{"paper_id":"4067f489-825a-4282-93ef-805489b3d1fc","body_text":"1 \n \n  Prediction of Procedural Pain during Endometrial Biopsy \n(Short running head: Pain Prediction during Endometrial Biopsy) \n \nHye Gyeong Jeong1, 4*, Jooyoung Kim2*, Sung il Choo3, Kidong Kim4, Banghyun Lee5, Soyeon Ahn6 \n \n1 Department of Obstetrics and Gynecology, Seoul National University College of Medicine, Seoul, Korea \n2 Department of Obstetrics and Gynecology,  Uijeongbu Eulji Medical Center, Eulji University, 712, Dongil -ro, \nUijeongbu-si, Gyeonggi-do, Republic of Korea \n3 Department of Obstetrics and Gynecology, Severance Hospital, Yonsei University College of Medicine, 50 -1 9 \nYonsei-ro, Seodaemun-gu, Seoul 03722, Korea \n4 Department of Obstetrics and Gynecology, Seoul National University Bundang Hospital, 82Gumi-ro, 173 Beon-\ngil, Bundang-gu, Seongnam-si, Gyeonggi-do, Korea 13620 \n5 Department of Obstetrics and Gynecology, Hallym University Kangdong Sacred Heart Hospital, 150, Seongan -\nro, Gangdong-gu, Seoul, Republic of Korea \n6 Division of Statistics, Medical Research Collaborating Center Seoul National University Bundang Hospital  \n \n*These authors contributed equally to this study. \n \nCorrespondence to: \nKidong Kim, MD, PhD \nDepartment of Obstetrics and Gynecology, Seoul National University Bundang Hospital, Seongnam-si, Gyeonggi-\ndo, Republic of Korea  \nTel: 82-31-787-7262 \nE-mail: kidong.kim.md@gmail.com \n \nType of manuscript: Original Research \n \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \nNOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.\n\n2 \n \nAbstract \nBackground/Objective \nEndometrial (EM) biopsy is a commonly-performed gynecological procedure that is associated with side effects \nsuch as discomfort and pain. The aim of the current study was to predict procedural pain during EM biopsy. \nMethods \nWe retrospectively reviewed the medical records of 100 women who underwent EM biopsy between July 2014 \nand November 2015 in an outpatient clinic of our hospital. Eighty-one patients were included in the final analysis \nafter excluding those who lacked pain data and those who were  sedated with midazolam . We examined the \nassociation of patient and clinician characteristics with procedural pain , and created  a prediction model using \ncharacteristics via multiple linear regression analysis. \nResults \nEighty-one women underwent EM biopsy ( dilatation and curettage, EM sampling). In univariable analysis, \nhistory of EM biopsy, endometrial thickness (EMT) and  training year  of operator (TY)  were significantly \nassociated with procedural pain . The initial multivariable model was fitted with significant predictors in a \nunivariable analysis. The p -value of EMT and TY was below the pre-defined threshold ( 0.2) and the final pain \nprediction model included EMT and TY . Furthermore, pain during the procedure was calculated by the following \nequation: pain score (numeric rating scale) = 7.364 + (-0.872) * EM thickness (cm) + (-1.033)*TY .  \nConclusion \nBoth endometrial thickness and training year of operator were useful predict ors of the severity of EM biopsy-\nrelated pain. \n \nKEY WORDS: endometrium, biopsy, procedural pain, outpatient clinic, decision support techniques \n  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n3 \n \nIntroduction \nEndometrial (EM) biopsy is a commonly performed outpatient procedure to evaluate the endometrium in patients \nwith abnormal uterine bleeding or abnormal findings  on sonography. According to a report in 2016, 21,889 EM \nbiopsies were performed in South Korea, of which 21071 (91.5%) were performed in outpatient clinics. (1). \nEM biopsy causes discomfort and pain. More than half of EM biopsied patients describe their experiences  as \n\"moderately\" or \"severely\" painful (2). Previous studies have demonstrated that  procedural pain is influenced by \nparity, pre -procedural anxiety,  menopausal status, history of vaginal delivery , provider experience, us e of a  \ntenaculum, and procedure time (3, 4). To reduce procedural pain, paracervical block, intrauterine anesthesia, oral \nmedications such as non -steroidal anti-inflammatory drugs or opioids, and intravenous sedation have been used. \nMany studies have concluded that these methods are effective for reduction of procedural pain (3, 5, 6). \nNevertheless, these methods to reduce procedural pain were not applicable to all patients due to many reasons.  \nFor example, intravenous sedation requires patient monitoring and a trained anesthetist as well as appropriate space \nfor recovery after the procedure, and these requirements increase the cost of EM biopsy. If the degree of procedural \npain can be predicted  prior to EM biopsy , we can implement more cost -effective methods of pain control . \nSpecifically, we can limit the need for intensive, resource demanding anesthesia and analgesia for women who \nexperience moderate or more severe pain. \nThis study aimed to build a predictive model for procedural pain during EM biopsy.  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n4 \n \nMaterials and methods \n \nPatients \nThis study was approved by the Institutional review board and the requirement for informed consent was waived \n(B-1606-349-111). We retrospectively reviewed the medical records of 100 women who visited the outpatient \nclinic of our institute and underwent EM biopsy between July 2014 and November 2015. Patient s without a pain \nrecord (n=15) and those sedated with midazolam (n=4) were excluded. Finally, a total of 81 patients were included \nin this study. \n \nVariables \nAge, menopause, parity, history of vaginal delivery, history of EM biopsy, presence of myoma and adenomyosis, \nuterus size and position , method of EM biopsy (dilatation and curettage or  EM sampling), EM thickness, gender \nand training year of residents who performed the procedu re, application of paracervical block , and information \nregarding maximal pain during procedure were retrieved from the medical records. \nAll procedures were performed by second or third year residents. The number of second and third year residents \nwas six and three, respectively. During the procedure, the anterior cervical lip was grasped with the tenaculum and \nthen uterine sound was inserted to the uterine fundus. The cervical os was dilated using a Hegar dilator. EM biopsy \nwas performed using a curette or sampler. Immediately after completing the procedure, the resident who performed \nthe EM biopsy recorded information regarding the maximal pain during the procedure and rated this on a 10-point \nscale (numeric rating scale , NRS). No anesthesia or analgesia was provided before the procedure except for \nparacervical block in some patients. \n \nAnalysis \nAll variables except maximal pain during procedure were converted to dichotomous variables. Maximal pain \nduring procedure was summarized into median and inter -quartile range (IQR). The association of variables with \nmaximal pain during procedure was examined using the Mann-Whitney test, and a p-value of <0.05 was considered \nsignificant. Cases with unknown values for a variable were excluded from the univariable analysis for that variable.  \nV ariables with p-value of <0.05 in univariable analysis were included into multivariable analysis. V ariable s \nincluded in  the final model were chosen using backward selection with a threshold of p -value = 0.2. Internal \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n5 \n \nvalidation of the model was performed using bootstrap analysis based on 1000 replications. \n Analysis was performed using R 3.3.0 version. \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n6 \n \nResults \n \nCharacteristics \n Characteristics of patients are summarized in Table 1. The median age of patients was 54 and the majority of \nwomen were menopausal (45/81). Most patients had a history of vaginal delivery but only 22 patients had a history \nof EM biopsy. Myoma and adenomyosis were present in 35 and 9 patients, respectively. The median uterus size \nwas 7 cm and most patients had an antero-verted uterus. Dilatation and curettage was more frequently performed \nthan EM sampling. The median EM thickness was 0.8 