Enamel matrix derivative as adjunctive to non-surgical periodontal therapy: a systematic review and meta-analysis of randomized controlled trials

Objectives To assess the potential additional benefit of the local application of enamel matrix derivative (EMD) on the clinical outcomes following non-surgical periodontal therapy (NSPT) (steps 1 and 2 periodontal therapy). Materials and Methods A systematic literature search was performed in several electronic databases, including Medline/PubMed, Embase, The Cochrane Register of Central Trials (CENTRAL), LILACS, and grey literature. Only randomized controlled clinical trials (RCTs) were eligible for inclusion. Clinical attachment level (CAL) change (primary outcome), probing pocket depth (PPD), and bleeding on probing (BoP) reductions (secondary outcomes) were evaluated. The Cochrane Risk of Bias tool (RoB 2.0) was used to assess the quality of the included trials. Weighted mean differences (WMDs) and 95% confidence intervals (CIs) between test and control sites were estimated using a random-effect model for amount of mean CAL and PPD change. Results Six RCTs were included for the qualitative analysis, while data from 4 studies were used for meta-analysis. Overall analysis of CAL gain (3 studies) and PPD reduction (4 studies) presented WMD of 0.14 mm (p = 0.74; CI 95% − 0.66; 0.94) and 0.46 mm (p = 0.25; CI 95% − 0.33; 1.26) in favor of NSPT + EMD compared to NSPT alone respectively. Statistical heterogeneity was found to be high in both cases (I2 = 79% and 87%, respectively). Conclusions Within their limitations, the present data indicate that the local application of EMD does not lead to additional clinical benefits after 3 to 12 months when used as an adjunctive to NSPT. However, due to the high heterogeneity among the studies, additional well-designed RCTs are needed to provide further evidence on this clinical indication for the use of EMD. Clinical relevance The adjunctive use of EMD to NSPT does not seem to additionally improve the clinical outcomes obtained with NSPT alone. Supplementary Information The online version contains supplementary material available at 10.1007/s00784-022-04474-1.


Introduction
Periodontitis is a chronic disease caused by bacterial biofilm which, when left untreated, leads to the destruction of the tooth-supporting apparatus ultimately leading to tooth loss [1][2][3][4][5][6][7]. The pathogenesis of periodontitis is driven by the complex host-biofilm interactions that result in the dysbiosis of the microbiome and the dysregulation of the host immune response [1-4, 6, 7].
Non-surgical periodontal therapy (NSPT) aims to remove supra-and subgingival hard and soft bacterial deposits and has been proven to be clinically effective by leading to substantial clinical improvements evidenced by clinical attachment level (CAL) gain and probing pocket depth (PPD) reduction [8]. Nevertheless, challenging anatomical sites with impaired access, such as furcation areas [9] or deep periodontal pockets [10], are difficult to be accessed and may further serve as shelter for residual subgingival calculus and bacterial deposits. This will in turn maintain the inflammation and lead to further loss of periodontal supporting tissues [11].
Various surgical procedures have been shown to be effective in additionally reducing the residual pockets persisting after NSPT and, in certain situations, to even effectively restore intrabony [12,13] and class II furcation defects [14,15].
Since its introduction more than 25 years ago, enamel matrix derivatives (EMD) have advanced to be generally accepted as a standard in regenerative periodontal therapy [16,17]. Animal and human histological studies [16,18,19] along with numerous clinical trials have proven the ability of surgical periodontal therapy in conjunction with EMD to facilitate periodontal wound healing/regeneration and enhance the short-and long-term clinical outcomes [17]. Particularly in deep intrabony defects, open flap debridement (OFD) + EMD has shown superior results compared to OFD alone [14,15].
A number of clinical studies have also evaluated the effect of EMD used in conjunction with OFD in periodontal pockets without an intrabony component (i.e., in so-called suprabony defects). The results revealed greater CAL gains following the additional use of EMD compared with the use of OFD alone [20][21][22][23][24]. This has been very recently also corroborated by a randomized clinical trial (RCT) with a 12-month follow-up [22]. The authors reported a significant benefit of OFD + EMD over OFD alone in terms of CAL gain (3.4 ± 0.6 mm vs 1.8 ± 0.6 mm), decrease in probing pocket depth (PPD) (3.9 ± 0.6 mm vs 3.2 ± 0.6 mm), and increase in gingival recession (0.5 ± 0.7 mm vs 1.4 ± 1.0 mm) favoring the use of EMD [22].
The effects of EMD are multifaceted and have been thoroughly investigated over the last two decades [17]; apart from its anti-inflammatory properties, it has been reported that EMD positively influence wound healing, prevent or retard epithelial cell migration, promote angiogenesis, and enhance proliferation of periodontal and osteoprogenitor cells and of fibroblasts [17].
When EMD were used in conjunction with NSPT, earlier studies have failed to reveal additional PPD reduction [25,26]. However, Wennström and Lindhe [26] observed a greater reduction of gingival inflammation (i.e., less bleeding on probing) following the use of EMD in conjunction with NSPT compared to the control group (i.e., SRP alone) after 3 weeks. More recently, 3 RCTs comparing NSPT with or without subgingival application of EMD have shown some additional clinical improvements following the use of EMD as compared to the control NSPT alone [27][28][29]. Even though the use of EMD in conjunction with NSPT appears to offer some interesting perspectives for the clinician, at present it is still unclear to what extent this new clinical indication is also supported by data from RCTs. According to the best of our knowledge, at present no systematic review has assessed the literature and pooled the effect estimates on the additional use of EMD to NSPT. Therefore, the aim of the present systematic review was to assess the potential additional benefit of the local application of EMD on the clinical outcomes following NSPT.

