INTRODUCTION

Leeway space is the difference in size of the mesiodistal crown widths of the primary canines and molars compared with that of their permanent successors (canine, first and second premolars; Figure 1).^1–4 The primary teeth typically possess a larger mesiodistal sum than the permanent teeth that replace them—especially in the mandibular arch, predominantly because of the primary second molar's larger mesiodistal crown dimension compared with the second premolar.⁵ Orthodontists often rely on leeway space, an anticipated gain of 1–2 mm per quadrant, in their treatment protocols to help resolve anterior crowding.^6–8

View Figure

• Download as Powerpoint
• Download Figure

It is unknown who first observed the space resulting from these crown-size differences, but Nance^9,10 commonly is cited as initially describing its clinical application. He coined the term “leeway” and labeled it a “fundamental fact concerning the human dentition.” Nance⁹ cited descriptive tooth sizes reported as early as 1890 by G.V. Black in the latter's first edition of Descriptive Anatomy.¹¹ Black's sample consisted of extracted teeth (sexes pooled) from an unreported number of cases. Although the source and method of measurement is poorly defined,¹² it is arguably the most commonly cited set of tooth dimensions, certainly so in the clinical literature. Using Black's figures, Nance reported leeway space averages of 0.9 mm per quadrant in the maxilla and 1.7 mm per quadrant in the mandible.

Leeway is not a constant, but differs among people. Nance⁹ described cases varying from 0.0 to 4.0 mm of mandibular leeway space. Numerous authors report different averages of leeway space, apparently depending on regional or population tooth crown differences plus sampling variation.¹³ In addition, Hille¹⁴ found that mandibular leeway space averaged 2.4 mm in girls, but was significantly smaller in boys (mean = 1.9 mm).

Northway et al.¹⁵ found that when maxillary primary first molars were lost prematurely, (1) maxillary primary second molars and permanent first molars drift mesially, (2) canines drift distally, (3) first premolars emerge more mesially, and (4) permanent maxillary canines emerge labially with risk of being blocked out. An effective solution, if treated in time, is to preserve the leeway space with, for example, a distal shoe space maintainer, lingual holding arch, or Nance appliance.^16,17 Leeway space in the mandibular arch is often more critical; there are fewer therapeutic options in this arch because of its limited potential for expansion, unstable labialization of incisors, and difficulty of molar distalization.¹⁸

Children in first-world countries have been experiencing secular trends over recent generations. Among the best-known examples are increase in stature,¹⁹ reduction in age at menarche,²⁰ and gain in body weight.²¹ The conventional explanations for these generational changes are centered on improved environment; diminished morbidity; and (principally) better, more dependable nutrition. Positive secular trends have been described for tooth crown sizes,^22–24 and larger teeth have been implicated as a contributor to dental crowding.^25–28 Tooth size–arch length discrepancies appear to have increased.²⁹ Additionally, the tempo of tooth emergence has quickened.^30,31 These observations suggest that environmental improvement might also affect leeway space, which is derived from differences among crown sizes.

For the face and teeth, much of our knowledge is derived from several human growth studies conducted in the 20th century. These are largely now complete.^32,33 Although they reflect children's growth of several generations ago, these studies are a principal source of information for many of today's clinical decisions. The purpose of the present study was to test for a secular trend, focusing on mandibular leeway space. The null hypothesis was that average leeway space has remained unchanged. We examined leeway space in the mandibular arch since it is larger and clinically more challenging than in the maxilla.

MATERIALS AND METHODS

An a priori power analysis was conducted assuming a factorial two-way analysis of variance (ANOVA), partitioning on cohort while controlling for sex in crown size.^34,35 Statistical power is the probability of rejecting the null hypothesis when it truly is false. Assuming a 0.5-mm difference in mandibular leeway space between cohorts (ie, minimum nontrivial effect size), with alpha settings of 0.05 and sample sizes of 45 per cohort (equal arms), expected power was 85%. To be conservative, we assumed that the interaction effect accounted for no variation.³⁴

Tests³⁶ indicated that variables were normally distributed, so inferential tests used two-way ANOVA,³⁷ with cohort and sex as fixed effects. Subject's sex was included to account for the tendency for boys to have larger crowns than do girls.³⁸ Statistical significance was set at the conventional level of 0.05, and tests were two-tailed.

A contemporary series (23 boys, 22 girls)—born between 1990 and 2000—was obtained by inspecting all early treatment cases in the University of Tennessee Health Science Center Graduate Orthodontic Clinic (IRB approval 14-03570-XM). Casts with bilateral fully erupted mandibular primary teeth were identified. Those individuals who received a second phase of treatment (when the premolars and permanent canines had emerged into functional occlusion) were selected for inclusion. This created a series of pairs of casts of the same subjects, one with primary teeth and the other with successors.

The earlier cohort used for comparison also consisted of 45 cases (23 boys, 22 girls), participants randomly selected from the Denver Growth Study³⁹ and born in the 1930s. Casts from successive ages were used to measure the primary and permanent teeth on the same persons.

All individuals in both cohorts were phenotypically normal American whites (based on photographs and patient records) with no congenitally missing teeth and no known syndrome or systemic condition that might affect growth. Cases with exfoliated or extracted teeth in the midarch, either primary or permanent, were eliminated. Teeth with marginal restorations or carious defects were excluded, as well as any distortions or irregularities in the model.

