early orthodontic treatment for class ii malocclusion ... · a result, correcting class ii...

SYSTEMATIC REVIEW

Early orthodontic treatment for Class IImalocclusion reduces the chance of incisaltrauma: Results of a Cochrane systematic review

Badri Thiruvenkatachari,a Jayne Harrison,b Helen Worthington,c and Kevin O'Briend

Manchester and Liverpool, United Kingdom

aNatiothodoKingdbHonoLiverpcProfeMancdProfeKingdAll auPotenAddreDentiM13Subm0889-Copyrhttp:/

In this article, we summarize the most clinically relevant findings of our recently updated Cochrane systematicreview into the treatment of Class II Division 1 malocclusion. Methods: A systematic review of the databaseswas performed to identify all randomized controlled trials evaluating early treatment with functional appliancesto correct Class II Division 1 malocclusion. Results: Three early treatment studies with data from 353 partici-pants were included in this review. The results showed no significant difference for any outcomes, exceptnew incidence of incisor trauma, which was significantly less for the early treatment group. The risk ratio analysisfor new incisor trauma showed that providing early treatment reduced the risk of trauma by 33% and 41% in thefunctional and headgear groups, respectively. However, when the numbers needed to treat were calculated,early treatment with functional appliances prevents 1 incidence of incisal trauma for every 10 patients (95%CI, 5-174), and headgear treatment prevents 1 incidence of incisal trauma for every 6 patients (95% CI, 3-23). Conclusions: Orthodontic treatment for young children, followed by a later phase of treatment whenthe child is in early adolescence, appears to reduce the incidence of new incisal trauma significantly comparedwith treatment that is provided in 1 phase when the child is in early adolescence. However, these data shouldbe interpreted with caution because of the high degree of uncertainty. There are no other advantages inproviding 2-phase treatment compared with 1 phase in early adolescence. (Am J Orthod DentofacialOrthop 2015;148:47-59)

In this article, we outline and discuss the most clini-cally relevant findings of our recently updatedCochrane systematic review into the treatment ofClass II malocclusion.1 This form of malocclusion affectsnearly a quarter of 12-year-olds in the United Kingdom 2

and 15% of 12- to 15-year-olds in the United States.3 Asa result, correcting Class II malocclusion is a commontreatment performed by orthodontists.

nal Institute of Health Research (NIHR) Academic Clinical Lecturer in or-ntics, School of Dentistry, University of Manchester, Manchester, Unitedom.rary senior lecturer/consultant orthodontist, Orthodontic Department,ool University Dental Hospital, Liverpool, United Kingdom.ssor, Cochrane Oral Health Group, School of Dentistry, University ofhester, Manchester, United Kingdom.ssor, School of Dentistry, University of Manchester, Manchester, Unitedom.thors have completed and submitted the ICMJE Form for Disclosure oftial Conflicts of Interest, and none were reported.ss correspondence to: Badri Thiruvenkatachari, JR Moore Bldg, School ofstry, University of Manchester, Oxford Rd, Manchester, United Kingdom9PL; e-mail, [email protected], August 2014; revised and accepted, January 2015.5406/$36.00ight � 2015 by the American Association of Orthodontists./dx.doi.org/10.1016/j.ajodo.2015.01.030

There has been extensive research into Class II treat-ment, and this was summarized in the previous versionof our Cochrane review, when we concluded that “earlytreatment of Class II malocclusion resulted in limitedadvantage when compared to providing treatment inone stage during adolescence.” Despite this high levelof evidence, these conclusions are still thought to becontroversial because the results of nonrandomizedinvestigations do not always agree with the results ofthe trials.4-10

Our review was originally published in 2008, and wehave updated it to ensure that additional research isincluded so that the conclusions remain contempo-rary.11 In this article, we outline the most important out-comes that are relevent to the timing of treatment.

MATERIAL AND METHODS

We identified studies using the Cochrane Oral HealthGroup's Trials Register (to April 17, 2013), the CochraneCentral Register of Controlled Trials (Cochrane Library2013, Issue 3), MEDLINE via OVID (1946 to April 17,2013), and EMBASE via OVID (1980 to April 17, 2013).Articles that were identified as part of the Cochrane

47

Delta:1_given nameDelta:1_surnameDelta:1_given nameDelta:1_surnamemailto:[email protected]://dx.doi.org/10.1016/j.ajodo.2015.01.030

48 Thiruvenkatachari et al

Oral Health Group's hand searching program wereobtained from the following journals:American Journalof Orthodontics and Dentofacial Orthopedics, AngleOrthodontist, European Journal of Orthodontics, Jour-nal of Orthodontics, and British Journal of Orthodon-tics. An example of the search strategy is shown in theAppendix. We included randomized controlled trialsthat looked at children or adolescents, or both, receivingorthodontic treatment to correct prominent maxillaryfront teeth. The participants had to be 16 years of ageor younger. We excluded trials that included participantswith cleft lip or palate or other craniofacial deformities orsyndromes. No language restrictions were placed on thestudies considered for inclusion in this review, and pub-lished or unpublished sources were considered. Forunpublished studies, an attempt was made to identifythem by contacting the first-named author of the trial re-ports. The study eligibility was assessed by 2 review au-thors (B.T. and K.O'B.) independently and in duplicate,and disagreements were resolved by discussions or clari-fications from the authors. Further details of themethod-ology that we used are included in the original review.1

For the purposes of this article, we reported on onlythe most clinically relevant subset of outcomes fromthe original review.1

The primary outcome measure was the prominenceof the maxillary front teeth (overjet).

The secondary outcomemeasures were the relationshipbetween the maxillary and mandibular jaws (cephalo-metric measurements), self-esteem and patient satisfac-tion (Piers Harris questionnaire), and incidence of incisaltrauma.

The risk of bias was evaluated according to theCochrane Collaboration's tool for assessing the risk ofbias, as described in the Cochrane Handbook for Sys-tematic Reviews of Interventions.12 This was assessedindependently by 2 authors (B.T. and another) againstthe following key criteria.

1. Sequence generation: This evaluation was based onexamination of the method used to generate theallocation sequence: eg, computer-generatedrandom numbers or random number tables.

