Prevention of alveolar osteitis with chlorhexidine: A meta-analytic review

Prevention of alveolar osteitis with chlorhexidine: A meta-analytic review Antonio Caso, DDS,a Li-Kuei Hung,b and O. Ross Beirne, DMD, PhD,c Seattle, Wash UNIVERSITY OF WASHINGTON

Objective. The objective of this study was to assess if chlorhexidine (CHX) rinse decreases the occurrence of alveolar osteitis (AO) following third molar removal. Study design. A literature search identified 7 randomized prospective clinical trials reporting incidence of AO following removal of mandibular third molars. Studies were combined into 2 groups and summary relative risks were calculated for each group. One group of studies evaluated rinsing on the day of surgery only and the second group of studies rinsed at least on the day of surgery and several days after surgery. Results. The relative risk for the single rinse group was 1.36 (95% confidence interval [CI] 0.80, 2.33), P [ .05, whereas for the multiple rinse group, the relative risk was 1.90 (95% CI 1.46, 2.47), P \ .05. Conclusion. Rinsing with CHX on the day of surgery and several days after may reduce the incidence of AO. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2005;99:155-9)

The reported incidence of alveolar osteitis (dry socket) following third molar extractions ranges from 1% to 30%.1 Alveolar osteitis (AO) can be a burden for both the patient and the surgeon. This complication may cause throbbing pain and loss of productivity because the patient typically requires multiple visits to the surgeon’s office. It can also be costly for the surgeon who will need to take clinic time to manage the patient’s symptoms. AO usually manifests itself 2 to 5 days after the surgery. The patient may experience fetid breath, and persistent and radiating pain that is not easily relieved by analgesics. The extraction socket may contain necrotic debris with absence of a blood clot and denuded alveolar bone. The etiology of AO is not firmly established, however the risk factors that increase the frequency of this complication include smoking,2 oral contraceptives,3 experience of the surgeon,2,4 and poor oral hygiene.4 There is evidence that bacteria in the wound contribute to fibrinolysis and subsequent loss of the blood clot.5 Many studies have focused on decreasing the frequency of AO by reducing the bacterial count in the oral cavity. Use of systemic,6,7 topical,8 and a This study was funded in part by the University of Washington Dental Alumni Association. a Chief Resident, Department of Oral and Maxillofacial Surgery, University of Washington, Seattle, Wash. b Dental Student, University of Washington, School of Dentistry, Seattle, Wash. c Professor and Chair, Department of Oral and Maxillofacial Surgery, University of Washington, Seattle, Wash. Received for publication Nov 20, 2003; returned for revision Feb 3, 2004; accepted for publication May 5, 2004. Available online 11 September 2004. 1079-2104/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.tripleo.2004.05.009

combination of antimicrobials9 has been reported in the literature. Chlorhexidine (CHX) rinses have been used to prevent AO with variable outcomes. The objective of this study was to evaluate the effect of CHX rinsing on the rate of AO following third molar removal. A systematic review of the literature was done to determine if the clinical evidence supports the use of CHX rinses to reduce AO. Valid randomized controlled clinical trials were identified using specific inclusion criteria and the data from the identified reports were reviewed and statistically compared. MATERIALS AND METHODS An exhaustive search of all the English and nonEnglish literature was performed using MEDLINE with the following keywords: chlorhexidine, third molar(s), and extraction(s). The initial search identified 12 studies. Using Science Citations and hand searching the references of the12 studies, another 578 reports were identified. Three examiners independently screened the abstracts of all the articles and selected 16 human clinical trials that evaluated the effect of CHX rinsing on AO. Studies selected for detailed analysis met the following criteria: d d

d

d

They were human clinical trials. They evaluated the effect of CHX rinse on the incidence of AO following extraction of mandibular third molars. They randomly assigned patients to control and CHX rinse groups. They used similar criteria to diagnose AO.

