INTRODUCTION primary role of pathological bacteria forINTRODUCTION primary role of pathological bacteria for


osteitis (AO), most commonly known as ´Dry Socket’ is a postoperative painful,
debilitating condition that occurs as a complication of tooth extraction in
permanent dentition1. The
frequency of AO in dental extraction normally ranges from 3% to 4%. However,
very high incidences, from 25% to 30% of all cases, occur after extraction of
impacted mandibular third molar.2

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

symptoms of AO appear after 24 to 48 hours of surgery as severe throbbing pain, that
can last for up to three weeks.3
The alveolar socket may contain partially or totally disintegrated blood clot
with or without fetid breath.4 Based
on different theories for the causative factors , numerous techniques have been
used for its prevention such as, the use of saline mouthwashes, topical
placement of antibiotics, antiseptic rinses, anti-fibrinolytic agents,
tranquilizer dressings, occlusive dressings as well as application of
polylactic acid.5 However, as the
primary role of pathological bacteria for the development of AO has been
constantly reported in the literature, the most effective method in the prevention
of AO has been the use of agents that systemically or locally reduce the
pathological microbes at the surgical site.6-7

efficacy of systemic antibiotic as well as topical antiseptic agents have been
studied and analyzed by different researchers. For example, Ramos et al8 have performed a systematic review and
meta-analysis and concluded that systemic antibiotics significantly reduce the
risk of dry socket after extraction of lower third molar. Similarly, Reza and
Khazaei9 have reviewed and
performed a meta-analysis to determine the efficacy of 0.2% chlorhexidine bio
adhesive gel for prevention of AO incidence after extraction of lower third
molar. However, I have not found any study that makes a comparison of efficacy
of systemic antibiotic and topical antiseptic agents for the prevention of AO.
The aim of this study is to review and analyze published studies and compare
the effectiveness of systemic antibiotics and topical antiseptics such as
chlorhexidine for the prevention of AO after extraction of lower third molar.


This review is
composed of publications that were published during 2006 to 2017, reporting on systemic
antibiotics and/or topical antiseptics in the prevention of AO after extraction
of lower third molar. Only studies that were based on randomized clinical
trials and had at least one control group have been included in the review. The
literature search was carried out with questions structured in the Patient,
Intervention, Comparison and Outcome (PICO) format.10-11

Eligibility Criteria

Patient: This review
includes the studies with individuals of any age and gender, who had lower
third molar extraction, regardless of how much were the degree of impaction.
All patients were otherwise healthy, without any underlying medical problem.

Intervention: The included
studies had tested the efficacy of different systemic antibiotics, administered
through oral or parental rout as well topical antiseptics in the form of oral
rinses or bio adhesive gel. However, the only antiseptic that was evaluated was
the chlorhexidine gluconate.

Comparison: All studies were
randomized clinical trials having a control group. The controls were the
placebo in the same form and shape of active regime.  However, the controls were in the same
patients (split-mouth technique) or in different patients.

 Outcome: This review includes
those studies who studied alveolar osteitis (dry socket) or infection or both
as post-operative complication, occurring after 3 to 5 days of surgery. It is
likely that investigators tested other outcomes such as inflammatory conditions
(pain, fever, size of mouth-opening etc.) along with alveolar osteitis or
infection. However, only alveolar osteitis or infection fragment was included
in this review.


Characteristics of studies

All included
studies were randomized clinical trials having at least one control where the control
group received placebo replica of active ingredient. However, in some studies
the control group did not receive any placebo12.
In such studies, the patients or extracted tooth sockets without any placebo
were considered as control. Similarly, the control was either an another group
of identical patients or it was the same patient where corresponding tooth in
the opposite side of the mouth (studies using split-mouth technique) was considered
as control. Though majority of the studies were double blinded, in some studies
the investigator’s blinding was not mentioned. However, there was one study
that was single-blinded.13 There
existed a large variation in the type and administration of experimental
material. For instance, the route of administration of antibiotics was either
oral or parental but the route of administration of chlorhexidine was always topical,
either in the gel or in the mouth rinse form. Similarly, the timing of
administration was either pre-operative where the drug was given before surgery
or post-operative when the drug was administered after the surgery. All
included studies tested a single regimen at one time, testing either antibiotic
or chlorhexidine gluconate. However, one study14
tested the effects of 0.2% chlorhexidine gluconate as well as 0.2%
chlorhexidine gluconate in combination with amoxicillin plus clavulanic acid in
two different groups. To avoid the complication, the results of 0.2%
chlorhexidine gluconate from first group were included in this review and
results from the second group were discarded. The outcome was alveolar
osteitis. The AO was established if there was sever pain after 24 to 48 hours
of extraction with exposed bony socket. Similarly, surgical wound infection was
diagnosed if there was purulent discharge from the wound along with symptoms of
the infection such as pain, fever and lymphadenopathy. 


