SYSTEMATIC REVIEW
Comparison of vacuum-formed and Hawley retainers: A systematic review Wenjia Mai,a Jin'an He,b Hongying Meng,a Yanping Jiang,a Chaoxiao Huang,a Min Li,a Kan Yuan,a and Na Kangc Nanning, Guangxi, China
Introduction: Hawley retainers (HRs) and vacuum-formed retainers (VFRs) are the 2 most commonly used retainers in orthodontics. However, the basis for selection of an appropriate retainer is still a matter of debate among orthodontists. In this systematic review, we evaluated the differences between VFRs and HRs. Methods: Electronic databases (PubMed, EMBASE, Cochrane Library, ISI Web of Science, LILACS, and Pro-Quest) were searched with no language restriction. The relevant orthodontic journals and reference lists were checked for all eligible studies. Two article reviewers independently screened the retrieved studies, extracted the data, and evaluated the quality of the primary studies. Results: A total of 89 articles were retrieved in the initial search. However, only 7 articles met the inclusion criteria. Some evidence suggested that no difference exists to distinguish between the HRs and VFRs with respect to changes in intercanine and intermolar widths after orthodontic retention. In terms of occlusal contacts, cost effectiveness, patient satisfaction, and survival time, there was insufficient evidence to support the use of VFRs over HRs. Conclusions: Additional highquality, randomized, controlled trials concerning these retainers are necessary to determine which retainer is better for orthodontic procedures. (Am J Orthod Dentofacial Orthop 2014;145:720-7)
R
etention is a critical phase of orthodontic treatment. After active orthodontic tooth movement, the teeth might be in an inherently unstable position and have a tendency to return to their pretreatment positions. Currently, the influences of the periodontal and gingival tissues, unstable positions of teeth, and continued skeletal growth are considered to be the major causes of relapse after removal of fixed appliances.1 To address this problem, retainers are used to prevent the teeth from returning to their former positions until gingival and periodontal reorganization and skeletal growth are essentially completed. Although many types of retainers are available, the Hawley retainer (HR) and the vacuum-formed retainer (VFR) are the 2 most commonly used clinical retainers. The HR was designed by Charles Hawley2 in 1919, has been used for nearly a century, and has become the
From the Department of Orthodontics, College of Stomatology, Guangxi Medical University, Nanning, Guangxi, China. a Postgraduate student. b Senior instructor. c Associate professor and director. All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest, and none were reported. Address correspondence to: Na Kang, Department of Orthodontics, College of Stomatology, Guangxi Medical University, Nanning, Guangxi, China 530021; e-mail, [email protected]. Submitted, July 2013; revised and accepted, January 2014. 0889-5406/$36.00 Copyright Ó 2014 by the American Association of Orthodontists. http://dx.doi.org/10.1016/j.ajodo.2014.01.019
720
most popular removable retention appliance. The alternative removable retainer is an invisible retainer that was designed in 1971 and has been referred to by the following names: VFR, clear overlay retainer, and Essix retainer.3 In this review, for simplicity, we considered any invisible retainer as a VFR, instead of the other names. The most compelling potential advantages attributed to this invisible retainer are not only its durability and esthetic qualities, but also its small size and cleanability. Consequently, the use of VFRs has increased exponentially in recent years. In the United Kingdom, VFRs have become the most commonly used retainers in National Health Service, hospital, and private practices.4 However, there is little clinical evidence to support the use of VFRs over conventional HRs. Several published studies have attempted to compare VFRs with HRs. Rowland et al5 conducted a prospective, randomized clinical trial and showed that VFRs were more effective than HRs in retaining the correction of the maxillary and mandibular labial segments. In addition, Demir et al6 also found that VFRs were more efficient in retaining the anterior mandibular teeth during a 1-year retention period. However, a recent retrospective, randomized, double-blind comparison study reported no statistically or clinically significant differences in the effectiveness of HRs and VFRs in maintaining specific arch-form features after orthodontic treatment.7 Other studies have compared these 2 appliances in terms of their cost-effectiveness, patient satisfaction,8,9 survival time,10
Mai et al
and occlusal contacts during retention.11,12 However, pertinent results were inconclusive, and some were unreliable; these studies could bias a clinician's understanding and mislead clinical practice. Thus, we conducted a critical systematic review to evaluate and compare the significant effects of VFRs and conventional HRs. This systematic review might provide clinical evidence to help an orthodontist decide which retainer is appropriate for a particular patient. MATERIAL AND METHODS
Randomized or quasi-randomized, controlled clinical trials were included in this review. Patients who had maxillary retainers, mandibular retainers, or both were included. There was no restriction regarding the type of active orthodontic treatment. The patients had to be followed for at least 6 months after completing their orthodontic treatment. However, patients with severe craniofacial deformities, cleft lip or palate, and poor periodontal status were excluded. For this review, VFRs or HRs were selected as the final retainers for the patients after active orthodontic treatment. Additionally, the retainers had to cover the teeth, at least from first molar to first molar. The primary outcomes included Little's index of irregularity,13 intercanine width, intermolar width, and arch length related to the effectiveness of HRs and VFRs. Secondary outcomes, including cost-effectiveness, patient satisfaction, survival time, and occlusal contacts for these 2 appliances, were extracted and collected. Adverse effects on the periodontal health of the teeth, such as gingival and periodontal diseases, were also evaluated. The following electronic databases were searched with no language restriction: Cochrane Central Register of Controlled Trials (CENTRAL; issue 1 of 12, January 2013), MEDLINE via PubMed (1960 to February 2013), EMBASE (1980 to February 2013), ISI Web of Science (1986 to February 2013), and LILACS (February 22, 2013). The search strategies are shown in Appendix I. In addition, Pro-Quest Dissertation and Thesis database (http://pqdt.lib.sjtu.edu.cn/AdvancedSearch.aspx) and Pro-Quest Science Journals (http://search.proquest. com/sciencejournals/advanced?accountid544440) were searched, with no limits set for the publication date. A manual search was performed of these journals: American Journal of Orthodontics and Dentofacial Orthopedics, Angle Orthodontist, European Journal of Orthodontics, and Journal of Orthodontics (all from 1980 to 2012). In addition, the conference proceedings and abstracts from the British Orthodontic Conference and the
721
European Orthodontic Conference were searched. The reference lists of potential clinical trials were checked to identify any additional studies, and an additional search to update the results was undertaken in July 2013. Two review authors (H.M. and Y.J.) independently screened the studies identified by the search strategies for relevance to this systematic review. Then the eligible studies were used independently for data extraction. Any disagreement between the 2 reviewers was resolved by discussion with another review author (J.H.) on the team. Data extraction was also performed independently by 2 reviewers (H.M. and Y.J.), and disagreements were resolved by discussion with a third reviewer (J.H.). Data from the included studies were entered on a customized data collection form for details, including study design, study participants' characteristics, course of interventions, and outcome measures. In addition, if any ambiguities or lack of data was discovered in the articles, we attempted to contact the authors by mail to obtain more information. Two reviewers (C.H. and M.L.) assessed the risk of bias in each included study independently. Disagreements were resolved by discussion with a third review author (J.H.), so that a consensus could be reached. This assessment followed the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions (version 5.1.0).14 Six specific domains were assessed: sample size calculation, random sequence generation, allocation concealment, blinding of measurement assessment, reporting of withdrawals, and the use of an intention-to-treat analysis. The overall risk of bias in each study was assessed using the following judgments: low, moderate, and high. Studies were categorized according to the following. 1.
2.
3.
Low risk of bias (plausible bias unlikely to seriously alter the results), if 5 or more domains were considered adequate. Moderate risk of bias (plausible bias that raises some doubt about the results), if 3 or more domains were recorded with “yes.” High risk of bias (plausible bias that seriously weakens confidence in the results), if the study recorded “yes” in less than 3 domains.