Registration of the study protocol
The study protocol was submitted to the PROSPERO international prospective register of systematic reviews hosted by the National Institute for Health Research (NIHR), University of York, UK, Center for Reviews and Dissemination and was allocated the identification number CRD42021258154.

Reporting format
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) were adopted throughout the process of the present systematic review [30].
Patient-related outcome measures (PROMs) such as pain, satisfaction, discomfort, quality of life indicators, and economic factors.

Study design
Randomized controlled trials (RCTs) with either splitmouth or parallel-arm designs were considered eligible for inclusion.

Focused questions
The following focused questions were adapted using the PICOS criteria [32]: • In patients with untreated periodontitis, does single delivery of EMD provide adjunctive effects on CAL change compared with non-surgical mechanical instrumentation alone? • In patients with untreated periodontitis, does single delivery of EMD provide adjunctive effects on PPD change compared with non-surgical mechanical instrumentation alone?

Inclusion criteria
The following inclusion criteria were applied:

Search strategy
Detailed search strategies were developed and appropriately revised for each database, considering the differences in controlled vocabulary and syntax rules. No language or publication date restrictions were applied.

Electronic search
The search strategies for 4 major databases are shown in Appendix 1.

Unpublished literature search
In order to further identify potential articles for inclusion, grey literature was searched in the register of clinical studies hosted by the US National Institutes of Health (www. clini caltr ials. gov), the multidisciplinary European database (www. openg rey. eu), the National Research Register, and Pro-Quest Dissertation Abstracts and Thesis databases (https:// about. proqu est. com).

Manual search
The reference lists of all identified eligible studies and other published systematic reviews were hand-searched in order to identify further eligible studies.

Study selection
Study selection was performed independently and in duplicate by the first 2 authors (A.R., J-C.I.) of the review, who were not blinded to the identity of the authors of the studies, their institutions, or the results of their research. Calibration among the reviewers was performed in the first 20 studies retrieved.
Study selection procedure comprised of title-and abstract-reading followed by full-text-reading stages. After exclusion of non-eligible studies, the full report of publications considered by either author as eligible for inclusion was obtained and assessed independently. In studies reporting on the same patient sample, only the publication with the longest follow-up was considered. Disagreements were resolved by discussion and consultation with the third author of the review (A.St.). Inter-observer agreement was assessed for the screening of full-text articles. A record of all decisions on study identification was kept.

Data collection
The first and third authors (A.R., A.St.) performed data extraction independently and in duplicate. Disagreements were resolved by discussion with the latest two authors.
Specifically designed excel collection forms were used to record the desired information.
The following data were collected: author/title/year of study, study affiliation data, design of the study, number/age/ gender/smoking status of participants in intervention and control groups, intervention applied, and outcome assessed with all relevant clinical and radiographic variables.
If stated, the sources of funding, trial registration, and publishing of the trial's protocol were recorded. This information was used to aid assessment of heterogeneity and the external validity of the included studies.
In case of missing data, it was attempted to contact the corresponding author. Studies without enough data for metaanalyses were kept in the systematic review but excluded from the meta-analyses.

