Comprehensive comparison of canine retraction using NiTi closed coil springs vs elastomeric chains:A split-mouth randomized controlled trial

Canine retraction and space closure is considered the most time-consuming phase in orthodontic treatment.¹  Acceleration of this step would reduce overall treatment time, improve patient cooperation, and decrease possible negative side effects.^2,3 

Manipulation of tooth biomechanics^1,4  and tissue reaction⁵  have been widely attempted to reduce treatment duration. Additionally, the rate and safety of different canine retraction methods⁶  and different force systems^7,8  have been intensely investigated.

The use of sliding mechanics for canine retraction has been frequently reported in the literature.⁹  This method reduces the chairside time compared to loop fabrication. Despite the fact that tooth movement along the arch wire is highly predictable, friction between the brackets and arch wires may bring about some limitation to the tooth movement.¹⁰ 

The wide use of NiTi coil springs for canine retraction can be attributed to their relatively constant force delivery,¹¹  hence, reducing the number of appliance reactivations. On the other hand, elastomeric chains deliver an interrupted force that provides periods of rest allowing for regeneration and better tolerance of the supporting tissues.^12,13  Its main weakness is the absorption of oral fluids leading to biodegradation and rapid force decay with a consequent need for frequent reactivation.¹⁴ 

Although the efficacy of coil springs and elastomeric chains has been studied frequently, few randomized controlled trials have compared the two methods for canine retraction directly.^15–19  A recent systematic review and meta-analysis²⁰  compared the efficacy and side effects of power chains and coil springs for space closure. The results showed a similar rate of retraction and insufficient data to compare side effects like pain, root resorption, and patient discomfort.

The aim of the present study was to compare the rate of canine retraction using NiTi closed coil springs and elastomeric chains comprehensively in orthodontic patients requiring first premolar extraction and maximum anchorage. The null hypothesis assumed that there would be no difference in the rate of canine retraction between the two methods. Other side effects were considered, including rotation, tipping, root resorption, anchorage loss, as well as associated pain.

This split-mouth randomized controlled trial was conducted at the Orthodontic Outpatient Clinic, Future University in Egypt, Egypt, between January 2018 and February 2019. The Institutional Review Board at FUE approved the study in April, 2017. Patients and parents who agreed to join the trial signed consent forms at the start of treatment.

Thirty-five consecutive patients seeking orthodontic treatment were screened. The inclusion criteria were patients with a full permanent dentition (not necessarily including third molars), patients indicated for bilateral first premolar extractions, and canine retraction in at least the maxillary arch with maximum anchorage required. Patients who reported any systemic disease or medication interfering with bone metabolism or patients with severe skeletal discrepancies indicated for orthognathic treatment were excluded from the study. Other exclusion criteria included pregnancy, patients with craniofacial deformities, or periodontal disease.

Conventional 0.022-inch Roth prescription brackets were used for all teeth except the second molars. The canines were bonded with vertical slot brackets (American Orthodontics, Sheboygan, Wis). The posterior segment was aligned while bypassing the incisors, reaching 0.017- by 0.025-inch stainless steel wires. Vertical power arms, 8 mm in length, were fabricated and inserted into the vertical slots of the canine brackets. Temporary anchorage devices (3M Unitek TAD, St. Paul, Minn., 8 by 1.6 mm) were placed between the roots of the second premolar and first molars in the maxilla and mandible (Figure 1). Patients were then referred for first premolar extractions and canine retraction was initiated within 2 weeks.

A NiTi closed coil spring (6 mm; Ormco, Orange, Calif.) was used for canine retraction on one side (Figure 1), while elastomeric chain (American Orthodontics) was used on the contralateral side (Figure 2). Both force delivery systems were extended between the inserted temporary anchorage devices (TADs) and the vertical power arms of the canine brackets. The retraction force was adjusted to 150 g using digital force gauge. A ligature wire was used to attach the coil spring to the TAD if the force exceeded 150 g. During monthly follow-up visits, the force delivered by the coil spring was measured and adjusted while the power chain was replaced to maintain constant force delivery.

Seven maxillary dental impressions were taken for each patient: immediately before canine retraction and every 4 weeks for 6 months. Digital models were obtained through laser scanning of plaster models using the R500 3Shape scanner (3Shape, Copenhagen, Denmark). Sagittal, horizontal, and frontal reference planes were constructed to orient the pre-retraction digital models and superimpose them on the medial points of the third rugae (Figure 3).

Measurements were taken using 3Shape Analyzer computer software (3Shape). All measurements were calculated as the difference between the pre-retraction model (M0) and the model taken at the end of 6 months of retraction (M6). Total canine retraction was represented by the perpendicular distance from the canine cusp tip to the frontal plane. Likewise, the mesial drift of the maxillary first molar was measured from the mesiobuccal cusp tip to the frontal plane. Canine rotation was measured as the angle between the projected line connecting the mesial and distal contact points of each canine and the frontal plane.

