Received: 18 January 2021 | Revised: 22 February 2021 | Accepted: 19 March 2021

DOI: 10.1111/jcpe.13466

ORIGINAL ARTICLE CLINICAL PERIODONTOLOGY

Autologous connective tissue graft or xenogenic collagen

matrix with coronally advanced flaps for coverage of
multiple adjacent gingival recession. 36-­month follow-­up of a
randomized multicentre trial

Maurizio S. Tonetti1,2 | Pierpaolo Cortellini2 | Daniele Bonaccini2 | Ke Deng3 |

Francesco Cairo2,4 | Mario Allegri2 | Gianpaolo Conforti2 | Filippo Graziani2,5 |
2 2 2,6 2,7
Adrian Guerrero | Jan Halben | Jacques Malet | Giulio Rasperini | Heinz Topoll2

1
Department of Oral and Maxillofacial
Implantology, Shanghai Key Laboratory of Abstract
Stomatology, National Clinical Research
Aim: To report the 36-­month follow-­up of a trial comparing the adjunct of a xenogenic
Centre for Stomatology, Shanghai Ninth
People Hospital, School of Medicine, collagen matrix (CMX) or connective tissue graft (CTG) to coronally advanced flaps
Shanghai Jiao Tong University, Shanghai,
(CAF) for coverage of multiple adjacent recessions.
China
2
European Research Group on
Material and methods: 125 subjects (61 CMX) with 307 recessions in 8 centres from
Periodontology, Genova, Italy the parent trial were followed-­up for 36 months. Primary outcome was change in
3
Division of Periodontology and Implant position of the gingival margin. Multilevel analysis used centre, subject and tooth as
Dentistry, Hong Kong University, Hong
Kong, China levels and baseline parameters as covariates.
4
Unit of Periodontology, Department Results: No differences were observed between the randomized and the follow-­up
of Surgery and Translational Medicine,
population. Average baseline recession was 2.6 ± 1.0 mm. 3-­year root coverage was
University of Florence, Florence, Italy
5
Department of Periodontology, 1.5 ± 1.5 mm for CMX and 2.0 ± 1.0 mm for CTG (difference of 0.32 mm, 95% CI
University of Pisa, Pisa, Italy from −0.02 to 0.65 mm). The upper limit of the confidence interval was over the non-­
6
U.F.R. of Odontology, Paris 5 -­Descartes
inferiority margin of 0.25 mm. No treatment differences in position of the gingival
University Paris, Paris, France
7
Department of Biomedical, Surgical and
margin were observed between 6-­and 36-­month follow-­up (difference 0.06 mm, 95%
Dental Sciences, University of Milano, CI −0.17 to 0.29 mm).
Milan, Italy
Conclusion: CMX was not non-­inferior with respect to CTG in multiple adjacent reces-
Correspondence sions. No differences in stability of root coverage were observed between groups and
Maurizio Tonetti, Department of Oral
and Maxillofacial Implantology, Shanghai
in changes from 6 to 36 months. Previously reported shorter time to recovery, lower
Jiao Tong University –­Pudong Campus, morbidity and more natural appearance of tissue texture and contour observed for
4F, Building 1, 115 Jinzun Road, 200125
Shanghai, China.
CMX in this trial are also relevant in clinical decision-­making.
Email: [email protected]
KEYWORDS
Funding information collagen matrix, coronally advanced flap, gingival recession, human, randomized controlled
This study was sponsored by the European clinical trial, root coverage
Research Group on Periodontology
(ERGOPerio). The study was financially
supported in part by an unrestricted

© 2021 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.

J Clin Periodontol. 2021;00:1–8.  wileyonlinelibrary.com/journal/jcpe | 1

2 | TONETTI et al.

grant from Geistlich Pharma AG,

Switzerland. The employed re-­generative Clinical Relevance
materials were a gift from Geistlich
Scientific rationale for the study: Stability of root coverage outcomes is critical to decision-­
Pharma AG, Switzerland. This research
was initiated by the investigators who making. CTG is considered to be the standard to obtain stability of outcomes. Little is known
independently performed all phases of about multiple recession sites.
the study including protocol development,
Principal findings: The results of this trial indicated that CMX was not non-­inferior to CTG at
experimental procedures, data analysis,
result interpretation and reporting 36 months. No differences between test and control, however, were observed in terms of sta-
bility of outcomes between 6 and 36 months. In this trial, the benefit of CTG was due mostly to
improved early outcomes.
Practical implications: The clinical indications for the use of first-­generation CMX are primarily
as follows: (i) cases with contraindication to autologous CTG harvesting from the palate and (ii)
cases where the patient and clinician are seeking to limit morbidity, shorten surgery and time
to recovery, obtain a more natural tissue texture and contour but are willing to accept a higher
chance of a less than optimal outcome in terms of root coverage.

