Digitally fabricated posterior ceramic crowns versus conventionally fabricated cast crowns: a split-mouth retrospective cohort study

Abstract

Background

This retrospective cohort study aimed to investigate the effectiveness of posterior ceramic crowns fabricated using digital technology compared to those fabricated via conventional elastomer impressions and casting methods at an academic institution.

Methods

A split-mouth design was employed, wherein each participant received both a posterior ceramic crown (lithium disilicate or zirconia) fabricated using digital workflows and a cast crown fabricated conventionally. Of 45 potential participants treated between 2015 and 2023, 31 were included after excluding anteriors (7), < 1-year follow-ups (5), and bridge abutments (2). Patient- and tooth-related data were collected. Wilcoxon signed-rank and chi-square tests, Kaplan-Meier survival analysis, and the log-rank test were conducted to compare the ceramic and cast crown groups.

Results

62 teeth from 15 men and 16 women were observed for 1–8 years. The mean observation periods did not differ between the ceramic and cast groups (3.3 versus 4 years; p = 0.123). No significant differences were found in the distributions of root canal treatments, clinical crown lengthening procedures, and post-insertions (p > 0.05), nor in the probability of tooth survival (p = 0.14) between the two groups. The insertion of the ceramic crowns was significantly expedited from making the final impression compared to the cast crowns (21 versus 41 days; p = 0.0006). There were no significant differences in the complication rates and tooth loss between the two groups (p > 0.05), with most extractions due to endodontic reasons.

Conclusion

The ceramic crowns demonstrated notable effectiveness over the cast crowns, completing the procedure in significantly fewer days, while resulting in a comparable prognosis with similar complication rates compared to the cast crowns.

Introduction

The success of complete-coverage restorations depends on multiple factors, including esthetics, mechanical strength, marginal adaptation, biological compatibility, and long-term clinical performance [1]. Ceramic crowns fabricated using computer-aided design and computer-aided manufacturing (CAD/CAM) technology offer better esthetics, and monolithic designs enhance mechanical properties by reducing the risk of chipping or fracture compared to porcelain-fused-to-metal (PFM) crowns [2, 3]. CAD/CAM technology enables in-office crown fabrication with fewer appointments, improving patient care and reducing long-term overhead. Additionally, digital impressions obtained via intraoral scanning enhance accuracy and patient comfort [4]. These advantages have driven the adoption of digital technology in private practice, despite the considerable initial investment and ongoing maintenance expenses [5].

To ensure dental students are prepared for contemporary clinical practice, integrating digital technologies into predoctoral education is essential [6]. Dental school clinics often face inefficiencies, such as waiting for faculty checks and lengthy procedures [6]. Incorporating digital technologies into clinic workflow can help address these issues while enhancing patient care and strengthening students’ theoretical knowledge and practical skills in complete-coverage restorations [6–9]. For example, digital scans can be evaluated during the same visit, minimizing the risk of impression failure and reducing the need for repeated appointments [10].

While CAD/CAM technology has been introduced in most dental schools, clinical studies evaluating CAD/CAM crowns completed by dental students remain limited. One study on posterior e.max CAD/CAM crowns inserted by dental students found that provider experience did not significantly affect the clinical performance of the crowns, with reported survival and success rates of 93% and 86.4%, respectively [3]. Additional studies have also reported high survival rates for CAD/CAM crowns completed by students [11, 12]. However, few have directly compared ceramic crowns to those fabricated using conventional impression and casting methods, and comprehensive evaluations of factors influencing their long-term prognosis are lacking.

The longevity of any crown is directly linked to the health and survival of the teeth crowned [13]. While mechanical and physical properties–such as marginal fits, occlusion, strength, and wear resistance–affect the prognosis of the crowns [14, 15], biological factors including dental caries, periodontal disease, and endodontic conditions also impact the health of the underlying teeth [13, 16, 17].

At the University of Maryland School of Dentistry (UMSOD), PFM crowns have traditionally been preferred for posterior restorations because of their strength and durability [1]. Since the introduction of CAD/CAM technology and an in-house milling system at UMSOD undergraduate clinic in 2015, the use of ceramic crowns for posterior teeth restorations has increased. This retrospective split-mouth design study aimed to investigate the effectiveness of posterior ceramic crowns fabricated using digital technology compared to those fabricated using conventional impression and casting methods. The study focused on the time from making the final impression to inserting crowns, complication rates, and tooth survival, all of which are key factors in patient care.

