Chronic periodontitis is an infectious disease of the oral cavity that causes progressive destruction of periodontal tissues, leading to structural changes like attachment loss, bone resorption, resulting in bony defects, and potential tooth loss if left untreated. Effective drugs, such as alendronate, rosuvastatin (RSV), atorvastatin, melatonin, and metformin (MF), have been used as adjuncts to scaling and root planning and require evaluation for their comparative effectiveness in treating bony defects in patients with chronic periodontitis. This study aims to compare the effectiveness of these drugs for treating such defects.
This network meta-analysis (NMA) was conducted following Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines and registered in PROSPERO (CRD42024600432). A comprehensive search of PubMed, Scopus, and Cochrane Library identified 11 eligible randomized clinical trials reporting changes in clinical attachment level (CAL) and bone fill (BF) at 6 months posttreatment. The NMA systematically compared treatment outcomes across different intervention groups.
MF was the most effective treatment for CAL and BF at 6 months. Ranking probabilities indicated that MF and RSV had the highest likelihood of being the most effective treatments.
These findings from the NMA suggest that MF may be an effective option for CAL improvement and BF. Further research is needed to validate these results and optimize treatment strategies for bony defects in chronic periodontitis.
Periodontitis, a chronic inflammatory disease of the supporting structures of the teeth, affects approximately 47% of U. S. adults aged 30 years and above (Eke
According to the American Academy of Periodontology, an intrabony defect is defined as a “periodontal defect within the bone surrounded by one, two, or three bony walls, or a combination thereof.”[
The goal of an effective periodontal treatment is to restore both the structural integrity and functional capacity of the affected periodontium. While scaling and root planning (SRP) is the primary treatment for periodontal disease, it may not fully eliminate pathogens in deep intrabony and interradicular defects, allowing infection to persist. To enhance clinical outcomes, adjunctive therapies such as local drug delivery (LDD) systems are often used in conjunction with SRP. Various pharmacological agents have been investigated for this purpose, including alendronate (ALN), rosuvastatin (RSV), atorvastatin (ATV), melatonin (ML), and metformin (MF), each offering distinct mechanisms of action that modulate the host response in periodontitis.
The host response in periodontal disease, while protective, can also cause tissue damage and bone resorption. Bisphosphonates such as ALN inhibit osteoclast-mediated bone resorption by disrupting the RANK/RANKL/OPG signaling pathway, which is essential for osteoclast differentiation and activation. ALN binds to hydroxyapatite in alveolar bone and is internalized by osteoclasts, where it inhibits farnesyl pyrophosphate synthase in the mevalonate pathway, leading to impaired GTPase prenylation and osteoclast apoptosis. It also reduces RANKL expression and enhances osteoblast activity, thereby promoting bone formation and maintaining alveolar bone integrity.[
Statins (e.g., RSV and ATV) are the competitive inhibitors of HMG-CoA reductase, primarily used for lipid-lowering therapy. They significantly reduce serum cholesterol levels and thereby lowers the risk of cardiovascular diseases. Beyond their lipid-lowering effects, they also exhibit notable anti-inflammatory properties and promote osteoblastic differentiation, and increase alkaline phosphatase activity, a recognized marker of osteoblastic function, indicating a potential role in bone health and regeneration.[
The addition of these pharmacological agents with host modulatory and potential antimicrobial effects as adjuncts notably enhanced periodontal status following nonsurgical periodontal treatment when compared to SRP alone.[
Earlier meta-analyses have been carried out to identify the therapy with the highest efficacy using direct evidence. However, these analyses were limited by their narrow scope of comparisons and inadequate statistical power, primarily due to the small number of head-to-head trials available. In addition, they were unable to compare multiple interventions simultaneously or incorporate indirect comparisons, which restricted the comprehensiveness of their findings. Based on this background, conducting a network meta-analysis (NMA) to assess efficacy outcomes holds significant clinical relevance. NMA in oral health research has been implemented as a method capable of integrating both direct and indirect comparisons among the studies included, the latter being comparisons not directly conducted within individual trials.[
The manuscript of this NMA has been prepared following the Cochrane Collaboration guidelines[
The goal of this review was to address the following focused questions regarding the use of 1% ALN, 1% RSV, 1.2% ATV, 1% ML, and 1% MF as locally delivered adjunctive drug agents in chronic periodontitis:
What is the comparative efficacy of these agents in improving clinical attachment levels (CALs) in chronic periodontitis patients? How do these agents compare in promoting bone fill (BF) at 6 months post-treatment?
