In order to fulfl the criteria for follow-up alterations in the subjective perception of ongoing pain using assessment, women had to have given birth vaginally to a self-report questionnaires [3]. Only few authors investigated full-term singleton with cephalic presentation. Non-age, pain perception in response to the standardised application multiparity, preterm delivery (< 37 weeks of pregnancy),

1 3 Archives of Gynecology and Obstetrics caesarean section, multiple pregnancy and left handedness In order to assess the infuence of psycho-social factors on constituted exclusion criteria. pain processing, we analysed the infuence of trait anxiety, child birth anxiety, the perception of social support, as well Experimental paradigm as a history of abuse and a history of chronic pain on the pressure pain threshold determined before CPT. Participants Pain sensitivity was measured at three time points: flled in the questionnaires after recruitment, i.e. 2–6 weeks 2–6 weeks before childbirth (prepartum measurement), before childbirth. In more detail, we applied the following 1–3 days (post-partum measurement) and 10–14 weeks after self-report questionnaires: childbirth (follow-up measurement). -General Health Questionnaire [23]: For the present At each of these time points, pressure pain thresholds investigation, only items measuring history of abuse (total (PPT) were assessed with a dolorimeter before and after of 4 questions addressing history of abuse) and chronic pain application of a longer lasting ongoing pain stimulus; the (one question) were used. If a history of abuse was identi- cold pressor test (CPT), as a standard to quantify CPM, for- fed by one or more of the four questions, regardless of the merly known as difuse noxious inhibitory control (DNIC). type of abuse (sexual, psychological, etc.), the parturient was With these parameters, we aimed to obtain a reliable meas- classifed as a person with history of abuse. ure for prepartum mechanical pain sensitivity (pre modu- lation) and the magnitude of pain modulation as well as • State-Trait Anxiety Inventory (STAI): The 20-item trait potential changes after delivery. In this manuscript, the version of the STAI was used to assess general trait anxi- abbreviation “PPT” as a measure of pain sensitivity refers ety. It is an introspective self-report psychological inven- to the single measurement of pressure pain threshold before tory comprising 20 items, which are scored on 4-point- the CPT. The PPT measurement after the CPT was only used forced-choice Likert-type response scales. Scores range in order to assess CPM. from 20 to 80, with higher scores suggesting greater lev- The principle of the quantifcation of conditioned pain els of anxiety [24, 25]. modulation is to use phasic test stimuli (such as a mechani- • Social Support Questionnaire (F-SozU K-22, short ver- cal pain stimulus) to calculate the diference in ratings or sion): A 22 item inventory to assess experienced and threshold before and after the conditioning stimulus, which anticipated social support. The sub-scales “emotional is likely to induce descending inhibition (or facilitation), support”, “practical support”, “social integration”, like the CPT [12]. This way the CPT served as the noxious “availability of a person of trust” and “satisfaction with (conditioning stimulus) and the dolorimeter assessed pain social support” sum up to a total score. The questions are threshold as the test stimulus. Extent of CPM was calculated answered on a 5-point Likert scale rated from “not true” as log10(PPTpre/PPTpost). Negative values indicate inhibi- to “true” (minimum score 1, maximum score 5) [26]. tion, positive values facilitation [21]. Log values were back- • “Wijma-Delivery Expectancy/Experience Questionnaire transformed to yield percentage change in PPT for visualisa- (W-DEQ, version A)”: This is a unidimensional 33 items tion of the results in Fig. 2. questionnaire rated on a 0/5 scale, ranging from “not at Pressure pain threshold was measured on the thenar emi- all” to “extremely” (minimum score 0, maximum score nence of the dominant right hand using the dolorimeter FDN 165). A higher score indicates a more intense fear of 100 (Wagner Instruments; Greenwich, CT, USA) according childbirth [8]. to the metronome-controlled protocol of repeated measure- ments by Pfau et al. [22]. During the CPT, participants had to immerse their non-dominant left hand for a maximum of Statistics 3 min in 0–1 °C cold ice water. Additionally to the pressure pain threshold, onset of pain, pain tolerance and duration In a frst step, short- and long-term infuences of child birth of pain were measured during the CPT [6]: onset of pain on pain processing were analysed by means of variance ana- was defned as the time from immersion to frst pain sensa- lytical designs with time as repeated-measurement factor tion. Pain tolerance was defned as the time from immersion (prepartum, post-partum, follow-up) separately for the dif- until withdrawal of the hand from the ice water. Duration ferent parameters of pain processing, i.e. phasic pressure of pain was defned as time from frst pain sensation until pain threshold (PPT) and tonic pain parameters, i.e. pain withdrawal of the hand. threshold, and pain tolerance, derived during CPT, as well All dare were log-transformed for further analyses. Con- as conditioned pain modulation (diference of pressure pain ditioned pain modulation was calculated by subtracting the threshold prior and after CPT, CPM). Degrees of freedom pressure pain threshold prior to the CPT from that after CPT were adjusted according to Greenhouse and Geisser if appro- with negative values indicating inhibitory conditioned pain priate. Post hoc analyses were done by pairwise comparison modulation. contrasting pain perception prepartum, post-partum and at