Seven studies met the criteria for detailed analysis.9-15 All 7 studies were randomized prospective clinical trials 155

OOOOE February 2005

156 Caso, Hung, and Beirne Table I. Studies not included in the meta-analysis and reason for exclusion Study

Reason for exclusion 16

Bonine Bragger et al17 Field et al18 Fotos et al19 Garibaldi et al20 Legarth et al21 Lilly et al3 MacGregor and Hart22 Wilhelm et al23

Not randomized Periodontal tissue, not third molars Not third molars Implanted into socket Treatment of dry socket Not randomized Not CHX Not dry socket No outcomes

CHX, chlorhexidine.

that reported the incidence of AO following removal of mandibular third molars. The reasons for excluding 9 of the studies are shown in Table I.3,16-23 The outcome measure reported in the 7 studies was the incidence of AO. Relative risks were calculated by dividing the percent incidence of AO in the control group by the percent incidence in the test group. Relative risks were calculated for each individual study and data from similar studies were combined to produce summary relative risks. The 95% confidence intervals (95% CI) were calculated and used to determine statistical significance. Because of differences in the protocols for rinsing with CHX, the 7 studies were separated into 2 groups that were examined separately. Group 1 included 2 studies that evaluated patients who rinsed with CHX only on the day of surgery and Group 2 included 5 studies that evaluated patients who used multiple rinses with CHX. The protocols for multiple CHX rinses were not identical. However, multiple rinses included at least 1 rinse immediately before removing the teeth followed by several days of rinsing after surgery. RESULTS The 7 studies that met the inclusion criteria represented a 25-year period from 1977 to 2002 (Table II). All studies were randomized, prospective clinical trials that reported outcomes for AO following removal of mandibular third molars. The methods of CHX administration and sample size for each study are described in Table II. Of note, Berwick and Lessin10 evaluated 2 different CHX protocols. These 2 different protocols are identified as Berwick and Lessin #1 and #2 in Table II. Neither protocol detected a reduction in AO with the use of CHX. Both protocols were examined separately and combined in the single CHX rinse group. Relative risks for each study were calculated and statistical significance of differences was determined using the 95% CI. Hermesch et al,12 Larsen,13 and Ragno and Szkutnik14 observed a statistically signifi-

cant difference in the incidence of AO between the patients who rinsed with CHX and the control patients. However, Berwick and Lessin,10 Krekmanov and Nordenram,9 Delilbasi et al,11 and Tjernberg15 did not observe a significant difference in the incidence of dry socket between the control and CHX groups. Fig 1 shows the relative risks and 95% CI intervals calculated for each individual study. Data synthesis was done for each of the 2 groups separately (Table II). The studies were weighted using sample size. The results from the Berwick and Lessin10 and the Krekmanov and Nordenram9 studies were combined in Group 1—rinse on the day of surgery only—and the results from Delilbasi et al,11 Hermesch et al,12 Larsen,13 Ragno and Szkutnik,14 and Tjernberg15 were combined in Group 2—CHX rinse at least on the day of surgery and several days following removal of the teeth. Data from the 7 studies were reported by number of patients, extraction sites, or both. Because Berwick and Lessin10 and Larsen13 reported outcomes using only extraction sites, the incidence of dry socket by extraction site was used for the combination of the data. The relative risk for the individual studies as well as the combined studies for Group 1 showed no significant benefit from using a single rinse with CHX (Figs 1 and 2). The combined relative risk for Group 1 was 1.36 (95% CI 0.80, 2.33), P [.05 (Fig 2). Three of the 5 studies in Group 2 showed that multiple rinses with CHX significantly reduced the incidence of AO (Fig 1). The combined relative risk for all 5 studies in Group 2 was 1.90 (95% CI 1.46, 2.47), P \ .05 (Fig 2). Two reviewers independently carried out a quality assessment of the clinical trials in Group 2. Quality was evaluated for the multiple CHX trials using the following criteria: patient stratification, precise definition of AO, masked taste of placebo, and blinded evaluation for AO (Table III). The 3 studies that demonstrated that multiple rinses with CHX reduced AO met all 4 of the quality criteria (Hermisch et al,12 Larsen,13 and Ragno and Szkutnik14) while the 2 studies that did not demonstrate an effect for CHX did not meet all 4 quality criteria. Combination of these 3 studies yielded a relative risk of 1.94 (95% CI 1.64, 2.29), P \ .05, which is consistent with the results from combining all 5 studies. Because extraction sites in the same patients are not independent variables, data analysis based on extraction site may confound the result in Group 2. Larsen was the only study in Group 2 that did not report AO using patients. To test the influence of Larsen’s study on the combined data, the 4 studies not including Larsen’s study were used to calculate the combined weighted relative ratio using incidence of AO in patients instead of extraction sites. The relative ratio for the 4 studies

