Simpson · The Cochrane database of systematic reviews 2022 · Systematic review and meta-analysis of randomized controlled trials · n=35 studies (3,249 participants)

Treatment of periodontitis for glycaemic control in people with diabetes mellitus.

Cited 145 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review and meta-analysis of randomized controlled trials

PubMed 35420698 · doi:10.1002/14651858.CD004714.pub4 · record verified 2026-08-28

What was done

Authors searched six bibliographic databases to September 2021 for randomized controlled trials (RCTs) lasting at least 3 months comparing subgingival instrumentation (with or without surgical or antimicrobial adjunctive therapy) against no active treatment or usual care (oral hygiene instruction or supragingival scaling) in patients with diabetes mellitus and periodontitis. The primary outcome was HbA1c level.

What was found

The review included 35 parallel-group RCTs with 3,249 participants (almost exclusively type 2 diabetes; most conducted in secondary care). Periodontal therapy reduced HbA1c compared to control: - At 3 to 4 months: absolute reduction in HbA1c of 0.43% (4.7 mmol/mol; 95% CI -0.59% to -0.28%; 30 studies, 2,443 analysed participants; moderate-certainty evidence). - At 6 months: absolute reduction in HbA1c of 0.30% (3.3 mmol/mol; 95% CI -0.52% to -0.08%; 12 studies, 1,457 participants). - At 12 months: absolute reduction in HbA1c of 0.50% (5.4 mmol/mol; 95% CI -0.55% to -0.45%; 1 study, 264 participants). Reported harms were mild or balanced between groups, but adverse events were not evaluated in most studies.

Why it matters

This Cochrane update shifts earlier conclusions, providing moderate-certainty evidence that nonsurgical periodontal treatment yields a clinically meaningful improvement in glycaemic control for individuals with diabetes.

Limits

Only 2 of 35 included studies were judged at low risk of bias (14 high risk, 19 unclear risk), though a low-risk sensitivity analysis supported the main findings. Data apply almost entirely to type 2 diabetes (only one trial included type 1 diabetes), 12-month follow-up was limited to a single trial, and most studies did not systematically evaluate adverse events, cost, or long-term diabetic complications.

Cited by