Hyrax, Hass Y Quad.pdf

ORIGINAL ARTICLE

Treatment response and stability of slow maxillary expansion using Haas, hyrax, and quad-helix appliances: A retrospective study Thuylinh Huynh,a David B. Kennedy,b Donald R. Joondeph,c and Anne-Marie Bollenc Rockville, Md, Vancouver, British Columbia, Canada, and Seattle, Wash Introduction: In this retrospective study, we evaluated the short- and long-term effects of slow maxillary expansion with Haas, hyrax, and quad-helix appliances on posterior crossbite (PXB) correction stability, and maxillary intermolar width and angulation, in the deciduous or early mixed dentition. Methods: The inclusion criteria were models and treatment notes of patients with PXB at the start of treatment (T1), after PXB correction (T2), and at least 2 years posttreatment (T3). Exclusion criteria were craniofacial anomalies, fixed appliance use, or more than 1 expander type. From 312 consecutive expansion patients, 74 Haas, 41 hyrax, and 45 quad-helix subjects were evaluated regarding PXB correction, intermolar width, and angulation and compared with published norms to separate treatment effects from growth. The mean ages at T1, T2, and T3 were 8, 9, and 13 years. Results: There were no significant differences in PXB correction stability or treatment response at T2 and T3 among the 3 expanders. Expansion increased intermolar width by 5 mm and tipped each molar by 2.3 . At least 2 years after expander removal, molar width decreased by 1.3 mm, and the molars uprighted by 6 . Compared with noncrossbite norms, PXB subjects had narrower intermolar width before treatment and greater width after expansion, and were slightly wider at least 2 years posttreatment. Both younger age at T1 and retainer use resulted in statistically greater intermolar width at T3. Conclusions: Eighty-four percent of PXB correction remained with about one third of the initial expansion lost; retainer use and early treatment provided increased intermolar width. Haas, hyrax, and quad-helix appliances were equally effective. Slow maxillary expansion altered the PXB patients’ maxillary widths from narrower to slightly wider than the widths of the noncrossbite norms. (Am J Orthod Dentofacial Orthop 2009;136:331-9)

P

osterior crossbite (PXB) occurs with a relatively narrower maxilla than mandible.1 The prevalence of PXB is 8% to 23% in the deciduous and mixed dentitions,2-5 with less than 16% incidence of self correction.6 More than 90% of children with PXB are unilateral in centric occlusion. The maxilla is usually symmetrical, with bilateral presentation in centric relation.5 In habitual occlusion, the child shifts the mandible to 1 side, called a functional shift, resulting in the unilateral PXB in centric occlusion.5,7,8 Children with unilateral PXB exhibit asymmetric occlusion and condylar position, and, after treatment, symmetry is a

Private practice, Rockville, Md. Private practice, Vancouver, British Columbia, Canada. c Professor, Department of Orthodontics, School of Dentistry, University of Washington, Seattle. Supported by the University of Washington Orthodontic Alumni Association. The authors report no commercial, proprietary, or financial interest in the products or companies described in this article. Reprint requests to: Anne-Marie Bollen, Department of Orthodontics, School of Dentistry, University of Washington, Box 357446, Seattle, WA 93195; e-mail, [email protected]. Submitted, April 2007; revised and accepted, August 2007. 0889-5406/$36.00 Copyright Ó 2009 by the American Association of Orthodontists. doi:10.1016/j.ajodo.2007.08.026 b

improved.7 By contrast, adults with untreated unilateral PXB have skeletal asymmetry, indicating that untreated unilateral PXB in a child might lead to mandibular asymmetric development in adults.9-14 Early treatment of PXB aims to expand the maxilla, eliminate the functional shift, and restore condylar and facial symmetry for normal occlusal development.7 Thus, PXB is usually considered an indication for early treatment.8,15 In growing PXB patients, maxillary expansion has long been the treatment of choice with intermolar width deficiency.1 The most popular modality is rapid maxillary expansion (RME).16,17 RME is usually defined as 2 turns per day (0.5 mm expansion) and has a cumulative force of approximately 100 N across the midpalatal suture.18,19 In theory, RME applies forces to the posterior teeth without giving sufficient time for tooth movement to occur, so that the force is transferred to the palatal suture, resulting in more sutural opening than dental expansion.18,20 Some RME studies reported side effects such as microtrauma of the midpalatal suture21 and relapse.22 Linder-Aronson and Lindgren23 reported about 55% expansion loss at 5 years postretention for RME without fixed appliances (FA). Spillane and McNamara24 reported a 20% loss of the 6 mm of initial 331

