TY - JOUR
AU - Roslan, Saub,
AB - Summary Objectives The aim of this study was to compare vacuum-formed thermoplastic retainers (VFRs) constructed on stone models (VFR-CV) and those constructed on three-dimensional (3D) printed models (VFR-3D) based on patients’ perspective and post-treatment stability. Study design The research was designed as a crossover, randomized control trial. Materials and methods Subjects comprised patients receiving fixed appliances at a teaching institution and indicated for VFRs. Post-treatment stone models were scanned with a structured-light scanner. A fused deposition modelling machine was used to construct acrylonitrile-butadiene-styrene (ABS)-based replicas from the 3D scanned images. VFRs were fabricated on the original stone and printed models. Analysis comprised independent t-tests and repeated measures analysis of variance. Randomization Subjects were allocated to two groups using Latin squares methods and simple randomization. A week after debond, subjects received either VFR-CV first (group A) or VFR-3D first (group B) for 3 months, then the interventions were crossed over for another 3 months. Blinding In this single-blinded study, subjects were assigned a blinding code for data entry; data were analysed by a third party. Outcome measures The primary outcome measured was oral health-related quality of life (OHRQoL) based on Oral Health Impact Profile-14 (OHIP-14). Secondary outcome was post-treatment stability measured using Little’s Irregularity Index (LII). Results A total of 30 subjects (15 in each group) were recruited but 3 dropped out. Analysis included 13 subjects from group A and 14 subjects from group B. Group A showed an increase in LII (P < 0.05) after wearing VFR-CV and VFR-3D, whereas group B had no significant increase in LII after wearing both VFRs. Both groups reported significant improvement in OHRQoL after the first intervention but no significant differences after the second intervention. LII changes and OHIP-14 scores at T2 and T3 between groups, and overall between the retainers were not significantly different. No harm was reported during the study. Conclusion VFRs made on ABS-based 3D printed models showed no differences in terms of patients’ OHRQoL and stability compared with conventionally made retainers. Registration NCT02866617 (ClinicalTrials.gov). Introduction Rapid prototyping (RP) is a method for reconstructing a three-dimensional (3D) structure based on its digital data. It is a highly versatile additive fabrication process involving repeated deposition ‘layer by layer’ using a print head-deposited liquid until a physical representation of the object is obtained (1). Several RP technologies exist with the same basic principle, differing mainly in the material and in the production method (2). RP has been used in a variety of medical applications including surgical planning, implant and tissue designing, construction of orthodontic study models, research and education, and training tool (3–6). Linear measurement studies have found varying results for the accuracy of reconstructed study models (4–6). Reconstructed study models have shown reduced surface details, causing difficulty in identification of landmarks, which in turn affects the accuracy of measurement on the reconstructed models (6). In terms of appliance construction, manufactured aligners such as Invisalign and ClearSmile construct a series of Essix-typed appliances to gradually align the teeth. These appliances have been made from models that were reconstructed from a series models with digitally aligned dentition. With increasing commercial accessibility, clinicians will have the option to add 3D printers to their practice armamentarium. However, evidence enabling reliance on this technology for current practice is still limited. The aim of this study was to evaluate the clinical acceptability of 3D printed working models made from commercially available 3D printers as an alternative to the stone models that are commonly used for fabrication of orthodontic appliances. Vacuum-formed thermoplastic retainers (VFRs) were chosen as the type of retainer needs to adapt well to all clinically relevant anatomical structures of the working model to be effective. Thus, a good fitting appliance would need a working model that replicates the patient’s dentition as accurately as possible. Our evaluation was based on the effect of the appliance both on patients’ perspective changes and of clinical importance. The objectives were to compare the oral health-related quality of life (OHRQoL) and stability in patients wearing VFRs constructed on conventional orthodontic stone models and reconstructed 3D working models at 6-month follow-up. Materials and methods Trial design This was a single-centre, single-blinded, crossover randomized controlled clinical study conducted at a teaching institution. Participants Eligible subjects were adults aged 18 and over who met the eligibility criteria of having fixed appliance treatment in both arches and indicated for VFR in the retention phase as part of their original treatment plan. Exclusion criteria were single-arch or sectional fixed appliances; hypodontia requiring tooth replacement on the retainer as a temporary measure; previous treatment with rapid maxillary expansion; indicated for double retention regime (such as VFRs fitted over fixed retainers); premature debond from the original fixed appliances course; intention to relocate within the study period; learning difficulties and inability to read written instructions/ questionnaire in English or Malay; cleft lip and/or palate; or orthognatic cases. The study took place at a teaching institution between December 2015 and October 2016. Interventions Fixed appliances were debonded by one operator (NHMT) as per university clinic protocol. Briefly, the fixed appliances were removed using bracket- and band-removing pliers. Any residual composite was removed using a tungsten carbide bur on a high-speed handpiece. The dentition was scaled using an ultrasonic scaler and polished using prophylaxis paste and a rubber cup bur on a slow handpiece. For each patient, three sets of alginate impressions (Hydrogum 5; Zhermack, Italy) were taken. Two sets of impressions were sent to the institution’s orthodontic laboratory for construction of a set of post-treatment study models for record keeping and a pair of VFR by the technician. This first non-interventional VFR (VFR-1) was issued on the same day of debond. Because of the limitation of our facility to construct the VFR-3D within a day, the interventional VFRs were standardized to be issued a week after debond. The VFR was constructed from an Erkodur blank (ERKODENT® Erich Kopp GmbH, Pfalzgrafenweiler, Germany) 1.0 mm in thickness using the manufacturer’s instructions. Each VFR was constructed by the same operator using the ERKOPRESS ES-200E (ERKODENT® Erich Kopp GmbH) machine at the orthodontic laboratory using the standard instructions provided by the manufacturer. Once the foil of the VFR had been thermoformed over the model and had sufficiently cooled, the retainer material was removed from the model. The excess material around the base of model was trimmed with scissors. The foil was then trimmed for 1−2 mm of facial gingival coverage, palatal/lingual coverage, and second molar occlusal coverage and finished with a laboratory handpiece and a variety of burs. Before insertion, the retainer was examined to ensure that the edges were smooth. Clinically, the VFRs were evaluated by the same operator intra-orally and adjusted as necessary. Subjects were instructed to wear their retainer full time, only to remove it when eating, brushing, and cleaning the teeth and appliance. The third set of impressions was used for the interventional study. It was cast within 24 hours in white stone (Elite Ortho; Zhermack, Italy). Because the institution did not have an intra-oral scanner, this stone model was scanned using a structured-light scanner (Maestro 3D; Age Solutions, Inc., Italy) to produce a 3D digital model