From 2017 to 2019, a total of 154 patients received 166 onlays treatments and were recalled for this retrospective study on clinical performance. Digital records showed that the proportion of PO onlays was the most (55.42%), for most of these teeth need to be repaired after root canal treatment because of the pulp inflammation caused by adjacent surface caries [18]. In this study, lithium disilicate glass-ceramic onlays showed satisfactory clinical performance in the returning visit. 95.76% of 166 onlays were grade A in proximal contact, and more than 98% reached A in the other observation criteria. Group O had the least grade A (89.23%). There were 5 onlays (7.69%) with grade C, including 2 cases with food impaction and 3 cases with the tooth fracture, while there were 7 onlays (7.61%) with grade C in PO group.
The kaplan-Meier survival rate analysis showed that the survival rate of O group and PO group decreased at the second year, but remained at a high level till the fourth year (exceeding 95%), which illustrated satisfactory survival rate in line with other studies [19, 20]. Although Georgia et al. [3] demonstrated that the mean survival rate of onlays and crowns were 93.50% and 95.38% respectively, some medium-term studies (2–5 years) indicated a survival rate of 91–100% of onlays [21]. In this study, the survival rate of onlays is 96.99%, which is comparable to that of full crown.
Log-rank test analysis showed that there was no significant difference in the survival rate among different kinds of onlays (O, PO and BO / LIO). This means that occlusal, proximal-occlusive and buccal-/ lingual-defects can all be repaired by lithium disilicate glass-ceramic onlay with ideal survival rate. Tavarez et al. [22] concluded that the preservation of the residual tooth tissue is the decisive factor for the fracture strength of the teeth. Compared with full crown restoration, the residual tooth tissue can be preserved to the greater extent by using onlay, which means onlays are more suitable for tooth defects. Interestingly, some previous studies even reported a better clinical performance of onlays than full crown, especially in survival rates [23, 24].
Two technical complications were observed from this study. One is onlay fracture. Ceramic fractures represented the most common technical complication [3]. Fracture of restoration can be caused by insufficient thickness of onlay. Murgueitio et al. [25] discovered that the thinner the thickness, the higher the risk of restoration fracture, when the thickness was less than 2mm. Della et al. [26] showed that the microcracks generated during polishing may expand under the excessive bite force, which will eventually lead to the fracture of the prosthesis. This fractured onlay happened in PO group in this study. The patient was a middle-aged man, and the tooth is a mandibular molar, thus excessive bite force may be the reason for this failure. When compared to full crowns, onlays showed better fracture resistance. Beier et al. [24] found that 64.86% of failures of crowns were attributed to ceramic fracture, and only 20.00% were of onlays’ failures. Another case failed for onlay debonded after being repaired for 1 year in PO group. When the margin of tooth defects was subgingival, gingival crevicular fluid or saliva can contaminate the adhesive interface during the bonding process, which lead to decreased bond strength and finally restoration debonded. Thus, using rubber dam to prevent saliva contamination, and cleaning lithium disilicate glass ceramic surface with orthophosphoric acid or re-etching with hydrofluoric acid can reach reliable bond strength [27].
Biological complications observed in this study were three teeth fracture failures, which are all happened in cracked teeth in O group. One is vital premolar, and the rest two are ET molars. But Log-rank test analysis showed that there was no statistical significance in survival rate of between the vital teeth and ET teeth, as well as cracked and uncracked teeth. The traditional viewpoint is that after endodontical treatment, the resistant strength of the tooth will decrease and easy to crack due to poor vascular nourishment. Although the repair is successfully completed, the risk of fracture is still higher than that of vital teeth [28]. Manhart J et al. [29] suggested that pulp vitality is not necessarily the main factor leading to restoration failure. The survival rate was significantly higher (97%) for cracked teeth receiving a full crown after endodontic treatment [30]. Nevertheless few studies reported the survival rate of cracked teeth receiving ceramic onlays. More studies are needed to reveal the long-term survival rate of cracked teeth with ceramic onlays. Fabbri et al. [31] showed tooth fracture accounted for 7.14% for crowns’ failure, and 25.00% in the study of Barnes et al. [32] Reich et al. [33] showed that tooth fracture accounted 4.54% for onlays’ failure. These studies demonstrated that less tooth fracture occurred in onlays than in full crown, which can be attributed to that onlay restoration preserved more tissue and provided teeth with better fracture resistance.
Food impaction was observed in 7 cases, in which there are two O onlays and five PO onlays. The two O cases and three of five PO cases were horizontal food impaction. The proximal contacts were of suitable tightness when checked by dental floss. The impaction may be related to the atrophy of gingival papilla between teeth. Only two cases were shown to have loose contact to the adjacent teeth in PO group, thus leading to vertical food impaction. When tooth defects were restored by onlays, most often onlays can preserve original adjacent relationship to the largest extent for the margin of onlays was located in the occlusal 1/3 of tooth. But during the tooth preparation of full crown, the original mesio- and distal-adjacency of the tooth were all destroyed. Food impaction can also result in secondary caries and caries accounted for 13.50% of the total failures of crowns [24]. The margin of onlays is visible, and easier to keep hygiene. Thus, secondary caries occurs rarely following onlay restoration.
In the survey of patient satisfaction, no patients were unsatisfied with the appearance of onlays. Except the 5 failure cases, only two patients were unsatisfied in comfort, one for severe food impaction and another for the pain during chewing. And 7 cases showed that the food impaction has influence on functional recovery, but the satisfactory rate was still over 90%. These all showed satisfactory performance of onlays in appearance restoration, comfort and function recovery. Furthermore, when introducing the restoration methods to patients, most of them tend to choose onlay which is more minimally invasive, less tooth tissue preparation.
In addition, the most often used material for posterior full crown is monolithic zirconia [34, 35]. Previous study suggested that enamel antagonized against monolithic zirconia resulting in more height loss when compared to lithium disilicate glass ceramic [36]. Thus, lithium disilicate glass-ceramic onlays also showed better performance in antagonist-friendly aspect.