2. Allocation concealment: These are the methods ofconcealing the allocation sequence from thoseassigning participants to the intervention groups.Did they use sealed envelopes, or was this doneremotely via an Internet site or telephoneallocation? These steps are taken so that the oper-ator in the study cannot influence the treatmentallocation.

3. Blinding of participants, personnel, and outcomeassessors: This ensures that the participants, clini-

July 2015 � Vol 148 � Issue 1 American

cians, and assessors are unaware of the interventionallocations. This is carried out to reduce the chancesthat the operators and the data handlers couldinfluence the results of the study. In orthodonticstudies, it is often difficult to conceal the treatmentallocation from participants and clinicians, but it isusually possible to blind the outcome assessors fordata collection and analysis.

4. Incomplete outcome data: This is an assessment ofthe possible effects of missing data caused by attri-tion or exclusion from analysis: eg, postrandomiza-tion dropouts. If there are many dropouts in atreatment group, this can introduce bias. This maybe an issue in orthodontic studies because ofrelatively high dropout rates as a result of the longduration of the studies.

5. Selective outcome reporting: Selectively reportingoutcomes: eg, chasing significance or not reportingharms. This is assessed to make sure that all data arereported, not just the outcomes that are consideredsignificant to the investigators.

6. Other sources of bias: Biases not covered elsewhere.

We summarized the overall risk of bias for each studyas low, unclear, or high.

Statistical analysis

The statistical analysis was performed according tothe statistical guidelines referenced in the CochraneHandbook for Systematic Reviews of Interventionsand facilitated by RevMan.12

Mean differences and 95% confidence intervals (CIs)were calculated for continuous data. Dichotomous out-comes were expressed as odds ratios (ORs) together with95% CIs. Any heterogeneity between trials was assessedwith the Cochran test and the I2 statistic. A meta-analysiswas performed on studies with similar comparisons thatreported the same outcome measures. We would haveused the random-effects models if there had beenmore than 3 studies in the meta-analysis and usedfixed-effect models if there were only 2 or 3 studies.When relevant, we calculated the numbers needed totreat and the risk ratio (RR).13

RESULTS

In our literature search, we initially identified a totalof 1572 records, of which 117 full-text records wereassessed. Of these, we excluded 57 articles; 10 additionalstudies were considered not relevant to this review.Seventeen trials were identified (published in 50 articles)(Fig 1); these included 3 early treatment trials and 14 latetreatment studies. The 3 early treatment trials are

Journal of Orthodontics and Dentofacial Orthopedics

Fig 1. Study flow diagram.

Thiruvenkatachari et al 49

included in this review. These studies were reported inseveral articles, so for ease of description, we have com-bined them into 3 broad descriptors: Florida,14-22 NorthCarolina,23-34 and United Kingdom mixed.35-38

Assessment of the risk of bias revealed the followingfor each study.

In the Florida studies, there was a high risk of bias in 2categories: bias in randomization method because theauthors used a stratified randomization procedure, butafter 3 years, the method was modified by allocationto groups (23% of the sample) because of the slow

American Journal of Orthodontics and Dentofacial Orthoped

recruitment rate; and attrition bias, since the dropoutrate was significantly higher for minority ethnic groups.

In the North Carolina studies, there was a high risk ofbias for blinding of the outcome assessment. The molarbands were left in place at the end of phase 1 data collec-tion, allowing the technician to identify these patients'treatment groups.

In the United Kingdommixed studies, there was a lowrisk of bias.

Further more detailed information on the risks of biasfor each study is given in Table I and Figure 2.

ics July 2015 � Vol 148 � Issue 1

Table I. Methodologic quality summary

Bias Authors' judgment Support for judgmentFlorida14-22

Random sequence generation (selection bias) High risk A stratified block randomization procedure was used“Subjects initially were selected in blocks of 6 and randomized to the

treatment protocols. This procedure of assigning subjects togroups only after a block had filled was modified in year 3, after werecognized slow entry rate and many partially filled blocks (23% ofthe sample) were randomized to groups”

Allocation concealment (selection bias) Unclear risk Not describedBlinding of outcome assessment (detection bias) Low risk “All cephalometric radiographs were encoded by the staff assistant

and then decoded for analysis”Incomplete outcome data (attrition bias) High risk Clear information on withdrawals. Dropouts, 24%. Number of

dropouts was approximately equal in each group, but the rate ofwithdrawal was significantly higher for subjects who were notwhite.

Selective reporting (reporting bias) Low risk All variables reportedOther biases Low risk No other sources of bias identified

North Carolina23-34

Random sequence generation (selection bias) Low risk “Randomization was performed within gender in blocks of sixpatients with Proc Plan in SAS”

Allocation concealment (selection bias) Unclear risk Not describedBlinding of outcome assessment (detection bias) High risk Because the molar bands were not removed at the end of phase 1, the

technician was not masked as to these patients' treatment groupIncomplete outcome data (attrition bias) Unclear risk Number of patients randomized in different groups not reported

192 randomized; 175 started, 166 finished phase 1; and 137 finishedphase 2

Dropout rates of 13.5% (low risk) for phase 1 and 28.6% (high risk)for phase 2. Reasons for dropouts reported, but not for eachtreatment group

Selective reporting (reporting bias) Low risk All variables reportedOther biases Low risk No other bias found

United Kingdom mixed35-38

Random sequence generation (selection bias) Low risk “The randomization was made at the start of the study with pre-prepared random number tables with a block stratification oncentre and sex”

Allocation concealment (selection bias) Low risk Randomization was carried out by using a central telephone line andminimization software

Blinding of outcome assessment (detection bias) Low risk Assessor blinded to outcomes. “The cephalograms and the studycasts were scored with the examiner unaware of the patient'sgroup”

Incomplete outcome data (attrition bias) Low risk Clear information on withdrawals, but rates different in each group:22/89 (25%) in the Twin-block group and 12/85 (14%) in thecontrol group.

