BACKGROUND:Modic changes (MC) are associated with low back pain (LBP). In this study, we compared changes in size and type of MC, after a single intravenous infusion of 5 mg zoledronic acid (ZA) or placebo, among chronic LBP patients with MC on magnetic resonance imaging (MRI), and evaluated whether the MRI changes correlate with symptoms. METHODS:All patients (N = 19 in ZA, 20 in placebo) had MRI at baseline (0.23-1.5 T) and at one year (1.5-3 T). We evaluated the level, type and volume of all the MC. The MC were classified into M1 (M1 (100%)), predominating M1 (M1/2 (65:35%)) or predominating M2 (M1/2 (35:65%)), and M2 (M2 (100%)). The first two were considered M1-dominant, and the latter two M2-dominant. Volumes of M1 and M2 were calculated separately for the primary MC, which was assumed to cause the symptoms, and the other MC. We analysed the one-year treatment differences in M1 and M2 volumes using analysis of covariance with adjustments for age, sex, body mass index, and smoking. The correlations between the MRI changes and the changes in LBP symptoms were analysed using Pearson correlations. RESULTS:In the ZA group, 84.2% of patients had M1-dominant primary MC at baseline, compared to 50% in the placebo group (p = 0.041). The primary MC in the ZA group converted more likely to M2-dominant (42.1% ZA, 15% placebo; p = 0.0119). The other MC (15 ZA, 8 placebo) were on average 42% smaller and remained largely M2-dominant. The M1 volume of the primary MC decreased in the ZA group, but increased in the placebo group (-0.83 cm3 vs 0.91 cm3; p = 0.21). The adjusted treatment difference for M1 volume was -1.9 cm3 (95% CI -5.0 to 1.2; p = 0.22) and for M2 volume 0.23 cm3 (p = 0.86). In the MC that remained M1-dominant, volume change correlated positively with increased symptoms in the placebo group, whereas the correlations were negative and weak in the ZA group. CONCLUSIONS:Zoledronic acid tended to speed up the conversion of M1-dominant into M2-dominant MC and decrease the volume of M1-dominant MC, although statistical significance was not demonstrated. TRIAL REGISTRATION:The registration number in ClinicalTrials.gov is NCT01330238 and the date of registration February 11, 2011.