in a binary stereolithographic (.stl) format via the Easy Dental Scan software (Age Solutions, Inc.). The digital models (.stl files) were edited through proprietary software (BioModroid; Centre of Biomedical and Technology Integration, Malaysia), which reduced the size of the study models to include only the required working area to minimize the production cost of 3D printed models. A fused deposition modelling (FDM) machine (UP Plus 2 3D Printer; 3D Printing Systems, Australia) was used to construct a replica of the original stone model using acrylonitrile-butadiene-styrene (ABS) plastic material with 0.15 mm layer thickness according to the manufacturer’s instructions. The interventional VFRs were constructed on these original stone models and on replicated 3D reconstructed models. The first interventional VFR was fitted within 1 week after debond (T1) and the second interventional VFR was fitted 3 months (T2) after debond. The duration of retainer wear was standardized: the patients were instructed to wear the VFRs 24 hours a day for the first week and remove them only for eating and brushing their teeth. After the first week, which is the start of the intervention, the retainer-wearing duration was reduced to 10 hours a day (7). All patients were provided with standardized oral and written instructions on how to use the retainers. Outcome measures The primary outcome measured was the OHRQoL of subjects because, since the removable appliance relies on patient for success, their perceptions were considered an important aspect to be investigated. Patients’ perspectives were measured using the Oral Health Impact Profile (OHIP), which measures problems related to oral health in terms of functional limitation, physical pain, psychological discomfort, physical disability, psychological disability, and handicap (8), as a result of wearing the appliance. The secondary outcome was stability. Retainers are meant to hold the dentition after active treatment, thus a good retainer would prevent relapse. Our study used the modified short version of the OHIP-14 for Malaysian adults (9). Responses to each item were scored on a five-point Likert scale measuring from never (score 0) to very often (score 4). The self-administered questionnaire was pre-tested and checked for face validity. After pre-testing, some of the OHIP items were replaced from the long version OHIP (10), which were more relevant for patients wearing removable retainers. The OHIP-14 items used in this study were as follows: ‘difficulties in speaking/pronunciation’ and ‘had bad breath caused by the retainer’ measuring functional limitation, ‘felt pain’ and ‘had ulcer in the mouth’ measuring physical pain, ‘felt discomfort’ and ‘felt shy’ measuring psychological discomfort, ‘had people misunderstand some of their words’ and ‘avoided smiling’ measuring physical disability, ‘had sleep disturbed’ and ‘concentration disturbed’ measuring psychological disability, ‘avoided going out’ and ‘had problems in carrying out daily activities’ measuring social disability, and ‘felt less confident with oneself’ and ‘had problems in carrying out work to the best of ability’ measuring handicap. Stability was measured using Little’s Irregularity Index (LII), which is an index used to measure the crowding or irregularities of mandibular anterior teeth (11). LII was measured using a manual vernier calliper, accurate to 0.5 mm, between each contact point from the mandibular canine-to-canine teeth to give a total of five measurements. Mesiodistal spacing was disregarded if the teeth were in the proper arch form, but when spacing was present with displacement or rotations, only the labiolingual displacement from proper arch form was recorded (11). Ten pairs of study models with mild crowding selected from the institution’s records were used for calibration. Two operators (NHMT and WNWH) measured these models twice, with a 2-week interval. Both operators’ reliabilities for measuring the LII were excellent with intra-class correlation coefficient at above 0.75 (12). The intra-examiner reliabilities were 0.893 (NHMT) and 0.965 (WNWH). The inter-examiner reliability was 0.963. In the study, the mandibular canine-to-canine teeth were directly measured intra-orally (11) by the calibrated operators (NHMT or WNWH) at all intervals: 1 week (T1), 3 months (T2), and 6 months (T3) after debond. An irregularity index score of less than 3.5 mm was considered minimal and clinically acceptable (13). In addition, compliance was also checked as it may confound the outcomes of the study. Compliance was measured using a global rating scale that asked if they wore the VFR 10 hours per day as instructed. If the patient admitted to wear the retainers more than 10 hours daily, the patient was considered to be compliant. OHRQoL, stability, and compliance were measured at T1, T2, and T3. Sociodemographic information was also recorded at T1. Sample size Sample size was calculated using PS software (version 3.0.17). To the best of our knowledge, no past study has examined the changes in OHIP during retention. Thus, clinically meaningful change in OHIP was not done for sample estimation. Sample size was estimated based on Student’s t-test for independent model using data from a past study on changes in LII (14). For this study, a minimum of 22 subjects per group were required to detect a clinically meaningful difference of 1.0 mm (SD = 1.0) between the groups based on an α significance level of 0.05 and a β significance level of 0.1 to achieve 90 per cent power. This sample size will also achieve a moderate Pearson product-moment correlation between the two groups. Because this is a crossover study, the same subject will become their own control, thus requiring half the sample size. In this clinical trial, 30 subjects were enrolled to compensate for possible dropouts. Randomization Potential subjects were recruited among patients undergoing orthodontic treatment in 2015 who were expected to have their fixed appliances removed and who matched the eligibility criteria. Subjects were invited to enrol by their treating clinicians if they met the required criteria. Subject group allocation for the two interventions was done using a repeated set of five 2 × 2 Latin squares method (15). In this method, each set will either start with group A (VFR from conventional study model) or group B (VFR from 3D reconstructed study model), followed by a crossover in the second order. The sequence of the first order was randomly allocated Supplementary Figure 1. Then, 30 subject numbers were arranged in ascending order for the five pairs of Latin squares. Thus, both interventions were balanced to have equal number of subjects. The patient enrolment order had a predetermined randomized subjects’ number generated using an online random number generator (www.random.org). Figure 1 shows the CONSORT flow chart. When subjects met the inclusion criteria and agreed to participate, they were given a predetermined subject number according to their enrolment order. The operator was concealed from the group allocation at enrolment. The allocation was revealed after referring to Supplementary Figure 1 to determine their intervention group. Initial data were collected at T1. After the first intervention, the first follow-up data were collected at T2. Subjects were then assigned to the second intervention. Then the final follow-up data were collected at T3. Figure 1. View largeDownload slide CONSORT flow chart. Figure 1. View largeDownload slide CONSORT flow chart. Blinding This study was a single-blinded study where each patient was assigned a blinding code, which was used for data entry and data analysis by a clerk who was not involved in the treatment. The double-blinded approach was not possible in this study due to some limitations: a single operator had to be responsible for the appliance construction due to limited numbers of dental technicians and there was insufficient funds to hire one. Owing to the subtle differences in the physical appearance between the two VFRs, it