1 | I NTRO D U C TI O N An initial study on isolated recession indicated that combining

CMX with a coronally advanced flap (CAF) led to good clinical out-
Stability of outcome after gingival recession coverage procedures comes that compared favourably with those obtained with the use
is a key component in clinical decision-­making. Recent systematic of autologous connective tissue grafts (McGuire & Scheyer, 2010).
reviews have highlighted the predictability of root coverage for sin- The clinical performance of CMX in a multicentre trial on sin-
gle tooth recession (Cairo et al., 2014; Chambrone et al., 2019) and gle tooth recessions (Jepsen et al., 2013) seems to enhance healing
the emerging evidence pointing to effective coverage of multiple outcomes of coronally advanced flaps and accumulating evidence
adjacent recessions (Graziani et al., 2014; Chambrone et al., 2019). points to benefits in medium-­term root coverage outcomes (Jepsen
The short-­term benefits, however, need to be substantiated by ro- et al., 2017). A possible mechanism for the medium-­term benefit of
bust prospective medium-­ to long-­term evaluations. While several CMX may be the observed increase in tissue thickness when added
reports have described long-­term outcomes, many of these studies to coronally advanced flaps in single tooth recessions (Stefanini et al.,
are retrospective. A recent systematic review has addressed the 2016). No additional benefit, however, has been observed adding
differential outcomes in terms of stability of position of the gingival CMX to CAF in multiple adjacent recessions (Rotundo et al., 2019).
margin after different types of root coverage procedures in single-­ Results of a non-­inferiority trial comparing CMX with autologous
rooted teeth (Dai et al., 2019). The results point to greater stabil- grafting from the palate in multiple adjacent recessions indicated
ity, defined as lack of significant difference between early-­ and that subjects treated with CMX had a quicker recovery after surgery,
medium-­term results, for procedures involving an increase in thick- but the trial failed to demonstrate non-­inferiority of CMX with re-
ness of the gingiva, such as those associated with the placement of spect to CTG (Tonetti et al., 2018). A follow-­up analysis focussing on
autologous connective tissue grafts (CTG) compared with coronal professional assessment of aesthetic outcomes at 6 months showed
displacement of the flap alone. Such understanding, which has long that CTG gave better results for the root coverage component of the
been an important component in clinical practice, has led to greater root coverage aesthetic score while CMX was better for marginal
emphasis on marginal tissue thickness and not only on the position tissue contour and tissue texture (Pelekos et al., 2019).
of the gingival margin as a component of success of root coverage The aim of this 3-­year follow-­up study was to compare the root
procedures. A recent network meta-­analysis has shown an asso- coverage outcomes of CTG and CMX and assess their stability be-
ciation between post-­operative gingival thickness, obtained with tween 6 and 36 months.
different biomaterials or grafts, and root coverage outcomes and
their stability over time (Tavelli et al., 2019; Barootchi et al., 2020).
In recent years, a lot of emphasis has been placed on biomaterials 2 | M ATE R I A L S A N D M E TH O DS
and biological agents to enhance root coverage outcomes and avoid
autologous tissue grafting from the palate. Among them, collagen 2.1 | Study design and population
matrices (CMX) have been extensively investigated.
Preclinical studies have indicated that the collagen matrix is This study reports root coverage outcomes at the 36-­month follow-­up
replaced with the subject own connective tissue with the desired of a non-­inferiority, randomized, controlled, parallel arm, standard of
histologic and functional characteristics (Thoma et al., 2012) and care-­controlled, assessor-­blind, multicentre, multinational and practice-­
leads to an augmentation in both the width and the thickness of the based trial (clincialtrial.gov registration NCT01440426). Ethical approval
band of keratinized tissue (Thoma et al., 2010; Thoma et al., 2011; was obtained by the Freiburg Ethics Committee International (FEKI
Vignoletti et al., 2011). code 011/1546 and 015/832) and by the competent local authority
TONETTI et al. | 3

