Toothbrush wear in relation to toothbrushing effectiveness

Abstract Objective To investigate to what extent the degree of toothbrush wear of 3‐month‐old manual toothbrushes influence plaque scores. Material and methods During a recently published study with a follow‐up of 1 year, all participants performed a similar basic home‐based oral hygiene regimen. Hence, they were instructed to brush for 2 minutes twice daily according to the Bass method technique and using a standard dentifrice containing sodium fluoride. Toothbrushes were turned in every 3‐month, and the degree of wear was scored. The mean plaque score data were additionally analysed and correlated with wear scores of the toothbrushes. Results For analysis, for each of 172 individual participants, a set of three identical, 3‐month‐old used toothbrushes were available. Toothbrush wear varied widely between participants. However, per patient, the 3‐month wear status of the three evaluated toothbrushes was strongly correlated (rho = 0.8, P < 0.0001). Participants who returned toothbrushes with extreme wear had significantly higher plaque scores than those who returned toothbrushes with no visible or light wear (P = 0.01). Conclusion Toothbrush wear per individual patient is fairly consistent. Toothbrushes with extreme wear were less effective than those with no or light wear. Therefore, bristle splaying appears to be a more appropriate measure of brush replacement time then the commonly used toothbrush age. Splaying of the outer tufts beyond the base of the toothbrush is a condition that indicates it is time to change the brush.

it loses its capacity to remove plaque effectively. This is most likely because filament tips that are bent will not adequately disrupt the plaque.
It is difficult to determine exactly when a toothbrush should be replaced. The American Dental Association recommends every 3-4 months or sooner if the bristles become frayed. 11 Toothbrush packaging sometimes includes the manufacturer's advice that the toothbrush should be discarded after 3 months. If a person brushes for 2 minutes, two times a day, 3 months may be equivalent to approximately 500 minutes of brushing per recommended lifetime of a toothbrush. 12 Although surveys among dental professionals show that replacement intervals of 2-3 months are recommended, [13][14][15] these suggestions do not seem to be based on firm scientific evidence. Interestingly, the lifespan proposed for a toothbrush appears to vary according to the person or organization suggesting it.
The criteria for replacing a toothbrush also differ. 16,17 It has been hypothesized most recently 18 that plaque removal decreases more due to a toothbrush's wear than to its age. In a study by Rosema et al, 18 the moment advocated for replacement was "when the outer tufts are splayed beyond the base of the toothbrush," as this was the state of wear at which a new brush always performed better than a worn one. This advice, however, was based on analyses of the brushes of only 45 participants.
To establish whether plaque score data would correlate with the wear score of the toothbrushes, and whether this would provide a basis for a recommendation when to replace a toothbrush, an explorative analysis of data obtained from a cohort of 267 participants who participated in a previous study comprising a 1-year period. 19 Clinical assessments were performed every 3 months, and the same type of fresh manual toothbrushes was provided for each period.
Toothbrushes were collected at each subsequent visit and stored for wear analysis.

| MATERIAL S AND ME THODS
The present study used plaque score data based on the modified Quigley and Hein 20 plaque index 21 (QHPI) obtained from a recent study 19  At screening, participants were asked to read and sign the informed consent form and were given a signed copy for their records.
In summary, to qualify for inclusion, the participants had to be ≥18 years of age, to have no systemic disorders, to have a minimum of 5 evaluable teeth per quadrant and to have moderate to advanced gingivitis (≥40% bleeding on marginal probing (BOMP)). 22,23 Exclusion criteria were open caries, Dutch Periodontal Screening Index (DPSI) scores ≥3 + , 24,25 orthodontic appliances or removable (partial) dentures and pregnancy.
All participants performed a similar basic oral hygiene regimen of brushing twice daily for 2 minutes with a fluoride-containing dentifrice for the full duration of the study. Table 1 and Figure 1 show detailed product information and instructions for use. Participants were instructed to brush according to the details provided in a written oral hygiene instruction leaflet describing the Bass method technique 26,27 and to brush 2-3 hours before all their appointments. 28 Participants were not allowed to use any other dental product or interdental cleaning aid during the study and/or to undergo dental prophylaxis during routine dental check-ups. At the first visit, participants handed in their used brushes. From that point onwards, each participant was provided with a new identical toothbrush on each subsequent visit (Table 1).
Among the cohort 19 that was followed at 3-month intervals ( Figure 2), the effect of the investigated interventions that had been provided at the start of the study on the clinically assessed parameters had worn off at the 4-month evaluation. Given that from that point onwards, no significant differences were found between groups, the toothbrush wear scores and mean plaque scores were used for all groups combined for this investigation.
Out of the original population, only those participants who returned their toothbrush at every occasion after 3 months were included for the analyses.