was not possible to blind the operator. Patients could not be blinded totally as they may have been able to see that the retainers were slightly different in terms of the surface finish of the appliances, although they were not informed which VFR was the control and which VFR was the intervention. Statistical analysis SPSS (version 20.0) and G*Power (version 3.1) were used for statistical analysis. Severity of OHRQoL was analysed by an additive scoring method by summing the response codes for the 14 OHIP items. The score could range from 0 to 56 with higher values indicating more severe impact. A high score indicated poorer OHRQoL whereas a low score indicated a good OHRQoL. There were no missing data because all subjects answered all items. Data were descriptively analysed using mean and standard deviation for continuous variables and percentages for categorical variables. The Fisher’s exact test was used to assess compliance. Repeated measures analysis of variance was conducted to determine changes in LII and OHIP-14 from the intervention at three time intervals where the participants were same in each group (16). If significant differences were detected, post hoc pairwise comparisons of the interventions with the preceding VFRs were then performed using the Bonferroni test. Independent t-test was used to assess differences in LII and OHIP-14 scores between VFR-CV and VFR-3D measured at T2, T3, and combination scores of different time intervals and groups. Multivariate linear regression analysis was conducted to assess if the type of retainers affected the outcome (i.e. OHIP-14 score and LII) after controlling for the potential confounders (i.e. compliance, time, and gender). For this analysis, the time included was at T2 and T3, when the subjects were undergoing the trial period. Gender was also included because OHIP scores were significantly different in the general Malaysian population (17). The level of significance was set at 0.05. Ethics and permission The study was approved by the medical ethics committee, Faculty of Dentistry, University of Malaya [DF CD1414/0091 (P)]. Written consent was obtained from participants before conduct of the study. Trial registration The trial was registered with ClinicalTrials.gov (NCT02866617). Results Participant flow The trial profile (Figure 1) shows the number of subjects randomly allocated to each group, allocation and crossover stages, and the number of dropouts and their reasons for dropping out. Recruitment Patients were recruited beginning in October 2015. Treatment commenced in December 2015 and was completed in October 2016. Among the 35 patients screened, 30 eligible subjects participated in the study. Baseline data The characteristics of the two groups are shown in Table 1. The baseline gender, ethnicity, level of education, compliance, age, LLI, and OHIP-14 scores between group A and group B were not statistically significant different (P < 0.05). Table 1. Sample characteristics Characteristics Distribution Total (N = 27) Group A (N = 13) Group B (N = 14) P-value N (%) N (%) N (%) Gender  Male 6 (22.2) 5 (38.5) 1 (7.1) 0.077∞  Female 21 (77.8) 8 (61.5) 13 (92.9) Ethnicity  Malay 12 (44.4) 4 (30.8) 8 (57.1) 0.605ф  Chinese 13 (48.1) 8 (61.5) 5 (35.7)  Indian 2 (7.4) 1 (7.7) 1 (7.1) Highest education level  Primary 1 (3.7) 0 (0) 1 (7.1) 0.808ф  Secondary 5 (18.5) 3 (23.1) 2 (14.3)  University 16 (59.3) 7 (53.8) 9 (64.3)  Others 5 (18.5) 3 (23.1) 2 (14.3) Compliance (T1)  Compliant 23 (82.2) 10 (76.9) 13 (92.9) 0.326∞  Non-compliant 4 (14.8) 3 (23.1) 1 (7.1) Mean (SD) Mean (SD) Mean (SD) Age 24.33 (5.79) 24.08 (6.87) 24.57(4.82) 0.829λ Little’s Irregularity Index (T1) 0.54 mm (0.44) 0.54 mm (0.48) 0.54 mm (0.41) 0.987λ OHIP-14 (T1) 14.9 (9.5) 14.6 (7.4) 15.2 (11.4) 0.872λ Characteristics Distribution Total (N = 27) Group A (N = 13) Group B (N = 14) P-value N (%) N (%) N (%) Gender  Male 6 (22.2) 5 (38.5) 1 (7.1) 0.077∞  Female 21 (77.8) 8 (61.5) 13 (92.9) Ethnicity  Malay 12 (44.4) 4 (30.8) 8 (57.1) 0.605ф  Chinese 13 (48.1) 8 (61.5) 5 (35.7)  Indian 2 (7.4) 1 (7.7) 1 (7.1) Highest education level  Primary 1 (3.7) 0 (0) 1 (7.1) 0.808ф  Secondary 5 (18.5) 3 (23.1) 2 (14.3)  University 16 (59.3) 7 (53.8) 9 (64.3)  Others 5 (18.5) 3 (23.1) 2 (14.3) Compliance (T1)  Compliant 23 (82.2) 10 (76.9) 13 (92.9) 0.326∞  Non-compliant 4 (14.8) 3 (23.1) 1 (7.1) Mean (SD) Mean (SD) Mean (SD) Age 24.33 (5.79) 24.08 (6.87) 24.57(4.82) 0.829λ Little’s Irregularity Index (T1) 0.54 mm (0.44) 0.54 mm (0.48) 0.54 mm (0.41) 0.987λ OHIP-14 (T1) 14.9 (9.5) 14.6 (7.4) 15.2 (11.4) 0.872λ ∞Fisher exact, фchi-square, and λindependent t-test; *P < 0.05. View Large Table 1. Sample characteristics Characteristics Distribution Total (N = 27) Group A (N = 13) Group B (N = 14) P-value N (%) N (%) N (%) Gender  Male 6 (22.2) 5 (38.5) 1 (7.1) 0.077∞  Female 21 (77.8) 8 (61.5) 13 (92.9) Ethnicity  Malay 12 (44.4) 4 (30.8) 8 (57.1) 0.605ф  Chinese 13 (48.1) 8 (61.5) 5 (35.7)  Indian 2 (7.4) 1 (7.7) 1 (7.1) Highest education level  Primary 1 (3.7) 0 (0) 1 (7.1) 0.808ф  Secondary 5 (18.5) 3 (23.1) 2 (14.3)  University 16 (59.3) 7 (53.8) 9 (64.3)  Others 5 (18.5) 3 (23.1) 2 (14.3) Compliance (T1)  Compliant 23 (82.2) 10 (76.9) 13 (92.9) 0.326∞  Non-compliant 4 (14.8) 3 (23.1) 1 (7.1) Mean (SD) Mean (SD) Mean (SD) Age 24.33 (5.79) 24.08 (6.87) 24.57(4.82) 0.829λ Little’s Irregularity Index (T1) 0.54 mm (0.44) 0.54 mm (0.48) 0.54 mm (0.41) 0.987λ OHIP-14 (T1) 14.9 (9.5) 14.6 (7.4) 15.2 (11.4) 0.872λ Characteristics Distribution Total (N = 27) Group A (N = 13) Group B (N = 14) P-value N (%) N (%) N (%) Gender  Male 6 (22.2) 5 (38.5) 1 (7.1) 0.077∞  Female 21 (77.8) 8 (61.5) 13 (92.9) Ethnicity  Malay 12 (44.4) 4 (30.8) 8 (57.1) 0.605ф  Chinese 13 (48.1) 8 (61.5) 5 (35.7)  Indian 2 (7.4) 1 (7.7) 1 (7.1) Highest education level  Primary 1 (3.7) 0 (0) 1 (7.1) 0.808ф  Secondary 5 (18.5) 3 (23.1) 2 (14.3)  University 16 (59.3) 7 (53.8) 9 (64.3)  Others 5 (18.5) 3 (23.1) 2 (14.3) Compliance (T1)  Compliant 23 (82.2) 10 (76.9) 13 (92.9) 0.326∞  Non-compliant 4 (14.8) 3 (23.1) 1 (7.1) Mean (SD) Mean (SD) Mean (SD) Age 24.33 (5.79) 24.08 (6.87) 24.57(4.82) 0.829λ Little’s Irregularity Index (T1) 0.54 mm (0.44) 0.54 mm (0.48) 0.54 mm (0.41) 0.987λ OHIP-14 (T1) 14.9 (9.5) 14.6 (7.4) 15.2 (11.4) 0.872λ ∞Fisher exact, фchi-square, and λindependent t-test; *P < 0.05. View Large Numbers analysed Among the 30 recruited subjects, 2 participants in group A and 1 participant in group B failed to complete the study and were considered as dropouts. There were no dropouts before randomization. However, one subject from each group was lost to follow-up as they had relocated during the study period, whereas one subject from group A did not wear his retainer at all, which caused relapsed such that the retainer could not fit at all and a new retainer had to be constructed based on the patient’s current malocclusion. Thus, the final analysis included 13 subjects from group A and 14 subjects from group B. The dropped-out subjects were removed from all analyses. Outcome Figure 2 (Supplementary Tables S9–S11), on impact of the VFRs on subjects’ OHRQoL, shows that both groups reported improvement in their OHIP-14 scores after wearing the VFR-CV and VFR-3D [P < 0.05, 95% confidence interval (CI) excluded zero] for the first 3 months and slight worsening of the OHIP-14 scores that was not statistically significant, after wearing the alternate VFRs for 3 months. The OHIP-14 scores of subjects who wore VFR-CV and VFR-3D were not significantly different at both T2 and T3. Overall, subjects’ OHIP-14 scores between the two retainers were not significantly different. Multivariate linear regression analysis (Table 2) showed that the OHRQoL was not significantly different between VFR-CV and VFR-3D after controlling for compliance, time, and gender. Table 2. Multiple linear regression for the relationship between the tested retainers (vacuum-formed retainers fabricated from conventional stone model (VFR-CV) and those fabricated from three-dimensional (3D) printed model) and the