for each centre. Subjects provided informed consent, and the study was using both the incisal edge and the natural or composite filling re-
performed according to the Declaration of Helsinki on experimentation constructed cement–­enamel junction (CEJ) as the reference point.
involving human subjects. Details of the inclusion and exclusion crite- Outcomes were assessed using changes in recession from the incisal
ria, randomization and allocation concealment have been previously edge as the reference; recessions were characterized using the CEJ
reported (Tonetti et al., 2018). In brief, subjects with the presence of as the reference. Depth of the gingival sulcus (PD) and width of the
a minimum of two adjacent recessions of the gingival margin requiring keratinized tissue (KT) were assessed clinically with a UNC15 probe.
surgical correction, no prior experience of root coverage surgery, ability The location of the mucogingival junction was assessed with the vis-
to achieve and maintain good oral hygiene and control gingivitis in the ual and functional method and supplemented by the histochemical
whole of the dentition (FMPS <25% and FMBS <25%) were invited to method in areas of unclear demarcation (Guglielmoni et al., 2001).
participate. Subjects presenting with (i) untreated periodontitis, (ii) per- Oral hygiene levels were assessed with the plaque control record
sistence of uncorrected gingival trauma from tooth brushing, (iii) Inter-­ (O'Leary et al., 1972), while gingival inflammation was assessed as
dental attachment loss greater than 1 mm or furcation involvement in percentage of sites with bleeding on probing (Tonetti et al., 1993).
the teeth to be treated, (iv) Presence of severe tooth malposition, ro-
tation or clinically significant super-­eruption, (v) self-­reported current
smoking exceeding 20 cigarettes/day or pipe or cigar smoking and (vi) 2.4 | Sample size
rheumatoid arthritis or known sensitization to collagen-­based medical
products, and/or presence of medical contraindications to elective sur- The detailed sample size calculation to detect a 0.25 mm non-­
gery were excluded. Indications for surgical intervention and required inferiority margin in recession reduction with CMX has been re-
prior therapy have been described in the parent trial. Eight of the original ported in the 6-­month trial. The size of this 36-­month follow-­up
13 study centres participated in this medium-­term follow-­up. Methods was limited to 8 centre who were available for the medium-­term
to control study bias have been described in the original report. To en- follow-­up.
sure consistency over time in measurements, examiners performed cali-
bration exercises in the clinical measurement of the primary outcome
(position of the gingival margin with regards to the cement–­enamel 2.5 | Statistical analysis
junction and the incisal edge of the tooth) and had to achieve an intra-­
examiner re-­producibility >98% within 1 mm (Cairo et al., 2016). Data were entered into a database and proofed for entry errors.
Descriptive statistics were summarized as means and standard devia-
tions for quantitative data and frequencies and percentages for quali-
2.2 | Interventions tative data. Multilevel analyses were performed with the treatment
(CMX versus CTG) as explicative variable. For the site outcome vari-
Surgical interventions, study devices and methods have been de- ables (for example root coverage), the three levels of the models were
scribed in detail (Tonetti et al., 2018). In brief, both groups received centre, patient and tooth. Baseline values were used as a covariate.
coronally advanced flaps with either autologous connective tissue For complete root coverage, a three-­level logistical model was tested
graft (standard of care control) or a xenogenic collagen matrix (CMX, using CEJ-­GM at baseline as a covariate. The intra-­class correlation
Geistlich Mucograft®, Geistlich Pharma AG). Based on the local anat- coefficients were calculated to estimate the variability among centres
omy and the location and distribution of the recessions, coronally ad- and the variability between measurements using the incisal edge and
vanced flaps included either rotated papillae flap or trapezoidal flap the CEJ as reference points to detect changes in position of the gin-
designs with or without vertical releasing incisions (Zucchelli & De gival margin. Estimates for the treatment effect, standard errors and
Sanctis, 2000; Cortellini et al., 2009; Cairo et al., 2016). The randomi- 95% confidence intervals were provided. The statistical software was
zation envelope was opened after completion of the preparation of MLwiN 2.21 (Centre for Multilevel Modelling, University of Bristol,
the recipient bed of the graft. Grafts and dried CMX were positioned UK), Stata 14 and JMP 13.0.0 (SAS Institute Inc.).
and sutured 1 mm apical to the cement–­enamel junction with 6–­0
braided resorbable polylactic sutures. Flaps were sutured with inter-
rupted (Seralene, Serag and Wiessner, Germany) and/or sling (e-­PTFE, 3 | R E S U LT S
W.L. Gore) 6–­0 and 7–­0 monofilament sutures attempting to fully
cover both CTG and CMX. Post-­operative procedures and instructions 3.1 | Study population and external validity
have been previously described in detail (Tonetti et al., 2018).
The CONSORT patient accountability diagram is displayed in Figure 1.
187 subjects with recession at 485 teeth were randomized and re-
2.3 | Clinical measures ceived the allocated intervention in 13 centres. 8 centres were avail-
able to participate into the 36-­month follow-­up. These had recruited
The position of the gingival margin was measured to the nearest 125 patients (64 allocated to CTG) for a total of 307 recessions
mm with a UNC15 periodontal probe (PCP-­UNC 15, Hu-­Friedy) (158 CTG). All subjects completed the 6-­month follow-­up. Over the
4 | TONETTI et al.