1 3 Archives of Gynecology and Obstetrics follow-up. A Bonferroni correction for multiple comparisons degree (N = 45). Four women did not specify their higher was accomplished. educational qualifcation (Table 1). Conditioned pain modulation was calculated, after log- transformation, by subtracting the pressure pain threshold Pain sensitivity and birth after CPT from that prior to the CPT with negative values, indicating inhibitory conditioned pain modulation. Comparing pain sensitivity prepartum, post-partum and at To analyse the infuence of chronic pain and a history follow-up rm-ANOVA revealed diferences for all param- of abuse, PPT measured before the application of the CPT eters of pain processing (Fig. 1, Table 2). Pressure pain was analysed in a 2 × 3 variance analytical design with the threshold, cold pain threshold and cold pain tolerance repeated measurement factor “time” (prepartum, post-par- increased signifcantly directly after birth (all p < 0.001). tum, follow-up) and the independent factors “abuse” (yes/ While cold pain parameters recovered on follow-up, pres- no) and “chronic pain” (yes/no), respectively. sure pain threshold remained increased above baseline To assess the relationships between endogenous pain (p < 0.001). modulation and changes in pain processing after birth, a Pearson’s correlation coefcient was calculated between Conditioned pain modulation CPM and the change in PPT between prepartum and post- partum. To analyse whether the change of phasic pain Pressure pain threshold increased after experimental cold threshold after birth was predicted by fear of delivery, trait pressure pain on average by 10%. This efect was independ- anxiety and social support, linear regression analyses were ent of whether PPT was measured prepartum, post-partum or calculated. at follow-up suggesting that endogenous pain inhibition was not afected by child birth (see Table 2 and Fig. 2).

Covariation between endogenous pain inhibition Results and change of pressure pain threshold after birth

Sample characteristics To explore whether the extent of endogenous pain inhibition before birth was related to the increase of the pain threshold Of the total sample of 131 participants, 40 parturients had after birth, we calculated Pearson’s correlation coefcient to be excluded from analyses due to secondary caesarean between experimental conditioned pain modulation pre- section (n = 31), incomplete data sets (n = 8) and left hand- partum with the change in pressure pain threshold between edness (n = 1). Hence, full data sets were available for 91 prepartum and post-partum values. Analysis revealed that a patients, so that only these were incorporated in further stronger endogenous pain inhibition was linked to a stronger analyses. The resulting sample had a mean age of 30.6 years increase in pain threshold after birth (r = 0.367, p < 0.001). (SD 4.53, range 19–41). Mean body mass indices (BMI) See Fig. 3. However, endogenous pain inhibition explained increased after birth (prepartum: 24.1 kg/m2, SD 5.61; post- with 13.5% only a small portion of the variance in the partum: 29.8 kg/m2, SD 5.78, t = − 26.7, p < 0.001). Two increase of the pain threshold after birth. expectant mothers lived alone, 2 with their parents and 87 with their partners. Fifteen women were single, 62 married Infuence of history of abuse and chronic pain and 14 lived in a co-habitational relationship. Highest edu- cational achievement reached after graduation (all recruited When comparing women with (N = 12) and without abuse women had a high school diploma) was a master school (N = 79), analysis of pain threshold revealed a diferen- diploma (N = 8), an apprenticeship (N = 29) or a university tial change over time [F(2, 177) = 3.7, p = 0.027, partial