332

Huynh et al

expansion 2.4 years postexpansion in a mixed RME sample that included other simultaneous treatments. A different modality is slow maxillary expansion (SME). SME is defined as 1 turn (0.25 mm of expansion) every second day for a Haas or hyrax appliance, or 1 molar width activation for a quad-helix, with 5 to 20 N of force.1,25 The theory is that the main resistance to the opening of the midpalatal suture is not the suture itself but the surrounding tissues such as the circummaxillary structures and midface sutures.18,25,26 In young patients, SME is said to provide the maximum rate at which the midface sutures can adapt, with minimum tearing and hemorrhaging compared with RME.18,21,25-27 Animal and histologic studies indicate that SME improves conservation of the suture and can produce a more stable result than RME.28-30 Some clinical studies also suggest that SME is more stable than RME25,31; however, these samples were small, and data collection was short term. Therefore, clinical SME studies with larger samples and long-term data are needed to validate its use. As age increases, the facial sutures become more interdigitated, especially after puberty.1,18,25,32 Sex also influences the maturation rate, with girls reaching skeletal maturity before boys.1 After puberty, greater force is required to open the sutures; this can surpass the capacity for physiologic adaptation.18,25,33 Various maxillary expander designs are believed to result in different skeletal-to-dental ratios of expansion and could lead to differences in treatment stability, but this assumption is not proven.34,35 To do so requires investigation of different SME appliances without other appliances to assess the pure effect of expansion type. Treatment effects and long-term outcomes have considerable clinical importance.22,34,36 Studies on early treatment of unilateral PXB concluded that canine grinding is a treatment of choice in the deciduous dentition.16,34 Most mixed dentition expansion studies have problems of small sample size, bias, confounding variables, lack of method analysis, lack of long-term data, no blinding for measurements, deficient statistical methods, and lack of controls.34 Lagravere et al36 reviewed SME without other simultaneous treatments and concluded that, without control groups, no strong conclusion can be made regarding the dental or skeletal changes after SME, but SME might be more stable than RME, and also expansion stability was related to age. In a maxillary expansion metaanalysis review, 4 long-term studies met the inclusion criteria with 2 studies of RME with FA,24,37 1 with RME alone,23 and 1 with SME alone.38 RME without FA seemed no better than SME, and the results of RME with FA do not have the pure effects of RME.22

American Journal of Orthodontics and Dentofacial Orthopedics September 2009

Therefore, a clinical SME study was needed with a large sample to evaluate the pure effect of expansion without supplementary appliances, considering growth, age, sex, and appliance design. We conducted a retrospective clinical study to evaluate the short- and longterm effects of SME on PXB correction stability, and maxillary intermolar width and angulation, in patients in the deciduous or early mixed dentition with PXB treated with a Haas, hyrax, or quad-helix appliance. MATERIAL AND METHODS

Our subjects were 312 consecutive, nonrandomized maxillary expansion patients, treated in the private orthodontic practice of an author (D.B.K.) in Vancouver, British Columbia, Canada, between January 1981 and November 2003. They were screened for inclusion in this retrospective study. Inclusion criteria were either unilateral or bilateral PXB at preexpansion (T1), treated with SME by using either a Haas, hyrax, or quad-helix appliance (Fig 1), and postexpansion and retention records with no PXB (T2). Records were taken a minimum of 2 years later (T3) in the permanent dentition. To prevent sample selection bias, patients who used a retainer between T2 and T3 were not excluded from the study. Subjects with surgical interventions, birth defects (clefts), or other simultaneous treatment, except for an anterior segmental archwire and a removable retainer, were excluded. The records and models were coded by someone not involved with the study. The coded models were given to the principal investigator (T.H.). Ages and dates were also obtained from the models. To prevent measurement bias, this investigator was blinded with regard to expander type, sex, and retainer use between T2 and T3. Models from T1, T2, and T3 were assessed separately; afterwards, they were sorted by expander type, sex, and retainer use. Coded records were reviewed in detail to verify expander type, expansion and retention protocol, retainer use, and any other treatment. SME was defined as no more than 1 turn every 2 days (0.25 mm every 2 days) for the Haas and the hyrax, or 1 molar width of activation for the quad-helix, until the PXB was mildly overcorrected so that the lingual incline of the mandibular buccal cusp contacted the buccal incline of the maxillary lingual cusp. The quadhelix was removed, activated, and recemented as needed. No subject was brought into buccal crossbite. The expander was left intraorally in a passive state for retention for a minimum of 6 months. A set of models in centric occlusion was taken after removal of the appliance (T2). At the minimum follow-up of 2 years later, progress models were taken in the early permanent dentition before further treatment if needed (T3).

Huynh et al

American Journal of Orthodontics and Dentofacial Orthopedics Volume 136, Number 3

333

Fig 1. SME appliances: A, Haas; B, hyrax; C, quad-helix.

Fig 2. Intermolar width was measured from the models as the intercentroid distance in millimeters between the maxillary first permanent molars.

Appliance choice and design were made independently of the crossbite severity by the treating clinician (D.B.K.). Quad-helices were used for expansion and when patients needed first permanent molar rotations corrected. Haas appliances were used when correction of digital habits was also needed. In the absence of these needs, either hyrax, Haas, or quad-helix appliances were used. The decision to use a removable retainer postexpansion was made by the clinician based on the patient’s needs. The study protocol was approved by the University of Washington’s Institutional Review Board. There were 2 components to the study as outlined. Part A included stability of crossbite correction. Models were evaluated by visual inspection for recurrence of PXB at T3. Subjects without models at T2, but with T3 models taken more than 2.5 years after T1, were included in part A. All subjects were relatively young with unerupted second permanent molars at T1 and T2; consequently, PXB of the second molars at T3 was not considered as relapse. Subjects with borderline recurrence of PXB at T3 were reviewed by a second investigator (A.B.), and consensus was reached. Part B included measurements of intermolar width and angulation, and comparison with norms. Intermolar

widths and angulations were measured for subjects with maxillary first permanent molars at T1, T2, and T3. Predicted normal intermolar width was calculated with a quadratic interpolation by using published norms matched with the sex and chronological age of the subjects (years and months).39 Comparisons of intermolar width were made. The following measurements were made from the models. 1.

