Xenogeneic collagen matrix versus connective tissue graft for the treatment of multiple gingival recessions: A systematic review and meta‐analysis

Abstract A systematic review and meta‐analysis was performed to understand the efficacy of xenogeneic collagen matrix (CMX) compared with connective tissue graft (CTG) for the treatment of multiple adjacent gingival recessions (MAGRs). A literature search was performed for published randomized controlled trials in adult patients (≥18 years old) with Multiple Adjacent Miller class I and II gingival recessions (MAGRs). The assessments included recession depth, Recession width, complete root coverage, mean root coverage, probing depth, clinical attachment level, and keratinized tissue width. Pooled data were analyzed using fixed‐ and random‐effects models, and Forest plots were constructed. Heterogeneity within studies was calculated to assess publication bias. Four randomized controlled trials were included based on the eligibility criteria. Although the recession depth, complete root coverage, and mean root coverage were significantly lower with CMX (p = .017 and p = .001, p = .001, respectively), there was no statistically significant difference in the Recession width between CMX and CTG (p = .203). CMX showed significantly lower Probing Depth than CTG (p = .023); however, no significant difference in clinical attachment level (p = .060) and keratinized tissue width (p = .052) was observed between the groups. Owing to the heterogeneity in the included studies, firm conclusions cannot be drawn regarding the noninferiority of CMX compared with CTG. Long‐term studies are therefore needed to conclusively establish the relative efficacy of CMX in MAGR.

incorrect occlusal relationships, thin periodontal phenotype, tooth eruption pattern, and mechanical trauma (Armitage, 1999). GR can be either localized or multiple adjacent gingival recession (MAGR) with or without loss of attached gingival tissue with ensuing tooth sensitivity due to exposed dentin. The exposed root surface is frequently associated with esthetic complaints, root hypersensitivity, mechanical root wear, cervical root caries, and difficulties to achieve optimal plaque control (Tonetti et al., 2014); these issues prompt patients to seek corrective treatments. The absence of adequate mucogingival complex can lead to localized inflammation predisposing to GR development (Ravipudi, Appukuttan, Prakash, & Victor, 2017).
Various surgical and nonsurgical options are available for the treatment of GR. When GR is minimal, adequate thickness of tissue, favorable plaque control, not affecting aesthetics or causing dentin hypersensitivity and/or root caries, no treatment is needed. However, deeper defects are managed by surgical techniques that have been proposed as treatment modalities for GR with various outcomes in accomplishing root coverage (Miller, 1985;Miller, 1988;Tatakis et al., 2015). Root coverage techniques for all types of GRs are performed either with the objective of increasing keratinized tissue (KT) alone or a combination of KT, tissue regeneration, and coronally advanced flap (CAF). Traditionally, in the presence of less KT near the recession sites, a soft tissue grafting such as connective tissue graft (CTG) along with CAF or free gingival graft are recommended.
However, if the width of the attached gingiva is adequate, CAF can be used alone (Pini-Prato et al., 2010). Also, gingival thickness has an impact on the presence of GR and the outcomes of root coverage procedures. Gingival thickness less than 1 mm had reduced root coverage compared with thick gingival flaps (Hwang & Wang, 2006). Whereas periodontal plastic procedures are often performed primarily to restore form and function of teeth and its associated gingival complex, the CAF together with CTG was found to provide and maintain complete root coverage in both short-and long-term periods (Lops et al., 2015). Indeed, a meta-analysis has indicated that CAF + CTG was more effective in root coverage at single GR with Miller class I and II compared to CAF alone or CAF + Guided Tissue Regeneration (Cairo, Nieri, & Pagliaro, 2014). However, this technique has some inherent challenges for patients with multiple recession defects especially when variation in root prominence, vestibular depth, and degree of recession is present. In addition, there is a degree of morbidity associated with CTG harvesting especially when the quantity of donor tissue is limited Wessel & Tatakis, 2008). Tunneling (TUN) technique gained popularity with clinicians by offering minimally invasive surgery with acceptable root coverage and better esthetic outcomes.
A recent review on the efficacy of TUN versus CAF indicated that the former was useful for both localized and MGARs, although, the latter was found to be associated with better root coverage .
Among these materials, the initial data evaluating xenogeneic collagen matrix (CMX) showed promising results in single recession defects.
CMX is a bilayer composed of an outer compacted layer designed to hold the suture and protect the defect and the inner porous matrix that promotes quick stabilization of blood clot and encouraging rapid vascularization and tissue integration (Ghanaati et al., 2011;Rocchietta, Schupbach, Ghezzi, Maschera, & Simion, 2012). CMX has been shown to promote regeneration of keratinized gingiva in both the width and thickness not only around natural tooth but also around dental implants (Sanz, Lorenzo, Aranda, Martin, & Orsini, 2009). A recent review (Atieh, Alsabeeha, Tawse-Smith, & Payne, 2016) reported that CMX had better outcomes than CAF alone in terms of root coverage. However, CMX performed less in overall clinical outcomes compared with CAF + CTG. CTG + CAF had a higher percentage of complete/mean root coverage and mean recession reduction than CMX. CMX showed higher mean root coverage, recession reduction, and gain in KT than CAF alone. No significant differences were reported in patient's aesthetic satisfaction between CMX and CTG.
Nevertheless, these findings were mostly related to isolated Miller class I and II marginal recession because there were limited reports of using CMX at multiple adjacent sites.
The goal of this systematic review was to compare the efficacy of CMX with CTG for the treatment of multiple adjacent Miller's Class I and Class II gingival recession (MAGR) in terms of clinical parameters and patient-related outcomes.

