Advertisement

Development and Internal Validation of Supervised Machine Learning Algorithms for Predicting Clinically Significant Functional Improvement in a Mixed Population of Primary Hip Arthroscopy

Published:January 15, 2021DOI:https://doi.org/10.1016/j.arthro.2021.01.005

      Purpose

      To (1) develop and validate a machine learning algorithm to predict clinically significant functional improvements after hip arthroscopy for femoroacetabular impingement syndrome and to (2) develop a digital application capable of providing patients with individual risk profiles to determine their propensity to gain clinically significant improvements in function.

      Methods

      A retrospective review of consecutive hip arthroscopy patients who underwent cam/pincer correction, labral preservation, and capsular closure between January 2012 and 2017 from 1 large academic and 3 community hospitals operated on by a single high-volume hip arthroscopist was performed. The primary outcome was the minimal clinically important difference (MCID) for the Hip Outcome Score (HOS)–Activities of Daily Living (ADL) at 2 years postoperatively, which was calculated using a distribution-based method. A total of 21 demographic, radiographic, and patient-reported outcome measures were considered as potential covariates. An 80:20 random split was used to create training and testing sets from the patient cohort. Five supervised machine learning algorithms were developed using 3 iterations of 10-fold cross-validation on the training set and assessed by discrimination, calibration, Brier score, and decision curve analysis on an independent testing set of patients.

      Results

      A total of 818 patients with a median (interquartile range) age of 32.0 (22.0-42.0) and 69.2% female were included, of whom 74.3% achieved the MCID for the HOS-ADL. The best-performing algorithm was the stochastic gradient boosting model (c-statistic = 0.84, calibration intercept = 0.20, calibration slope = 0.83, and Brier score = 0.13). Of the initial 21 candidate variables, the 8 most important features for predicting the MCID for the HOS-ADL included in model training were body mass index, age, preoperative HOS-ADL score, preoperative pain level, sex, Tönnis grade, symptom duration, and drug allergies. The algorithm was subsequently transformed into a digital application using local explanations to provide customized risk assessment: https://orthoapps.shinyapps.io/HPRG_ADL/.

      Conclusions

      The stochastic boosting gradient model conferred excellent predictive ability for propensity to gain clinically significant improvements in function after hip arthroscopy. An open-access digital application was created, which may augment shared decision-making and allow for preoperative risk stratification. External validation of this model is warranted to confirm the performance of these algorithms, as the generalizability is currently unknown.

      Level of Evidence

      IV, Case series.
      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 Arthroscopy
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect

      References

        • Novikov D.
        • Cizmic Z.
        • Feng J.E.
        • Iorio R.
        • Meftah M.
        The historical development of value-based care: How we got here.
        J Bone Joint Surg Am. 2018; 100: e144
        • Nwachukwu B.U.
        • Hamid K.S.
        • Bozic K.J.
        Measuring value in orthopaedic surgery.
        JBJS Rev. 2013; 1
        • Ring D.
        • Bozic K.J.
        Value-based healthcare: The value of considering patient preferences and circumstances in orthopaedic surgery.
        Clin Orthop Relat Res. 2016; 474: 633-635
        • McCreary D.L.
        • Dugarte A.J.
        • Vang S.
        • et al.
        Patient-level value analysis: An innovative approach to optimize care delivery.
        J Orthop Trauma. 2019; 33: S49-S52
        • Winegar A.L.
        • Moxham J.
        • Erlinger T.P.
        • Bozic K.J.
        Value-based healthcare: Measuring what matters-engaging surgeons to make measures meaningful and improve clinical practice.
        Clin Orthop Relat Res. 2018; 476: 1704-1706
        • Nwachukwu B.U.
        • Beck E.C.
        • Kunze K.N.
        • Chahla J.
        • Rasio J.
        • Nho S.J.
        Defining the clinically meaningful outcomes for arthroscopic treatment of femoroacetabular impingement syndrome at minimum 5-year follow-up.
        Am J Sports Med. 2020; 48: 901-907
        • Flores S.E.
        • Sheridan J.R.
        • Borak K.R.
        • Zhang A.L.
        When do patients improve after hip arthroscopy for femoroacetabular impingement? A prospective cohort analysis.
        Am J Sports Med. 2018; 46: 3111-3118
        • Levy D.M.
        • Kuhns B.D.
        • Chahal J.
        • Philippon M.J.
        • Kelly B.T.
        • Nho S.J.
        Hip arthroscopy outcomes with respect to patient acceptable symptomatic state and minimal clinically important difference.
        Arthroscopy. 2016; 32: 1877-1886
        • Nwachukwu B.U.
        • Chang B.
        • Adjei J.
        • et al.
        Time required to achieve minimal clinically important difference and substantial clinical benefit after arthroscopic treatment of femoroacetabular impingement.
        Am J Sports Med. 2018; 46: 2601-2606
        • Beck E.C.
        • Nwachukwu B.U.
        • Kunze K.N.
        • Chahla J.
        • Nho S.J.
        How can we define clinically important improvement in pain scores after hip arthroscopy for femoroacetabular impingement syndrome? Minimum 2-year follow-up study.
        Am J Sports Med. 2019; 47: 3133-3140
        • Nwachukwu B.U.
        • Chang B.
        • Fields K.
        • et al.
        Defining the "substantial clinical benefit" after arthroscopic treatment of femoroacetabular impingement.
        Am J Sports Med. 2017; 45: 1297-1303
        • Martin R.L.
        • Philippon M.J.
        Evidence of validity for the hip outcome score in hip arthroscopy.
        Arthroscopy. 2007; 23: 822-826
        • Martin R.L.
        • Philippon M.J.
        Evidence of reliability and responsiveness for the hip outcome score.
        Arthroscopy. 2008; 24: 676-682
        • Kunze K.N.
        • Leong N.L.
        • Beck E.C.
        • Bush-Joseph C.A.
        • Nho S.J.
        Hip arthroscopy for femoroacetabular impingement improves sleep quality postoperatively.
        Arthroscopy. 2019; 35: 461-469
        • Cancienne J.
        • Kunze K.N.
        • Beck E.C.
        • Chahla J.
        • Suppauksorn S.
        • Nho S.J.
        Influence of cigarette smoking at the time of surgery on postoperative outcomes in patients with femoroacetabular impingement: A matched-pair cohort analysis.
        Am J Sports Med. 2019; 47: 1138-1144
        • Kunze K.N.
        • Beck E.C.
        • Nwachukwu B.U.
        • Ahn J.
        • Nho S.J.
        Early hip arthroscopy for femoroacetabular impingement syndrome provides superior outcomes when compared with delaying surgical treatment beyond 6 months.
        Am J Sports Med. 2019; : 2038-2044
        • Krishnamoorthy V.
        • Kunze K.
        • Beck E.
        • et al.
        Radiographic prevalence of symphysis pubis abnormalities and clinical outcomes in patients with femoroacetabular impingement syndrome.
        Am J Sports Med. 2019; 47: 1467-1472
        • Chandrasekaran S.
        • Darwish N.
        • Close M.R.
        • Suarez-Ahedo C.
        • Lodhia P.
        • Domb B.G.
        Minimum 2-year outcomes of arthroscopic management of symptomatic hip labrum tears in patients with global acetabular overcoverage.
        Arthroscopy. 2017; 33: 1514-1520
        • Frank R.M.
        • Kunze K.N.
        • Beck E.C.
        • Neal W.H.
        • Bush-Joseph C.A.
        • Nho S.J.
        Do female athletes return to sports after hip preservation surgery for femoroacetabular impingement syndrome? A comparative analysis.
        Orthop J Sports Med. 2019; 7 (2325967119831758-2325967119831758)
        • Ramkumar P.N.
        • Haeberle H.S.
        • Bloomfield M.R.
        • et al.
        Artificial intelligence and arthroplasty at a single institution: Real-world applications of machine learning to big data, value-based care, mobile health, and remote patient monitoring.
        J Arthroplasty. 2019; 34: 2204-2209
        • Helm J.M.
        • Swiergosz A.M.
        • Haeberle H.S.
        • et al.
        Machine learning and artificial intelligence: Definitions, applications, and future directions.
        Curr Rev Musculoskelet Med. 2020; 13: 69-76
        • Nwachukwu B.U.
        • Beck E.C.
        • Lee E.K.
        • et al.
        Application of machine learning for predicting clinically meaningful outcome after arthroscopic femoroacetabular impingement surgery.
        Am J Sports Med. 2020; 48: 415-423
        • 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): The TRIPOD Statement.
        Br J Surg. 2015; 102: 148-158
        • Luo W.
        • Phung D.
        • Tran T.
        • et al.
        Guidelines for developing and reporting machine learning predictive models in biomedical research: A multidisciplinary view.
        J Med Internet Res. 2016; 18: e323
        • Copay A.G.
        • Eyberg B.
        • Chung A.S.
        • Zurcher K.S.
        • Chutkan N.
        • Spangehl M.J.
        Minimum clinically important difference: Current trends in the orthopaedic literature, part II: Lower extremity: A systematic review.
        JBJS Rev. 2018; 6: e2
        • Hardt J.
        • Herke M.
        • Leonhart R.
        Auxiliary variables in multiple imputation in regression with missing X: A warning against including too many in small sample research.
        BMC Med Res Methodol. 2012; 12: 184
        • Karhade A.V.
        • Shah A.A.
        • Bono C.M.
        • et al.
        Development of machine learning algorithms for prediction of mortality in spinal epidural abscess.
        Spine J. 2019; 19: 1950-1959
        • De Silva A.P.
        • Moreno-Betancur M.
        • De Livera A.M.
        • Lee K.J.
        • Simpson J.A.
        Multiple imputation methods for handling missing values in a longitudinal categorical variable with restrictions on transitions over time: A simulation study.