A cone beam computed tomography (CBCT) scan (Acteon X-mind Trium CBCT machine, La Ciotat, France) was obtained before the start of canine retraction and 6 months later. In accordance with the ALARA (as low as responsibly achievable) guidelines,²¹  a medium CBCT field of view was used. The digital imaging and communications in medicine (DICOM) images were imported into Invivo Dental 5 software (version 5.3.1, Company, Santa Clara, Calif.) and 3D images were constructed (Figure 4).

The pre- and post-retraction CBCTs were compared. The total amount of canine retraction and canine rotation was measured using the same landmarks used for the analysis of the digital models. Canine tipping was calculated as the change in the angle between the long axis, from cusp tip to root apex, and the frontal plane. Root resorption was measured as the difference in the length of the canines from the cusp tips to the root apices.

The patients reported the presence or absence of dental pain for the first 10 days after each activation. The patients were asked to record the intensity of the pain on a 100 mm visual analogue scale.²² 

Intrarater reliability was evaluated by remeasuring six randomly selected digital models and CBCTs and another investigator measured these records for interrater reliability.

A priori sample size was determined from the data reported by Dixon et al.¹³  using the G*Power software. Twenty-eight observations were required in each group for a study power of 0.8 and an 0.05 alpha error.

In Microsoft Office Excel Mac (version 16.24; Microsoft, Redmond, Wash.), the right quadrants in 32 arches were equally and randomly assigned to one of the two interventions. The contralateral quadrant received the other intervention. The allocation ratio was 1:1.

It was not possible to mask the patients or the orthodontist providing the treatment. However, the outcome assessor was masked to the intervention.

The data were analyzed using SPSS (version 17, Chicago, Ill.), Descriptive statistics were reported for all outcomes. The Shapiro-Wilk test evaluated data normality. All variables except pain followed a Gaussian distribution. Independent t-test was used to compare the mean differences of the two groups for monthly and overall canine retraction rate as well as overall dental change at the end of 6 months. The Mann-Whitney test was applied to compare the pain intensity across the two groups. The number of days with pain for the first 10 days after each activation were described by proportions and percentages and compared using the Z score test. Tests were two tailed and the significance level was set at P < .05.

The progress of patient selection and recruitment is shown in Figure 5. The CBCT for one patient was distorted causing some missing data. These measurements were statistically imputed.

Intra- and interobserver agreement for angular and linear measurements on the CBCT and digital models was 0.99 (intraclass correlation coefficient).

There was no statistical difference between the groups for monthly canine retraction rate. The average monthly rate was 0.79 ± 0.138 mm and 0.86 ± 0.14 mm for the NiTi coil spring and the elastomeric chain, respectively (Table 1). The total amount of retraction for pooled maxillary and mandibular canines was 4.44 ± 2.22 mm and 4.33 ± 1.31 mm for the NiTi coil spring and the elastomeric chain groups, respectively (Table 2), while it was 4.73 ± 1.64 mm and 5.17 ± 1.71 mm for the maxillary canine retraction (Table 1).

At 6 months, the groups were similar for canine tipping, rotation, and root resorption as well as for maxillary first molar mesial drift (Table 3).

No statistical difference between the groups was reported for pain intensity. However, patients in the coil spring group reported significantly less days with pain (Table 4). About 70% of patients reported pain only for the first 2 months.

Many previous studies have evaluated the rate of canine retraction.^{15,17 23} The lack of evidence to support a single, fast method with minimal unwanted tooth movement and discomfort has been reported.⁶  Hence, there was a recommendation for further studies with more methodological rigor and patient-related outcomes.²⁰ 

Sliding mechanics has been shown to be a controlled and predictable method of space closure. Coil springs and elastomeric chains have been used in a multitude of studies. Yet, only eight primary studies^{8,13,15–19,23}  compared the efficacy of coil springs and elastomeric chains for canine retraction and space closure. Various outcomes including the total distance of retraction, the percent of closed extraction spaces within a given period of time, as well as weekly, monthly, and overall retraction rate have been measured. The mean differences for the monthly rates reported by Nightingale and Jones,⁸  Bokas and Woods,¹⁵  Khanmasjedi et al.,¹⁶  Talwar and Bhat,¹⁷  Davidovic et al.,²³  and Chaudhari and Tarvade¹⁹  were 0.05, 0.17, 0.23, 0.28, 0.21, and 0.25 mm, respectively.

