Reliability study of Mahajan's classification of gingival recession: A pioneer clinical study.

BACKGROUND: Gingival recession defects (GRD) are one of the most commonly encountered conditions for which periodontal treatment is required. Several attempts have been made to classify GRD. Among all the classifications Millers classification for gingival recessions is the most accepted. Recently, several limitations and drawbacks have been pointed out in Miller's classification system. The aim of the present study is to test the reliability of the recently proposed Mahajan's Modification of the Miller's Classification for GRD.
MATERIALS AND METHODS: All 15 males and 11 females between the ages of 22 and 55 years (mean age, 37.15 year) with at least one buccal gingival recession were consecutively recruited by the same periodontist in order to test the reliability of the new classification of GRD. The classification was tested by two examiners blinded to the data collected by the other examiner. Intra-rater and inter-rater agreement was assessed.
RESULTS: The new classification system of gingival recessions was tested in a total of 175 gingival recessions in 26 patients. The intraclass correlation coefficient for inter-rater agreement was 0.90, showing an almost perfect agreement between the examiners.
CONCLUSION: The newly proposed classification system eliminates the drawbacks and limitations associated with Miller's classification system and can be used to classify GRD reliably.

INTRODUCTION

Gingival recession is defined as the displacement of the marginal tissue apical to the cement enamel junction (CEJ).[1] Gingival recession has been found to affect populations of industrialized[234] and non-industrialized countries[56] and is prevalent among all age groups.[2] Numerous factors may result in gingival recession.[7] Untreated recession sites in patients are more likely to progress than sites treated with gingival augmentation procedures,[8] therefore, precise diagnosis and appropriate treatment are mandatory to prevent further loss of attachment and progression of gingival recession. Although much emphasis has been given on the treatment aspect of gingival recession defects (GRD) and many techniques have been developed to obtain a predictable root coverage,[789101112] there is still much research, which is required to be done to evolve a comprehensive system to diagnose and classify GRD. Classification systems are necessary in order to provide a framework to scientifically study the etiology, pathogenesis and treatment of diseases in an orderly fashion. In addition, such systems give clinicians a way to organize the health-care needs of their patients. The importance of developing new improved classification system lies in the fact that recent studies have pointed out some inherent limitations and drawbacks in the already popular classification systems.[13141516] Although most of the clinicians are aware about the clinical implications of GRD, but until date there is no consensus on the various classifications used to classify GRD therefore, considering all these aspects, the present study was carried out to test the reliability of a new classification system for classifying GRD evolved from the popular Miller's classification system.

MATERIALS AND METHODS

Subjects showing at least one buccal gingival recession were consecutively recruited by the same periodontist (AM) in order to test the reliability of the new classification of GRD. All patients were recruited in the out-patient Department of Periodontics, Himachal Pradesh Government Dental College and Hospital, Shimla, India and signed a written informed consent in accordance with the Helsinki declaration of 1975 as revised in 2000.

Inclusion criteria

The presence of a buccal recession defect at one or more teeth, irrespective of the amount of clinical attachment loss at the interproximal sites

Completion of causal-related therapy when necessary.

Exclusion criteria

Medically compromised subjects who are unable to participate in the study.

Determination of sample size

In order to establish the appropriate number of gingival recessions needed to verify the agreement among three clinicians of the proposed classification system, an a priori sample size calculation was performed. The sample size was calculated using a minimal acceptance level of intraclass correlation coefficient (ICC) of 0.80 with an alternative hypothesis of 0.90, two operators, a = 0.05 and b = 0.0118. Using these parameters, the required number of recessions was 114.

Assessment of agreement

The assessment of agreement was carried out at the out-patient Department of Periodontics, Himachal Pradesh Government Dental College and Hospital, Shimla, India. Patients presenting at least one gingival recession were consecutively enrolled by one examiner (AM). Two periodontal examiners (DK examiner n. 1 and AK-examiner n. 2) were recruited for the study. Both examiners were informed about and trained and calibrated on the use of the proposed classification system and were blinded with respect to the evaluation of the first author (AM). All needed clarifications were provided before the study. The examiners evaluated each selected gingival recession twice, independently and blindly. The examiners rated the recessions using the above-mentioned classification system. There was no time restriction during the procedure.

The following clinical variables were recorded to assess the new classification system:

Mucogingival junction identification

Intraoral periapical radiographs were advised to assess the radicular bone level at the site of gingival recession.

Gingival thickness

GT was evaluated and categorized into thick or thin on a site level as described by De Rouck et al.[17] This evaluation was based on the transparency of the periodontal probe through the gingival margin while probing the sulcus at the midfacial aspect of the teeth. If the outline of the underlying periodontal probe could be seen through the gingival, it was categorized as thin; if not, it was categorized as thick.