Reasons for exclusion specified (unpublished data)Control group: 4 refused to consent to phase 2 treatment, 1 withdrew

due to illness, 3 had multiple DNAs with no final records, 1 movedaway or lost contact, 2 had Twin-blocks fitted in phase 1 in error, 1had a sore mouth and required treatment in phase 1

Treatment group: 2 moved away or lost contact, 9 hadmultiple DNAswith no follow-up records, 4 did not start because eligibilitycriteria were not met, 5 refused to continue, 1 had poor oral health,1 was removed from study because of health problems

Selective reporting (reporting bias) Low risk All variables reportedOther biases Low risk Groups appeared similar at baseline

DNA, Did not attend.


Three trials reported on early treatment for Class II Di-vision 1 malocclusion, and all were included in the meta-analyses. Detailed characteristics are described in Table II.


We looked at the outcomes of early (2-phase) inter-vention vs adolescent (1-phase) treatment at the conclu-sion of treatment.


Fig 2. Risk of bias summary for each included study.


Three trials with data from 343 participantscompared treatment for younger children with a func-tional appliance vs treatment for adolescents.14-38

When we evaluated the effects of treatment at ayoung age with a functional appliance (phase 1)followed by further treatment during adolescence(phase 2), at the conclusion of all treatment, wefound no statistically significant differences in finaloverjet (P 5 0.18), final ANB (P 5 0.92), peerassessment rating (PAR) score (P 5 0.34), andself-concept score (P 5 0.60) (Fig 3). However, theincidence of new incisal trauma showed statisticallysignificant results in favor of early treatment with afunctional appliance (P 5 0.04) compared with 1-phase treatment during adolescence only (OR, 0.59;95% CI, 0.35-0.99) (Fig 4). When the risk of traumawas evaluated, 29% (54 of 185) of patients who had1 course of treatment in adolescence had new traumacompared with only 20% (34 of 172) of patients hav-ing incisal trauma in the early treatment group. Thedifference in the risk was 9.7%. The RR analysisshowed that providing early treatment reduced therisk of new trauma by 33% of what it was when treat-ment was delayed and provided in 1 course in adoles-cence (RR, 0.67; 95% CI, 0.46-0.98). Importantly,calculating the measure of treatment impact—ie, the


number needed to treat—showed that early treatmentwith a functional appliance prevents 1 incidence ofnew incisal trauma for every 10 patients treated(95% CI, 5.4-175).

Two trials, with data from 285 children, comparedtreatment for young children who used headgear withadolescent (1-phase) treatment.14-34 There were nostatistically significant effects of an early course ofheadgear treatment at a young age followed bytreatment in adolescence with respect to final overjet(P 5 0.20), final ANB (P 5 0.32), and PAR score(P 5 0.16) (Fig 5) compared with treatment in adoles-cence only. However, the incidence of new incisaltrauma showed a statistically significant reduction inthe early treatment group (P 5 0.009), with an ORof 0.47 (95% CI, 0.27-0.83) (Fig 6). The adolescenttreatment group showed nearly twice the incidence ofnew incisal trauma, with 39% (47 of 120) comparedwith 23% (27 of 117) for the young group with earlyheadgear treatment. The RR showed a risk reductionof 41% for patients having early headgear treatment(RR, 0.59; 95% CI, 0.39-0.87). When we consideredthe numbers needed to treat, we found that early treat-ment with headgear prevented 1 incidence of incisaltrauma for every 6 patients treated (95% CI, 3.64-23.22).

When we evaluated any effect of early treatment witheither a functional appliance or headgear, we found nosignificant differences in final overjet (Fig 7), final ANB(Fig 5), PAR score (Fig 5), or the incidence of incisaltrauma (Fig 8) between the 2 interventions.14-38

DISCUSSION

The results of this updated review showed that foroverjet reduction, PAR score, and skeletal change,there are no benefits of early treatment, and the resultsof the original systematic review have not changed.However, the addition of new data showed that earlytreatment, with either a functional appliance or head-gear, resulted in a reduction in incisal trauma. Thisillustrates the value of continuously updating system-atic reviews.

It is worthwhile to consider the important outcomesin which early treatment had an effect in terms of theirability to reduce treatment uncertainty.

For the psychosocial impact, the United Kingdomstudy was the only one that reported in sufficient detailon self-concept.35,37 The results from this studysuggested that although self-concept and self-esteemimproved initially for patients having treatment early,at the end of phase 1 treatment, this effect was notmaintained to the end of phase 2 treatment in


Table II. Characteristics of studies

Study CharacteristicsFlorida14-22

Methods Location: University of FloridaNumber of centers: 1Recruitment period: not statedFunding source: funded by NIH (DE08715)Trial design: randomized parallel group study over 10 years

Participants Inclusion criteria: third or fourth grade at school, at least bilateral 1/2 cusp Class II molars or 1 side\1/2 cusp Class II if other side was greater than 1/2 cusp Class II. Fullyerupted permanent first molars, emergence of not more than 3 permanent canines or premolars, and positive overbite and overjet

Exclusion criteria: not willing to undergo orthodontic treatment or to be randomly allocated to treatment type. Poor general health, active dental or periodontal pathologyAge at baseline: mean, 9.6 yearsScreened child population (360) then referred to clinic for treatmentNumber randomized: 325 randomized; 277 started treatment: 95, 100, and 82 in bionator, headgear, and control groups, respectivelyNumber evaluated: end of treatment phase (1), 79/95, 92/100, and 78/82; end of retention phase, 75/95, 85/100, and 75/82; and end of follow-up (II), 70/95, 81/100,

and 74/82 in bionator, headgear, and control groups, respectivelyInterventions Group A: bionator appliance

Group B: cervical pull headgear with removable biteplaneGroup C: delayed treatment control3 phases of treatment: 2 years of early treatment plus 6 months retention plus further 6 months follow-up

Outcomes OverjetSkeletal discrepancyDental alignment measured with the PAR index

Notes Duration of randomized treatment: 2 years initiallySample size calculation not reported

North Carolina23-34

Methods Location: North CarolinaNumber of centers: 1Recruitment period: August 1988 to November 1993Funding source: grants from NIH, and Orthodontic Fund, Dental Foundation of North CarolinaTrial design: parallel group randomized controlled trial with 2 treatment phases