Statistical analysis
Clinical heterogeneity was assessed by examining the participant types, interventions, and outcomes of each study. Ideally, a meta-analysis would have been performed if studies with similar comparisons reported comparable outcome measures. Risk ratio values would have been calculated along with 95% confidence intervals (CIs)
American Journal of Orthodontics and Dentofacial Orthopedics
June 2014 Vol 145 Issue 6
Mai et al
722
for dichotomous data. Mean differences and 95% CIs would have been used for continuous data. Before each meta-analysis, chi-square and I-square (I2) tests for homogeneity would have been undertaken.15 Heterogeneity would have been considered to exist with a P\0.10 or an I2 value greater than 50%.16 The random-effects model would have been used to incorporate heterogeneity among studies; otherwise, the fixed-effects model would have been used. When necessary, a sensitivity analysis would have been conducted to test the robustness of the findings for the meta-analysis. RESULTS
During the initial search, 89 articles were deemed potentially relevant to the review; 58 were rejected, including duplicates. Then we assessed the titles and abstracts of 33 articles, of which 19 were excluded. The primary reasons for rejection are shown in the PRISMA flowchart (Appendix II). The full texts of 14 articles were assessed extensively, and 7 studies were finally excluded.4,8,17-21 Five studies were cross-sectional studies that lacked a comparator,4,8,17-19 and 2 articles were systematic reviews (Table I).20-21 Therefore, only 7 articles met the inclusion criteria, and their characteristics are summarized in Table II.5-7,9,10,12,22 All 7 included studies were parallel group studies. Five studies were randomized, controlled trials (RCTs),5,7,9,10,22 and 2 were controlled clinical trials.6,12 The changes in intercanine and intermolar widths and Little's index of incisor irregularity were evaluated in 4 studies.5-7,22 However, 1 study compared VFRs and HRs combined with mandibular fixed lingual retainers; the measurements were made only in the mandible.22 In view of the high heterogeneity in the intervention groups, this study was excluded (Table I). Therefore, only 3 studies with high homogeneity were found among the selected articles and used for the data aggregation and analysis.5-7 Of the included articles, the same research group published 2 articles.5,9 Although these articles had the same sample of subjects, the authors used different analyses and obtained different outcomes. The main outcomes of the study of Hichens et al9 were the cost-effectiveness and patient satisfaction between the retainer groups. Detailed assessments of the bias risk from the included studies are shown in Table III. After we assessed the quality of these studies according to the assessment criteria, we deemed the overall bias risk of the 3 studies to be low5,9,10; another 3 trials were considered to have a moderate bias risk6,7,12; and the remaining study was classified as having a high bias risk.22 Of all the included studies, the authors of 4 trials conducted an a priori sample size calculation.5,6,9,10 In these
June 2014 Vol 145 Issue 6
Table I. Characteristics of excluded studies Study 5 studies (references 4, 8, 17, 18, 19) 2 studies (references 20, 21) 1 study (reference 22)
Reason for exclusion These were cross-sectional studies These were systematic reviews This study had high clinical heterogeneity with respect to the intervention groups
studies, the generation of a random sequence was considered adequate because they used computergenerated random sequence,10 blocked randomization method,5,9 or throw of dice.7 However, only 1 study used adequate allocation concealment, in which both patients and researchers were blinded to randomization during the participants' enrollment.10 Although we contacted the authors of 2 studies with regard to this issue, we remained unclear about the allocation concealment in these studies.6,22 In this review, only the blinding of outcomes was assessed. Because the retainers were visible to both patients and orthodontists, blinding of participants and personnel was not feasible in the trials. Four articles were evaluated to have a low risk of bias for outcome blinding.5,6,10,12 Although no blinding was used in 1 study, the outcomes and outcome measurements were not likely to have been influenced by the lack of blinding.10 In 6 studies, the number of dropouts and the reasons for withdrawal were clearly described,5-7,9,10,22 but an intention-to-treat analysis was conducted only in 3 trials.5,9,10 There were no withdrawals in 1 study.12 The stability of the orthodontic patients' teeth with HRs and VFRs after active orthodontic tooth movement was reported in 4 studies.5-7,22 The details of the outcomes and conclusions are described in Table IV. Three studies reported no significant differences in the ability of HRs and VFRs to retain the dentition in terms of intercanine and intermolar widths.5-7 No significant differences in arch lengths were reported in 2 studies.6,7 One study was not appropriate for the quantitative analysis because it had high clinical heterogeneity with respect to the intervention groups.22 Unfortunately, we were unable to pool the data of the remaining 3 studies for a meta-analysis because they used different statistical methods and had high heterogeneity with regard to the data sizes.5-7 Although we attempted to contact the authors to request the primary data, unfortunately, we received no additional information. One study, which involved 30 participants, compared the changes in the number of occlusal contacts between HRs and VFRs after 3 months of retainer use.12 The 30
American Journal of Orthodontics and Dentofacial Orthopedics
Mai et al
723
Table II. Summary of studies included in the systematic review Study Methods Rowland et al,5 2007 RCT The subjects were observed for 6 mo after debond
Barlin et al,7 2011
Participants 397 patients: male, 156; female, 241 Mean ages, 15 y (1.5) for VFR and 14.8 y (1.8) for HR RCT 82 patients: male, 29; Observed for 2, 6, and12 mo female, 53 after debond Mean age for all subjects, 14.9 y (2.7)
Demir et al,6 2012
CCT Retention time was 1 y
Hichens et al,9 2007
RCT The subjects were observed for 6 mo after debond
Sun et al,10 2011
RCT Retention time was 12 mo
Sauget et al,12 1997
CCT Retention time was 3 mo
Xu et al,22 2011
RCT Retention time for 12 mo
Interventions Outcomes VFR: 201 patients, 24 h/d for Changes in intercanine and the first week, then 12 h/d intermolar widths; Little's HR: 196 patients, 24 h/d for index of incisor 3 mo, then 12 h/d for irregularity for both another 3 mo groups VFR: 40; HR: 42 Changes in intercanine and All subjects wearing retainers intermolar widths; Little's 24 h/d for 12 mo except index of incisor for cleaning irregularity and arch lengths for both groups VFR: 22; HR: 20 Irregularity index, All the subjects wearing intercanine width, and retainers 24 h/d for 12 mo lengths of both arches except during meals
42 patients: male, 12; female, 30 Mean ages, 13.8 y (3.1) for VFR and 12.9 y (2.5) for HR 397 patients: male, 156; VFR: 201 patients, 24 h/d for female, 241 the first week, then 12 h/d Mean ages, 15 y (1.5) for VFR HR: 196 patients, 24 h/d for and 14.8 y (1.8) for HR 3 mo, then 12 h/d for another 3 mo 120 participants: male, 60; VFR: 59; HR: 61 female, 60 Wearing retainers full time, Mean age of all subjects, except during meals 14.7 y 30 participants: male, 11; VFR: 13, wearing retainers female, 19 full time for first 3 days Mean ages, 19.6 y for VFR (except during meals) and and 18.8 y for HR nightly thereafter HR: 17 (except 2 had maxillary HR only), wearing retainers full time except during meals 45 participants: male, 16; VFR: 25, wearing full time, female, 29 except during meals and Mean ages, 13.6 y for VFR brushing and 15.2 y for HR HR: 20, wearing nightly, but combined with mandibular fixed lingual retainers
Cost-effectiveness and patient satisfaction between the retainer groups Survival times of the 2 types of retainers
Occlusal contacts of the maxillary and mandibular teeth with the 2 types of retainers
Changes in overbite, overjet, intercanine and intermolar widths, Little's index, and calculus index scores for both groups in the mandible
CCT, Controlled clinical trial.
