Introduction

Background

Computer-aided design/computer-aided manufacturing (CAD/CAM) technology is becoming popular since its introduction in the 1980s1. The laboratory or chairside digital technology has encompassed a wide range of dental treatment modalities such as crowns, onlays, inlays, veneers implant-supported restoration, and even full mouth rehabilitations2. The benefits that this technology offer to the dental community has been tremendous, however not been fully investigated in depth, especially from a clinical perspective3. The CAD/CAM technology has made it possible for clinicians to digitize implant positions using intraoral scanners and help in planning and designing restoration virtually4. Manufacture of the restoration has been simplified with the invention of the milling machine, which undeniably saves chairside time5.

What is already known and what is the need of this review?

Numerous studies have shown the benefits of digital impressions compared to conventional impression from a patient’s perspective6,7,8,9. Joda et al.6 has reported that digital impression is more preferred by patients. Schepke et al.7 and Wismeijer et al.8 in their clinical trials have also concluded higher patient preference for digital impression. Yuzbasioglu et al.9 have also reported significant differences in the total working time between digital and conventional impressions, where digital impressions recorded shorter duration. However, little evidence has been found on the clinical outcomes of fixed restorations fabricated from the digital impression. Clinical outcomes can be interpreted as the outcomes after the dental treatment which are observed and assessed clinically by the dental practitioner. For example, success rate, survival rate, fit (marginal, internal, occlusal) of the restoration, marginal leakage, gaps on the margin, bone loss, periodontal probing depth, bleeding on probing, plaque index, and other possible complications and failures. The authors have decided to include two types of prosthesis: implant-supported and tooth-supported fixed prosthesis in order to widen the scope of the population in this study. Hopefully, through this systematic review, clinicians will be able to provide more inclusive information, backed by evidence-based dentistry to patients when it comes to treatment planning for fixed prosthodontics.

Focus question (PICOS)

P – population: patients treated with the implant-supported prosthesis and tooth-supported fixed prosthesis

I – intervention: digital impressions or intraoral scanning

C – control: conventional or analogue impression

O – outcomes: clinical outcomes (success rate, survival rate, marginal fit, internal fit, occlusal fit, failures, marginal leakage, internal gap, marginal gap, periodontal probing depth, bleeding on probing, plaque index)

S: Study type: Randomized controlled trials

Materials and methods

Search strategy

The research question was formulated, based on the PICO framework as follows: Compared to conventional impression, does digital impression provide better clinical outcomes in implant-supported and tooth-supported fixed prosthesis?

The literature search strategy followed PRISMA 2020 guidelines for systematic review and meta-analysis. Databases used for literature search were PubMed and Cochrane Library. Keywords used during the search were “implant-supported prosthesis”, “tooth-supported fixed prosthesis”, “crown”, “bridge”, “implant”, “digital impression”, “intraoral scanner”, “conventional impression”, “clinical outcomes”, “success rate”, “survival rate”, “marginal fit”, “internal fit”, “occlusal fit”, “failure”, “marginal leakage”, “internal gap”, “marginal gap”, “periodontal probing depth”, “bleeding on probing”, and “plaque index”. BOOLEAN operators (AND, OR) were used to produce a more focused search result. The following combination of search terms were used:

  • Implant-supported prosthesis AND clinical outcomes

  • Implant OR crown OR bridge AND clinical outcomes

  • Implant OR crown OR bridge AND success rate

  • Implant OR crown OR bridge AND survival rate

  • Implant OR crown OR bridge AND failure

  • Implant OR crown OR bridge AND marginal fit

  • Implant OR crown OR bridge AND occlusal fit

  • Implant OR crown OR bridge AND internal fit

  • Implant OR crown OR bridge AND bone loss

  • Implant OR crown OR bridge AND digital impression

  • Implant OR crown OR bridge AND intraoral scanner

  • Tooth-supported fixed prosthesis AND clinical outcomes

  • Fixed prosthe* AND outcomes

  • Fixed prosthe* AND fail*

  • Fixed prosthe* AND survival

  • Fixed bridge AND digital impression

  • Implant OR crown OR bridge AND marginal gap OR internal gap

  • Implant OR crown OR bridge AND intraoral scanner AND probing depth OR bleeding on probing OR plaque index

  • Implant OR crown OR bridge AND conventional impression

  • Implant OR crown OR bridge AND digital impression

Review registry and ethical approval

The research was approved by the ethical committee of International Medical University (Research ID: IMU BDS I-01-2020(02).