In the current study, the difference between the two groups (0.02 mm) was the smallest compared to previous studies, which ranged between 0.05 mm and 0.28 mm. However, previous results consistently demonstrated that the differences were not clinically significant. This can be explained by the similar biologic response of the periodontium despite the different methods of force application.

The rate of canine retraction in the present study was similar to that reported by Chaudhari and Tarvade¹⁹  and Dixon et al.¹³  for the coil spring (0.81 ± 0.51 mm) and the elastomeric chain (0.58 ± 0.3 mm). The highest rates of retraction were reported by Bokas and Woods¹⁵  (1.85 mm) and Khanmasjedi et al.¹⁶  (1.67 ± 0.39 mm) for the coil spring group and 1.68 mm and 1.89 ± 0.36 mm for the elastomeric chain, respectively. Similar rates were produced with different arch wire diameters.^15–17 

Force magnitudes ranging between 50 and 300 g have been used effectively in canine retraction.^1,9,10  Most clinicians activate the spring to produce 100–250 g according to the manufacturers' instructions for most NiTi closed coil springs.

For elastomeric chains, some investigators recommended the use of higher initial forces to compensate for the loss of elasticity and force biodegradation. Nightingale and Jones⁸  reported that biodegradation was clinically much lower than the amounts expected. Greater initial forces (300–450 g) were associated with more force biodegradation and were not correlated to the amount of canine retraction.⁸ 

In the current study, all patients required maximum retraction of the anterior segment and TADs were used for direct anchorage. The average mesial drift of the posterior segment was <0.05 mm during 6 months in both groups (Table 2). These results were similar to those reported by Al Suleiman and Shehadah¹⁸  using miniscrews. The use of TADs as direct anchorage during canine retraction is often advocated for patients requiring maximum anchorage.

During tooth movement, force application away from the center of resistance results in unwanted tipping and rotation. The correction of side effects could prolong the overall treatment time. These outcomes should be considered in relation to the rate of canine retraction and space closure.

In this study, canine tipping was similar between the two methods. The amount of tipping was significantly reduced by using the vertical power arm. Tipping was less than that observed by Al Suleiman and Shehadah.¹⁸  Yet the position of the power arm caused cheek lacerations related to the coil springs in two mandibular quadrants. Canine rotations in the present study were almost four times (Table 2) the amount reported by Al Suleiman and Shehadah¹⁸  at 3.43° ± 1.38 and 3.32° ± 1.42 for the coil spring and elastomeric chain, respectively.

Pain and root resorption have been associated with orthodontic tooth movement. These outcomes have not been investigated in studies comparing NiTi closed coil springs and elastomeric chains for canine retraction. In the current study, the amount of root resorption observed was in agreement with the results of a previous systematic review²⁴  (0.3–12.83 mm) for tooth movement with conventional brackets. Although the elastomeric chains produced intermittent forces, there was no difference in the amount of root resorption between the two groups. This may be explained by the findings of Nightingale and Jones,⁸  in which the difference between force biodegradation in the two groups was not as significant as generally expected.

Pain was investigated using the visual analogue scale,²²  which has been validated and is commonly used for patient reported outcomes. The severity of pain was similar between the two groups. Yet, patients reported significantly fewer days of pain in the quadrants in which NiTi coil springs were used for retraction. Since the retraction force systems were similar for both methods, this may have been due to less activation adjustments required by the coil spring. However, due to the subjective²⁵  and variable nature of pain, further investigations are needed for conclusive results.

An attempt was made to reduce performance and assessor bias. One clinician treated all the patients with a similar protocol except for the method of retraction. The retraction force was standardized using a gauge. It was not possible to mask patients to the retraction methods. Nevertheless, this had no impact on the results as retraction did not rely on patient cooperation. On the other hand, the outcome assessor was masked to the intervention and assessed de-identified digital models and CBCTs.

The changes in the maxillary arches were measured on digital models scanned from dental casts. Lemos et al.²⁶  reported the accuracy and reliability of the 3Shape scanners used in this study.

CBCTs were used to assess changes in the mandible due to the difficulty in superimposing subsequent models. Baumgaertel et al.²⁷  showed that measurements constructed from CBCT scans were accurate and reliable. After the ALARA guideline,²¹  the CBCT scans were taken using a medium field of view where images were confined to the borders of the upper and lower arches.

Unlike other studies comparing the NiTi closed coil spring and the elastomeric chain, this study investigated all relevant outcomes to allow a comprehensive comparison regarding efficiency and adverse effects.

• There was no clinical or statistical difference in canine retraction rate, tipping, rotation, or root resorption between the NiTi closed coil spring and the elastomeric chains to recommend one method over the other.

• Significantly fewer days with pain were reported for the NiTi closed coil spring. Further studies are needed to investigate this finding.