consecutive patients were assigned alternatively to 1 of 2 retainer groups at the orthodontic clinic at the University of the Pacific School of Dentistry, San Francisco, Calif. Impression material was used to record the occlusal contacts of each patient at debonding, at retainer delivery, and after a 3-month retention. The anteroposterior contacts were also assessed in both groups. The numbers of total contacts and posterior contacts were significantly higher in the HR group than in the VFR group from retainer delivery to 3-month retention, and from debonding to 3-month retention. The authors of 1 study compared the satisfaction of 397 participants who were randomly assigned to 2 intervention groups (HR, 196; VFR, 201), and the participants completed a total of 350 patient satisfaction questionnaires
(Hawley, 168; VFR, 182) at 6 months.9 The dropout rate was 11.8%. The main outcomes were the frequency of responses to the satisfaction-related questions in the questionnaires. A statistically significant result for the factor of embarrassment was found more often in the HR group than in the VFR group (P 5 0.005), particularly in terms of speech and esthetics. More patients felt better in the VFR group than in HR group when the retainers were compared with fixed appliances (P \0.001). However, there was no statistically significant difference in the amount of time a retainer was worn away from home. The authors of 1 study investigated the costeffectiveness of 3 groups, including the National Health Service, orthodontic practices, and patients in the United Kingdom.9 The researchers further investigated
American Journal of Orthodontics and Dentofacial Orthopedics
June 2014 Vol 145 Issue 6
Mai et al
724
Table III. Assessment of risk of bias Study Rowland et al,5 2007 Barlin et al,7 2011 Demir et al,6 2012 Hichens et al,9 2007 Sun et al,10 2011 Sauget et al,12 1997 Xu et al,22 2011
Design RCT RCT CCT RCT RCT CCT RCT
Sample size calculation Yes No Yes Yes Yes No No
Random sequence generation Yes Yes Unclear Yes Yes No Unclear
Allocation concealment Yes No Unclear Yes Yes No Unclear
Blinding of measurement Yes Unclear Yes No Yes Yes Unclear
Withdrawals reported Yes Yes Yes Yes Yes – Yes
ITT Yes No Unclear Yes Yes – No
Risk of bias Low Moderate Moderate Low Low Moderate High
Yes, Low risk of bias; no, high risk of bias; unclear, unclear or unknown risk of bias. ITT, Intention to treat; CCT, controlled clinical trial.
Table IV. Effectiveness of HRs and VFRs Study Rowland et al,5 2007
Demir et al,6 2012 Barlin et al,7 2011
Intercanine Intermolar Arch width width length Little's index Conclusions NS NS – Effective (P 5 0.013) VFRs were more effective than HRs at holding NS NS – Effective (P \0.01) corrections of the maxillary and mandibular labial segments Maxillary NS NS NS NS No statistically significant difference was found Mandibular NS NS NS NS in the effectiveness between these 2 retainers Maxillary NS NS NS NS There was no statistically or clinically significant Mandibular NS NS NS NS difference in the measured arch width, arch length, or modified Little's index over a 12-mo period Arch Maxillary Mandibular
Effective, VFRs were more effective than HRs. NS, No statistically significant difference between HRs and VFRs.
62 subjects (HR, 41; VFR, 21) who attended extra appointments for problems with their retainers with regard to their travel costs, child-care costs, patient fees, and lost income. The HR patients had statistically significantly greater costs than did the VFR patients, with a mean difference of V2.15 (95% CI, –V2.90-V7.57). Greater costs were demonstrated with respect to the National Health Service and the orthodontic practices for the HR groups than the VFR groups, with a statistically significant result and with mean differences of V31.35 (95% CI, V28.06-V34.68) and V32.60 (95% CI, V30.58-V34.67), respectively. Therefore, the cost-effectiveness analysis of the retainers showed that VFRs were more cost-effective than HRs from 3 perspectives over a 6-month retention period. A study conducted by Sun et al10 investigated the survival time of HRs and VFRs over a 1-year follow-up of 120 adolescent patients who were randomly assigned to 1 of 2 intervention groups. The survival times of the maxillary and mandibular retainers were assessed separately. There was no statistically significant difference between the 2 groups with either the maxillary retainers (P 5 0.254) or the mandibular retainers (P 5 0.188). Another study by Hichens et al9 also investigated the rates of retainer breakage and loss in the 2 retainer
June 2014 Vol 145 Issue 6
groups after a 6-month retention period. Interestingly, more subjects broke their retainers in the HR group than in the VFR group, with a statistically significant difference (P\0.001). However, no difference was found in the loss rates of the retainers between the groups. Unfortunately, we were unable to retrieve the original data regarding the breakage and loss rates from a study after attempting to contact the authors.10 Therefore, we did not perform a data analysis of these 2 trials. DISCUSSION
In this systematic review, 5 RCTs and 2 controlled clinical trials satisfied our inclusion criteria and were analyzed after searching and assessing the quality and the data extraction methods. Most studies did not use adequate randomization and allocation concealment. Although an intention-to-treat analysis is often recommended as the least biased way to estimate the intervention effects in randomized trials,23 only 3 studies reported usage of an intention-to-treat analysis for the incomplete outcome data.5,9,10 A CONSORT flow diagram is suggested as an appropriate way to improve the quality of data reported from parallel-group randomized trials,24
American Journal of Orthodontics and Dentofacial Orthopedics
Mai et al
but just 3 studies used this method.5,9,10 Therefore, more prospective research in this area should be conducted and reported in accordance with the CONSORT guidelines, so that more high-quality RCTs will be eligible for future meta-analyses. Unfortunately, no studies were eligible for a metaanalysis in this systematic review. The effectiveness of HRs and VFRs was evaluated in 3 studies.5-7 All reported no statistically significant differences with respect to changes in the intercanine and intermolar widths during the follow-up periods. The comparison of occlusal contacts between HRs and VFRs was assessed in 1 study after a 3-month retention period.12 Sauget et al12 found that the number of occlusal contacts with HRs was significantly more than with VFRs, especially the posterior contacts. These results were supported in a study by Hoybjerg et al.25 With respect to HRs, the authors of several studies also found a statistically significant increase in the number of posterior contacts after a retention period with HRs.11,26,27 These findings showed that HRs allowed relative vertical movement of the posterior teeth during the retention stage. Therefore, HRs might be a better choice for patients who did not have good posterior settling after active orthodontic treatment. With respect to VFRs, a recent study by Dincer and Isik Aslan28 reported that changes in the occlusal contacts were not observed at the end of a 9-month retention with VFRs, whereas there was a significant increase after a 2.5-year follow-up. These results could be interpreted with the following possibilities. Because of the construction of VFRs, which cover the occlusal surfaces of the teeth, no posterior contacts increased significantly during retention. By removing the retainers, the continued mobility of the teeth might contribute to continual vertical settling in a long-term follow-up. In addition, the authors also concluded that both ideal and nonideal posterior contacts increased in the long term, and the numbers of nonideal contacts were greater than those of ideal contacts.28 Consequently, the results of the included study should be interpreted with caution because of a short retention time—only 3 months—and no long-term follow-up.12 The authors of 1 study investigated patient satisfaction and found that wearing a VFR causes less embarrassment than wearing an HR (P 5 0.005).9 A similar result was demonstrated in a study by Pratt et al,8 who found that patients were more compliant with VFRs than with HRs after the fixed appliances were debonded. A possible explanation was that VFRs are made of transparent plastic and are more esthetically pleasing and lighter than HRs, which have a stainless steel wire across the labial surface of the anterior teeth. Also, speech articulation might be another aspect with respect to patient satisfaction.