2.

3.

Intermolar width (W) is the intercentroid distance in millimeters between the maxillary first permanent molars, using the method described by Moyers et al39 (Fig 2). Intermolar angulation is the angle of intersecting lines tangent to the mesiobuccal and mesiolingual cusp tips of the maxillary right and left first permanent molars as described by Handelman40 and Handelman et al.41 The angulation changes from T1 to T2 and T1 to T3 were calculated as half of the angular change: a negative number for molar tipping to the buccal aspect, and a positive number for uprighting, with T1 as the baseline (Fig 3).

334

Huynh et al

American Journal of Orthodontics and Dentofacial Orthopedics September 2009

Table III. Means and standard deviations for age, intermolar width, and intermolar angulation at T1 for subjects with first permanent molars at T1

Fig 3. Angulation changes from T1 to T2 and T1 to T3 were calculated as half of the angular change: a negative number for molar tipping to the buccal aspect and a positive number for uprighting, using T1 as the baseline. T(2,1) 5 (154 – 166 )/2 5 6 (tipping). Table I.

Included subjects

Expander type

Part A: PXB correction (n)

Part B: intermolar width and angulation (n)

74 41 45 160

57 32 43 132

Haas Hyrax Quad-helix Total

Table II.

T1 (PXB–) T2 (PXB1) RME used FFA, 2 3 4, headgear, Nance Craniofacial anomaly Other expander (RA and W arch) Missing models Incomplete record T2-T3 \ 2 y T1-T3 \ 2.5 y

Subjects (n) 12 9 15 20 5 15 59 39 10 5

, no crossbite present; 1, crossbite present.

5.

Intermolar expansion (E): E(2,1), E(3,1), E(3,2) reflect the intermolar expansion between T1, T2, and T3. For example, E(2,1) 5 width at T2 – width at T1, or the intermolar expansion at T2. Elapsed time was calculated as T3 minus T2, the time between T2 and T3 in months.

Quad-helix (n 5 43)

Hyrax (n 5 32)

All (n 5 132)

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

8.1 (1.1) 9.2 (1.3) 13.2 (1.7) 41.4 (2.5)

8.3 (1.0) 9.4 (1.3) 13.3 (1.6) 42.6 (2.5)

7.8 (1.1) 8.8 (1.3) 13.0 (1.3) 42.8 (2.6)

8.1 (1.1) 9.2 (1.3) 13.2 (1.5) 42.1 (2.6)

156.1 (6.8)

154.3 (9.9)

155.7 (7.2)

155.4 (8.0)

Percentages and numbers of subjects by sex, those using a retainer or no retainer, and those experiencing relapse at T3 as a total group and as a no-retainer group

Table IV.

Excluded subjects

Exclusion reason

4.

Age at T1 (y) Age at T2 (y) Age at T3 (y) Width at T1 (mm) Angulation at T1 ( )

Haas (n 5 57)

Female Male Retainer No retainer Total relapse Relapse, no retainer

Haas (n 5 74)

Quad-helix (n 5 45)

Hyrax (n 5 41)

All (n 5 160)

Percent (n)

Percent (n)

Percent (n)

Percent (n)

68.9 (51) 31.1 (23) 10.8 (8) 89.2 (66) 13.5 (10) 15.2 (10)

73.3 (33) 26.7 (12) 8.9 (4) 91.1 (41) 20.0 (9) 17.1 (7)

63.4 (26) 36.6 (15) 26.8 (11) 73.2 (30) 14.6 (6) 16.7 (5)

69.2 (110) 31.3 (50) 14.4 (23) 85.6 (137) 15.6 (25) 16.1 (22)

the null hypothesis of no difference between observed and predicted widths by using 1-sample t tests. We rejected the null hypothesis if the P value was less than 0.05 because it results in a 5% type 1 error probability. Twenty subjects were randomly chosen, and intermolar width and angulation were remeasured and the measurement error calculated according to Dahlberg’s formula.42 Measurement errors were insignificant. Standard deviations for width measurements ranged from 0.04 to 0.16 mm with correlation coefficients of 0.99 to 0.999. Angular measurements showed standard deviations of 0.95 to 1.5 with correlation coefficients of 0.91 to 0. 97.

Statistical analysis

Regression models were used to evaluate associations between variables. We found that these models were not significantly affected by outliers. Expander type, sex, and retainer use were adjusted to eliminate confounding by these variables. The expansion amount at T2, tipping at T2, and elapsed time were also adjusted. Differences between the observed intermolar widths and predicted normal values were tested with

RESULTS

Of the 312 consecutive expansion patients, 160 PXB subjects satisfied the inclusion criteria (Table I). The reasons for excluding the other 152 patients are given in Table II. Twenty-eight subjects from part A were excluded in part B (20 because of unerupted first molars at T1, and 8 because of missing T2 models) (Tables I and II).