Study selection and eligibility criteria

Study design selection was randomized controlled trials published by dental journals from January 2011 to September 2022. The obtained publications were imported into a reference management software (Endnote X9) and duplicates were removed. The studies were screened according to the inclusion and exclusion criteria. The inclusion criteria were as follows: (a) randomized control trials; (b) studies on tooth-supported fixed prosthesis and implant-supported prosthesis. The exclusion criteria were (a) in vitro or in situ studies; (b) case control studies.

Screening of studies

The first round of screening was based on titles and abstracts. A second round of screening was conducted by reviewing the full-text articles for the selected studies by 2 reviewers (SK and NJ). For each paper, a discussion between the two reviewers was held to maintain consensus. Cohen’s Kappa score was determined to measure the inter-rater reliability and the score was 0.8, indicating good agreement. The number of excluded studies including the reason for omission was documented.

Data extraction and collection

The tool used for data extraction was Cochrane Review Manager Software (RevMan 5.4). If multiple publications of the same trial/study are obtained, only the most recent, informative, or relevant data were included from these publications. Principle characteristics of the included studies were reviewed and summarized by two reviewers (SK and NJ).

Clinical outcomes identified

The relevant data to be extracted from the included studies were as follows: year of publication, sample size, mean age (range), settings of study, intraoral scanner system, type of restorations, total no. of implants (if placed), follow-up period, clinical outcomes (success rate/survival rate/marginal fit/occlusal fit/internal fit/marginal gap/internal gap/marginal leakage/failures/complications/periodontal probing depth/bleeding on probing/plaque index).

Risk of bias assessment

Bias risks were evaluated for all included studies in accordance with the criteria proposed by the Cochrane Handbook10. Quality and risk bias assessment of the RCTs was done using the Cochrane collaboration tool.

Meta-analysis scoring criteria assessment & synthesis of result

The eligibility of articles for meta-analysis was based on a modified version of the Newcastle-Ottawa scale (Table 1). All the articles that fulfilled all scoring scale criteria received a score of 3 and were selected for the meta-analysis. All articles that scored less than 3 were excluded for the reasons stated. A statistical method of inverse variance of random effect model with 95% confidence interval was used. Cochrane Review Manager 5.4 was used to derive the forest plots.

Table 1 Scoring criteria for meta-analysis article selection.
Full size table

Results

Study selection

The literature search has resulted in a total of 2250 studies from Cochrane Library and a total of 427 studies from PubMed (Fig. 1). The search results were imported into the reference management software programmer (Endnote X9) and after being combined, the duplicates were eliminated, leaving 714 articles for further review. The titles and abstracts of these studies were further screened and during this step, 615 articles had to be excluded, resulting in 99 papers potentially eligible for inclusion. Reasons for exclusion were irrelevant/inadequate data, ongoing studies, and different study designs. The full texts of these studies were retrieved and further screened according to the inclusion and exclusion criteria. A total of eight studies fulfilled the inclusion criteria for this systematic review.

Fig. 1: PRISMA flowchart
figure 1

Electronic search results.

Full size image

Summary and characteristics of the study

In total, there were eight randomized controlled clinical trials that have been selected for the systematic review. These studies were published between January 2011 and December 2022. Two studies, Gjelvold et al.11 and Zeltner et al.12 reported results on tooth-supported single crowns. Meanwhile, Benic et al.13 reported results on fixed dental prostheses. Ahberg et al.14 reported results for both tooth-supported single crowns and fixed dental prostheses. Three studies - Mühlemann et al.15, Mangano et al.16, Joda et al.17 have reported results on single implant-supported crowns while Cappare et al.18 reported results on the implant-supported prosthesis. A summary of the characteristics of the included studies is presented in Tables 2 and 3.

