This in vitro study was conducted in the Department of Conservative Dentistry and Endodontics at Drs Sudha and Nageswara Rao Siddhartha Institute of Dental Sciences, Gannavaram, Andhra Pradesh, India. The study was performed for the time period of one month i.e., in July 2016. Human maxillary first premolars which were extracted atraumatically for orthodontic reasons were collected and observed under microscope at 10X magnifications. Teeth with cracks were excluded and finally forty sound teeth were selected [4]. The power of the study was 80%. The teeth were stored in 0.5% chloramine solution (Oxford Lad Chem., India) before use. The roots of the teeth were embedded in acrylic resin 2 mm below the cementoenamel junction and parallel to the long axis of teeth. Periodontal ligament simulation was not performed.

Among 40 teeth, 10 teeth were randomly selected for the control group, i.e., Sound teeth (Group I) and remaining teeth were divided into three groups based on the cavity designs. MOD cavities were prepared using FG 271 and FG169L carbide burs (SS White, Lake wood, USA). In Group II-MOD Inlay, cavity preparations were done with occlusal cavity width kept at 1/3^rd the intercuspal distance, the cavity pulpal floor depth was 1.5 mm at the central groove area and each wall with a taper of 3°. In the proximal box, gingival seat was placed 1 mm above the cementoenamel junction and mesiodistal width of the seat was kept to 1 mm [Table/Fig-1]. In Group III-Partial Onlay, preparations were similar to Group II but the palatal cusp was reduced by 1.5 mm with a collar of 1 mm [Table/Fig-2]. In Group IV -Complete Onlay preparations were similar to Group III with palatal cusp reduction by 1.5 mm with a collar of 1 mm width and buccal cusp reduction of 1 mm with collar of 1 mm [Table/Fig-3, 4].

[Table/Fig-1]:

Group II - MOD Inlay cavity preparation: a) Occlusal view; b) Proximal view; c) Zirconia restoration.

[Table/Fig-2]:

Group III - MOD Onlay with buccal cusp preservation and palatal cusp reduction (Partial Onlay); a) Occlusal view; b) Proximal view; c) Zirconia restoration.

[Table/Fig-3]:

Group IV - MOD Onlay with buccal and palatal cusp reduction (Complete Onlay): a) Occlusal view; b) Proximal view; c) Zirconia restoration.

[Table/Fig-4]:

Schematic diagram of occlusal and proximal view of Complete Onlay (Group IV). IC-Intercuspal distance; CEJ- Cementoenamel Junction

[Table/Fig-5]:

Load application using Universal Testing Machine (UTM) for testing fracture resistance.

Restorations were fabricated with monolithic partially sintered zirconia CAD (SAGEMAX- NexxZr) using Computer Aided Design/ Computer Aided Manufacturing (CAD/CAM) i.e., Delcam Dent CAD and Roland milling machine. All the restorations were cemented using Multilink Automix (Ivoclar Vivadent, Schaan, Lichtenstein) according to the manufacturers recommendation. Prior to cementation, the fitting surface of the restorations were sand blasted with air borne particle abrasion with 27 µm aluminium oxide at 30 psi followed by the application of silane coupling agent (Metal/Zirconia Primer) for 180 seconds. Multilink primer A and B were mixed and applied to the cavity preparations. The restorations were finally cemented and light cured on all restored surfaces for 40 seconds.

The samples were stored in water at 37°C for 21 days. The fracture resistance of the samples was tested using UTM (Instron, UK). The load was applied on the center of the buccal cusp slope at an angle of 30° to the long axis of the tooth with a steel ball of diameter 3.5 mm and a crosshead speed of 0.5 mm per minute [Table/Fig-5]. The applied force was measured in Newtons (N).

All the fractured specimens were observed under stereomicroscope (2X magnification) to visualize the fracture lines in the tooth and the restoration [Table/Fig-6].

[Table/Fig-6]:

Representative images of the observed failure modes in four groups: a) Type I failure pattern; b) Type II failure pattern; c) Type V failure pattern.

Failure patterns were evaluated and classified into five categories [8].

Fracture resistance of the samples was analysed using parametric one way ANOVA test, unpaired t-test and Tukey test. Evaluation of failure patterns was performed using non parametric Chi-square test. The analysis was done using Statistical Package for Social Sciences (SPSS), version #16. The level of significance was considered when p-value was <0.01 and <0.05.

The mean fracture resistance of the experimental groups and their standard deviation are shown in [Table/Fig-7]. Highest fracture resistance was presented in Group IV (Complete Onlay) and it showed statistical significant difference (p<0.01) [Table/Fig-8, 9]. Group I (Sound teeth) showed significantly higher values than Group II (MOD Inlay) and Group III (Partial Onlay). However, there is no statistical significant difference between Groups I, II and III [Table/Fig-10]. According to the failure pattern shown in [Table/Fig-11], Group II (MOD Inlay) and Group III (Partial Onlay) presented mixed type of failures. Group IV (Complete Onlay) demonstrated 70% Type I failure.

