Chlorhexidine for prevention of alveolar osteitis: a randomised clinical trial

Abstract Objective To determine the effectiveness of chlorhexidine 0.12% mouthwash (CHX) after tooth extraction for the prevention of alveolar osteitis (AO). Material and methods We conducted a double-blind randomised clinical trial stratified by risk factors. We enrolled a cohort of 822 patients who underwent dental extractions, and were considered to be at risk of developing AO (previous surgical site infection, traumatic extraction, and tobacco smoking). After extraction, patients were randomly allocated for CHX group or placebo group, matched by risk factors. The primary outcome was clinical diagnosis of AO: increasing postoperative pain for 4 d within and around the socket, and total or partial breakdown of the blood clot in the socket with or without bone exposure. Results Follow-up was completed by 744 participants (372 chlorhexidine and 372 placebo). We detected no significant differences between the two groups at baseline. After completed follow-up, risk factors were equally distributed between the two groups. Overall incidence of OA was 4.97%, in which 27 participants treated with placebo (7.26%) and 10 participants treated with CHX (2.69%) developed AO. CHX reduced the incidence of AO by 63% [Absolute Risk Reduction: 4.57 (95% CI 1.5-7.7), Number Needed to Treat: 21.88 (95% CI 13.0-69.3), Fisher's exact test: p=0.006]. No adverse effects were reported. Conclusion The use of chlorhexidine 0.12% mouthwash after tooth extraction is safe and effective in reducing the incidence of AO in high-risk patients.


Introduction
Patients who undergo dental extractions are at risk of developing post-operative complications, and the most common is the alveolar osteitis (AO) 12 . Our group 8 recently reported an incidence of 6.4% of AO, and we determined by a logistic regression model that previous surgical site infection, tobacco smoking and traumatic extraction are risk factors for developing AO.
These risk factors explain why there were changes in prevalence ranging from 3.9% up to 29.6% for the third molar 10,23 .
With a risk model that predicts the development of AO, it is possible to implement preventive health care to those individuals at high risk. Thus, clinicians can help their patients to avoid the severe pain caused by this complication 17 , which traditionally receives symptomatic treatment of uncertain effectiveness 16 .
Hence, clinicians can reduce health care costs and provide comfort for patients 9 .
It has been proposed that chlorhexidine 0.12% mouthwash can be used after extraction for the prevention of AO 10,13,18 . Chlorhexidine is the most widely used antiseptic in dentistry because its broad-spectrum antibacterial effectiveness is well established 2 , so it can be implemented as a simple and inexpensively public health policy.
However, the clinical trials that support chlorhexidine 0.12% mouthwash are inconclusive, showing methodological weaknesses and having a high risk of bias 5,9,24 . Therefore, the need to conduct randomized clinical trials of better quality and including risk factors is imperative.
Here, we conducted a double-blind randomised clinical trial stratified by risk factors to determine the effectiveness of CHX after tooth extraction for the prevention of AO.   To avoid the risk of having more patients in a group, we stored black envelopes in a box containing a paper with the letter C for chlorhexidine or P for placebo (half of each). The envelopes were chosen for each patient after the extraction and transported to another room (without opening them); they were read only by one of the authors, who then distributed the chlorhexidine or placebo accordingly. For each patient who was assigned to a group, the subsequent patient who arrived with the same risk factors was matched to the opposite group, and the respective envelope was discarded (to ensure homogeneity of groups).

Outcome measures
We recorded age (years), gender (male or female), tooth location (mandibular or maxillary), diagnosis or previous surgical site infection (yes or no, as described), smoking (smoker or non-smoker, as described) and traumatic extraction (yes or no, as described) before tooth extraction for each patient.

Sample size
We estimated sample size using data published previously by Halabi, et al. 8 (2012), expecting an incidence reduction of two-thirds. The P A expected incidence of disease (AO) in the placebo group was 6.14%, while the P B expected incidence of disease (AO) in the CHX group was 2.05%. Additionally, k groups ratio of sample sizes between groups was 1:1.
The power of the study was set at 80% (β=0.20), with α=0.05 as the significance level. Based on these parameters, we applied the following equation 19 :

Statistics
We performed the statistical analysis by DH using R 3.3.1 (R Foundation for Statistical Computing, Vienna, Austria). We used the Fisher's exact test to detect significant differences in the incidence of AO between CHX and placebo groups, and also to analyse gender, tooth location, patient smoking, previous infection and traumatic extraction variables. We used unpaired t-test was to detect significant differences in age between CHX and placebo groups. We determined the incidence of AO for both groups and calculated the Number Needed to Treat (NNT). For all tests, statistical differences were determined to be significant at p<0.05.