cm. Most of the procedures were performed by female, 2nd \nyear residents and paracervical block was performed in over half of patients. The median NRS of maximal pain \nduring the procedure was 4. \n \nUnivariable, multivariable analysis and predictive model \n Based on univariable analysis, a history of EM biopsy , EM thickness , and training year of residents who \nperformed the procedure were associated with maximal pain during the procedure (Table 1). Therefore, the initial \nmodel was constructed using three variables (history of EM biopsy, EM thickness , and training year of residents \nwho performed the procedure). Because the p -value of history of EM biopsy was over the pre-defined threshold, \nthe final model was constructed using EM thi ckness and training year of residents who performed the procedure  \n(NRS = 7.364 + (-0.872) * EM thickness (cm) + (-1.033) * training year of residents who performed the procedure \n(2 vs 3)). \n In validation using 1000 times bootstrapping, the mean squared error and mean absolute error of the final model \n(0.14, 0.16) were better than those of the initial model (0.33, 0.35). The calibration plot of the final model is \nillustrated in Figure 1.  \n \n  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n7 \n \nDiscussion \n \n We successfully built a predictive model for procedural pain during EM biopsy using EM thickness and training \nyear of residents who performed the procedure . Our model will be useful to help predict procedural pain during \nEM biopsy and therefore decide whether patients require additional ane sthetic or analgesic intervention s. \nAccording to our model, a thinner endometrium is expected to be associated with a more painful procedure. This \nsuggests that more attention should be given to pain management when an operator performs a biopsy of a thin \nendometrium. Additionally, our model showed that the level of experience of residents also influences procedural \npain, which highlights the importance of appropriately training residents to perform this common p rocedure.  \n The results of this study are partially discordant with those of other studies regarding predictors of procedural \npain during EM biopsy. For example,  in contrast to the results of our study, a previous study reported that \npostmenopausal women tend to have more severe pain during EM biopsy , and that a history of vaginal delivery \nwas associated with procedural pain (3). A different study also reported a positive correlation between endometrial \nthickness and pain (<5 mm vs. ≥5 mm) (6), which is contradictory to our results. Consistent with our results, \nprevious studies have demonstrated that the skill or experience of the operator is a predictor of procedural pain (7). \nIn addition in keeping with our results, it has been previously demonstrated that the method of EM biopsy , for \nexample curette versus Pipelle biopsy, was not associated with a significant difference in pain scores (8).  \n To the best of our knowledge, this is the first study to identify predictors of procedural pain during EM biopsy. \nNevertheless, there are also several limitations worth noting. First, this was a single center study and as a result, \nthe number of patients examined was small . Second, this study was vulnerable to many biases because of its \nretrospective nature. Third, external validation was not performed.  \n In conclusion, we successfully developed a predictive model for procedural pain  during EM biopsy using EM \nthickness and training year of residents who performed the procedure. We believe this model will help clinicians \nto manage procedural pain more effectively.   \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n8 \n \nAcknowledgements \n \n \nConflict of interest \nNo potential conflict of interest relevant to this article was reported. \n  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n9 \n \nReferences \n1. Health Insurance Review & Assessment Service. Endometrial biopsy in outpatient clinic (2016) [Internet] \nGangwon(KR): Health Insurance Review & Assessment Service; c2016 [cited 2018 Feb 15]. Available from: \nhttp://opendata.hira.or.kr/op/opc/olapDiagBhvInfo.do. \n2. Dogan E, Celiloglu M, Sarihan E, Demir A. Anesthetic effect of intrauterine lidocaine plus naproxen \nsodium in endometrial biopsy. Obstetrics and gynecology. 