Antibiotic treatment in patients with chronic low back pain and vertebral bone edema (Modic type I changes): a double-blind randomized clinical controlled trial of efficacy (Albert et al. 2013 Eur Spine J) ALAN TAM PGY3 DIVISION OF PHYSIATRY, UNIVERSITY OF TORONTO MARCH 28 TH, 2014 Aim of current study Test efficacy of Modic antibiotic spine therapy (MAST) in patients with chronic low back pain, and with new Modic 1 changes in vertebrae adjacent to previously herniated disc 1
Background Systematic review of prevalence of Modic changes published in 2008 (Jensen et al.) Non clinical population (non-lbp, general, working): 13 studies median = 6% Clinical population: 45 studies median = 43% Infection hypothesized to be a cause of bone edema underlying Modic type I Previous pilot study, significant clinical effect of Antibiotic therapy on LBP with Modic type I changes (n=29) Patient Recruitment Patients recruited from 2 centres in Denmark Inclusion criteria Age 18-65 MRI confirmed disc herniation at L3-4/L4-5/L5-S1 within last 6-24 months Lower back pain >6 months Exclusion criteria Allergy to antibiotics Current pregnancy or lactation Any kidney disease Pending litigation 2
Individuals recruited into the study had repeat MRI (at Spine centre of Southern Denmark) Modic findings coded by 1 experienced research radiologist (high intra and inter tester reliability) On pain questionnaire, required Numeric Pain Rating Scale > 6 (from 0-10) Average pain score of: Current pain, average pain in last 2 weeks, worst pain in last 2 weeks Study Protocol Randomization performed by Central Pharmacy Patient and assessor blinded to randomization (until after 1 year follow-up) Active antibiotic regimen 3 independent infectious disease experts consulted with culture results of previous studies All recommended Amoxicillin-clavunate Treatment with Bioclavid tablets (Amox-clav 500mg/125mg) q8h for 100 days 3
Sample Size Based on results of previous pilot study Median reduction of RMDQ 3.7 (used this as minimum reduction detection in calculation) Standard deviation = 4 Placebo effect = 1.4 Alpha = 0.05 Power = 0.90 Estimated dropout rate = 10% Sample size 65 in each group based on two-group comparison (Antibiotic vs placebo) Dose-response part of study (1 vs 2 tabs of Antibiotic) included due to request by Danish Medical authorities Study not designed nor powered to test dose-response Study flow Baseline and 1 year Physical exam (performed by same blinded observer) Self-reported questionnaire MRI (same protocol, same staff, same machine) Blood samples 100 days Patients mailed questionnaire 4
Randomization groups 4 dosage groups A One Bioclavid tablet (n=45) Number of possible endplates with Modic type B One one placebo changes, adjacent tablet to the previous (n=36) herniation C Two bioclavid tablets (n=45) Distribution of size of Modic changes minute changes = size 1 D Two placebo tablets (n-36) Antibiotics group n = 90 Placebo group n = 72 p value of difference Age (years) 44.7 SD 10.3 45.5 SD 9.2 Gender (female) 58.2 % 58.2 % Smokers 33 % 32.8 % 97 % 92.2 % 10.4 % 28.8 % 0.007 Continuous pain for more than 2 years 52.6 % 52.2 % Disease-specific disability-rmdq Median, 15 [lower; upper quartile] [11; 18] Back pain (0 10) 6.7 Median [lower; upper quartile] [5.3; 7.7] Compliance measured via medication diaries Leg pain (0 10) 5.3 94.8% in Median placebo, [lower; upper quartile] 94.0% in [2.3; antibiotic 7] Number of hours with back pain 448 Median [lower; upper quartile] [364; 448] 15 [12; 18] 6.3 [4.7; 8] 4.0 [1; 7] 448 [392; 448] Previous disc herniation surgery 51.9 % 40.3 % Days with sick leave Mean 51 SD 92 42 SD 80 Bothersomeness Median, [lower; upper quartile] 7 [6; 8] 8 [5; 9] Number with constant pain 75.3 % 73.1 % EQ-5D, thermometer Median, [lower; upper quartile] 59 [40; 70] 60 [40; 75] Co-interventions Mild analgesics as required Patient s individual usual treatment of anti-inflammatory/pain-relieving medication No standardization of adjuvant pain medication Education session Explained why should not exercise during study period No method of tracking whether patients did/did not exercise 5