The following PICOT framework was used to guide the inclusion and exclusion of studies for the aforementioned focused questions.[
Population (P): Patients undergoing treatment for chronic periodontitis exhibiting intrabony or interradicular defects Intervention (I): Use of locally delivered adjunctive drugs (e. g., ALN, RSV, ATV, ML, and MF) in conjunction with SRP Comparison (C): Adjunctive drugs were compared with each other and with placebo gel Outcome (O): Improvement in clinical and radiographic parameters, including BF and gain in CAL around the treated teeth Time (T): Follow-up duration of 6 months’ posttreatment.
The PubMed/MEDLINE, Wiley Online Library, Google Scholar, Scopus, Embase, and Cochrane Central Register of Controlled Trials (CENTRAL) databases were searched up to October 2023, with articles taken after publication year 2016. The outcomes of interest were CAL and BF reported at 6 month post-treatment.
Three authors performed a literature search of titles and abstracts relevant to the PICOT question across the following databases: PubMed/MEDLINE, Wiley Online Library, and Google Scholar and included articles published after 2016. A combination of keywords, Mesh terms, and Boolean operators (AND, OR, and NOT) were used in the search.
The terms included:
(1) Periodontitis AND Alendronate/Rosuvastatin/Atorvastatin/Metformin/Melatonin (2) Alendronate/Rosuvastatin/Atorvastatin/Metformin/Melatonin AND Adjunctive periodontal therapy (3) Alendronate/Rosuvastatin/Atorvastatin/Metformin/Melatonin AND interradicular defects OR furcation involvement (4) Alendronate/Rosuvastatin/Atorvastatin/Metformin/Melatonin AND Intrabony defects.
The following key terms were utilized for searching the remaining electronic databases.
bone regeneration; NMA; alendronate; metformin; melatonin; rosuvastatin; atorvastatin; synonyms for these terms.
(1) Randomized clinical trials (RCTs) performed on humans (2) Patients presenting with Grade II furcation and intrabony defects confirmed by radiographic and clinical evidence and treated with locally delivered ALN, RSV, ATV, ML, MF, and placebo with articles after publication year 2016 (3) Articles with a follow-up period of 6 months (4) Only studies published in English were included (5) Studies reporting clinical parameters: CAL and BF.
(1) (2) Non-English (3) Animal studies (4) Case reports (5) Case series (6) Reviews (7) Conference abstracts (8) Patients undergoing systemic treatment with any of the following agents: ALN, RSV, ATV, MF, or ML.
For clarity and ease of interpretation in network geometry and result presentation, each drug intervention included in this review was assigned a group label. These labels were consistently used in subsequent network diagrams and statistical analyses.
To facilitate clarity in the presentation of network geometry and treatment comparisons, each drug intervention was assigned a group label. ALN was designated as Group A, RSV as Group B, ATV as Group C, ML as Group D, MF as Group E, and Placebo as Group F. These group labels were used consistently throughout the NMA for ease of interpretation in figures and statistical comparisons.
Essential information regarding title, authors, published year, interventions, comparators, time periods, and no. of studies was extracted. The outcomes of interest were CAL, BF. Data extraction was performed independently by two authors (S. K. and S. R.), with discrepancies resolved through discussion with a third reviewer (P. B). CAL was assessed using a periodontal probe, typically a UNC-15 probe, by measuring the distance from the cementoenamel junction to the base of the periodontal pocket, while BF was evaluated radiographically, with several studies employing cone-beam computed tomography for quantitative analysis at baseline and 6-month follow-up period.
