Advertisement

Prediction of breakthrough pain during labour neuraxial analgesia: comparison of machine learning and multivariable regression approaches

Published:August 25, 2020DOI:https://doi.org/10.1016/j.ijoa.2020.08.010

      Highlights

      • Predictive models for breakthrough pain were developed.
      • Retrospective data of neuraxial labour analgesia from a single centre were used.
      • Machine learning (random forest and XGBoost) and logistic regression were used.
      • The models performed similarly, with area-under-the-curve 0.763–0.772.
      • The model derived using logistic regression required the fewest variables.

      Abstract

      Introduction

      Risk-prediction models for breakthrough pain facilitate interventions to forestall inadequate labour analgesia, but limited work has used machine learning to identify predictive factors. We compared the performance of machine learning and regression techniques in identifying parturients at increased risk of breakthrough pain during labour epidural analgesia.

      Methods

      A single-centre retrospective study involved parturients receiving patient-controlled epidural analgesia. The primary outcome was breakthrough pain. We randomly selected 80% of the cohort (training cohort) to develop three prediction models using random forest, XGBoost, and logistic regression, followed by validation against the remaining 20% of the cohort (validation cohort). Area-under-the-receiver operating characteristic curve (AUC), sensitivity, specificity, and positive and negative predictive values (PPV and NPV) were used to assess model performance.

      Results

      Data from 20 716 parturients were analysed. The incidence of breakthrough pain was 14.2%. Of 31 candidate variables, random forest, XGBoost and logistic regression models included 30, 23, and 15 variables, respectively. Unintended venous puncture, post-neuraxial analgesia highest pain score, number of dinoprostone suppositories, neuraxial technique, number of neuraxial attempts, depth to epidural space, body mass index, pre-neuraxial analgesia oxytocin infusion rate, maternal age, pre-neuraxial analgesia cervical dilation, anaesthesiologist rank, and multiparity, were identified in all three models. All three models performed similarly, with AUC 0.763–0.772, sensitivity 67.0–69.4%, specificity 70.9–76.2%, PPV 28.3–31.8%, and NPV 93.3–93.5%.

      Conclusions

      Machine learning did not improve the prediction of breakthrough pain compared with multivariable regression. Larger population-wide studies are needed to improve predictive ability.

      Keywords

      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'

      Subscribe:

      Subscribe to International Journal of Obstetric Anesthesia
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect

      References

        • Anim-Somuah M.
        • Smyth R.M.
        • Cyna A.M.
        • Cuthbert A.
        Epidural versus non-epidural or no analgesia for pain management in labour.
        Cochrane Database Syst Rev. 2018; 5 (CD000331)
        • Chan J.J.I.
        • Gan Y.Y.
        • Dabas R.
        • et al.
        Evaluation of association factors for labor episodic pain during epidural analgesia.
        J Pain Res. 2019; 12: 679-687
        • Sng B.L.
        • Tan M.
        • Yeoh C.J.
        • et al.
        Incidence and risk factors for epidural re-siting in parturients with breakthrough pain during labour epidural analgesia: a cohort study.
        Int J Obstet Anesth. 2018; 34: 28-36
        • Agaram R.
        • Douglas M.J.
        • McTaggart R.A.
        • Gunka V.
        Inadequate pain relief with labor epidurals: a multivariate analysis of associated factors.
        Int J Obstet Anesth. 2009; 18: 10-14
        • Eappen S.
        • Blinn A.
        • Segal S.
        Incidence of epidural catheter replacement in parturients: a retrospective chart review.
        Int J Obstet Anesth. 1998; 7: 220-225
        • Paech M.J.
        • Godkin R.
        • Webster S.
        Complications of obstetric epidural analgesia and anaesthesia: a prospective analysis of 10,995 cases.
        Int J Obstet Anesth. 1998; 7: 5-11
        • Hess P.E.
        • Pratt S.D.
        • Lucas T.P.
        • et al.
        Predictors of breakthrough pain during labor epidural analgesia.
        Anesth Analg. 2001; 93: 414-418
        • Hood D.D.
        • Dewan D.M.
        Anesthetic and obstetric outcome in morbidly obese parturients.
        Anesthesiology. 1993; 79: 1210-1218
        • Melzack R.
        • Kinch R.
        • Dobkin P.
        • Lebrun M.
        • Taenzer P.
        Severity of labour pain: Influence of physical as well as psychologic variables.
        Can Med Assoc J. 1984; 130: 579-584
        • Le Coq G.
        • Ducot B.
        • Benhamou D.
        Risk factors of inadequate pain relief during epidural analgesia for labour and delivery.
        Can J Anaesth. 1998; 45: 719-723
        • Pan P.H.
        • Bogard T.D.
        • Owen M.D.
        Incidence and characteristics of failures in obstetric neuraxial analgesia and anesthesia: a retrospective analysis of 19,259 deliveries.
        Int J Obstet Anesth. 2004; 13: 227-233
        • Sng B.L.
        • Zhang Q.
        • Leong W.L.
        • Ocampo C.
        • Assam P.N.
        • Sia A.T.
        Incidence and characteristics of breakthrough pain in parturients using computer-integrated patient-controlled epidural analgesia.
        J Clin Anesth. 2015; 27: 277-284
        • Khalilia M.
        • Chakraborty S.
        • Popescu M.
        Predicting disease risks from highly imbalanced data using random forest.
        BMC Med Inform Decis Mak. 2011; 11: 51
        • Liu N.
        • Koh Z.X.
        • Chua E.C.
        • et al.
        Risk scoring for prediction of acute cardiac complications from imbalanced clinical data.
        IEEE J Biomed Health Inform. 2014; 18: 1894-1902
        • Oh S.
        • Lee M.S.
        • Zhang B.T.
        Ensemble learning with active example selection for imbalanced biomedical data classification.
        IEEE/ACM Trans Comput Biol Bioinform. 2011; 8: 316-325
        • Handelman G.S.
        • Kok H.K.
        • Chandra R.V.
        • Razavi A.H.
        • Lee M.J.
        • Asadi H.
        Edoctor: Machine learning and the future of medicine.
        J Intern Med. 2018; 284: 603-619
        • Cruz J.A.
        • Wishart D.S.
        Applications of machine learning in cancer prediction and prognosis.
        Cancer Inform. 2007; 2: 59-77
        • Menden M.P.
        • Iorio F.
        • Garnett M.
        • et al.
        Machine learning prediction of cancer cell sensitivity to drugs based on genomic and chemical properties.
        PLoS ONE. 2013; 8e61318
        • Guan W.J.
        • Jiang M.
        • Gao Y.H.
        • et al.
        Unsupervised learning technique identifies bronchiectasis phenotypes with distinct clinical characteristics.
        Int J Tuberc Lung Dis. 2016; 20: 402-410
        • Howard R.
        • Rattray M.
        • Prosperi M.
        • Custovic A.
        Distinguishing asthma phenotypes using machine learning approaches.
        Curr Allergy Asthma Rep. 2015; 15: 38
        • Weng S.F.
        • Reps J.
        • Kai J.
        • Garibaldi J.M.
        • Qureshi N.
        Can machine-learning improve cardiovascular risk prediction using routine clinical data?.
        PLoS ONE. 2017; 12e0174944
        • Liu N.
        • Koh Z.X.
        • Goh J.
        • et al.
        Prediction of adverse cardiac events in emergency department patients with chest pain using machine learning for variable selection.
        BMC Med Inform Decis Mak. 2014; 14: 75
        • Liu N.
        • Lin Z.
        • Cao J.
        • et al.
        An intelligent scoring system and its application to cardiac arrest prediction.
        IEEE Trans Inf Technol Biomed. 2012; 16: 1324-1331
        • Collins G.S.
        • Reitsma J.B.
        • Altman D.G.
        • Moons K.G.
        Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (tripod).
        Ann Intern Med. 2015; 162: 735-736
        • Breiman L.
        Random forests.
        Machine Learning. 2001; 45: 5-32
      1. Chen T, Xgboost, G.C. A scalable tree boosting system, Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016;785-94.