Table 1 Sample descriptives for AM SD SE Range parturients (N = 91) Age 30.6 4.53 0.48 19–41 BMI 24.1 5.61 0.59 15–45 Trait anxiety (STAI; score range 20–80) 34.3 8.58 0.91 20–62 Social support 4.5 0.37 0.38 3.4–5 (F-SozU-K22; score range 1–5) Fear of Childbirth W-DEQ (score range 0–165) 100.2 18.46 1.94 49–160

AM arithmetic mean, SD standard deviation, SE standard error

1 3 Archives of Gynecology and Obstetrics

Fig. 1 a–c Mean and standard error for pressure pain threshold (PPT) and parameters of the cold pressure test (CPT) prepartum, post-partum and at follow-up (*p < 0.05, ***p < 0.001 for post hoc pairwise comparisons)

Table 2 Results of rm-ANOVA for PPT and CPT parameters together with post hoc pairwise comparisons between prepartum, post-partum and follow-up Bonferroni correction for multiple testing

* 2 df1, df2 F p Partial η Prepartum vs Prepartum vs Post-partum post-partum follow-up vs follow-up

Pressure pain threshold kPa 2.0; 180.0 23.16 < 0.001 0.205 < 0.001 < 0.001 1.000 Pain tolerance ST s 1.9; 171.2 9.81 < 0.001 0.098 < 0.001 0.102 0.058 Onset of pain SB s 1.9; 172.9 10.92 < 0.001 0.108 < 0.001 1.000 0.002 Duration of pain STSB s 1.9; 172.3 6.21 0.003 0.065 0.004 0.049 0.648 Conditioned pain modulation ΔkPa 1.9; 175.0 < 0.001 .998 < 0.001 – – –

* Greenhouse–Geisser corrected

η2 = 0.040]. See Fig. 4. Pain threshold in women with difer signifcantly in regard to age (abuse: 30.75, without abuse was not afected by birth (all p > 0.140). In con- abuse: 30.63, t = − 0.1, p = 0.934), social support (abuse: trast in women without abuse, pain threshold was higher 4.32, without abuse: 4.56, t = 2.2, p = 0.120), fear of birth at post-partum and follow-up compared to prepartum (both (W-DEQ) (abuse: 101.42, without abuse: 99.99, t = − 0.3, p < 0.001), while no diferences were found between post- p = 0.804) or conditioned pain modulation (abuse: − 0.047, partum and follow-up (p = 0.356). Analyses did not reveal without abuse: − 0.046, t = 0.7, p = 0.946). a diference between groups independent of the measure- ment time [F(1, 89) = 0.45, p = 0.504, partial η2 = 0.005]. Infuence of anxiety, fear of birth and social support The interpretability of the main efect of the factor “time” on pain threshold was restricted due to the higher-order interaction efect [F(2, 177) = 5.51, p = 0.005, partial η2 = 0.058]. Women The increase of pressure pain threshold from prepartum to with a history of abuse reported a higher trait anxiety than post-partum was predicted by general trait anxiety (STAI women without a history of abuse (abuse: 41.17, without b = − 0.005, p = 0.001, r = 0.357, r2 = 0.128), but not by abuse: 33.29, t = − 3.1, p = 0.003). Both groups did not fear of birth (W-DEQ: b = − 0.0002, p = 0.826, r = 0.023,

1 3 Archives of Gynecology and Obstetrics

Fig. 4 Mean and standard error of pressure pain threshold at prepar- tum, post-partum and follow-up for women with and without history of abuse

r2 = 0.001) nor social support (F_SOZU: b = 0.040, p = 0.259, r = 0.120, r2 = 0.014): Higher trait anxiety pre- dicted a smaller increase of pain threshold after birth (less inhibition) or even a decrease (facilitation). See Fig. 5. Fig. 2 Mean and standard error for conditioned pain modulation When comparing women with (N = 20) and without (CPM) prepartum, post-partum and at follow-up measurement, chronic pain (N = 71), analysis of pain threshold revealed respectively calculated as log10(PPTpre/PPTpost). Negative values no diferential change of pressure pain threshold over time indicate inhibition, positive values facilitation [F(2, 178) = 0.89, p = 0.411, partial η2 = 0.010; main effect “chronic pain”: F(1, 89) = 0.25, p = 0.617, par- tial η2 = 0.003; main efect “time”: F(2, 178) = 15.87, p < 0.001, partial η2 = 0.151]. Hence, in contrast to trait anxiety, neither chronic pain, fear of delivery nor social support predicted the extent of changes in pain threshold after birth.