F I G U R E 1 CONSORT patient
Randomized
accountability diagram
187 subjects, 485 teeth

Allocation

Allocated to CTG (n= 95 subjects, 243 teeth) Allocated to CMX (n= 92 subjects, 242 teeth)

Received allocated intervention (n= 95) Received allocated intervention (n= 92)
Did not receive allocated intervention (n= 0) Did not receive allocated intervention (n= 0)

6-month Follow-
Up & Analysis

Lost to follow-up (n= 0) Lost to follow-up (n= 0)

Analysed (13 centres n= 95 subjects, 243 teeth) Analysed (13 centres n= 92 subjects, 242 teeth)

36-month Follow-
Up Study

8 centres participating n= 64 subjects, 158 8 centres participating n= 61 subjects, 149

teeth teeth

36-month Follow-
Up & Analysis

Lost to follow-up (n= 19 subjects, 46 teeth lost Lost to follow-up (n= 17 subjects, 42 teeth lost
to follow-up) to follow-up)
Analysed (n= 45 subjects, 112 teeth) Analysed (n= 44 subjects, 107 teeth)

TA B L E 1 Study population TA B L E 2 Location of treated teeth

CTG CMX CTG CMX

N = 64 N = 61 Tooth type N = 158 N = 149

Age (years) 39.1 ± 10.5 41.2 ± 10.0 Maxillary incisors 21 (13) 29 (19)
Females 37 (58%) 37 (61%) Maxillary canines 39 (25) 41 (28)
Smokers 12 (19%) 13 (21%) Maxillary pre-­molars 58 (37) 44 (30)
Baseline OHIP14 values 9.0 ± 6.2 9.1 ± 6.5 Maxillary molars 11 (7) 4 (3)
Full Mouth Plaque Scores 13.3 ± 7.8 13.2 ± 7.6 Mandibular incisors 9 (6) 5 (3)
Full Mouth Bleeding Score 7.6 ± 7.4 6.2 ± 5.6 Mandibular canines 8 (5) 8 (5)
Dentine sensitivity Air 43 (67%) 40 (66%) Mandibular pre-­molars 12 (8) 17 (11)
Test positive Mandibular molars 0 (0) 1 (1)
Dentine sensitivity Yeaple 23 (36%) 31 (51%)
Note: Frequency (percentage).
Test positive

Note: Patient-­level baseline characteristics (mean ± SD or frequency and (70%, 112 teeth) and 44 subjects in the CMX group (72%, 107 teeth)
percentage). Means ± SD. were available for follow-­up. The patient characteristics, the position
Abbreviations: FMBS, full mouth bleeding score; FMPS, Full mouth
of the treated teeth and the local condition at teeth with recession
plaque score.
that were included are shown in Tables 1–­3, respectively. No differ-
ences were observed comparing the parent population (6-­month trial)
36 months, 19 patients allocated to CTG and 17 allocated to CMX and the current population (data not shown).
were lost to follow-­up due to lack of availability for the examina-
tion, relocation to distant areas or inability of the study team to con-
tact them. No subject was lost due to known study-­related reasons. 3.2 | Root coverage and sensitivity at 36 months
Subjects gave initial consent for the 6-­month trial and were later asked
to consent to an extension to 36-­month, this may have contributed to Table 4 shows the adjusted comparisons in root coverage outcomes
the loss of some subjects. At 36 months, 45 subjects in the CTG group at 36 months between test and control arising from the multilevel
TONETTI et al. | 5

TA B L E 3 Baseline characteristics of teeth with recessions

p < .001) and bigger increases in KT width (0.56 mm more than CTX,
CTG CMX 95% CI 0.23–­0.89 mm, p < .001). No significant differences were ob-
Variable N = 158 N = 149 served comparing test and control treatments in terms of changes in
Distance from CEJ-­GM (gingival 2.6 ± 1.0 2.6 ± 1.0 the position of the gingival margin with respect of the CEJ (p = .067,
margin) mm NS) or changes in PD (p = .084, NS). No difference in odds ratios of
Distance from incisal edge to-­GM 11.8 ± 1.6 12.2 ± 1.9 complete root coverage at 36 months was observed (p = .061).
mm
The different results obtained using the incisal edge of the tooth
Probing depth mm 1.4 ± 0.5 1.4 ± 0.6 or the CEJ (Table 4) was further explored. Firstly, the results of the
Width of keratinized tissue mm 2.8 ± 1.3 2.6 ± 1.2 model using the CEJ as the reference were confirmed with a sensi-
Max inter-­dental clinical 0.2 ± 0.4 0.3 ± 0.6 tivity analysis using a multiple imputation method for missing values
attachment loss mm
that assigned 20 different values for every missing observation. The
Presence of inter-­dental clinical 29 (18%) 32 (21%) results confirmed those of the main analysis (0.31 mm in favour of
attachment loss
CTG, with a 95% CI of −0.01 to 0.62, superiority p = .055). With both
Local plaque score 0 (0%) 3 (2%)
the main and the sensitivity analysis, the non-­inferiority hypothesis
Local bleeding on probing 0 (0%) 0 (0%) cannot be rejected and the superiority of CTG over CMX cannot be
Cervical filling 0 (0%) 15 (10%) excluded. Secondly, the intra-­class correlation between the results
Cervical filling removed, if present 0 (0%) 11 (73%) obtained with the incisal edge and the CEJ as references for deter-
Cervical caries 2 (1%) 0 (0%) mining changes in the position of the gingival margin were calculated
Cervical caries treated, if caries 2 (100%) 0 (0%) as follows: 0.73 (95% CI 0.64 to 0.77).
present No differences in the frequency of dentine sensitivity were ob-
Cervical caries filled, if caries 2 (100%) 0 (0%) served among groups (Table 5).
present
Presence of CEJ abrasion 66 (42%) 63 (42%)
CEJ abrasion re-­constructed with 56 (85%) 50 (79%) 3.3 | Stability of root coverage between 6 and
adhesive re-­construction, if 36 months
present

