Volume 81 • Number 12

Root Coverage Esthetic Score After

Treatment of Gingival Recession:
An Interrater Agreement Multicenter
Study
Francesco Cairo,* Michele Nieri,* Marcello Cattabriga,† Pierpaolo Cortellini,‡ Sergio De Paoli,§
Massimo De Sanctis,i Alberto Fonzar,¶ Luca Francetti,# Mauro Merli,** Giulio Rasperini,††
Maurizio Silvestri,‡‡ Leonardo Trombelli,§§ Giovanni Zucchelli,ii and Giovan Paolo Pini-Prato*

Background: The root coverage esthetic score (RES) sys-

tem was proposed for evaluating esthetic outcomes of root
coverage procedures. The aim of this multicenter study is to
assess the interrater agreement of the RES among expert peri-
odontists.
Methods: Eleven periodontists were selected in different
clinical centers. Each operator had ‡15 years of experience

T
he effectiveness of root coverage
in mucogingival surgery. Each periodontist was trained to procedures is quantified by the
use RES before the beginning of the study. Subsequently, level of the gingival margin with
baseline and post-treatment (6 months) photographs of 41 respect to the cemento-enamel junction
Class I and II gingival recessions in 41 patients were separately (CEJ) after surgery. Complete root cov-
given to each operator who evaluated the outcomes according erage (CRC) is achieved when the gin-
to the RES method. A two-way random interclass correlation gival margin is located coronally to the
coefficient and 95% confidence interval (CI) were used to as- CEJ. The mean recession reduction and
sess the global interrater agreement for RESs. the mean percentage of root coverage
Results: The total interrater agreement for RESs was 0.92 are also used to offer clinicians a descrip-
(95% CI: 0.88 to 0.95), which indicated that an almost perfect tive efficacy of the tested techniques in
agreement was achieved. clinical trials. A systematic review (SR)1
Conclusion: Tested individually by a group of periodontists, described the possible deceptive nature
the RES seems to be a reliable method for assessing the es- of the percentage of root coverage when
thetic outcomes of root coverage procedures. J Periodontol used to assess the final success of the
2010;81:1752-1758. procedure in clinical studies. Other SRs2-4
suggested that CRC should be considered
KEY WORDS
the main treatment outcome of these
Esthetics, dental; gingival recession; plastic surgery; gingival procedures. A study5 that evaluated the
recession/therapy; validation study. esthetic perception of simulated root
coverage procedures in photographs
* Department of Periodontology, University of Florence, Florence, Italy. showed that patients, dentists, and peri-
† Private practice, Rome, Italy.
‡ Tuscany Academy of Dental Research, Florence, Italy. odontists consider CRC the most impor-
§ Private practice, Ancona, Italy. tant outcome.
i Department of Periodontology, University of Siena, Siena, Italy.
¶ Private practice, Campoformido-Udine, Italy. Nevertheless, the position of the gingi-
# Department of Health Technologies, Galeazzi Orthopedic Institute University of Milan,
Milan, Italy.
val margin after surgery is restrictive and
** Private practice, Rimini, Italy. may not be adequate for assessing the
†† Unit of Periodontology, Dental Clinic, Department of Surgical, Reconstructive, and
Diagnostic Sciences, University of Milan. overall esthetic outcome of the treatment.
‡‡ Private practice, Pavia, Italy. In fact, an impaired esthetic appearance
§§ Research Center for Study of Periodontal Disease, University of Ferrara, Ferrara, Italy.
ii Department of Periodontology, University of Bologna, Bologna, Italy. occurs in cases of complete failure or
partial root coverage and when CRC is
achieved along with an irregular profile
of the gingival margin, poor color match,