Huynh et al

American Journal of Orthodontics and Dentofacial Orthopedics Volume 136, Number 3

Logistic regression for association between risk of relapse and appliance, sex, age, retainer use, elapsed time, and expansion amount at T2

Table V.

Appliance Haas (relative to quad-helix) Hyrax (relative to quad-helix) Male (relative to female) Age at T1 Retainer use (relative to no retainer use) Elapsed time (T2-T3) Expansion at T2 (W2-W1) Tipping at T2

OR

95% CI

P

0.68 0.33 0.68 0.93 0.94

0.23, 1.96 0.06, 1.30 0.20, 2.01 0.56, 1.51 0.18, 3.77

0.47* 0.13* 0.51* 0.76* 0.94*

0.91 0.85 0.88

0.60, 1.30 0.62, 1.11 0.76, 1.01

0.63* 0.26* 0.07*

OR, Odds ratio. *Not significant.

Table VII. Results of linear regression model for association between the changes in intermolar width from T1 to T2 and appliance, sex, and age Coefficient Appliance Haas (relative to Quad-helix) Hyrax (relative to Quad-helix) Male (relative to female) Age at baseline (T1)

0.06 0.49 0.20 0.09

Haas (n 5 57)

Quad-helix (n 5 43)

Hyrax (n 5 32)

All (n 5 132)

Mean (SD) Mean (SD) Mean (SD) Mean (SD) Width at T1 (mm) 41.4 (2.5) Width at T2 (mm) 46.6 (2.7) Expansion at 5.3 (1.8) T2 (mm) Width at T3 (mm) 45.3 (2.8) Expansion at 4.0 (2.2) T3 (mm)

42.6 (2.5) 47.7 (2.8) 5.1 (2.1)

42.8 (2.6) 47.5 (2.8) 4.7 (1.3)

42.1 (2.6) 47.2 (2.8) 5.1 (1.8)

46.0 (2.6) 3.4 (1.8)

46.4 (3.0) 3.6 (1.6)

45.8 (2.8) 3.7 (1.9)

Sample demographics are given in Table III. There were no statistical differences in age, sex, intermolar width, and intermolar angulation among the 3 expander types at T1. This confirmed an unbiased assignment to the appliances. The average ages at T1, T2, and T3 were 8, 9, and 13 years (Table III). Of the 160 patients, 23 used a removable retainer between T2 and T3 (Table IV); 69% of the sample were girls, and 31% were boys (Table IV). The overall PXB relapse in 160 subjects was 15.6% (Table IV); of the137 subjects without a retainer, the probability of relapse was slightly higher: 16.1%. Relapse rates were lowest for Haas, followed by hyrax, and then quad-helix, but the differences were not statistically significant (Table V). Appliance type, sex, age of treatment, retainer use, elapsed time, expansion amount at T2, and tipping at T2 were not associated with relapse at T3 (Table V). Although relapse of PXB occurred in only 15% of the patients, Table VI shows that only 3.7 mm of an average 5-mm expansion at T2 remained at T3. No statis-

95% CI

0.85, 0.74 1.44, 0.46 0.54, 0.94 0.43, 0.24

P

0.89* 0.31* 0.59* 0.58*

*Not significant.

Table VIII. Results of linear regression model for association between change in intermolar width from T1 to T3 and appliance, sex, age, and time between T2 and T3 Coefficient

Means and standard deviations for age, intermolar widths, and expansions for subjects with first permanent molars at T1

Table VI.

335

Appliance Haas (relative to Quad-helix) Hyrax (relative to Quad-helix) Male (relative to female) Age at baseline(T1) Retainer use (relative to no retainer use) Elapsed time (T2-T3)

95%Cl

P

0.45 0.15 0.19 0.42 0.98

0.30, 1.19 1.04, 0.75 0.53, 0.90 0.76, 0.09 0.03, 1.92

0.24* 0.74* 0.60* 0.01† 0.04†

0.12

0.11, 0.35

0.29*

†

*Not significant; significant.

tically significant association was found between the intermolar width change and expander type, sex, and age (Table VII). At T3, intermolar width change was significantly negatively associated with age at treatment and positively associated with retainer use (P\0.05) but not significantly associated with appliance type, sex, and elapsed time between T2 and T3 (P .0.05). All other variables being equal, for each year earlier that treatment started, the patients had 0.42 mm more expansion in the long term. Similarly, other variables being equal, the retainer group maintained 0.98 mm more intermolar width in the long term, compared with the nonretainer group (Table VIII). Because expander type had no effect on intermolar widths at T2 and T3, all 3 subsamples were grouped together (n 5 132) to characterize the effect of SME. Paired t tests indicated that the 3.7 mm of expansion from T1 to T2 was highly statistically significant, as was the relapse of 1.3 mm in width between T2 and T3 (P \0.0001). On average, the molars tipped 2.3 during treatment and then uprighted 6 after removal of the expander (Table IX). No significant association was found between intermolar angulation from T1 to T2 and appliance type, sex, and age (P .0.05) (Table X). Intermolar