Note: Means ±SD or frequency (percentage). Table 6 shows the tooth-­based differences in root coverage param-
eters between 6 and 36 months. This analysis was based on the 89
models. The main non-­inferiority analysis indicates that CMX fails to subjects available for both the 6-­month and 36-­month follow-­up.
reach the stipulated margin of 0.25 mm with respect to CTG. A post No differences were observed between the ITT and this popula-
hoc superiority interpretation of the data was also done due to the tion both in terms of baseline characteristics and oral hygiene and
loss of power resulting from the reduced sample size. In this context, gingival inflammation parameters (data not shown). An increase in
CTG would have resulted in significantly greater estimated changes recession was observed for both groups. A multilevel model compar-
in positions of the gingival margin with respect to the incisal edge ing test and control treatments showed 0.06 mm (95% CI −0.17 to
of the tooth (0.56 more coronal than CTX, 95% CI 0.25–­0.88 mm, 0.29 mm, superiority p = .63) greater recession in the CMX group.

TA B L E 4 A36-­month clinical outcomes

Estimated
CTG CMX difference
Variable N = 112 N = 107 (Odds ratio*) 95% CI

Changes in CEJ-­GM (mm) 2.0 ± 1.0 1.5 ± 1.5 0.32 −0.02; 0.65
Changes in IE-­GM (mm) 1.8 ± 1.3 1.5 ± 1.3 0.56 0.25; 0.88
Changes in PD (mm) −0.3 ± 0.7 0.0 ± 0.8 −0.17 −0.37; 0.02
Changes in KT (mm) 0.5 ± 1.0 0.0 ± 1.2 0.56 0.23; 0.89
Complete root coverage N (%) 66 (59%) 47 (44%) 2.17* 0.96; 4.91

Note: Multilevel model: Centre, patient, tooth.

Multilevel model estimating clinical outcomes taking into account clustering of multiple
teeth in a single patient (surgery) and patients within a specific study centre. Changes in
CEJ-­GM = changes in the distance from the cement–­enamel junction to the gingival margin.
Changes in IE-­GM (mm) = changes in the distance between the incisal edge of the tooth and
the gingival margin. They are estimates of root coverage. PD = probing depth, KT = width of
keratinised tissue. Data are expressed as means (SD) in mm.
6 | TONETTI et al.

TA B L E 5 Patient-­based tooth
CTG CMX
sensitivity at 36 months
Variable N = 45 N = 43 Odds Ratio 95% CI p-­value

Air Test (positive) 8 (18%) 10 (23%) 1.38 0.27; 5.13 .6308

Yeaple Test 6 (13%) 5 (12%) 0.92 0.42; 2.04 .8380
(positive)

Note: Odds ratios are calculated using baseline sensitivity as a covariate.

TA B L E 6 Stability of outcomes in the test and control groups with measurements that used the CEJ as the reference point to as-
between 6 and 36 months (tooth-­based analysis) sess variation in the position of the gingival margin. This observation

CTG CMX is potentially important. In this trial, reconstruction of the CEJ was
Variables N = 112 N = 107 performed at sites with cervical abrasion; it is possible that difficul-
ties in recognition of a clearly demarcated CEJ within the composite
Increase in recession from CEJ 0.2 ± 0.6 0.3 ± 0.9
mm (SD) filling might have led to uncertainty about this location and increased
Increase in recession from 0.3 ± 0.7 0.2 ± 0.6 measurement errors in particular in cases with more exposure of the
incisal edge mm (SD) cervical reconstruction. Difficulties in the detection of the CEJ in root
Difference in PPD mm 0.0 ± 0.6 −0.1 ± 0.6 coverage procedures have been well recognized and may be even
Changes in inter-­dental CAL mm 0.0 ± 0.2 a
0.3 ± 1.2 more significant at sites where composite reconstruction is performed

Changes in KT mm 0.0 ± 0.7 a

0.0 ± 0.8 (Cairo et al., 2020). This anticipated issue was the rationale for intro-
duction of the incisal edge of the tooth (rather than only the CEJ) as
CRC at 6 and 36 months (%) 63 (56) 44 (41)
the reference point for assessing changes in position of the gingival
CRC not 6 and not 36 months 34 (30) 44 (41)
(%) margin. The divergent results were further explored by assessing the