725
Haydar et al29 found that patients experienced articulation problems while wearing HRs during the first week. However, to date, no RCT has evaluated and compared speech articulation differences between VFRs and HRs. Therefore, a speech articulation analysis between the 2 retainers should be considered in future studies. In addition, Hichens et al9 conducted an economic evaluation and found that VFRs were more costeffective than HRs during retention. However, because of variability across study settings, this study, which was conducted in the United Kingdom, cannot be representative of costs in other countries. More RCTs concerning the economic evaluations of retainers in different regions of the world are needed to confirm this finding. When comparing the survival times between the 2 retainers, the thickness and materials of the VFRs and the compliance of patients should be carefully considered. First, although the study by Sun et al10 used 0.75-mm VFRs, until now, no uniform standards have been established for the thickness of VFRs. In some other studies, VFRs of 0.75, 1, and 1.5 mm were used.5,28,30 Second, the different thermoplastic materials should be considered to have an effect on VFR wear resistance and durability. Gardner et al31 found that TR material, a hard polyethylene terephthalate glycol copolymer, demonstrated greater wear resistance than the other 2 materials, which were softer, polypropylene-based thermoplastics. Furthermore, although no study indicated a correlation between retainer damage and patient compliance, we found increased compliance and increased wear with VFRs in our clinical practice. However, it remains unanswered whether increased compliance could decrease the VFR survival time. Further studies concerning these 3 points are necessary. Two systematic reviews were identified in our review.20,21 The article by Ge et al21 focused only on the outcomes pertaining to the irregularity of incisors and on intercanine and intermolar widths to compare the effectiveness of HRs and VFRs, whereas our review evaluated the outcomes including not only clinical effectiveness, but also occlusal contacts, cost-effectiveness, patient satisfaction, and survival time. The findings regarding the effectiveness of retainers were similar in our opinion21; however, the outcomes pertaining to irregularity should be interpreted with caution because of the poor quality of the included studies. Because of insufficient evidence, we cannot suggest that VFRs are more effective than HRs; furthermore, high-quality RCTs are necessary. Another review was partly relevant to our review concerning occlusal contacts between HRs and VFRs.20 The findings were consistent in our opinion because HRs were found to have more occlusal contacts than VFRs, but these findings also should be interpreted
American Journal of Orthodontics and Dentofacial Orthopedics
June 2014 Vol 145 Issue 6
Mai et al
726
with caution because of the poor quality of the included studies. Although evidence regarding these 2 retainers is currently insufficient, it does not indicate that this evidence should not be used in clinical practice. Our review comprehensively and objectively evaluated the differences between HRs and VFRs. Through this review, we believe it is both easy and useful for orthodontists and orthodontic patients to understand these 2 retainers; however, we also think that an inadequate number of studies has been conducted on these 2 retainers and thus believe that this necessitates further research. CONCLUSIONS
Some evidence suggested that there are no differences with respect to changes in intercanine and intermolar widths between HRs and VFRs after orthodontic retention. However, at present, there is insufficient evidence that VFRs are more effective than HRs. In terms of occlusal contacts, cost-effectiveness, patient satisfaction, and survival time, there is also a lack of sufficient evidence of differences between these 2 retainers. In addition, no evidence concerning periodontal health was retrieved from the studies. Therefore, this systematic review suggests that further high-quality RCTs regarding the differences between HRs and VFRs during orthodontic retention are necessary to determine which retainer is the better selection for orthodontists. REFERENCES 1. Melrose C, Millett DT. Toward a perspective on orthodontic retention? Am J Orthod Dentofacial Orthop 1998;113:507-14. 2. Hawley CA. A removable retainer. Int J Orthod 1919;5:291-305. 3. Ponitz RJ. Invisible retainers. Am J Orthod 1971;59:266-72. 4. Singh P, Grammati S, Kirschen R. Orthodontic retention patterns in the United Kingdom. J Orthod 2009;36:115-21. 5. Rowland H, Hichens L, Williams A, Hills D, Killingback N, Ewings P, et al. The effectiveness of Hawley and vacuum-formed retainers: a single-center randomized controlled trial. Am J Orthod Dentofacial Orthop 2007;132:730-7. 6. Demir A, Babacan H, Nalcaci R, Topcuoglu T. Comparison of retention characteristics of Essix and Hawley retainers. Korean J Orthod 2012;42:255-62. 7. Barlin S, Smith R, Reed R, Sandy J, Ireland AJ. A retrospective randomized double-blind comparison study of the effectiveness of Hawley vs vacuum-formed retainers. Angle Orthod 2011;81:404-9. 8. Pratt MC, Kluemper GT, Lindstrom AF. Patient compliance with orthodontic retainers in the postretention phase. Am J Orthod Dentofacial Orthop 2011;140:196-201. 9. Hichens L, Rowland H, Williams A, Hollinghurst S, Ewings P, Clark S, et al. Cost-effectiveness and patient satisfaction: Hawley and vacuum-formed retainers. Eur J Orthod 2007;29:372-8. 10. Sun J, Yu YC, Liu MY, Chen L, Li HW, Zhang L, et al. Survival time comparison between Hawley and clear overlay retainers: a randomized trial. J Dent Res 2011;90:1197-201.
June 2014 Vol 145 Issue 6
11. Basciftci FA, Uysal T, Sari Z, Inan O. Occlusal contacts with different retention procedures in 1-year follow-up period. Am J Orthod Dentofacial Orthop 2007;131:357-62. 12. Sauget E, Covell DA Jr, Boero RP, Lieber WS. Comparison of occlusal contacts with use of Hawley and clear overlay retainers. Angle Orthod 1997;67:223-30. 13. Little RM. The irregularity index: a quantitative score of mandibular anterior alignment. Am J Orthod 1975;68:554-63. 14. Higgins JPT GSe. Cochrane handbook for systematic reviews of interventions (version 5.1.0, updated March 2011). The Cochrane Collaboration 2011. Available at: www.cochrane-handbook.org. Accessed February 26, 2013. 15. Higgins JP, Thompson SG. Quantifying heterogeneity in a metaanalysis. Stat Med 2002;21:1539-58. 16. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557-60. 17. Mollov ND, Lindauer SJ, Best AM, Shroff B, Tufekci E. Patient attitudes toward retention and perceptions of treatment success. Angle Orthod 2010;80:468-73. 18. Valiathan M, Hughes E. Results of a survey-based study to identify common retention practices in the United States. Am J Orthod Dentofacial Orthop 2010;137:170-7. 19. Pratt MC, Kluemper GT, Hartsfield JK Jr, Fardo D, Nash DA. Evaluation of retention protocols among members of the American Association of Orthodontists in the United States. Am J Orthod Dentofacial Orthop 2011;140:520-6. 20. Littlewood SJ, Millett DT, Doubleday B, Bearn DR, Worthington HV. Retention procedures for stabilising tooth position after treatment with orthodontic braces. Cochrane Database Syst Rev 2006;CD002283. 21. Ge ZL, Jiao X, Tian JH, Yang KH. Clinical effectiveness of vacuumformed versus Hawley retainers: a systematic review. Chin J Evid Based Med 2012;12:596-601. 22. Xu XC, Li RM, Tang GH. Clinical evaluation of lingual fixed retainer combined with Hawley retainer and vacuum-formed retainer. Shanghai Kou Qiang Yi Xue 2011;20:623-6. 23. Newell DJ. Intention-to-treat analysis: implications for quantitative and qualitative research. Int J Epidemiol 1992;21:837-41. 24. Moher D, Schulz KF, Altman D. The CONSORT statement: revised recommendations for improving the quality of reports of parallelgroup randomized trials. JAMA 2001;285:1987-91. 25. Hoybjerg AJ, Currier GF, Kadioglu O. Evaluation of 3 retention protocols using the American Board of Orthodontics cast and radiograph evaluation. Am J Orthod Dentofacial Orthop 2013; 144:16-22. 26. Haydar B, Ciger S, Saatci P. Occlusal contact changes after the active phase of orthodontic treatment. Am J Orthod Dentofacial Orthop 1992;102:22-8. 27. Sari Z, Uysal T, Basciftci FA, Inan O. Occlusal contact changes with removable and bonded retainers in a 1-year retention period. Angle Orthod 2009;79:867-72. 28. Dinc¸er M, Isik Aslan B. Effects of thermoplastic retainers on occlusal contacts. Eur J Orthod 2010;32:6-10. 29. Haydar B, Karabulut G, Ozkan S, Aksoy AU, Ciger S. Effects of retainers on the articulation of speech. Am J Orthod Dentofacial Orthop 1996;110:535-40. 30. Pascual AL, Beeman CS, Hicks EP, Bush HM, Mitchell RJ. The essential work of fracture of thermoplastic orthodontic retainer materials. Angle Orthod 2010;80:554-61. 31. Gardner GD, Dunn WJ, Taloumis L. Wear comparison of thermoplastic materials used for orthodontic retainers. Am J Orthod Dentofacial Orthop 2003;124:294-7.
American Journal of Orthodontics and Dentofacial Orthopedics
Mai et al
APPENDIX I: SEARCH STRATEGIES
MEDLINE (via PubMed) search strategy 1. 2. 3. 4. 5. 6. 7.
exp ORTHODONTICS/ orthodontic*.mp. or/1-2 (retention or retain*).mp. (Vacuum formed or Vacuum*).mp. (thermoplastic* or invisible* or clear*).mp. Essix*.mp.
727
8. Hawley*.mp. 9. 5 or 6 or 7 10. 8 and 9 11. 4 and 10 12. 3 and 11 *indicates truncation. .mp. indicates a search of title, original title, abstract, name of substance word, and subject heading word.
