Spyridon N. Papageorgiou and Theodore Eliades
In the last decade several systematic reviews of clinical studies comparing orthodontic aligners with fixed appliances have emerged. However, they all present methodological issues that can introduce bias and hamper their ability to draw robust evidence-based recommendations, including among others: assessment of oral hygiene but not efficacy, lack of an a priori design/pre-registered protocol, language bias, inclusion of nonrandomized studies with uncontrolled confounding, inadequate handing of the studies’ risk of bias, lack of quantitative data synthesis (meta-analysis), improper data synthesis methods, and being outdated. This chapter presents a critical appraisal on the clinical performance of orthodontic aligners based on currently available studies founded on the principles of evidence-based medicine. According to currently existing clinical evidence from randomized trials and matched nonrandomized studies on mostly adult patients with mild to severe malocclusions treated with or without extractions it seems that orthodontic treatment with aligners is associated with inferior treatment outcome compared to fixed appliances. Treatment duration is not directly influenced by choice of appliance alone and patient-related or treatment-related factors might come into play.
The use of sequential clear aligners has seen a remarkable surge in the last decades and, following considerable technical developments, has been widely adopted by both orthodontic specialists and general dentists alike. Fueled by aggressive marketing campaigns from various manufacturers of aligner systems, a growing interest has been reported, especially among adult patients, for such methods of invisible orthodontics.1 , 2 A survey of Australian orthodontists in 2013 indicated that 73% of responders had used aligners to treat at least one case in the last year, with a median of 8 aligner cases/year.3 A similar survey among Irish orthodontists in 2014 reported that 19% of them often used aligners to treat adult patients.4 A large 2014 survey among orthodontist in the United States5 revealed that 89% of them had treated at least one case with aligners (compared to 76% in 2008) and treated a median of 22 cases/year with aligners (compared to 12 cases/year in 2008). Responding orthodontists who used aligners employed them in a variety of cases, with the most common diagnostic category for aligner treatment being: Class I with moderate crowding (94%), space closure (78%), Class II (68%), lower incisor extraction (47%), Class I with severe crowding (37%), and Class III (49%), while only few orthodontists used aligners for premolar extraction cases (9–18%). Interestingly, responders in 2014 considered 90% of their aligner cases successful (compared to 80% in 2008), but also saw about 10% of aligner cases with relapse (same as in 2008). Additionally, another survey among members of the European Aligner Society indicated that 45% of orthodontists believed that aligners limit orthodontic treatment outcomes (even though the respective percentage among general dentists was only 5%).6 These data might indicate that the initial surge of aligner treatment during its early years of fame might have now given its place to a more mature evaluation of this treatment modality, based on long-term evaluations of previously treated patients.
In any case, it is imperative that any treatment modality offered to orthodontic patients as an alternative is based on both the doctor’s clinical expertise and solid evidence on the clinical performance on this modality. Unfortunately, contrary to many medical fields, it is commonly seen in orthodontics that novel marketed products and treatment approaches are clinically adopted based on advertisement without the appropriate clinical evidence to back any claims made by the manufacturers.7 , 8 Good clinical practice, however, obligates that any treatment decision between the treating orthodontist and the patient is done after meticulous discussion of all available treatment options and evidence-based notions about their efficacy and adverse effects. Ideally, these should be based on well-designed and well-reported comparative clinical trials on human patients and systematic reviews/meta-analyses thereof.9 , 10 Ample empirical evidence has now been gathered about the importance of proper study design and the role that various methodological characteristics can play in introducing bias.11 – 17
In the last decade, several systematic reviews of clinical studies comparing orthodontic aligners with fixed appliances have emerged.18 – 27 However, they all present methodological issues that can introduce bias and hamper their ability to draw robust evidence-based recommendations, including among others: assessment of oral hygiene but not efficacy,18 , 23 , 24 lack of an a priori design/preregistered protocol,19 – 22 , 26 , 27 language bias,20 , 22 , 25 inclusion of nonrandomized studies with uncontrolled confounding,19 , 20 , 22 , 25 – 27 inadequate handing of the studies’ risk of bias,19 – 22 , 25 – 27 lack of quantitative data synthesis (meta-analysis),19 , 20 , 22 , 25 , 27 outdated data synthesis methods,21 , 26 and outdated literature searches.19 – 21 Therefore, clinical practice ought to be informed by a critical appraisal of currently available studies according to the principles of evidence-based medicine.