| Selection of studies
Two independent reviewers (M.T. and A.A.) screened the titles and abstracts initially, then, full-text articles were analyzed to decide whether the studies met the inclusion criteria. Disagreement between the reviewers was resolved through discussion until consensus was reached. Cohen's Kappa score was used to assess inter-reviewer agreement of selection process (McHugh, 2012). The reasons for excluding studies were recorded.

| Data synthesis
Eligible studies underwent data extraction and validity assessment.
Predesigned extraction forms were developed to retrieve and assess essential information such as title, authors, year, study location, study design, method of randomization, duration of study, allocation concealment, blinding (participants, investigators, and outcome assessors) length of observation period, and reported clinical outcomes.
Data synthesis was preformed through organizing data in an evidence table, and a descriptive summary was created to determine study characteristics, study quality, and results. Descriptive statistical analysis according to the mean values was used to evaluate the outcomes of test and control groups. Any disagreements were resolved by discussion.

| Risk of bias assessment
The assessments of the risk of bias for the included clinical trials were performed using the Cochrane Collaboration's tool for assessing risk of bias (Higgins et al., 2011). The analysis of each clinical trial was based on the following seven main domains: random sequence generation, allocation concealment, blinding participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting, and other sources of bias. The risk of bias was graded as low, high, or unclear for each domain based on the criteria defined in the Cochrane Handbook for Systematic Reviews of Interventions version 5.1.0 (Higgins and Green, 2011).

| Data analysis
Data were analyzed using MedCalc for Windows version 15.0 (MedCalc Software, Ostend, Belgium). Descriptive statistics (mean and standard deviation) were used to describe the quantitative outcome variables (RD, RW, CRC, MRC, PD, CAL, and KTW). Metaanalysis was carried out by combining the mean difference values for each of the seven outcome variables. The relative risk (RR) and standardized mean difference (SMD) as a pooled effect and a cut-off values of 0.2 as small effect, 0.5 as medium effect, and 0.8 and more as larger effect (Cohen's rule) were used to report the overall effect, and student's t test was used to report the statistical significance.
Cochran's Q, which is the weighted sum of squares on a standardized scale, was used to identify the presence of heterogeneity along with I 2 , which its percentage (0% to 100%) of observed total variation across included studies in meta-analysis, due to real heterogeneity rather than chance. A value of I 2 , which is greater than 50%, was used to indicate increasing levels of unexplained variability in the effect sizes. Both the fixed and random effect models were used to obtain the pooled estimates of all the outcome variables. Based on the values of I 2 , appropriate overall effect (RR and SMD) was used to report its statistical significance. The results of different studies with its 95% confidence intervals (CI) and the overall effect (under the fixed and random effects model) with 95% CI were illustrated in Forest plots. A p value of <.05 and 95% CI were used to report statistical significance and its precision.

| RESULTS
The screening process according to Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines (Moher et al., 2009) is shown in Figure 1. A total of 1,392 studies were identified based on the search terminology from the various search engines. However, most of the studies (N = 1272) were duplicate results. The remaining 120 articles were screened, and 116 were excluded owing to lack of relevance and criteria applied. Assessments were performed for four included studies. The kappa value for inter-reviewer agreement for potentially relevant articles was 0.95 for full text articles (Cohen, 1968).

