メタルポストとファイバーポスト-歯内療法を受けた歯の修復にはどちらが臨床的に優れているか？

Metal vs fibre posts - which is clinically superior for the restoration of endodontically treated teeth?

抄録

データソース Medline/PubMed, Web of Science, Scopus, Cochrane Library の各データベース。メタルポスト（MP）とファイバーポスト（FP）を直接比較した無作為化対照試験（RCT）および前向き臨床試験（PCT）を選択。試験には、歯内療法で単冠か固定式部分床義歯（ブリッジ）を装着した永久歯の患者10人以上を含み、最低1年間フォローアップした。主要アウトカムでは、FPとMPの破損率の差を比較し、サブグループ分析では、部位（前歯／後歯）、MPのタイプ（キャストポストコア／プリフォームドMP）、最も頻繁に起こる破損モード（デボンド／ルーツフラクチャー）の比較を行った。意見の相違は、3人目の査読者と協議し、コンセンサスを得た。偏りリスクはCochrane risk of bias tool（RCT）またはROBINS-I tool（PCT）を用いて評価した。回答の質は、GRADE（Grading of Recommendations Assessment, Development And Evaluation）手法で評価した。メタ分析（MA）は、Mantel-Haenszel法を用いて実施した。試験の異質性は、ランダム効果モデルおよび固定効果モデルを用いて評価した。結果 重複を排除した後、1,026件の論文を評価し、タイトルと要旨でスクリーニングを行った。このうち、全文評価を受けた論文は21件で、適格基準を満たした論文は10件であった。適格な研究には、6件のRCTと4件のPCTが含まれた。704名の参加者に対し、合計844本の歯内療法治療歯（ETT）がラジアルポストで修復された。450本のETTがFPで修復され、391本のETTがMPで修復された。参加者の平均年齢は42.7歳（33.95～55.45歳）で、平均追跡期間は50.95ヵ月（12～154ヵ月）であった。ほとんどの研究で追跡期間中の失敗が報告されていたが、MAでは失敗率に関してFPとMPの間に有意差はなかった（P = 0.39; RR: 0.82 mm; CI: 0.52-1.29 ）。サブ解析では、前方領域と後方領域で破損率に差はなく、FPとキャストポスト、コアとプレハブMPを比較した場合にも差はなかった。歯根破折とポストデボンディングが最も一般的な破損モードであったが、これらの破損の中で、FPとMPの間に差は認められなかった。結論 FPとMPの間の破損率に差があることを示す証拠は確認されなかった。これはMPのタイプやアーチ内の位置とは無関係であった。RCTとPCTの報告にはばらつきがあり、さらに質の高い研究が必要である。

Data sources Medline/PubMed, Web of Science, Scopus and Cochrane Library databases. Grey literature searches (OpenGrey, ProQuest databases), hand searches in the reference list of eligible studies and relevant journals.Study selection Randomised controlled trials (RCTs) and prospective clinical trials (PCTs) with direct comparisons between metal posts (MPs) and fibre posts (FPs). Trials contained a minimum of ten patients and endodontically treated permanent teeth that had received either single crowns or fixed partial dentures (bridges) and followed up for a minimum of one year. The primary outcome compared the difference in failure rates between FPs and MPs, with subgroup analysis comparing location (anterior/posterior), type of MP (cast post core/preformed MP) and most frequent modes of failure (debond/root fracture).Data extraction and synthesis Study selection, data collection and risk of bias assessments were completed independently by two reviewers. Disagreements were discussed with a third reviewer to reach a consensus. The risk of bias was evaluated using the Cochrane risk of bias tool (RCTs) or the ROBINS-I tool (PCTs). The quality of responses was assessed using the Grading of Recommendations Assessment, Development And Evaluation (GRADE) approach. The meta-analysis (MA) was performed using the Mantel-Haenszel method. Trial heterogeneity was assessed using random- and fixed-effects models. Inter-examiner agreement during the database search and study selection process was checked using the kappa statistic.Results After duplicates were removed, 1,026 articles were assessed and screened by title and abstract. Of these, 21 articles underwent full-text evaluation, with ten articles meeting the eligibility criteria. Eligible studies included six RCTs and four PCTs. A total of 844 endodontically treated teeth (ETT) were restored with intra-radicular posts in 704 participants. Four hundred and fifty-three ETT were restored with FPs and 391 with MPs. The mean age of participants was 42.7 (33.95-55.45) years, with a mean follow-up period of 50.95 (12-154) months. Most studies reported failures during the follow-up period, but the MA revealed no significant difference between FPs and MPs in terms of failure rates (P = 0.39; RR: 0.82 mm; CI: 0.52-1.29). Sub-analysis showed no difference in failure rates between anterior and posterior regions and no difference when comparing FPs to cast post and core vs prefabricated MPs. Root fractures and post debondings were the most common modes of failure, but within these failures, no difference was observed between FPs and MPs.Conclusions No evidence was identified for a difference in failure rates between FPs and MPs. This was independent of the type of MP and position within the arch. Reporting of RCTs and PCTs was variable and further high-quality studies are needed.