Reproducibility and validity of anterior implant esthetic indices: A review.

BACKGROUND: The purpose of this review was to narrate about the reproducibility and validity of different indices evaluating esthetic aspects in anterior single implant-supported restorations.
MATERIALS AND METHODS: An electronic search of Medline, Scopus, Embase, Cochrane Central, and Web of Science databases was performed using the keywords "dental implants," "anterior esthetics," "esthetic score," and "esthetic index." Besides, a manual search of dental implant journals was carried out.
RESULTS: The electronic search revealed 932 titles. After further review, 14 articles fulfilled the eligibility criteria and were included in this review. Because of the heterogeneity of the study designs, interventions, and parameters used for assessment of esthetics, no meta-analysis could be performed.
CONCLUSION: Many indices have been proposed for the evaluation of the esthetic aspects of single implant-supported reconstructions in the anterior maxilla. All of them have some advantages and drawbacks that this review pointed out. The evidence level of studies used for the validation of these indices is poor. It is necessary to achieve a consensus on the tools for assessment of the esthetic aspect and perform evidence-based studies to validate an appropriate index. Copyright:

INTRODUCTION

Replacement of single or multiple missing teeth with implant-supported prosthesis is a dependable treatment modality with success rates as high as 95%.[123] With changing times, success criteria for anterior implants have changed from mere survival and function to esthetics.[4] A combination of a visually pleasing prosthesis and healthy, harmoniously scalloped peri-implant soft tissues is the key to achieve adequate esthetics in anterior implant restorations.[5] Objective assessment done by professionals and by patients minimizes the subjective nature of esthetics. Esthetic indices not only give insight about the esthetic outcome but also evaluate the outcomes as a function of time regarding the stability of implant-supported restoration.

Jemt's papilla index (PI)[6] was the first to assess the papilla fill of the anterior implant restoration objectively though it considered only one factor for esthetic assessment. Annibali et al.,[7] in their review, had found that this index was the most commonly reported until 2009 for evaluation of esthetics in anterior single implant crowns. The Implant Esthetic Score (IAS) proposed by Testori et al.[8] comprised five variables: presence and stability of the mesiodistal papilla, ridge stability buccopalatally, texture and color of the peri-implant soft tissue as well as gingival contour. Taking into account seven different variables (height of mesial and distal papilla, level, contour, color, and texture of soft tissue and alveolar process deficiency) to objectively assess the peri-implant soft tissue, Furhauser et al.[9] proposed the Pink Esthetic Score (PES). Mesiodistal width, incisal edge position, labial surface convexity, color, translucency and surface characteristics of the crown along with labial margin, interproximal embrasures, surface contour, and color and texture of the labial mucosa are the nine variables considered by Meijer et al.[10] in the “Implant Crown Esthetic Index” (ICAI). The major drawback of this index is its scoring pattern providing five points for gross deviation; hence, the single gross deviation can result in poor esthetic results. Belser et al.[11] developed a new complete esthetic index that combines the PES with White Esthetic Score (PES/WES), which objectively assessed the peri-implant soft tissues and featured inherent to the restoration. The five PES parameters are mesial papilla, distal papilla, the curvature of the facial mucosa, level of the facial mucosa, and root convexity/soft-tissue color and texture. The WES parameters are tooth form, outline, and volume of the crown, the color of the crown, surface texture, and translucency, and characterization. In 2010, Juodzbalys and Wang[12] developed and validated a Complex Esthetic Index (CEI) for evaluating esthetics of implant-supported restorations and peri-implant soft and hard tissues. This comprehensive index is composed of Soft-tissue index (S), Predictive index (P), and Restoration index (R). Hosseini and Gotfredsen[13] established the Copenhagen Index Score (CIS), which was composed of six scores: crown morphology, color match, symmetry/harmony, mucosal discoloration, and PI mesially and distally. Encompassing three subjective and six objective criteria, Tettamanti et al.[14] proposed the Peri-Implant and Crown Index and further compared that with PES/WES and ICAI. The Implant Restoration Esthetic Index, a novel index developed by Li et al.,[15] included six soft-tissue parameters and six crown-related parameters. The authors compared it with another widely used index (PES/WES).