Participants Inclusion criteria: children with mixed dentition, with all permanent teeth developing, with growth potential throughout phase 1 of treatment. Overjet.7mm, all incisorserupted, second molars not erupted

Exclusion criteria: clinically obvious facial asymmetry, cleft or syndrome, more than 2 SD from normal vertical proportionality, and prior orthodontic treatmentAge group: mean, 9.4 years (SD, 1.0 year)Screened child population (2164) then referred to clinic for treatmentNumbers randomized: 192 randomized, 175 started treatmentNumbers evaluated: 53, 52, and 61 at the end of phase 1, and 39, 47, and 51 at the end of phase 2 for bionator, headgear, and control groups, respectively

Interventions Group A (n5 53): functional appliance—modified bionator with the bite taken with 4-6 mm of protrusion and minimal vertical opening. Reactivation of appliance whennecessary was by construction of a new appliance

Group B (n5 52): headgear—combination headgear with supershort outer bow, adjusted to deliver 8-10 oz to the head cap, with neck strap force just sufficient to preventbuccal flaring of maxillary molars

All appliances delivered within 1 month of patients' initial records being takenGroup C (n 5 61): control (observation only)

52Thiruvenkatachariet

al

July2015�

Vol148�

Issue1

American

JournalofOrthodontics

andDentofacialO

rthopedics

Table II. Continued

Study Characteristics

Outcomes Skeletal growth changes; maxilla, mandible, skeletal relationship, dental relationshipNotes Duration of intervention: phase 1, 15 months, and phase 2, 25.5, 30.1, and 34.5 months for functional, headgear, and control groups, respectively

Frequency of treatment visits: every 6-8 weeks for active treatment groups and every 6 months for control groupSample size calculation: sample size of 40 per group was calculated as necessary to detect a mean difference between any 2 groups equivalent to the doubling in

annualized change of SNPg (with alpha 5 0.01 and power of 0.90)Patients were rerandomized at the end of phase 2 for different clinicians

United Kingdom35-38

Methods Location: United KingdomNumber of centers: 13Recruitment period: March 1997 to June 1998.Funding source: Medical Research Council (99410454)Trial design: randomized parallel group trial

Participants Inclusion criteria: children in the mixed dentition with overjet greater than 7mm and willingness of the patient and a parent to participate in the study. The patients had tobe in the mixed dentition with at least the permanent incisors and first molars erupted, but there was no age criterion

Exclusion criteria: craniofacial syndromesAge at baseline: average ages were 9.7 (SD, 0.98) years for the treatment group and 9.8 (SD 0.94) years for the control groupNumber randomized: 174Number evaluated: 127

Interventions ComparisonGroup A: Twin-block early treatment: randomized, 89; completed, 67Group B: Twin-block delayed treatment: randomized, 85; completed, 73

Outcomes (trauma notnoted)

OverjetSkeletal discrepancy measured by the Pancherz analysisDental alignment measured with the PAR indexDuration of treatment

Notes Duration of intervention: phase 1, 15 months; phase 2, early treatment group, 14 months (435 days), late treatment group, 24 months (744 days).Sample size calculation: “This showed that the mean duration of treatment for patients who had later treatment after early treatment was 25 months (SD, 11). It was

decided that a meaningful difference between the treatment duration for children who did, or did not, receive early treatment was 6 months. To give a study a power of80% and an alpha of 0.05, the sample size needed to be 60 in each group”

Thiruvenkatacharietal

53

American

JournalofOrthodontics

andDentofacialO

rthopedicsJuly

2015�Vol148�

Issue1

Fig 4. Forest plot representing the incidence of new incisal trauma in patients receiving early treat-ment with a functional appliance compared with 1-phase treatment during adolescence only. M-H,Mantel-Haenszel.

Fig 3. Forest plot representing the effects of treatment at a young age with a functional appliance(phase 1) followed by a further treatment during adolescence (phase 2) and final overjet, final ANB,PAR score, and self-concept. IV, Instrumental variables.


adolescence. We can, therefore, conclude that providingtreatment did not make a positive impact in the longterm. Unfortunately, it appears that the effect of earlyorthodontic treatment diminishes with time. Neverthe-less, we do not know the effect of the increase inself-esteem that occurred after early intervention; thismay have clinical importance, particularly if a child issubjected to excessive teasing or bullying.


All 3 early treatment studies reported on the inci-dence of new incisal trauma.14,27,36 The Floridastudy used the modified Ellis classification with thehelp of study models, photographs, and x-rays.27 Inthe North Carolina study, the authors used the modi-fied National Health and Nutritional Examination Sur-vey (NHANES III), which was clinically assessed at eachstage of the study.14 Although the scores ranged from


Fig 6. Forest plot representing new incisal trauma in the early treatment and adolescent treatmentgroups. M-H, Mantel-Haenszel.

Fig 5. Forest plot representing effects of an early headgear treatment at a young age followed by treat-ment in adolescence with respect to final overjet, final ANB, and PAR score. IV, Instrumental variables.


0 to 7 for both studies, the scoring criteria weredifferent. In the United Kingdom study, the incidenceof trauma was simply recorded as “yes” or “no,” withno information on the severity of the injury.36 In allthese studies, a reliability assessment was performedfor the assessors.

The meta-analyses showed a statistically and clini-cally significant reduction in incisal trauma after earlytreatment with both functional appliances and head-gear. However, we need to interpret this finding withcaution and consider all the data, particularly the RRsand the numbers needed to treat.

The RR is a measure of an event happening in onegroup compared with the risk of the event happeningin another group. For example, the RR of 0.67 that wereported for early treatment with a functional appliancemeans that the risk of incisal trauma is reduced by 33%


in the early treatment group, when compared with theadolescent treatment group. This suggests thatproviding treatment early is likely to have a benefit.However, when we examined the confidence intervals,these were wide (0.46-0.98). This means that the riskreduction could be as high as 54% or as low as 2%. Asa result, there is a high degree of uncertainty aboutthis finding.

Another useful way to interpret these data is toconsider the numbers needed to treat. This is the numberof patients that one needs to treat to prevent 1 addi-tional adverse outcome: in this case, 1 episode oftrauma. When this was calculated for early treatmentwith a functional appliance, it showed that we neededto treat 10 patients to prevent 1 episode of trauma.The 95% CIs (5-175) again showed a high degree ofuncertainty.