To this end, a systematic review was designed a priori based on the Cochrane guidelines,28 registered in PROSPERO (CRD42019131589), and is reported according to the PRISMA statement.29 Eight databases (MEDLINE through PubMed, Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, Cochrane Database of Abstracts of Reviews of Effects, Scopus, Virtual Health Library, and Web of Knowledge) were searched up to April 25, 2019, without any restrictions for publication date, language, or type with the following search strategy: (orthodon* OR malocclusion* OR “tooth movement” OR “fixed appliances”) AND (aligner* OR “clear aligner” OR “clear aligners” OR “ClearCorrect” OR “Invisalign” OR “Orthocaps” OR “TwinAligner”).
Eligible for inclusion were randomized trials comparing adolescent/adult patients with any kind of malocclusion receiving full-arch comprehensive treatment with either orthodontic aligners or any kind of fixed appliances. Due to the scarcity of randomized trials on the subject, nonrandomized studies were also included, with the requirement that the populations to be compared were matched regarding baseline malocclusion severity with objective measures such as the Peer Assessment Rating (PAR) index30 or the Discrepancy Index (DI)31 from the American Board of Orthodontics (ABO). The PAR index uses seven criteria: tooth alignment (referring to dental crowding), right and left buccal segment relationship (sagittal, vertical, and transverse assessments), overjet, overbite, and centerline (midline discrepancies). Each difference from the norm has points attributed to the severity of the discrepancy. Once tabulated and weighted according to the United States or United Kingdom weightings, an overall score for the malocclusion is calculated. The ABO DI scores 12 target disorders: overjet, overbite, anterior open bite, lateral open bite, crowding, occlusal relationship, lingual posterior cross-bite, buccal posterior cross-bite, ANB angle, mandibular plane inclination, lower incisor inclination, and a category “other” that includes complexes such as Bolton discrepancy, shortened roots, deep curve of Spee, traumatic injuries, bimaxillary protrusion cases with critical anchorage need, and craniofacial dysmorphologies. Similarly to the PAR index, scores are assigned and tabulated to reflect the malocclusion severity. Matching was judged adequate when the Cohen’s d for PAR or ABO DI between aligner and fixed appliance group at baseline was up to 0.3. The primary outcome for this review was the outcome of comprehensive orthodontic treatment judged with objective and reliable measures such as the PAR index and the ABO’s Objective Grading System (ABO-OGS) for dental casts and panoramic radiographs.32 The ABO-OGS rates the final occlusion after appliance removal with eight criteria that contribute to ideal intercuspation and function: alignment, marginal ridges, buccolingual inclination, overjet, occlusal contacts, occlusal relationships, interproximal contacts, and root angulation. Best occlusion and alignment receive a score of 0 points, while for each parameter that deviates from the ideal, 1 or 2 penalty points are added. The greater the posttreatment ABO-OGS score, the more the final treatment result deviates from ideal occlusion, while a case can also be classified as “successful” or “failed” according to their ABO criteria for score lower or higher than 30 points. Secondary outcomes included treatment duration, as well as adverse effects such as loss of periodontal support, external apical root resorption (EARR), gingival recession, and uncontrolled proclination of the lower incisors during treatment.
Study selection, data extraction, and risk of bias assessment were performed by three independent assessors. The risk of bias of included studies was assessed according to Cochrane guidelines with the RoB 2.0 tool for randomized trials33 and the ROBINS-I (“Risk Of Bias In Nonrandomised Studies-of Interventions”) tool for nonrandomized studies.34 Mean differences (MDs) for continuous outcomes and relative risks (RRs) for binary outcomes and their corresponding 95% confidence intervals (CIs) were pooled with random-effects meta-analysis (using a restricted maximum likelihood variance estimator35), with p < 0.05 considered significant, and presented in contour-enhanced forest plots.36 Relative/absolute heterogeneity was assessed with I 2 and tau,2 respectively, and incorporated into random-effects 95% predictions to quantify expected treatment effects in a future clinical setting.37 The overall quality of clinical recommendations (confidence in effects estimates) for each of the main outcomes was rated using the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) approach38 using an improved Summary of Findings table format39 and guidance on how to combine randomized and nonrandomized studies.40
The electronic literature search up to April 2019 yielded 1,376 hits, while 7 additional studies were manually identified through checking of the reference or citation lists of identified reports (Fig. 9.1). After applying the review’s eligibility criteria, 11 publications pertaining to 11 unique studies (4 randomized and 7 retrospective nonrandomized)41 – 51 published as journal papers or dissertation/theses were included (Table 9.1). The included studies were conducted in university clinics (n = 6; 55%), private practices (n = 4; 36%), or hospitals (n = 1; 9%) and originated from six different countries (Canada, China, Ireland, Italy, South Korea, and the United States). A total of 446 and 443 patients were treated with aligners and fixed appliances, respectively, with a median total sample of 66 patients per included study (range 19–200 patients per study). Out of the 7 studies reporting on patient sex, 215 of the 661 patients in total were male (33%), while the mean patient age out of the 9 studies reporting this was 28.0 years.