Maximum mesiodistal crown diameters of the mandibular teeth were measured in both the left and right quadrants on plaster dental casts using a digital-readout sliding caliper (Mitutoyo, Aurora, Ill ). The beaks of the caliper had been machined to fit well into the dental embrasures. Measurements were made in a standardized manner⁴⁰ and recorded to the nearest 0.01 mm even though the caliper's readout was precise to 0.005 mm. All data were acquired by the senior author. Initial testing showed no significant side difference, so the arithmetic means of the left and right homologues were used for subsequent analysis. A subset of the casts (n = 20 casts, 240 paired measurements), both primary and permanent, were remeasured after a washout period to estimate intraobserver reliability.^41,42

RESULTS

Intraobserver repeatability was high. No variable showed a systematic difference between measurement sessions, and Dahlberg's d_i was less than 0.1 mm (mean = 0.07 mm), making measurement error appreciably less than the observed cohort differences. Cronbach's alpha was 0.998, and the intraclass correlation between measurements (mixed model, fixed observer) was also 0.998 (95% CI: 0.998–0.999), which was highly significant (P < .001).

Neither the primary canine nor primary first molar differed in size between cohorts (Table 1). Despite these findings, the cumulative mandibular primary canine plus first and second primary molar size (c + m1 + m2) was slightly larger in the recent cohort (23.53 mm earlier vs 23.83 mm recent cohort, a mean difference of 0.30 mm; P = .009), principally due to larger second primary molars (m2) in the recent cohort (Figure 2).

Table 1. Results of Two-factor Analysis of Variance for Mandibular Teeth^a

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

When we compared the permanent dentitions, we found that the first and second premolars were each significantly larger in the recent cohort, resulting in a significant difference of leeway space. Leeway space averaged 2.45 mm per quadrant in the earlier cohort and 2.03 mm in the recent group (sexes pooled), though sample variability remained the same (Table 2).

Table 2. Descriptive Statistics of Tooth Crown Sizes by Cohort and Sex (mm)

View Fullscreen

The increases in tooth size were at the expense of leeway space. Tests of this are the association between tooth size and leeway space (Figure 3). Table 3 lists the results of three ANOVA tests, considering the effects of cohort and sex on the association between crown size and mandibular leeway space. Of note, none of the interaction effects was significant statistically. Figure 3A shows that the leeway space regressed on primary tooth size (ie, sum of c + m1 + m2). The association was positive and statistically significant (P = .0166), so the larger the primary teeth, the larger the predicted leeway space. This is intuitive in that larger primary teeth preserve more arch space that contributes to leeway space. The earlier cohort had a higher y-intercept (a = −3.58), but the regression coefficient—the change in leeway space per unit of primary-tooth size (c + m1 + m2)—was less. In the recent cohort, the y-intercept was lower (a = −7.41), but the regression slope was steeper. In clinical practice, though, the difference in slopes (0.26 vs 0.40; Figure 3A) is unlikely to be noticeable. Regression coefficients are listed in Table 4.

View Figure

• Download as Powerpoint
• Download Figure

Table 3 Results of Three-way Analyses of Variance A. Leeway Space Predicted by Primary Tooth Sizes

View Fullscreen

Table 4. Linear Regression Equations A. Mandibular Primary Tooth Size (c + m1 + m2) Predicting Leeway Space (per Cohort)

View Fullscreen

Figure 3B shows that permanent tooth size (C + P1 + P2) was negatively associated with leeway space; an increase of 1 mm in C + P1 + P2 predicted that the leeway space was reduced by 0.51 mm. Cohort and sex are combined since they differed insignificantly (Table 3B).

Primary and permanent tooth sizes were positively correlated (Figure 3C); children with large primary teeth were likely to have large permanent teeth. The correlation coefficient between c + m1 + m2 and C + P1 + P2 was 0.599 (95% CL = 0.447 and 0.717, respectively; r² = 0.395). While this sample is small (n = 45), the association shows the weak predictive power in mixed dentition analyses. Statistical significance is attainable with adequate sample sizes, but clinical precision cannot be improved because of the biologic limit of the association.

DISCUSSION

Tooth size is regulated by the size of the pulp cavity, which is established before deposition of mineralized tissue.^43,44 Statistically significant differences (Tables 1 and 2) were found between cohorts for the two premolars. These crowns mineralize perinatally,⁴⁵ after the in utero formation of most primary crowns.^46,47 One explanatory scenario is that the prenatal environment—depending principally on maternal physiology⁴⁸—has remained generally static across the two cohorts, but childhood nutrition has improved (and morbidity has lessened),⁴⁹ thus promoting larger postnatal permanent tooth development,^50,51 while primary tooth sizes remain unchanged.

As leeway space seems to be less dependable now than in the past, orthodontists should consider relying more on alternate solutions to resolve anterior crowding. Solutions can involve extraction therapy or greater use of interproximal reduction. Not all tooth types were measured, but the positive associations among tooth sizes⁵² suggest that other permanent teeth will also contribute to a greater space requirement in the typical patient. Large tooth size per se is a risk factor for malocclusion.^25–28

A limitation of the study is that results extrapolated from only two localized samples of American whites were used to interpret childhood conditions in general. It remains to be seen whether geographic differences affect interpretation. In addition, we compared data from this study from 45 subjects in each cohort. The study may promote interest in further examining this potential secular trend in other samples. Also, secular trends do not occur globally or synchronously across a population. They reflect environmental changes. Responses seem largely to have been completed in first-world countries,⁵³ but may be ongoing elsewhere as living conditions improve.

CONCLUSIONS

Comparing an earlier cohort of American whites (born in the 1930s) with a recent cohort (born in the 1990s) to test for a secular change in leeway space showed that

• Mandibular leeway space ranged from 0.0 mm to 3.3 mm in the recent cohort, with an average of 2.0 mm per quadrant.

• Mandibular leeway space was lower by an average of 0.42 mm per quadrant between cohorts—a clinically and statistically significant finding. This difference is mostly attributed to a secular trend for larger premolars.