336

Huynh et al

American Journal of Orthodontics and Dentofacial Orthopedics September 2009

Table IX. Means and standard deviations for ages, and intermolar angulations for subjects having first permanent molars at T1

Age at T1 (y) Angulation at T1 ( ) Age at T2 (y) Angulation at T2 ( ) Tipping at T2 ( ) (T2,1) Age at T3 (y) Angulation at T3 ( ) Uprighting T2-T3 ( ) (T3,2) Uprighting at T3 ( ) (T3,1)

Haas (n 5 57)

Quad-helix (n 5 43)

Hyrax (n 5 32)

All (n 5 132)

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

8.1 (1.1) 156.1 (6.8) 9.2 (1.3) 150.5 (9.4) 2.8 (3.7) 13.2 (1.7) 163.1 (7.1) 6.3 (4.5) 3.5 (4.1)

8.3 (1.0) 154.3 (9.9) 9.4 (1.3) 151.5 (11.3) 1.4 (4.2) 13.3 (1.6) 163.6 (8.9) 6.1 (4.7) 4.7 (3.3)

7.8 (1.1) 155.7 (7.2) 8.8 (1.3) 150.3 (9.1) 2.7 (3.8) 13.0 (1.3) 163.0 (7.4) 6.3 (4.4) 3.6 (4.0)

8.1 (1.1) 155.4 (8.0) 9.2 (1.3) 150.8 (9.9) 2.3 (3.9) 13.2 (1.5) 163.2 (7.8) 6.2 (4.5) 3.9 (3.8)

Table X. Results of linear regression of association between intermolar angulation from T1 to T2 and appliance, sex, and age, with linear effects of appliance relative to quad-helix Coefficient Appliance Haas Hyrax Male Age at baseline (T1)

1.25 1.03 0.01 0.40

95% CI

2.83, 0.32 2.92, 0.85 1.46, 1.47 0.27, 1.07

Table XI. Results of linear regression of association between intermolar angulation from T1 to T3 and appliance, sex, age, and time elapsed (T2-T3), with linear effects of appliance relative to the quad-helix

P

0.12* 0.28* 0.99* 0.24*

*Not significant.

Coefficient Appliance Haas Hyrax Male Age at baseline (T1) Retainer use Elapsed time (T2-T3)

1.1 1.05 0.23 0.3 0.14 0.54

95% CI

2.63, 2.87, 1.23, 0.38, 2.08, 0.08,

0.42 0.77 1.69 0.99 1.79 1.01

P

0.15* 0.25* 0.76* 0.38* 0.88* 0.02†

*Not significant; †significant.

angulation change from T1 to T3 was significantly positively associated with elapsed time (P \0.05), but not with expander type, sex, or age at T1 (P .0.05). All other variables being equal, for each year longer than 2 years out of the appliance, the molar uprighted 0.54 more (Table XI). Because expander type had no effect on intermolar angulation, the 3 subsamples were pooled (n 5 132) to show the effect of SME. Paired t tests indicated that 2.3 of tipping during treatment (T1-T2) and overall 3.9 of uprighting (T1-T3) were highly statistically significant (P \0.0001). Table XII shows that the correlations between intermolar angulation changes with intermolar width changes were statistically significant. Subjects with greater molar tipping tended to have greater expansion at T2 and T3 (P \0.0001); subjects with greater tipping at T2 tended to have greater tipping at T3 and greater expansion at T3. Relative to age- and sex-specific normal values, intermolar widths were statistically significantly narrower than the predicted normal values at T1 (P \0.05) and significantly wider than the predicted normal values at both T2 and T3 (P \0.0001) (Fig 4, Table XIII).

DISCUSSION

To our knowledge, this is the first SME study with a large sample and several types of fixed expanders with long-term data but without other simultaneous treatments that could contribute to the expansion effect. Based on the similarity of the 3 expander subgroups at T1, the large sample, and the blinded measurements, our data strongly support the conclusion that the Haas, hyrax, and quad-helix are equally effective for PXB correction, intermolar expansion, and intermolar angulation. This finding should carry more weight than previous studies because of the large sample and the long-term follow-up. Our data support the assessmemt of Harrison and Ashby16 that there is no evidence of a difference in treatment effect between bonded vs banded expanders, or quad-helix vs removable expander in SME. Petren et al34 also reached a similar conclusion. A likely explanation is that, in young patients, sutures open easily to allow expansion under light forces regardless of expander type.18,33 Since there was no difference among the 3 expanders, we will discuss them together as SME.

Huynh et al

American Journal of Orthodontics and Dentofacial Orthopedics Volume 136, Number 3

Correlations between tipping and expansion with 95% confidence intervals

Table XII.