As far as complexity of the treated cases is concerned, only six studies (55%) reported this with either the PAR index (n = 3; 27%) or the ABO DI (n = 3; 27%). Eight of the studies (73%) performed nonextraction treatment, one study (9%) both extraction and nonextraction treatment, and one study (9%) extraction treatment. The majority of studies (9/11 studies; 82%) reported on conventional comprehensive treatment, while 1 study (9%) reported on orthodontic treatment of patients with history of periodontal disease and 1 study (9%) reported on combined orthodontic/orthognathic treatment. Details of the aligner treatment were only partly reported among the included studies, with only two studies (18%) reporting the number of aligners, four studies (36%) reporting on “refinement” rate (i.e., the midcourse re-evaluation and planning of additional aligners), and two studies (18%) on the actual amount of interproximal enamel reduction performed during treatment in both groups.
The included randomized trials presented several issues that increased their risk for bias (Table 9.2). Two trials were in high risk of bias due to problems in the randomization process, deviations from intended interventions, missing outcome data, and outcome measurement. The remaining two trials were in low risk of bias, except for the fact that no a priori trial protocol could be found to rule out selective reporting. The included nonrandomized studies were in considerably higher risk of bias, with five of them presenting moderate risk of bias, one of them serious risk of bias, and one of them critical risk of bias (Table 9.3). Their main shortcomings pertained to confounding, selection of participants into the study, deviations from intended interventions, outcome measurement, and selection of the reported result.
The included studies reported on a wide spectrum of treatment outcomes, with only three studies41 , 45 , 47 reporting on the complete ABO-OGS score including all eight components, as well as failure of the case to pass the ABO criteria for adequate occlusal results (ABO-OGS score < 30 points). One study reported on the ABO-OGS score of seven out of eight components (excluding root angulation)42 and also excluded scoring the second molars without any justification. One study also reported solely on two of the eight ABO-OGS components49—namely, marginal ridges and buccolingual inclination. Three studies used the PAR index48 , 50 , 51 and reported either posttreatment PAR scores or PAR reductions. Eight studies reported on treatment duration,41 , 46 – 51 though considerable variation in the reported results was seen. Finally, single studies reported on periodontal probing depth, alveolar bone loss, EARR, lower incisor inclination, and gingival recessions.
The ABO-OGS enables the objective and precise evaluation of the outcome of comprehensive orthodontic treatment including the fine details of a balanced ideal occlusion. The meta-analysis combining the results of the three existing studies measuring the total ABO-OGS score at debond is presented in Fig. 9.2. On average, meta-analysis of three studies indicated that orthodontic treatment with aligners was associated with a statistically significant reduction in the finishing quality according to ABO-OGS compared to fixed appliances (MD: 9.9 points; 95% CI: 3.6–16.2 points; p = 0.002). Considerable heterogeneity was seen among the three included studies (I 2 = 84%), which meant that several patient- or treatment-related factors might play a role in the actual final occlusal result. However, existing heterogeneity influenced only the precise calculation of the difference between aligners and fixed appliances, as one study indicated a moderate difference and the other two indicated a large one. It did not, however, influence the direction of the effect, as all three studies showed that fixed appliances were significantly associated with better treatment results than aligners.
Fig. 9.2 Contour-enhanced forest plot on the comparison of total ABO-OGS scores posttreatment between aligners and fixed appliances. ABO-OGS, American Board of Orthodontics Objective Grading System; AL, aligner; CI, confidence interval; FX, fixed appliance; M, mean; MD, mean difference; N, number of patients; SD, standard deviation. Contours correspond to different effect magnitude and the red dotted line corresponds to 95% random-effects prediction.
