Factors associated with orthodontic pain

Abstract Background Pain experienced at teeth during orthodontic treatment varies largely over time, with the reasons for its interindividual variability being largely unknown: age, sex, clinical activations, psychosocial factors and genetic polymorphisms of candidate genes are putative factors that may account to explain this variability. We aimed to investigate the effect of clinical, demographic, psychological and genetic factors on pain levels experienced during fixed orthodontic treatment. Methods A convenience sample of 183 patients undergoing full‐fixed orthodontic treatment were recruited. Participant's pain levels were assessed seven times over a three‐day period via a smartphone app. Clinical, demographic and psychological data were collected via questionnaires. This included the Pain Catastrophising Scale (Child version), the Corah Dental Anxiety Scale and the State and Trait Anxiety Inventory. Participants provided a DNA sample either in the form of blood or saliva, which were used for genotyping COMT gene rs6269, rs4680, rs4646310, NR3C1 gene rs2963155 and the HTR2A gene rs9316233. Results Bond ups had the greatest influence on perceived levels of pain experienced on teeth during orthodontic treatment, accounting for over 20% of total variance in pain response. High‐pain responders had higher scores on pain catastrophising (magnification subscale). Self‐reported pain during fixed orthodontic treatment was not influenced by sex, age, time into treatment, anxiety, nor by polymorphisms of COMT, HTR2A or NR3C1 genes. Conclusions Pain on teeth resulting from orthodontic fixed appliances is stronger during bonds‐up and in patients with high catastrophising scores. Demographics, type of clinical activations and the genetic polymorphisms investigated in this research had little or no impact on perceived pain levels.

ising scores being best treated as a continuous variable as opposed to a categorical variable. 22 The COMT gene is implicated in pain perception and is found on the long arm of chromosome 22 and codes of the enzyme catecholamine-O-Methyltransferase, which is involved in the degradation of catecholamines (eg dopamine, adrenaline and noradrenaline) as well as catechol oestrogens and other substances which have a catechol structure. It has been suggested that multiple Single Nucleotide Polymorphisms (SNPs) have a synergistic effect on COMT enzyme activity, producing a much more pronounced effect than any singular SNP. 23 Diatchenko et al 2005 separated subjects into three major haplotypes, LPS (low-pain sensitivity), APS (intermediate pain sensitivity) and HPS (high-pain sensitivity). SNPs of the COMT gene have been associated with oro-facial pain as well as differences in pain perception. 24,25 The HTR2A gene is located on chromosome 13 and codes for a serotonin G protein-coupled membrane receptor. It is involved in the serotoninergic system with serotonin being the main neurotransmitter. The serotoninergic system has many functions, such as learning and memory, but perhaps is most well known as a contributor to a feeling of well-being and happiness. Whilst the HTR2A has not been as extensively studied as the COMT gene, certain SNPs has been associated with musculoskeletal pain 26 and temporomandibular joint disorders. 27 The NR3C1 gene is located on chromosome 5 and codes for a glucocorticoid receptor. This receptor is involved in the primary endocrine stress axis in humans, which plays an important role in pain perception. Though the role of this gene and its exact function has not been well studied, SNPs of the NR3C1 gene have been associated with chronic fatigue syndrome 28 and oro-facial pain. 27 In a pilot study, we investigated the association between certain genetic polymorphisms and the patient's pain experience during fixed orthodontic treatment. 29 The preliminary findings indicated that individuals with the genotype AA of the COMT gene (rs464310) and CG of the HTR2A gene (rs93116233) could experience significantly more pain during fixed orthodontic treatment.
We hypothesised that patient's self-reported pain levels during fixed orthodontic treatment are influenced by clinical factors, psychological factors, such as anxiety and pain catastrophising, as well as certain single nucleotide polymorphisms (SNP) of the COMT, HTR2A and NR3C1 genes. Eligible participants completed a self-reported questionnaire, which contained basic demographic information (sex, age and   ethnicity), participants also completed a seven-page questionnaire which included the Pain Catastrophizing scale for Children,   the Corah Dental Anxiety Scale and the State-Trait Inventory for Children. For DNA collection, blood samples were taken in participants who were willing, whilst a saliva sample was taken in participants who were unwilling to provide blood samples. The blood samples included a 10mL EDTA tube whilst 10mL saliva samples were taken with Genotek TM Oragene-500 kits. All DNA samples were stored in a refrigerator set at 3℃ and transported weekly to Merriman Laboratories (University of Otago) for storage, DNA extraction and genotyping.

