Class II Division 1 malocclusion treatment with extraction of maxillary first molars: Evaluation of treatment and post‐treatment changes by the PAR Index

Abstract Objective To investigate occlusal result and post‐treatment changes after orthodontic extraction of maxillary first permanent molars in patients with a Class II division 1 malocclusion. Setting and Sample Retrospective longitudinal study in a private practice, with outcome evaluation by an independent academic hospital. Ninety‐six patients (53 males, 43 females) consecutively treated by one orthodontist with maxillary first permanent molar extraction were studied, divided into three facial types, based on pre‐treatment cephalometric values: hypodivergent (n = 18), normodivergent (n = 21) and hyperdivergent (n = 57). Methods Occlusal outcome was scored on dental casts at T1 (pre‐treatment), T2 (post‐treatment) and T3 (mean follow‐up 2.5 ± 0.9 years) using the weighted Peer Assessment Rating (PAR) Index. The paired sample t test and one‐way ANOVA followed by Tukey's post hoc test were used for statistical analysis. Results PAR was reduced by 95.7% and 89.9% at T2 and T3, respectively, compared with the start of treatment. The largest post‐treatment changes were found for overjet and buccal occlusion. Linear regression analysis did not reveal a clear effect (R‐Square 0.074) of age, sex, PAR score at T1, incremental PAR score T2‐T1, overjet and overbite at T1, and facial type on the changes after treatment (incremental PAR score T3‐T2). Conclusions The occlusal outcome achieved after Class II division 1 treatment with maxillary first permanent molar extractions was maintained to a large extent over a mean post‐treatment follow‐up of 2.5 years. Limited changes after treatment were found, for which no risk factors could be discerned.


| INTRODUC TI ON
A great variety of treatment options exists for the treatment of Class II malocclusions, including facial orthopaedic, functional, non-extraction and extraction procedures. Treatment options depend on the type and severity of the malocclusion, age and facial growth status of the patient, educational background of the orthodontist and treatment preferences of the patient. 1 In young patients, growth modification with functional appliances or extra-oral traction is often the treatment of choice, although nowadays the concept of long-term growth modification of the mandible and maxilla is questioned. 2 6 Several studies have reported the short-term results of this procedure. [7][8][9] Using the PAR index to measure occlusal outcome in 100 consecutive patients at the end of active treatment, 73% were in the 'greatly improved' and 27% in the 'improved' group. 7 There were no patients in the 'worse or no different' group. The cephalometric analysis revealed that this type of treatment had only a minimal bite-closing effect, while no significant differences for change in mandibular plane angle were found between different facial types.
The patients showed a flattening of the profile and an increase in the nasolabial angle, comparable to the soft-tissue outcomes of other extraction modalities, as reported in systematic reviews. 10,11 To our knowledge, no previous studies have reported on stability of Class II division 1 malocclusion treatment with maxillary first permanent molar extractions. As post-treatment changes occur mostly in the first 2 years, 12-14 the aim of this study was to evaluate occlusal results of Class II division 1 treatment with extraction of maxillary first permanent molars after a mean follow-up period of 2.5 years, in a large group of consecutively treated patients.

| Subjects
This was a retrospective, longitudinal, one-group outcome study in a private practice, with outcome evaluation by an independent academic hospital. The research was conducted in accordance with the Helsinki Declaration with regard to research on human subjects. All parents and patients agreed to have their patient records used in the study and gave signed informed consent. Ethical approval was not needed, as this was an observational study using anonymized, routinely collected health data. Dental casts of all patients were made at T1 (pre-treatment), T2 (post-treatment) and T3 (follow-up). The standard recall schedule was 2 years after treatment, and the minimum follow-up was set at 24 months with a deviation of −20%. To study the effect of treatment for different facial types, the sample was divided into three groups,

| Treatment method
Treatment with fixed appliances started 2 weeks after the extraction of the maxillary first permanent molars. In case of a deep bite, the extractions were delayed, with initial placement of an upper bite plate and a fixed appliance in the lower arch. Second maxillary molars were fully erupted before the extractions were carried out. All These buccal retention wires were removed after complete eruption of the maxillary third molars.