The fracture resistance of human maxillary premolar with different MOD cavity design was assessed in the present study. Premolar teeth are an interesting option because these teeth have been reported to show a high incidence of fractures in the clinical situation. Fracture of buccal cusps of premolars occur more often compared to palatal cusps which was supported by Cavel WT et al., who reported that high percentage of fractures upto 60% occurred on the non functional buccal cusp while 40% fractures on the functional palatal cusp [9]. The fracture load was applied on the buccal cusp in order to simulate extreme eccentric forces [10].

In the present study, the cavities were more conservative considering the high fracture toughness and superior properties of zirconia. The occusal cavity width was maintained at 1/3rd the intercuspal distance and occlusal depth of 1.5 mm from the central groove. Palatal cusp was reduced by 1.5 mm in both Partial Onlay and Complete Onlay groups and buccal cusp by 1 mm. The gingival seat was located 1 mm above the CEJ as the margins placed below CEJ affected the stability of the restorations and also bonding.

The Sound teeth presented mean fracture resistance of 997.84 N, similar to values found in other studies, which ranged between 882 N and 1568 N [11]. Statistically there was no significant difference in the fracture resistance of Sound teeth and Inlay (Group II). This might be attributed to the minimal cavity preparation in the present study. Ragauska A et al., concluded that ceramic inlays prepared with 1/2 or 1/3rd of the intercuspal distance can recover tooth strength near to that of intact teeth [12]. According to Cubas GB et al., inlay restorations required less invasive preparation and should be preferred as they are more conservative since there is no occlusal reduction and they present fracture resistance similar to sound teeth [2].

Cuspal coverage is commonly recommended to protect the weakened cusp and to enhance the fracture resistance. In the present study palatal cusp coverage i.e., Partial Onlay (Group II) restored the fracture resistance of premolar to that of Sound teeth. These results are in accordance with a study conducted by Kantardzic I et al., who concluded that palatal cusp reduction is preferred for ideal stress distribution than inlay and complete onlay [5]. On the contrary, Cubas GB et al., suggested that partial and complete onlays with cuspal coverage has no influence on improving the fracture resistance of premolars [2]. There was no significant difference in fracture resistance values between Inlay (Group II) and Partial Onlay (Group III). It might be because of adhesively bonded zirconia restoration in both the groups which had increased the fracture resistance similar to that of Sound teeth.

Complete onlays with both the buccal and palatal cusps involved had showed a mean fracture resistance of 1843.6 N, highest of all the groups. The main reason may be related to the transformation toughening phenomenon of zirconia which requires a higher load to fracture the material and minimal onlay preparation with both cusps involved so that there is equal stress distribution in the cavity which resulted in highest fracture resistance compared to inlay, partial onlay and sound teeth. These findings were supported by Mynampati P et al., where RCT treated premolars restored with monolithic zirconia complete onlays showed a very high fracture resistance (1524.4 N) compared to sound teeth (980 N) [8]. This proves the reinforcement capability of monolithic zirconia complete onlays. Salies SG et al., reported that the fracture resistance of teeth restored with MOD Type III gold inlays with cuspal coverage is enhanced to a level much greater than the sound teeth i.e., the absorption energy of sound teeth was 7.58 kg.cm and sound teeth with gold onlay was 21.41 kg.cm. Similarly the same can be applied to the present study as the reinforcement of the tooth by zirconia complete onlays was much higher than the sound teeth [13].

In the present study, none of the samples in the experimental groups showed Type III failure (Fracture restricted to the restoration alone) i.e., cohesive failure alone in zirconia restorations. This might be attributed to sub critical crack growth and transformation toughening properties of zirconia [14]. The results obtained in the study are contradictory to the values obtained by Soares SJ et al., where the failure mode analysis revealed that fractures of the ceramic restoration occurred before the tooth structure fractured where they used IPS Empress (Leucite reinforced glass ceramic) [15].

According to the outcome of the present study, Type V mode of failure i.e., 40% and 50% was observed in Inlay and Partial Onlay groups respectively. One possible explanation is the non homogeneous stress distribution in the tooth because of the cusps involvement resulting in predominance of catastrophic failure. The findings were supported by Jiang W et al., who reported that ceramic inlay restorations result in higher stress levels in the internal surfaces of the preparation [16]. These results are in accordance with those of the current study where Onlay restored tooth (Group IV) presented 30% of Type V failures and 70% of Type I failures. The occurrence of Type I failure pattern might be because of bond failure in the cementation (adhesive) or the fracture in the internal surface of tooth, where more amount of stresses are concentrated, which had occurred before any visible cracks observed in the restoration or external surface of the tooth. However, none of the onlays got debonded. These findings are similar to the study done by Mynampati P et al., where they found 90% Type I failure and 10% Type V failure in premolars restored with zirconia onlays [8].

The study consists of smaller sample size and the marginal fit of the restorations was not considered. Future studies with larger sample size can be undertaken to evaluate different aesthetic materials with different cavity designs.

Although the MOD inlays and partial onlays almost reinforced the tooth structure to that of sound teeth, complete onlays showed maximum reinforcement of the tooth structure. Hence monolithic zirconia can be chosen as the ideal material to complete onlays because of the reinforcing effect on the tooth. With the overall improvement in the aesthetics of zirconia, it could probably be the best alternative to all metal and all ceramic onlays in aesthetic areas.