APPENDIX II: PRISMA FLOW CHART OF THIS SYSTEMATIC REVIEW
American Journal of Orthodontics and Dentofacial Orthopedics
June 2014 Vol 145 Issue 6
Comparison of vacuum-formed and Hawley retainers: A systematic review Wenjia Mai,a Jin'an He,b Hongying Meng,a Yanping Jiang,a Chaoxiao Huang,a Min Li,a Kan Yuan,a and Na Kangc Nanning, Guangxi, China
Introduction: Hawley retainers (HRs) and vacuum-formed retainers (VFRs) are the 2 most commonly used retainers in orthodontics. However, the basis for selection of an appropriate retainer is still a matter of debate among orthodontists. In this systematic review, we evaluated the differences between VFRs and HRs. Methods: Electronic databases (PubMed, EMBASE, Cochrane Library, ISI Web of Science, LILACS, and Pro-Quest) were searched with no language restriction. The relevant orthodontic journals and reference lists were checked for all eligible studies. Two article reviewers independently screened the retrieved studies, extracted the data, and evaluated the quality of the primary studies. Results: A total of 89 articles were retrieved in the initial search. However, only 7 articles met the inclusion criteria. Some evidence suggested that no difference exists to distinguish between the HRs and VFRs with respect to changes in intercanine and intermolar widths after orthodontic retention. In terms of occlusal contacts, cost effectiveness, patient satisfaction, and survival time, there was insufficient evidence to support the use of VFRs over HRs. Conclusions: Additional highquality, randomized, controlled trials concerning these retainers are necessary to determine which retainer is better for orthodontic procedures. (Am J Orthod Dentofacial Orthop 2014;145:720-7)
R
etention is a critical phase of orthodontic treatment. After active orthodontic tooth movement, the teeth might be in an inherently unstable position and have a tendency to return to their pretreatment positions. Currently, the influences of the periodontal and gingival tissues, unstable positions of teeth, and continued skeletal growth are considered to be the major causes of relapse after removal of fixed appliances.1 To address this problem, retainers are used to prevent the teeth from returning to their former positions until gingival and periodontal reorganization and skeletal growth are essentially completed. Although many types of retainers are available, the Hawley retainer (HR) and the vacuum-formed retainer (VFR) are the 2 most commonly used clinical retainers. The HR was designed by Charles Hawley2 in 1919, has been used for nearly a century, and has become the
From the Department of Orthodontics, College of Stomatology, Guangxi Medical University, Nanning, Guangxi, China. a Postgraduate student. b Senior instructor. c Associate professor and director. All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest, and none were reported. Address correspondence to: Na Kang, Department of Orthodontics, College of Stomatology, Guangxi Medical University, Nanning, Guangxi, China 530021; e-mail, [email protected]. Submitted, July 2013; revised and accepted, January 2014. 0889-5406/$36.00 Copyright Ó 2014 by the American Association of Orthodontists. http://dx.doi.org/10.1016/j.ajodo.2014.01.019
720
most popular removable retention appliance. The alternative removable retainer is an invisible retainer that was designed in 1971 and has been referred to by the following names: VFR, clear overlay retainer, and Essix retainer.3 In this review, for simplicity, we considered any invisible retainer as a VFR, instead of the other names. The most compelling potential advantages attributed to this invisible retainer are not only its durability and esthetic qualities, but also its small size and cleanability. Consequently, the use of VFRs has increased exponentially in recent years. In the United Kingdom, VFRs have become the most commonly used retainers in National Health Service, hospital, and private practices.4 However, there is little clinical evidence to support the use of VFRs over conventional HRs. Several published studies have attempted to compare VFRs with HRs. Rowland et al5 conducted a prospective, randomized clinical trial and showed that VFRs were more effective than HRs in retaining the correction of the maxillary and mandibular labial segments. In addition, Demir et al6 also found that VFRs were more efficient in retaining the anterior mandibular teeth during a 1-year retention period. However, a recent retrospective, randomized, double-blind comparison study reported no statistically or clinically significant differences in the effectiveness of HRs and VFRs in maintaining specific arch-form features after orthodontic treatment.7 Other studies have compared these 2 appliances in terms of their cost-effectiveness, patient satisfaction,8,9 survival time,10
Mai et al
and occlusal contacts during retention.11,12 However, pertinent results were inconclusive, and some were unreliable; these studies could bias a clinician's understanding and mislead clinical practice. Thus, we conducted a critical systematic review to evaluate and compare the significant effects of VFRs and conventional HRs. This systematic review might provide clinical evidence to help an orthodontist decide which retainer is appropriate for a particular patient. MATERIAL AND METHODS
Randomized or quasi-randomized, controlled clinical trials were included in this review. Patients who had maxillary retainers, mandibular retainers, or both were included. There was no restriction regarding the type of active orthodontic treatment. The patients had to be followed for at least 6 months after completing their orthodontic treatment. However, patients with severe craniofacial deformities, cleft lip or palate, and poor periodontal status were excluded. For this review, VFRs or HRs were selected as the final retainers for the patients after active orthodontic treatment. Additionally, the retainers had to cover the teeth, at least from first molar to first molar. The primary outcomes included Little's index of irregularity,13 intercanine width, intermolar width, and arch length related to the effectiveness of HRs and VFRs. Secondary outcomes, including cost-effectiveness, patient satisfaction, survival time, and occlusal contacts for these 2 appliances, were extracted and collected. Adverse effects on the periodontal health of the teeth, such as gingival and periodontal diseases, were also evaluated. The following electronic databases were searched with no language restriction: Cochrane Central Register of Controlled Trials (CENTRAL; issue 1 of 12, January 2013), MEDLINE via PubMed (1960 to February 2013), EMBASE (1980 to February 2013), ISI Web of Science (1986 to February 2013), and LILACS (February 22, 2013). The search strategies are shown in Appendix I. In addition, Pro-Quest Dissertation and Thesis database (http://pqdt.lib.sjtu.edu.cn/AdvancedSearch.aspx) and Pro-Quest Science Journals (http://search.proquest. com/sciencejournals/advanced?accountid544440) were searched, with no limits set for the publication date. A manual search was performed of these journals: American Journal of Orthodontics and Dentofacial Orthopedics, Angle Orthodontist, European Journal of Orthodontics, and Journal of Orthodontics (all from 1980 to 2012). In addition, the conference proceedings and abstracts from the British Orthodontic Conference and the
721
European Orthodontic Conference were searched. The reference lists of potential clinical trials were checked to identify any additional studies, and an additional search to update the results was undertaken in July 2013. Two review authors (H.M. and Y.J.) independently screened the studies identified by the search strategies for relevance to this systematic review. Then the eligible studies were used independently for data extraction. Any disagreement between the 2 reviewers was resolved by discussion with another review author (J.H.) on the team. Data extraction was also performed independently by 2 reviewers (H.M. and Y.J.), and disagreements were resolved by discussion with a third reviewer (J.H.). Data from the included studies were entered on a customized data collection form for details, including study design, study participants' characteristics, course of interventions, and outcome measures. In addition, if any ambiguities or lack of data was discovered in the articles, we attempted to contact the authors by mail to obtain more information. Two reviewers (C.H. and M.L.) assessed the risk of bias in each included study independently. Disagreements were resolved by discussion with a third review author (J.H.), so that a consensus could be reached. This assessment followed the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions (version 5.1.0).14 Six specific domains were assessed: sample size calculation, random sequence generation, allocation concealment, blinding of measurement assessment, reporting of withdrawals, and the use of an intention-to-treat analysis. The overall risk of bias in each study was assessed using the following judgments: low, moderate, and high. Studies were categorized according to the following. 1.
2.
3.
Low risk of bias (plausible bias unlikely to seriously alter the results), if 5 or more domains were considered adequate. Moderate risk of bias (plausible bias that raises some doubt about the results), if 3 or more domains were recorded with “yes.” High risk of bias (plausible bias that seriously weakens confidence in the results), if the study recorded “yes” in less than 3 domains.