Correlation Expansion T2, tipping T2 Expansion T3, tipping T3 Tipping T2, tipping T3 Tipping T2, expansion T3

0.35 0.51 0.32 0.23

95% CI

Table XIII. Differences between predicted and observed intermolar values at T1, T2, and T3

P

0.49, 0.19 \0.0001* 0.63, –0.37 \0.0001* 0.15, 0.46 0.0002* 0.38, –0.06 0.0092*

337

Mean T1 T2 T3

0.456 3.89 0.861

95% CI 0.868, 0.045 3.434, 4.347 0.398, 1.323

P 0.03* \0.0001† \0.0001†

*Significant; †very significant.

P values are based on t tests. *Very significant.

Fig 4. Relative to age- and sex-specific normal values, intermolar widths were statistically significantly narrower than the predicted normal values at T1 (P\0.05) and significantly wider than the predicted normal values at both T2 and T3 (P \0.0001).

Regarding PXB correction, after an average of 4 years after retention, SME with its relapse rate 15.6% was quite stable in the hierarchy of stability.1 The prevalence of crossbites was significantly higher in girls than in boys but similarly distributed among the 3 expander types. This distribution was consistent with the findings of previous studies about the sex differences of PXB prevalence.43,44 Sex in our data did not alter the outcome; this was probably because the subjects were prepubertal at T1 (age, 8 years). Sexual dimorphism in growth is typically expressed only after puberty.45 Intermolar width at T3 was inversely related to age at T1. When the patients were treated at a younger age, width at T3 was the greatest. As a patient ages, the sutures become more interdigitated, possibly making them more prone to some expansion loss.18,19,25,33 This finding supports early treatment for better longterm results. By including subjects using retainers (n 5 23), we avoided sample bias and confirmed a better long-term expansion result with retention, as indicated by previously studies.22,25,38 The retainer group had 0.98 mm more intermolar width at T3 than did the nonretainer

group. This was about 20% of the 5-mm expansion at T2 and 27% of the 3.7 mm of remaining expansion at T3. This finding was consistent with the trend in part A; the nonretainer subgroup had a worse relapse rate than the total group. The few (n 5 23) retainer subjects was insufficient to follow a binomial distribution, but, indirectly, it appears that the retainer subjects were slightly less prone to relapse than those without retainers. Previously, Hesse et al7 used 61 PXB patients from this sample at T1 and T2; although they used different landmarks for intermolar width, the same 5-mm expansion between T1 and T2 was found. The meta-analysis of Schiffman and Tuncay22 included 4 other studies with at least 1-year postretention data, 3 RME studies,23,24,37 and 1 SME study.38 Spillane and McNamara24 and Moussa et al37 showed higher expansion of 5.5 mm using RME with FA and retainers, but their various appliances confuse interpretation. LinderAronson and Lindgren23 used RME alone and achieved 3.6 mm of long-term expansion at T3 compared with an initial 8-mm expansion at T2. This suggests that the end result of RME is no better than SME. Using SME, Boysen et al38 found that 3.6 mm of expansion remained after 5 mm of initial expansion at T2, with either a quadhelix or a removable expander; the quad-helix caused greater expansion and buccal translation, whereas the removable expander gave more buccal tipping and long-term relapse. Using meta-analysis, Schiffman and Tuncay22 found that an average of 2.4 mm of expansion remained after 3.88 mm of expansion at least a year out of retention. However, their subjects were 3 years older on average than ours. Our study predicted 0.44 mm less long-term expansion for each year of age at T1. Thus, we would expect about 1.2 mm less final intermolar width than in the study of Schiffman and Tuncay; this was the case. The significant buccal molar tipping at T2 was consistent with the notion that dental tipping occurs during expansion.46 We found 2.3 of tipping after expansion and retention; this is mild compared with 3.7 of molar tipping reported for RME with the hyrax.47 The 6 of molar uprighting from T2 to T3 could be a result of

338

Huynh et al

expansion relapse and natural occlusal changes. Marshall et al48 found 3.3 of molar uprighting naturally from the mixed to the permanent dentition without orthodontic treatment. Interestingly, for every year beyond 2 years, there was 0.54 more uprighting, and the residual buccal tipping at T3 compared with T2 implied that not all dental tipping relapses. No previous study either supports or refutes these findings. The pattern of facial growth is established early in life and rarely changes significantly without treatment.1 At T1, the PXB subjects had intermolar widths about 0.5 mm narrower than the norms; we presumed that they would continue to be narrower than the treated norms at T2 and T3. However, at T2 and T3, the treated subjects were significantly wider than the norms. Due to normal growth, Hesby et al49 found only 2.3 mm of maxillary intermolar width increase in 36 untreated Class I subjects from ages 7.6 to 12.9 years using different landmarks. That finding was consistent with the norms we used for 8- to 13-year-olds.50 In the long term, we found a 3.7-mm intermolar width increase from 8 to 13 year olds in treated PXB patients. Skieller51 showed that the rate of growth at the midpalatal suture was significantly greater than growth during the follow-up period, indicating that expansion stimulated growth at the suture. Statistically, we can conclude that SME altered the maxillary width of PXB patients from narrow to at least normal in the long term. As a retrospective study, the level of evidence is limited, even though appliance distribution appeared to be unbiased. Furthermore, an untreated PXB control group is needed to evaluate the potentially altered growth of PXB patients. But such a study or randomized controlled trial might not be ethical and realistic. Another limitation was that our sample had a mixed ethnic makeup, whereas the norms were from white subjects. Similarly, a direct comparison between RME and SME is difficult, because a valid study would require comparable RME and SME treated samples, having the same pretreatment condition. Most RME studies used older patient samples and involved FA in addition to RME.17,32,37 Our sample included patients with both unilateral and bilateral PXB. It might be valuable to separate the 2 types of PXB and quantify the differences, if any, between these patients. Further comparisons between subjects in the deciduous, early, and late mixed dentition might provide useful information. CONCLUSIONS

Based on this SME study, we concluded the following. 1.

