INTRODUCTION

Modic type 1 changes (MC1) are inflammatory vertebral bone marrow lesions and associate with chronic low back pain (CLBP). Patients with MC1 profit from intraosseous basivertebral nerve ablation. However, MC1 are often not detected because MRI are not done due to the risk of patient catastrophizing and cost reasons. Therefore, MC1 biomarkers are needed. Increased serum C-reactive protein (CRP) has inconsistently been associated with MC1. It remains unclear if CRP is linked to local inflammation in MC1. The aim of this study was to show face-validity of CRP as a potential biomarker for MC1. We hypothesized that CRP is locally increased in MC1 bone marrow and correlates to blood CRP.

METHODS

This study was approved by the local Ethics Commission and performed in accordance with the Declaration of Helsinki. Thirteen MC1 patients undergoing lumbar spinal fusion at MC1 levels were included. Exclusion: infections, tumors, prior spinal fusion. Patients reported back and leg pain on a 10-point Visual-Analogue-Score (VAS.back,VAS.leg) and completed a Oswestry-Disability-Index (ODI). Two radiologists independently scored MC1 lesion size (0-4), Pfirmann grade of disc degeneration (0-5), and endplate degeneration score (0-6). Bone marrow aspirates from MC1 and control bone marrow were taken intraoperatively through pedicle screw trajectories before screw insertion (Fig. 1). Bone marrow plasma and cells were collected by centrifugation. Bone marrow plasma CRP was measured with ELISA. Blood CRP was measured as routine pre-operative clinical examination. CRP in MC1 was compared to intra-patient control bone marrow with t-tests. Blood CRP was correlated to MC1 CRP and to the difference of MC1 to intra-patient control bone marrow CRP (Spearman correlation). Expression of CRP by bone marrow cells was analyzed with quantitative real-time PCR (TaqMan).

RESULTS

Median blood CRP was 4 mg/L (0.2-52.9 mg/L). All values were below 8.0 mg/L except one patient, which was HLA-B27 positive. Mean bone marrow plasma CRP was higher in MC1 compared to intra-patient control (p=0.076; MC1: 49.7±52.6 mg/L; control 36.8±35.2 mg/L) (Fig.2). Blood CRP correlated with MC1 bone marrow CRP (rho=0.90, p<0.0001) and with the difference MC1-control bone marrow CRP (rho=0.82, p=0.002) (Fig.3). A blood CRP threshold between 3.1-4.1 mg/L detected with 100% accuracy increased CRP concentrations in MC1 bone marrow. Exclusion of the HLA-B27 patient did not affect correlation (rho=0.77, p=0.005). Patient demographic (age, sex, weight, height, BMI, smoker, VAS.back, VAS.leg, ODI) and radiologic data (MC1 size, disc degeneration, endplate score) did not correlated with blood or bone marrow CRP. Bone marrow cells did not express CRP.

DISCUSSION

Blood CRP correlated to increased CRP in MC1 bone marrow suggesting specific accumulation of CRP in MC1 as a result of local inflammation. Therefore, blood CRP in CLBP patients might be a MC1 biomarker with high face-validity to support physicans’ decision on lumbar spine MRI for CLBP patients. Limitations: blood CRP does not correlate with MC1 size, pain, and disability and hence does not reflect clinical severity of MC1. Robustness of CRP with respect to comorbidities (e.g. osteoarthritis) should be tested in larger studies to substantiate validity of CRP as MC1 biomarker.