Fig 8. Forest plot representing effect of early treatment with either a functional appliance or headgearand the incidence of incisal trauma. M-H, Mantel-Haenszel.

Fig 7. Forest plot representing effect of early treatment with either a functional appliance or headgearand final overjet, final ANB, and PAR score. IV, Instrumental variables.


Similar data were found for the effect of earlytreatment with headgear; the risk reduction was41% (RR, 0.59; 95% CI, 0.39-0.87), with numbersneeded to treat of 6 (95% CI, 3-23). Although thesefigures show a lower degree of uncertainty, it is stillconsiderable.

In 2 of the 3 studies, the majority of the injuries werescored as minor fractures involving the enamel only.14,27

In the Florida study, 80% of the patients had enamel-only fractures, 19% had enamel and dentin fractures,and 1 patient had pupal involvement.27 In the NorthCarolina study, at the end of phase 2, most new injurieswere craze lines or enamel-only fractures.14 One patient


had enamel and dentin fracture that required resin resto-ration, and another patient required endodontic treat-ment for a pulpal injury. In the United Kingdom trial,the authors reported only the presence or absence ofincisor trauma. However, all 3 studies reported a signif-icant number of patients with incisor trauma before thestart of the trial.

It is interesting that in the North Carolina andthe Florida studies, the incidences of trauma were notstatistically significantly different between boys andgirls.14,27 In the North Carolina study, the differencewas statistically significant for the control group, withsignificantly more boys having incisal trauma.14



However, no study provided data on sex, and so a meta-analysis was not possible in this review.

It would have been useful to have additional infor-mation on the sporting activities of the patientsinvolved in the trials, but none of the 3 studies re-ported on this and also whether mouthguards were rec-ommended.

It is also relevant to consider that although the re-view showed that early treatment may reduce trauma,other factors such as cost, patient compliance, andrisks associated with lengthy treatment should beconsidered before deciding on treatment timing. Thecost savings of preventing incisor trauma must bebalanced against the greater cost of providingtreatment in 2 phases.

Whereas the findings on the reduction of trauma areencouraging, it is clear that they should be interpreted inrelation to the high level of uncertainty. As a result, wesuggest that the prevention of trauma should not bethe only reason for routinely providing early treatmentfor Class II malocclusion. This decision should be takenas part of a risk evaluation that considers the size ofoverjet and the child's engagement in activities that in-crease the risk of trauma.

For the dental and skeletal outcomes, analyses ofthe PAR scores, cephalometric measurements, andoverjet reduction showed no statistically or clinicallysignificant effect of providing early orthodontic treat-ment. This was not surprising because this was theconclusion of the 3 large-scale randomized trials, andthese findings have not changed since the original re-view. It is therefore clear that in terms of morphologicoutcomes of orthodontic treatment, there are no ad-vantages to early treatment.

The result of the assessment of the risk of bias isimportant because it showed that the quality of 2 ofthe 3 long-term (2-phase) studies were classified as hav-ing a higher risk of bias.14-34 This may be a concern, butthese studies were planned and completed some yearsago, and the Cochrane tool reflects current researchpractices. Furthermore, these were long-term (10 years)studies; as a consequence, subject dropout was unavoid-able. These results should be accepted but interpreted inthe light of this classification.

The results from this systematic review do not sup-port the widespread provision of early orthodontic treat-ment for children with Class II malocclusion in terms ofdentoskeletal outcomes. However, they do provide datathat should be used to inform discussions on whetherparents wish their children to have early orthodontictreatment with the aim of reducing the chances of incisaltrauma, particularly in groups that may be vulnerable tothis problem.


CONCLUSIONS

1. There are no advantages in providing a 2-phasetreatment compared with 1 phase in early adoles-cence except for a potential reduction in the inci-dence of new incisal trauma.

2. Orthodontic treatment for young children, followedby a later phase of treatment when the child is inearly adolescence, appears to reduce the incidenceof new incisal trauma significantly compared withtreatment in 1 phase when the child is in earlyadolescence. However, these data should be inter-preted with caution because of the high degree ofuncertainty.

REFERENCES

1. Thiruvenkatachari B, Harrison JE, Worthington HV, O'Brien KD.Orthodontic treatment for prominent upper front teeth (Class IImalocclusion) in children. Cochrane Database Syst Rev 2013;(11):CD003452.

2. Office for National Statistics. National children's dental healthsurvey; 2003. Available at: http://www.ons.gov.uk/ons/guide-method/method-quality/specific/health-methodology/dental-health/dental-health-of-children/index.html. AccessedJuly 1, 2014.

3. Proffit WR, Fields HW Jr, Moray LJ. Prevalence of malocclusionand orthodontic treatment need in the United States: estimatesfrom the NHANES III survey. Int J Adult Orthod Orthognath Surg1998;13:97-106.

4. Baccetti T, Franchi L, Kim LH. Effect of timing on the outcomes of1-phase nonextraction therapy of Class II malocclusion. Am J Or-thod Dentofacial Orthop 2009;136:501-9.

5. Bailleau A, Aknin JJ, Gebeile-Chauty S. One phase or two-phase orthodontic treatment: comparisons. Orthod Fr 2012;83:289-96.

6. Bollen AM. Interceptive orthodontic treatment of Class II mal-occlusions is slightly more successful with headgear/biteplanethan with bionator treatment. J Evid Based Dent Pract 2011;11:58-9.

7. Franchi L, Pavoni C, Faltin K Jr, McNamara JA Jr, Cozza P. Re:response to: Long-term skeletal and dental effects and treatmenttiming for functional appliances in Class II malocclusion. Angle Or-thod 2013;83:933.

8. Ghislanzoni LT, Toll DE, Defraia E, Baccetti T, Franchi L. Treatmentand posttreatment outcomes induced by the mandibular advance-ment repositioning appliance; a controlled clinical study. AngleOrthod 2011;81:684-91.