outcome controlling for the effect of compliance, time and gender. OHIP, Oral Health Impact Profile-14 [M]; LII, Little’s Irregularity Index; VFR-CV, vacuum-formed retainer fabricated from conventional stone model; VFR-3D, VFR fabricated from 3D printed model Model Predictors Unstandardized coefficients Standardized coefficients t P-value 95% Confidence interval for B B Std. error β Lower Upper OHIP (Constant) 14.72 2.31 6.39 0.000 10.09 19.36 Retainer type −0.80 1.75 −0.06 −0.46 0.651 −4.30 2.71 Compliance −8.02 2.10 −0.47 −3.82 0.000* −12.24 −3.80 Time −2.34 2.14 −0.13 −1.10 0.279 −6.64 1.96 LII Gender 1.56 1.76 0.11 0.89 0.379 −1.98 5.10 (Constant) 1.15 0.26 4.47 0.000 0.63 1.67 Retainer type 0.15 0.20 0.10 0.79 0.433 −0.24 0.55 Compliance −0.66 0.24 −0.37 −2.82 0.007* −1.13 −0.19 Time 0.38 0.24 0.21 1.61 0.115 −0.10 0.86 Gender 0.32 0.20 0.21 1.63 0.109 −0.07 0.72 Model Predictors Unstandardized coefficients Standardized coefficients t P-value 95% Confidence interval for B B Std. error β Lower Upper OHIP (Constant) 14.72 2.31 6.39 0.000 10.09 19.36 Retainer type −0.80 1.75 −0.06 −0.46 0.651 −4.30 2.71 Compliance −8.02 2.10 −0.47 −3.82 0.000* −12.24 −3.80 Time −2.34 2.14 −0.13 −1.10 0.279 −6.64 1.96 LII Gender 1.56 1.76 0.11 0.89 0.379 −1.98 5.10 (Constant) 1.15 0.26 4.47 0.000 0.63 1.67 Retainer type 0.15 0.20 0.10 0.79 0.433 −0.24 0.55 Compliance −0.66 0.24 −0.37 −2.82 0.007* −1.13 −0.19 Time 0.38 0.24 0.21 1.61 0.115 −0.10 0.86 Gender 0.32 0.20 0.21 1.63 0.109 −0.07 0.72 Model OHIP: R = 0.54; R2 = 0.29; adjusted R2 = 0.24; F (4, 49) = 5.06; P = 0.002*. Model LII: R = 0.47; R2 = 0.22; adjusted R2 = 0.16; F (4, 49) = 3.49; P = 0.014*. *P < 0.05. View Large Table 2. Multiple linear regression for the relationship between the tested retainers (vacuum-formed retainers fabricated from conventional stone model (VFR-CV) and those fabricated from three-dimensional (3D) printed model) and the outcome controlling for the effect of compliance, time and gender. OHIP, Oral Health Impact Profile-14 [M]; LII, Little’s Irregularity Index; VFR-CV, vacuum-formed retainer fabricated from conventional stone model; VFR-3D, VFR fabricated from 3D printed model Model Predictors Unstandardized coefficients Standardized coefficients t P-value 95% Confidence interval for B B Std. error β Lower Upper OHIP (Constant) 14.72 2.31 6.39 0.000 10.09 19.36 Retainer type −0.80 1.75 −0.06 −0.46 0.651 −4.30 2.71 Compliance −8.02 2.10 −0.47 −3.82 0.000* −12.24 −3.80 Time −2.34 2.14 −0.13 −1.10 0.279 −6.64 1.96 LII Gender 1.56 1.76 0.11 0.89 0.379 −1.98 5.10 (Constant) 1.15 0.26 4.47 0.000 0.63 1.67 Retainer type 0.15 0.20 0.10 0.79 0.433 −0.24 0.55 Compliance −0.66 0.24 −0.37 −2.82 0.007* −1.13 −0.19 Time 0.38 0.24 0.21 1.61 0.115 −0.10 0.86 Gender 0.32 0.20 0.21 1.63 0.109 −0.07 0.72 Model Predictors Unstandardized coefficients Standardized coefficients t P-value 95% Confidence interval for B B Std. error β Lower Upper OHIP (Constant) 14.72 2.31 6.39 0.000 10.09 19.36 Retainer type −0.80 1.75 −0.06 −0.46 0.651 −4.30 2.71 Compliance −8.02 2.10 −0.47 −3.82 0.000* −12.24 −3.80 Time −2.34 2.14 −0.13 −1.10 0.279 −6.64 1.96 LII Gender 1.56 1.76 0.11 0.89 0.379 −1.98 5.10 (Constant) 1.15 0.26 4.47 0.000 0.63 1.67 Retainer type 0.15 0.20 0.10 0.79 0.433 −0.24 0.55 Compliance −0.66 0.24 −0.37 −2.82 0.007* −1.13 −0.19 Time 0.38 0.24 0.21 1.61 0.115 −0.10 0.86 Gender 0.32 0.20 0.21 1.63 0.109 −0.07 0.72 Model OHIP: R = 0.54; R2 = 0.29; adjusted R2 = 0.24; F (4, 49) = 5.06; P = 0.002*. Model LII: R = 0.47; R2 = 0.22; adjusted R2 = 0.16; F (4, 49) = 3.49; P = 0.014*. *P < 0.05. View Large Figure 2. View largeDownload slide Comparison between vacuum-formed retainers fabricated from conventional stone model (VFR-CV) and those fabricated from 3D printed model (VFR-3D) in terms of the effect on the impact on oral health-related quality of life. Figure 2. View largeDownload slide Comparison between vacuum-formed retainers fabricated from conventional stone model (VFR-CV) and those fabricated from 3D printed model (VFR-3D) in terms of the effect on the impact on oral health-related quality of life. Table 3 (Supplementary Tables S6 and S7) shows the severity of impact by domains between VFR-CV and VFR-3D worn by the same subjects. Group A reported significant differences in functional limitations and psychological discomfort domains. However, post hoc analysis showed that only functional limitation improved by 1.4 points after wearing VFR-CV in the first 3 months. Group B reported significant differences in functional limitations, psychological discomfort, physical disability, social disability, and handicap domains. However, post hoc analysis showed that only functional limitations and physical disability were statistically significant improved by 2.1 and 1.7 points, respectively, after wearing VFR-3D in the first 3 months. Neither group showed significant differences in OHIP scores between VFR-CV and VFR-3D as measured in changes between T2 and T3. Table 3. Severity of Oral Health Impact Profile-14 by domain between the two retainers by subjects of the same group. VFR-1, standard VFR given during transition from debond to the start of the intervention; VFR-CV, VFR fabricated from conventional stone model; VFR-3D, VFR fabricated from three-dimensional (3D) printed model; CI, confidence interval; SD, standard deviation Group A Group B Domain T1 (VFR-1) (n = 13) T2 (VFR-CV) (n = 13) T3 (VFR-3D) (n = 13) P-valueβ; η2 Post hoc (T2-T1) Post hoc (T3-T2) T1 (VFR-1) (n = 14) T2 (VFR-3D) (n = 14) T3 (VFR-CV) (n = 14) P valueβ; η2 Post hoc (T2-T1) Post hoc (T3-T2) Mean (SD) Mean (SD) Mean (SD) Mean difference (SD); P valueχ; 95% CI; Cohen’s d Mean difference (SD); P valueχ; 95% CI; Cohen’s d Mean (SD) Mean (SD) Mean (SD) Mean difference (SD); P-valueχ; 95% CI; Cohen’s d Mean difference (SD); P valueχ; 95% CI; Cohen’s d Functional limitation 3.4 (1.8) 2.0 (1.4) 2.4 (1.6) 0.001*; 0.4 −1.4 (0.4); 0.008*; −2.4, −0.4; 0.9 0.4 (0.4); 0.879; −0.6, 1.4; 0.3 3.5 (2.4) 1.4 (1.5) 1.9 (2.0) 0.001*; 0.4 −2.1 (0.5); 0.004*; −3.5, −0.7; 1.0 0.4 (0.6); 1.000; −1.1, 2.0; 0.3 Physical pain 2.5 (1.3) 1.5 (1.6) 1.8 (1.0) 0.060; 0.2 — — 2.1 (1.7) 1.0 (1.4) 1.4 (1.7) 0.063; 0.2 — Psychological discomfort 2.7 (1.8) 1.5 (1.3) 1.7 (1.4) 0.017*; 0.3 −1.2 (0.5); 0.086; −2.6, 1.5; 0.7 0.2 (0.4); 1.000; −0.9, 1.3; 0.1 2.5 (2.1) 1.4 (1.2) 1.6 (1.6) 0.041*; 0.2 −1.1 (0.5); 0.088; −2.4, 0.1; 0.6 0.2 (0.4); 1.000; −0.9, 1.3; 0.1 Physical disability 2.2 (1.6) 1.2 (1.4) 1.3 (1.3) 0.061; 0.2 — — 2.7 (2.4) 1.0 (1.0) 1.7 (1.9) 0.005*; 0.3 −1.7 (0.6); 0.027*; −3.3, −0.2; 0.8 0.7 (0.4); 0.350; −0.5, 1.9; 0.4 Psychological disability 1.5 (1.6) 0.6 (0.9) 1.2 (2.1) 0.235; 0.1 — — 1.4 (1.5) 0.9 (1.4) 1.1 (1.8) 0.514; 0.1 — — Social disability 1.3 (1.4) 0.5 (0.8) 0.8 (1.5) 0.246; 0.1 — — 1.6 (1.6) 0.5 (0.8) 0.9 (1.4) 0.048*; 0.2 −1.1 (0.4); 0.056; −2.1, 0.2; 0.8 0.4 (0.4); 1.000; −0.9, 1.6; 0.3 Handicap 1.1 (1.1) 0.8 (1.2) 1.0 (1.0) 0.613; 0 — — 1.4 (1.9) 0.2 (0.6) 0.8 (1.3) 0.034*; 0.2 −1.1 (0.5); 0.109; −2.5, 0.2; 0.7 0.6 (0.3); 0.263; −0.3, 1.4; 0.5 Group A Group B Domain T1 (VFR-1) (n = 13) T2 (VFR-CV) (n = 13) T3 (VFR-3D) (n = 13) P-valueβ; η2 Post hoc (T2-T1) Post hoc (T3-T2) T1 (VFR-1) (n = 14) T2 (VFR-3D) (n = 14) T3 (VFR-CV) (n = 14) P valueβ; η2 Post hoc (T2-T1) Post hoc (T3-T2) Mean (SD) Mean (SD) Mean (SD) Mean difference (SD); P valueχ; 95% CI; Cohen’s d Mean difference (SD); P valueχ; 95% CI; Cohen’s d Mean (SD) Mean (SD) Mean (SD) Mean difference (SD); P-valueχ; 95% CI; Cohen’s d Mean difference (SD); P valueχ; 95% CI; Cohen’s d Functional limitation 3.4 (1.8) 2.0 (1.4) 2.4 (1.6) 0.001*; 0.4 −1.4 (0.4); 0.008*; −2.4, −0.4; 0.9 0.4 (0.4); 0.879; −0.6, 1.4; 0.3 3.5 (2.4) 1.4 (1.5) 1.9 (2.0) 0.001*; 0.4 −2.1 (0.5); 0.004*; −3.5, −0.7; 1.0 0.4 (0.6); 1.000; −1.1, 2.0; 0.3 Physical pain 2.5 (1.3) 1.5 (1.6) 1.8 (1.0) 0.060; 0.2 — — 2.1 (1.7) 1.0 (1.4) 1.4 (1.7) 0.063; 0.2 — Psychological discomfort 2.7 (1.8) 1.5 (1.3) 1.7 (1.4) 0.017*; 0.3 −1.2 (0.5); 0.086; −2.6, 1.5; 0.7 0.2 (0.4); 1.000; −0.9, 1.3; 0.1 2.5 (2.1) 1.4 (1.2) 1.6 (1.6) 0.041*; 0.2 −1.1 (0.5); 0.088; −2.4, 0.1; 0.6 0.2 (0.4); 1.000; −0.9, 1.3; 0.1 Physical disability 2.2 (1.6) 1.2 (1.4) 1.3 (1.3) 0.061; 0.2 — — 2.7 (2.4) 1.0 (1.0) 1.7 (1.9) 0.005*; 0.3 −1.7 (0.6); 0.027*; −3.3, −0.2; 0.8 0.7 (0.4); 0.350; −0.5, 1.9; 0.4 Psychological disability 1.5 (1.6) 0.6 (0.9) 1.2 (2.1) 0.235; 0.1 — — 1.4 (1.5) 0.9 (1.4) 1.1 (1.8) 0.514; 0.1 — — Social disability 1.3 (1.4) 0.5 (0.8) 0.8 (1.5) 0.246; 0.1 — — 1.6 (1.6) 0.5 (0.8) 0.9 (1.4) 0.048*; 0.2 −1.1 (0.4); 0.056; −2.1, 0.2; 0.8 0.4 (0.4); 1.000; −0.9, 1.6; 0.3 Handicap 1.1 (1.1) 0.8 (1.2) 1.0 (1.0) 0.613; 0 — — 1.4 (1.9) 0.2 (0.6) 0.8 (1.3) 0.034*; 0.2 −1.1 (0.5); 0.109; −2.5, 0.2; 0.7 0.6 (0.3); 0.263; −0.3, 1.4; 0.5 βRepeated measures analysis of variance; *P < 0.05; χPost hoc comparison with Bonferroni correction showing only differences between the vacuum-formed retainer (VFR) with the preceding VFR (Bold, P < 0.05). View Large Table 3. Severity of Oral Health Impact Profile-14 by domain between the two retainers by subjects of the same group. VFR-1, standard VFR given during transition from debond to the start of the intervention; VFR-CV, VFR fabricated from conventional stone model; VFR-3D, VFR fabricated from three-dimensional (3D) printed model; CI, confidence interval; SD, standard deviation Group A Group B Domain T1 (VFR-1) (n = 13) T2 (VFR-CV) (n = 13) T3 (VFR-3D) (n = 13) P-valueβ; η2 Post hoc (T2-T1) Post hoc (T3-T2) T1 (VFR-1) (n = 14) T2 (VFR-3D) (n = 14) T3 (VFR-CV) (n = 14) P valueβ; η2 Post hoc (T2-T1) Post hoc (T3-T2) Mean (SD) Mean (SD) Mean (SD) Mean difference (SD); P valueχ; 95% CI; Cohen’s d Mean difference (SD); P valueχ; 95% CI; Cohen’s d Mean (SD) Mean (SD) Mean (SD) Mean difference (SD); P-valueχ; 95% CI; Cohen’s d Mean difference (SD); P valueχ; 95% CI; Cohen’s d Functional limitation 3.4 (1.8) 2.0 (1.4) 2.4 (1.6) 0.001*; 0.4 −1.4 (0.4); 0.008*; −2.4, −0.4; 0.9 0.4 (0.4); 0.879; −0.6, 1.4; 0.3 3.5 (2.4) 1.4 (1.5) 1.9 (2.0) 0.001*; 0.4 −2.1 (0.5); 0.004*; −3.5, −0.7; 1.0 0.4 (0.6); 1.000; −1.1, 2.0; 0.3 Physical pain 2.5 (1.3) 1.5 (1.6) 1.8 (1.0) 0.060; 0.2 — — 2.1 (1.7) 1.0 (1.4) 1.4 (1.7) 0.063; 0.2 — Psychological discomfort 2.7 (1.8) 1.5 (1.3) 1.7 (1.4) 0.017*; 0.3 −1.2 (0.5); 0.086; −2.6, 1.5; 0.7 0.2 (0.4); 1.000; −0.9, 1.3; 0.1 2.5 (2.1) 1.4 (1.2) 1.6 (1.6) 0.041*; 0.2 −1.1 (0.5); 0.088; −2.4, 0.1; 0.6 0.2 (0.4); 1.000; −0.9, 1.3; 0.1 Physical disability 2.2 (1.6) 1.2 (1.4) 1.3 (1.3) 0.061; 0.2 — — 2.7 (2.4) 1.0 (1.0) 1.7 (1.9) 0.005*; 0.3 −1.7 (0.6); 0.027*; −3.3, −0.2; 0.8 0.7 (0.4); 0.350; −0.5, 1.9; 0.4 Psychological disability 1.5 (1.6) 0.6 (0.9) 1.2 (2.1) 0.235; 0.1 — — 1.4 (1.5) 0.9 (1.4) 1.1 (1.8) 0.514; 0.1 — — Social disability 1.3 (1.4) 0.5 (0.8) 0.8 (1.5) 0.246; 0.1 — — 1.6 (1.6) 0.5 (0.8) 0.9 (1.4) 0.048*; 0.2 −1.1 (0.4); 0.056; −2.1, 0.2; 0.8 0.4 (0.4); 1.000; −0.9, 1.6; 0.3 Handicap 1.1 (1.1) 0.8 (1.2) 1.0 (1.0) 0.613; 0 — — 1.4 (1.9) 0.2 (0.6) 0.8 (1.3) 0.034*; 0.2 −1.1 (0.5); 0.109; −2.5, 0.2; 0.7 0.6 (0.3); 0.263; −0.3, 1.4; 0.5 Group A Group B Domain T1 (VFR-1) (n = 13) T2 (VFR-CV) (n = 13) T3 (VFR-3D) (n = 13) P-valueβ; η2 Post hoc (T2-T1) Post hoc (T3-T2) T1 (VFR-1) (n = 14) T2 (VFR-3D) (n = 14) T3 (VFR-CV) (n = 14) P valueβ; η2 Post hoc (T2-T1) Post hoc (T3-T2) Mean (SD) Mean (SD) Mean (SD) Mean difference (SD); P valueχ; 95% CI; Cohen’s d Mean difference (SD); P valueχ; 95% CI; Cohen’s d Mean (SD) Mean (SD) Mean (SD) Mean difference (SD); P-valueχ; 95% CI; Cohen’s d Mean difference (SD); P valueχ; 95% CI; Cohen’s d Functional limitation 3.4 (1.8) 2.0 (1.4) 2.4 (1.6) 0.001*; 0.4 −1.4 (0.4); 0.008*; −2.4, −0.4; 0.9 0.4 (0.4); 0.879; −0.6, 1.4; 0.3 3.5 (2.4) 1.4 (1.5) 1.9 (2.0) 0.001*; 0.4 −2.1 (0.5); 0.004*; −3.5, −0.7; 1.0 0.4 (0.6); 1.000; −1.1, 2.0; 0.3 Physical pain 2.5 (1.3) 1.5 (1.6) 1.8 (1.0) 0.060; 0.2 — — 2.1 (1.7) 1.0 (1.4) 1.4 (1.7) 0.063; 0.2 — Psychological discomfort 2.7 (1.8) 1.5 (1.3) 1.7 (1.4) 0.017*; 0.3 −1.2 (0.5); 0.086; −2.6, 1.5; 0.7 0.2 (0.4); 1.000; −0.9, 1.3; 0.1 2.5 (2.1) 1.4 (1.2) 1.6 (1.6) 0.041*; 0.2 −1.1 (0.5); 0.088; −2.4, 0.1; 0.6 0.2 (0.4); 1.000; −0.9, 1.3; 0.1 Physical disability 2.2 (1.6) 1.2 (1.4) 1.3 (1.3) 0.061; 0.2 — — 2.7 (2.4) 1.0 (1.0) 1.7 (1.9) 0.005*; 0.3 −1.7 (0.6); 0.027*; −3.3, −0.2; 0.8 0.7 (0.4); 0.350; −0.5, 1.9; 0.4 Psychological disability 1.5 (1.6) 0.6 (0.9) 1.2 (2.1) 0.235; 0.1 — — 1.4 (1.5) 0.9 (1.4) 1.1 (1.8) 0.514; 0.1 — — Social disability 1.3 (1.4) 0.5 (0.8) 0.8 (1.5) 0.246; 0.1 — — 1.6 (1.6) 0.5 (0.8) 0.9 (1.4) 0.048*; 0.2 −1.1 (0.4); 0.056; −2.1, 0.2; 0.8 0.4 (0.4); 1.000; −0.9, 1.6; 0.3 Handicap 1.1 (1.1) 0.8 (1.2) 1.0 (1.0) 0.613; 0 — — 1.4 (1.9) 0.2 (0.6) 0.8 (1.3) 0.034*; 0.2 −1.1 (0.5); 0.109; −2.5, 0.2; 0.7 0.6 (0.3); 0.263; −0.3, 1.4; 0.5 βRepeated measures analysis of variance; *P < 0.05; χPost hoc comparison with Bonferroni correction showing only differences between the vacuum-formed retainer (VFR) with the preceding VFR (Bold, P < 0.05). View Large Table 4, on the severity of impact by domains between VFR-CV and VFR-3D at T2 and T3, shows that the subjects did not report significant differences between the severities of the OHIP scores for all domains at both intervals. Overall, subjects’ OHRQoL by all domains between the two retainers were not statistically different. Table 4. Severity of Oral Health Impact Profile-14 between the two retainers by domain. VFR-CV, vacuum-formed retainer fabricated from conventional stone model; VFR-3D, VFR fabricated from three-dimensional printed model; SD, standard deviation T2 (3 months) T3 (6 months) Overall Domain VFR-CV (n = 13) VFR-3D (n = 14) P valueф; Cohen’s d VFR-CV (n = 14) VFR-3D (n = 13) P valueф; Cohen’s d VFR-CV (n = 27) VFR-3D (n = 27) P valueф; Cohen’s d Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Functional limitation 2.0 (1.4) 1.4 (1.5) 0.311; 0.4 1.9 (2.0) 2.4 (1.6) 0.464; 0.8 1.9 (1.7) 1.9 (1.6) 0.935; 0 Physical pain 1.5 (1.6) 1.0 (1.4) 0.347; 0.4 1.4 (1.7) 1.8 (1.0) 0.544; 0.3 1.5 (1.6) 1.4 (1.2) 0.779; 0.1 Psychological discomfort 1.5 (1.3) 1.4 (1.2) 0.829; 0.1 1.6 (1.6) 1.7 (1.4) 0.833; 0.1 1.5 (1.4) 1.5 (1.3) 1.000; 0 Physical disability 1.2 (1.4) 1.0 (1.0) 0.614; 0.2 1.7 (1.9) 1.3 (1.3) 0.533; 0.2 1.5 (1.7) 1.1 (1.1) 0.395; 0.3 Psychological disability 0.6 (0.9) 0.9 (1.4) 0.504; 0.2 1.1 (1.8) 1.2 (2.1) 0.988; 0.1 0.9 (1.4) 1.0 (1.7) 0.734; 0.1 Social disability 0.5 (0.8) 0.5 (0.8) 0.898; 0 0.9 (1.4) 0.8 (1.5) 0.985; 0.1 0.7 (1.1) 0.7 (1.2) 0.907; 0 Handicap 0.8 (1.2) 0.2 (0.6) 0.158; 0.6 0.8 (1.3) 1.0 (1.0) 0.639; 0.2 0.8 (1.3) 0.6 (0.9) 0.533; 0.2 T2 (3 months) T3 (6 months) Overall Domain VFR-CV (n = 13) VFR-3D (n = 14) P valueф; Cohen’s d VFR-CV (n = 14) VFR-3D (n = 13) P valueф; Cohen’s d VFR-CV (n = 27) VFR-3D (n = 27) P valueф; Cohen’s d Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Functional limitation 2.0 (1.4) 1.4 (1.5) 0.311; 0.4 1.9 (2.0) 2.4 (1.6) 0.464; 0.8 1.9 (1.7) 1.9 (1.6) 0.935; 0 Physical pain 1.5 (1.6) 1.0 (1.4) 0.347; 0.4 1.4 (1.7) 1.8 (1.0) 0.544; 0.3 1.5 (1.6) 1.4 (1.2) 0.779; 0.1 Psychological discomfort 1.5 (1.3) 1.4 (1.2) 0.829; 0.1 1.6 (1.6) 1.7 (1.4) 0.833; 0.1 1.5 (1.4) 1.5 (1.3) 1.000; 0 Physical disability 1.2 (1.4) 1.0 (1.0) 0.614; 0.2 1.7 (1.9) 1.3 (1.3) 0.533; 0.2 1.5 (1.7) 1.1 (1.1) 0.395; 0.3 Psychological disability 0.6 (0.9) 0.9 (1.4) 0.504; 0.2 1.1 (1.8) 1.2 (2.1) 0.988; 0.1 0.9 (1.4) 1.0 (1.7) 0.734; 0.1 Social disability 0.5 (0.8) 0.5 (0.8) 0.898; 0 0.9 (1.4) 0.8 (1.5) 0.985; 0.1 0.7 (1.1) 0.7 (1.2) 0.907; 0 Handicap 0.8 (1.2) 0.2 (0.6) 0.158; 0.6 0.8 (1.3) 1.0 (1.0) 0.639; 0.2 0.8 (1.3) 0.6 (0.9) 0.533; 0.2 фIndependent t-test; *P < 0.05. View Large Table 4. Severity of Oral Health Impact Profile-14 between the two retainers by domain. VFR-CV, vacuum-formed retainer fabricated from conventional stone model; VFR-3D, VFR fabricated from three-dimensional printed model; SD, standard deviation T2 (3 months) T3 (6 months) Overall Domain VFR-CV (n = 13) VFR-3D (n = 14) P valueф; Cohen’s d VFR-CV (n = 14) VFR-3D (n = 13) P valueф; Cohen’s d VFR-CV (n = 27) VFR-3D (n = 27) P valueф; Cohen’s d Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Functional