Human Molar Size Variation: Wisdom Teeth Just Don't Care?

Abstract

The inhibitory cascade model (ICM) is a tenet of mammalian molar morphogenesis that predicts that a larger developing first molar (M1) results in a smaller and later‐forming adjacent second (M2) and third (M3) molar. Recently, we have shown that late‐developing M3s are also at higher risk of impaction, a common dental health concern worldwide. The mechanisms delaying M3 onset time are unclear.Here we tested the hypothesis that M1 crown size predicts M3 developmental timing and M3 eruption, based on a sample of young Western Canadian dental patients. We also tested the extent to which contemporary modern human molars conform to the ICM. We sampled retrospective cone beam computed tomography scans and panoramic radiographs taken for clinical diagnostic purposes of multi‐ethnic patients aged 13‐24 years residing in four Western Canadian municipalities. We studied 99 patients, sampling 327 quadrants from 57 girls and 42 boys, at two distinct times that allowed us to know if an M3 had properly erupted. Using Horos and Xoran i‐Cat software, we scored permanent molar development using a 9‐stage classification after Demirjian's system. We measured mesiodistal lengths of M1, M2 and M3 crowns. In SPSS v.25, we calculated a Generalized Estimating Equation model through a multivariable linear regression to test if M1 crown length (“size”) predicted M3 developmental timing and eruption status. We also tested other variables including jaw type (maxillary; mandibular). Significance was set to p<0.05. Unexpectedly, we identified 13 molar size ratio patterns occurring at different frequencies (e.g., M1>M2>M3 in only 31.6% of cases) that reflected whether molars were located in the maxilla versus the mandible. This pattern diversity included up to four patterns in the same person's mouth. Further, M3 size was predicted not by M1 size (Wald Chi‐Square [Wald X2]=0.701, p=0.40) but by M2/M1 ratio and absolute M2 size (Wald X2=14.04, p=0.000). Relative to M1 crown length, if M2 crown length increased by 10%, the adjacent M3 crown lengthened by 7% (±3%), indicating that M2 increases “faster” than M3. Crown length ratios (M3:M1, M2:M1) did not predict M3 impaction (Wald X2=0.124, p=0.73) or M3 developmental stage (M2:M1, Wald X2=0.228, p=0.63; M3:M1, Wald X2=3.253, p=0.07). We conclude that M1 crown length does not predict early versus late M3 development or M3 eruption status. Thus, the ICM does not appear to be useful for predicting M3 impaction risk. Contemporary modern humans present several distinct molar size ratio patterns, indicating that Homo sapiens molar size is subject to oral quadrant‐specific and jaw type‐specific effects beyond inhibition from a developing M1.