To comprehensively investigate the associations of subclinical thyroid dysfunction with BMD and fractures at various sites. Comprehensive electronic and manual searches of databases were systematically conducted to identify prospective cohort studies from the inception of the databases to May 2019. The summary results for fractures and BMDs at various sites were calculated by relative risks (RRs) and weighted mean differences (WMDs) with corresponding 95% confidence intervals (CIs) using the random-effects model. Seventeen prospective cohorts from 24 studies were identified and 313,557 individuals were recruited in a final analysis. The summary RR indicated that subclinical hyperthyroidism was associated with an increased risk of any fracture (RR, 1.17; 95% CI, 1.08-1.26; P < 0.001), hip fracture (RR, 1.27; 95% CI, 1.09-1.48; P = 0.003), spine fracture (RR, 1.97; 95% CI, 1.31-2.97; P = 0.001), and non-spine fracture (RR, 1.19; 95% CI, 1.04-1.37; P = 0.014). However, there were no significant associations of subclinical hypothyroidism with the risk of any fractures (P = 0.166), hip fracture (P = 0.068), spine fracture (P = 0.818), and non-spine fracture (P = 0.277). Finally, subclinical hyperthyroidism was associated with lower distal forearm BMD in women, and ultradistal forearm BMD in both men and women, whereas subclinical hypothyroidism was associated with higher femur neck BMD in women. Subclinical hyperthyroidism could induce additional risk on fractures at any, hip, spine, and non-spine, whereas subclinical hypothyroidism did not have any impact on fractures. Moreover, BMD at the lower distal and ultradistal forearms might be affected by subclinical hyperthyroidism, and higher femur neck BMD could be affected by subclinical hypothyroidism.