General Session: Innovative Technologies II - Hall F

Presented by: M.F. Gornet

Author(s):

M.F. Gornet(1), J.C. Lotz(2), R.K. Eastlack(3), J. Peacock(4), F.W. Schranck(5), J. Claude(4), A.G. Copay(5)

(1) The Orthopedic Center of St. Louis, St Louis, MO, United States
(2) University of California San Francisco, Department of Orthopedic Surgery, Redwood City, CA, United States
(3) Scripps Clinic, Department of Orthopedic Surgery, San Diego, CA, United States
(4) Nocimed, Inc, Redwood City, CA, United States
(5) SPIRITT Research, St Louis, MO, United States

Abstract

Purpose: We have previously shown that non-invasive single voxel MR Spectroscopy (SVS) measurement of disc chemistry correlates with +/- provocative discography results (PD+/-) for painful (P) versus non-painful (NP) discs in chronic, severe low back pain patients (LBP). This current study extends on our previous work, and further evaluates for the first time whether in vivo disc SVS data in LBP patients predicts lumbar fusion and fusion+TDR surgical treatment (Sx) outcomes.

Methods: SVS was performed, single center, in 607 lumbar discs from 129 LBP patients via an investigational protocol developed using a commercial Siemens SVS pulse sequence and 3T Verio MR scanner. A custom investigational post-processor was used to enhance spectral quality and calculate acid and disc structural matrix biomarkers and related scoring algorithms. 262 (43%) discs received PD: 118 PD+, 144 PD-. 81 PD+ discs (plus 14 non-PD discs that were physician-diagnosed as painful by other criteria) and 101 PD- discs were used as P and NP controls, after excluding 19 discs with poor spectral quality and 61 discs with voxels not meeting optimal geometry (≥1cc, ≤4mm height). Ratios of acidity biomarkers to structural integrity biomarkers were used to determine a degenerative pain score: DP-SCORE Total and Normalized (to highest Total in-patient). DP-SCORE ranges (high=DP+, low=DP-) were trained to correlate to the P and NP discs (plus intermediate DPmild range). Standard statistics were calculated, with 10-fold repeat cross-validation performed for generalizability. Sx success outcomes (≥15 point ODI reduction) in 52 Sx patients (after 4 technical exclusions) were evaluated at 6-12 months follow-up and correlated with pre-surgical DP-SCORES of treated and non-treated discs. PD (generally SIS guidelines) and Sx (P control discs) were performed per physician discretion.

Results: DP-SCORES differed significantly between P and NP discs (p< 0.01; Figure 1). DP+/- versus P/NP correlations were (X%/average cross-validation %): Total SVS Accuracy= 85%/83% (90% for non-herniated discs), Sensitivity= 81%/80%, and Specificity= 89%/87%. Fusion surgery results correlated with pre-surgical SVS measurements (Table 1: Sx success rates were 92% (22/24) when treating all DP+ discs in-patient and 100% (5/5) in those 2-level surgeries; versus only 57% (4/7) when treating DP- discs.

Conclusions: Non-invasive, objectively quantitative SVS-derived DP-SCORES and DP+/- results accurately distinguish P from NP discs, especially in non-herniated axial LBP patients. 6-12 month Sx success rates were very high for patients treated at all DP+ discs, versus low for treated DP- discs. These data suggest a potentially valuable new approach to help doctors, in combination with other available information, better diagnose and evaluate treatment options toward more successful outcomes in LBP patients.

DP-SCORE

Table