Reproducibility is the degree of concordance of multiple measurements of the same subject, carried out under different conditions. Intra-observer and inter-observer agreement values are considered to predict the reproducibility of an index. Different studies exist concerning esthetic assessment;[141617181920] however, there is no uniform consensus among clinicians on this topic. As far as we know, this is the first review specifically aiming at comparing the reproducibility and validity of several esthetic indices of anterior single implant-supported restorations. A secondary aim was the identification of the strength and the most common drawbacks of existing indices, to help in designing specific tools, validated by specific evidence-based studies.

MATERIALS AND METHODS

An electronic search in Medline, Scopus, Embase, Cochrane Central, and Web of Science databases was conducted using the keywords “dental implants,” “anterior esthetics,” “esthetic score,” and “esthetic index” in combination to one another. The search was done for studies reporting on single-tooth implant-supported restorations in the maxillary esthetic zone for which esthetic evaluation was done in the form of esthetic index or esthetic score.

Inclusion criteria

Studies with a minimum sample size of ten patients that have proposed a new esthetic index and validated the reproducibility of the same or studies that have checked the reproducibility and validity of esthetic indices already present in the literature were included.

Exclusion criteria

Studies with lack of reproducibility assessment, lack of esthetic assessment by the clinician, lack of explicit mention of anterior single-tooth implant, case series with less than ten patients, lack of adjacent natural tooth, and no index being used for esthetic assessment were excluded [Table 1]. No limitation was set regarding the study design and the language.

Table 1

Exclusion criteria

Study	Reasons for exclusion
Al-Dosari et al.[21]	Lack of reproducibility assessment
Altay[22]	Lack of reproducibility assessment
Chang et al.[23]	Lack of reproducibility assessment
Chang et al.[24]	Esthetic index was not used for assessment
Cosyn and De Rouck 2009[25]	Esthetic index was not used for assessment
Cosyn et al.[26]	Lack of reproducibility assessment
Cosyn et al.[27]	Lack of reproducibility assessment
den Hartog et al.[28]	Lack of reproducibility assessment
Dueled et al.[29]	No natural contralateral tooth, no esthetic index was used
Evans and Chen[30]	Lack of reproducibility assessment
Fava[31]	Lack of reproducibility assessment
Gotfredsen[4]	None of the esthetic indexes was used for assessment
Hall et al.[32]	Lack of reproducibility assessment
Hartlev et al.[33]	Lack of reproducibility assessment
Jones and Martin[34]	Assessment was not done by clinician
Lai et al.[35]	Lack of reproducibility assessment
Luo et al.[36]	Lack of reproducibility assessment
Meijndert et al.[37]	Lack of reproducibility assessment
Misje et al.[38]	None of the esthetic indexes was used for assessment
Petsos[39]	Lack of reproducibility assessment
Suphanantachat et al.[40]	None of the esthetic indexes was used for assessment
Testori et al.[8]	Case report
Vanlıoğlu[41]	Lack of reproducibility assessment

In addition, implant dentistry-related journals were manually searched to include all pertinent publications: Clinical Oral Implants Research, Journal of Prosthetic Dentistry, International Journal of Oral and Maxillofacial Implants, Journal of Prosthodontics, International Journal of Prosthodontics, Journal of Periodontology, Clinical Implant Dentistry and Related Research, The International Journal of Periodontics and Restorative Dentistry, and European Journal of Oral Implantology.

The search was performed independently by two reviewers (GS and SP). All titles obtained by the electronic search were screened for meeting the inclusion criteria. The titles containing insufficient information for inclusion were selected for the abstract evaluation. Articles were selected for evaluation of full text when synonyms related to esthetic evaluation were present in the abstract. Full texts of carefully chosen articles were appraised entirely before the final inclusion. Any disagreement about inclusion was resolved after discussion with a third reviewer (SWP).