Outcomes Self-reported questionnaires Global perceived effect Roland Morris Disability Questionnaire (RMDQ) LBP pain rating scale Questionnaire characteristics The patients compare their baseline status with their status at follow-up, measured on a 7-point Likert scale A disease-specific disability questionnaire in which the patient answers 23 yes/no questions. The scale width is 0 23, where high scores are worst Three 11-point box scales measuring current pain, the worst within the last 2 weeks and usual pain within the last 2 weeks. These three scores are measured and averaged for both leg and lumbar pain independently Hours with LBP during the last 4 weeks EQ-5D: Thermometer Days with sick leave Bothersomeness Number of days during the last 28 days (4 weeks) the participant had experienced LBP (0 28 days), and, on an typical day, how many of the hours awake they experienced LBP (0 16 h). The number of days and hours are multiplied (a 0 448 scale) Quality-adjusted health status EQ-5D on a vertical Thermometer (1 100), 100 is best The number of days within the last year the participant was on sick-leave support from the government Measured on a 11-point box scale, where 0 = none, 10 = my life is worthless due to my back pain Constant pain Pain that can vary during the day but is always present MRI Volume of the vertebrae with Modic type 1 change, (1 4) Serum analysis Leukocytes, neutrophils, eosinophils, basophils, lymphocytes, monocytes, P/S creatinine, lactate dehydrogenase, alkaline phosphatase, C-reactive protein Statistics Differences in binary variables Fisher s exact test Continuous variables tested with 3 tests Unpaired T-test Unpaired T-test (adjusted for age and gender) Mann-Whitney test (if not normally distributed data) Intention to treat analysis 6
Results Antibiotics group n = 90 Placebo group n = 72 p value of difference Age (years) 44.7 SD 10.3 45.5 SD 9.2 Gender (female) 58.2 % 58.2 % Smokers 33 % 32.8 % Number of possible endplates with Modic type one changes, adjacent to the previous 97 % 92.2 % herniation Distribution of size of Modic changes minute changes = size 1 10.4 % 28.8 % 0.007 Continuous pain for more than 2 years 52.6 % 52.2 % Disease-specific disability-rmdq Median, [lower; upper quartile] 15 [11; 18] 15 [12; 18] Back pain (0 10) Median [lower; upper quartile] 6.7 [5.3; 7.7] 6.3 [4.7; 8] Leg pain (0 10) Median [lower; upper quartile] 5.3 [2.3; 7] 4.0 [1; 7] Number of hours with back pain Median [lower; upper quartile] 448 [364; 448] 448 [392; 448] Previous disc herniation surgery 51.9 % 40.3 % Days with sick leave Mean 51 SD 92 42 SD 80 Bothersomeness Median, [lower; upper quartile] 7 [6; 8] 8 [5; 9] Number with constant pain 75.3 % 73.1 % EQ-5D, thermometer Median, [lower; upper quartile] 59 [40; 70] 60 [40; 75] Antibiotic group (n=90) Placebo group (n=72) Modic change size = 1 10.4% 28.8% Modic change size = 2 to 4 89.6% 71.2% The degree of Modic changes initially is different between the 2 treatment groups The Antibiotic group had more severe changes on their MRI at baseline Did this translate into a difference clinically at baseline? The two groups had non-statistically significant difference in RMDQ and pain scores What might you infer from this data? The degree of Modic change on a MRI may not correlate with severity of back pain Is comparing the number of patients (between the two groups) that have improvement in severity of Modic changes appropriate? If the absolute number of endplates, then no If the percentage change in each group, it may be an appropriate measure 7
