To carry out a meta-analysis in order to assess the influencing factors on retention loss and marginal discoloration of cervical restorations made of composites and glass ionomer (derivates).
The literature was searched for prospective clinical studies on cervical restorations with an observation period of at least 18 months.
Fifty clinical studies involving 40 adhesive systems matched the inclusion criteria. On average, 10% of the cervical fillings were lost and 24% exhibited marginal discoloration after 3 years. The variability ranged from 0% to 50% for retention loss and from 0% to 74% for marginal discoloration. Hardly any secondary caries was detected. When linear mixed models with a study and experiment effect were used, the analysis revealed that the adhesive/restorative class had the most significant influence, with 2-step self-etching adhesive systems performing best and 1-step self-etching adhesive systems performing worst; 3-step etch-and-rinse systems, glass ionomers/resin-modified glass ionomers, 2-step etch-and-rinse systems and polyacid-modified resin composites were ranked in between. Restorations placed in teeth whose dentin/enamel had been prepared/roughened showed a statistically significant higher retention rate than those placed in teeth with unprepared dentin ( p < 0.05). Beveling of the enamel and the type of isolation used (rubberdam/cotton rolls) had no significant influence.
The clinical performance of cervical restorations is significantly influenced by the type of adhesive system used and/or the adhesive class to which the system belonged and whether the dentin/enamel is prepared or not. 2-Step self-etching- and 3-step etch&rinse systems shall be chosen over 1-step self-etching systems and glass ionomer derivates. The dentin (and enamel) surface shall be roughened before placement of the restoration.
About a quarter of the population do have non-carious cervical lesions, and the lesions are significantly more prevalent at older ages (<50%), with premolars being the most affected teeth . There is consensus that the etiology is multifactorial with mechanical-abrasive (toothbrush/toothpaste) and erosive (acids from food and beverages) processes. Occlusal overloading and/or eccentric movements may be found as co-factors because of some confounding effects, but are unlikely causal factors . There is no biological reason to restore non-carious lesions other than esthetics. Under some circumstances, when the lesion has significantly compromised tooth structure or is progressing at a fast rate, can restoring the lesion prevent further tooth damage. A questionnaire conducted among US general practitioners in the 90s revealed that more than half of the dentists interviewed did not restore non-carious cervical lesions .
Loss of retention and marginal discoloration are still the main shortcomings of cervical restorations (Class V) placed with adhesive technology . Both clinical incidents compromise the esthetic appearance, especially if they occur in anterior teeth. The prevalence of retention loss rises sharply with increasing observation periods . On the other hand, Class V non-carious non-retentive lesions are frequently used to clinically evaluate the effectiveness of adhesive systems.
In non-carious cervical lesions, restoratives are placed with either preparation of dentin and/or enamel. In contrast to intact or caries-affected dentin, non-carious cervical lesions exhibit a high degree of sclerosis and have a high amount of minerals, which renders the establishment of a hybrid layer more difficult . Some clinicians roughen the dentin and enamel with a diamond bur and/or bevel the enamel margin to improve the bond to the hard tissues. The results of some clinical studies on the topic of preparation are inconclusive . The influence of absolute versus relative isolation of the treatment field is another topic that is subject to controversy. A meta-analysis revealed no influence of the type of isolation on the survival rate in posterior composite restorations . A systematic analysis of clinical studies on cervical restorations has, to date, not been carried to investigate these operative aspects.
The type of adhesive system and/or the belonging to a specific class of adhesives as proposed by Van Meerbeek may play an important role on the longevity of the restoration. A review of clinical trials on the effectiveness of adhesive systems in non-carious cervical lesions measured as retention loss arrived at the conclusion that glass ionomer cements had the lowest rate of retention loss and 1-step self-etching adhesives the highest .
The American Dental Association (ADA) previously defined an adhesive system to be adequate and acceptable for clinical use (“full acceptance”) if the retention rate of restorations placed in non-carious lesions is higher than 90% after an observation period of 1.5 years . Many of the newer adhesive systems, especially the 1-step self-etching systems, would not have received ADA acceptance. The ADA acceptance program was abandoned by the end of 2008 .
The goal of the present study was to assess the influence of the following factors on the clinical outcome:
operative technique: beveling of enamel, preparation/roughening of dentin/enamel, absolute versus relative isolation;
the type of adhesive system and/or restorative material;
The following hypotheses were examined:
Beveling of enamel and/or preparation/roughening of dentin/enamel results in less retention loss and marginal discoloration.
The type of isolation does not influence the clinical outcome.
The type of adhesive system or restorative material has an influence on the performance of cervical restorations.
Materials and methods
Selection of clinical trials on Class V restorations
Prospective clinical studies on Class V restorations were searched in MEDLINE (search period 12/2008) and IADR abstracts (1994–2008). The search words were “Class V” or “cervical” or “abfraction lesion” and “clinical”. The inclusion criteria were as follows:
Prospective clinical trial involving at least one adhesive system in Class V cavities.
Minimal duration of 18 months.
The study had to report about the following outcome variables: retention, marginal discoloration, marginal integrity, secondary caries.
The study had to report on the operative technique (beveling of enamel, preparation/roughening of dentin/enamel, isolation).
If a clinical trial investigated the effect of etching the enamel by comparing the results with those of etch&rinse adhesives, only the data of the etching group were selected.
The restorative materials and adhesive systems (AS) were grouped as follows:
1-step self-etching AS;
2-step self-etching AS;
2-step etch&rinse AS;
3-step etch&rinse AS;
polyacid-modified resin composites (PMRC);
resin-modified glass/glass ionomer cements (RMGIC/GIC).
The following clinical outcomes were retained: R = 100 − (% of retention loss), MD = 100 − (% of marginal discoloration) and MI = 100 − (% of detectable margins). Since most experiments had 0% of secondary caries, this outcome was not considered. Following Heintze et al. , in vivo performance was summarized by combining three clinical outcomes into one clinical index: CI = (4 × R + 2 × MD + 1 × MI)/7. Our goal was to model deterioration of each outcome Y (where Y = R , MD, MI or CI) along time t (with data available on t = 12, 18, 24 and 36 months), accounting for the fact that Y = 100 at t = 0.
A first look at the data revealed roughly linear deterioration with time. However, a simple regression model Y = 100 − βt + error, equivalent to (100 − Y )/ t = β + error, was not adequate, the error being not normally distributed (data not shown). Thus, we applied a square root transformation and opted for model
100 − Y t = β + error
for which normality could be assumed (especially for Y = CI, data not shown). Since this model implies Y = 100 − ( β + error) 2 t , the slope −( β + error) 2 characterizing deterioration was forced to be negative, which makes sense, and the quantity − β 2 was interpreted as median slope of deterioration.
The goal was then to assess how the factors beveling, preparation, rubberdam and class of AS (the latter with six levels) affected deterioration. These factors were treated as fixed effects in our model, while we included a random “study effect”, to account that partly the same patients were used in a given study, and a random “experiment effect”, to account for the correlation between measurements within a given experiment. Thus, we considered a linear mixed model where
β = β 0 + β B + β P + β R + β A 1 + β A 2 + β A 3 + β A 4 + β A 5 + study effect + experiment effect.