RESULTS

Searches of PubMed, Scopus, Embase, Cochrane Central, and Web of Science databases generated 932 articles. The last electronic search was performed on February 20, 2019. After removing the duplicates, 740 titles were retrieved. Of which, 263 abstracts were selected for further evaluation. After a meticulous screening of abstracts, 37 articles were subjected to full-text assessment. Finally, 14 studies that validated or compared the reproducibility of the esthetic index in a single anterior implant-supported crown were selected for inclusion [Figure 1]. A total of 23 studies were excluded with proper reasons for exclusion [Table 1]. Standardized descriptive tables were prepared to present the relevant data, which was extracted by two reviewers independently (GS and SKP).

Figure 1

Search strategy flow diagram. n – Number of articles

All characteristics of included studies are presented in Table 2. This review included 14 studies evaluating 641 implants in 597 patients with a minimum follow-up duration of 1 year. The details of the studies, including the nature of the assessor, assessment interval, mode of assessment, and the criterion of the rating scale for each index are recorded for included studies.

Table 2

Characteristics of the studies

Study and year	Number of patients/implants	Follow up period (in months)	Assessor	Assessment interval (in days)	Method	Esthetic index	Rating scale
Jemt 1997[6]	21/25	18	NR	11	Photographs	PI	5-point rating scale 0- no papilla 1 - <half of height 2 - half or more 3 - complete papilla fill 4 - hyperplastic papilla
Fürhauser et al., 2005[9]	30/30	15-143	20 assessors (5 oral surgeons, 5 prosthodontists, 5 orthodontists, 5 dental students)	28	Photographs	PES	0-1-2 scoring system, maximum PES=14
Meijer et al., 2005[10]	24/24	NR	4 assessors (2 oral surgeons, 2 prosthodontists)	14	Photographs	ICAI	No deviation- score 0 Slight deviation - score 1 Major deviation - score 5
Gehrke et al., 2008[16]	30/30	NR	15 assessors (3 general dentists, 3 oral surgeons, 3 orthodontists, 3 postgraduate students in implant dentistry, 3 lay people)	28	Photographs	PES	0-1-2 scoring system
Gehrke et al., 2009[17]	23/23	NR	10 assessors (2 general dentists, 2 prosthodontists, 2 oral surgeons, 2 orthodontists, 2 dental technicians)	28	Photographs	ICAI	No deviation - score 0 Slight deviation - score 1 Major deviation - score 5
Belser et al., 2009[11]	45/45	24-48	2 assessors (prosthodontist)	NR	Photographs dental casts	PES/WES	3-point rating scale Score of 2, 1, or 0 Maximum PES/WES=20
Cho et al., 2010[18]	41/41	102	8 assessors (2 periodontists, 2 prosthodontists, 2 orthodontists and 2 senior dental students)	28	Photographs dental casts	PES/WES	3-point rating scale Score of 2, 1, or 0
Juodzbalys and Wang 2010[12]	50/50	NR	2 assessors (oral surgeons)	14	Direct visualization Periapical radiographs	CEI	Adequate (rating 20%), compromised (rating 10%) deficient (rating 0%)
Vilhjálmsson 2011[19]	50/56	12	2 assessors (blinded)	NR	Photographs	PES ICAI mod ICAI	For mod ICAI No deviation - score 0 Slight deviation - score 1, major/gross deviation - 2
Hosseini and Gotfredsen 2012[13]	34/66	NR	13 assessors (2 prosthodontists and 11 dental students)	7	Photographs	CIS	Four-point rating scale 1. Excellent 2. Satisfactory 3. Moderate 4. Poor
Vaidya et al., 2015[20]	20/20	NR	14 assessors (2 orthodontists, 2 prosthodontists, 2 oral surgeons, 2 periodontists, 2 dental technicians, 2 dental assistants, 2 postgraduate students)	7	Photographs	PES/WES mod ICAI	For PES/WES 3-point rating scale Score 2, 1, or 0 Maximum PES/WES=20 For mod ICAI No deviation - score 0 Slight deviation - score 1 Major deviation - score 2
Tettamanti et al., 2016[14]	15/15	12	40 assessors (10 prosthodontist, 10 orthodontist, 10 general dentists, 10 lay people)	14	Photographs dental casts	PICI PES/WES ICAI	100 mm Visual Analog scale 0-600 patient 0 patient=0% 300 patient=50% 600 patient=100%
Li et al., 2017[15]	25/27	24	20 assessors (10 graduate students, 10 prosthodontists)	14	Intraoral facial occlusal photograph	IREI PES/WES	100 mm Visual Analog Scale 0-600 patient Threshold score of 400
Hof et al., 2018[42]	189/189	NR	5 assessors (1 general practitioner, 1 prosthodontist, 1 surgeon, 1 orthodontist, 1 student)	28	Photographs	PI PES ICAI PES/WES CEI IAS SES Rompen index	As per the scoring scale defined by each index.

