Abstract
Purpose
Surgical treatment of herniated lumbar intervertebral disks is a common procedure worldwide. However, recurrent herniated nucleus pulposus (re-HNP) may develop, complicating outcomes and patient management. The purpose of this study was to utilize machine-learning (ML) analytics to predict lumbar re-HNP, whereby a personalized risk prediction can be developed as a clinical tool.
Methods
A retrospective, single center study was conducted of 2630 consecutive patients that underwent lumbar microdiscectomy (mean follow-up: 22-months). Various preoperative patient pain/disability/functional profiles, imaging parameters, and anthropomorphic/demographic metrics were noted. An Extreme Gradient Boost (XGBoost) classifier was implemented to develop a predictive model identifying patients at risk for re-HNP. The model was exported to a web application software for clinical utility.
Results
There were 1608 males and 1022 females, 114 of whom experienced re-HNP. Primary herniations were central (65.8%), paracentral (17.6%), and far lateral (17.1%). The XGBoost algorithm identified multiple re-HNP predictors and was incorporated into an open-access web application software, identifying patients at low or high risk for re-HNP. Preoperative VAS leg, disability, alignment parameters, elevated body mass index, symptom duration, and age were the strongest predictors.
Conclusions
Our predictive modeling via an ML approach of our large-scale cohort is the first study, to our knowledge, that has identified significant risk factors for the development of re-HNP after initial lumbar decompression. We developed the re-herniation after decompression (RAD) profile index that has been translated into an online screening tool to identify low–high risk patients for re-HNP. Additional validation is needed for potential global implementation.
Similar content being viewed by others
References
Papagiannis GI, Triantafyllou AI, Konstantina YG et al (2019) Biomechanical factors could affect lumbar disc reherniation after microdiscectomy. J Orthop Sports Med 1:46–50
Weinstein JN, Lurie JD, Tosteson TD et al (2006) Surgical vs nonoperative treatment for lumbar disk herniation: the Spine patient outcomes research trial (SPORT) observational cohort. JAMA 296:2451–2459
Pinto RZ, Maher CG, Ferreira ML et al (2012) Drugs for relief of pain in patients with sciatica: systematic review and meta-analysis. BMJ 344:e497
Chou R, Atlas SJ, Stanos SP, Rosenquist RW (2009) Nonsurgical interventional therapies for low back pain: a review of the evidence for an American Pain Society clinical practice guideline. Spine 34:1078–1093
Radcliff K, Hilibrand A, Lurie JD et al (2012) The impact of epidural steroid injections on the outcomes of patients treated for lumbar disc herniation: a subgroup analysis of the SPORT trial. J Bone Joint Surg Am 94:1353
Virk SS, Diwan A, Phillips FM et al (2017) What is the rate of revision discectomies after primary discectomy on a national scale? Clin Orthop Relat Res 475:2752–2762
Weinstein JN, Tosteson TD, Lurie JD et al (2006) Surgical vs nonoperative treatment for lumbar disk herniation: the Spine patient outcomes research trial (SPORT): a randomized trial. JAMA 296:2441–2450
Grasso G, Torregrossa F, Landi A (2019) Prevention of lumbar reherniation by the intraoperative use of a radiofrequency bipolar device: a case-control study. J Craniovertebr Junction Spine 10:94–99
Thomé C, Klassen PD, Bouma GJ et al (2018) Annular closure in lumbar microdiscectomy for prevention of reherniation: a randomized clinical trial. Spine J 18:2278–2287
Carragee EJ, Han MY, Suen PW, Kim D (2003) Clinical outcomes after lumbar discectomy for sciatica: the effects of fragment type and anular competence. J Bone Joint Surg Am 85:102–108
Wera GD, Marcus RE, Ghanayem AJ, Bohlman HH (2008) Failure within one year following subtotal lumbar discectomy. J Bone Joint Surg Am 90:10–15
Rogers LA (1988) Experience with limited versus extensive disc removal in patients undergoing microsurgical operations for ruptured lumbar discs. Neurosurg 22:82–85
O’Sullivan MG, Connolly AE, Buckley TF (1990) Recurrent lumbar disc protrusion. Br J Neurosurg 4:319–325
Lebow RL, Adogwa O, Parker SL et al (2011) Asymptomatic same-site recurrent disc herniation after lumbar discectomy: results of a prospective longitudinal study with 2-year serial imaging. Spine 36:2147–2151