        BMC Med Res Methodol. 2019; 19: 14
        • Resche-Rigon M.
        • White I.R.
        Multiple imputation by chained equations for systematically and sporadically missing multilevel data.
        Stat Methods Med Res. 2018; 27: 1634-1649
        • Lee K.J.
        • Carlin J.B.
        Multiple imputation in the presence of non-normal data.
        Stat Med. 2017; 36: 606-617
        • Guyon I.
        • Weston J.
        • Barnhill S.
        • Vapnik V.
        Gene selection for cancer classification using support vector machines.
        Machine Learning. 2002; 46: 389-422
        • Kunze K.N.
        • Karhade A.V.
        • Sadauskas A.J.
        • Schwab J.H.
        • Levine B.R.
        Development of machine learning algorithms to predict clinically meaningful improvement for the patient-reported health state after total hip arthroplasty.
        J Arthroplasty. 2020; 35: 2119-2123
        • Steyerberg E.W.
        • Vergouwe Y.
        Towards better clinical prediction models: Seven steps for development and an ABCD for validation.
        Eur Heart J. 2014; 35: 1925-1931
        • Steyerberg E.W.
        • Vickers A.J.
        • Cook N.R.
        • et al.
        Assessing the performance of prediction models: A framework for traditional and novel measures.
        Epidemiology. 2010; 21: 128-138
        • Van Calster B.
        • McLernon D.J.
        • van Smeden M.
        • Wynants L.
        • Steyerberg E.W.
        • TGEdtapmotS initiative
        Calibration: The Achilles heel of predictive analytics.
        BMC Med. 2019; 17: 230
        • Brier G.W.
        • Allen R.A.
        Verification of weather forecasts.
        in: Malone T.F. Compendium of meterology. American Meteorological Society, Boston, MA1951: 841-848
        • Vickers A.J.
        • Elkin E.B.
        Decision curve analysis: A novel method for evaluating prediction models.
        Med Decis Making. 2006; 26: 565-574
        • Vickers A.J.
        • Cronin A.M.
        • Elkin E.B.
        • Gonen M.
        Extensions to decision curve analysis, a novel method for evaluating diagnostic tests, prediction models and molecular markers.
        BMC Med Inform Decis Mak. 2008; 8: 53
        • Ribeiro M.T.
        • Singh S.
        • Guestrin C.
        Model-agnostic interpretability of machine learning.
        Cornell University, 2019
        • Myers T.G.
        • Ramkumar P.N.
        • Ricciardi B.F.
        • Urish K.L.
        • Kipper J.
        • Ketonis C.
        Artificial intelligence and orthopaedics: An introduction for clinicians.
        J Bone Joint Surg Am. 2020;
        • Karnuta J.M.
        • Golubovsky J.L.
        • Haeberle H.S.
        • et al.
        Can a machine learning model accurately predict patient resource utilization following lumbar spinal fusion?.
        Spine J. 2020; 20: 329-336
        • Thio Q.
        • Karhade A.V.
        • Ogink P.T.
        • et al.
        Can machine-learning techniques be used for 5-year survival prediction of patients with chondrosarcoma?.
        Clin Orthop Relat Res. 2018; 476: 2040-2048
        • Karnuta J.M.
        • Navarro S.M.
        • Haeberle H.S.
        • Billow D.G.
        • Krebs V.E.
        • Ramkumar P.N.
        Bundled care for hip fractures: A machine-learning approach to an untenable patient-specific payment model.
        J Orthop Trauma. 2019; 33: 324-330
        • Navarro S.M.
        • Wang E.Y.
        • Haeberle H.S.
        • et al.
        Machine learning and primary total knee arthroplasty: Patient forecasting for a patient-specific payment model.
        J Arthroplasty. 2018; 33: 3617-3623
        • Ramkumar P.N.
        • Navarro S.M.
        • Haeberle H.S.
        • et al.
        Development and validation of a machine learning algorithm after primary total hip arthroplasty: Applications to length of stay and payment models.
        J Arthroplasty. 2019; 34: 632-637
        • Santiso S.
        • Casillas A.
        • Perez A.
        The class imbalance problem detecting adverse drug reactions in electronic health records.
        Health Informatics J. 2019; 25: 1768-1778
        • Mirza B.
        • Wang W.
        • Wang J.
        • Choi H.
        • Chung N.C.
        • Ping P.
        Machine learning and integrative analysis of biomedical big data.
        Genes (Basel). 2019; 10
        • Cvetanovich G.L.
        • Weber A.E.
        • Kuhns B.D.
        • et al.
        Hip arthroscopic surgery for femoroacetabular impingement with capsular management: Factors associated with achieving clinically significant outcomes.
        Am J Sports Med. 2018; 46: 288-296
        • Kunze K.N.
        • Nwachukwu B.U.
        • Beck E.C.
        • et al.
        Preoperative duration of symptoms is associated with outcomes 5 years after hip arthroscopy for femoroacetabular impingement syndrome.
        Arthroscopy. 2020; 36: 1022-1029
        • Leong N.L.
        • Clapp I.M.
        • Neal W.H.
        • Beck E.
        • Bush-Joseph C.A.
        • Nho S.J.
        The influence of pain in other major joints and the spine on 2-year outcomes after hip arthroscopy.
        Arthroscopy. 2018; 34: 3196-3201
        • Beck E.C.
        • Nwachukwu B.U.
        • Chahla J.
        • et al.
        Patients with borderline hip dysplasia achieve clinically significant outcome after arthroscopic femoroacetabular impingement surgery: A case-control study with minimum 2-year follow-up.
        Am J Sports Med. 2019; 47: 2636-2645
        • Sochacki K.R.
        • Jack 2nd, R.A.
        • Bekhradi A.
        • Delgado D.
        • McCulloch P.C.
        • Harris J.D.
        Are self-reported medication allergies associated with worse hip outcome scores prior to hip arthroscopy?.
        Arthroscopy. 2018; 34: 1856-1861
        • Lansdown D.A.
        • Ukwuani G.
        • Kuhns B.
        • Harris J.D.
        • Nho S.J.
        Self-reported mental disorders negatively influence surgical outcomes after arthroscopic treatment of femoroacetabular impingement.
        Orthop J Sports Med. 2018; 6 (2325967118773312)
        • Liljensoe A.
        • Laursen J.O.
        • Bliddal H.
        • Soballe K.
        • Mechlenburg I.
        Weight loss intervention before total knee replacement: A 12-month randomized controlled trial.
        Scand J Surg. 2019; (1457496919883812)
        • Li J.T.
        • Markus P.J.
        • Osmon D.R.
        • Estes L.
        • Gosselin V.A.
        • Hanssen A.D.
        Reduction of vancomycin use in orthopedic patients with a history of antibiotic allergy.
        Mayo Clin Proc. 2000; 75: 902-906
        • Hacken B.
        • Rogers A.
        • Chinchilli V.
        • Silvis M.
        • Mosher T.
        • Black K.
        Improvement in knee osteoarthritis pain and function following bariatric surgery: 5-year follow-up.
        Surg Obes Relat Dis. 2019; 15: 979-984
      1. Kunze KN, Polce EM, Rasio J, Nho SJ. Machine learning algorithms predict clinically significant improvements in satisfaction after hip arthroscopy [published online December 20, 2020]. Arthroscopy. https://doi.org/10.1016/j.arthro.2020.11.027.