PI – Papilla index; PES – Pink Esthetic Score; ICAI – Implant Crown Esthetic Index; PES/WES – Pink Esthetic Score/White Esthetic Score; mod-ICAI – Modified Implant Crown Esthetic Index; CEI – Complex Esthetic Index; CIS – Copenhagen Index Score; PICI – Peri-Implant and Crown Index; IREI – Implant Restoration Esthetic Index; IAS – Implant Esthetic Score; SES – Subjective Esthetic Score; NR – Not reported

The number of examiners and assessment intervals were different in all studies. The mode of assessment was heterogeneous and varied. Ten of 14 studies had used photographs alone[691013151617192042] as the mode of assessment, whereas photographs in combination with the diagnostic cast were used in other three studies[111418] Only one study[12] had used periapical radiographs as well as direct visualization on a patient's mouth. The PES/WES was the most repeatedly used index in the included studies (n = 6), followed by ICAI used by five studies. Objective evaluation of esthetic indices has been done using either a numerical scale (NS) or a Visual Analog Scale (VAS).

The summary of findings of all the included studies is presented in Table 3. The mean values of the esthetic score, intra-observer agreement, and inter-observer agreement were tabulated for a comprehensive interpretation. No statistical analysis could be carried out across the indices for measuring the variation in inter- and intra-observer agreement because of heterogeneous nature of indices in terms of assessment, assessor's expertise, and evaluation criteria.