Primary outcome Antibiotic baseline n = 90 Antibiotic 100 days Antibiotic 1-year n = 76 follow-up n = 77 Placebo baseline n = 72 Placebo 100 days n = 67 Placebo 1-year follow-up n = 67 p value for difference between placebo and antibiotic groups at 1-year follow-up Disease-specific disability-rmdq Median [lower; upper quartile] 15.0 [11; 18] 11.5 [7; 14] 7.0 [4; 11] 15.0 [12; 18] 14.0 [11; 18] 14.0 [8; 18] 0.0001 Back pain (0 10) Median [lower; upper quartile] 6.7 [5.3; 7.7] 5.0 [2.7; 6.7] 3.7 [1.3; 5.8] 6.3 [4.7; 8] 6.3 [3.7; 7.7] 6.3 [4; 7.7] 0.0001 Secondary outcomes Antibiotic baseline n = 90 Antibiotic 100 days n = 76 Antibiotic 1-year follow-up n = 77 Placebo baseline n = 72 Placebo 100 days n = 67 Placebo 1-year follow-up n = 67 p value for difference between placebo and antibiotic groups at 1-year follow-up Leg pain (0 10) Median [lower; upper quartile] 5.3 [2.3; 7] 3.0 [1; 5.7] 1.7 [0; 4.2] 4.0 [1; 7] 4.3 [1; 7] 4.3 [1; 6.3] 0.0004 Number of hours with back pain Median [lower; upper quartile] 448 [364; 448] 180 [16; 136] 64 [4; 280] 448 [392; 448] 200 [28; 392] 448 [224; 448] 0.0001 Days with sick leave 51.0 last year SD 92 Mean * 18.9 SD 61 42.0 SD 80 * 45.4 SD 90 0.064 Bothersomeness scale (0 10) Median [lower; upper quartile] 7 [6; 8] * 3 [2; 5] 8 [5; 9] * 6 [4; 8] 0.0001 EQ-5D (1 100) 100 is best Thermometer 59 Median [lower; [40; 70] upper quartile] 65 [40; 79] 75 [54; 90] 60 [40; 75] 60 [40; 75] 60 [39; 74] 0.0014 General improvement in % * * 39 SD 38.4 * * 1.8 SD 31.7 0.0001 8
Modic Changes Baseline Placebo: 130/134 (97%) adjacent endplates demonstrated Modic changes Antibiotic: 142/154 (92%) adjacent endplates demonstrated Modic changes Grade 1 Modic changes: 28.8% in placebo, 10.4% in antibiotic (p=0.006) 1 year follow-up 10 patients in each group demonstrated no Modic changes Significant decrease in volume observed in antibiotic group, with volume 2-4 reduced to volume 1 (p=0.05) Reduction not observed in placebo group Antibiotic baseline n = 90 Antibiotic 1-year followup n = 77 Placebo baseline n = 72 Placebo 1-year follow-up n = 67 Observed volume volume 16 1, minute size 29 31 24 0.05 Observed volume volume 126 2 4, moderate/large size 113 99 96 0.07 P value for difference between placebo and antibiotic groups at 1- year follow-up Dose response There was a trend towards a positive dose response relationship with double dose antibiotics appearing to be more efficacious; however, this was not statistically significant as the trial was not powered for this comparison (results not shown) 9
Adverse events More common in antibiotic group (65% vs 23% in placebo) Mainly low-grade GI issues (loose BM, increased flatus, burping) Loose BM for more than 3 weeks more common in antibiotic group (27%) vs placebo (11%) One case of severe vomiting (blood in vomit) in placebo group No difference in adverse effects between single-dose and double-dose antibiotic (no numbers provided) Causes of (Chronic) Low back pain 10
Modic changes Anatomy and Physiology Vertebral end plate signal change seen on MRI Changes likely a response of bone marrow to degenerative process involving the disk Modic Type I changes on MRI 11
Classifying/Quantifying Modic Changes Nordic Modic Protocol 1 = endplate only 2 = up to 25% of vertebrae 3 = up to 50% of vertebrae 4 = above 50% of vertebrae Chung et al. (2004) Lumbar endplate level Location within endplate (anterior, middle, posterior) Karchevsky et al. (2005) Lumbar endplate level Location within endplate (15 segments) Differential Diagnosis on MRI Spinal Infection Spondylodiscitis hyperintense on T2, hypointense on T1 (also paravertebral soft-tissue edema or epidural mass effect) Erosions of vertebral body and endplates with intervertebral disc infection Schmorl s node Hypointense T1 and hyperintense T2 Tumour Metastasis most common neoplastic lesion in spinal column Rarely involves the disc 12