limitation 2.0 (1.4) 1.4 (1.5) 0.311; 0.4 1.9 (2.0) 2.4 (1.6) 0.464; 0.8 1.9 (1.7) 1.9 (1.6) 0.935; 0 Physical pain 1.5 (1.6) 1.0 (1.4) 0.347; 0.4 1.4 (1.7) 1.8 (1.0) 0.544; 0.3 1.5 (1.6) 1.4 (1.2) 0.779; 0.1 Psychological discomfort 1.5 (1.3) 1.4 (1.2) 0.829; 0.1 1.6 (1.6) 1.7 (1.4) 0.833; 0.1 1.5 (1.4) 1.5 (1.3) 1.000; 0 Physical disability 1.2 (1.4) 1.0 (1.0) 0.614; 0.2 1.7 (1.9) 1.3 (1.3) 0.533; 0.2 1.5 (1.7) 1.1 (1.1) 0.395; 0.3 Psychological disability 0.6 (0.9) 0.9 (1.4) 0.504; 0.2 1.1 (1.8) 1.2 (2.1) 0.988; 0.1 0.9 (1.4) 1.0 (1.7) 0.734; 0.1 Social disability 0.5 (0.8) 0.5 (0.8) 0.898; 0 0.9 (1.4) 0.8 (1.5) 0.985; 0.1 0.7 (1.1) 0.7 (1.2) 0.907; 0 Handicap 0.8 (1.2) 0.2 (0.6) 0.158; 0.6 0.8 (1.3) 1.0 (1.0) 0.639; 0.2 0.8 (1.3) 0.6 (0.9) 0.533; 0.2 T2 (3 months) T3 (6 months) Overall Domain VFR-CV (n = 13) VFR-3D (n = 14) P valueф; Cohen’s d VFR-CV (n = 14) VFR-3D (n = 13) P valueф; Cohen’s d VFR-CV (n = 27) VFR-3D (n = 27) P valueф; Cohen’s d Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Functional limitation 2.0 (1.4) 1.4 (1.5) 0.311; 0.4 1.9 (2.0) 2.4 (1.6) 0.464; 0.8 1.9 (1.7) 1.9 (1.6) 0.935; 0 Physical pain 1.5 (1.6) 1.0 (1.4) 0.347; 0.4 1.4 (1.7) 1.8 (1.0) 0.544; 0.3 1.5 (1.6) 1.4 (1.2) 0.779; 0.1 Psychological discomfort 1.5 (1.3) 1.4 (1.2) 0.829; 0.1 1.6 (1.6) 1.7 (1.4) 0.833; 0.1 1.5 (1.4) 1.5 (1.3) 1.000; 0 Physical disability 1.2 (1.4) 1.0 (1.0) 0.614; 0.2 1.7 (1.9) 1.3 (1.3) 0.533; 0.2 1.5 (1.7) 1.1 (1.1) 0.395; 0.3 Psychological disability 0.6 (0.9) 0.9 (1.4) 0.504; 0.2 1.1 (1.8) 1.2 (2.1) 0.988; 0.1 0.9 (1.4) 1.0 (1.7) 0.734; 0.1 Social disability 0.5 (0.8) 0.5 (0.8) 0.898; 0 0.9 (1.4) 0.8 (1.5) 0.985; 0.1 0.7 (1.1) 0.7 (1.2) 0.907; 0 Handicap 0.8 (1.2) 0.2 (0.6) 0.158; 0.6 0.8 (1.3) 1.0 (1.0) 0.639; 0.2 0.8 (1.3) 0.6 (0.9) 0.533; 0.2 фIndependent t-test; *P < 0.05. View Large Figure 3 (Supplementary Tables S9–S11) shows the effect of the VFRs on stability. Group A had significantly (P < 0.05, 95% CI excluded zero) increased LII after wearing the VFR-CV for 3 months by 0.35 mm (SD = 0.12) and after wearing the subsequent VFR-3D for the subsequent 3 months by 0.58 mm (SD = 0.19). Group B also showed a similar trend, but the increases were not significantly different after wearing both interventional retainers compared with the preceding retainers. The LII of subjects who wore VFR-CV and VFR-3D were not significantly different at both T2 and T3. Overall, the LII levels between the two retainers were not significantly different. Multivariate linear regression analysis (Table 2) showed that the LII was not significantly different between VFR-CV and VFR-3D after controlling for compliance, time, and gender. Figure 3. View largeDownload slide Comparison between vacuum-formed retainers fabricated from conventional stone model (VFR-CV) and those fabricated from 3D printed model (VFR-3D) in terms of their effect on stability. Figure 3. View largeDownload slide Comparison between vacuum-formed retainers fabricated from conventional stone model (VFR-CV) and those fabricated from 3D printed model (VFR-3D) in terms of their effect on stability. Figure 4, on compliance, shows a general trend of reduced compliance over time. For both groups, compliance at T2 of the VFR to the first retainer received was not significantly different but compliance at T3 was significantly reduced (P < 0.05) compared to that at T2. Compliance levels between VFR-CV and VFR-3D were not significant at T2 and T3. However, the multivariate linear regression analysis (Table 2) showed that compliance was a significant modifying factor impacting the OHRQoL (P < 0.001; 95% CI = −12.2, −3.8) and influencing stability (P = 0.007; 95% CI = −1.13, −0.19). Those who were compliant had lower OHIP scores (by 8.0 units) and lower LII (by 0.66 mm). Figure 4. View largeDownload slide Compliance assessment from the first standard vacuum-formed retainers received (VFR-1) at T1 to the interventions at T2 and T3 by subjects from both groups. Figure 4. View largeDownload slide Compliance assessment from the first standard vacuum-formed retainers received (VFR-1) at T1 to the interventions at T2 and T3 by subjects from both groups. Harms No serious harms were observed. No adverse events or side effects were reported. Discussion The study demonstrated that 3D printed models made using ABS filaments by the FDM printers (UP Plus 2 3D Printer; 3D Printing Systems) with 0.15 mm layering thickness can be used as working models to produce a VFR that is comparable to a conventionally made VFR in terms of its ability to maintain post-treatment changes without significantly affecting patient’s OHRQoL. Limitations There were several limitations to this study. This study was single-blinded, where each patient was assigned a blinding code, which was used for data entry by a clerk who was not involved in the treatment. The double-blind design was not possible in this study. Furthermore, the 3D printed models had lines reflecting the layers of printed material added on top of each other. The VFR adopted not only the anatomical shapes of the dentition but also the lines of the 3D printed model surfaces (Figure 5). Owing to the subtle differences in the physical appearances of the two VFRs, it was not possible to blind both the operator and patients. Even though the patients were not informed which were the controlled VFR and studied VFR, there could still be potential bias from the inability to blind them in this study. Figure 5. View largeDownload slide Observable differences between the two retainers. Figure 5. View largeDownload slide Observable differences between the two retainers. On the other hand, the crossover method gave an added benefit over conventional parallel-group trial design in that each patient served as his/her own control, thus requiring a lower sample size to meet the same criteria in terms of type I and type II error risks (18). This eliminated problems in comparing study and control groups with regard to confounding variables (e.g. age, sex, and ethnicity), thus compensating for unsuccessful randomization (19). However, a wash-out period, commonly done in crossover studies, between the two trials could not be performed because of the nature of the retention requirement. This limitation may potentially have caused a carryover effect from the first trial onto the second trial because participants may not have been given enough time to forget about their experience from the last trial. We felt it was not ethical to ask participants not to wear retainers during a wash-out period as it is not standard practice to leave the patient without retainers after debond. It was also not appropriate for the patient to wear other types of retainer during a wash-out period as they may have ‘carryover’ effects as well. In addition, a past study showed physiological impairment with VFR returned to normal within a week of insertion (20). Thus, with a 3-month period of retainer wear, we considered that there was sufficient time lapse for the participants to relate their experience only to the current trial. As for stability, it is unlikely to have a carryover effect since teeth movements only occur with direct force while retainers are passive appliances used to hold the teeth in the current position. Because dental movements only occur with direct force, any displacement observed is due to relapse, possibly from poor fitting retainers and unlikely to be due to the carryover effect from the previous retainer. In terms of relapse, there might be differences in risk of relapse between the first 3 months compared with the second 3 months after debond. However, in this study, both interventions were experienced by both groups at both time points. In addition, the multivariate linear regression analysis supported that time was not a modifying factor to the outcome of this study. A recent systematic review (21) has found that most retention and stability studies reported their outcome at 3 months, 6 months, and 1 year after debond. In this study, the 3-month review period was considered sufficient