The same conclusion was drawn when looking at the proportion of cases being finished to an acceptable quality according to the ABO standards—i.e., the proportion of patients having less than 30 ABO-OGS points at debond (Fig. 9.3). Meta-analysis of three studies indicated that treatment with aligners was associated with significantly increased probability for the case to be considered a “failure” according to the ABO standards (i.e., that the case has an ABO-OGS score more than 30) compared to fixed appliances (RR: 1.6; 95% CI: 1.2–2.0; p < 0.001). No considerable heterogeneity across studies was seen, which reported a small to moderate increase in the rate of suboptimal finishing quality. On absolute terms, this is translated to a 60.6% “failing” grade according to ABO for the aligner group compared to a 38.9% rate for the fixed appliance group (Fig. 9.4). This is in turn translated to a number needed to treat of 5, which means that every fifth case treated with aligners instead of fixed appliances would fail the ABO examination, but would get a “passing” grade if it was treated with fixed appliances, which is a potentially clinically relevant effect.
Fig. 9.3 Contour-enhanced forest plot on the comparison of proportion of “passing” cases according to the ABO examination (cases with ABO-OGS score lower than 30 points) posttreatment between aligners and fixed appliances. ABO-OGS, American Board of Orthodontics Objective Grading System; AL, aligner; CI, confidence interval; FX, fixed appliance; N, number of patients; RR, relative risk. Contours correspond to different effect magnitude and the red dotted line corresponds to 95% random-effects prediction.
Looking at the comparative performance for each separate component of ABO-OGS between aligners and fixed appliances gives a more precise image about the occlusal aspects mostly affected by the treatment modality (Fig. 9.5). Overall, meta-analyses of three studies indicated that five of the eight aspects of the occlusion were significantly better finished with fixed appliances than with aligners: buccolingual inclination (MD: 0.8 point; 95% CI: 0.5–1.1 point; p < 0.001), occlusal contacts (MD: 3.1 points; 95% CI: 0.6–5.6 points; p = 0.02), occlusal relationship (MD: 1.0 point; 95% CI: 0.6–1.4 points; p < 0.001), overjet (MD: 1.8 points; 95% CI: 0.6–3.0 points; p = 0.002), and root angulation (MD: 0.8 point; 95% CI: 0.5–.1 point; p < 0.001). It has been reported that it is considerably more difficult to control root movement with aligners compared to fixed appliances, especially without the use of attachments.2 , 51 , 52 The third generation of aligners presumably allows for improved movement of this type by adding ellipsoid precision attachments that are able to produce couples creating root movement,2 which remains to be tested experimentally. On the other side, three ABO-OGS components (alignment, marginal ridges, and interproximal contacts) gave very similar results for both modalities. This is not surprising, since aligners are known to consistently produce adequate space closure of up to 6 mm by progressively tipping teeth into spaces in small increments and can successfully straighten dental arches by derotating teeth, especially when composite attachments are bonded.52 – 54 Looking carefully at the effect magnitude, it is obvious that the clinical relevance for each separate criterion is questionable, as small to moderate differences between aligners and fixed appliances are seen on average. However, when adding all these differences for each criterion, a clinically relevant worse finishing outcome is seen with aligners overall (as seen in Fig. 9.2 and Fig. 9.3).
Fig. 9.5 Composite contour-enhanced forest plot illustrating the summary results of eight meta-analyses (each with three studies and 297 patients) for the comparison of each separate ABO-OGS component between orthodontic aligners and fixed appliances. ABO-OGS, American Board of Orthodontics Objective Grading System; CI, confidence interval; MD, mean difference. Contours correspond to different effect magnitude and the red dotted lines correspond to 95% random-effects predictions.