Ebeling U, Kalbarcyk H, Reulen HJ (1989) Microsurgical reoperation following lumbar disc surgery. timing, surgical findings, and outcome in 92 patients. J Neurosurg 70:397–404
Shimia M, Babaei-Ghazani A, Sadat BE et al (2013) Risk factors of recurrent lumbar disk herniation. Asian J Neurosurg 8:93–96
Ambrossi GLG, McGirt MJ, Sciubba DM et al (2009) Recurrent lumbar disc herniation after single-level lumbar discectomy: incidence and health care cost analysis. Neurosurg 65:574–578 (discussion 578)
Shepard N, Cho W (2019) Recurrent lumbar disc herniation: a review. Global Spine J 9:202–209
Parker SL, Grahovac G, Vukas D et al (2013) Cost savings associated with prevention of recurrent lumbar disc herniation with a novel annular closure device: a multicenter prospective cohort study. J Neurol Surg A Cent Eur Neurosurg 74:285–289
Belykh E, Krutko AV, Baykov ES et al (2017) Preoperative estimation of disc herniation recurrence after microdiscectomy: predictive value of a multivariate model based on radiographic parameters. Spine J 17:390–400
Mallow GM, Siyaji ZK, Galbusera F, et al (2020) Intelligence-based spine care model: a new era of research and clinical decision-making. Global Spine J 2192568220973984
Chen T, Guestrin C (2016). XGBoost: A Scalable Tree Boosting System. In: proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining. Association for Computing Machinery, New York, NY, USA, pp 785–794
Pedregosa F, Varoquaux G, Gramfort A et al (2011) Scikit-learn: machine learning in Python. The J Mach Learn Res. 12:2825–2830
Thio QCBS, Karhade AV, Ogink PT et al (2018) Can machine-learning techniques be used for 5-year survival prediction of patients with chondrosarcoma? Clin Orthop Relat Res 476:2040–2048
Karhade AV, Thio QCBS, Ogink PT et al (2019) Predicting 90-Day and 1-year mortality in spinal metastatic disease: development and internal validation. Neurosurg 85:E671–E681
Thio QCBS, Karhade AV, Ogink PT et al (2020) Development and internal validation of machine learning algorithms for preoperative survival prediction of extremity metastatic disease. Clin Orthop Relat Res 478:322–333
Steyerberg EW, Vergouwe Y (2014) Towards better clinical prediction models: seven steps for development and an ABCD for validation. Eur Heart J 35:1925–1931
Lipton ZC, Elkan C, Narayanaswamy B (2014) Thresholding Classifiers to Maximize F1 Score. arXiv [stat.ML]
Breiman L (2001) Random forests. Mach Learn 45:5–32
Rodríguez-Pérez R, Bajorath J (2020) Interpretation of machine learning models using shapley values: application to compound potency and multi-target activity predictions. J Comput Aided Mol Des 34:1013–1026
Lundberg S, Lee S-I (2017) A unified approach to interpreting model predictions. arXiv [cs.AI]
Kim JM, Lee SH, Ahn Y et al (2007) Recurrence after successful percutaneous endoscopic lumbar discectomy. Minim Invasive Neurosurg 50:82–85
Staartjes VE, de Wispelaere MP, Vandertop WP, Schröder ML (2019) Deep learning-based preoperative predictive analytics for patient-reported outcomes following lumbar discectomy: feasibility of center-specific modeling. Spine J 19:853–861
Chun S-W, Lim C-Y, Kim K et al (2017) The relationships between low back pain and lumbar lordosis: a systematic review and meta-analysis. Spine J 17:1180–1191
Yao Y, Liu H, Zhang H et al (2017) Risk factors for recurrent herniation after percutaneous endoscopic lumbar Discectomy. World Neurosurg 100:1–6
Meredith DS, Huang RC, Nguyen J, Lyman S (2010) Obesity increases the risk of recurrent herniated nucleus pulposus after lumbar microdiscectomy. Spine J 10:575–580
Callaghan JP, McGill SM (2001) Intervertebral disc herniation: studies on a porcine model exposed to highly repetitive flexion/extension motion with compressive force. Clin Biomech 16:28–37
Wilder DG, Pope MH, Frymoyer JW (1988) The biomechanics of lumbar disc herniation and the effect of overload and instability. J Spinal Disord 1:16–32
Ghezelbash F, Shirazi-Adl A, Plamondon A et al (2017) Obesity and obesity shape markedly influence Spine biomechanics: a subject-specific risk assessment model. Ann Biomed Eng 45:2373–2382
Moliterno JA, Knopman J, Parikh K et al (2010) Results and risk factors for recurrence following single-level tubular lumbar microdiscectomy. J Neurosurg Spine 12:680–686