2004;103(2):347-51. \n3. Ireland LD, Allen RH. Pain Management for Gynecologic Procedures in the Office. Obstetrical & \ngynecological survey. 2016;71(2):89-98. \n4. Cicinelli E. Hysteroscopy without anesthesia: review of recent literature. J Minim Invasive Gynecol. \n2010;17(6):703-8. \n5. Vigneault L, Turgeon AF, Cote D, Lauzier F, Zarychanski R, Moore L, et al. Perioperative intravenous \nlidocaine infusion for postoperative pain control: a meta-analysis of randomized controlled trials. Canadian journal \nof anaesthesia = Journal canadien d'anesthesie. 2011;58(1):22-37. \n6. Kosus N, Kosus A, Demircioglu RI, Simavli SA, Derbent A, Keskin EA, et al. Transcervical intrauterine \nlevobupivacaine or lidocaine infusion for pain control during endometrial biopsy. Pain research & management. \n2014;19(2):82-6. \n7. Hubacher D, Reyes V , Lillo S, Zepeda A, Chen PL, Croxatto H. Pain from copper intrauterine device \ninsertion: randomized trial of prophylactic ibuprofen. American journal of obstetrics and gynecology. \n2006;195(5):1272-7. \n8. Leclair CM, Zia JK, Doo m CM, Morgan TK, Edelman AB. Pain experienced using two different \nmethods of endometrial biopsy: a randomized controlled trial. Obstetrics and gynecology. 2011;117(3):636-41. \n \n  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n10 \n \nTable 1. Baseline characteristics and univariable analysis between variables and maximal pain during the \nprocedure (n = 81) \nVariables N \nmaximal pain during procedure, \nNRS, median (IQR) \np-value \nAge, years    \n<54 40 4 (3-7) 0.28 \n≥54 41 4 (3-5)  \nMenopause    \nYes 45 5 (3-6) 0.37 \nNo 36 4 (3-5)  \nParity    \n0 5 7 (3-8) 0.61 \n≥1  22 4 (4-6)  \nUnknown 54   \nHistory of vaginal delivery    \nYes 19 4 (3-4.5) 0.11 \nNo 8 7 (4-8)  \nUnknown 54   \nHistory of EM biopsy    \nYes 22 5.5 (4-7) 0.01 \nNo 59 4 (3-5)  \nUnknown 0   \nPresence of myoma    \nYes 35 5 (3-6) 0.57 \nNo 41 4 (3-6)  \nUnknown 5   \nPresence of adenomyosis    \nYes 9 4 (4-6) 0.66 \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n11 \n \nNo 67 4 (3-6)  \nUnknown 5   \nUterus size, cm     \n>7 40 4 (3-6) 0.79 \n≤7 34 4 (3-5)  \nUnknown 7   \nUterus position    \nAntero-verted 48 4 (3-6) 0.05 \nRetro-verted 20 5.5 (4-7)  \nUnknown 13   \nMethod of EM biopsy    \nDilatation and curettage 56 4 (3-6) 0.34 \nEM sampling 25 5 (3-7)  \nEM thickness, cm    \n≥0.8 24 3.5 (3-4) 0.02 \n<0.8 35 5 (3.5-6)  \nUnknown 22   \nGender of residents who performed \nthe procedure \n   \nMale 19 5 (3-6.5) 0.98 \nFemale 62 4 (3-6)  \nTraining year of residents who \nperformed the procedure \n   \n2nd year 65 4.5 (3-6) 0.01 \n3rd year 16 3.5 (1.5-4)  \nParacervical block    \nYes 48 4 (3-5) 0.98 \nNo 33 5 (3-5.5)  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n12 \n \nMaximal pain during procedure  4 (3-6)  \n \nCases with unknown value were excluded from the analysis \nNRS = Numeric rating scale; IQR = interquartile range; EM = endometrial \n \n  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n13 \n \nTable 2. Multivariable analysis and model development \nV ariables in each model Coefficient P-value 95% CI \nInitial model    \nIntercept 7.092 <0.01 4.214, 9.97 \nHistory of EM biopsy 0.765 0.24 -0.523, 2.052 \nEM thickness -0.852 0.05 -1.703, -0.002 \nTraining year of residents who performed the procedure -0.99 0.12 -2.256, 0.276 \nFinal model    \nIntercept 7.364 <0.01 4.514, 10.214 \nEM thickness -0.872 0.05 -1.725, -0.019 \nTraining year of residents who performed the procedure -1.0334 0.11 -2.301, 0.235 \nEM = endometrial \n \n  \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint \n\n14 \n \nFigure 1. Calibration plot of the final model \n \n \n \nAll rights reserved. No reuse allowed without permission. \nperpetuity. \npreprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.24.21254143doi: medRxiv preprint","source_license":"CC0","license_restricted":false}