Looking at the occlusal outcome of treatment through meta-analyses of two studies using the PAR index gives a slightly different picture (Table 9‑4). Overall, no statistically significant difference in the posttreatment PAR scores between aligners and fixed appliances was seen (MD: 0 points; 95% CI: –2.0 to 2.0 points; p = 0.98). Contrary to that, treatment with orthodontic aligners was associated with a significantly smaller reduction in PAR scores (significantly worse treatment efficacy) compared to fixed appliances (MD: –2.9 points; 95% CI: –5.0 to –0.8 points; p = 0.007). However, even though the effect was statistically significant, the magnitude of this difference was small, which makes its clinical relevance questionable. Results of a single study48 indicated that aligners were worse in terms of reduction for the PAR component for upper anteriors (MD: –1.0 point; 95% CI: –1.9 to –0.1 point; p = 0.02) and overbite (MD: –1.0 point; 95% CI: –1.9 to –0.2 points; p = 0.02) compared to fixed appliances (Table 9.5). Again, differences between aligners and fixed appliances for these outcomes might be statistically significant, but probably are not clinically relevant. On the other side, the proportion of patients experiencing a great improvement in their PAR scores through treatment (PAR reduction of at least 22 points or PAR score of 0 posttreatment) was significantly smaller with aligners than with fixed appliances (RR: 0.5; 95% CI: 0.3–0.9; p = 0.02; Table 9.5). This corresponds to absolute risk for great PAR improvement of 22.9 and 45.8% for aligners and fixed appliance, respectively (Fig. 9.6). The number needed to treat is again 5, which is translated as every fifth case treated with aligners instead of fixed appliances not experiencing a great reduction in PAR scores through treatment, which would be seen if the case had been treated with fixed appliances, and denotes a potentially clinically relevant effect. This discrepancy between the results of the ABO-OGS and the PAR index can be explained by obvious differences between the two tools. The PAR index was developed to assess in a systematic manner the outcome of orthodontic treatment in order to be incorporated in both quality assessment measures of orthodontic care and scientific research. It, however, provides a vague assessment of the occlusion and disregards aspects such as tooth inclination, remaining spaces, and alignment of the posterior dental arch, which are important variable for board examination cases.32 It does not provide a detailed assessment of the position of each tooth and relationship with its neighbors within an ideal dental arch as the ABO-OGS does, which was developed in order to assess the fine details expected to be seen in a meticulously finished case in all three planes (first, second, and third order). Reported limitations of the PAR index55 include, among others, a low weighting for overbite scores and high weighting for overjet scores.56 Indeed, posttreatment PAR scores do not correlate significantly with posttreatment ABO-OGS scores.57 , 58 Subsequently, the PAR index has been widely used to also assess the baseline severity of a case. However, the PAR index to this end does not take into account aspects such as skeletal discrepancies/cephalometric values, developmental tooth anomalies, ectopic teeth, or soft-tissues relationships and again does not correlate well with the ABO DI.57
Fig. 9.6 Illustration of the expected absolute risk for a case to experience a great improvement in its PAR score (PAR reduction of at least 22 points or PAR score of 0 posttreatment) when treated with aligners or fixed appliances, according to the results of a single included study. PAR, peer assessment rating.
Considerable variation was seen in the effect of treatment modality on treatment duration. Meta-analysis of seven studies indicated that on average no definite conclusions can be drawn regarding treatment duration with either aligners or fixed appliances (MD: –0.6 month; 95% CI: –3.7 to 2.6 months; p = 0.73). Extreme heterogeneity was seen across studies (I 2 = 94%), which makes the ability to synthesize existing studies questionable (Fig. 9.7). Specifically, two studies reported statistically significant reduction in treatment duration with aligners and two studies reported statistically significant increase in treatment duration with aligners, while the remaining three studies did not find statistically significant differences. Furthermore, exclusion of a study assessing combined orthodontic/orthognathic treatment47 instead of only orthodontic treatment did not improve the results (six studies; MD: –0.1 month; 95% CI: –3.5 to 3.4 months; I 2 = 95%). Nor was the situation improved by limiting the meta-analysis to only randomized trials (two studies; MD: 2.69 months; 95% CI: –5.0 to 10.4 months; I 2 = 96%) or to only studies with nonextraction treatment (five studies; MD: 0.6 month; 95% CI: –3.2 to 4.4 months; I 2 = 96%). Therefore, it is logical to assume that treatment duration is influenced by many other confounding variables and that the choice of appliance alone does not show a consistent effect on treatment duration.
Fig. 9.7 Contour-enhanced forest plot on the comparison of treatment duration in months between aligners and fixed appliances. AL, aligner; CI, confidence interval; FX, fixed appliance; M, mean; MD, mean difference; N, number of patients; SD, standard deviation. Contours correspond to different effect magnitude and the red dotted line corresponds to 95% random-effects prediction.
