A number of factors - including preparation design, occlusal forces, oral hygiene, and the restoration materials used - influence the success of fixed dental prostheses. Nevertheless, the key factor for success is the choice of suitable luting cement and cementing procedure.1 Cementing is the final stage for indirect restoration and the most critical step for ensuring the long-term success of the dental restoration.1

Dental luting cement is a substance that bonds juxtaposed surfaces together and provides thermal, electrical, and chemical insulation. The main purpose of such bond is to achieve marginal sealing, adaptation, and stable adhesion between the different types of substrate, as well as adequate retention and resistance.1234

Radiopacity is one of the main requirements for luting cements, particularly when used for cementing indirect restorations, as radiographs enable dentists to complete their diagnoses.5 In a radiograph, a sufficiently radiopaque material allows proper assessment of its bond with the dental tissues, particularly in the areas of difficult access.2 The radiopacity also assists the detection of secondary caries under the restoration6 and marginal defects and enables the observation of periodontal effects of the cement overhangs. Therefore, when the luting agent is not radiopaque enough, it is impossible to detect voids,6 inadequate marginal adaptation, and interfacial gaps.6789101112

For many years in the past, zinc phosphate cement had been the most commonly used luting material.131415 However, despite its good radiopacity - Attar et al.16 showed it to be the most radiopaque cement - it has certain undesirable properties, such as solubility in the oral environment, which have prompted the search for an alternative with ideal characteristics. In recent years, the use of resin cements and resin-modified glass ionomer cements has increased significantly. 517 This is largely due to the evolution of adhesive systems. These materials have the great advantage of adhering not only to the tooth structure but also to ceramics, compound resins and metal alloys. They widely used in restorations with little frictional retention, owing to their high bonding strength.21318 However, despite the scientific advances they bring to dentistry, few studies have addressed their radiopacity.

The purposes of the present study were to compare the radiopacity of six resin cements used in dentistry with that of human enamel and dentin, verify whether they meet the requirements of the ISO 4049/2000 standard and ANSI/ADA specification No. 27/1993, and assess whether their radiopacity is influenced by the thickness of the cement employed.

Six resin cements in this study came onto the market recently. Their chemical composition and characteristics are shown in Table 1. In order to standardize the specimens, a stainless steel wedge was made to have 3 steps of thicknesses (0.5, 1 and 1.5 mm thick), and a light silicon mould was cast from this wedge. The moulds were filled with the cements, covered with microscopy slides to remove excess material and allow the passage of light, and cured with the LED curing lamp (Bluephase C8, IvoclarVivadent) with a power of 800 mW/cm² and wavelength of 380 - 515 nm, as verified by a radiometer. The duration of the photopolymerization was that recommended by the manufacturers.

The thickness of each specimen was checked with a digital calibrator. Specimens of more than the desired thickness (± 0.01 mm) were polished to get the correct size and were then cleaned with 70% ethyl alcohol. For each cement, 3 specimens in the form of wedges of the previously mentioned thicknesses were prepared.

To prepare of tooth specimens, sections of a third molar were cut with a low-speed diamond disc (Bredent, Senden, Germany) to obtain 0.5, 1 and 1.5 mm (± 0.01 mm) mesiodistal specimens of enamel and dentin. The final thickness of each tooth specimen was adjusted by sandpapering and checked with a digital calibrator, allowing a critical tolerance level of ± 0.01 mm. The tooth specimen was then stored in saline solution at room temperature.

The aluminum step wedge employed was machined from a single block of aluminum (Alcoa Inespal, Avilés, Asturias, Spain) with a purity of 99.889% Al, 0.0202% Si, 0.0687% Fe and 0.0001% Cu. It had 10 steps or thickness levels, rising in 0.5 mm increments, and its overall dimensions were 10 mm wide by 20 mm long.

Afterwards, each specimen was X-rayed 5 times at 65 kV and 10 mA for 0.30 seconds at a focal object distance of 30 cm, with the cement specimen, the aluminum step wedge and the three enamel-dentin specimens placed on the Sopix 2 digital sensor (Satelec, La Ciotat, France) (Fig. 1).

The images were then imported into Digora for Windows 2.5 software (Soredex, Tuusula, Finland) in order to measure the specimens' radiopacity in pixels, using the density measurement tool. Each thickness of each cement was measured at 5 different positions. Only the regions that were free of air bubbles or other defects were analyzed (Fig. 2, Fig. 3). Using a similar procedure, the radiopacity of the enamel and dentin was also measured in 5 different regions of each thickness specimen, with 15 exposures in each region.

The density of each of the 10 thicknesses of the aluminum step wedge was measured by 15 random radiographs, each yielding 5 radiopacity values. A density calibration curve was constructed from the mean values for each step of the aluminum wedge, making it possible to convert the mean radiopacity value of each luting cement into millimeters of aluminum.

Statistical analysis of the data collected in this study was performed with SPSS software at a 95% confidence level. A t-test for equality of means was used to examine the differences in radiopacity between the different materials, and an ANOVA, with post-hoc tests, was carried out to study the influence of the thickness of the materials on the radiopacities measured in this study.