9. Silvestrini-Biavati A, Alberti G, Silvestrini Biavati F, Signori A,Castaldo A, Migliorati M. Early functional treatment in Class II di-vision 1 subjects with mandibular retrognathia using Frankel IIappliance. A prospective controlled study. Eur J Paediatr Dent2012;13:301-6.

10. Wigal TG, Dischinger T, Martin C, Razmus T, Gunel E, Ngan P. Sta-bility of Class II treatment with an edgewise crowned Herbst appli-ance in the early mixed dentition: skeletal and dental changes. AmJ Orthod Dentofacial Orthop 2011;140:210-23.

11. Harrison JE, O'Brien KD, Worthington HV. Orthodontic treatmentfor prominent upper front teeth in children. Cochrane DatabaseSyst Rev 2007;CD003452.


http://refhub.elsevier.com/S0889-5406(15)00441-2/sref1http://refhub.elsevier.com/S0889-5406(15)00441-2/sref1http://refhub.elsevier.com/S0889-5406(15)00441-2/sref1http://refhub.elsevier.com/S0889-5406(15)00441-2/sref1http://www.ons.gov.uk/ons/guide-method/method-quality/specific/health-methodology/dental-health/dental-health-of-children/index.htmlhttp://www.ons.gov.uk/ons/guide-method/method-quality/specific/health-methodology/dental-health/dental-health-of-children/index.htmlhttp://www.ons.gov.uk/ons/guide-method/method-quality/specific/health-methodology/dental-health/dental-health-of-children/index.htmlhttp://refhub.elsevier.com/S0889-5406(15)00441-2/sref2http://refhub.elsevier.com/S0889-5406(15)00441-2/sref2http://refhub.elsevier.com/S0889-5406(15)00441-2/sref2http://refhub.elsevier.com/S0889-5406(15)00441-2/sref2http://refhub.elsevier.com/S0889-5406(15)00441-2/sref3http://refhub.elsevier.com/S0889-5406(15)00441-2/sref3http://refhub.elsevier.com/S0889-5406(15)00441-2/sref3http://refhub.elsevier.com/S0889-5406(15)00441-2/sref4http://refhub.elsevier.com/S0889-5406(15)00441-2/sref4http://refhub.elsevier.com/S0889-5406(15)00441-2/sref4http://refhub.elsevier.com/S0889-5406(15)00441-2/sref5http://refhub.elsevier.com/S0889-5406(15)00441-2/sref5http://refhub.elsevier.com/S0889-5406(15)00441-2/sref5http://refhub.elsevier.com/S0889-5406(15)00441-2/sref5http://refhub.elsevier.com/S0889-5406(15)00441-2/sref6http://refhub.elsevier.com/S0889-5406(15)00441-2/sref6http://refhub.elsevier.com/S0889-5406(15)00441-2/sref6http://refhub.elsevier.com/S0889-5406(15)00441-2/sref6http://refhub.elsevier.com/S0889-5406(15)00441-2/sref7http://refhub.elsevier.com/S0889-5406(15)00441-2/sref7http://refhub.elsevier.com/S0889-5406(15)00441-2/sref7http://refhub.elsevier.com/S0889-5406(15)00441-2/sref7http://refhub.elsevier.com/S0889-5406(15)00441-2/sref8http://refhub.elsevier.com/S0889-5406(15)00441-2/sref8http://refhub.elsevier.com/S0889-5406(15)00441-2/sref8http://refhub.elsevier.com/S0889-5406(15)00441-2/sref8http://refhub.elsevier.com/S0889-5406(15)00441-2/sref8http://refhub.elsevier.com/S0889-5406(15)00441-2/sref9http://refhub.elsevier.com/S0889-5406(15)00441-2/sref9http://refhub.elsevier.com/S0889-5406(15)00441-2/sref9http://refhub.elsevier.com/S0889-5406(15)00441-2/sref9http://refhub.elsevier.com/S0889-5406(15)00441-2/sref10http://refhub.elsevier.com/S0889-5406(15)00441-2/sref10http://refhub.elsevier.com/S0889-5406(15)00441-2/sref10


12. Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD,et al. The Cochrane Collaboration's tool for assessing risk of bias inrandomised trials. BMJ 2011;343:d5928.

13. Pandis N. The effect size. Am J Orthod Dentofacial Orthop 2012;142:739-40.

14. Chen DR, McGorray SP, Dolce C, Wheeler TT. Effect of early Class IItreatment on the incidence of incisor trauma. Am J Orthod Dento-facial Orthop 2011;140:e155-60.

15. Dolce C, McGorray SP, Brazeau L, King GJ, Wheeler TT. Timing ofClass II treatment: skeletal changes comparing 1-phase and 2-phase treatment. Am J Orthod Dentofacial Orthop 2007;132:481-9.

16. Johnson PD, Cohen DA, Aiosa L, McGorray S, Wheeler T. Attitudesand compliance of pre-adolescent children during early treatmentof Class II malocclusion. Clin Orthod Res 1998;1:20-8.

17. Keeling SD, Garvan CW, King GJ, Wheeler TT, McGorray S. Tempo-romandibular disorders after early Class II treatment with bionatorsand headgears: results from a randomized controlled trial. SeminOrthod 1995;1:149-64.

18. Keeling SD, Wheeler TT, King GJ, Garvan CW, Cohen DA,Cabassa S, et al. Anteroposterior skeletal and dental changes afterearly Class II treatment with bionators and headgear. Am J OrthodDentofacial Orthop 1998;113:40-50.

19. King GJ, McGorray SP, Wheeler TT, Dolce C, Taylor M. Comparisonof peer assessment ratings (PAR) from 1-phase and 2-phase treat-ment protocols for Class II malocclusions. Am J Orthod DentofacialOrthop 2003;123:489-96.

20. King GJ, Wheeler TT, McGorray SP. Randomised prospective clin-ical trial evaluating early treatment of Class II malocclusions [ab-stract]. Eur J Orthod 1999;21:445.

21. Wheeler TT, McGorray SP, Dolce C, Taylor MG, King GJ. Effective-ness of early treatment of Class II malocclusion. Am J Orthod Den-tofacial Orthop 2002;121:9-17.