Additionally, results of a single study48 indicated that aligners were more efficient in terms of PAR reduction/month of treatment compared to fixed appliances (MD: 0.4 point/month; 95% CI: 0.1–0.7 point/month; p = 0.01). However, as the same study reported that aligners were overall associated with smaller reductions in the PAR scores than fixed appliances, looking at the PAR reduction/month outcome might be misleading.
As far as adverse effects of treatment are concerned, a single identified study on EARR51 reported that significantly smaller percentage of the incisors’ root was resorbed during aligner treatment compared to fixed appliances (MD: –1.8%; 95% CI: –2.4 to –1.3%; p < 0.001; Table 9.5). The same was seen for the various subgroups according to tooth type (central vs. lateral incisor) and jaw (maxilla vs. mandible), but the effect magnitude was on average very small and probably of no clinical relevance. It must be stressed here also that evaluation of EARR during treatment is complicated, since many risk factors come into play, including the patient’s genetic predisposition toward EARR,59 the chosen mechanotherapy,60 the duration of treatment,61 and the actual amount of tooth movement (and especially apical movement).59 A carefully conducted retrospective nonrandomized study taking confounders such as baseline severity through ABO DI, genetic polymorphisms, and absolute apical displacement into account concluded that treatment with orthodontic aligners results in similar amounts of EARR compared to fixed appliances. Therefore, it might be prudent to check if any significant differences in EARR reported in the literature are not rather due to actually teeth being moved less around with aligners.
Additionally, treatment with aligners was not associated, in a single included study46 with significantly lower proclination of the lower incisors compared to fixed appliances (MD: –1.9°; 95% CI: –4.1 to 0.3°; p = 0.10). However, it must be noted that a very small sample was included, which makes the study probably underpowered to identify such a small difference of 1.9° between groups, if this really exists.
Furthermore, no significant difference in the development of gingival recessions 2 years after treatment with aligners or fixed appliances was seen in another single study (MD: 0.9; 95% CI: 0.3–2.7; p = 0.86).50 It might be expected that choice of appliance alone might not directly influence the development of gingival recession. Even if appliance choice was associated with increased anterior anchorage loss/incisor proclination (which was not seen), this would not necessarily translate to increased risk of gingival recession.62 , 63 Although orthodontic treatment on average increases the risk for gingival recessions,64 its precise etiology is multifactorial with risk factors including periodontal disease, mechanical trauma, patient age, smoking, and induction of bone dehiscences by positioning the teeth beyond the alveolar plate.65 – 72
Finally, limited evidence on the effect of appliance choice on loss of periodontal attachment was provided by a single identified study,44 which assessed orthodontic alignment of anterior teeth in adult patients with previous history of treated periodontal disease. After retrieving raw data from the author and matching the study’s groups for baseline status, no differences between aligners and fixed appliances were seen for periodontal probing depth (MD: 0 mm; 95% CI: –0.4 to 0.4 mm; p = 1.00) or alveolar bone levels (MD: 0.1 mm; 95% CI: –0.4 to 0.6 mm; p = 0.69). On the other side, fixed appliances were significantly quicker in repositioning the patients’ migrated anterior teeth compared to aligners (3.9 vs. 6.0 months; MD: –2.1 months; 95% CI: –3.7 to –0.5 months; p = 0.01). It must be noted here that although previous systematic reviews of mostly compromised studies have reported that aligners might be associated with facilitation of better oral hygiene than fixed appliances,18 , 23 , 45 a recent randomized clinical trial73 found no significant consistent advantage in terms of plaque index, gingival index, or periodontal bleeding index between patients treated with aligners and fixed appliances. It seems, therefore, that fixed appliances also can be compatible with proper oral hygiene.
Our confidence in the clinical recommendations that can be formulated based on the quality of evidence using the GRADE framework is presented in Table 9.6. We can say with moderate certainty that compared to treatment with fixed appliances, treatment with orthodontic aligners: (1) leads to worse finishing quality (higher ABO-OGS scores), (2) leads to greater proportion of treated cases that would not pass the ABO examination criteria (ABO-OGS score > 30 points), and (3) makes little to no difference in the development of gingival recessions. This means that future research could have an important impact, which might change current estimates of effect. The main reason for downgrading the quality of evidence pertained to the inclusion of nonrandomized studies that (although being matched) had methodological issues that could introduce some bias. Therefore, even though a potentially large clinical difference in posttreatment ABO-OGS scores between aligners and fixed appliances was seen, this cannot be used as basis to upgrade our confidence in these estimates, as heterogeneity across studies precludes precise effect quantification (Fig. 9.2).