| Occlusal outcome
For assessment of the results, the dental casts were randomly placed on a table and identified by only a non-traceable number. The scoring was performed by one observer (CL) calibrated in the use of the Peer Assessment Rating (PAR) Index, who was not involved in the treatment.
Occlusal outcome was scored on the dental casts at T1 (pre-treatment), T2 (post-treatment) and T3 (2 or more years post-treatment) using the PAR Index. 16 The PAR Index consists of the sum of seven subcomponent scores: upper anterior segment, lower anterior segment, left and right buccal occlusion, overjet, overbite and centre line.
Weighted PAR scores (British weightings) were used, which means that the individual scores for overjet were multiplied by 6, overbite by 2 and centre line by 4. A weighted PAR score of 0 means good alignment and higher scores indicate the level of irregularity. The degree of success of the orthodontic treatment is reflected by the percentage reduction in the total weighted PAR score. The PAR subcomponent 'anterior cross bite' was excluded because this sample consisted of Class II division 1 patients and only one patient scored on this item. It concerned an end-to-end position of two lateral incisors (patient number 65). Nomograms were used to visualize the degree of improvement following treatment and to visualize the degree of final improvement between the starting condition and 2 years post-treatment. In these nomograms, 17 the degree of change of the weighted PAR score is divided into three categories: worse or no different (cases with less than 30% reduction), improved (patients with ≥30% reduction) and greatly improved (generally a reduction of 22 weighted PAR points or more).
The weighted PAR scores were used to evaluate treatment outcome, treatment efficiency, operator experience and the change after treatment. Treatment efficiency was defined as the treatment efficiency index (TEI) according to Janson et al 18 as the PAR reduction between T1 and T2 divided by treatment duration (in months).
Furthermore, the weighted PAR scores of the three vertical facial types were compared.
To determine the error of the method, the same observer re-assessed 21 series of models (at T1, T2 and T3) 2 weeks after the first assessment.
The eruption status of the third molars at T3 was evaluated on the dental casts and the radiographs (orthopantomogram and/or lateral cephalogram).

| Statistics
The statistical analysis was performed using SPSS version 22 for Windows (IBM, North Castle, USA). For the overall PAR score, the reliability coefficients between the two measurements were calculated as Pearson's correlation coefficients. Paired sample t tests were applied to identify systematic differences between the first and second measurement. The duplicate measurement error (DME) was calculated as the SD of the difference between two observations divided by √2. The intra-observer reliability for the PAR subcomponents was calculated using weighted kappa statistics. A kappa less than 0 reflects 'poor', 0 to 0.20 'slight', 0.21 to 0.4 'fair', 0.41 to 0.60 'moderate', 0.61 to 0.8 'substantial' and above 0.81 'almost perfect' agreement. 19 Outcomes are presented as a variable with a mean and ±SD. The paired sample t test was applied to analyse the changes in the PAR score between T1 and T2, T2 and T3, and T1 and T3.
One-way ANOVA followed by Tukey's post hoc test was applied to test for differences in the TEI between the three facial types.
Linear regression analysis was applied to analyse the effects of the independent variables age, sex, PAR score at T1, incremental PAR score T2-T1, overjet and overbite at T1, and facial type at T1 on the incremental PAR score T3-T2 (the dependent variable).
A P value of <.05 was considered to indicate statistical significance.

| Subjects
All 99 patients finished their treatment, and no treatments were dis-

| Error of the method
For the overall PAR score, the Pearson's correlation coefficient was 0.998. The duplicate measurement error (DME) was 0.638 PAR points. The mean difference between the first and second measurement was 0.159 PAR points (95% CI, −0.068…0.386) which was not statistically significant (P = .167). The kappa values for the weighted subcomponents ranged from 0.833 to 1.000, reflecting almost perfect agreement.