22. Wortham JR, McGorray S, Taylor M, Dolce C, King DJ,Wheeler TT. Arch dimension changes following phase I andphase II orthodontic class II treatment [abstract 1131]. JDent Res 2001;80:177.

23. Almeida MA, Phillips C, Kula K, Tulloch C. Stability of the palatalrugae as landmarks for analysis of dental casts. Angle Orthod1995;65:43-8.

24. Brin I, Tulloch JF, Koroluk L, Philips C. External apical root resorp-tion in Class II malocclusion: a retrospective review of 1- versus 2-phase treatment. Am J Orthod Dentofacial Orthop 2003;124:151-6.

25. Dann C 4th, Phillips C, Broder HL, Tulloch JF. Self-concept,Class II malocclusion, and early treatment. Angle Orthod1995;65:411-6.


26. Ehmer U, Tulloch CJ, Proffit WR, Phillips C. An international com-parison of early treatment of angle Class-II/1 cases. Skeletal effectsof the first phase of a prospective clinical trial. J Orofac Orthop1999;60:392-408.

27. Koroluk LD, Tulloch JF, Phillips C. Incisor trauma and early treat-ment for Class II Division 1 malocclusion. Am J Orthod DentofacialOrthop 2003;123:117-26.

28. Proffit WR, Tulloch JF. Preadolescent Class II problems: treat nowor wait? Am J Orthod Dentofacial Orthop 2002;121:560-2.

29. Tulloch JF. Early versus late treatment for Class II malocclusions[abstract]. Eur J Orthod 1999;21:453.

30. Tulloch JF, Phillips C, Koch G, Proffit WR. The effect of earlyintervention on skeletal pattern in Class II malocclusion: a ran-domized clinical trial. Am J Orthod Dentofacial Orthop 1997;111:391-400.

31. Tulloch JF, Phillips C, Proffit WR. Benefit of early Class II treat-ment: progress report of a two-phase randomized clinical trial.Am J Orthod Dentofacial Orthop 1998;113:62-72.

32. Tulloch JF, Proffit WR, Phillips C. Influences on the outcome ofearly treatment for Class II malocclusion. Am J Orthod DentofacialOrthop 1997;111:533-42.

33. Tulloch JF, Proffit WR, Phillips C. Outcomes in a 2-phase random-ized clinical trial of early Class II treatment. Am J Orthod Dentofa-cial Orthop 2004;125:657-67.

34. Tulloch JF, Rogers L, Phillips C. Early results from a randomizedclinical trial of growth modification in Class II malocclusion [ab-stract]. J Dent Res 1992;72:523.

35. O'Brien K, Macfarlane T, Wright J, Conboy F, Appelbe P, Birnie D,et al. Early treatment for Class II malocclusion and perceived im-provements in facial profile. Am J Orthod Dentofacial Orthop2009;135:580-5.

36. O'Brien K, Wright J, Conboy F, Appelbe P, Davies L, Connolly I,et al. Early treatment for Class II Division 1 malocclusion withthe Twin-block appliance: a multi-center, randomized, controlledtrial. Am J Orthod Dentofacial Orthop 2009;135:573-9.

37. O'Brien K, Wright J, Conboy F, Chadwick S, Connolly I, Cook P,et al. Effectiveness of early orthodontic treatment with theTwin-block appliance: a multicenter, randomized, controlled trial.Part 2: psychosocial effects. Am J Orthod Dentofacial Orthop2003;124:488-95.

38. O'Brien K, Wright J, Conboy F, Sanjie Y, Mandall N, Chadwick S,et al. Effectiveness of early orthodontic treatment with theTwin-block appliance: a multicenter, randomized, controlled trial.Part 1: dental and skeletal effects. Am J Orthod Dentofacial Orthop2003;124:234-43.


