Decision tree analysis to better control treatment effects in spinal cord injury clinical research

Restricted access

OBJECTIVE

The aim of this study was to use decision tree modeling to identify optimal stratification groups considering both the neurological impairment and spinal column injury and to investigate the change in motor score as an example of a practical application. Inherent heterogeneity in spinal cord injury (SCI) introduces variation in natural recovery, compromising the ability to identify true treatment effects in clinical research. Optimized stratification factors to create homogeneous groups of participants would improve accurate identification of true treatment effects.

METHODS

The analysis cohort consisted of patients with acute traumatic SCI registered in the Vancouver Rick Hansen Spinal Cord Injury Registry (RHSCIR) between 2004 and 2014. Severity of neurological injury (American Spinal Injury Association Impairment Scale [AIS grades A–D]), level of injury (cervical, thoracic), and total motor score (TMS) were assessed using the International Standards for Neurological Classification of Spinal Cord Injury examination; morphological injury to the spinal column assessed using the AOSpine classification (AOSC types A–C, C most severe) and age were also included. Decision trees were used to determine the most homogeneous groupings of participants based on TMS at admission and discharge from in-hospital care.

RESULTS

The analysis cohort included 806 participants; 79.3% were male, and the mean age was 46.7 ± 19.9 years. Distribution of severity of neurological injury at admission was AIS grade A in 40.0% of patients, grade B in 11.3%, grade C in 18.9%, and grade D in 29.9%. The level of injury was cervical in 68.7% of patients and thoracolumbar in 31.3%. An AOSC type A injury was found in 33.1% of patients, type B in 25.6%, and type C in 37.8%. Decision tree analysis identified 6 optimal stratification groups for assessing TMS: 1) AOSC type A or B, cervical injury, and age ≤ 32 years; 2) AOSC type A or B, cervical injury, and age > 32–53 years; 3) AOSC type A or B, cervical injury, and age > 53 years; 4) AOSC type A or B and thoracic injury; 5) AOSC type C and cervical injury; and 6) AOSC type C and thoracic injury.

CONCLUSIONS

Appropriate stratification factors are fundamental to accurately identify treatment effects. Inclusion of AOSC type improves stratification, and use of the 6 stratification groups could minimize confounding effects of variable neurological recovery so that effective treatments can be identified.

ABBREVIATIONS AIS = American Spinal Injury Association Impairment Scale; AOSC = AOSpine classification; CCI = Charlson Comorbidity Index; ISNCSCI = International Standards for the Neurological Classification of Spinal Cord Injury; ISS = Injury Severity Score; RHSCIR = Rick Hansen Spinal Cord Injury Registry; SCI = spinal cord injury; TMS = total motor score.

Article Information

Correspondence Marcel F. Dvorak: University of British Columbia, Vancouver, BC, Canada. marcel.dvorak@ubc.ca.

INCLUDE WHEN CITING Published online June 14, 2019; DOI: 10.3171/2019.3.SPINE18993.

Disclosures Dr. Noonan: employee of the Rick Hansen Institute. Dr. Fisher: consultant for Medtronic and NuVasive, and royalties from Medtronic. Dr. Dea: consultant for Stryker and Baxter, and direct stock ownership in Medtronic.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Decision tree approach results with AIS grade at admission as the dependent variable and age, sex, anatomical region (cervical vs thoracic), and AOSC injury morphology types as the independent variables. Sex and mechanism of injury were also analyzed but did not improve the determination of homogeneous subgroups of injury severity using AIS grade. The 6 terminal nodes, or final classification (most homogeneous) groups, identified the final boxes for each branch of the tree (nodes 4–9). Figure is available in color online only.

  • View in gallery

    Results of the decision tree analysis, with AIS grade at admission as the dependent variable, demonstrating the 6 terminal nodes, or final classification (most homogeneous) groups, identified. Figure is available in color online only.

  • View in gallery

    Change in TMS for the final classification (most homogeneous) groups (6 nodes) identified in the decision tree modeling. The blue line (top) represents the mean TMS at discharge and the red line (bottom) the mean TMS at admission. The difference in mean TMS between admission and discharge is indicated by the green arrow for node (subgroup) 2. See Fig. 1 for details of the subgroups. Figure is available in color online only.

