Abstract Purpose Inflammation could play a key role in tissue damage and bone metabolism. The modified dietary inflammatory score (M-DIS) is a validated tool to estimate the inflammatory potential of the diet. In the present study, we evaluate the associations between the M-DIS and bone mineral density (BMD) in a senior Mediterranean population with overweight/obesity and metabolic syndrome. Methods Baseline cross-sectional association between the M-DIS and bone mineral density was assessed in 1134 participants of the multicenter PREDIMED-Plus trial (aged 55–75 with overweight/obesity and metabolic syndrome). BMD was measured using Dual-energy X-ray Absorptiometry scans and participants answered a food frequency questionnaire to determine the M-DIS. BMD was categorized as low BMD when T score was equal or lower than -1 and normal BMD in another case. Associations between BMD and M-DIS were evaluated by using linear and logistic regressions adjusted by other co-variates. Results Participants in the top tertile of the M-DIS had a lower BMD at total femur [β (95% CI) − 0.02 (− 0.04, − 0.01)], trochanter areas [β (95% CI) − 0.03 (− 0.05, − 0.01)] and lumbar spine area [β (95% CI) − 0.03 (− 0.07, 0.01)] (but in the last case, measures were less precise and hence not statistically significant) compared to those in the lower M-DIS tertile. Multiple logistic regression analyses showed that the odds of the total femur and femoral trochanter osteopenia/osteoporosis were higher in participants in the top tertile compared to those in the lowest tertile of M-DIS [OR (95% CI) 1.71 (1.12, 2.64), P for trend 0.015; 2.02 (1.29, 3.21), P for trend 0.002, respectively]. Conclusion A high pro-inflammatory diet, measured by the M-DIS, is associated with lower BMD in a senior Mediterranean population with metabolic syndrome.