OOOOE Volume 99, Number 2

Caso, Hung, and Beirne 157

Table II. Studies included in the meta-analysis Study Group 1 Berwick and Lessin #110 Berwick and Lessin #210 Krekmanov and Nordenram9 Group 2 Delilbasi et al11 Hermesch et al12 Larsen13 Ragno and Szkutnik14 Tjernberg15

Method of CHX administration

CHX sample size (patients)

Control sample size (patients)

CHX sample size (extractions)

Control sample size (extractions)

Day of surgery rinse, intraop normal saline irrigation Day of surgery rinse, intraop CHX irrigation Day of surgery rinse only

20

20

40

38

20

20

39

38

37

36

37

36

Day of surgery rinse, 1 wk after surgery Rinse 1 wk before, 1 wk after surgery Rinse 1 wk before, 1 wk after surgery Day of surgery rinse, 1 wk after surgery Rinse 5 days before, 1 wk after surgery

62

59

62

59

136

135

239

240

73

67

144

134

40

40

80

80

30

30

30

30

CHX, chlorhexidine.

Fig 2. Combined calculated relative risk with 95% confidence intervals for alveolar osteitis. Fig 1. Individual calculated relative risk with 95% confidence intervals for alveolar osteitis.

was 1.35 (95% CI 1.13,1.60), P \ .05. The combined results using only the 4 studies that report incidence of AO in patients still demonstrated that multiple rinses with CHX reduced the incidence of AO. Publication bias can have a significant influence on the conclusions from meta-analytic or systematic reviews. A funnel plot can sometimes reveal if there is more frequent publication of studies with only positive or negative outcomes.24 Publication bias for the multiple rinse group is difficult to eliminate but funnel plots were done separately using the standard error and number of patients participating in the study. There were only 5 points on the funnel plots but there did not appear to be significant asymmetry in the plot. This is consistent with the hypothesis that there was no publication bias for reporting of the outcomes of studies evaluating the use of multiple CHX rinses.

Table III. Quality evaluation of multirinse studies

Study Delilbasi et al11 Hermesch et al12 Larsen13 Ragno and Szkutnik14 Tjernberg15

Precise Masked Blinded Patient definition taste of evaluation stratification of AO placebo of AO X X X

X X X X X

X X X

X X X X ?

AO, Alveolar osteitis.

DISCUSSION The studies in Group 1 analyzed individually and together did not detect a reduction in AO with a single rinse with CHX on the day of surgery. In Group 2, 3 out of 5 studies showed that multiple rinses with CHX significantly reduced the incidence of AO. When the 2 studies that did not observe a reduction in AO with multiple CHX rinses were combined with the 3 that did

OOOOE February 2005

158 Caso, Hung, and Beirne Table IV. Criteria for alveolar osteitis Study Berwick and Lessin10 Delilbasi et al11 Hermesch et al12 Krekmanov and Nordenram9 Larsen13 Ragno and Szkutnik14 Tjernberg15