Methods

The study report followed the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines from the Enhancing the Quality and Transparency Of Health Research (AQUATOR) [18].

Study design and data collection

The screening population was identified from the electronic health record (EHR) database from 2015 to 2023 using the Code on Dental Procedures and Nomenclature (CDT) for single porcelain/ceramic and PFM/metal crowns. The study included participants (> 20 years old) who had both ceramic and PFM/metal complete-coverage single crowns inserted by dental students at UMSOD. Exclusion criteria were as follows: (1) anterior crowns because of different functional and esthetic requirements, (2) less than one-year follow-ups to ensure sufficient time to observe complications and failure, and (3) bridge abutments.

This study focused on the use of digital technology in clinic workflow and combined lithium disilicate (e.max) and zirconia crowns into the ceramic group, despite their different mechanical properties, to evaluate the overall effectiveness of ceramic crowns against PFM/metal crowns. Out of 45 participants who met the inclusion criteria, 14 were excluded: 7 with anterior crowns, 5 with less than one-year follow-ups, and 2 with bridge abutments. A total of 31 participants were finalized for this study, including 31 ceramic crowns and 31 PFM/metal crowns. Figure 1A illustrates a flow chart explaining the inclusion and exclusion criteria and Fig. 1B shows one of the participants in this study.

Fig. 1.

Recruitment of participants. (A) A flowchart of the study. (B) One of the 31 participants in this study. A green arrow indicates a cast crown. A yellow arrow indicates a ceramic crown

At UMSOD, once a ceramic crown is determined, a dental student prepares a tooth for the crown and makes the final impression using an intraoral scanner (Omnicam, Dentsply Sirona, Charlotte, NC, US) under faculty supervision. Lithium disilicate crowns are designed and fabricated in-house using a CEREC InLab MXCL Milling Unit (Dentsply Sirona, Charlotte, NC, US) with a CAD/CAM block (IPS e.max; Ivoclar Vivadent, Amherst, NY, US) by dental students and faculty. Zirconia crowns are fabricated using an in-lab milling machine by an external dental laboratory using the digital impression made by dental students. The ceramic crown insertion is performed at a separate visit using an adhesive luting composite (Variolink; Ivoclar Vivadent, Inc., Amherst, NY, US) for e.max crowns and a glass ionomer cement (FujiCEM 2; GC America, Alsip, IL, US) for zirconia crowns.

For PFM/metal crowns, dental students make the final impression using elastomers after preparing the tooth under faculty supervision. These crowns are then fabricated by an external dental laboratory. Before inserting PFM crowns, a metal coping try-in is performed prior to porcelain layering. PFM/metal crowns are inserted using a glass ionomer cement (FujiCEM 2).

All patients receive post-cementation instructions and are scheduled for follow-up. Long-term follow-up is conducted during routine recall visits, including clinical and radiographic assessments as needed.

From the participants’ records, the demographic information, medical, and social histories of the participants were collected. Data on the following tooth- and procedure-related variables were obtained: reasons for complete-coverage crowns, elapsed time (in years) since crown insertion, restoration location (premolar, first molar, or second molar), status of root canal treatment (RCT) (yes/no), history of crown lengthening procedure (CLP) (yes/no), presence of a post (yes/no), duration (in days) from making the final impression to crown insertion, and types of opposing dentition. All study outcomes were objectively recorded in the dental records, and no blinding was applied during outcome assessment.

For the survival analysis, the primary outcome was tooth survival. A failure event was defined as the extraction of the crowned tooth for any reason. Biological (e.g., caries, pulpal or periodontal disease) and biomechanical (e.g., crown dislodgement, crown fracture, tooth fracture) complications associated with the underlying teeth and crowns were recorded, along with any ensuing treatments. Tooth extractions, if occurred, were documented, including the elapsed time (in years) from crown insertion and reasons for extractions. The reasons for tooth extractions were categorized into restorative (recurrent caries or coronal tooth fracture) and endodontic (root fracture, failed RCT, pulpitis, or apical periodontitis) failure. Figure 2 shows examples of critical complications resulting in tooth loss.