  • View in gallery

    Relative percentage of AOSC type (A, B, or C) by mechanism of injury. Figure is available in color online only.

References

  • 1

    Baker SPO’Neill BHaddon W JrLong WB: The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 14:1871961974

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Biering-Sørensen FDeVivo MJCharlifue SChen YNew PWNoonan V: International Spinal Cord Injury Core Data Set (version 2.0)—including standardization of reporting. Spinal Cord 55:7597642017

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Bracken MBShepard MJCollins WFHolford TRYoung WBaskin DS: A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. Results of the Second National Acute Spinal Cord Injury Study. N Engl J Med 322:140514111990

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Bracken MBShepard MJHellenbrand KGCollins WFLeo LSFreeman DF: Methylprednisolone and neurological function 1 year after spinal cord injury. Results of the National Acute Spinal Cord Injury Study. J Neurosurg 63:7047131985

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Bracken MBShepard MJHolford TRLeo-Summers LAldrich EFFazl M: Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal cord injury. Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial. National Acute Spinal Cord Injury Study. JAMA 277:159716041997

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Breiman LFriedman JHOlshen RAStone CJ: Classification and Regression Trees. Boca Raton: Chapman & Hall/CRC1984

  • 7

    Chan WMMohammed YLee IPearse DD: Effect of gender on recovery after spinal cord injury. Transl Stroke Res 4:4474612013

  • 8

    Charlson MEPompei PAles KLMacKenzie CR: A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40:3733831987

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9

    Chen YTang YAllen VDeVivo MJ: Fall-induced spinal cord injury: external causes and implications for prevention. J Spinal Cord Med 39:24312016

  • 10

    Dvorak MFNoonan VKFallah NFisher CGRivers CSAhn H: Minimizing errors in acute traumatic spinal cord injury trials by acknowledging the heterogeneity of spinal cord anatomy and injury severity: an observational Canadian cohort analysis. J Neurotrauma 31:154015472014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Elizei SSKwon BK: The translational importance of establishing biomarkers of human spinal cord injury. Neural Regen Res 12:3853882017

  • 12

    Fawcett JWCurt ASteeves JDColeman WPTuszynski MHLammertse D: Guidelines for the conduct of clinical trials for spinal cord injury as developed by the ICCP panel: spontaneous recovery after spinal cord injury and statistical power needed for therapeutic clinical trials. Spinal Cord 45:1902052007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Fehlings MGVaccaro AWilson JRSingh AW Cadotte DHarrop JS: Early versus delayed decompression for traumatic cervical spinal cord injury: results of the Surgical Timing in Acute Spinal Cord Injury Study (STASCIS). PLoS One 7:e320372012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Hastie TTibshirani RFriedman J: The Elements of Statistical Learning. Data Mining Inference and Predictioned 2. New York: Springer2011 pp 501520

    • Search Google Scholar
    • Export Citation
  • 15

    Kass GV: An exploratory technique for investigating large quantities of categorical data. Appl Stat 29:1191271980

  • 16

    Kirshblum SCWaring WBiering-Sorensen FBurns SPJohansen MSchmidt-Read M: Reference for the 2011 revision of the international standards for neurological classification of spinal cord injury. J Spinal Cord Med 34:5475542011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Krueger HNoonan VKTrenaman LMJoshi PRivers CS: The economic burden of traumatic spinal cord injury in Canada. Chronic Dis Inj Can 33:1131222013

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    Kurpad SMartin ARTetreault LAFischer DJSkelly ACMikulis D: Impact of baseline magnetic resonance imaging on neurologic, functional, and safety outcomes in patients with acute traumatic spinal cord injury. Global Spine J 7 (3 Suppl):151S174S2017

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19

    Kwon BKStammers AMTBelanger LMBernardo AChan DBishop CM: Cerebrospinal fluid inflammatory cytokines and biomarkers of injury severity in acute human spinal cord injury. J Neurotrauma 27:6696822010

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Kwon BKStreijger FFallah NNoonan VKBélanger LMRitchie L: Cerebrospinal fluid biomarkers to stratify injury severity and predict outcome in human traumatic spinal cord injury. J Neurotrauma 34:5675802017

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    Lammertse DP: Clinical trials in spinal cord injury: lessons learned on the path to translation. The 2011 International Spinal Cord Society Sir Ludwig Guttmann Lecture. Spinal Cord 51:292013