Denuded socket

Necrotic debris

Fetid breath

X X X

X X X

X

X

X X X

X X X

detect a reduction, multiple rinses with CHX still significantly reduced the incidence of dry socket. Based on an assessment of quality, the 3 studies that detected a reduction in AO were the only studies that met all 4 quality criteria. The combined sample population for Group 1 was 153 patients and for Group 2 was 671 patients. The sample population in Group 1 was low and may have been inadequate to reveal an effect of CHX on AO. However, individual and combined examination of the study outcomes from Group 1 gave the same results—CHX administration once on the day of surgery had no significant benefit in the prevention of AO. Selecting studies that have similar protocols for rinsing with CHX and diagnosing AO posed a challenge in our study. Although the 7 studies were divided into 2 groups based on similarity of CHX administration, the methods within the 2 groups were not identical. For instance, in the single rinse group, Berwick and Lessin10 incorporated intraoperative lavage with both CHX and saline. Krekmanov and Nordenram9 included cleaning of the teeth and rinsing of gingival pockets in addition to a preoperative CHX rinse. As shown in Table II, there were differences in CHX administration within Group 2 as well. In addition to difference in CHX rinsing, the criteria used to diagnose AO were not exactly the same for each study (Table IV). Even though the use of CHX was not identical for each study, we were able to divide the studies identified for detailed analysis into 2 groups based on the number of CHX rinses. One group examined the efficacy of a single CHX rinse and the second group examined the efficacy of multiple CHX rinses for the prevention of AO. In addition to differences in CHX administration and criteria for the diagnosis of AO, there were other differences in study design between the individual articles. Studies varied in their criteria for patient selection, presurgical patient preparation, intraoperative medications, and postoperative analgesia regimes. For example, Delilbasi et al11 excluded all patients from their study who were taking oral contraceptives. Tjernberg15 saw patients in the test group 5 days prior to tooth removal for scaling and plaque removal and

Loss of blood clot

Pain not relieved by analgesics

X X X X X

X X X X X X

Trismus

Paper point test for blood

X

X

Larsen13 administered dexamethasone intra-operatively to all patients. These are confounding variables and are acknowledged as inherent limitations of this metaanalytic review. Despite the fact that not all of the individual studies in Group 2 observed a benefit from multiple rinses with CHX, the combined analysis of the 5 studies supports the results of Hermesch et al,12 Larsen,13 and Ragno and Szkutnik.14 These 3 studies ranked the highest on the qualitative analysis. Tjernberg15 may not have observed a significant reduction in AO with multiple uses of CHX because of the small sample size and low incidence of AO. He evaluated only 60 patients and his subsequent confidence intervals were relatively wide. Tjernberg also had a very restrictive definition of AO, which included checking for fresh blood with a paper point. Only 1 patient in his test group was reported to have developed AO using these criteria. CONCLUSION It could not be determined if a single rinse with CHX on the day of surgery significantly reduced the incidence of AO. However, the results of this study support the conclusion that rinsing at least on the day of surgery and for several days following the removal of the teeth reduces the incidence of AO associated with the extraction of mandibular third molars. The minimum number of days of rinsing needed to produce this effect could not be determined by this analysis. Further studies using standard definition for AO, stratification of patients by gender, difficulty of extraction, smoking, etc. are needed to determine the best use for CHX rinses in patients having mandibular third molar removed. REFERENCES 1. Report of a workshop on the management of patients with third molar teeth. J Oral Maxillofac Surg 1994;52:1102-12. 2. Larsen PE. Alveolar osteitis after surgical removal of impacted mandibular third molars. Identification of the patient at risk. Oral Surg Oral Med Oral Pathol 1992;73(4):393-7. 3. Lilly GE, Osbon DB, Rael EM, Samuels HS, Jones JC. Alveolar osteitis associated with mandibular third molar extractions. J Am Dent Assoc 1974;88(4):802-6. 4. Sisk AL, Hammer WB, Shelton DW, Joy ED Jr. Complications following removal of impacted third molars: the role of the

OOOOE Volume 99, Number 2

5.

6. 7. 8. 9. 10.

11.

12.