Table 3

Summary of findings

Study and year	Esthetic score	Intra-observer agreement	Inter-observer agreement	Statistical analysis	Reproducibility
Jemt 1997[6]	PI 1, PI 2 Mesial 1.44 2.48 Distal 1.52 2.46	Mean difference between values was 0.11 (SD=0.53)	NR	Sign test P<0.001
Fürhauser et al., 2005[9]	Mean PES 1=9.46Mean PES 2=9.24	No statistically significant difference (P=0.6379)	NR	Chi-square test ANOVA and Tukey’s test (P<0.05)	Good
Meijer et al., 2005[10]	NR	Surgeons=67.1-84.7%Prosthodontists=86.1-86.6%	Surgeons=74.1%Prosthodontists=81.5% between different examiners=70.4%-82.4%	Cohen’s (κ)	High
Gehrke et al., 2008[16]	Mean PES 1=9.28Mean PES 2=9.57	All specialty group=70.5%Orthodontists=73.5%lay people=65.9%	Spearman’s R=0.58, P=0; Z=3.383599, P=0.000716	Spearman’s rank correlation coefficient Wilcoxon signed-rank test	Good
Gehrke et al., 2009[17]	Mean ICAI 1=10.4Mean ICAI 2=9.73	Cohen’s κ=0.49 (P<0.001).	A minimum agreement (Cohen’s K=0.11-0.37, observer agreement: 40.2-66.3%).	Cohen’s κ	Questionable
Belser et al., 2009[11]	Mean total PES/WES=14.7±1.18Mean PES=7.8±0.88Mean WES=6.9±1.47	NR	NR	Linear regression analysis was conducted	Suitable
Cho et al., 2010[18]	Mean total PES/WES=11.19±3.59Mean PES=5.17±2.29Mean WES=6.02±1.96	Very good and moderate agreements, Cohen’s κ=0.526-0.941	NR	Weighted Cohen’s κ Kruskal-Wallis analysis	Suitable
Juodzbalys and Wang 2010[12]	NR	Cohen’s Soft tissue: Predictive: Restoration κExaminer 1-0.9: 0.91: 0.92Examiner 2-0.86: 0.84: 0.84	Examiner1: Examiner 2Soft tissue 86% - 78%Predictive 82% - 66%Restoration 64% - 62%	Weighted Cohen’s kappa (k)	Reproducible
Vilhjálmsson 2011[19]	PES=8 ICAI=9 Mod-ICAI=7	Intra-examiner 1:Intra-examiner 2PES=0.67-1 - PES=0.62-1ICAI=0.64-1 - ICAI=0.62-1	PES=0.62-1ICAI=0.62-1	Weighted Cohen’s κ Spearman rank correlation coefficients	Reproducible
Hosseini and Gotfredsen 2012[13]	NR	Cohen’s k=0.63-0.67	Cohen’s k=0.42-0.51	Cohen’s κ and Cronbach’s α, Spearman correlation coefficients	Useful
Vaidya et al., 2015[20]	Mean PES/WES=14.42, mean modified ICAI=13.46	PES/WES for both the examiners except for the crown surface texture and translucency=(κ=0.30-0.89) modified-ICAI except the mesiodistal dimension of crown=(κ=0.39-1.00)	Lowest agreement prosthodontists (4-28%), rest of the groups low to-moderate agreement (20-80%) more inter-observer agreement with mod-ICAI, than with PES/WES	Cohen κ Kruskal-Wallis Test Mann-Whitney U-test	Both PES/WES and ICAI are reliable
Tettamanti et al., 2016[14]	1 2PICI - 64.69 67.30ICAI - 19.45 20.9 PES/ WES 66.05 67.77	PES/WES - 31 examiners (κ 0.41-1) ICAI - 15 examiners (κ 0.41-1)	No significant differences	Cohen’s Kappa	PES/WES -Highest Reproducibility ICAI-lowest
Li et al., 2017[15]	1 2IREI - 70.73-68.78 PES/ WES - 63.90-64.26	Graduate student group=0.961 prosthodontist=0.952 (P<0.05)	The inter-observer reliability was acceptable, with ICCs of 0.649 and 0.667, respectively (P<0.05)	ICC	Valid and reliable
Hof et al., 2018[42]	PES=9.06 PI=1.94 ICA=4.35 PES/WES=15.1 CEI=75.3 IAS=5.47 SES=1.87 Rompen=1.88	The highest level of intra-rater reproducibility showed PI (ρˆ intra r=0.90) PES (ρˆ intra r=0.88) and CEI (r intra r=0.92) The lowest level of intra-rater reproducibility showed Rompen (ρˆ=intra r=0.70), SES (ρˆ=intra r=0.75), and ICA (r= intra r=0.75)	The highest inter-rater reliability ρˆ inter) values CEI (ρˆ= inter r=0.70), PES (ρˆ inter r=0.66), PI (r= inter r=0.64), and IAS (ρˆ inter r=0.62) The worst inter-rater reliability SESρˆ inter r=0.43 Rompenρˆ inter r=0.44	Inter-rater correlation coefficients, Pearson’s product-moment correlation

PI – Papilla index; PES – Pink Esthetic Score; ICAI – Implant Crown Esthetic Index; PES/WES – Pink Esthetic Score/White Esthetic Score; mod-ICAI – Modified Implant Crown Esthetic Index; CEI – Complex Esthetic Index; CIS – Copenhagen Index Score; PICI – Peri-Implant and Crown Index; IREI – Implant Restoration Esthetic Index; IAS – Implant Esthetic Score; SES – Subjective Esthetic Score; NR – Not reported; ICC – Interclass correlation coefficients; P – Probability value

DISCUSSION

With an increased demand for esthetics, the perspective for the success of implant-supported restorations has changed from mere osseointegration to natural appearance of crowns. The International Team for Implantology (ITI) consensus statement of 2014 has suggested the use of esthetic indices for objective assessment of factors that contribute to esthetic outcomes.[43] The esthetic index serves as a checklist to confirm the vital parameters of esthetic implant restorations. There have been multiple studies suggesting novel indices. On critical evaluation of those studies, this review has attempted to summarize the reproducibility and validity of those indices. Reproducibility of the esthetic index is influenced by intra- and inter-observer variations, the interval between the first and second assessments, observer's specialization, methods of evaluation, and scoring parameters.