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Martin ARDe Leener BCohen-Adad JCadotte DWKalsi-Ryan SLange SF: A novel MRI biomarker of spinal cord white matter injury: T2*-weighted white matter to gray matter signal intensity ratio. AJNR Am J Neuroradiol 38:126612732017

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Noonan VKFingas MFarry ABaxter DSingh AFehlings MG: Incidence and prevalence of spinal cord injury in Canada: a national perspective. Neuroepidemiology 38:2192262012

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24

    Noonan VKKwon BKSoril LFehlings MGHurlbert RJTownson A: The Rick Hansen Spinal Cord Injury Registry (RHSCIR): a national patient-registry. Spinal Cord 50:22272012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Paquet JRivers CSKurban DFinkelstein JTee JWNoonan VK: The impact of spine stability on cervical spinal cord injury with respect to demographics, management, and outcome: a prospective cohort from a national spinal cord injury registry. Spine J 18:88982018

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Sekhon LHFehlings MG: Epidemiology, demographics, and pathophysiology of acute spinal cord injury. Spine (Phila Pa 1976) 26 (24 Suppl):S2S122001

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27

    Sipski MLJackson ABGómez-Marín OEstores IStein A: Effects of gender on neurologic and functional recovery after spinal cord injury. Arch Phys Med Rehabil 85:182618362004

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Steeves JDLammertse DCurt AFawcett JWTuszynski MHDitunno JF: Guidelines for the conduct of clinical trials for spinal cord injury (SCI) as developed by the ICCP panel: clinical trial outcome measures. Spinal Cord 45:2062212007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Tanadini LGHothorn TJones LALammertse DPAbel RMaier D: Toward inclusive trial protocols in heterogeneous neurological disorders: prediction-based stratification of participants with incomplete cervical spinal cord injury. Neurorehabil Neural Repair 29:8678772015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30

    Tanadini LGSteeves JDHothorn TAbel RMaier DSchubert M: Identifying homogeneous subgroups in neurological disorders: unbiased recursive partitioning in cervical complete spinal cord injury. Neurorehabil Neural Repair 28:5075152014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31

    Tator CH: Review of treatment trials in human spinal cord injury: issues, difficulties, and recommendations. Neurosurgery 59:9579872006

  • 32

    Tee JWChan PCHRosenfeld JVGruen RL: Dedicated spine trauma clinical quality registries: a systematic review. Global Spine J 3:2652722013

  • 33

    Thibault-Halman GRivers CSBailey CSTsai ECDrew BNoonan VK: Predicting recruitment feasibility for acute spinal cord injury clinical trials in Canada using national registry data. J Neurotrauma 34:5996062017

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 34

    Vaccaro ARKoerner JDRadcliff KEOner FCReinhold MSchnake KJ: AOSpine subaxial cervical spine injury classification system. Eur Spine J 25:217321842016

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35

    Vaccaro AROner CKepler CKDvorak MSchnake KBellabarba C: AOSpine thoracolumbar spine injury classification system: fracture description, neurological status, and key modifiers. Spine (Phila Pa 1976) 38:202820372013

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 36

    Waring WP IIIBiering-Sorensen FBurns SDonovan WGraves DJha A: 2009 review and revisions of the International Standards for the Neurological Classification of Spinal Cord Injury. J Spinal Cord Med 33:3463522010

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37

    Wilson JRArnold PMSingh AKalsi-Ryan SFehlings MG: Clinical prediction model for acute inpatient complications after traumatic cervical spinal cord injury: a subanalysis from the Surgical Timing in Acute Spinal Cord Injury Study. J Neurosurg Spine 17 (1 Suppl):46512012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 38

    Wilson JRCadotte DWFehlings MG: Clinical predictors of neurological outcome, functional status, and survival after traumatic spinal cord injury: a systematic review. J Neurosurg Spine 17 (1 Suppl):11262012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 39

    Wilson JRDavis AMKulkarni AVKiss AFrankowski RFGrossman RG: Defining age-related differences in outcome after traumatic spinal cord injury: analysis of a combined, multicenter dataset. Spine J 14:119211982014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 225 225 27
Full Text Views 53 53 10
PDF Downloads 37 37 7
EPUB Downloads 0 0 0

PubMed

Google Scholar