13. 14. 15. 16.

experience of the surgeon. J Oral Maxillofac Surg 1997;44(11): 855-9. Blum IR. Contemporary views on dry socket (alveolar osteitis): a clinical appraisal of standardization, aetiopathogenesis and management: a critical review. Int J Oral Maxillofac Surg 2002; 31(3):309-17. Mitchell DA. A controlled clinical trial of prophylactic tinidazole for chemoprophylaxis in third molar surgery. Br Dent J 1986;160(8):284-6. Rood JP, Murgatroyd J. Metronidazole in the prevention of Ôdry socket.Õ Br J Oral Surg 1979;17(1):62-70. Hall HD, Bildman BS, Hand CD. Prevention of dry socket by local application of tetracycline. J Oral Surg 1971;29(1):35-7. Krekmanov L, Nordenram A. Postoperative complications after surgical removal of mandibular third molars. Effects of penicillin V and chlorhexidine. Int J Oral Maxillofac Surg 1986;15(1):25-9. Berwick JE, Lessin ME. Effects of a chlorhexidine gluconate oral rinse on the incidence of alveolar osteitis in mandibular third molar surgery. J Oral Maxillofac Surg 1990;48(5):444-8; discussion 449. Delilbasi C, Saracoglu SH, Kesken A. Effects of 0.2% chlorhexidine gluconate and amoxicillin plus clavulanic acid on the prevention of alveolar osteitis following mandibular third molar extractions. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2002;94(3):301-4. Hermesch CB, Hilton TJ, Biesbrock AR, Baker RA, Cain-Hamlin J, McClanahan SF, et al. Perioperative use of 0.12% chlorhexidine gluconate for the prevention of alveolar osteitis: efficacy and risk factor analysis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998;85(4):381-7. Larsen PE. Use of chlorhexidine to prevent alveolar osteitis. J Oral Maxillofac Surg 1990;48(11):1244-5. Ragno JR Jr, Szkutnik AJ. Evaluation of 0.12% chlorhexidine rinse on the prevention of alveolar osteitis. Oral Surg Oral Med Oral Pathol 1991;72(5):524-6. Tjernberg A. Influence of oral hygiene measures on the development of alveolitis sicca dolorosa after surgical removal of mandibular third molars. Int J Oral Surg 1979;8(6):430-4. Bonine FL. Effect of chlorhexidine rinse on the incidence of dry socket in impacted mandibular third molar extraction sites.

Caso, Hung, and Beirne 159

17.

18. 19.

20. 21. 22. 23. 24.

Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995;79(2): 154-7; discussion 157-8. Bragger U, Schild U, Lang NP. Effect of chlorhexidine (0.12%) rinses on periodontal tissue healing after tooth extraction. (II). Radiographic parameters. J Clin Periodontol 1994;21(6): 422-30. Field EA, Nind D, Varga E, Martin MV. The effect of chlorhexidine irrigation on the incidence of dry socket: a pilot study. Br J Oral Maxillofac Surg 1988;26(5):395-401. Fotos PG, Koorbusch GF, Sarasin DS, Kist RJ. Evaluation of intra-alveolar chlorhexidine dressings after removal of impacted mandibular third molars. Oral Surg Oral Med Oral Pathol 1992; 73(3):383-8. Garibaldi JA, Greenlaw J, Choi J, Fotovatjah M. Treatment of post-operative pain. J Calif Dent Assoc 1995;23(4):71-2, 74. Legarth J, Munster R, Swendsen J. Effect of chlorhexidine on development of dry socket after surgical removal of mandibular third molars. Tandlaegebladet 1977;81:451-3. MacGregor AJ, Hart P. The topical effect of chlorhexidine on third molar wounds and their complications. J Oral Surg 1971; 29:481-5. Wilhelm RJ, Sutley SH, Quigley NC. Antimicrobial management of third molars: survey results for military dentists. Gen Dent 1996;44(6):538-43. Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997; 315:629-34.

Reprint requests: O. Ross Beirne, DMD, PhD University of Washington School of Dentistry Department of Oral and Maxillofacial Surgery Box 357134 1959 NE Pacific St. Seattle, WA 98195-7134 [email protected]