Based on the assessment of above clinical parameters the modified classification system given by Mahajan, 2010, was applied as given below:

Class I: GRD not extending to the MGJ [Figure 1].

Figure 1

Mahajan's Class I gingival recession defects

Class II: GRD extending to the MGJ/beyond it [Figure 2].

Figure 2

Mahajan's Class II gingival recession defects

Class III: GRD with bone or soft-tissue loss in the interdental area up to cervical 1/3 of the root surface and/or malpositioning of the teeth [Figure 3].

Figure 3

Mahajan's Class III gingival recession defects

Class IV: GRD with severe bone or soft-tissue loss in the interdental area greater than cervical 1/3rd of the root surface and/or severe malpositioning of the teeth [Figure 4].

Figure 4

Mahajan's Class IV gingival recession defects

The examiners were also asked to write down the prognosis based on the following guidelines.

Prognosis

BEST: Class I and Class II with thick gingival profile.

GOOD: Class I and Class II with thin gingival profile.

FAIR: Class III with thick gingival profile.

POOR: Class III and Class IV with thin gingival profile.

Statistical analysis

Descriptive statistics with mean and standard deviation (minimum; maximum) were performed. The two-way random ICC and 95% confidence interval (CI) were used to assess the intra-rater and inter-rater agreement among the two periodontal examiners for the Recession type and Recession class. In addition, inter-rater agreement (ICC and 95% CI) were also assessed among the two periodontal examiners and the first author (AM). The level of agreement was evaluated according to the six-level nomenclature given by Landis and Koch:

Poor agreement: 0.00

Slight agreement: 0.00-0.20

Fair agreement: 0.21-0.40

Moderate agreement: 0.41-0.60

Substantial agreement: 0.61-0.80

Almost perfect agreement: 0.81-1.00.

RESULTS

A total of 175 gingival recession sites in 26 subjects (15 male and 11 female) in the age range of 22-55 (mean age 37.15 ± 8.9) were enrolled in order to test the reliability of a new classification system of GRD. Kappa statistics were performed to analyze intra-rater and inter-rater agreement among the three examiners. The kappa statistics for intra-rater agreement ranged from 0.86 (DK) to 0.89 (AK) (almost perfect agreement) [Tables 1 and 2].

Table 1

Crosstab and corresponding ϰ-value intra-operator for clinical examiner-1 (DK) (1.00, 2.00, 3.00 and 4.00 are classes according to Mahajan's classification)

Table 2

Crosstab and corresponding ϰ-value intra-operator for clinical examiner-2 (AK) (1.00, 2.00, 3.00 and 4.00 are classes according to Mahajan's classification)

ICC and 95% CI between the two examiners, i.e. inter-rater agreement (DK-AK) were calculated for GRD. It was 0.9 [Table 3], showing an almost perfect agreement, whereas on comparing examiner 1 (DK) and examiner 2 (AK) with the primary author (AM) the ICC was 0.78, which is again a substantial degree of agreement between the three examiners [Tables 4 and 5].

Table 3

Crosstab and corresponding ϰ-value inter-operator for clinical examiner-1 and 2 (DK-AK) (1.00, 2.00, 3.00 and 4.00 are classes according to Mahajan's classification)

Table 4

Crosstab and corresponding ϰ-values interoperator for clinical examiner-1 and primary author (DK-AM) (1.00, 2.00, 3.00 and 4.00 are classes according to Mahajan's classification)

Table 5

Crosstab and corresponding ϰ-values inter-operator for clinical examiner-2 and primary author (AK-AM). (1.00, 2.00, 3.00 and 4.00 are classes according to Mahajan's classification)

DISCUSSION

Classifications, defined as “systematic arrangements in groups or categories according to established criteria”[18] have been conceived to facilitate the comprehension of the great amount of factors and information involved in complex systems. In periodontology classifications are widely used to categorize defects due to periodontitis according to their etiology, diagnosis, treatment and prognosis. Gingival recessions are frequent lesions and due to aesthetic reasons, patients have always requested treatment.