      2. Polce EM, Kunze KN, Fu M, et al. Development of supervised machine learning algorithms for prediction of satisfaction at two years following total shoulder arthroplasty [published online September 30, 2020]. J Shoulder Elbow Surg. https://doi.org/10.1016/j.jse.2020.09.007.

        • Kunze K.N.
        • Polce E.M.
        • Sadauskas A.J.
        • Levine B.R.
        Development of machine learning algorithms to predict patient dissatisfaction after primary total knee arthroplasty.
        J Arthroplasty. 2020; 35: 3117-3122
        • Kunze K.N.
        • Karhade A.V.
        • Sadauskas A.J.
        • Schwab J.H.
        • Levine B.R.
        Development of machine learning algorithms to predict clinically meaningful improvement for the patient-reported health state after total hip arthroplasty.
        J Arthroplasty. 2020; 35: 2119-2123
        • Karhade A.V.
        • Ogink P.T.
        • Thio Q.
        • et al.
        Development of machine learning algorithms for prediction of prolonged opioid prescription after surgery for lumbar disc herniation.
        Spine J. 2019; 19: 1764-1771
        • Karhade A.V.
        • Ogink P.T.
        • Thio Q.
        • et al.
        Machine learning for prediction of sustained opioid prescription after anterior cervical discectomy and fusion.
        Spine J. 2019; 19: 976-983