CRC 6 but not at 36 months (%) 12 (11) 16 (15)

intra-­class correlation of measurements using the incisal margin or the
CEJ as the reference point to assess changes in position of the gingival
CRC not 6 but yes at 36 months 3 (3) 3 (3)
(%) margin. This analysis revealed only moderate agreement between the
a two measurements (ICC 95% CI range of 0.64 to 0.77). More research
N = 110, please see text for multilevel model reporting significance of
treatment effect. is needed to improve accuracy of measurements of root coverage out-
comes, in particular for trials incorporating re-­construction of the CEJ.
Both test and control cases showed a degree of relapse between
3.4 | Adverse events 6 and 36 months, but the inter-­group difference was small and not sig-
nificant. This observation raises interesting hypotheses. CTG is gen-
Investigators reported no study-­related adverse events during the erally considered to be the gold standard for the combination of two
follow-­up period of the study indicating that both treatments were advantages: (i) better early healing thanks to superior potential arising
safe and well tolerated. from improved vascularization and survival of soft tissues on the root
surface and (ii) greater stability over time due to enhanced resistance
to relapse due to increased tissue thickness (Tavelli et al., 2019). The
4 | DISCUSSION observations from the parent trial at 6 months (Tonetti et al., 2018) are
in broad agreement with the expected early benefits in root coverage
The results of the present report together with the previous ones outcomes of CTG in multiple recession. The lack of an intergroup dif-
(Tonetti et al., 2018; Pelekos et al., 2019) of this trial show im- ference between 6 and 36 months in terms of position of the gingival
proved root coverage outcomes for both test and control treatments. margin reported in this study may be interpreted as an indication that
Interpretation of the inter-­group difference is affected by the loss of CMX and CTG might have similar effect on relapse. The low power of
power arising from the fact that only 8 study centres were available for the present study, however, does not allow drawing of firm conclusions.
the 3-­year follow-­up. The original non-­inferiority hypothesis, there- CTG was statistically significantly better than CMX in terms of in-
fore, cannot be properly assessed; results, however, point to the fact crease in KT width, the clinical relevance, if any, of the observed 0.5 mm
that CMX was not non-­inferior with respect to CTG. An exploratory difference needs to be better understood. Indeed, in recent years,
superiority analysis showed divergent effects with respect to different the interest in KT width in periodontal plastic surgery procedures has
root coverage parameters: At 36 months, the position of the gingi- been questioned and greater attention has been drawn to thickness of
val margin with respect to the incisal edge of the tooth was statis- the soft tissue margin, or periodontal phenotype (Jepsen et al., 2018;
tically significantly more coronal in the CTG group than in the CMX Barootchi et al., 2020). A limitation of the present study is the lack of
group. The size of the effect was just above the clinically significant measurement of tissue thickness and its variations over time.
difference stipulated for the definition of the non-­inferiority margin Comparing outcomes of the present trial with previous stud-
of this study (0.5 mm). The results, however, were not confirmed ies shows somehow lower frequency of complete root coverage.
TONETTI et al. | 7