Fig. 2.

Examples of tooth loss in one participant. (A) Apical periodontitis with root fracture was developed 1.4 years following crown insertion on the mandibular left second molar. (B) Pain associated with apical periodontitis and root fracture was developed 5.3 years following crown insertion on the maxillary left second molar

Statistical analysis

A post hoc power analysis was conducted using the tooth as the unit of analysis, consistent with our split-mouth design. With a total of 62 teeth (31 matched pairs), the study achieved approximately 80% power to detect a mean difference of 7 days in the elapsed time between final impression and crown insertion.

Descriptive statistics were prepared for each variable. The data were expressed as mean ± standard deviation or count. Wilcoxon signed-rank and chi-square tests were used to compare the cast and ceramic groups. The Kaplan-Meier curve was used to estimate the probability of survival for all teeth. The log-rank test was conducted to compare the survival of the cast and ceramic groups. Data analysis was performed with a software program, GraphPad Prism version 10.2.2 (San Diego, CA, USA); p < 0.05 was considered significant.

Results

A total of 62 teeth from 15 men and 16 women were included. The mean age of 31 participants at the time of crown insertions was 57 ± 14 years. Three participants were diagnosed with diabetes (9.7%) and there was no current smoker. All participants exhibited either gingivitis (14) or periodontitis (17) and received appropriate periodontal treatments. Table 1 summarizes the reasons for the 62 crowns. 89% of crowns were rendered because of restorative purposes, including 24 extensive fillings and caries, 12 recurrent caries on existing crowns, and 19 RCT. The crowns were inserted on the 27 first molars, followed by on the 18 second molars, and on the 17 premolars. Among the 31 cast crowns, 27 were PFM and 4 were metal crowns. The 31 ceramic crowns included 19 e.max and 12 zirconia crowns.

Table 1.

Clinical reasons for complete-coverage crowns

Reasons for rendering crowns (count)
1. Extensive filling or caries (24)
2. Root canal treated teeth (19)
3. Recurrent caries on existing crowns (12)
4. Coronal tooth fracture (4)
5. Defective crowns, such as overhanging or open margin (2)
6. Loose contact (1)

Table 2 summarizes the characteristics of teeth with ceramic (the ceramic group) or PFM/metal (the cast group) crowns during the observed period ranging from 1 to 8 years. The mean observation period was 4 years for the cast group and 3.3 years for the ceramic group with no significant difference between the two groups (median difference: 0.3 years, 97% confidence interval (CI) [-0.2 to 2.0], p = 0.18). Some teeth had pre-existing treatments including 24 RCTs, 7 CLPs, and 7 post insertions. There were no significant differences in the distribution of the aforementioned treatments, crown locations, or opposing dentitions between the two groups (p > 0.05). The mean elapsed time from making the final impression to crown insertion was 41 ± 19 days for the cast group and 21 ± 17 days for the ceramic group; the ceramic crowns were completed in a significantly shorter time (median difference: 16 days, 97% CI [6 to 29], p = 0.0006).

Table 2.

Characteristics of the cast and ceramic groups. SD = standard deviation; rct = root Canal treatment; clp = crown lengthening procedure

Parameters		Cast (n = 31)	Ceramic (n = 31)	p value
Location	1st molar	12	15	0.53 a
	2nd molar	11	7
	Premolar	8	9
RCT	Yes	11	13	0.60 a
	No	20	18
CLP	Yes	3	4	0.69 a
	No	28	27
Post	Yes	2	5	0.23 a
	No	29	26
Duration in days* (mean ± SD/ median)		40.5 ± 19.4 / 30	21.3 ± 16.8/ 16	< 0.01 b
Opposing Dentition	Tooth	30	28	0.5 a
	RPD	1	2
	Implant	0	1
Elapsed Year (mean ± SD/ range)		4.0 ± 2.1/ 1–8	3.3 ± 1.6/ 1–8	0.12 b