Discussion

Assessment of pain parameters revealed that all pain enti- ties underwent signifcant changes comparing measurements before with measurements directly after birth. In general, PPT increased after delivery and stayed on a higher level, whereas endogenous pain inhibition remained almost unal- tered. Most notably, the increase of pain threshold from baseline to post-partum was predicted by general trait anxi- ety but not by fear of childbirth nor social support. Trauma- tised women did not show the overall trend of reduced pain sensitivity after delivery. Fig. 3 Covariation between conditioned pain modulation prepartum Persistence of a reduced pain sensitivity (prepartum to and change in pain perception after birth follow-up) was found for the pressure pain threshold (both

1 3 Archives of Gynecology and Obstetrics

Fig. 5 Covariation of a trait anxiety (STAI), b fear of child birth (WDEC) and c social support (F-SozU) with short-time change in pressure pain threshold from prepartum to post-partum before and after CPT), as well as pain tolerance and dura- increases pain sensitivity and, in line with our results, vari- tion of pain while processing tonic pain stimuli during the ous authors found a correlation between increased anxiety CPT. The remaining pain entities returned to baseline val- and augmented labour pain [3, 16–18]. In contrast to our ues a few weeks after delivery. Only few studies exist on fndings Saito et al., using a CPT, found that patients with experimentally measured pain sensitivity in the context of fear of birth tolerated pain signifcantly shorter than did childbirth. Ohel et al. assessed PPT in n = 40 women at women without fear both during in the pregnancy and in the term before labour, during the active phase of labour and in post-partum period [15]. the early post-partum period [4]. The authors also found a Chronic pain showed no impact on pain threshold altera- signifcant rise in PPT during labour when compared with tions. This stands in contrast with existing literature suggest- both thresholds before and after labour. However, as we did ing that women having experienced chronic or severe pain not assess PPT during delivery a direct comparison between tend to have reduced pain sensitivity after birth compared to these two studies is not possible. Saito et al. compared pain parturients without such a history of pain [3, 20, 27]. tolerance in women with (n = 20) and without (n = 20) fear A wide variety of social variables have been studied in of childbirth and found, analogously to our results, that in all relationship to the pain of childbirth, and their infuence on participants (independently of the level of fear of birth) pain a woman’s perception of pain during labour is a well-known tolerance was lower during compared to after pregnancy clinical phenomenon [3]. To the best of our knowledge, this [15]. Further indicators for a sustainable change in pain sen- is the frst investigation to assess social support via a stand- sitivity after vaginal birth are various investigations in which ardised questionnaire. In contrast to the assumption that authors found that pain perception was lower in parous com- greater social support leads to a reduced pain sensitivity, no pared to nulliparous women [5–7]. Up to this point of time, signifcant efect was found. existing literature does not permit a fnal conclusion. Besides anxiety, a history of abuse also had an impact Assessment of the psycho-social determinants anxiety, on pain thresholds. Women with a history of abuse showed social support and fear of birth revealed that the increase of no diferential change of pain sensitivity as there were pain threshold before birth to post-partum was predicted by no PPT alterations in contrast to women without history general trait anxiety (higher trait anxiety/smaller increase of of abuse, with a signifcant post-partum increase of pain pain threshold after birth), but not by fear of birth nor social threshold. Arguably, a history of abuse constitutes an addi- support. An explanation for this unexpected efect could be tional strain leading to a less alterable, higher pain sensi- that childbirth burdens anxious women more, hence leading tivity. Soet et al. demonstrated that a history of traumatic to a prolonged reduction in pain sensitivity compared to less sexual experience constitutes a predictor for traumatic anxious parturients. It is assumed that anxiety commonly birth experience, which goes in line with our fndings [9,