| Outcome
The results for the PAR index at T1, T2 and T3 are presented in Table 1 and

| Operator experience
To determine the operator experience, the first 20 treated patients were compared with the last 20 treated patients with regard to the weighted PAR score at T2. For the first 20 patients, the weighted PAR score at T1 was 28.4 (SD 7.1), and for the last 20 patients, it was 26.6 (SD 7.1). The changes in the PAR scores for the first and last 20 patients between T1 and T2 were −26.95 (SD 6.70) and −25.95 (SD 7.62), respectively, and not significantly different from each other (independent samples t test P = .662). Operator experience had also no effect on the change in the PAR score between T2 and T3, which amounted to 1.6 (SD 3.82) PAR points for the first 20 patients and 1.2 (SD 1.88) for the last 20 patients (P = .677). In both groups, 8 of 20 patients (40%) showed a change in the PAR score after treatment.
The Treatment Efficiency Index (TEI) for the total group was 3.35 (SD 0.85). We also compared TEI for the first and last 20 treated patients between T1 and T2. The first group had a TEI of 2.88 (SD 0.65), and the more recently treated group had a TEI of 3.87 (SD 0.70). This difference was highly significant (P < .001), indicating a greater PAR reduction per treatment month in the more recently treated group. TA B L E 1 Mean weighted PAR scores. Means and SD for the total PAR score and the subcomponents before treatment (T1), after treatment (T2) and after a mean follow-up of 2.5 y after treatment (T3)

| Maxillary third molars
In 52 patients (54.2%), a buccal retention wire was placed at the first and second lower molar in cases of non-occlusion of the mandibular second molars at T2 (n = 4 on one side, n = 48 at two sides). At T3 in 11 patients (11.5%), these wires were still present (n = 4 on one side, n = 7 at two sides). In 80 patients (83.3%), both maxillary third molars were present at T3. In 8 patients, one of the molars was erupted at that point of time. The 24 as yet unerupted maxillary third molars were checked on the X-rays, and 23 of them had a good prognosis for eruption. The prognosis of 1 molar was doubtful.

| D ISCUSS I ON
We   35 To optimize the quality of the dentition, extraction of poorly conditioned maxillary first permanent molars is a good option for Class II malocclusion treatment, if the second and third molars are of good quality. The third molars have also been shown to have a better prognosis for normal eruption when the first molars are extracted 8 which was also confirmed in the present study as only one third molar showed a bad prognosis and 83.3% of the third molars had already erupted at the follow-up observation.

| Limitations
This study was a longitudinal cohort study for which the data were collected in a private practice. All treatments were carried out by the same orthodontist. As compared to a multicentre, multi-operator trial, the single-centre, one-operator study design is less favourable for the generalizability of the results. As compared to the randomized controlled trial, the retrospective study design has wellknown drawbacks. Not all clinical orthodontic research questions can be studied, however, in randomized controlled trials. For example, ethical concerns-extraction version non-extraction therapymay limit the application of the most rigorous design.
As this was a one-group longitudinal study, we are not able to determine the contribution of physiological changes to the treatment and post-treatment changes we found. An earlier study, however, showed that the PAR score in non-treated individuals between 12 and 22 years of age remained the same, irrespective of the Angle classification, although clinically relevant changes were found in individual cases. 36 We are aware of the limitations of the PAR index. Dental variables, like proclined lower incisors and retroclined upper incisors, are not represented in the rating. Furthermore, the PAR index uses a weighting system for several subcomponents of the index.
Overjet, for example, has a weighting of 6 in the British weighting system, which adds considerably to a high pre-treatment PAR score in a sample of Class II division 1 cases with a large overjet, as in our study. For that reason, it is easier to realize remarkable changes in the PAR score when the initial PAR score is high. 37 A recent study in a Chinese population suggested that different Angle classifications may need different weightings. 38 This supports the discussion, for more than two decades, of the British weightings overemphasizing overjet and insufficiently weighting overbite. 39 Some have suggested extending the PAR index with a score for sufficient torque, good axial control of lower incisors and the irregularity index. 27,40 Overall quality of the treatment manifested as the presence of root resorption, gingival recessions, white spots, dysfunctions, facial aesthetics, and patient satisfaction and quality of life were not measured in this study.

| CON CLUS ION
The occlusal outcome achieved after Class II division 1 treatment with maxillary first permanent molar extractions was maintained to