* Elapsed days between final impression/scan and crown insertion

^a Chi-square test: ^b Wilcoxon signed rank test

Table 3 summarizes the complications reported among the surviving teeth, showing no significant difference between the two groups (p = 0.15). In the cast group, 4 complications were reported: 2 recurrent caries (treated with a ceramic crown and silver diamine fluoride), and 2 endodontic issues (treated with RCTs). In the ceramic group, 3 complications were noted: coronal tooth fracture (no treatment), crown dislodgment (treated with a PFM crown), and recurrent caries (treated with amalgam). During the observed period, the tooth loss rate was 11.3%, with 7 teeth extracted: 2 second molars in the cast group, and 4 second molars and 1 first molar in the ceramic group. There was no significant difference in tooth loss between the groups (p = 0.49). Most extractions were due to endodontic reasons, except for one second molar in the ceramic group, which was extracted due to pain from an apical cyst originating from a third molar. No further treatments were performed after the extractions.

Table 3.

Complications among the surviving teeth and reasons for tooth loss in the cast and ceramic groups. RCT = root Canal treatment

The surviving teeth (n = 55)	Cast	Ceramic	P value
Reported Complications (number)	4	3	0.15 a
Endodontic (Apical abscess, Apical Periodontitis)	2	0
Restorative	2	3
The extracted teeth (n = 7)	Cast	Ceramic	P value
Occurrence (number)	2	5	0.49 a
Elapsed years (mean ± SD)	2.7 ± 1.8	3.2 ± 1.7
Endodontic (Failed RCT, Pain, Root fracture)	2	4
Apical cyst	0	1

^a Chi-square test

Figure 3A shows the Kaplan-Meier survival curve with 95% CI for survival of crowned teeth; at 5 years, the survival probability was 85% in all crowns. No significant difference in the probability of survival was observed between the cast and ceramic groups based on the log-rank test (Figs. 3B and 90% versus 80%, Log-rank test, χ² = 2.15, df = 1, p = 0.14). The hazard ratio for tooth loss in cast crowns compared to ceramic crowns was 0.3, 95% CI [0.07 to 1.4], indicating a non-significant trend toward a lower risk of tooth loss with cast crowns.

Fig. 3.

Survival analyses. (A) Kaplan-Meier survival curve with 95% confidence interval for all crowns (N = 62). (B) Kaplan-Meier survival curves for the ceramic (n = 31) and cast (n = 31) groups. Log-rank (Mentel-Cox) test, χ² = 2.15, degree of freedom = 1, p = 0.14

Discussion

In clinical studies, inter-individual variability, which includes participants’ medical and dental conditions, can affect treatment outcomes [19]. Our split-mouth design approach significantly reduces inter-individual variability in the estimates of treatment results by accounting for participants’ predisposing factors, such as oral hygiene, diet habits, or occlusion, which could otherwise impact the long-term survival of any restorations [17].

Existing clinical studies on the success and survival rates of ceramic crowns by dental students mainly utilized outdated camera systems like the CEREC Bluecam, which required spray powder that could distort the images of tooth preparations [3, 11]. One study scanned die stone models using the Bluecam system, with crowns made by facility lab-technicians [3]. In our study, dental students used the Omnicam scanner without spray powder to make the chairside final impression and participated in fabricating the in-house e.max crowns.

Unlike private practice, where the procedure is typically completed in a single visit, the completion of ceramic crowns in academic settings is usually divided into two appointments. The first appointment involves preparation, scanning, and temporization, while the second appointment focuses on the design, fabrication, and insertion of the crown. This approach provides students with more opportunities for hands-on experience while they learn to use this technology.

The integration of digital technology, such as intraoral scanners and in-office milling units, has significantly streamlined the process. When scanning and milling are performed at the school, the turnaround time is reduced as the crown can be fabricated on-site, eliminating the need for communication with a third-party dental laboratory. On the other hand, lab-manufactured crowns, while potentially offering higher precision and customization, involve additional steps and communication, which can extend the overall timeline from final impression to crown insertion.

The mean elapsed time from making the final impression to inserting crowns was twice as long for the cast group compared to the ceramic group. The process for PFM crowns at UMSOD involved an additional step—a metal coping try-in. Although the metal coping try-in was necessary to ensure a proper fit and for teaching purposes, this extra step required more time and coordination with a dental laboratory, leading to a longer overall treatment time.