doi: 10.1902/jop.2010.100278

1752
J Periodontol • December 2010
alteration of the position in the mucogingival junction
(MGJ), and presence of scar tissue.4 Therefore, CRC,
per se, cannot be considered the sole factor for defining
the full success of the treatment of gingival recession.
An SR4 on root coverage procedures showed that
only few randomized clinical trials evaluated the es-
thetic outcomes of the treatment, and a meta-analysis
could not be performed because few and only hetero-
geneous data are available. Recently, two different
methods were proposed for the assessment of the cos-
metic outcomes of root coverage procedures. A five-
point ordinal improvement scale (i.e., poor, fair, good,
very good, and excellent) was suggested in a before–
after panel scoring system.6 Furthermore, the root
coverage esthetic score (RES) system was introduced
to evaluate the overall esthetic outcome after root
coverage procedures.7 This score is based on the
evaluation of five variables: the level of the gingival
margin, marginal contour, soft tissue surface, position
of the MGJ, and gingival color. RES values vary from
0 (final residual recession equal to or higher than the
baseline recession) to 10 (CRC associated with the
fulfillment of the other four variables). RESs were fur-
ther used to evaluate the esthetic results of localized
recessions in a study8 that compared the use of a sub-
epithelial connective tissue graft versus an acellular
dermal matrix allograft seeded with autologous gingi-
val fibroblasts. Even if the preliminary proposal of the
RES seems promising, to our knowledge, its interrater
agreement has not been statistically tested.
The aim of this study is to assess the interrater
agreement of the RES among expert periodontists
in different clinical centers.
MATERIALS AND METHODS
Prior to starting the study, the primary periodontist
(FC) collected the baseline and 6-month post-treat-
ment photographs of single Miller Class I and II gin-
gival recessions from 87 patients (33 male and 54
female, aged 19 to 57 years; mean age: 33.5 – 9.1
years) who were treated from January 2007 to June
2008.9 The defects were treated with different root
coverage procedures at the Department of Periodon-
tology, University of Florence, by three operators
(FC, Jana Mervelt, and Sandro Cincinelli, Department
of Periodontology, University of Florence, Florence,
Italy) with different levels of clinical experience. Before
any procedure, each patient signed a written informed
consent form in accordance with the Helsinki Declara-
tion of 1975, as revised in 2000. No ethical committe
approval was required by the University of Florence
for this type of study.
Selection and Management of Photographs
Pretreatment and 6-month follow-up photographs at
a magnification ratio of 1:1 were selected. Regular
Cairo, Nieri, Cattabriga, et al.
photography with no custom-built frame was used.
A standard shooting protocol was applied positioning
the target camera perpendicular to the long axis of the
experimental tooth. A preliminary sample of 87 pre-
and post-treatment pictures in 87 different patients
were initially screened and evaluated for their quality
by the same periodontist (FC). Comparable pre- and
post-treatment images were selected after the assess-
ment of slide quality. Pictures with image distortion,
different framing, color contrast, and a lack of visibil-
ity of CEJ and/or MGJ were excluded from the study.
Of the previous screened sample of 87 pre- and post-
treatment pictures, a final sample of 41 gingival reces-
sions in 41 different patients was selected by the same
periodontist (FC) and used for this study.
Selected preoperative and postoperative slides
were digitized at 300 dots per inch with a professional
scanner. All images were displayed using specific
software.¶¶ Care was taken to avoid image distortion
and to maintain the original proportions. A stent to ori-
ent the single slide was placed over the scanner to im-
prove the scanning procedure. In addition, no color
adjustment was performed, and the original bright-
ness and contrast was maintained. Finally, care was
taken to standardize a similar image size for both pre-
and post- treatment photographs included in each
slide.
Each preoperative and postoperative image for
each corresponding clinical case was imported and
paired in a file for a slide presentation.## A file*** that
contained a table with identification numbers corre-
sponding to each clinical case was also prepared for
data collection.
Selection of Clinical Centers and Training
of Operators
Eleven periodontists (MC, PC, SDP, MDS, AF, LF, MM,
GP, MS, LT, GZ) in different clinical centers were se-
lected to participate in the study. All of them were
experienced operators in mucogingival surgery and
active members of the Italian Society of Periodontol-
ogy. In addition, all had published articles on muco-
gingival surgery in international periodontal journals
and had ‡15 years of clinical experience.
Prior to starting the study, each operator was
trained on the use of RES by the same periodontist
(FC). According to the RES system, each variable
and corresponding value of the score was explained.
Preliminary clinical cases were provided along with
the corresponding RES values to show its applicabil-
ity. All clarifications were provided before the study.
¶¶ Adobe Photoshop, version 7.0, Adobe Systems Europe, Uxbridge, U.K.
## PowerPoint, Microsoft, Redmond, WA.
*** Excel, Microsoft.
1753
Esthetic Evaluation After Root Coverage Volume 81 • Number 12