Two independent observers independently scanned the abstracts and later the preselected full-text articles.
For the risk of bias across studies:
The included studies were evaluated for bias following the Cochrane Handbook of Systematic Reviews (SR) guidelines[
A frequentist random-effects model was used for the NMA to estimate relative treatment effects and generate SUCRA values. A network plot was constructed to depict relationships among treatment methods, with nodes representing treatments and connecting lines indicating direct comparisons. Node size reflected the number of studies, and line thickness denoted data volume for each comparison. Inconsistency between direct and indirect evidence was assessed using overall inconsistency and node-splitting analyses; a
Following an extensive electronic search, a total of 151 articles were identified, specifically 78 from Pubmed/ Medline, 26 from Wiley Online Library and 47 from Google Scholar. After removing 53 duplicates and excluding 98 articles for other reasons, 45 records were screened on the basis of titles and abstracts. Full-text assessment was performed on 24 articles based on the inclusion criteria and 11 articles were finally selected for this present NMA [
Flowchart (Preferred Reporting Items for Systematic Reviews and Meta-analyses format) of the screening and selection process.
The included studies compared ALN vs placebo (
Study characteristics
Eleven of the included RCTs were considered to have a low risk of bias,[
Risk of Bias Assessment in Studies on the Efficacy of Pharmacological Treatments for Bony Defects in Chronic Periodontitis.
Pooled estimates, individual study outcomes, and clinical recommendation for regenerative procedures (RPs).
A total of 11 eligible RCTs were included in the study. The drugs were categorized as A, B, C, and so on. Networks were created for each outcome: CAL and BF which included only adjunctive drugs compared with other drugs or placebo.
We conducted a NMA comparing the effectiveness of 6 treatments for CAL and BF using mean differences observed at 6 months of treatments.
Network Plot for (i) clinical attachment level (II) bone fill The size of the six nodes, each representing a treatment, reflects the number of studies associated with that treatment, while the thickness of the lines connecting two nodes represents the volume of relevant data for those comparisons. (a) Alendronate; (b) Rosuvastatin; (c) Atorvastatin; (d) Melatonin; (e) Metformin; (f) Placebo.
To explore the loop inconsistency, we fit the side-splitting models.
Inconsistency test assessing agreement between direct and indirect treatment comparisons for clinical attachment level
Using the frequentist consistency model, MF, ALN, RSV, and ATV demonstrated statistically significant improvements in CAL, with
Results under multivariate meta-analysis for clinical attachment level
Network Forest Plot of Treatment Effects with Pooled Estimates for (i) clinical attachment level; (II) bone fill (a) Alendronate; (b) Rosuvastatin; (c) Atorvastatin; (d) Melatonin; (e) Metformin; (f) Placebo; Study 5: Sahu Ipshita 25; Study 6: Vidushi Sheokand26; Study 9: Avani R. Pradeep27; Study 10: Nitesh Kumar Sharma33; Study 13: Debopriya Chatterjee28; Study 14: A R Pradeep29; Study 15: A R Pradeep30; Study 20: A R Pradeep32; Study 21: Dipika Mitra34; Study 22: Dileep P35; Study 24: Noopur P. Go31.
Treatments were ranked after the NMA using estimated probabilities (%) of each treatment achieving each rank, assuming the maximum parameter indicates the best performance. The analysis was based on 10,000 draws, accounting for parameter uncertainty.
Effect sizes and SUCRA values for treatments based on network meta-analysis for clinical attachment level; bone fill
Cumulative rank probability of Treatments: (i) clinical attachment level, (II) bone fill.
Similar to CAL, loop inconsistency in BF was also explored by fitting side-splitting models.
Inconsistency test assessing agreement between direct and indirect treatment comparisons for bone fill
Under the frequentist consistency model, MF, ALN, RSV, and ATV yielded statistically significant improvements in BF, with
Results under multivariate meta-analysis for bone fill
Overall, MF emerges as the top choice, providing the most promising outcomes based on effect sizes and SUCRA values.