There have been numerous attempts to classify GRD. One of the first classification on gingival recession was given by Sullivan and Atkins in 1968.[19] In a classical article, soft-tissue defects at mandibular incisors were divided into four classes: “Narrow,” “wide,” “shallow” and “deep.” After that Mlinek et al.[20] in1973 identified “shallow-narrow” defects as the recession <3 mm, while “deep-wide” defects were recessions >3 mm. In 1985 Miller[21] proposed four classes of marginal tissue recessions based on both level of the gingival margin with respect to the MGJ and the underlying interdental alveolar bone, which is the most commonly used classification world-wide until date. In Class I, the recession did not extend to the MGJ, while in Class II the gingival margin reached MGJ, both showing no loss of interproximal bone. In the Class III recession defect, the gingival margin was located to or beyond the MGJ with interproximal bone loss and/or tooth malpositioning. Finally, Class IV showed serious interproximal bone loss and/or severe tooth malpositioning. In 1997 Smith[22] also proposed a compound index of recession to assess both vertical and horizontal extent of the defect. The degree of the horizontal component was expressed as a value ranging from 0 to 5 depending on the severity of CEJ exposure while the vertical extent of recession was measured in millimeters using a periodontal probe on a 0-9 range. In 1998, Nordland and Tarnow[23] were proposed a classification system for loss of papillary height and classified into three classes on the basis of height of interdental papilla and CEJ. After these classifications, some other modification were also given by authors like Mahajan and Pini-prato et al.[24] P. Prato classified gingival recessions into four classes based on presence and absence of CEJ and surface discrepancy. Recently, Cairo et al.[14] have also proposed and tested the reliability of a classification system for GRD.

Despite the availability of various classification systems Miller's classification is still the most widely used system for classifying GRD. Until recently, Miller's classification system was considered as the best classification system for classifying GRD, but with the advancements in diagnosis and treatment of gingival recessions drawbacks of Miller's system have been pointed out by some researchers.[13141516] Therefore, in the absence of a quality classification system to classify GRD, the need for a updated classification system is immense, hence the present system for classifying GRD is proposed after testing it on 175 clinical sites. The results of our classification system were evaluated and vis-a-vis compared with the popular Miller's classification system based on the criteria given by Murphy[25] in 1997, who suggested that a classification system must have the following characteristics.

Usefulness

“Usefulness can be construed at several different levels. Not the least is practicality, even crass practicality.” While Miller's classification system can be considered useful in terms of classifying GRD, but its usefulness is limited when it comes to the determination of gingival profile, which is an indispensable factor in determining and classifying GRD according to their prognosis. These drawbacks have been eliminated in the present classification system.

Exhaustivenesss

“An ideal classification should be exhaustive, that is, accommodate naturally every member of the group.” Miller's classification is not exhaustive because it does not consider all cases of recession. For example, a marginal tissue recession with inter-proximal bone loss, which does not extend to the MGJ is not classified. In fact, this recession cannot be included in Class I because of inter-proximal bone loss and it cannot be categorized in Class III because the gingival margin does not extend to the MGJ. In the present system, there is a clear demarcation between the recession groups. Gingival recessions with only soft-tissue loss are in Class I and Class II while with interproximal bone loss are in Class III and IV.

Disjointness

“No particular case should fall into more than one class.” The difference between Miller's Class III and IV lies only in the extent of the severity of the GRD. The classification system doesn't mention any objective criteria to assess the severity of bone/soft-tissue loss. In the present classification system, clear distinction has been made between different classes of gingival recession based on the objective criteria to avoid any confusion.

Simplicity

“The most convenient classifications are simple for practical applications as a large number of subclasses may be inconvenient.” Due to lack of exhaustiveness and disjointness in Miller's classification system, it is not possible for the clinician to effectively communicate his findings to other colleagues; which results in confusion among different clinicians resulting in difference of opinion in diagnosis and prognosis of the condition. Furthermore, the studies on the reliability and validity of Miller's classification are lacking. The proposed system of classification given by us keeping in mind is based on the popular Miller's classification system, which has been modified to make the proposed system more simple, user-friendly, in addition, the modified classification system has shown almost perfect inter and intra-rater agreement.

Based on the above findings, it can be concluded that to extent the proposed classification system by Mahajan satisfies the majority criteria, which are considered essential for a good classification system; also by modifying the current popular classification system and using the same as the basis for the proposed classification system the author has tried help the clinicians who are accustomed of using the Millers classification for classifying GRD's who may find it difficult to suddenly shift over to a totally new classification system.

The Mahajan's classification system also recommends the use of radiographs for the purpose of diagnoses, the use of radiographs may add to some extra time, radiation exposure and increase in the cost of treatment, but since radiographs also add to the objectivity and precision in diagnosing the condition their use is justifiable, also it's a routine practice in modern periodontal therapy to utilize radiographic aids in diagnosing various periodontal bone defects, furcation defects etc., prior to reconstructive periodontal surgery, so there is no harm in utilizing radiographs which aid in diagnosing the GRD in a better way. One possible theoretical limitation of the proposed classification system may be that it doesn't classify palatal/lingual recession defects, since majority cases who report for treatment are concerned about the facial/buccal gingival recessions the Mahajan's classification system is practical, simple and exhaustive. The authors recommend future studies to assess the validity of the proposed classification system.

SUMMARY

The study suggests the use of Mahajan's modification of Miller's classification for gingival recessions as it has all the qualities of a reliable classification system and it eliminates the limits and drawbacks of Millers classification.