Interpretation of this finding needs to take into account at least 5 as- treatment of single maxillary gingival recession with non-­c arious
pects: (i) the focus on multiple adjacent recessions; (ii) the inclusion cervical lesion. A randomized controlled clinical trial. Journal of
Clinical Periodontology, 47, 362–­371.
of initial RT2 recessions in the population (interdental CAL of 1 mm
Cairo, F., Cortellini, P., Pilloni, A., Nieri, M., Cincinelli, S., Amunni, F.,
accepted); (iii) the inclusion of areas with non-­carious cervical lesions, Pagavino, G., & Tonetti, M. S. (2016). Clinical efficacy of coro-
cervical fillings and caries; (iv) the re-­storative approach to the man- nally advanced flap with or without connective tissue graft for
agement of the CEJ defects; and (v) the stringent multicentre design. the treatment of multiple adjacent gingival recessions in the aes-
thetic area: a randomized controlled clinical trial. Journal of Clinical
The present observations, together with the results of the parent trial
Periodontology, 43, 849–­856.
(Tonetti et al., 2018), assist in clinical decision-­making for multiple reces- Cairo, F., Nieri, M., & Pagliaro, U. (2014). Efficacy of periodontal plas-
sion coverage (Tonetti et al., 2014). They confirm that application of CTG tic surgery procedures in the treatment of localized facial gingival
combined with a specifically designed CAF is probably the best approach recessions. A systematic review. Journal of Clinical Periodontology,
41(Suppl 15), S44–­S62.
for root coverage at multiple adjacent recessions (Tonetti et al., 2018).
Chambrone, L., Ortega, M. A. S., Sukekava, F., Rotundo, R., Kalemaj,
Use of a CMX, and avoidance of a CTG donor site, results in shorter Z., Buti, J., & Prato, G. P. P. (2019). Root coverage procedures
time to recovery, less post-­operative morbidity (Tonetti et al., 2018) and for treating single and multiple recession-­t ype defects: An up-
more natural tissue texture and contour (Pelekos et al., 2019). CTG or dated Cochrane systematic review. Journal of Periodontology, 90,
1399–­1422.
CMX results in similar stability of outcomes between 6 and 36 months.
Cortellini, P., Tonetti, M., Baldi, C., Francetti, L., Rasperini, G., Rotundo,
Care must also be exerted when using CMX or collagen-­based products R., Nieri, M., Franceschi, D., Labriola, A., & Prato, G. P. (2009). Does
in subjects with auto-­immune diseases such as rheumatoid arthritis or placement of a connective tissue graft improve the outcomes of
known sensitization to collagen-­based medical products. At this point, coronally advanced flap for coverage of single gingival recessions
in upper anterior teeth? A multi-­centre, randomized, double-­blind,
however, a full cost-­benefit assessment cannot be done as the relative
clinical trial. Journal of Clinical Periodontology, 36, 68–­79.
weights of clinically assessed root coverage, aesthetic scores, decreased
Dai, A., Huang, J. P., Ding, P. H., & Chen, L. L. (2019). Long-­term stability
morbidity, faster recovery, surgery duration and monetary cost of the of root coverage procedures for single gingival recessions: A sys-
CMX device remain unclear. More research is necessary in this area. tematic review and meta-­analysis. Journal of Clinical Periodontology,
46, 572–­585.
Graziani, F., Gennai, S., Roldan, S., Discepoli, N., Buti, J., Madianos, P.,
AC K N OW L E D G E M E N T S
& Herrera, D. (2014). Efficacy of periodontal plastic procedures
Authors wish to recognize the clinical staff of the participating cen- in the treatment of multiple gingival recessions. Journal of Clinical
tres for their excellent clinical work that made this study possible. Periodontology, 41(Suppl 15), S63–­S76.
Guglielmoni, P., Promsudthi, A., Tatakis, D. N., & Trombelli, L. (2001).
Intra-­and inter-­examiner reproducibility in keratinized tissue width
C O N FL I C T O F I N T E R E S T
assessment with 3 methods for mucogingival junction determina-
Authors report no conflict of interest related to this study. tion. Journal of Periodontology, 72, 134–­139.
Jepsen, K., Jepsen, S., Zucchelli, G., Stefanini, M., de Sanctis, M., Baldini,
AU T H O R C O N T R I B U T I O N S N., Greven, B., Heinz, B., Wennstrom, J., Cassel, B., Vignoletti, F.,
& Sanz, M. (2013). Treatment of gingival recession defects with a
MT and PC conceived the study, wrote the protocol, supervised
coronally advanced flap and a xenogeneic collagen matrix: a mul-
quality control, analysed and interpreted the data and wrote the ticenter randomized clinical trial. Journal of Clinical Periodontology,
manuscript. KD and DB assisted in data collection and analysis and 40, 82–­89.
monitored the quality of the data. FC, MA, GC, FG, AG, JH, JM, GR, Jepsen, K., Stefanini, M., Sanz, M., Zucchelli, G., & Jepsen, S. (2017).
Long-­term stability of root coverage by coronally advanced flap
HT contributed to final study design, were the centre PIs and sur-
procedures. Journal of Periodontology, 88, 626–­633.
geons, commented and approved manuscript. Jepsen, S., Caton, J. G., Albandar, J. M., Bissada, N. F., Bouchard, P.,
Cortellini, P., Demirel, K., de Sanctis, M., Ercoli, C., Fan, J., Geurs,
DATA AVA I L A B I L I T Y S TAT E M E N T N. C., Hughes, F. J., Jin, L., Kantarci, A., Lalla, E., Madianos, P. N.,
Matthews, D., McGuire, M. K., Mills, M. P., … Yamazaki, K. (2018).
Data for this study are not available.
Periodontal manifestations of systemic diseases and developmen-
tal and acquired conditions: Consensus report of workgroup 3 of
ORCID the 2017 World Workshop on the Classification of Periodontal
Maurizio S. Tonetti https://orcid.org/0000-0002-2743-0137 and Peri-­Implant Diseases and Conditions. Journal of Clinical
Francesco Cairo https://orcid.org/0000-0003-3781-1715 Periodontology, 45(Suppl 20), S219–­S229.
McGuire, M. K., & Scheyer, E. T. (2010). Xenogeneic collagen matrix with
Filippo Graziani https://orcid.org/0000-0001-8780-7306
coronally advanced flap compared to connective tissue with coro-
Adrian Guerrero https://orcid.org/0000-0003-2093-6079 nally advanced flap for the treatment of dehiscence-­t ype recession
defects. Journal of Periodontology, 81, 1108–­1117.
REFERENCES O'Leary, T. J., Drake, R. B., & Naylor, J. E. (1972). The plaque control re-
cord. Journal of Periodontology, 43, 38.
Barootchi, S., Tavelli, L., Zucchelli, G., Giannobile, W. V., & Wang, H. L.
Pelekos, G., Lu, J. Z., Ho, D. K. L., Graziani, F., Cairo, F., Cortellini, P., &
Gingival phenotype modification therapies on natural teeth: A net-
Tonetti, M. S. (2019). Aesthetic assessment after root coverage
work meta-­analysis. Journal of Periodontology, 91, 1386–­1399.
of multiple adjacent recessions with coronally advanced flap with
Cairo, F., Cortellini, P., Nieri, M., Pilloni, A., Barbato, L., Pagavino, G., &
adjunctive collagen matrix or connective tissue graft: Randomized
Tonetti, M. (2020). Coronally advanced flap and composite resto-
clinical trial. Journal of Clinical Periodontology, 46, 564–­571.
ration of the enamel with or without connective tissue graft for the
8 | TONETTI et al.