Risk of bias assessment
Risk of Bias 2.0 tool (RoB) was used to assess the quality of the included randomized studies [33]. Risk of bias assessment was performed independently and in duplicate by the first and third authors (A.R., A.St.) for the primary outcomes. Any concern was resolved by discussion with the last two authors.

Data analysis
Meta-analyses were conducted with included studies reporting similar interventions and comparable outcomes in homogeneous samples, i.e., in the case of limited heterogeneity. For continuous variables, mean differences, and standard deviations were used to summarize the data from each study. Mean differences and 95% CIs were calculated across studies. Data were analyzed with Review Manager 5.4 (Review Manager (RevMan), Version 5.4, The Cochrane Collaboration, Copenhagen, 2020).

Heterogeneity
Clinical and methodological heterogeneity were assessed by examining the characteristics of the studies, the similarity between the types of participants, the interventions, and the outcomes as specified in the inclusion criteria for considering studies for this review. Statistical heterogeneity was assessed using a chi 2 test and the I 2 statistic.

Subgroup analyses
In the case of sufficient data, subgroup analyses to explore the influence of study characteristics such as age, gender, and/or jaw were planned to be conducted.

Sensitivity analysis
We explored whether or not the analysis of studies stratified by design or by risk of bias (i.e., overall low risk versus high risk) yielded similar or different results.

Unit of analysis issues
We anticipated that some of the included studies presented data from repeated or paired observations on participants, which could lead to unit of analysis errors. In such cases, we followed the advice provided in Section 9.3.4 of the Cochrane Handbook for Systematic Reviews of Interventions [33].

Assessment of reporting bias
Potential reporting biases including publication bias, multiple (duplicate reports) publication bias, and language bias in this review were reduced by conducting an accurate and at the same time a sensitive search of multiple sources with no restriction on language. A search for ongoing trials was conducted too. In the presence of more than 10 studies in a meta-analysis, the possible presence of publication bias would have been investigated for the primary outcome.

Search
Through electronic search, a total of 607 records were identified. After duplicate removal, 160 records underwent abstract screening. No publications were retrieved from the manual searches. Upon exclusion of one-hundred and fiftyone articles based on their abstracts, nine articles remained for full-text evaluation. After exclusion of three articles [34][35][36] based on full-text analysis, 6 articles [25,[27][28][29]37, 38] remained for qualitative (n = 6) and quantitative analysis (n = 4). Calibration among authors indicated complete agreement (i.e., k-score = 1, agreement = 100%) for the study selection process. Details of the search process are provided in Fig. 1.

Studies' samples
Sample sizes varied from 16 to 49 patients. A power calculation was reported in 2 RCTs [27,28]. The age of the patients in all included studies ranged from 18 to 85 years. All studies, except 2, reported on gender distribution [37,38]. All studies included smoking patients, except one study [38].

Intervention/Comparison
Details of the interventions are described in Table 1. Briefly, all the included studies clearly reported on the NSPT performed under local anesthesia whether with hand and/or ultrasonic instrumentation. Five studies [25,27,29,37,38] reported on NSPT with or without the use of EMD  in non-treated periodontitis patients, while one study [28] showed clinical results after retreatment of selected sites. The application of EMD in test sites was performed always once, expect Schallhorn and co-workers [29] who re-applied EMD after 2 to 3 weeks. One study [37] included 4 groups of which two received NSPT and adjunctive systemic antibiotics (AB) either with or without the use of EMD. Inclusion of test and control sites varied among the studies. One study included all sites ≥ 6 mm in a parallel group design [27]; another study included only one inter-proximal site with an intrabony defect ≥ 2 mm and PPD ≥ 5 mm per patient [37]. Others included two sites per patient, one experimental and one control [25] or at least two teeth with similar PPD ≥ 5 mm and ≤ 8 mm, BoP in a split-mouth design [28]. Two studies included at least 2 pockets per contralateral quadrants, also in a split-mouth design [29,38].