Statistical analysis
Clinical heterogeneity was assessed by examining the participant types, interventions, and outcomes of each study. Ideally, a meta-analysis would have been performed if studies with similar comparisons reported comparable outcome measures. Risk ratio values would have been calculated along with 95% confidence intervals (CIs)
American Journal of Orthodontics and Dentofacial Orthopedics
June 2014 Vol 145 Issue 6
Mai et al
722
for dichotomous data. Mean differences and 95% CIs would have been used for continuous data. Before each meta-analysis, chi-square and I-square (I2) tests for homogeneity would have been undertaken.15 Heterogeneity would have been considered to exist with a P\0.10 or an I2 value greater than 50%.16 The random-effects model would have been used to incorporate heterogeneity among studies; otherwise, the fixed-effects model would have been used. When necessary, a sensitivity analysis would have been conducted to test the robustness of the findings for the meta-analysis. RESULTS
During the initial search, 89 articles were deemed potentially relevant to the review; 58 were rejected, including duplicates. Then we assessed the titles and abstracts of 33 articles, of which 19 were excluded. The primary reasons for rejection are shown in the PRISMA flowchart (Appendix II). The full texts of 14 articles were assessed extensively, and 7 studies were finally excluded.4,8,17-21 Five studies were cross-sectional studies that lacked a comparator,4,8,17-19 and 2 articles were systematic reviews (Table I).20-21 Therefore, only 7 articles met the inclusion criteria, and their characteristics are summarized in Table II.5-7,9,10,12,22 All 7 included studies were parallel group studies. Five studies were randomized, controlled trials (RCTs),5,7,9,10,22 and 2 were controlled clinical trials.6,12 The changes in intercanine and intermolar widths and Little's index of incisor irregularity were evaluated in 4 studies.5-7,22 However, 1 study compared VFRs and HRs combined with mandibular fixed lingual retainers; the measurements were made only in the mandible.22 In view of the high heterogeneity in the intervention groups, this study was excluded (Table I). Therefore, only 3 studies with high homogeneity were found among the selected articles and used for the data aggregation and analysis.5-7 Of the included articles, the same research group published 2 articles.5,9 Although these articles had the same sample of subjects, the authors used different analyses and obtained different outcomes. The main outcomes of the study of Hichens et al9 were the cost-effectiveness and patient satisfaction between the retainer groups. Detailed assessments of the bias risk from the included studies are shown in Table III. After we assessed the quality of these studies according to the assessment criteria, we deemed the overall bias risk of the 3 studies to be low5,9,10; another 3 trials were considered to have a moderate bias risk6,7,12; and the remaining study was classified as having a high bias risk.22 Of all the included studies, the authors of 4 trials conducted an a priori sample size calculation.5,6,9,10 In these
June 2014 Vol 145 Issue 6
Table I. Characteristics of excluded studies Study 5 studies (references 4, 8, 17, 18, 19) 2 studies (references 20, 21) 1 study (reference 22)
Reason for exclusion These were cross-sectional studies These were systematic reviews This study had high clinical heterogeneity with respect to the intervention groups
studies, the generation of a random sequence was considered adequate because they used computergenerated random sequence,10 blocked randomization method,5,9 or throw of dice.7 However, only 1 study used adequate allocation concealment, in which both patients and researchers were blinded to randomization during the participants' enrollment.10 Although we contacted the authors of 2 studies with regard to this issue, we remained unclear about the allocation concealment in these studies.6,22 In this review, only the blinding of outcomes was assessed. Because the retainers were visible to both patients and orthodontists, blinding of participants and personnel was not feasible in the trials. Four articles were evaluated to have a low risk of bias for outcome blinding.5,6,10,12 Although no blinding was used in 1 study, the outcomes and outcome measurements were not likely to have been influenced by the lack of blinding.10 In 6 studies, the number of dropouts and the reasons for withdrawal were clearly described,5-7,9,10,22 but an intention-to-treat analysis was conducted only in 3 trials.5,9,10 There were no withdrawals in 1 study.12 The stability of the orthodontic patients' teeth with HRs and VFRs after active orthodontic tooth movement was reported in 4 studies.5-7,22 The details of the outcomes and conclusions are described in Table IV. Three studies reported no significant differences in the ability of HRs and VFRs to retain the dentition in terms of intercanine and intermolar widths.5-7 No significant differences in arch lengths were reported in 2 studies.6,7 One study was not appropriate for the quantitative analysis because it had high clinical heterogeneity with respect to the intervention groups.22 Unfortunately, we were unable to pool the data of the remaining 3 studies for a meta-analysis because they used different statistical methods and had high heterogeneity with regard to the data sizes.5-7 Although we attempted to contact the authors to request the primary data, unfortunately, we received no additional information. One study, which involved 30 participants, compared the changes in the number of occlusal contacts between HRs and VFRs after 3 months of retainer use.12 The 30
American Journal of Orthodontics and Dentofacial Orthopedics
Mai et al
723
Table II. Summary of studies included in the systematic review Study Methods Rowland et al,5 2007 RCT The subjects were observed for 6 mo after debond
Barlin et al,7 2011
Participants 397 patients: male, 156; female, 241 Mean ages, 15 y (1.5) for VFR and 14.8 y (1.8) for HR RCT 82 patients: male, 29; Observed for 2, 6, and12 mo female, 53 after debond Mean age for all subjects, 14.9 y (2.7)
Demir et al,6 2012
CCT Retention time was 1 y
Hichens et al,9 2007
RCT The subjects were observed for 6 mo after debond
Sun et al,10 2011
RCT Retention time was 12 mo
Sauget et al,12 1997
CCT Retention time was 3 mo
Xu et al,22 2011
RCT Retention time for 12 mo
Interventions Outcomes VFR: 201 patients, 24 h/d for Changes in intercanine and the first week, then 12 h/d intermolar widths; Little's HR: 196 patients, 24 h/d for index of incisor 3 mo, then 12 h/d for irregularity for both another 3 mo groups VFR: 40; HR: 42 Changes in intercanine and All subjects wearing retainers intermolar widths; Little's 24 h/d for 12 mo except index of incisor for cleaning irregularity and arch lengths for both groups VFR: 22; HR: 20 Irregularity index, All the subjects wearing intercanine width, and retainers 24 h/d for 12 mo lengths of both arches except during meals
42 patients: male, 12; female, 30 Mean ages, 13.8 y (3.1) for VFR and 12.9 y (2.5) for HR 397 patients: male, 156; VFR: 201 patients, 24 h/d for female, 241 the first week, then 12 h/d Mean ages, 15 y (1.5) for VFR HR: 196 patients, 24 h/d for and 14.8 y (1.8) for HR 3 mo, then 12 h/d for another 3 mo 120 participants: male, 60; VFR: 59; HR: 61 female, 60 Wearing retainers full time, Mean age of all subjects, except during meals 14.7 y 30 participants: male, 11; VFR: 13, wearing retainers female, 19 full time for first 3 days Mean ages, 19.6 y for VFR (except during meals) and and 18.8 y for HR nightly thereafter HR: 17 (except 2 had maxillary HR only), wearing retainers full time except during meals 45 participants: male, 16; VFR: 25, wearing full time, female, 29 except during meals and Mean ages, 13.6 y for VFR brushing and 15.2 y for HR HR: 20, wearing nightly, but combined with mandibular fixed lingual retainers
Cost-effectiveness and patient satisfaction between the retainer groups Survival times of the 2 types of retainers
Occlusal contacts of the maxillary and mandibular teeth with the 2 types of retainers
Changes in overbite, overjet, intercanine and intermolar widths, Little's index, and calculus index scores for both groups in the mandible
CCT, Controlled clinical trial.