| ME THOD
Immediately following an orthodontic appointment participants were issued an Android smartphone (Vodafone Smart Prime 6, with a 5" colour display) and shown how to use an Android smartphone application (MyBraces Experience) on the issued phone.
The app was developed as part of this study to record the severity of pain following an orthodontic adjustment appointment.
The app included a series of pain surveys, which asked about the participants' analgesic consumption, resting pain at the teeth and pain at the teeth immediately after chewing a piece of chewing gum twenty times (Wrigley Extra Wrigley, Chicago, Illinois, USA). The pain surveys used did ask whether participants experienced any pain around the temporomandibular joint (TMJ) or headaches.
Participant reported their pain severity on a digital sliding visual analogue scale (VAS) measuring 9.3cm long, and participants were asked to drag a small circular dot (1.5mm in diameter) on the VAS to rate the severity of pain they felt at the time. The VAS was labelled 'no pain at all' on the left-hand side and 'worst pain imaginable' on the right side. Each participant was required to fill out the pain survey outline by app seven times, that is at (1) baseline on day one (immediately after an orthodontic adjustment); (2) 8pm on day one; (3) 8am on day two; (4) 8pm on day two; (5) 8am on day three; (6) 8pm on day three; and (7) 8am on day four. The orthodontic adjustment was made between 10am and 4pm of the same day the survey started.
The app sent screen and audio alerts to participants when it was time to complete the pain survey; the app allowed participants to enter data up to three hours before and up to three hours after the specified time; if no data were entered within this time period, a missing score was recorded. When all pain survey sessions were completed, the answers were stored on a local database (SQLite, https://www.sqlite.org/). Once all the sessions were completed, the data were retrieved from the local database and compiled into a single comma-separated values file. The file was emailed using a JavaMail API to a Google Mail account. If no internet connection was available, the app notified the user to connect to the internet, and the email was sent.
Adjustment details at the orthodontic adjustment appointments were recorded and coded into five mutually exclusive categories as follows: 1. no arch wires changed +/-minor bends in arch wire; 2.

| Outcomes
The primary outcome measurements were pain at the teeth at rest and after chewing gum in the 72 hours following an orthodontic adjustment. The collected VAS scores were expressed as percentages (0-100). Because VAS scores were not normally distributed, the cumulative score of pain levels assessed across the three days was calculated and defined henceforth in this manuscript as pain integral.
This was compared against a participant details of adjustment, their psychological scores (Pain catastrophising, rumination, magnification, helplessness, A-State and A-Trait scores and Dental Anxiety scores), and their genotypes and haplotypes. Type I error was set at 0.05 and the proportion of explained variance (R 2 ) was calculated for each factor. Comparison of means was performed using Student's t test if the data were normally distributed, or using the Mann-Whitney U test if the data were not normally distributed. Comparisons of proportions and the chi-square test. One-Way ANOVA was used to assess differences in pain severity across genotype and haplotypes.

| Sample
A total of 183 patients were recruited for this study, the mean age of participants was 14.9 years (SD 1.5 years) and ranged between 11.9 and 18.6 years. There were a greater number of female participants

| Orthodontic adjustment
All 183 participants' data were used for this part of the analysis.
The VAS ratings for resting pain and chewing pain over the 72 hours showed a peak the day after the clinical activations and were markedly reduced on day 4 ( Figure 1). reported having a headache during the survey, whilst 12 (6.5%) participants reported having pain in the TMJ area. This pain, however, was consistently reported throughout the survey by only one participant.

| Psychological Factors
A total of 177 participants' data were used for psychological analysis, six participants were excluded due to either missing data or incomplete questionnaires. High-pain responders showed a pattern to have a higher total pain catastrophising, rumination, helplessness, magnification, A-State and A-Trait scores when compared to lowpain responders. However, only magnification scores were significantly higher the high-pain responders' group, when compared to the low-pain responders' group (2.8 vs1.4; p=0.048).
No statistical difference in the dental anxiety scale score was found between high-pain responders, average pain responders or low-pain responders (p=0.966). Pain catastrophising, rumination, helplessness, magnification, A-State, A-Trait and dental anxiety scale scores did not differ when comparing average pain responders to high-pain responders nor to low-pain responders (Table 1).

| Genetics data
Of the total 179 DNA samples collected, only 172 participants' data were used for genetic analysis, and eleven were excluded due of the COMT gene at position rs4646310 in one study participant was associated with higher pain levels compared with the AG and GG genotypes, but the difference could not be statistically evaluated. No statistically significant difference was found between the remaining SNPs of the COMT, HTR2A and NR3C1 genes (F ≤ 3.0; p ≥ 0.05, Table 2).

| D ISCUSS I ON
After an orthodontic adjustments, our study participants experienced mild-to-moderate levels of pain, which steadily increased following the adjustment and peaked at 8am of day 2 and then grad-  21,35 ; however, these studies were not conducted in relation to orthodontic treatment. The lack of significant difference in A-State and A-Trait levels between the two pain groups was consistent with a previous study conducted at the University of Otago which also did not find A-State and A-Trait scores to differ between high and low-pain responders after the placement of orthodontic separators. 36 Dental anxiety scores were almost identical between the high-and low-pain responders, with both groups have a mean DAS score of 7.8. Most literature cite that dental anxiety levels were positively related to the patient's pain experiences during routine dental procedures. 18  Overall, this study has found that participants perceive the greatest level of pain after an initial bond up, which accounted for 20% of the reported variance. We were unable to find any statistically significant findings between pain catastrophising, A-State, A-Trait, Dental anxiety scores or SNPs of the COMT, NR3C1 and HTR2A genes and participants self-reported pain levels during orthodontic treatment.
A larger sample size will be required to establish a clearer link between genetic markers and pain experienced during orthodontic treatment, and the current analysis for the genetic data should be considered exploratory, due to limited power and a relatively small convenience sample.
In conclusion, pain at teeth during orthodontic treatment with fixed appliances is stronger during bonds-up and in patients with high catastrophising scores. Sex, age, type of clinical activations and the genetic polymorphisms investigated in this research had little or no impact on perceived pain levels.

CO N FLI C T O F I NTE R E S T
The authors have nothing to disclose.

AUTH O R CO NTR I B UTI O N S
All the authors listed have made substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; have been involved in drafting the manuscript or revising it critically for important intellectual content; gave final approval of the version to be published; and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.