SRP remains the gold-standard treatment for chronic periodontitis. However, when used as an adjunct, pharmacologic agents play a crucial role in enhancing CAL gain and BF, thereby improving the management of bony defects. In addition, there are no previous studies utilizing a NMA to evaluate the efficacy and comparative ranking of these agents in periodontal therapy, highlighting the need for further research in this area.
CAL and BF are the pivotal indicators of periodontal health and treatment success. CAL reflects the structural integrity of the periodontal attachment, while BF signifies the regeneration of the underlying bone. Both parameters are crucial for restoring functional capacity, minimizing disease progression, and improving long-term tooth retention. All the studies share the common conclusion that pharmacologic agents are effective in improving CAL and BF parameters compared to placebo, making them valuable adjuncts in the treatment of chronic periodontitis.
Our study employed a mixed-model NMA incorporating both direct and indirect comparisons. This approach enabled treatment ranking based on clinical and radiographic outcomes, with indirect comparisons enhancing the robustness of the findings. The review adhered to Cochrane Collaboration guidelines and the PRISMA-NMA framework, systematically identifying 11 RCTs published after 2016. Risk of bias was assessed using Cochrane guidelines to ensure study reliability.
The evaluation of soft (CAL) and hard tissue (BF) parameters confirmed MF was most effective for CAL gain and BF. These findings highlight the targeted regenerative potential of various pharmacological agents in periodontal therapy. We are further planning to conduct a research incorporating these pharmacologic agents alongside biologic agents such as PRF, aiming to provide a global ranking and perform both direct and indirect comparisons of their efficacy.
Most included studies had short follow-up periods, with only 6 months of data for most trials. Variations in study designs, such as differences in treatment methods (e.g., open flap debridement [OFD] vs. LDD), also contributed to inconsistencies. For example, study no. 21 on MF[
Furthermore, the limited number of included RCTs, despite involving multiple interventions, may restrict the robustness of indirect comparisons within the NMA framework.
Small sample sizes and the lack of direct comparisons among treatment groups further reduced the statistical power and generalizability of the findings.
A study by Avani R. Pradeep (2016)[
Pradeep
Similarly, a study by Dipika
However, Dileep
Alice
Another study by Wang
A study by Claudia Arena
Another study by Ru-Yeu Liu
A study by G. Cecoroet
A study by Z. Akram
The ability of these pharmacologic agents to enhance bone regeneration through their anti-inflammatory and osteogenic effects demonstrates their potential to improve CAL gain and BF. These findings contributed to the comparative analysis and ranking of these agents, providing a thorough understanding of their effectiveness in managing bony defects.
MF consistently emerged as the most effective treatment for both CAL and BF improvement, with the highest SUCRA values. RSV and ALN also demonstrated strong efficacy, ranking just below MF in both CAL and BF assessments. In addition, ML and ATV were found to enhance BF, although more studies are needed to compare its effectiveness with other LDD agents. In contrast, the placebo consistently ranked the lowest in treatment efficacy, confirming that adjunct drug therapies significantly enhance outcomes. These findings support the potential of MF, RSV, and ALN as effective adjuncts in the management of bony defects in chronic periodontitis, with MF being the preferred option based on efficacy and statistical significance.
Further research is needed to validate these findings and refine treatment approaches for managing bony defects in chronic periodontitis. Future RCTs should include larger sample sizes and extended follow-up periods to assess the long-term efficacy of pharmacologic agents.
Nil.
The authors of this manuscript declare that they have no conflicts of interest, real or perceived, financial or nonfinancial in this article.
The authors thank Dr. Vishwajeet Singh Assistant Professor, Department of Biostatistics, AIIMS Nagpur and Dr. Savitha. S. Junior Resident, Department of Community medicine, AIIMS Nagpur for their contribution in carrying out the statistical analysis.