Kara B, Tulum Z, Acar U (2005) Functional results and the risk factors of reoperations after lumbar disc surgery. Eur Spine J 14:43–48
Huang W, Han Z, Liu J et al (2016) Risk factors for recurrent lumbar Disc herniation: a systematic review and meta-analysis. Medicine 95(2):e2378
Siccoli A, Schröder ML, Staartjes VE (2020) Association of age with incidence and timing of recurrence after microdiscectomy for lumbar disc herniation. Eur Spine J. https://doi.org/10.1007/s00586-020-06692-1
Abdu RW, Abdu WA, Pearson AM et al (2017) Reoperation for recurrent intervertebral disc herniation in the Spine patient outcomes research trial: analysis of rate, risk factors, and outcome. Spine 42:1106–1114
Wilke H-J, Ressel L, Heuer F et al (2013) Can prevention of a reherniation be investigated? Establishment of a herniation model and experiments with an anular closure device. Spine 38:E587–E593
Miwa S, Yokogawa A, Kobayashi T et al (2015) Risk factors of recurrent lumbar disk herniation: a single center study and review of the literature. J Spinal Disord Tech 28:E265–E269
Saftić R, Grgić M, Ebling B, Splavski B (2006) Case-control study of risk factors for lumbar intervertebral disc herniation in croatian island populations. Croat Med J 47:593–600
Rihn JA, Hilibrand AS, Radcliff K et al (2011) Duration of symptoms resulting from lumbar disc herniation: effect on treatment outcomes: analysis of the Spine patient outcomes research trial (SPORT). J Bone Joint Surg Am 93:1906–1914
Movassaghi K, Basques BA, Louie PK et al (2019) The Duration of symptoms does not impact clinical outcomes following lumbar decompression surgery. Spine 44:305–308
Ahsan K, Najmus-Sakeb HA et al (2012) Discectomy for primary and recurrent prolapse of lumbar intervertebral discs. J Orthop Surg 20:7–10
Cinotti G, Roysam GS, Eisenstein SM, Postacchini F (1998) Ipsilateral recurrent lumbar disc herniation. a prospective, controlled study. J Bone Joint Surg Br 80:825–832
Morgan-Hough CVJ, Jones PW, Eisenstein SM (2003) Primary and revision lumbar discectomy. a 16-year review from one centre. J Bone Joint Surg Br 85:871–874
Protopsaltis TS, Lafage R, Smith JS et al (2018) The lumbar pelvic angle, the lumbar component of the T1 pelvic angle, correlates with HRQOL, PI-LL mismatch, and it predicts global alignment. Spine 43:681–687
Rothenfluh DA, Mueller DA, Rothenfluh E, Min K (2015) Pelvic incidence-lumbar lordosis mismatch predisposes to adjacent segment disease after lumbar spinal fusion. Eur Spine J 24:1251–1258
Obeid I, Bourghli A, Boissière L et al (2014) Complex osteotomies vertebral column resection and decancellation. Eur J Orthop Surg Traumatol 24(Suppl 1):S49-57
Obeid I, Boissière L, Vital J-M, Bourghli A (2015) Osteotomy of the spine for multifocal deformities. Eur Spine J 24(Suppl 1):S83-92
Chang H-K, Chang H-C, Wu J-C et al (2016) Scoliosis may increase the risk of recurrence of lumbar disc herniation after microdiscectomy. J Neurosurg Spine 24:586–591
Wu W, Chen Y, Yu L et al (2019) Coronal and sagittal spinal alignment in lumbar disc herniation with scoliosis and trunk shift. J Orthop Surg Res 14:264
Endo K, Suzuki H, Tanaka H et al (2010) Sagittal spinal alignment in patients with lumbar disc herniation. Eur Spine J 19:435–438
Obeid I, Berjano P, Lamartina C et al (2019) Classification of coronal imbalance in adult scoliosis and spine deformity: a treatment-oriented guideline. Eur Spine J 28:94–113
Matsui H, Ohmori K, Kanamori M et al (1998) Significance of sciatic scoliotic list in operated patients with lumbar disc herniation. Spine 23:338–342
Beretta L, Santaniello A (2016) Nearest neighbor imputation algorithms: a critical evaluation. BMC Med Inform Decis Mak 16(Suppl 3):74
Batista GE, Bazzan ALC, Monard MC (2003). Balancing training data for automated annotation of keywords: a case study. In: WOB, pp 10–18
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interests
The authors have no competing or financial conflicts of interest in relation to this work.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Harada, G.K., Siyaji, Z.K., Mallow, G.M. et al. Artificial intelligence predicts disk re-herniation following lumbar microdiscectomy: development of the “RAD” risk profile. Eur Spine J 30, 2167–2175 (2021). https://doi.org/10.1007/s00586-021-06866-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00586-021-06866-5