Rotundo, R., Genzano, L., Patel, D., D'Aiuto, F., & Nieri, M. (2019). coronally advanced flaps for coverage of multiple adjacent gingival
Adjunctive benefit of a xenogenic collagen matrix associated with recession: Randomized trial assessing non-­inferiority in root cover-
coronally advanced flap for the treatment of multiple gingival re- age and superiority in oral health-­related quality of life. Journal of
cessions: A superiority, assessor-­blind, randomized clinical trial. Clinical Periodontology, 45, 78–­88.
Journal of Clinical Periodontology, 46, 1013–­1023. Tonetti, M. S., Jepsen, S., & WORKING GROUP 2 OF THE EUROPEAN
Stefanini, M., Jepsen, K., de Sanctis, M., Baldini, N., Greven, B., Heinz, WORKSHOP ON, P. (2014). Clinical efficacy of periodontal plas-
B., Wennstrom, J., Cassel, B., Vignoletti, F., Sanz, M., Jepsen, S., & tic surgery procedures: consensus report of Group 2 of the
Zucchelli, G. (2016). Patient-­reported outcomes and aesthetic eval- 10th European Workshop on Periodontology. Journal of Clinical
uation of root coverage procedures: a 12-­month follow-­up of a ran- Periodontology, 41(Suppl 15), S36–­S 43.
domized controlled clinical trial. Journal of Clinical Periodontology, Tonetti, M. S., Pini-­Prato, G., & Cortellini, P. (1993). Periodontal regen-
43, 1132–­1141. eration of human intrabony defects. IV. Determinants of healing
Tavelli, L., Barootchi, S., Cairo, F., Rasperini, G., Shedden, K., & Wang, H. response. Journal of Periodontology, 64, 934–­940.
L. (2019). The effect of time on root coverage outcomes: a network Vignoletti, F., Nunez, J., Discepoli, N., de Sanctis, F., Caffesse, R., Munoz,
meta-­analysis. Journal of Dental Research, 98, 1195–­1203. F., Lopez, M., & Sanz, M. (2011). Clinical and histological healing of
Thoma, D. S., Hammerle, C. H., Cochran, D. L., Jones, A. A., Gorlach, C., a new collagen matrix in combination with the coronally advanced
Uebersax, L., Mathes, S., Graf-­Hausner, U., & Jung, R. E. (2011). flap for the treatment of Miller class-­I recession defects: an exper-
Soft tissue volume augmentation by the use of collagen-­based ma- imental study in the minipig. Journal of Clinical Periodontology, 38,
trices in the dog mandible –­a histological analysis. Journal of Clinical 847–­855.
Periodontology, 38, 1063–­1070. Zucchelli, G., & de Sanctis, M. (2000). Treatment of multiple recession-­
Thoma, D. S., Jung, R. E., Schneider, D., Cochran, D. L., Ender, A., Jones, type defects in patients with esthetic demands. Journal of
A. A., Gorlach, C., Uebersax, L., Graf-­Hausner, U., & Hammerle, Periodontology, 71, 1506–­1514.
C. H. (2010). Soft tissue volume augmentation by the use of
collagen-­based matrices: a volumetric analysis. Journal of Clinical
Periodontology, 37, 659–­666.
How to cite this article: Tonetti MS, Cortellini P, Bonaccini D,
Thoma, D. S., Villar, C. C., Cochran, D. L., Hammerle, C. H., & Jung, R. E.
(2012). Tissue integration of collagen-­based matrices: An experi-
et al. Autologous connective tissue graft or xenogenic
mental study in mice. Clinical Oral Implants Research, 23, 1333–­1339. collagen matrix with coronally advanced flaps for coverage of
Tonetti, M. S., Cortellini, P., Pellegrini, G., Nieri, M., Bonaccini, D., Allegri, multiple adjacent gingival recession. 36-­month follow-­up of a
M., Bouchard, P., Cairo, F., Conforti, G., Fourmousis, I., Graziani, F., randomized multicentre trial. J Clin Periodontol. 2021;00:1–8.
Guerrero, A., Halben, J., Malet, J., Rasperini, G., Topoll, H., Wachtel,
https://doi.org/10.1111/jcpe.13466
H., Wallkamm, B., Zabalegui, I., & Zuhr, O. (2018). Xenogenic col-
lagen matrix or autologous connective tissue graft as adjunct to