consecutive patients were assigned alternatively to 1 of 2 retainer groups at the orthodontic clinic at the University of the Pacific School of Dentistry, San Francisco, Calif. Impression material was used to record the occlusal contacts of each patient at debonding, at retainer delivery, and after a 3-month retention. The anteroposterior contacts were also assessed in both groups. The numbers of total contacts and posterior contacts were significantly higher in the HR group than in the VFR group from retainer delivery to 3-month retention, and from debonding to 3-month retention. The authors of 1 study compared the satisfaction of 397 participants who were randomly assigned to 2 intervention groups (HR, 196; VFR, 201), and the participants completed a total of 350 patient satisfaction questionnaires
(Hawley, 168; VFR, 182) at 6 months.9 The dropout rate was 11.8%. The main outcomes were the frequency of responses to the satisfaction-related questions in the questionnaires. A statistically significant result for the factor of embarrassment was found more often in the HR group than in the VFR group (P 5 0.005), particularly in terms of speech and esthetics. More patients felt better in the VFR group than in HR group when the retainers were compared with fixed appliances (P \0.001). However, there was no statistically significant difference in the amount of time a retainer was worn away from home. The authors of 1 study investigated the costeffectiveness of 3 groups, including the National Health Service, orthodontic practices, and patients in the United Kingdom.9 The researchers further investigated
American Journal of Orthodontics and Dentofacial Orthopedics
June 2014 Vol 145 Issue 6
Mai et al
724
Table III. Assessment of risk of bias Study Rowland et al,5 2007 Barlin et al,7 2011 Demir et al,6 2012 Hichens et al,9 2007 Sun et al,10 2011 Sauget et al,12 1997 Xu et al,22 2011
Design RCT RCT CCT RCT RCT CCT RCT
Sample size calculation Yes No Yes Yes Yes No No
Random sequence generation Yes Yes Unclear Yes Yes No Unclear
Allocation concealment Yes No Unclear Yes Yes No Unclear
Blinding of measurement Yes Unclear Yes No Yes Yes Unclear
Withdrawals reported Yes Yes Yes Yes Yes – Yes
ITT Yes No Unclear Yes Yes – No
Risk of bias Low Moderate Moderate Low Low Moderate High
Yes, Low risk of bias; no, high risk of bias; unclear, unclear or unknown risk of bias. ITT, Intention to treat; CCT, controlled clinical trial.
Table IV. Effectiveness of HRs and VFRs Study Rowland et al,5 2007
Demir et al,6 2012 Barlin et al,7 2011
Intercanine Intermolar Arch width width length Little's index Conclusions NS NS – Effective (P 5 0.013) VFRs were more effective than HRs at holding NS NS – Effective (P \0.01) corrections of the maxillary and mandibular labial segments Maxillary NS NS NS NS No statistically significant difference was found Mandibular NS NS NS NS in the effectiveness between these 2 retainers Maxillary NS NS NS NS There was no statistically or clinically significant Mandibular NS NS NS NS difference in the measured arch width, arch length, or modified Little's index over a 12-mo period Arch Maxillary Mandibular
Effective, VFRs were more effective than HRs. NS, No statistically significant difference between HRs and VFRs.
62 subjects (HR, 41; VFR, 21) who attended extra appointments for problems with their retainers with regard to their travel costs, child-care costs, patient fees, and lost income. The HR patients had statistically significantly greater costs than did the VFR patients, with a mean difference of V2.15 (95% CI, –V2.90-V7.57). Greater costs were demonstrated with respect to the National Health Service and the orthodontic practices for the HR groups than the VFR groups, with a statistically significant result and with mean differences of V31.35 (95% CI, V28.06-V34.68) and V32.60 (95% CI, V30.58-V34.67), respectively. Therefore, the cost-effectiveness analysis of the retainers showed that VFRs were more cost-effective than HRs from 3 perspectives over a 6-month retention period. A study conducted by Sun et al10 investigated the survival time of HRs and VFRs over a 1-year follow-up of 120 adolescent patients who were randomly assigned to 1 of 2 intervention groups. The survival times of the maxillary and mandibular retainers were assessed separately. There was no statistically significant difference between the 2 groups with either the maxillary retainers (P 5 0.254) or the mandibular retainers (P 5 0.188). Another study by Hichens et al9 also investigated the rates of retainer breakage and loss in the 2 retainer
June 2014 Vol 145 Issue 6
groups after a 6-month retention period. Interestingly, more subjects broke their retainers in the HR group than in the VFR group, with a statistically significant difference (P\0.001). However, no difference was found in the loss rates of the retainers between the groups. Unfortunately, we were unable to retrieve the original data regarding the breakage and loss rates from a study after attempting to contact the authors.10 Therefore, we did not perform a data analysis of these 2 trials. DISCUSSION
In this systematic review, 5 RCTs and 2 controlled clinical trials satisfied our inclusion criteria and were analyzed after searching and assessing the quality and the data extraction methods. Most studies did not use adequate randomization and allocation concealment. Although an intention-to-treat analysis is often recommended as the least biased way to estimate the intervention effects in randomized trials,23 only 3 studies reported usage of an intention-to-treat analysis for the incomplete outcome data.5,9,10 A CONSORT flow diagram is suggested as an appropriate way to improve the quality of data reported from parallel-group randomized trials,24
American Journal of Orthodontics and Dentofacial Orthopedics
Mai et al
but just 3 studies used this method.5,9,10 Therefore, more prospective research in this area should be conducted and reported in accordance with the CONSORT guidelines, so that more high-quality RCTs will be eligible for future meta-analyses. Unfortunately, no studies were eligible for a metaanalysis in this systematic review. The effectiveness of HRs and VFRs was evaluated in 3 studies.5-7 All reported no statistically significant differences with respect to changes in the intercanine and intermolar widths during the follow-up periods. The comparison of occlusal contacts between HRs and VFRs was assessed in 1 study after a 3-month retention period.12 Sauget et al12 found that the number of occlusal contacts with HRs was significantly more than with VFRs, especially the posterior contacts. These results were supported in a study by Hoybjerg et al.25 With respect to HRs, the authors of several studies also found a statistically significant increase in the number of posterior contacts after a retention period with HRs.11,26,27 These findings showed that HRs allowed relative vertical movement of the posterior teeth during the retention stage. Therefore, HRs might be a better choice for patients who did not have good posterior settling after active orthodontic treatment. With respect to VFRs, a recent study by Dincer and Isik Aslan28 reported that changes in the occlusal contacts were not observed at the end of a 9-month retention with VFRs, whereas there was a significant increase after a 2.5-year follow-up. These results could be interpreted with the following possibilities. Because of the construction of VFRs, which cover the occlusal surfaces of the teeth, no posterior contacts increased significantly during retention. By removing the retainers, the continued mobility of the teeth might contribute to continual vertical settling in a long-term follow-up. In addition, the authors also concluded that both ideal and nonideal posterior contacts increased in the long term, and the numbers of nonideal contacts were greater than those of ideal contacts.28 Consequently, the results of the included study should be interpreted with caution because of a short retention time—only 3 months—and no long-term follow-up.12 The authors of 1 study investigated patient satisfaction and found that wearing a VFR causes less embarrassment than wearing an HR (P 5 0.005).9 A similar result was demonstrated in a study by Pratt et al,8 who found that patients were more compliant with VFRs than with HRs after the fixed appliances were debonded. A possible explanation was that VFRs are made of transparent plastic and are more esthetically pleasing and lighter than HRs, which have a stainless steel wire across the labial surface of the anterior teeth. Also, speech articulation might be another aspect with respect to patient satisfaction.