Cohen's kappa is an accepted statistical measure to assess intra-observer and inter-observer reliability,[44] which was used by most of the included studies.[1012131417181920] The interval between the first and second assessments of esthetic index ranged from 7 to 28 days. A uniform interval between the first and second evaluations would have increased the homogeneity and reduced bias of included studies. Retained memory of the observer after the first assessment may negatively influence the judgment of the second evaluation when the interval is very short.

Multiple authors showed the influence of the observer's specialization on an objective assessment of index reproducibility.[91016171820] It has been observed that orthodontists were more critical in rating the esthetics for implant restoration likely due to their inclination toward natural esthetics than the restorative esthetics.[91718] We may opine that an unbiased assessor who is unaware of difficulties in achieving peri-implant esthetics is required for assessment of the esthetic index.

The esthetic indices reported so far have evaluated the esthetics either by photographs alone or in combination with dental casts. CEI[12] is the only index that used direct visualization on the patient and intraoral periapical radiographs. Before CEI, none of the esthetic indexes had taken into consideration the underlying hard tissue upon which the soft-tissue appearance is mainly dependent. Horizontal bone level and interproximal bone height are strongly correlated to the stability and appearance of the peri-implant soft tissue.[4546] Chen et al.[47] showed that the thin tissue biotype had more implant soft-tissue recession than the thick tissue biotype. They also found that buccally placed implants had shown three-folds more recession than lingually placed implants. Assessment on photographs runs the risk of a different angle of exposure and difference in brightness and contrast. Again, some anatomical features are more discernible in three-dimensional structures such as alveolar process convexity, which are difficult to interpret in photographs. While most authors used a NS, few authors[1415] used VAS, which has added to the random nature of the data.

Tettamanti et al.[14] rated PES/WES as the fastest and easiest index. They compared three objective indices and found the highest intra-observer agreement with PES/WES and lowest with ICAI; therefore, they conclude that the use of ICAI was questionable. Gehrke et al.[17] also concluded the same about ICAI. Vilhjálmsson et al.[19] also compared three objective indices and found the highest intra-observer reproducibility with modified ICAI. The modified ICAI gives fewer penalty points for gross deviation, thus overcoming the demerit of ICAI. Hof et al.,[42] in a recent study, had compared eight objective indices and discovered the highest intra-rater reproducibility for the PI, PES, and CEI while the lowest intra-rater reproducibility was recorded for ICAI.

Most of the included studies have evaluated esthetics from the clinician's perspective except few,[111314151942] which have included the patient's as well as clinician's perspective. In general, it can be noted that clinicians tend to evaluate the esthetic outcomes of restoration more critically than patients.[48]

As included, articles have used different indices, and the heterogeneous nature of parameters being taken any suggestive comparison between data was not possible. The retrospective nature of all included studies excludes a higher level of evidence, which further encourages to conduct prospective randomized controlled trials with adequately explained inclusion and exclusion criteria.

Clinical significance

This study compares the esthetic indices used for anterior single implant-supported restorations and gives an insight to the readers for using indices in their implant dentistry practice to judge the level of esthetics achieved at the end of the treatment procedure. Evaluation of esthetic outcomes adds as a success criterion for the restoration of anterior implants.

CONCLUSION

Many indices have been proposed for the evaluation of the esthetic aspects of single implant-supported reconstructions in the anterior maxilla. All of them have some advantages and drawbacks that this review pointed out. The evidence level of studies used for the validation of these indices is poor. It is necessary to achieve a consensus on the tools for assessment of the esthetic aspect and perform evidence-based studies to validate an appropriate index.

By comprehensively reviewing several indices, the authors believe that PES/WES index proposed by Belser et al.[11] is the most user-friendly index with scoring criteria straightforward and easy to use. This index was used in most of the clinical trials on anterior implant esthetics the authors came across during the review process.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.