Results
From April 2013 to December 2015, we recruited 822 participants. Out of these, 744 met inclusion criteria and completed the follow-up. They were allocated into two groups of 372 for treatment of chlorhexidine 0.12% or placebo. Figure 1 shows the flow diagram of participants.
As seen in Table 1 Once all participants completed the follow-up, we observed that the risk factors were equally distributed between the two groups, without statistically significant differences (see details in Table 2).
We diagnosed 37 cases of AO, with an overall prevalence of 4.97%. In the group treated with See more details in Table 3.
No patient had hypersensitivity to chlorhexidine, dysgeusia or tooth pigmentation.

Discussion
We conducted a randomised, controlled trial to assess the effectiveness of postoperative chlorhexidine 0.12% mouthwash to reduce the prevalence of postextraction AO, and we contrasted it with a placebocontrolled group. We presented new findings in which chlorhexidine treatment reduces the incidence of AO by 63% in high-risk patients, with strong statistical power.
Subgroup analysis showed no difference in the incidence of AO and location of the extracted tooth.
Mandibular teeth developed the same rate of AO as that of the maxillary teeth, and anterior teeth showed  Therefore, our results can be extrapolated to any patient who requires an extraction, regardless of the location of the tooth.
AO incidence in this study was quite similar to that which we reported previously: 4.97% versus 6.4% 8 .
We suspect that the incidence might have been higher in the placebo group on this occasion (incidence was 7.26%), since the study sample consisted of patients with an increased risk of developing alveolar osteitis. Hermesch, et al. 10 (1998) reported that chlorhexidine 0.12% mouthwash reduced to 38% the prevalence of AO after extraction of impacted mandibular third molars (RR 0.62; 95% CI, 0.40 to 0.96). In contrast, Delilbasi, et al. 7 (2002) found that 0.2% chlorhexidine had no statistically significant effect in reducing the prevalence of AO after mandibular third molar removal (RR 0.88; 95% CI, 0.45 to 1.72). However, these studies have a high risk of bias 6,9,24 .
Our study has several strengths compared to those reported in the literature. To minimize selection bias, we randomly included only the patients who had the risk factors that we reported previously 8 . Thus, the preventive effect was studied in a group known to have a high risk of developing AO, and not in persons who have not had chances to develop AO, regardless of receiving treatment or not.
We assigned participants to each group randomly, blindly and matched by risk factors. Thus, both groups had similar distribution in the risk of developing AO, and the results cannot be taken by a simple imbalance in the risk to develop AO in any of the groups.
Additionally, we did not observe losses regarding follow-up or treatment withdrawals, minimising attrition bias risk. This was because we designed a protocol of four phone calls encouraging the patient to attend the clinical control, and our local community clinics allow to "retain" patients.
Studies that we compared previously used a model of impacted mandibular third molars. We consider that this model does not differ from ours, in which we included the extraction of any tooth in the mouth. The only difference is that in the extraction of impacted third molars, the risk factor "traumatic extraction" is exacerbated, which explains why the prevalence of AO in these studies is higher. In addition, we controlled the tooth location, which gives us the advantage that our results can be extrapolated to a wider population, and not be limited to only those procedures with a high level of tissue damage (i.e. mandibular third molar extraction).
Nonetheless, this study has some weaknesses.
Firstly, it is difficult to ensure that patients have  followed the full treatment, because they administered themselves in their homes. To reduce the risk of poor treatment procedure, we phoned the patients once a day for the first 4 d to remind them to use the rinse.
Additionally, we required them to bring the bottles of mouthwash to the dental control, to ensure that it was fully used. If this had not been the case, the patient would have been discarded from the follow-up.
We used sterile water as placebo, which could have been a risk if the patient recognized it as such, but in the reinforcement phone calls, we did not detect this situation in any patient. Secondly, the operational definition of smokers was changed in relation to our previous study 8 (i.e. consumption of ≥5 cigarettes 24 h after extraction, to consumption of ≥5 cigarettes 24 h before extraction). To include the patient as a smoker we had to know if they smoked previously.
It has been proposed that intra-alveolar application of chlorhexidine 0.2% gel may be an effective treatment to prevent AO 11,20,21 . However, the evidence supporting this preventive therapy has the same inconsistencies and weaknesses as the studies of chlorhexidine 0.12% mouthwash 6 . There is no biological plausibility that this form of treatment can be more effective, because a concentration of 0.2% shows no additional antibacterial benefits than the concentration of 0.12% 3 , and the substantivity of the chlorhexidine mouthwash is sufficient to maintain its effect for 12 h 4 . There is no evidence that chlorhexidine may have a negative effect on haemostasis, and the mode of use as a mouthwash does not interfere with clot formation, since our application protocol begins 24 h after the extraction, when the clot has already been stabilized and begins to be replaced with granulation tissue 1 .
With the strengths and weaknesses of our study, we conclude that the use of chlorhexidine 0.12% mouthwash after tooth extraction is highly effective compared with placebo mouthwash in preventing AO in high-risk patients. Clinical trials evaluating preventive effects should consider the risk factors in the study design to minimise risk of bias.