For PXB correction, SME has a stability rate of 84%.

American Journal of Orthodontics and Dentofacial Orthopedics September 2009

2. 3. 4. 5.

6.

Haas, hyrax, and quad-helix produce similar results. No sex effects were observed. Long-term results were improved by early treatment and retainer use. During treatment, the molars tipped about 2.3 , and, after removal of the expander, the molars uprighted about 6 , with not all dental tipping lost in the long term. Expansion altered the maxillary width of PXB patients from narrower than the norms at T1 to at least normal at T2 and T3.

We thank Sue Herring, Rebecca Hubbar, Lindsay Kennedy, Dr Kennedy’s office staff, Greg Huang, Terry Wallen, Jessica Lee, and Audrey Isaacson. REFERENCES 1. Profitt W. Contemporary orthodontics. St Louis: Mosby; 2000. 2. Egermark-Eriksson I, Carlsson GE, Magnusson T, Thilander B. A longitudinal study on malocclusion in relation to signs and symptoms of cranio-mandibular disorders in children and adolescents. Eur J Orthod 1990;12:399-407. 3. Heikinheimo K, Salmi K. Need for orthodontic intervention in five-year-old Finnish children. Proc Finn Dent Soc 1987;83: 165-9. 4. Kutin G, Hawes RR. Posterior cross-bites in the deciduous and mixed dentitions. Am J Orthod 1969;56:491-504. 5. Thilander B, Myrberg N. The prevalence of malocclusion in Swedish schoolchildren. Scand J Dent Res 1973;81:12-21. 6. Lindner A. Longitudinal study of the effect of early interceptive treatment in 4-year old children with unilateral cross-bite. Scand J Dent Res 1989;97:432-8. ˚ rtun J, Joondeph DR, Kennedy DB. Changes in con7. Hesse KL, A dylar postition and occlusion associated with maxillary expansion for correction of functional unilateral posterior crossbite. Am J Orthod Dentofacial Orthop 1997;111:410-8. 8. Kennedy DB, Osepchook M. Unilateral posterior crossbite with mandibular shift: a review. J Can Dent Assoc 2005;71:569-73. 9. Miyawaki S, Tanimoto Y, Araki Y, Katayama A, Kuboki T, TakanoYamamoto T. Movement of the lateral and medial poles of the working condyle during mastication in patients with unilateral posterior crossbite. Am J Orthod Dentofacial Orthop 2004;126:549-54. 10. Mupparapu M, Parisi E, DeRossi SS. Temporomandibular joint disc disfigurement and abnormal thickening of the posterior band. Gen Dent 2003;51:256-8. 11. Riolo ML, Brandt D, TenHave TR. Associations between occlusal characteristics and signs and symptoms of TMJ dysfunction in children and young adults. Am J Orthod Dentofacial Orthop 1987;92:467-77. 12. Nebbe B. Transverse skeletal and dental asymmetry in adults with unilateral lingual posterior crossbite. Am J Orthod Dentofacial Orthop 2005;127:15-6. 13. Motegi E, Miyazaki H, Ogura I, Konishi H, Sebata M. An orthodontic study of temporomandibular joint disorders. Part 1: epidemiological research in Japanese 6-18 year olds. Angle Orthod 1992;62:249-56. 14. Pinto AS, Buschang PH, Throckmorton GS, Chen P. Morphological and positional asymmetries of young children with functional

American Journal of Orthodontics and Dentofacial Orthopedics Volume 136, Number 3

15. 16. 17.

18. 19.

20. 21.

22. 23. 24. 25.

26. 27.

28.

29. 30. 31.

32. 33. 34.