We can say with low confidence that compared to treatment with fixed appliances treatment with orthodontic aligners: (1) leads to lower treatment efficacy (smaller PAR reduction through treatment), (2) leads to greater proportion of treated cases seeing a great improvement (PAR reduction of at least 22 points or PAR score of 0 posttreatment), (3) leads to greater EARR, and (4) makes little to no difference in proclination of the lower incisors during treatment. The main reason for downgrading the quality of evidence pertained to the inclusion of nonrandomized studies with serious/critical methodological issues that most probably introduce bias. This was especially seen in the retrospective study of Gu et al48 that selectively reported data from what might be regarded as “good” cases, while excluding patients with issues of compliance or oral hygiene. This means that further research in terms of well-designed studies is very likely to have an important impact, which is likely to change our current estimates of effect.
Finally, we have very low confidence on the currently observed effect of treatment modality choice (aligners vs. fixed appliances) on treatment duration. This has to do with the fact that very heterogeneous results were seen across existing studies that could not be explained by either clinical heterogeneity, study design, or incorporation of extractions in the treatment plan. Therefore, any estimate about the average difference in duration between aligners and fixed appliances is very uncertain and future well-designed studies should be based on careful selection of cases matched for baseline severity and take into account potential confounders such as case severity, incorporation of extractions, amount of interproximal enamel reduction, and quality of the final occlusal outcome.
Nevertheless, it is important to point out that several threats to the validity of currently available clinical recommendations exist. For one, methodological issues existed for all included studies that might influence conclusions, and this is especially the case for included retrospective nonrandomized studies.11 – 14 Furthermore, most meta-analyses were based predominantly on small trials, which might affect their results.74 Additionally, the small number of trials that were ultimately included in the meta-analyses and their incomplete reporting of results and potential confounders such as level of case severity, oral hygiene, compliance, use of bonded attachments, number of aligners, rate of refinement need, or amount of interproximal enamel reduction precluded the conduct of many analyses for subgroups and meta-regressions that might enable identification of patient subgroups for which aligners might be equally or even more appropriate treatment alternative compared to fixed appliances.
According to currently existing clinical evidence from randomized trials and matched nonrandomized studies on mostly adult patients with mild to severe malocclusions treated with or without extractions, it seems that orthodontic treatment with aligners is associated with worse treatment outcome compared to fixed appliances. On the other side, aligners might be associated with a small decrease in EARR during treatment, while there seems to be little to no difference on proclination of lower incisors and development of gingival recessions. Treatment duration is not defined by choice of appliance alone and patient- or treatment-related factors might come into play.
Since the preparation and submission of this chapter, additional data were made available, which explains small differences in the results of this chapter compared to the subsequent journal paper,74 and a publication notice was issued.76 However, the study’s results and final conclusions in all instances remain practically the same with no threat to their validity.
11 Papageorgiou SN, Kloukos D, Petridis H, Pandis N. Publication of statistically significant research findings in prosthodontics & implant dentistry in the context of other dental specialties. J Dent. 2015;43(10):1195-1202.
13 Papageorgiou SN, Höchli D, Eliades T. Outcomes of comprehensive fixed appliance orthodontic treatment: A systematic review with meta-analysis and methodological overview. Korean J Orthod. 2017;47(6):401-413.
15 Papageorgiou SN, Xavier GM, Cobourne MT, Eliades T. Registered trials report less beneficial treatment effects than unregistered ones: a meta-epidemiological study in orthodontics. J Clin Epidemiol. 2018;100:44-52.
16 Sideri S, Papageorgiou SN, Eliades T. Registration in the international prospective register of systematic reviews (PROSPERO) of systematic review protocols was associated with increased review quality. J Clin Epidemiol. 2018;100:103-110.
22 Aldeeri A, Alhammad L, Alduham A, Ghassan W, Shafshak S, Fatani E. Association of orthodontic clear aligners with root resorption using three-dimension measurements: A systematic review. J Contemp Dent Pract. 2018;19(12):1558-1564.