to allow patient adaptation to the new retainers to measure the outcomes. Our study protocol set the minimum threshold of part-time retainer wear at 10 hours daily as this regime has been found to be as effective as wearing retainers full time (7). The digital models produced by the Maestro 3D scanner were found to be clinically comparable to conventional stone models (22). When previous models were printed with the Zprinter 450, there was reduced morphologic detail of the 3D reconstructed models (6). The working model in this study, however, used ABS filaments printed with the UP Plus 2 FDM because our preliminary work showed that the 3D printed model was better able to withstand the heat and pressure of the VFR construction in comparison to the models made by the Zprinter. The ABS model build layer thickness of 0.15 mm was similar to that of RepRap 3D (5) but slightly thicker than that of the Zprinter of 0.089–0.102 mm (6). The ease to fit the VFR on subjects in this study suggests that the shape and size of the teeth and arch forms of the 3D printed models were similar to those of the original casts. The effective use of removable appliances requires the patient’s compliance, which was difficult to assess because of various factors that can affect compliance (23). The patient’s acceptance towards the appliance is important; to promote the success of the treatment the patient must agree to wear the appliance according to instructions. Self-reporting of compliance relies on trust, incurring the risk that a patient’s report may not correlate with the actual hours worn (24). On the other hand, although various devices have been introduced as an attempt to objectively evaluate compliance (25, 26), high cost, potential to increased bulk of the VFR, and inadequate information on their reliability and accuracy, we decided not to use such devices for this study. OHIP-14 was used because it specifically measures items that were important to this study. It is preferred to the long version (10), which may contain insignificant items. Thus, the short version ensures content validity of the questionnaire (9). Furthermore, having a smaller pool of items is less exhaustive for patients to complete. However, it should be noted that the short version only limits two most frequently reported items per domain (9). Therefore, other items which may also be of value were not measured. Interpretation A patient-reported outcome measure-based approach was used to determine how well the subjects accepted the VFR-3D in comparison to the conventionally made VFR-CV. Kang and Kang (27) have shown that retainer use affected the OHRQoL. Their subjects’ OHIP-14 scores were similar to our subjects’, even though they did not specify their type of retainers. In this study, both VFRs affected the subjects’ OHRQoL to a similar degree. OHRQoL improved after the first 3 months of wearing either VFR, particularly in the functional limitation domain. In the first 3 months, subjects who wore VFR-3D also reported significant improvement in the physical disability domain, unlike those who had VFR-CV in the first 3 months. The initial impacts related to ‘difficulties in speaking/pronunciation’ and ‘had bad breath caused by the retainer’ measured by functional limitation, ‘had people misunderstood some of their words’ and ‘avoided smiling’ measured by physical disability indicate that subjects had problems to adapt to the first VFR that they received at the first week. Improvement in the OHRQoL suggests that our subjects were able to adapt well to the interventional VFRs over the first 3-month period, and this situation was the same in OHRQoL from the second interventional VFRs over the subsequent 3-month period. Both interventional VFRs were reported to give similar impacts at both time points as well as overall, which suggests that the VFR-3D impacted the subjects’ OHRQoL to a similar degree as conventional VFRs. Despite the slight differences in appearances, subjects did not report differences in the impacts that measured psychological discomfort (such as ‘felt shy’), social disability (such as ‘avoided going out’), or handicap (such as ‘felt less confident’). VFRs are known to be more effective in maintaining mandibular anterior segment than are Hawley retainers (28, 29). This difference in effectiveness is due to the manner in which the VFR holds the teeth individually compared to Hawley retainers, which do not. For effective control of the labial segments, the working models should replicate the dentition as closely as possible so that the VFR made will fit well over the dentition. Our study showed gradual increase in the LII over time, but the differences were not clinically significant (13). Although group A had a greater increase in LII after wearing VFR-3D for 3 months than the increase after wearing VFR-CV for 3 months, they also had much lower compliance during the period of wearing the VFR-3D. Thus, the increase in LII could have been due to poorer compliance during the second intervention. The increases in LII by group B after wearing both retainers for 3 months were not statistically significant. Furthermore, this study found no significant differences between VFR-CV and VFR-3D in the LII changes over the first 3-month observation period and over the subsequent 3-month one, and overall after including all samples. This suggests that there are no differences in the fit of the VFR made on the ABS model compared to that made on conventional models, such that the occlusion was maintained in a similar manner in terms of irregularities and stability during the retention period. Our study found compliance affected the outcome. Non-compliant subjects were more likely to report higher impacts of the retainers on their OHRQoL. Their inability to persevere wearing the retainers according to instructions may have been attributed to the difficulties experienced with VFRs, regardless of the type of retainers, compared to the compliant subjects. In terms of stability, without the active use of the retainers, the position of the corrected dentition was not maintained. Thus, poor compliance increased the probability of relapse in this study. Nevertheless, there was no difference in compliance between the two groups at both time points. Both types of retainers had similar chances of impacting on the subjects’ OHRQoL and affecting their post-treatment stability because the duration of wear was comparable. In view of the findings, the null hypothesis was not rejected. This study could not find differences between VFR constructed on 3D printed model and VFR conventionally made on stone model in terms of the retainers’ effect on the impact on patients’ OHRQoL and post-treatment stability. Conclusion VFR fabricated on 3D printed model showed no differences in terms of patients’ OHRQoL and stability compared to conventionally made retainers. The ABS 3D printed model reconstructed by the UP Plus 2 3D Printer can be used as a working model for appliance construction. Availability of data The dataset supporting the conclusion of this article is available in the University of Malaya Repository (UM eprints ID: 18967 in http://eprints.um.edu.my/18967/) Supplementary material Supplementary data are available at European Journal of Orthodontics online. Supplementary data also includes severity of the overall OHIP-14 scores (Table S1), OHIP scores by domain (Table S5) and LII (Table S8) within all subjects, and the multivariate linear regression analysis with the domains of OHIP as the dependent variables (Table S12). Funding This work was supported by the University of Malaya Clinical Postgraduate Research Grant (PPPC/C1-2015/DGD/12). Acknowledgements The authors would like to thank Noor Lide Abu Kassim and Najihah Lokman for statistical advice. Special gratitude to Alena Sanusi for helpful comments. Conflict of interest The authors declare no conflict of interests. References 1. Groth , C. , Kravitz , N.D. , Jones , P.E. , Graham , J.W. and Redmond , W.R . ( 2014 ) Three-dimensional printing technology . Journal