725
Haydar et al29 found that patients experienced articulation problems while wearing HRs during the first week. However, to date, no RCT has evaluated and compared speech articulation differences between VFRs and HRs. Therefore, a speech articulation analysis between the 2 retainers should be considered in future studies. In addition, Hichens et al9 conducted an economic evaluation and found that VFRs were more costeffective than HRs during retention. However, because of variability across study settings, this study, which was conducted in the United Kingdom, cannot be representative of costs in other countries. More RCTs concerning the economic evaluations of retainers in different regions of the world are needed to confirm this finding. When comparing the survival times between the 2 retainers, the thickness and materials of the VFRs and the compliance of patients should be carefully considered. First, although the study by Sun et al10 used 0.75-mm VFRs, until now, no uniform standards have been established for the thickness of VFRs. In some other studies, VFRs of 0.75, 1, and 1.5 mm were used.5,28,30 Second, the different thermoplastic materials should be considered to have an effect on VFR wear resistance and durability. Gardner et al31 found that TR material, a hard polyethylene terephthalate glycol copolymer, demonstrated greater wear resistance than the other 2 materials, which were softer, polypropylene-based thermoplastics. Furthermore, although no study indicated a correlation between retainer damage and patient compliance, we found increased compliance and increased wear with VFRs in our clinical practice. However, it remains unanswered whether increased compliance could decrease the VFR survival time. Further studies concerning these 3 points are necessary. Two systematic reviews were identified in our review.20,21 The article by Ge et al21 focused only on the outcomes pertaining to the irregularity of incisors and on intercanine and intermolar widths to compare the effectiveness of HRs and VFRs, whereas our review evaluated the outcomes including not only clinical effectiveness, but also occlusal contacts, cost-effectiveness, patient satisfaction, and survival time. The findings regarding the effectiveness of retainers were similar in our opinion21; however, the outcomes pertaining to irregularity should be interpreted with caution because of the poor quality of the included studies. Because of insufficient evidence, we cannot suggest that VFRs are more effective than HRs; furthermore, high-quality RCTs are necessary. Another review was partly relevant to our review concerning occlusal contacts between HRs and VFRs.20 The findings were consistent in our opinion because HRs were found to have more occlusal contacts than VFRs, but these findings also should be interpreted
American Journal of Orthodontics and Dentofacial Orthopedics
June 2014 Vol 145 Issue 6
Mai et al
726
with caution because of the poor quality of the included studies. Although evidence regarding these 2 retainers is currently insufficient, it does not indicate that this evidence should not be used in clinical practice. Our review comprehensively and objectively evaluated the differences between HRs and VFRs. Through this review, we believe it is both easy and useful for orthodontists and orthodontic patients to understand these 2 retainers; however, we also think that an inadequate number of studies has been conducted on these 2 retainers and thus believe that this necessitates further research. CONCLUSIONS
Some evidence suggested that there are no differences with respect to changes in intercanine and intermolar widths between HRs and VFRs after orthodontic retention. However, at present, there is insufficient evidence that VFRs are more effective than HRs. In terms of occlusal contacts, cost-effectiveness, patient satisfaction, and survival time, there is also a lack of sufficient evidence of differences between these 2 retainers. In addition, no evidence concerning periodontal health was retrieved from the studies. Therefore, this systematic review suggests that further high-quality RCTs regarding the differences between HRs and VFRs during orthodontic retention are necessary to determine which retainer is the better selection for orthodontists. REFERENCES 1. Melrose C, Millett DT. Toward a perspective on orthodontic retention? Am J Orthod Dentofacial Orthop 1998;113:507-14. 2. Hawley CA. A removable retainer. Int J Orthod 1919;5:291-305. 3. Ponitz RJ. Invisible retainers. Am J Orthod 1971;59:266-72. 4. Singh P, Grammati S, Kirschen R. Orthodontic retention patterns in the United Kingdom. J Orthod 2009;36:115-21. 5. Rowland H, Hichens L, Williams A, Hills D, Killingback N, Ewings P, et al. The effectiveness of Hawley and vacuum-formed retainers: a single-center randomized controlled trial. Am J Orthod Dentofacial Orthop 2007;132:730-7. 6. Demir A, Babacan H, Nalcaci R, Topcuoglu T. Comparison of retention characteristics of Essix and Hawley retainers. Korean J Orthod 2012;42:255-62. 7. Barlin S, Smith R, Reed R, Sandy J, Ireland AJ. A retrospective randomized double-blind comparison study of the effectiveness of Hawley vs vacuum-formed retainers. Angle Orthod 2011;81:404-9. 8. Pratt MC, Kluemper GT, Lindstrom AF. Patient compliance with orthodontic retainers in the postretention phase. Am J Orthod Dentofacial Orthop 2011;140:196-201. 9. Hichens L, Rowland H, Williams A, Hollinghurst S, Ewings P, Clark S, et al. Cost-effectiveness and patient satisfaction: Hawley and vacuum-formed retainers. Eur J Orthod 2007;29:372-8. 10. Sun J, Yu YC, Liu MY, Chen L, Li HW, Zhang L, et al. Survival time comparison between Hawley and clear overlay retainers: a randomized trial. J Dent Res 2011;90:1197-201.
June 2014 Vol 145 Issue 6
11. Basciftci FA, Uysal T, Sari Z, Inan O. Occlusal contacts with different retention procedures in 1-year follow-up period. Am J Orthod Dentofacial Orthop 2007;131:357-62. 12. Sauget E, Covell DA Jr, Boero RP, Lieber WS. Comparison of occlusal contacts with use of Hawley and clear overlay retainers. Angle Orthod 1997;67:223-30. 13. Little RM. The irregularity index: a quantitative score of mandibular anterior alignment. Am J Orthod 1975;68:554-63. 14. Higgins JPT GSe. Cochrane handbook for systematic reviews of interventions (version 5.1.0, updated March 2011). The Cochrane Collaboration 2011. Available at: www.cochrane-handbook.org. Accessed February 26, 2013. 15. Higgins JP, Thompson SG. Quantifying heterogeneity in a metaanalysis. Stat Med 2002;21:1539-58. 16. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557-60. 17. Mollov ND, Lindauer SJ, Best AM, Shroff B, Tufekci E. Patient attitudes toward retention and perceptions of treatment success. Angle Orthod 2010;80:468-73. 18. Valiathan M, Hughes E. Results of a survey-based study to identify common retention practices in the United States. Am J Orthod Dentofacial Orthop 2010;137:170-7. 19. Pratt MC, Kluemper GT, Hartsfield JK Jr, Fardo D, Nash DA. Evaluation of retention protocols among members of the American Association of Orthodontists in the United States. Am J Orthod Dentofacial Orthop 2011;140:520-6. 20. Littlewood SJ, Millett DT, Doubleday B, Bearn DR, Worthington HV. Retention procedures for stabilising tooth position after treatment with orthodontic braces. Cochrane Database Syst Rev 2006;CD002283. 21. Ge ZL, Jiao X, Tian JH, Yang KH. Clinical effectiveness of vacuumformed versus Hawley retainers: a systematic review. Chin J Evid Based Med 2012;12:596-601. 22. Xu XC, Li RM, Tang GH. Clinical evaluation of lingual fixed retainer combined with Hawley retainer and vacuum-formed retainer. Shanghai Kou Qiang Yi Xue 2011;20:623-6. 23. Newell DJ. Intention-to-treat analysis: implications for quantitative and qualitative research. Int J Epidemiol 1992;21:837-41. 24. Moher D, Schulz KF, Altman D. The CONSORT statement: revised recommendations for improving the quality of reports of parallelgroup randomized trials. JAMA 2001;285:1987-91. 25. Hoybjerg AJ, Currier GF, Kadioglu O. Evaluation of 3 retention protocols using the American Board of Orthodontics cast and radiograph evaluation. Am J Orthod Dentofacial Orthop 2013; 144:16-22. 26. Haydar B, Ciger S, Saatci P. Occlusal contact changes after the active phase of orthodontic treatment. Am J Orthod Dentofacial Orthop 1992;102:22-8. 27. Sari Z, Uysal T, Basciftci FA, Inan O. Occlusal contact changes with removable and bonded retainers in a 1-year retention period. Angle Orthod 2009;79:867-72. 28. Dinc¸er M, Isik Aslan B. Effects of thermoplastic retainers on occlusal contacts. Eur J Orthod 2010;32:6-10. 29. Haydar B, Karabulut G, Ozkan S, Aksoy AU, Ciger S. Effects of retainers on the articulation of speech. Am J Orthod Dentofacial Orthop 1996;110:535-40. 30. Pascual AL, Beeman CS, Hicks EP, Bush HM, Mitchell RJ. The essential work of fracture of thermoplastic orthodontic retainer materials. Angle Orthod 2010;80:554-61. 31. Gardner GD, Dunn WJ, Taloumis L. Wear comparison of thermoplastic materials used for orthodontic retainers. Am J Orthod Dentofacial Orthop 2003;124:294-7.
American Journal of Orthodontics and Dentofacial Orthopedics
Mai et al
APPENDIX I: SEARCH STRATEGIES
MEDLINE (via PubMed) search strategy 1. 2. 3. 4. 5. 6. 7.
exp ORTHODONTICS/ orthodontic*.mp. or/1-2 (retention or retain*).mp. (Vacuum formed or Vacuum*).mp. (thermoplastic* or invisible* or clear*).mp. Essix*.mp.
727
8. Hawley*.mp. 9. 5 or 6 or 7 10. 8 and 9 11. 4 and 10 12. 3 and 11 *indicates truncation. .mp. indicates a search of title, original title, abstract, name of substance word, and subject heading word.
APPENDIX II: PRISMA FLOW CHART OF THIS SYSTEMATIC REVIEW
American Journal of Orthodontics and Dentofacial Orthopedics
June 2014 Vol 145 Issue 6