Firstly, the independent specimens t-test for equality of means was used to discover any statistically significant differences between the mean radiopacity of each cement and that of the enamel and dentin at the three thicknesses studied. At every thickness level, statistically significant differences (P < .05) between the radiopacities of the enamel and dentin and those of all the cements were encountered. The results also demonstrate that the Calibra, Block Out, Ultrabond Plus, Insure and Rely X Ultimate cements were significantly more radiopaque than the enamel and dentin at each of the study thicknesses, while the radiopacity of the Variolink Veneer cement with medium value 0 shade was lower than that of the tooth specimen at all the thicknesses studied.

To verify whether the materials comply with current standards, the independent specimens t-test for equality of means was used. The mean radiopacity of all the materials differed significantly from that of aluminum (P < .001). Observing the Mean Differences column values, it may also be stated that Calibra, Block Out, Ultrabond Plus, Insure and Rely X Ultimate show a positive difference compared to the value for aluminum - in other words, their mean radiopacity is higher than that of aluminum - while the mean radiopacity of the Variolink Veneer cement with medium value 0 shade is significantly lower than that of aluminum at all the thicknesses tested. This demonstrates that all the cements except Variolink Veneer cement with medium value 0 shade comply with the current standards.

The present study also examined whether the radiopacity values presented statistically significant differences depending on the thickness of the cement (Table 2). For all the materials studied, the ANOVA and Welch statistic findings provide statistical evidence that the radiopacity of the cements depends to a significant degree on its thickness, as the radiopacity results fall into three groups, significantly differentiated by thickness, and the thicker the material, the greater its radiopacity (Fig. 4).

In the present study, the radiopacity of six resin luting cements was compared with human enamel and dentin and with the mean values of the aluminum step wedge employed as a standard control material. The latter is essential for comparison with radiopacity measurements performed by other researchers. The radiopacity of dentin and enamel specimens can vary, but pure aluminum provides a constant reference value.69

It should be pointed out that it is difficult to compare the findings of the present study with those of previous researchers who examined older luting cements. Advances in materials are constantly being made and the present study examined cements that have only recently come on the market.

Enamel is a radiopaque structure, as it has a 90% mineral content, whereas dentin is less mineralized (75%) and therefore less radiopaque.19

Van Dijken19 showed that the radiopacity of dentin is approximately equivalent to that of aluminum of the same thickness, while enamel has approximately twice the radiopacity of aluminium. These findings agree with the values reported by Salzedas et al.11 The present study yielded similar results for the three thicknesses examined, as the enamel values were approximately double those of dentin.

Some authors consider that to allow correct diagnosis, these restoration materials should be at least as radiopaque as the same thickness of dentin, and have highlighted the importance of using sections of dental tissues as a secondary standard.6712 Papers by Altintas et al. 6 and Furtos et al. 20 present similarities with the present study as regards encountering similar or higher radiopacity to that of dentin in most of the cements they studied.

However, other researchers consider that restoration materials need to be slightly more radiopaque than enamel.912

Pedrosa et al.10 state that materials with higher levels of radiopacity than enamel facilitate diagnosis as the contrast differentiates the dental structures, whereas radiopacity values between those of enamel and dentin or lower than in dentin tend to confuse the examiner, so there is a greater possibility of false positive diagnoses of secondary caries. Along the same lines, Takuma Tsuge4 studied 8 resin cements and reported that 5 were more radiopaque than enamel and 3 less radiopaque, as was found with the Variolink Veneer cement Medium Value 0 shade in the present study.

According to the ISO and ANSI/ADA standards,2122 the minimum acceptable radiopacity for a restoration material is equivalent to that of the same thickness of aluminium. The results of the present study fell within the radiopacity requirements required for cementing agents included in the standards issued by the ISO 4049 and ANSI/ADA,2122 that it to say, the radiopacity of any thickness of the resin cement materials studied is significantly higher than that of aluminum, with the exception of Variolink Veneer Medium Value 0, which is less radiopaque than the standard.

According to the bibliography, the radiopacity of resin cement is typically attributed to its inorganic load, with little contribution from the matrix.23720

The physical properties of dental cements can vary considerably as a result of differences in the quantity and quality of their chemical components. Including elements with a high atomic weight among the filler particles of a resin-based cement helps to increase the radiopacity of these materials.313

Furtos et al.20 verified that the cements with the highest filler content were the most radiopaque (a direct correlation between filler content and radiopacity), and that level of radiopacity was significantly higher than that this of enamel. However, the cement with the lowest percentage load was still significantly more radiopaque than dentin.

Information on the correct, detailed chemical composition of resin cements is scarce and most manufacturers consider these data confidential. Owing to the limited data available on this aspect of the materials studied, it has not proved possible to construct a correlation between filler content, organic matrix and type of radiopacity.

At each of the study thicknesses, the Calibra, Block Out, Ultrabond Plus, Insure and Rely X Ultimate cements are significantly more radiopaque than enamel and dentin, while Variolink Veneer Medium Value 0 is significantly less radiopaque than either at all these thicknesses. With the exception of Variolink Veneer Medium Value 0, all the cements studied comply with the ISO and ANSI/ADA standards at all three thicknesses. Furthemore, the radiopacity of all these luting cements depends to a significant degree on their thickness: the thicker the material, the greater its radiopacity.