unilateral posterior crossbite. Am J Orthod Dentofacial Orthop 2001;120:513-20. Turpin DL. Good time for discussion of early treatment. Am J Orthod Dentofacial Orthop 2000;118:247. Harrison JE, Ashby D. Orthodontic treatment for posterior crossbites. Cochrane Database Syst Rev 2001:CD000979. McNamara JA Jr, Baccetti T, Franchi L, Herberger TA. Rapid maxillary expansion followed by fixed appliances: a long-term evaluation of changes in arch dimensions. Angle Orthod 2003;73:344-53. Bell RA. A review of maxillary expansion in relation to rate of expansion and patient’s age. Am J Orthod 1982;81:32-7. Henry RJ. Slow maxillary expansion: a review of quad-helix therapy during the transitional dentition. ASDC J Dent Child 1993;60:408-13. Haas AJ. Long-term posttreatment evaluation of rapid palatal expansion. Angle Orthod 1980;50:189-217. Akkaya S, Lorenzon S, Ucem TT. Comparison of dental arch and arch perimeter changes between bonded rapid and slow maxillary expansion procedures. Eur J Orthod 1998;20:255-61. Schiffman PH, Tuncay OC. Maxillary expansion: a meta analysis. Clin Orthod Res 2001;4:86-96. Linder-Aronson S, Lindgren J. The skeletal and dental effects of rapid maxillary expansion. Br J Orthod 1979;6:25–9. Spillane LM, McNamara JA Jr. Maxillary adaptation to expansion in the mixed dentition. Semin Orthod 1995;1:176-87. Hicks EP. Slow maxillary expansion. A clinical study of the skeletal versus dental response to low-magnitude force. Am J Orthod 1978;73:121-41. Mew J. Relapse following maxillary expansion: a study of twentyfive consecutive cases. Am J Orthod 1983;83:56-61. Bell RA. The effects of maxillary expansion using a quad-helix appliance during the deciduous and mixed dentitions. Am J Orthod 1981;79:152-61. Cotton LA. Slow maxillary expansion: skeletal versus dental response to low magnitude force in Macaca mulatta. Am J Orthod 1978;73:1-23. Ohshima O. Effects of lateral expansion force on the maxillary suture in cynomolgus monkey. J Osaka Dent Univ 1972;6:11-50. Storey E. Tissue response to the movement of bones. Am J Orthod 1973;64:229-47. Mossaz-Joelson K, Mossaz CF. Slow maxillary expansion: a comparison between banded and bonded appliances. Eur J Orthod 1989;11:67-76. Baccetti T, Franchi L, Cameron CG, McNamara JA Jr. Treatment timing for rapid maxillary expansion. Angle Orthod 2001;71:343-50. Melsen B. Palatal growth studied on human autopsy material. A histologic microradiographic study. Am J Orthod 1975;68:42-54. Petren S, Bondemark L, Soderfeldt B. A systematic review concerning early orthodontic treatment of unilateral posterior crossbite. Angle Orthod 2003;73:588-96.

Huynh et al

339

35. Turpin DL. Dealing with posterior crossbite in young patients. Am J Orthod Dentofacial Orthop 2004;126:531-2. 36. Lagravere MO, Major PW, Flores-Mir C. Skeletal and dental changes with fixed slow maxillary expansion treatment: a systemic review. J Am Dent Assoc 2005;136:194-9. 37. Moussa R, O’Reilly MT, Close JM. Long-term stability of rapid palatal expander treatment and edgewise mechanotherapy. Am J Orthod Dentofacial Orthop 1995;108:478-88. 38. Boysen B, La Cour K, Athanasiou AE, Gjessing PE. Threedimensional evaluation of dentoskeletal changes after posterior cross-bite correction by quad-helix or removable appliances. Br J Orthod 1992;19:97-107. 39. Moyers RE, van der Linden FPGM, Riolo ML, McNamara JA Jr. Standards of human occlusal development. Ann Arbor: Center for Human Growth and Development, The University of Michigan; 1976. 40. Handelman CS. Nonsurgical rapid maxillary alveolar expansion in adults: a clinical evaluation. Angle Orthod 1997;67: 291-305. 41. Handelman CS, Wang L, BeGole EA, Haas AJ. Nonsurgical rapid maxillary expansion in adults: report on 47 cases using the Haas expander. Angle Orthod 2000;70:129-44. 42. Dahlberg G. Statistical methods for medical and biological students. New York: Interscience Publications; 1940. 43. Helm S. Malocclusion in Danish children with adolescent dentition: an epidemiologic study. Am J Orthod 1968;54:352-66. 44. Erdinc AE, Ugur T, Erbay E. A comparison of different treatment techniques for posterior crossbite in the mixed dentition. Am J Orthod Dentofacial Orthop 1999;116:287-300. 45. Broadbent BH Jr, Golden WH. The value of an assessment of skeletal maturity in orthodontic diagnosis. Am J Phys Anthropol 1971;35:409-10. 46. Karaman AI. The effects of nitanium maxillary expander appliances on dentofacial structures. Angle Orthod 2002;72: 344-54. 47. Adkins MD, Nanda RS, Currier GF. Arch perimeter changes on rapid palatal expansion. Am J Orthod Dentofacial Orthop 1990; 97:194-9. 48. Marshall S, Dawson D, Southard KA, Lee AN, Casko JS, Southard TE. Transverse molar movements during growth. Am J Orthod Dentofacial Orthop 2003;124:615-24. 49. Hesby RM, Marshall SD, Dawson DV, Southard KA, Casko JS, Franciscus RG, et al. Transverse skeletal and dentoalveolar changes during growth. Am J Orthod Dentofacial Orthop 2006; 130:721-31. 50. Moorrees CF. Growth and development in orthodontics. Curr Opin Dent 1991;1:609-21. 51. Skieller V. Expansion of the midpalatal suture by removable plates, analysed by the implant method. Trans Eur Orthod Soc 1964;143-57.