Esthetic Assessment With RES Score failure of the root coverage procedure) independent
After the completion of training, the file containing pre- of color, the presence of scarring, the gingival margin,
and post-surgical pictures and the corresponding data or MGJ. A score of 0 was also assigned when a partial
collection file was provided to evaluate the selected or total loss of interproximal papilla (a black triangle)
clinical cases according to RES indications. Each clini- occurred after the treatment. The esthetic evaluation
cian assessed the photographs separately in the corre- with RES of two different clinical cases is presented
sponding clinical center. No time limitation was given in Figures 2 through 5. The completed data collection
to the raters to present their personal evaluations. file with the corresponding RES values was returned
The RES system 7 evaluated five variables at to the primary author (FC).
6-months after surgery: the level of the gingival mar-
Statistical Analyses
gin, marginal tissue contour, soft tissue texture, MGJ
Descriptive statistics with the median, quartiles, and
alignment, and gingival color (Fig. 1). Because CRC
box plot of each rater were performed.
was the treatment goal, whereas other variables were
secondary endpoints, 60% of the RES value was at-
tributed to the level of gingival margin, and 40% was
attributed to the other four variables.
1) Gingival margin level (GM): Score 0: failure of
root coverage (gingival margin apical or equal to
the baseline recession), Score 3: partial root coverage,
Score 6: complete root coverage with no detectable
CEJ; 2) Marginal Tissue Contour (MTC): Score 0:
irregular gingival margin (it does not follow CEJ);
Score 1: proper marginal contour/scalloped gingival
margin (following CEJ); 3) Soft tissue texture
(STT): Score 0: presence of scar formation and/or
keloid-like appearance; Score 1: absence of scar or
keloid formation; 4) Muco-gingival junction (MGJ):
Score 0 : MGJ not aligned with MGJ on adjacent teeth;
Score 1: MGJ aligned with MGJ on adjacent teeth; and
5) Gingival Color: Score 0 : color of tissue differs from
gingival color on adjacent teeth; Score 1: normal color Figure 2.
and integration with the adjacent soft tissues. A 3-mm gingival recession at the lateral left incisor.
The best esthetic score was 10. A score of 0 was as-
signed when the final gingival margin position was
equal or apical to the previous recession depth (i.e.,

Figure 1.
Single RES variables involved in the esthetic assessment: gingival margin
level (GM), marginal tissue contour (MTC), soft tissue texture (STT),
MGJ, and gingival color (GC).
Figure 3.
Defect was treated by means of coronally advanced flap technique. CRC
was achieved 6 months after therapy. The final RES value was 10 (CRC =
6, scalloped gingival margin = 1, scar formation = 1, MGJ alignment = 1,
and gingival color = 1), with full esthetic success.