| Wear assessment
In our analysis, the degree of wear of the toothbrushes that had been collected was evaluated on a 5-point scale ( Figure 3) according to the method described by Conforti et al 29 The wear ratings were screened independently by three calibrated examiners (GVA, TA B L E 1 Following regimen groups were designed and described using the TIDieR checklist 45

Allocated
Brushing twice daily a for 2 min with a fluoride-containing dentifrice b during the study.

| Data analysis
The unit of analysis was the participant. Mean plaque scores per individual, per time point, were used as the main response variable in the analysis to establish whether these were correlated with wear scores. SPSS a was used to perform the statistical analyses.
The Spearman's Rho correlation coefficient of brush-wear scores was calculated for the toothbrushes used by the same individual for 3 months. These correlations were interpreted according to the suggestions by Evans. 30 The brush-wear score was assessed per toothbrush, and the plaque score means were calculated for each brush-wear category.
These scores were compared using the ANOVA test. Post-testing was performed to determine the origin of observed differences using independent t-tests between the wear groups. The P-values were corrected for multiple comparisons using the Bonferroni correction and were considered statistically significant if the P-values were <0.05.

| RE SULTS
A complete case analysis of three toothbrushes and corresponding plaque score was available for toothbrushes collected at the designated time points from 172 of the 267 enrolled participants of the original study. Participants from the control I group of the original study only returned for their final assessment, and no intermediate assessment was performed. Therefore, they could not contribute to the present data set (N = 44). Furthermore, there were dropouts (N = 16) and participants that did not return all of their toothbrushes (N = 35).
These were excluded from the present study which only assessed those with a complete data set at the 7-month assessment, 10-month assessment and final assessment.
Thus, 516 identical toothbrushes were available for analyses.
All toothbrushes were assessed for wear by three independent calibrated examiners who had a high interexaminer reproducibility score (0.95 Cronbach's alpha). Figure 4 shows the number of toothbrushes graded per wear score.
With respect to the influence of the degree of wear after 3 months on plaque removal, there was a significant (P < 0.0001) but weak positive correlation (Rho = 0.223). Figure 4 shows that subjects who had toothbrushes with extreme wear (score 4) had significantly higher plaque scores (Plaque index, PI = 1.98) than those with a brush with no visible wear (PI = 1.71) or with light wear (PI = 1.80).
Additionally, the scatterplot in Figure 5 shows that there is a wide range within the five wear score groups.
During the experimental period, three toothbrushes were pro-

| D ISCUSS I ON
Although individuals were rather consistent in the degree of wear they induced after 3 months, the present study shows that wear varied widely between individuals. With respect to toothbrushing efficacy, it seems that the age of a toothbrush should not be the factor guiding replacement. Instead, the level of wear appeared to be more important. This is consistent with the conclusion of Rosema et al 18 It has also been shown that toothbrush bristles that spread apart take on permanent curvatures. 31 Variation in the degree of wear is most likely caused by differing toothbrushing forces and techniques amongst individuals. 32 The individual manner of brushing seems to be of more importance than the length of time the brush is in use in the development of wear. 8,32 The most obvious aspect of brush wear is bristle splaying whereby the bristles spread apart and take on a permanent curvature. Several methods have been used for the measurement of this phenomenon, including the angle of bending of the outside bristles, 32 29 although being subjective and qualitative, is a quick means of ranking brushes in various stages of deterioration. Therefore, these methods appear to be suitable not only for research, but also for quality control, the setting of standards, and for substantiation of advertising claims.
Studies comparing manual vs power toothbrushes have shown that in power toothbrush users, bristle splaying was less than among those using a manual brush. 35 Furthermore, also quality issues of optically comparable brushes are apparent with this method of scoring where differences in susceptibility to splaying. 36 Consequently depending on the configuration of the filaments (tufts) and the quality of the bristles, the durability of toothbrushes will vary.  16 Previous studies suggest that a toothbrush's cleaning ability decreases as the filaments become worn. 17 Kreifeldt et al 16