http://refhub.elsevier.com/S0889-5406(15)00441-2/sref11http://refhub.elsevier.com/S0889-5406(15)00441-2/sref11http://refhub.elsevier.com/S0889-5406(15)00441-2/sref11http://refhub.elsevier.com/S0889-5406(15)00441-2/sref12http://refhub.elsevier.com/S0889-5406(15)00441-2/sref12http://refhub.elsevier.com/S0889-5406(15)00441-2/sref13http://refhub.elsevier.com/S0889-5406(15)00441-2/sref13http://refhub.elsevier.com/S0889-5406(15)00441-2/sref13http://refhub.elsevier.com/S0889-5406(15)00441-2/sref14http://refhub.elsevier.com/S0889-5406(15)00441-2/sref14http://refhub.elsevier.com/S0889-5406(15)00441-2/sref14http://refhub.elsevier.com/S0889-5406(15)00441-2/sref15http://refhub.elsevier.com/S0889-5406(15)00441-2/sref15http://refhub.elsevier.com/S0889-5406(15)00441-2/sref15http://refhub.elsevier.com/S0889-5406(15)00441-2/sref16http://refhub.elsevier.com/S0889-5406(15)00441-2/sref16http://refhub.elsevier.com/S0889-5406(15)00441-2/sref16http://refhub.elsevier.com/S0889-5406(15)00441-2/sref16http://refhub.elsevier.com/S0889-5406(15)00441-2/sref17http://refhub.elsevier.com/S0889-5406(15)00441-2/sref17http://refhub.elsevier.com/S0889-5406(15)00441-2/sref17http://refhub.elsevier.com/S0889-5406(15)00441-2/sref17http://refhub.elsevier.com/S0889-5406(15)00441-2/sref18http://refhub.elsevier.com/S0889-5406(15)00441-2/sref18http://refhub.elsevier.com/S0889-5406(15)00441-2/sref18http://refhub.elsevier.com/S0889-5406(15)00441-2/sref18http://refhub.elsevier.com/S0889-5406(15)00441-2/sref19http://refhub.elsevier.com/S0889-5406(15)00441-2/sref19http://refhub.elsevier.com/S0889-5406(15)00441-2/sref19http://refhub.elsevier.com/S0889-5406(15)00441-2/sref20http://refhub.elsevier.com/S0889-5406(15)00441-2/sref20http://refhub.elsevier.com/S0889-5406(15)00441-2/sref20http://refhub.elsevier.com/S0889-5406(15)00441-2/sref21http://refhub.elsevier.com/S0889-5406(15)00441-2/sref21http://refhub.elsevier.com/S0889-5406(15)00441-2/sref21http://refhub.elsevier.com/S0889-5406(15)00441-2/sref21http://refhub.elsevier.com/S0889-5406(15)00441-2/sref22http://refhub.elsevier.com/S0889-5406(15)00441-2/sref22http://refhub.elsevier.com/S0889-5406(15)00441-2/sref22http://refhub.elsevier.com/S0889-5406(15)00441-2/sref23http://refhub.elsevier.com/S0889-5406(15)00441-2/sref23http://refhub.elsevier.com/S0889-5406(15)00441-2/sref23http://refhub.elsevier.com/S0889-5406(15)00441-2/sref24http://refhub.elsevier.com/S0889-5406(15)00441-2/sref24http://refhub.elsevier.com/S0889-5406(15)00441-2/sref24http://refhub.elsevier.com/S0889-5406(15)00441-2/sref25http://refhub.elsevier.com/S0889-5406(15)00441-2/sref25http://refhub.elsevier.com/S0889-5406(15)00441-2/sref25http://refhub.elsevier.com/S0889-5406(15)00441-2/sref25http://refhub.elsevier.com/S0889-5406(15)00441-2/sref26http://refhub.elsevier.com/S0889-5406(15)00441-2/sref26http://refhub.elsevier.com/S0889-5406(15)00441-2/sref26http://refhub.elsevier.com/S0889-5406(15)00441-2/sref27http://refhub.elsevier.com/S0889-5406(15)00441-2/sref27http://refhub.elsevier.com/S0889-5406(15)00441-2/sref28http://refhub.elsevier.com/S0889-5406(15)00441-2/sref28http://refhub.elsevier.com/S0889-5406(15)00441-2/sref29http://refhub.elsevier.com/S0889-5406(15)00441-2/sref29http://refhub.elsevier.com/S0889-5406(15)00441-2/sref29http://refhub.elsevier.com/S0889-5406(15)00441-2/sref29http://refhub.elsevier.com/S0889-5406(15)00441-2/sref30http://refhub.elsevier.com/S0889-5406(15)00441-2/sref30http://refhub.elsevier.com/S0889-5406(15)00441-2/sref30http://refhub.elsevier.com/S0889-5406(15)00441-2/sref31http://refhub.elsevier.com/S0889-5406(15)00441-2/sref31http://refhub.elsevier.com/S0889-5406(15)00441-2/sref31http://refhub.elsevier.com/S0889-5406(15)00441-2/sref32http://refhub.elsevier.com/S0889-5406(15)00441-2/sref32http://refhub.elsevier.com/S0889-5406(15)00441-2/sref32http://refhub.elsevier.com/S0889-5406(15)00441-2/sref33http://refhub.elsevier.com/S0889-5406(15)00441-2/sref33http://refhub.elsevier.com/S0889-5406(15)00441-2/sref33http://refhub.elsevier.com/S0889-5406(15)00441-2/sref34http://refhub.elsevier.com/S0889-5406(15)00441-2/sref34http://refhub.elsevier.com/S0889-5406(15)00441-2/sref34http://refhub.elsevier.com/S0889-5406(15)00441-2/sref34http://refhub.elsevier.com/S0889-5406(15)00441-2/sref35http://refhub.elsevier.com/S0889-5406(15)00441-2/sref35http://refhub.elsevier.com/S0889-5406(15)00441-2/sref35http://refhub.elsevier.com/S0889-5406(15)00441-2/sref35http://refhub.elsevier.com/S0889-5406(15)00441-2/sref36http://refhub.elsevier.com/S0889-5406(15)00441-2/sref36http://refhub.elsevier.com/S0889-5406(15)00441-2/sref36http://refhub.elsevier.com/S0889-5406(15)00441-2/sref36http://refhub.elsevier.com/S0889-5406(15)00441-2/sref36http://refhub.elsevier.com/S0889-5406(15)00441-2/sref37http://refhub.elsevier.com/S0889-5406(15)00441-2/sref37http://refhub.elsevier.com/S0889-5406(15)00441-2/sref37http://refhub.elsevier.com/S0889-5406(15)00441-2/sref37http://refhub.elsevier.com/S0889-5406(15)00441-2/sref37


APPENDIX

SEARCH STRATEGY FOR MEDLINE (OVID)

1. exp Orthodontics/2. (appliance$ adj5 (function$ or remova$ or

fix$)).mp.3. (orthodontic$ and (brace$ or band$ or

wire$)).mp.4. (orthodontic$ and (extract$ or remov$)).mp.5. (orthodontic$ and (headgear$ or “head gear$” or

head-gear$)).mp.6. (device$ adj5 (function$ or remova$ or fix$)).mp.7. ((appliance$ or device$) adj5 (intraoral or “intra

oral” or intra-oral or extraoral or “extra oral” or ex-tra-oral)).mp.

8. (activator adj appliance$).mp.9. (Frankel or “twin$ block$” or FR-II).mp.


10. ((growth adj3 modif$) and (jaw$ or maxilla$ ormandible$ or mandibular)).mp.

11. (two-phase and (treatment or therapy) and (ortho-dontic$ or malocclusion$)).mp.

12. ((orthopedic$ or orthopaedic$) and (dental or or-thodontic$ or facial)).mp.

13. or/1-1214. Malocclusion, Angle Class II/15. Retrognathism/16. ((“class II” or “class 2”) adj3 malocclusion$).mp.17. (posterior adj3 occlusion$).mp.18. (distoclusion$ or disto-occlusion$ or distocclu-

sion$).mp.19. retrognath$.mp.20. (prominent adj3 upper adj3 teeth).mp.21. (overjet$ or “over jet$” or over-jet$).mp.22. or/14-2123. 13 and 22


Early orthodontic treatment for Class II malocclusion reduces the chance of incisal trauma: Results of a Cochrane systemati ...Material and methodsStatistical analysis

ResultsDiscussionConclusionsReferencesAppendix

early orthodontic treatment for class ii malocclusion ... · a result, correcting class ii...

Documents