1754
J Periodontol • December 2010 Cairo, Nieri, Cattabriga, et al.

Figure 4. Figure 5.
A 4-mm gingival recession at the upper left maxillary cuspid. Defect was treated by means of a connective tissue graft procedure with
double papilla-flap technique. CRC was achieved 6 months after therapy.
The final RES value was 7 (CRC = 6, scalloped gingival margin = 1,
The two-way random intraclass correlation coeffi- presence of scar formation = 0, poor MGJ alignment = 0, and gingival
color = 0). Even if CRC was achieved, the case shows an impaired final
cient (ICC) and 95% confidence interval (CI) were
esthetic result.
used to assess the global interrater agreement for
the RES. The interrater agreement of the single RES
variables (the level of the gingival margin, marginal
contour, soft tissue surface, MGJ position, and gin-
gival color) was also conducted. These statistical Table 1.
analyses were performed with software.†††
Median and Quartiles of RES of Each
A six-level nomenclature was used to assess the
level of agreement:10 poor agreement = <0.00, slight Examiner
agreement = 0.00 to 0.20, fair agreement = 0.21 to
0.40, moderate agreement = 0.41 to 0.60, substantial Rater Median Lower Quartile Upper Quartile
agreement = 0.61 to 0.80, and almost perfect agree- 11 8 4 10
ment = 0.81 to 0.92.
7 7 4 10
RESULTS 6 7 4 9.5
The defects were treated by three different operators
(FC, Jana Mervelt, and Sandro Cincinelli) at the De- 9 7 4 9
partment of Periodontology, University of Florence, 4 7 4 9
by means of the coronally advanced flap procedure
alone11 or in combination with a connective tissue 2 7 4 9
graft12 or enamel matrix protein,13 double papilla flap 8 7 4 9
plus connective tissue graft,14 epithelialized free gin-
gival graft,15 and lateral positioned flap.16 The base- 1 7 4 8
line recession depth (mean – SD) was 3.4 – 1.3 mm 5 7 3.5 8
(minimum: 2 mm; maximum: 7 mm), and the mean
final residual recession was 1.2 – 1.8 mm (minimum: 10 7 4 7
0 mm; maximum: 5 mm). Twenty-one gingival reces- 3 6 4 8
sions resulted in CRC.
Table 1 shows the median and quartiles of RES of
each examiner. Slight differences in terms of median marginal contour, soft tissue surface, MGJ position,
individual RES values were detected among the raters and gingival color) are shown in Table 2. The assess-
but did not exceed two points. The box plot of the eval- ment of the level of the gingival margin was the more
uation of all involved clinicians is shown in Figure 6. reliable variable (ICC = 0.92), whereas the assessment
The two-way random ICC (95% CI) used to assess
the global interrater agreement for RES was 0.92
††† R software, version 2.9.2, package ‘‘irr,’’ The R Foundation for Sta-
(0.88 to 0.95). The results of the single RES variable tistical Computing, Institute for Statistics and Mathematics, Vienna,
reliability (assessment of the level of gingival margin, Austria.

1755
Esthetic Evaluation After Root Coverage Volume 81 • Number 12

Table 2. literature, few studies17,18 attempted to evaluate the

esthetic outcomes of root coverage procedures with
Results of Single RES-Variable Reliability
different approaches. In a comparative study,18 a four-
point scale (poor, fair, good, and excellent) was used
Variable ICC 95% CI
to rate the overall esthetic satisfaction by a masked
Level of gingival margin 0.92 0.88 to 0.95 clinician and masked patient. More recently, a similar
approach using a qualitative scale (poor, fair, good,
Marginal tissue contour 0.63 0.52 to 0.74
very good, and excellent) was tested by professional
MGJ 0.54 0.43 to 0.67 and non-professional operators.6
The RES system was proposed in a preliminary
Soft tissue texture 0.49 0.38 to 0.63
study to evaluate the esthetic outcomes of root cover-
Gingival color 0.61 0.50 to 0.73 age techniques.7 The rationale for the RES was the ob-
servation that the achievement of CRC per se cannot
be considered a full esthetic suc-
cess. Even when CRC is obtained,
esthetics can be impaired if an ir-
regular profile of the gingival mar-
gin, poor color match, alteration
of the MGJ position, and presence
of scar tissue are detectable after
treatment. Because the final target
is the full esthetic success, the
achievement of CRC along with
soft tissue anatomy comparable
and indistinguishable from adja-
cent tissue should be considered
the treatment goal in studies deal-
ing with root coverage procedures.
The present study was aimed
at evaluating the interrater agree-
ment of the RES in different clinical
centers. The results show that RES
is a reliable method for assessing
the esthetic outcomes of root cov-
erage procedures; an almost per-
fect agreement was reached with
an ICC = 0.92. Based on the classi-
fication of Landis and Koch,10 the
RES may be considered a good
instrument to evaluate the es-
thetic outcomes of root coverage
Figure 6. procedures.
Box plot showing evaluations of RES by different raters. Red lines inside the bars identify medians,
A possible hypothesis to ex-
whereas black lines correspond to mean values.
plain the high ICC value of the
RES among the raters is the skill
of the MGJ (ICC = 0.54) and soft tissue texture (ICC = of experienced periodontists to assess root cover-
0.49) showed less agreement. age outcomes on slides. Furthermore, a positive ef-
fect of the preliminary training should be taken into
account. In fact, one study reported a positive effect
DISCUSSION of training in improving the reliability of an observer
In the last decade, several sophisticated techniques evaluation to assess the quality of medical care.19 A
were published for the treatment of single and multiple similar effect of observer training was reported in
gingival recessions. Even if the achievement of the a study that evaluated esthetics of root coverage
CRC is the main treatment outcome,2-4 this variable outcomes using a qualitative scale.6 Even if a fair to
may be restrictive and unable to describe the final moderate reliability was observed when trained and
esthetic outcome of the surgery. In the periodontal untrained periodontists were compared, an improved