The survival of teeth crowned and the survival/success of the crown itself are related but distinct concepts. The survival of teeth crowned refers to the overall longevity of the tooth that has received a crown, which can be affected by any underlying dental complications on teeth and crowns [13, 20, 21]. The survival/success of the crown itself pertains to the stability and functionality of the crown, which is influenced by the mechanical and physical properties of the crown. A successful crown remains intact, fits well, and performs its function without issues such as chipping, cracking, or fracturing of crowns [2]. No such complications from the crowns were observed in this study.

The majority of the crowns in this study were inserted because of dental caries. Restoring extensive caries requires considerable effort before fabricating crowns; 38.7% of teeth received RCT, 11.3% underwent CLPs, and 11.3% had posts inserted. CLPs reduce crestal bone support for the affected teeth. However, the 5-year survival probabilities for all crowns in this study and those with CLPs in another study [13] are similar (85% versus 87%). Therefore, CLPs may not significantly affect the survival of crowned teeth. No significant difference was observed in the success of single crowns on endodontically treated teeth restored with and without posts [22].

In contrast, RCT substantially affects the survival of crowned teeth. The overall estimated 10-year survival rate is 82% for endodontically treated teeth [23]. The survival of endodontically treated teeth depends on several factors. Teeth restored with complete-coverage crowns shortly after RCT and teeth with RCT performed by endodontists have a higher survival rate [24]. The location of teeth in the maxilla or mandible does not significantly affect their survival following RCT, but molars generally have a lower chance of survival compared to other tooth types [24].

Although complete-coverage crowns following RCT significantly improve the survival of endodontically treated teeth [25], one of the most common biological complications following crown insertions is the endodontic conditions [26, 27]. In this study, of the 8 endodontic complications observed, 6 (5 second molars and 1 first molar) resulted in tooth extractions. This may indicate that anatomical root and canal variations in second molars make RCT difficult [26]. All 7 extracted teeth (6 second molars and 1 first molar) were not replaced by any prostheses. Since the first molars bear 90% chewing efficiency [27], patients might not experience a noticeable deficiency in function.

While e.max and zirconia crowns are widely used in routine practices, material selection is often based on clinicians’ preference and case-specific factors rather than specific contraindications, as their clinical performance has shown comparable outcomes [21]. At UMSOD, zirconia crowns are used in cases of limited interocclusal space. In this study, 4 teeth with zirconia crowns and 1 tooth with e.max crown were extracted. Although the survival rate for the e.max crown was higher than that for the zirconia crown, all 5 extractions were due to endodontic failure. Therefore, differences in mechanical properties between zirconia and lithium disilicate may not significantly affect the prognosis of the crowns.

This study has several limitations. While the current sample size was adequate to detect a meaningful difference in the primary outcome, conducting the study at a single academic institute may not reflect broader clinical practices. Although all treatments were performed by students under faculty supervision, there was variability in preparation quality, impression accuracy, and follow-up times. Due to the retrospective nature of the study, potential confounding factors–such as parafunctional habits, bruxism, the quality of RCT, and detailed crown evaluations–were not controlled, as these conditions or all relevant parameters were not consistently documented in patient records. To ensure standardized assessments in the future studies, the United States Public Health Service (USPHS) criteria [28] or other validated evaluation tools should be employed. Despite these limitations, the split-mouth design improved internal validity of the study findings by minimizing inter-subject variability.

Conclusion

The ceramic crowns via CAD/CAM technology demonstrated notable effectiveness over the cast crowns, completing the procedure in significantly fewer days while achieving a comparable prognosis compared to the cast crowns at an academic institution. Future research should involve larger, more diverse samples across multiple institutions to improve generalizability by controlling confounding factors.

Author contributions

SR collected data and participated in writing the manuscript. KK collected data and participated in writing the manuscript. SO designed the study, collected data, carried out the statistical analysis, and participated in writing the manuscript. All authors have read and approved the manuscript, are aware of this submission, and agree with its publication.reasonable request.

Funding

There is no fund related to this study.

Data availability

The data of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board (IRB) at University of Maryland, Baltimore (UMB) (HP-00108993). Obtaining informed consent was waived by the IRB at UMB due to the absence of direct interactions with participants.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.