Third Molar Status and Loss of Adjacent Second Molars

Objectives: To determine if second molars (2^nd-M) adjacent to erupted or unerupted third molars (3^rd-M) are at greater risk for loss than those adjacent to absent 3^rd-M spaces.
Methods: Participants were 965 men (baseline age 48±8 years) in the Department of Veterans Affairs (VA) Dental Longitudinal Study followed since 1968. They are not VA patients but receive medical and dental care in the private sector. Status of 3,008 3^rd-M/2^nd-M tooth pairs was assessed clinically and confirmed radiographically at each examination. Alveolar bone loss as percent of CEJ-root apex distance (ABL), any decay or restoration (DFT) and bleeding on probing (BOP) were recorded. For unerupted 3^rd-M, additional measurements taken from radiographs included depth of impaction based on the nature of overlying tissue, proximity to 2^nd-M, and angle of inclination of the long axes of the 3^rd-M/2^nd-M pair. Smoking status was assessed. Hazards ratios (HR) and 95% confidence intervals (CI) of 2^nd-M tooth-loss were estimated from Cox proportional hazards regression models for clustered data. Covariates were baseline age and time-varying values of smoking (cigarette, cigar, none), 2^nd-M ABL>20%, DFT, and BOP. The reference group was 2^nd-M adjacent to absent 3^rd-M.
Results: Baseline 3^rd-M status did not significantly predict 2^nd-M tooth loss independently of 2^nd-M pathology (ABL, DFT, BOP), age, and smoking status. Overall, hazards for loss of 2^nd-M adjacent to erupted 3^rd-M (HR=0.9, CI=0.8-1.1) or to unerupted 3^rd-M (HR=1.3, CI=0.9-1.7) were not significantly increased. Among 2^nd-M adjacent to unerupted 3^rd-M, neither depth of impaction nor proximity to 3^rd-M altered the HR, however, hazards of loss were higher in a subset (27%) in which the adjacent unerupted 3^rd-M axis was oriented >45° to the 2^nd-M (HR=1.9, CI=1.2-3.0).
Conclusions: The majority of second molars adjacent to erupted or long-standing unerupted third molars are not at increased risk of loss.