      2. Louppe G, Wehenkel L, Sutera A, Geurts P. Understanding variable importances in forests of randomized trees. Adv Neural Inf Process Systems 2013;431-9.

        • Nattino G.
        • Finazzi S.
        • Bertolini G.
        A new test and graphical tool to assess the goodness of fit of logistic regression models.
        Stat Med. 2016; 35: 709-720
        • Christodoulou E.
        • Ma J.
        • Collins G.S.
        • Steyerberg E.W.
        • Verbakel J.Y.
        • Van Calster B.
        A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models.
        J Clin Epidemiol. 2019; 110: 12-22
        • Desai R.J.
        • Wang S.V.
        • Vaduganathan M.
        • Evers T.
        • Schneeweiss S.
        Comparison of machine learning methods with traditional models for use of administrative claims with electronic medical records to predict heart failure outcomes.
        JAMA Netw Open. 2020; 3e1918962
      3. Pua YH, Kang H, Thumboo J, et al. Machine learning methods are comparable to logistic regression techniques in predicting severe walking limitation following total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 2020;28:3207–16.

        • Heesen M.
        • Van de Velde M.
        • Klohr S.
        Meta-analysis of the success of block following combined spinal-epidural vs epidural analgesia during labour.
        Anaesthesia. 2014; 69: 64-71
        • Panni M.K.
        • Segal S.
        Local anesthetic requirements are greater in dystocia than in normal labor.
        Anesthesiology. 2003; 98: 957-963
        • Selin L.
        • Wallin G.
        • Berg M.
        Dystocia in labour - risk factors, management and outcome: a retrospective observational study in a swedish setting.
        Acta Obstet Gynecol Scand. 2008; 87: 216-221
        • Ben-Haroush A.
        • Yogev Y.
        • Bar J.
        • Glickman H.
        • Kaplan B.
        • Hod M.
        Indicated labor induction with vaginal prostaglandin e2 increases the risk of cesarean section even in multiparous women with no previous cesarean section.
        J Perinat Med. 2004; 32: 31-36
        • Tan H.S.
        • Sng B.L.
        • Sia A.T.H.
        Reducing breakthrough pain during labour epidural analgesia: an update.
        Curr Opin Anaesthesiol. 2019; 32: 307-314
        • Gambling D.
        • Berkowitz J.
        • Farrell T.R.
        • Pue A.
        • Shay D.
        A randomized controlled comparison of epidural analgesia and combined spinal-epidural analgesia in a private practice setting: pain scores during first and second stages of labor and at delivery.
        Anesth Analg. 2013; 116: 636-643
        • D'Angelo R.
        New techniques for labor analgesia: PCEA and CSE.
        Clin Obstet Gynecol. 2003; 46: 623-632
        • Haydon M.L.
        • Larson D.
        • Reed E.
        • Shrivastava V.K.
        • Preslicka C.W.
        • Nageotte M.P.
        Obstetric outcomes and maternal satisfaction in nulliparous women using patient-controlled epidural analgesia.
        Am J Obstet Gynecol. 2011; 205: e271-e276
        • van der Vyver M.
        • Halpern S.
        • Joseph G.
        Patient-controlled epidural analgesia versus continuous infusion for labour analgesia: a meta-analysis.
        Br J Anaesth. 2002; 89: 459-465
        • Sng B.L.
        • Zeng Y.
        • de Souza N.N.A.
        • et al.
        Automated mandatory bolus versus basal infusion for maintenance of epidural analgesia in labour.
        Cochrane Database Syst Rev. 2018; 5: CD011344
        • Boogmans T.
        • Vertommen J.
        • Valkenborgh T.
        • Devroe S.
        • Roofthooft E.
        • Van de Velde M.
        Epidural neostigmine and clonidine improves the quality of combined spinal epidural analgesia in labour: a randomised, double-blind controlled trial.
        Eur J Anaesthesiol. 2014; 31: 190-196
        • Goodman P.
        • Mackey M.C.
        • Tavakoli A.S.
        Factors related to childbirth satisfaction.
        J Adv Nurs. 2004; 46: 212-219