1 3 Archives of Gynecology and Obstetrics

19]. Also Leeners et al. found women with a history of Compliance with ethical standards abuse to experience labour more dramatic with a signif- cantly higher necessity for anaesthesia to reduce pain com- Funding There was no funding for this investigation. pared to controls [9]. One might assume that women with a reduced psycho- Conflict of interest All authors state that there is no confict of interest. logical burden (less anxious, more social support, less fear Ethical approval All procedures performed in studies involving human of birth, no history of abuse or chronic pain) experience a participants were in accordance with the ethical standards of the insti- greater reduction in pain sensitivity after birth compared to tutional and/or national research committee and with the 1964 Helsinki women with a less stable psycho-social pattern. Our results declaration and its later amendments or comparable ethical standards. partly support this assumption, in terms of anxiety and his- This article does not contain any studies with animals performed by any of the authors. tory of abuse. Finally, the following limitations of the investigation have Informed consent Informed consent was obtained from all individual to be considered: We did not include an objective measure participants included in the study. of the painfulness of the birth itself. Future studies should include parameters such as duration of birth, foetal birth weight, degree of perineal laceration, e.g. in order to clas- sify severity of perceived birth trauma. Moreover, the fnd- References ings concerning women with a trauma history have to be interpreted with caution: the sample size was small and we 1. Ryding EL, Lukasse M, Kristjansdottir H, Steingrimsdottir T, did not apply a detailed questionnaire such as the “Child- Schei B, Bidens study g (2016) Pregnant women’s preference for hood Traumatisation Questionnaire” to assess the type and cesarean section and subsequent mode of birth—a six-country cohort study. J Psychosom Obstet Gynaecol 37(3):75–83. https:// severity of childhood adversity [28]. Nevertheless, our data doi.org/10.1080/0167482X.2016.1181055 suggest that childhood traumatisation has a strong impact 2. Bastos MH, Furuta M, Small R, McKenzie-McHarg K, Bick D on pain processing characterised by a dysfunction of those (2015) Debriefng interventions for the prevention of psychologi- biological mechanisms protecting women from aversive cal trauma in women following childbirth. Cochrane Database Syst Rev. https://doi.org/10.1002/14651858.CD007194.pub2 childbirth aspects. Similarly, a more elaborated assessment 3. Lowe NK (2002) The nature of labor pain. Am J Obstet Gynecol of chronic pain may have allowed to detect a detailed rela- 186(5 Suppl Nature):S16-24 tion between a history of pain and pain perception. 4. Ohel I, Walfsch A, Shitenberg D, Sheiner E, Hallak M (2007) A rise in pain threshold during labor: a prospective clinical trial. Pain 132(Suppl 1):S104–S108. https://doi.org/10.1016/j. pain.2007.05.007 5. Sheiner E, Sheiner EK, Shoham-Vardi I (1998) The relation- Conclusion ship between parity and labor pain. Int J Gynaecol Obstet 63(3):287–288 6. Hapidou EG, DeCatanzaro D (1992) Responsiveness to laboratory Inter-individual diferences in the strength of endogenous pain in women as a function of age and childbirth pain experience. pain inhibition, trait anxiety and a history of childhood mal- Pain 48(2):177–181 treatment afect the strength of short-term alterations in post- 7. Melzack R, Kinch R, Dobkin P, Lebrun M, Taenzer P (1984) partal pain processing. Both biological and psycho-social Severity of labour pain: infuence of physical as well as psycho- logic variables. Can Med Assoc J 130(5):579–584 factors in this study appeared to exert their efects as gener- 8. Alehagen S, Wijma B, Lundberg U, Wijma K (2005) Fear, pain alised trait variables and not as situation specifc (no efect and stress hormones during childbirth. J Psychosom Obstet of fear of childbirth) or state variables (no change in CPM Gynaecol 26(3):153–165 across measurement times). In general, these determinants 9. Leeners B, Gorres G, Block E, Hengartner MP (2016) Birth experiences in adult women with a history of childhood sexual were signifcant but explained only a rather small proportion abuse. J Psychosom Res 83:27–32. https://doi.org/10.1016/j. of the variability in responsivity of postpartal pain process- jpsychores.2016.02.006 ing. The clinical relevance of these associations should be 10. Gintzler AR, Bohan MC (1990) Pain thresholds are elevated dur- investigated by a comprehensive multidisciplinary assess- ing pseudopregnancy. Brain Res 507(2):312–316 11. Beecher HK (1946) Pain in men wounded in battle. Ann Surg ment of childbirth experience itself during labour, which 123(1):96–105 was not the focus of the present study. 12. Skovbjerg S, Jorgensen T, Arendt-Nielsen L, Ebstrup JF, Carstensen T, Graven-Nielsen T (2016) Conditioned pain modu- Author contributions SB wrote the manuscript. SL was involved in lation and pressure pain sensitivity in the adult Danish general data analysis. KH collected the data. NK was involved in data manage- population: the DanFunD study. J Pain. https://doi.org/10.1016/j. ment. UB edited the manuscript. RDT was involved in project develop- jpain.2016.10.022 ment and manuscript editing. MS was involved in project development 13. Ossipov MH, Dussor GO, Porreca F (2010) Central modulation and manuscript editing. CS was involved in project development and of pain. J Clin Invest 120(11):3779–3787. https://doi.org/10.1172/ manuscript editing. JCI43766