1756
J Periodontol • December 2010
interrater agreement was observed among trained
observers (k = 0.68).6
The analysis of the RES single variables suggested
some considerations. The level of the gingival margin
with respect to the CEJ that reveals partial root cover-
age or CRC proved to be the more reliable variable
among raters (ICC = 0.92) (Table 2). An easy explana-
tion of this finding is related to the relative simplicity of
visualizing and comparing the position of the gingival
margin on the pre- and post-treatment pictures. How-
ever, raters showed disagreements in the evaluation of
some clinical cases. A possible reason may be related
to the difficultly in identification of CEJ. A recent study
showed that the evaluation of the position of CEJ in
teeth with gingival recession may raise some concerns
among different operators.20
On the other hand, the assessment of the MGJ po-
sition (ICC = 0.54) showed less agreement among
raters. This may have been due to the difficulty in eval-
uating the alignment of MGJ with adjacent teeth after
therapy on photos because the borderline between
the keratinized tissue and the alveolar mucosa was
often unclear in the clinical examinations as well.
Furthermore, the tissue texture score resulted in
moderate agreement (ICC = 0.49) because the use
of photographs might have influenced the examiner’s
self-perception during the assessment of soft tissue
characteristics. However, both variables showed an
acceptable agreement among raters scored as mod-
erate by the classification of Landis and Koch.10
Limits of the present study may be related to the
use of photographs as opposed to a direct clinical
assessment of the patient. A patient evaluation by
means of the RES should be useful for completing
the overall esthetic assessment of root coverage pro-
cedures. In addition, the weight attributed to CRC
in the RES may have influenced the final results. Fi-
nally, possible factors that may influence the esthetic
evaluation after root coverage need to be explored
in further studies.
CONCLUSION
This study shows that RES seems to be a reliable
method for assessing the esthetic outcomes of root
coverage procedures.
ACKNOWLEDGMENTS
The study was self-founded by the authors and their
institutions. The authors report no conflicts of interest
related to this study.
REFERENCES
1. Greenwell H, Bissada NF, Henderson RD, Dodge JR.
The deceptive nature of root coverage results. J
Periodontol 2000;71:1327-1337.
Cairo, Nieri, Cattabriga, et al.
2. Pagliaro U, Nieri M, Franceschi D, Clauser C, Pini-
Prato GP. Evidence-based mucogingival therapy.
Part 1: A critical review of the literature on root
coverage procedures. J Periodontol 2003;74:709-
740.
3. Clauser C, Nieri M, Franceschi D, Pagliaro U, Pini-
Prato GP. Evidence-based mucogingival therapy. Part
2: Ordinary and individual patient data meta-analyses
of surgical treatment of recession using complete root
coverage as the outcome variable. J Periodontol 2003;
74:741-756.
4. Cairo F, Pagliaro U, Nieri M. Treatment of gingival
recession with coronally advanced flap procedures: A
systematic review. J Clin Periodontol 2008;35(Suppl.