24 Lu H, Tang H, Zhou T, Kang N. Assessment of the periodontal health status in patients undergoing orthodontic treatment with fixed appliances and Invisalign system: A meta-analysis. Medicine (Baltimore). 2018;97(13):e0248
29 Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol. 2009;62(10):e1-e34.
39 Carrasco-Labra A, Brignardello-Petersen R, Santesso N, et al. Improving GRADE evidence tables part 1: a randomized trial shows improved understanding of content in summary of findings tables with a new format. J Clin Epidemiol. 2016;74:7-18.
40 Schünemann HJ, Cuello C, Akl EA, et al. GRADE Working Group. GRADE guidelines: 18. How ROBINS-I and other tools to assess risk of bias in nonrandomized studies should be used to rate the certainty of a body of evidence. J Clin Epidemiol. 2019;111:105-114.
41 Djeu G, Shelton C, Maganzini A. Outcome assessment of Invisalign and traditional orthodontic treatment compared with the American Board of Orthodontics objective grading system. Am J Orthod Dentofacial Orthop. 2005;128(3):292-298, discussion 298.
42. Fetouh O. Comparison of Treatment Outcome of Invisalign® and Traditional Fixed Orthodontics by Model Analysis Using ABO Objective Grading System. New York, NY: State University of New York at Buffalo; 2009
44 Han JY. A comparative study of combined periodontal and orthodontic treatment with fixed appliances and clear aligners in patients with periodontitis. J Periodontal Implant Sci. 2015;45(6):193-204.
45 Li W, Wang S, Zhang Y. The effectiveness of the Invisalign appliance in extraction cases using the the ABO model grading system: a multicenter randomized controlled trial. Int J Clin Exp Med. 2015;8(5):8276-8282.
47. Robitaille P. Traitement combiné d’orthodontie et de chirurgie orthognatique avec Invisalign®: revue de la durée de traitement et des résultats obtenus [MSc thesis]. Montreal: University of Montreal; 2016
48 Gu J, Tang JS, Skulski B, et al. Evaluation of Invisalign treatment effectiveness and efficiency compared with conventional fixed appliances using the Peer Assessment Rating index. Am J Orthod Dentofacial Orthop. 2017;151(2):259-266.
50 Lanteri V, Farronato G, Lanteri C, Caravita R, Cossellu G. The efficacy of orthodontic treatments for anterior crowding with Invisalign compared with fixed appliances using the Peer Assessment Rating Index. Quintessence Int. 2018;49(7):581-587.
52 Simon M, Keilig L, Schwarze J, Jung BA, Bourauel C. Treatment outcome and efficacy of an aligner technique–regarding incisor torque, premolar derotation and molar distalization. BMC Oral Health. 2014;14:68
53 Kravitz ND, Kusnoto B, BeGole E, Obrez A, Agran B. How well does Invisalign work? A prospective clinical study evaluating the efficacy of tooth movement with Invisalign. Am J Orthod Dentofacial Orthop. 2009;135(1):27-35.
57 Deguchi T, Honjo T, Fukunaga T, Miyawaki S, Roberts WE, Takano-Yamamoto T. Clinical assessment of orthodontic outcomes with the peer assessment rating, discrepancy index, objective grading system, and comprehensive clinical assessment. Am J Orthod Dentofacial Orthop. 2005;127(4):434-443.
58 Hong M, Kook YA, Baek SH, Kim MK. Comparison of treatment outcome assessment for Class I malocclusion patients: Peer Assessment Rating versus American Board of Orthodontics – Objective Grading System. J Korean Dent Sci. 2014;7:6-15.
59 Iglesias-Linares A, Sonnenberg B, Solano B, et al. Orthodontically induced external apical root resorption in patients treated with fixed appliances vs removable aligners. Angle Orthod. 2017;87(1):3-10.
60 Iliadi A, Koletsi D, Eliades T. Forces and moments generated by aligner-type appliances for orthodontic tooth movement: A systematic review and meta-analysis. Orthod Craniofac Res. 2019;22(4):248-258.
61 Samandara A, Papageorgiou SN, Ioannidou-Marathiotou I, Kavvadia-Tsatala S, Papadopoulos MA. Evaluation of orthodontically induced external root resorption following orthodontic treatment using cone beam computed tomography (CBCT): a systematic review and meta-analysis. Eur J Orthod. 2019;41(1):67-79.