1 3 Archives of Gynecology and Obstetrics

14. Nir RR, Yarnitsky D (2015) Conditioned pain modulation. Curr Quantitative sensory testing in the German Research Network Opin Support Palliat Care 9(2):131–137. https://doi.org/10.1097/ on Neuropathic Pain (DFNS): reference data for the trunk and SPC.0000000000000126 application in patients with chronic postherpetic . Pain 15. Saisto T, Kaaja R, Ylikorkala O, Halmesmaki E (2001) Reduced 155(5):1002–1015. https://doi.org/10.1016/j.pain.2014.02.004 pain tolerance during and after pregnancy in women sufering 23. Kamping S (2016) Pain modulation by contextual factors in maso- from fear of labor. Pain 93(2):123–127 chists: the role of the insula and parietal operculum. Schmerz 16. Waldenstrom U, Bergman V, Vasell G (1996) The complexity 30(2):187–189. https://doi.org/10.1007/s00482-015-0090-5 of labor pain: experiences of 278 women. J Psychosom Obstet 24. Laux LGP, Schafner P, Spielberger CD (1981) STAI—Das State- Gynaecol 17(4):215–228 Trait-Angstinventar—Theoretische Grundlagen und Handanwei- 17. Reading AE, Cox DN (1985) Psychosocial predictors of labor sungen. Beltz Testgesellschaft, Weinheim pain. Pain 22(3):309–315 25. Gros DF, Antony MM, Simms LJ, McCabe RE (2007) Psycho- 18. Lowe NK (1989) Explaining the pain of active labor: the impor- metric properties of the State-Trait Inventory for Cognitive and tance of maternal confdence. Res Nurs Health 12(4):237–245 Somatic Anxiety (STICSA): comparison to the State-Trait Anxi- 19. Soet JE, Brack GA, DiIorio C (2003) Prevalence and predictors ety Inventory (STAI). Psychol Assess 19(4):369–381. https://doi. of women’s experience of psychological trauma during childbirth. org/10.1037/1040-3590.19.4.369 Birth 30(1):36–46 26. Dunkel D, Antretter E, Frohlich-Walser S, Haring C (2005) Evalu- 20. Bajaj P, Bajaj P, Madsen H, Moller M, Arendt-Nielsen L (2002) ation of the short-form social support questionnaire (SOZU-K-22) Antenatal women with or without can be characterized in clinical and non-clinical samples. Psychother Psychosom Med by generalized or segmental in late pregnancy. J Pain Psychol 55(5):266–277. https://doi.org/10.1055/s-2004-834746 3(6):451–460 27. Niven C, Gijsbers K (1984) A study of labour pain using the 21. Yarnitsky D, Bouhassira D, Drewes AM, Fillingim RB, Granot McGill Pain Questionnaire. Soc Sci Med 19(12):1347–1351 M, Hansson P, Landau R, Marchand S, Matre D, Nilsen KB, Stub- 28. Klinitzke G, Romppel M, Hauser W, Brahler E, Glaesmer H haug A, Treede RD, Wilder-Smith OH (2015) Recommendations (2012) The German Version of the Childhood Trauma Question- on practice of conditioned pain modulation (CPM) testing. Eur J naire (CTQ): psychometric characteristics in a representative sam- Pain 19(6):805–806. https://doi.org/10.1002/ejp.605 ple of the general population. Psychother Psychosom Med Psychol 22. Pfau DB, Krumova EK, Treede RD, Baron R, Toelle T, Birklein F, 62(2):47–51. https://doi.org/10.1055/s-0031-1295495 Eich W, Geber C, Gerhardt A, Weiss T, Magerl W, Maier C (2014)

1 3