Statistical analysis

The extracted
data from selected studies were analyzed using statistical software R (version
3.3.2)15 and package ‘metafor.’16 The effect of treatment was analyzed with
a random effect model, using the Relative Risk (RR) as effect size of
individual studies. Treatment antibiotic or antiseptic were used as a modifier.
A funnel plot was designed to check the precision and/or existence of
publication bias of the selected studies.


Out of 31
potentially eligible full-text studies, 16 were selected for quantitative
analysis. However, some studies17,18 used
more than 1 group to check the efficacy of the treatment. In such cases, every
group was taken as a single study, thus making 20 suitable trials for
meta-analysis.  Figure 1 shows the
selected studies and the flow chart of the selection process. Fifteen studies
failed to fulfill the inclusion criteria and, therefore, were excluded. The
excluded studies and the reason for their exclusion are summarized in appendix

Analysis of efficacy

This quantitative
analysis included 2136 tooth extractions in antibiotic group and 868 tooth
extractions in antiseptic group, thus making total 3004 dental extractions for
meta-analysis. Out of 2136 dental extractions in antibiotic group, 1089 cases
received antibiotics with 53 infections (4.87%) and 1047 were in control group
with 91 infections/dry sockets (8.69%). Similarly, 439 cases were treated with
antiseptics where 43 cases developed infections/dry sockets (9.79%) whereas 101
resulted with infections out of 429 extractions in control group (23.54%). The
overall RR was 0.49, with 95% confidence interval, ranging from 0.35 to 0.68.
This was statistically significant (P < 0.0001) and different from RR=1, indicating that the overall effect of all treatment was significant and both antibiotics and antiseptics can prevent AO. The overall effect of antibiotics and antiseptics are plotted in figure 2 and results of random effect model are shown in table 2. Table 2. Random effect model for overall effect of all studies   Model Results estimate Standard Error (SE) Z-value P-value ci.ub -0.7110 0.1672 -4.2530 <.0001 -1.0386 -0.3833   Test for Heterogeneity Q (df = 19) P-value 29.5734 0.0575   Random-Effects Model (k = 20; tau^2 estimator: REML) I^2 (total heterogeneity / total variability) 32.12% H^2 (total variability / sampling variability)) 1.47     Subgroup Analysis: Types of Treatment Based on type of treatment, the 20 clinical trials were classified into two main groups, antibiotic and antiseptic. The antiseptic group included 8 clinical trials where topical antiseptic, the chlorhexidine gluconate, was tested either in the gel or liquid form.     Figure 2. Forest plot of the overall effect of antiseptics and antibiotics                The substance was topically applied either pre or post operatively. The antibiotic group comprised of 12 clinical trials, that were testing systemic antibiotics, administered via oral or parental rout either pre or post operatively. In all clinical trials either a placebo substance was used in control groups or patients were considered as control when neither an antiseptic/antibiotic nor placebo substance was used.                       Table 3.  Mixed effect model for the comparison of antiseptics and antibiotics Model Results   estimate Standard Error (SE) Z-value P-value ci.ub Intercept -0.5239 0.2242 -2.3369 0.0194     -0.9633 -0.0845 Treatment (T Antiseptic) -0.3503      0.3202 -1.0942 0.2739 -0.9778 0.2772 Test for Residual Heterogeneity QE (df = 18) P-value 25.3386 0.1159 Test of Moderators (coefficient(s) 2) QM (df = 1) P-value 1.1973 0.2739 Mixed-Effects Model (k = 20; tau^2 estimator: REML) I^2 (residual heterogeneity / unaccounted variability) 32.12% H^2 (unaccounted variability / sampling variability) 1.47 R^2 (amount of heterogeneity accounted for) 18.55%   As shown in table 3, subgroup antiseptic and antibiotic treatment yielded similar results as total effect of both treatment, i.e. both antiseptic and antibiotic are effective in reducing AO. However, when both groups were compared with each other, a statistically less significant result (P = 0.28) was obtained. It means there was no significant difference between antiseptic and antibiotic treatment regarding prevention of AO after third molar surgery.   Subgroup Analysis: Types of Control In this meta-analysis, the included clinical trials used two types of control groups. The first group (Different) comprised 13 trials where the placebo was given to different patients with similar surgeries. The second group (Same) included 7 trials where control was in the same patient. In these patients, the lower third molars were extracted using split-mouth technique. The corresponding tooth on the opposite side in same patient was considered as control. When compared with each other, a less significant result (P = 0.69) was generated. It means both control types have approximately similar effects on the analysis in the effectiveness of the treatments. The results of mixed effect model for the comparison of the control types are shown in table 4.                                                                                  Figure 3. Forest plot of the subgroup analysis of the type of treatment.                 Table 4.  