of Clinical Orthodontics , 48 , 475 – 485 . Google Scholar PubMed WorldCat 2. Murugesan , K. , Anandapandian , P.A. , Sharma , S.K. and Vasantha Kumar , M . ( 2012 ) Comparative evaluation of dimension and surface detail accuracy of models produced by three different rapid prototype techniques . Journal of Indian Prosthodontic Society , 12 , 16 – 20 . Google Scholar Crossref Search ADS PubMed WorldCat 3. Rengier , F. , Mehndiratta , A. , von Tengg-Kobligk , H. , Zechmann , C.M. , Unterhinninghofen , R. , Kauczor , H.U. and Giesel , F.L . ( 2010 ) 3D printing based on imaging data: review of medical applications . International Journal of Computer Assisted Radiology and Surgery , 5 , 335 – 341 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Keating , A.P. , Knox , J. , Bibb , R. and Zhurov , A.I . ( 2008 ) A comparison of plaster, digital and reconstructed study model accuracy . Journal of Orthodontics , 35 , 191 – 201 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Kasparova , M. , Grafova , L. , Dvorak , P. , Dostalova , T. , Prochazka , A. , Eliasova , H. , Prusa , J. and Kakawand , S . ( 2013 ) Possibility of reconstruction of dental plaster cast from 3D digital study models . BioMedical Engineering Online , 12 , 49 . Google Scholar Crossref Search ADS PubMed WorldCat 6. Wan Hassan , W.N. , Yusoff , Y. and Mardi , N.A . ( 2017 ) Comparison of reconstructed rapid prototyping models produced by 3-dimensional printing and conventional stone models with different degrees of crowding . American Journal of Orthodontics and Dentofacial Orthopedics , 151 , 209 – 218 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Thickett , E. and Power , S . ( 2010 ) A randomized clinical trial of thermoplastic retainer wear . European Journal of Orthodontics , 32 , 1 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat 8. Slade , G.D. and Spencer , A.J . ( 1994 ) Development and evaluation of the Oral Health Impact Profile . Community Dental Health , 11 , 3 – 11 . Google Scholar PubMed WorldCat 9. Saub , R. , Locker , D. and Allison , P . ( 2005 ) Derivation and validation of the short version of the Malaysian Oral Health Impact Profile . Community Dentistry and Oral Epidemiology , 33 , 378 – 383 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Saub , R. , Locker , D. , Allison , P. and Disman , M . ( 2007 ) Cross-cultural adaptation of the Oral Health Impact Profile (OHIP) for the Malaysian adult population . Community Dental Health , 24 , 166 – 175 . Google Scholar PubMed WorldCat 11. Little , R.M . ( 1975 ) The irregularity index: a quantitative score of mandibular anterior alignment . American Journal of Orthodontics , 68 , 554 – 563 . Google Scholar Crossref Search ADS PubMed WorldCat 12. Roberts , C.T. and Richmond , S . ( 1997 ) The design and analysis of reliability studies for the use of epidemiological and audit indices in orthodontics . British Journal of Orthodontics , 24 , 139 – 147 . Google Scholar Crossref Search ADS PubMed WorldCat 13. Little , R.M. , Riedel , R.A. and Artun , J . ( 1988 ) An evaluation of changes in mandibular anterior alignment from 10 to 20 years postretention . American Journal of Orthodontics and Dentofacial Orthopedics , 93 , 423 – 428 . Google Scholar Crossref Search ADS PubMed WorldCat 14. Jäderberg , S. , Feldmann , I. and Engström , C . ( 2012 ) Removable thermoplastic appliances as orthodontic retainers—a prospective study of different wear regimens . European Journal of Orthodontics , 34 , 475 – 479 . Google Scholar Crossref Search ADS PubMed WorldCat 15. Bradley , J.V . ( 1958 ) Complete counterbalancing of immediate sequential effects in a Latin square design . Journal of the American Statistical Association , 53 , 525 – 528 . Google Scholar Crossref Search ADS WorldCat 16. Crowder , M.J. , and Hand , D.J . ( 1990 ) Multivariate analysis . In Analysis of Repeated Measures . CRC Press , Florida, US , pp. 60 – 70 . Google Preview WorldCat COPAC 17. Zainab , S. , Saub , S. , Yaw , S.L. , Khairiyah , A.M. and Jamalludin , A.R . ( 2013 ) Oral health-related quality of life . In National Oral Health Survey of Adults 2010 (NOHSA 2010) . Oral Health Division Ministry of Health Malaysia , Putrajaya, Malaysia , pp. 387 – 410 . Google Preview WorldCat COPAC 18. Wellek , S. and Blettner , M . ( 2012 ) On the proper use of the crossover design in clinical trials: part 18 of a series on evaluation of scientific publications . Deutsches Arzteblatt International , 109 , 276 – 281 . Google Scholar PubMed WorldCat 19. Thiese , M.S . ( 2014 ) Observational and interventional study design types; an overview . Biochemia Medica , 24 , 199 – 210 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Wan , J. , Wang , T. , Pei , X. , Wan , Q. , Feng , W. and Chen , J . ( 2017 ) Speech effects of Hawley and vacuum-formed retainers by acoustic analysis: a single-center randomized controlled trial . The Angle Orthodontist , 87 , 286 – 292 . Google Scholar Crossref Search ADS PubMed WorldCat 21. Littlewood , S.J. , Millett , D.T. , Doubleday , B. , Bearn , D.R. and Worthington , H.V . ( 2016 ) Retention procedures for stabilising tooth position after treatment with orthodontic braces . Cochrane Database Systematic Review , 1 , CD002283 . WorldCat 22. Wan Hassan , W.N. , Othman , S.A. , Chan , C.S. , Ahmad , R. , Ali , S.N. and Abd Rohim , A . ( 2016 ) Assessing agreement in measurements of orthodontic study models: digital caliper on plaster models vs 3-dimensional software on models scanned by structured-light scanner . American Journal of Orthodontics and Dentofacial Orthopedics , 150 , 886 – 895 . Google Scholar Crossref Search ADS PubMed WorldCat 23. Lee , S.J. , Ahn , S.J. and Kim , T.W . ( 2008 ) Patient compliance and locus of control in orthodontic treatment: a prospective study . American Journal of Orthodontics and Dentofacial Orthopedics , 133 , 354 – 358 . Google Scholar Crossref Search ADS PubMed WorldCat 24. Cureton , S.L. , Regennitter , F.J. and Yancey , J.M . ( 1993 ) The role of the headgear calendar in headgear compliance . American Journal of Orthodontics and Dentofacial Orthopedics , 104 , 387 – 394 . Google Scholar Crossref Search ADS PubMed WorldCat 25. Ackerman , M.B. , McRae , M.S. and Longley , W.H . ( 2009 ) Microsensor technology to help monitor removable appliance wear . American Journal of Orthodontics and Dentofacial Orthopedics , 135 , 549 – 551 . Google Scholar Crossref Search ADS PubMed WorldCat 26. Schott , T.C. , Ludwig , B. , Glasl , B.A. and Lisson , J.A . ( 2011 ) A microsensor for monitoring removable-appliance wear . Journal of Clinical Orthodontics , 45 , 518 – 520 ; quiz 516. Google Scholar PubMed WorldCat 27. Kang , J.M. and Kang , K.H . ( 2014 ) Effect of malocclusion or orthodontic treatment on oral health-related quality of life in adults . Korean Journal of Orthodontics , 44 , 304 – 311 . Google Scholar Crossref Search ADS PubMed WorldCat 28. Rowland , H. , Hichens , L. , Williams , A. , Hills , D. , Killingback , N. , Ewings , P. , Clark , S. , Ireland , A.J. and Sandy , J.R . ( 2007 ) The effectiveness of Hawley and vacuum-formed retainers: a single-center randomized controlled trial . American Journal of Orthodontics and Dentofacial Orthopedics , 132 , 730 – 737 . Google Scholar Crossref Search ADS PubMed WorldCat 29. Demir , A. , Babacan , H. , Nalcacı , R. and Topcuoglu , T . ( 2012 ) Comparison of retention characteristics of Essix and Hawley retainers . Korean Journal of Orthodontics , 42 , 255 – 262 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2018. Published by Oxford University Press on behalf of the European Orthodontic Society. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Comparing retainers constructed on conventional stone models and on 3D printed models: a randomized crossover clinical study
JF - The European Journal of Orthodontics
DO - 10.1093/ejo/cjy063
DA - 2019-08-08
UR - https://www.deepdyve.com/lp/oxford-university-press/comparing-retainers-constructed-on-conventional-stone-models-and-on-3d-KENBQFIn5D
SP - 370
VL - 41
IS - 4
DP - DeepDyve
ER -