| Study design and patient features
The age of the participants ranged from 18 to 50 years with a followup period ranging from 3 months to 5 years. All included studies (
The remaining did not mention which arches were treated. Regarding the type of teeth treated, two studies included anterior teeth, premolars, and molars (Cieślik-Wegemund et al., 2016;Pietruska et al., 2019); whereas, the remaining studies did not mention the type of teeth treated in a detailed manner.

| Type of interventions
GRs were surgically treated by CTG + CAF in control groups of all included clinical trials. Autologous CTGs were harvested from the palate using the trap door technique whenever feasible and the de-epithelialized free gingival graft in cases with insufficient tissue thickness as described in most articles (Aroca et al., 2013;Cieślik-Wegemund et al., 2016;Tonetti et al., 2018). The CTG was immediately harvested after tunnel preparation by using either a modified distal wedge procedure (Azzi, Etienne, & Carranza, 1998) or the single-incision technique (Hürzeler & Weng, 1999) depending on anatomical considerations. If needed, the harvested graft was trimmed using a N°15 blade to achieve an optimal thickness of 1-1.5 mm.
Recessions were also surgically treated with a variation of techniques. One study performed CAF (Tonetti et al., 2018) whereas FIGURE 1 Study identification flow chart according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses   Aroca et al., 2013;Pietruska et al., 2019). Only one study reported the application of TUN to cover both grafts (Cieślik-Wegemund et al., 2016). The flaps were positioned coronally to the cementoenamel junction by means of suspended sutures placed above the contact point (Azzi et al., 1998).

| Postoperative care
In week (Tonetti et al., 2018). In terms of antibiotic use, Tonetti et al. (2018) reported the use of antibiotics; however, specific details were not mentioned. Another study used Augmentin 625 mg for 7 days due to their university regulations (Aroca et al., 2013). With regard to analgesics use, one study reported that patients were given analgesics 50-mg Cataflam for 3 days (Aroca et al., 2013); whereas in another study, patients were given 600-mg ibuprofen or 500-mg paracetamol (Tonetti et al., 2018). Finally, one trial reported instructing their patients to use analgesics when needed, but details were not mentioned (Pietruska et al., 2019).

| Risk of bias assessment
The results of bias assessment of the included studies are presented in A summary of meta-analysis for secondary outcome variables along with the corresponding Forest plots is presented in Table 5 and Figure 3.b, respectively.    (Slade, 1997). They reported that time of recovery was 1.8 days shorter in CMX group compared with CTG group. This difference was shown to be statistically significant. Finally, they also reported that CMX group surgery was 15.7 min shorter (95% CI from 11.9 to 19.6, p < .0001) and less painful as reported by patient (11.9

| Aesthetics, healing, and pain
VAS units, 95% CI from 4.6 to 19.1, p = .0014) in CMX subjects (Tonetti et al., 2018). A summary is presented in (Table 2).    the efficacy of any one type of procedure . Collagen matrix of porcine and bovine origin has been developed to be a safe alternative material, which provides regeneration of gingival tissues and promotes wound healing (Sanz et al., 2009;Thoma, Sancho-Puchades, Ettlin, Hämmerle, & Jung, 2012 It seems that CTG provide better KTW, although this was not significant (p = .052) probably due to the limited number of studies included. A possible explanation for the difference in KTW is the lack of cells of the CMX (Yu, Tseng, & Wang, 2018). However, it is important to note that CMX was found to be completely incorporated into the adjacent host connective tissues in the absence of a significant inflammatory response. The healing was characterized by the formation of new cementum and new connective tissue attachment in the apical aspect of the defect and by a junctional epithelium in its most coronal third. When compared with CAF alone, both techniques rendered similar clinical outcomes. Although the CMX graft attained more tissue regeneration, with a shorter epithelium and a larger new-cementum formation (Vignoletti et al., 2011).
The authors are aware of the limitations of the current review.
Only four randomized controlled trials were found with matched criteria. We believe that clinical experience and availability of materials contributed to this limitation. In addition, heterogenicity was obvious among included studies with different surgical techniques and variation of reported outcomes preventing a firm conclusion.
The studies identified in the current meta-analysis were not conclusive about the superiority of one procedure over the other.
However, in spite of clinical outcomes compared with CTG in the treatment of MAGR, CMX may be considered as noninferior alternative avoiding the need for second area of surgery and shortening the procedure time with reduced patient's discomfort postoperatively.