Modic Changes and Low Back Pain 2 reviews concluded positive association between Modic changes and low back pain Those with LBP were more likely to have Modic changes (in cross-sectional and prospective studies) Natural history of Modic changes Conflicting studies regarding progression/conversion between different Modic types In general, conversion from Modic I to Modic II has been observed (most often L4/5, L5/S1 level) Previously believed that Modic I was unstable state, and Modic II was stable Newer studies have reported Modic II conversion to Modic I Potential Mechanisms of Modic changes Pathogenic mechanisms unclear 1) Biomechanics Endplates calcify with aging and replacement of bone, alters structure and may lead to uneven distribution of loads across entire disc (contribute to endplate fissure) Loss of nucleus pulposus (herniation, dehydration from degeneration) can increase shear forces on endplates, result in microfractures Microfractures and fissures may be source of Modic changes 2) Biochemical Many studies have found higher levels of pro-inflammatory mediators (IL-6, IL-8, PGE2, TNF) in patients with Modic changes and LBP 13
Inflammation and Modic Changes Prospective study by Rannou et al. (2007, Arthritis Rheum.) looked a high-sensitivity CRP from chronic low back pain patients with Modic I, Modic II and no Modic changes on MRI Modic I group had significant higher hs-crp compared to Modic II and no Modic change groups Infection and back pain Infection one potential cause of Type 1 Modic changes Several studies culturing samples from surgeries for herianted lumbar disks have found high incidence of infection (Proprionibacterium acnes, Corynebacterium propinquam, Staphylococus) Rates of positive culture ranged 19% -71% Small uncontrolled study (32 patients) with chronic low back pain (following lumbar disc herniation) and treated with Amox-Clav for 90 days (Albert et al. 2008) Statistically and clinically significant improvement in outcome 14
Possible interventions No unanimous conclusions regarding minimally invasive surgery for treating degenerative disc disease in Modic changes Stabilization and fusion surgery has been suggested, but no evidence to support this (and also because underlying mechanism is unknown) Spinal steroid injections more effective for treating lumbar DDD than Modic changes Investigator s conclusions Antibiotics could be considered as a treatment option for this special subgroup of patients with CLBP and Modic type 1 changes after a lumbar disc herniation when all other treatment options have failed What do we think about this statement? Who would be appropriate? 15
Determining who might benefit Confirmed Modic type I changes on MRI Operative management population Intra-operative culture Referral to ID if positive (Amox-clav is broad-spectrum) Non-operative management population Chronic, functionally limiting pain Positive inflammatory markers, with autoimmune and malignancy ruled out Rule out other organ dysfunction that cause elevated inflammatory markers (Cardiac disease, Respiratory infection, Bladder infection, osteomyelitis) More than 1 possible mechanism More confirmatory work in other populations and studies on improved protocols as well as the background science should be encouraged Small randomized trail published March 2014 of patients with Modic changes on MRI found zoledronic acid to be effective in treatment chronic low back pain 16