Primary outcome
All studies, except Jentsch et al. [28], reported on CAL changes. Detailed information regarding CAL changes are summarized in Table 2. CAL change among the studies ranged from 0.75 to 2.91 mm. Although all studies reported higher values for CAL change for the test compared to the control groups, none of the studies reported on a significant difference among test and control groups. Graziani and co-workers [27] were the only ones distinguishing between sites ≤ 5 mm and ≥ 6 mm. Of the two papers [29,37] that presented CAL change after a follow-up time of 12 months, only one [37] presented 6-month and 12-month results demonstrating a stable CAL after NSPT with no further changes from the 6-month to the 12-month follow-up. Mombelli and co-workers [37] included additional two groups receiving systemic antibiotics or a placebo. No statistical difference was discerned when the groups with or without EMD were compared with each other irrespective of the antibiotic therapy they received (i.e., groups NSPT + AB and NSPT vs NSPT + AB + EMD and NSPT + EMD). However, the sites treated with both EMD and AB showed a synergistic effect gaining the largest amount of CAL and when comparing this group with the other three a significant difference favoring AB + EMD was found for all 3 time points that were assessed (i.e., 2, 6, and 12 months).

Secondary outcomes
All studies reported on the secondary outcome PPD change. Details are reported in Table 2. While 4 studies reported on average PPD change for all sites, two studies distinguished between ≤ 4 mm or ≤ 5 mm and ≥ 6 mm [27,29].  Sites > 6 mm demonstrated a significantly better result when EMD was used (Graziani et al. [27]). For mean PPD change, Jentsch et al. [28] observed an additional benefit of 0.79 ± 1.3 mm for the test group compared to the control. Two studies evaluated the frequency of pocket closure [28,29] reporting on significantly better results for test groups. BoP was determined by four studies of which 2 observed a significant difference favoring test conditions [28,29].

Meta-analysis
Meta-analysis could only be performed for CAL change at a 3-month follow-up, with data collected from 3 included studies [25,27,38] and for PPD change at a 3-6-month follow-up with data from 4 studies [25,27,28,38]. Random effects model was implemented. The meta-analysis revealed only a trend in favor of SRP + EMD versus SRP alone: a non-statistically significant CAL gain of 0.14 mm (p = 0.74; CI 95% − 0.66; 0.94) in favor of SRP + EMD compared to SRP alone was discovered. With respect to PPD, again, a non-statistically significant additional PPD reduction of 0.46 mm (p = 0.25; CI 95% − 0.33; 1.26) in favor of SRP + EMD compared to SRP alone was identified. Statistical heterogeneity was high in both cases (I 2 = 79% and 87%, respectively). None of the other investigated outcomes as well as the CAL and PPD changes at additional time points (i.e., 6-12 months) could be evaluated due to methodological and clinical heterogeneity. Details of the forest plots with respect to the primary (i.e., CAL change) and secondary outcome (i.e., PPD change) are reported in Figs. 2 and 3.

Risk of bias assessment
Detailed risk of bias assessment is demonstrated in Table 2.

