The objective was to determine the association between specific foot lesions and culling in dairy cows. Using 5 trained professional hoof trimmers, data from 6,513 cows in 157 herds were recorded for analysis. During the study period, 1,293 cows (19.9%) were culled. Infectious lesions were most frequent in nonculled cows, whereas hoof horn lesions were most common in the culled cows. Median time to culling was 188 d [95% confidence interval (CI): 175-198 d] for cows without a lesion and 157 d (CI: 149-168 d) for cows with a lesion. Time from hoof trimming to culling was used to model the association between foot lesions and culling hazard. The final multivariate Cox proportional hazards model included heifers, infectious lesions, white line lesions, hemorrhages, sole ulcers, other lesions, and free-stall housing as covariates. Results of the final model showed that infectious hoof lesions had no significant association with culling. Yet, the hazard ratios for white line lesions, ulcers, and hemorrhage were 1.72 (CI: 1.39-2.11), 1.26 (CI: 1.05-1.52), and 1.36 (CI: 1.16-1.59), respectively. The association with culling for the grouped variable "other lesions" was time dependent and decreased with time. These results illustrate that there were significant associations with cow productivity for hoof horn lesions found at routine hoof trimming and that emphasis should be placed on proper treatment and earlier detection of these foot lesions.