ex-
plained that tapering will result in reduction in filament diameter, and thus, the brush will become softer and remove less plaque.
However, a recent systematic review 41 16,44 Different types of commercially available toothpastes influence the deterioration of the bristle tip morphology. Factors related to the abrasive toothpaste such as type, size and shape of the abrasive particles greatly influence the friction force generated by the toothbrush. 33 Extra soft toothbrushes appeared to be most susceptible to bristle wear. 45 The American Dental Association (ADA) guidelines on manual toothbrushes 46 suggest that, to claim that one brush is better than the other, there should be a minimum absolute difference of 15% in plaque scores. Although of the level of mean plaque scores in our study was statistically significant between the wear score extremes categories (0 and 4), the maximum observed absolute difference of 13.6% was close, but did not exceed this limit.
Given the guidelines from the ADA, in our study, toothbrushes with a brush-wear score of 0 had no clinically relevant benefit over toothbrushes with a brush-wear score of 4. However, the ADA has developed their guidelines around (randomized) controlled clinical trials, whereas the present observational study clearly showed that higher visible wear scores corresponded with higher plaque scores. The observed 13.6% difference in plaque scores deserves further research in order to establish the impact this will have on gingival inflammation in order to establish its clinical relevance.
One possible explanation for the relatively low maximum absolute difference is the study design. To avoid the risk of increased bleeding resulting from toothbrushing, 28 plaque scores were assessed 2-3 hours after brushing. This is contrast to Rosema et al 18 where plaque scores were assessed just before and immediately after brushing. Their study design was more experimental, whereas the present study was designed to evaluate effectiveness in an intervention under more or less ordinary day-to-day circumstances. Likewise, the level of plaque present after brushing is clinically of more relevance than the plaque reduction itself.
On average, the amount of plaque removed by toothbrushes with wear score 4 was significantly different from that removed by brushes with wear score ≤1. It therefore seems prudent to advise patients to replace their toothbrush before it reaches wear score 2, when outer tufts are splayed beyond the base of the toothbrush. This is in accordance with a previous study by Rosema et al 18

but in
contrast with older study's 2,47 who found no significant differences with between new and 3-month-old toothbrushes; however, these studies did not report on wear scores.
A problem associated with toothbrushes is that they are overthe-counter products for which no special instruction is given to the potential users when they purchase such an oral hygiene prod- for the toothbrush to be replaced. 48,49 But as observed by Hill and Kreifeldt 40 , it seems to be difficult for user's to judge the state of their own brush by only a picture. A short but concise explanation appears to be an important addition which is a responsibility that could be in the hands of the dental care professional. 2. Another limitation is that brushes were used for a restricted period of 3 months. It has been shown that during extended use, bristles become thin near their tips and take on a bent, matted appearance. This is probably the result of abrasive reduction in diameter, fatigue and the gradual accumulation of permanent strain. 16 Both matting and bristle tapering, as components of brush wear, contribute to loss of effectiveness, although matting rather than tapering appears to be the primary cause. 16 3. The wear index described by Conforti et al 29 is an subjective tool.

| LI M ITATI O N S
4. Habits such as "chewing" the brush head whilst brushing could also have contributed to the differing appearances of the worn toothbrushes.

| CON CLUS ION
Toothbrush wear per individual patient is fairly consistent.
Toothbrushes with extreme wear were less effective than those with no or light wear. Therefore, bristle splaying appears to be a more appropriate measure of brush replacement then the commonly used toothbrush age. It is suggested that the threshold at which a brush loses efficacy is when the outer tufts are splayed beyond the base of the toothbrush.

| Scientific rationale for the study
Advice varies on how frequently a toothbrush should be replaced.
There are no data on how consistently an individual causes wear to his or her toothbrush.

| Principal findings
After 3 months of use, toothbrush wear per patient was strongly correlated. Toothbrushes with extreme wear were less effective than those with no or light wear.

| Practical implications
Equating brush wear (and, presumably, loss of effectiveness) with brush age in use is not justified. Advice on replacing toothbrushes should be based mainly on bristle flaring rather than on a "fixed" period of usage. We recommend that a manual toothbrush should be discarded when its outer tufts are splayed beyond the toothbrush base.
Dental professionals should be aware of these differences, both in durability and in cleaning performance, when recommending brushes to their patients.

ACK N OWLED G EM ENTS
The authors wish to thank Nienke Hennequin-Hoenderdos, Eveline van der Sluijs, Claire Berchier, Sam Supranoto (SCS), Guylaine van Anraat (GVA) and Paula Versteeg for their help during this trial and their assistance with study coordination.