Discussion
The present systematic review has assessed the level of evidence on the potential added benefit of the adjunctive use of EMD in NSPT compared to NSPT alone. The present study is, to our knowledge, the first systematic review on this controversially discussed topic. More specifically, based on the inclusion criteria, 6 RCTs with at least 3-month follow-up were retrieved and data from 4 of them could be mathematically combined. When focusing on the obtained results, minimal changes in terms of CAL change (3 studies) (i.e., 0.14 mm) and PPD reduction (4 studies) (i.e., 0.46 mm) were noted, failing to detect an additional clinical benefit of EMD application following NSPT. The main goal of periodontal therapy is to eliminate periodontal pockets through a non-surgical and/or surgical approach [39]. Even though scaling and root planing does still represent the "gold standard," initial supra-and subgingival instrumentation might leave residual pockets of at least 5 mm [40] which do require further treatment [41]. Consequently, as stated by Sanz-Sanchez and co-workers, root surfaces re-instrumentation might be recommended for shallow pockets (i.e., PPD up to 5 mm), while periodontal surgery should be performed in sites with PPD ≥ 6 mm [41]. Between these two treatment options (i.e., non-surgical re-instrumentation and periodontal surgery), several adjuncts to NSPT have been used to increase the chance of pocket closure (i.e., PPD ≤ 4 mm), such as local delivery of chlorhexidine chip [42], doxycycline [43], and tetracycline [44]; local application of lasers or photodynamic therapy [45,46]; and hyaluronic acid [47], and to reduce patients' morbidity. Nevertheless, it must be pointed out that none of the proposed techniques has shown additional beneficial clinical outcomes [48]. One the contrary, EMD application might provide a further treatment option to reduce the need for surgery as clearly shown by Jentsch and co-workers [28] and Schallhorn and co-workers [29]. Teeth with furcation involvement often exhibit residual pockets after NSPT or even experience exacerbation of probing depths in supportive therapy. However, none of the included RCTs reported on the closure of involved furcations. Furthermore, most studies did not report on inclusion of furcation sites.
The application of EMD in conjunction with a surgical access was proven to result in periodontal regeneration (i.e., formation of periodontal ligament, root cementum, and bone) in both animal [49] and human histological studies [18,19], while its flapless application has shown controversial results. More specifically, histological evidence of periodontal regeneration has been reported on 3 out of 4 examined teeth 6-month post-treatment by Mellonig and coworkers [50], while Sculean and co-workers [36] revealed a healing pattern predominantly characterized by formation of a long junctional epithelium. Irrespective of the histologically healing features, the application of EMD has been shown to have anti-inflammatory and antibacterial effects [51,52] as well as to promote early wound healing (i.e., neo-angiogenesis, periodontal ligament, and fibroblast cell migration) [53]; all these characteristics might explain the positive short-term results (i.e., 3 weeks up to 3 months) reported by some studies [27,29].
Several patient-related factors such as smoking status [54], self-performed plaque control [55], and adherence to a SPT regime [56] have been proven to influence the outcomes of NSPT. Smokers respond less favorably to periodontal therapy with less PPD reduction and CAL gain than non-smokers [57,58]. In the present study, only one article clearly excluded smokers [38]. Most of the studies, however, excluded heavy smokers. Furthermore, most of the studies did not report details on the full-mouth plaque score values which might have had an impact on the clinical outcomes. With respect to the SPT regime, all studies except Wyganowska et al. [38] clearly stated that patients were enrolled in a strict supportive periodontal therapy protocol consisting of oral hygiene measure reinforcement and supra-and subgingival instrumentation, if deemed necessary.
One additional aspect that has to be discussed is related to the 3-D defect morphology associated with deep pockets, which has never been clearly reported in any of the included articles, even though it is well known that deep and narrow infrabony defects do have higher self-regenerative potential than horizontal defects [59]. This aspect is even more important when one considers that the flapless application of EMD in intrabony defects has led to encouraging clinical results [34].
It has been documented that to maximize the efficacy of EMD during periodontal surgery, a careful drying of the planed root surfaced has to be performed before its application [60]: consequently, due to the difficulties to perform an accurate bleeding control through a minimal access to the root surfaces through the periodontal sulcus, the real effect of EMD in this clinical scenario might be difficult to be estimated. Further studies are warranted to investigate the efficacy of non-surgical application of EMD also with respect to costs. Nevertheless, EMD presents a further tool in the toolbox during non-surgical periodontal therapy and in certain cases may prevent surgical therapy and thus may reduce costs.
This study does present some limitations such as the low number of included studies and included patients. Moreover, due to the high heterogeneity of the included studies, a number of additional clinically relevant data such as the BoP values and longer evaluation periods (i.e., 12-month follow-up) were missing.

Conclusion
Within their limits, the present data have failed to show that the application of EMD as adjunct to NSPT may lead to additional clinical benefits in terms of CAL gain and PPD reduction when compared to NST alone. However, due to the high heterogeneity among the included studies, additional well-designed RCTs need to be performed to evaluate the use of EMD in various clinical scenarios, such as initial periodontal therapy or maintenance.
Author contribution Andrea Roccuzzo, Jean-Claude Imber, and Alexandra Stähli contributed to the study conception and design, and Andrea Roccuzzo, Jean-Claude Imber, and Alexandra Stähli to the material preparation and data collection. Dimitrios Kloukos performed the statistical analysis. Giovanni E. Salvi and Anton Sculean critically revised the manuscript and supervised the review process. The first draft of the manuscript was written by Andrea Roccuzzo, Jean-Claude Imber, and Alexandra Stähli and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding Open access funding provided by University of Bern The present study was funded by the author's own institutions. A.R. is the recipient of a 3-year scholarship from the Clinical Research Foundation (CFR) for the Promotion of Oral Health, Brienz, Switzerland.

Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations
Ethics approval Due to the nature of the present study with no patients' involvement, ethical committee approval was not sought.

Consent to participate
To perform the present study, informed consent was not required.

Conflict of interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.