Mixed effect model for the comparison of the control types   Model Results   Estimate Standard Error (SE) Z-value P-value ci.ub Intercept -0.5974 0.2672 -2.2360 0.0254 -1.1210 -0.0738 Treatment (T Antiseptic) -0.3426 0.3327 -1.0298 0.3031 -0.9947 0.3094 Control Type ( T Same) 0.1343 0.3373 0.3981 0.6905 -0.5268 0.7954 Test for Residual Heterogeneity QE (df = 17) P-value 25.0788 0.0930 Test of Moderators (coefficient(s) 2,3) QM (df = 2) P-value 1.1799 0.5544 Mixed-Effects Model (k = 20; tau^2 estimator: REML) I^2 (residual heterogeneity / unaccounted variability) 34.61% H^2 (unaccounted variability / sampling variability) 1.53 R^2 (amount of heterogeneity accounted for) 2.92%       DISCUSSION In this meta-analysis, the results from 20 randomized clinical trials were analyzed. From the analysis, it is evident that topical antiseptic (chlorhexidine gluconate) reduces the risk of dry socket/infection by 58% (RR = 42; 95%CI; 0.28-0.68) and antibiotics by 44% (RR= 56; 95% CI; 0.34-0.91). Though the infection rate in the antiseptic group was twice as high (9.79%) as the infection rate in the antibiotic group (4.87%), a similar pattern was also observed in their control groups, respectively. The infection rate was almost three times higher (23.54%) in the antiseptic controls than the antibiotic controls (8.69%). The possible reason for higher value in the control may be due to variation in the criteria used to define 'infection' in these clinical trials. However, this resulted in a comparatively lower RR in the antiseptic group compared to antibiotic group, indicating that antiseptics are relatively more effective than antibiotics. However, when both groups were compared within the mixed effect model, no significant difference (P = 0.28) was observed.  No complaints or major side effects of chlorhexidine were reported in 8 clinical trials. Only Delilbasi et al19 reported staining of oral tissues, an alteration of the taste and bad taste of the solution in around 30% patients. Similarly, the reported adverse reaction after the use of antibiotics were diarrhea, gastric pain, nausea and vaginal candidiasis20-21. However, like antiseptic group, most of the selected clinical trials in antibiotic group failed to report the associated adverse effects. Even though there is no clear evidence of adverse effects of antibiotics in the selected studies, their indiscriminate use is not free from hazards. Their extensive use is associated with antimicrobial resistance, secondary infection, allergic reaction and drug-related toxicity22. One study reported 9 cases of nausea and 21 with diarrhea after use of amoxicillin in combination with clavulanic acid23. Similarly, the common side effects of chlorhexidine gluconate were staining of tongue and dental fillings and soreness of oral mucosa.24-25 These effects are comparatively mild, local and easy to manage. Most of the post-operative complications are associated with bacterial contamination, it seems reasonable to prescribe antibiotics to prevent or reduce the incidence of AO.  On the other hand, the incidence of dry socket is relatively low and usually not life-threatening. Moreover, in this analysis some studies26-27 have not recommended the routine use antibiotics in third molar surgery. This raises the question if and how antibiotics should be used to prevent AO. The purpose this meta-analysis is to collect all available evidence to prevent the alveolar osteitis in third molar surgery. The complications associated with third molar surgery especially with AO is very painful and disabling, that affects the quality of life and productivity of the patients.28 Thus, the cost associated with these complications is much higher than the cost of antibiotic. In this scenario it seems reasonable to promote the prophylactic use of antibiotics from the cost-benefit perspective only. Moreover, the findings of this meta-analysis indicate no difference among the topical antiseptics and systemic antibiotics in the prevention of AO. Thus, it makes sense to prefer topical antiseptics over systemic antibiotics because of their mild, local and comparatively less adverse effects. Regarding other published meta-analysis, Torres et al29 concluded that systemic antibiotics can reduce the risk of post-operative complications by approximately 57%. (RR = 0.43, 95% CI, P < 0.0001). However, in current meta-analysis, the risk-reduction of AO using antibiotics was 44% (RR = 56, 95% CI, P < 0.0001), a lower efficacy observed than the efficacy of the antibiotics used in Ramos's meta-analysis. Similarly, a meta-analysis conducted by Zeitler DL30   & Flotra L et al31 have concluded that 0.2% chlorhexidine gel, placed in the tooth socket after removal of third molar, can reduce the risk of alveolar osteitis by approximately 72%. Despite a highly significant reduction of AO after use of antiseptics/antibiotics, the results of this analysis should be interpreted carefully because of the limitations regarding their generalizability. As current research has shown32-33, bacterial invasion is not the sole cause of AO. Rather it is a healing disturbance due to disintegration of blood clot at the extraction site and bacterial infection could be one of many factors that lead to AO34. Moreover, the experience of the surgeon and the complexity of the surgery due to angulation of the tooth within the alveolar bone as well as the duration of the treatment also plays an important role for the development of AO. CONCLUSION On the basis of the results presented in this analysis, it is concluded that the prophylactic use of both antiseptic and antibiotics can substantially reduce the risk of AO after extraction of lower third molar. However, these antimicrobial agents are not significantly different from each other regarding the risk-reduction of AO. On the other hand, systemic antibiotics exert more adverse effects than topical antiseptics. So, because of the potential systemic adverse effects and increasing trend of drug resistance towards antibiotics, it is recommended that the topical antiseptics should be considered as viable alternative over systemic antibiotics.