8):136-162.
5. Rotundo R, Nieri M, Mori M, Clauser C, Prato GP.
Aesthetic perception after root coverage procedure. J
Clin Periodontol 2008;35:705-712.
6. Kerner S, Katsahian S, Sarfati A, et al. A comparison
of methods of aesthetic assessment in root coverage
procedures. J Clin Periodontol 2009;36:80-87.
7. Cairo F, Rotundo R, Miller PD, Pini Prato GP. Root
coverage esthetic score: A system to evaluate the
esthetic outcome of the treatment of gingival recession
through evaluation of clinical cases. J Periodontol 2009;
80:705-710.
8. Jhaveri HM, Chavan MS, Tomar GB, Deshmukh VL,
Wani MR, Miller PD Jr. Acellular dermal matrix seeded
with autologous gingival fibroblasts for the treatment
of gingival recession: A proof-of-concept study. J Peri-
odontol 2010;81(2):616-625.
9. Miller PD Jr. A classification of marginal tissue re-
cession. Int J Periodontics Restorative Dent 1985;5(2):
8-13.
10. Landis JR, Koch GG. The measurement of observer
agreement for categorical data. Biometrics 1977;33:
159-174.
11. Allen EP, Miller PD Jr. Coronal positioning of existing
gingiva: Short term results in the treatment of shallow
marginal tissue recession. J Periodontol 1989;60:316-
319.
12. Zucchelli G, Amore C, Sforzal NM, Montebugnoli L, De
Sanctis M. Bilaminar techniques for the treatment of
recession-type defects. A comparative clinical study.
J Clin Periodontol 2003;30:862-870.
13. Modica F, Del Pizzo M, Roccuzzo M, Romagnoli R.
Coronally advanced flap for the treatment of buccal
gingival recessions with and without enamel matrix
derivative. A split-mouth study. J Periodontol 2000;
71:1693-1698.
14. Harris RJ. The connective tissue with partial thickness
double pedicle graft: The results of 100 consecutively-
treated defects. J Periodontol 1994;65:448-461.
15. Miller PD Jr. Root coverage using the free soft tissue
autograft following citric acid application. III. A suc-
cessful and predictable procedure in areas of deep-
wide recession. Int J Periodontics Restorative Dent
1985;5(2):14-37.
16. Espinel MC, Caffesse RG. Lateral positioned pedicle
sliding flap-revised technique in the treatment of local-
ized gingival recessions. Int J Periodontics Restorative
Dent 1981;1(5):42-51.
17. Wang HL, Bunyaratavej P, Labadie M, Shyr Y, MacNeil
RL. Comparison of 2 clinical techniques for treatment
of gingival recession. J Periodontol 2001;72:1301-
1311.
1757
Esthetic Evaluation After Root Coverage
18. Aichelmann-Reidy ME, Yukna RA, Evans GH, Nasr
HF, Mayer ET. Clinical evaluation of acellular allograft
dermis for the treatment of human gingival recession.
J Periodontol 2001;72:998-1005.
19. Koran LM. The reliability of clinical methods, data and
judgments (second of two parts). N Engl J Med 1975;
293:695-701.
20. Pini-Prato G, Franceschi D, Cairo F, Nieri M, Rotundo
R. Classification of dental surface defects in areas
1758
Volume 81 • Number 12
of gingival recession. J Periodontol 2010;81:885-
890.
Correspondence: Dr. Francesco Cairo, via Mino Celsi 16,
53100 Sant’Andrea, Siena, Italy. E-mail: cairofrancesco@
virgilio.it.
Submitted May 9, 2010; accepted for publication June 14,
2010.