Table 2 Characteristics of included studies (tooth-supported fixed prosthesis).
Full size table
Table 3 Characteristics of included studies (implant-supported fixed prosthesis).
Full size table

Risks of bias in included studies

Quality and risk bias assessment of the randomized controlled trials are summarized in Fig. 2. The Cochrane Collaboration tool showed an overall low risk of bias in all the included studies. Gjelvold et al.11, Zeltner et al.12, Benic et al.13 and Cappare et al.14 did not report sufficient information on allocation concealment. Benic et al.15 reported a high risk of confounding bias and the risk was minimized by selecting a study design with intrasubject comparison. Joda et al.17 reported a high risk of selection bias, performance bias, and operator bias. Judgments included for the biases were potential allocation distortion, blinding was not applicable because of the trial design and only a single operator conducted the assessment.

Fig. 2: Risk bias assessment of included studies.
figure 2

Using Cochrane Collaboration tool.

Full size image

Meta-analysis

Three articles on implant-supported prosthesis were selected for this meta-analysis as they fulfilled the scoring criteria for meta-analysis. Out of eight articles, five articles were excluded from this meta-analysis and the reasons for the exclusion are tabulated in Table 4. The summary of the findings and the forest plot is shown in Fig. 3. Studies on tooth-supported fixed prosthesis were excluded due to not fulfilling the scoring criteria.

Table 4 Reasons of exclusion of studies for meta-analysis.
Full size table
Fig. 3: Forest plot.
figure 3

Comparing marginal bone loss between conventional impression group and digital impression group.

Full size image

Based on Fig. 3, conventional and digital impression techniques were considered as control and test groups, respectively. The result of this meta-analysis demonstrated that, there were no significant differences (P > 0.05) in marginal bone loss between conventional and digital impression techniques after ≥1 year of follow-ups. However, the overall forest plot shows an inclination towards the control group which signifies that there is higher marginal bone loss in the implant-supported prosthesis in the control (conventional) group. At 95% of a confidence interval, heterogeneity: Tau2 = 0.00; Chi2 = 1.19, df = 2 (P = 0.5), I2 = 0 and test for overall effect, Z = 1.47 (P = 0.14).

Discussion

This systematic review and meta-analysis were conducted to analyze the clinical outcomes of implant-supported prosthesis and tooth-supported fixed prostheses fabricated from digital impression technique and conventional impression technique. The term ‘clinical outcomes’ is quite broad; hence the authors of this review have decided to include a few clinical interpretations to aid the literature search. They were success rate, survival rate, marginal fit, occlusal fit, internal fit, marginal gap, internal gap, marginal leakage, failures, complications, periodontal probing depth, bleeding on probing, and plaque index. There were already systematic reviews done on clinical outcomes pertaining to quality of life and patient-related outcomes, hence the authors have not decided to include these two criteria in this study.

A similar systematic review was published by Özcan et al.19 in 2020 but it mainly focused on the clinical performance of partial and full-coverage fixed dental restorations fabricated from hybrid polymer and ceramic CAD/CAM materials. By conducting this systematic review, the authors hope that it will be able to provide valuable insight to clinicians who are interested in adopting digital impressions in their clinical practice. More inclusive information can be communicated to the patients on the longevity of restorations fabricated from digital and conventional impressions. Major influencing factors for survival of implant-supported prosthesis are implant mobility, swelling, or pain at the surgical site at the time of examination. For the success of the implant-supported prosthesis, marginal bone loss after one year of loading is an important deciding factor.

In this systematic review, one study has included the success rate data which was Mangano et al.16 in 2018 where they reported a 92% success rate in both digital and conventional groups after one year. Meanwhile, for the implant-crown survival rate, three studies (Mangano et al.16, Joda et al.17, Cappare et al.18) have reported 100% survival rates in both digital and conventional groups, showing no significant differences. Other than that, one study by Mühlemann et al.15 in 2020 showed a lower survival rate of 97.4% in the digital group.

The studies on implant-supported prosthesis by Mühlemann et al.15 and Joda et al.17 have also reported additional clinical parameters which were peri-implant crestal bone loss/marginal bone loss, biological outcomes (bleeding on probing, plaque scoring) and technical outcomes (fractures, chipping). For marginal bone loss, there were no significant findings reported between the digital and conventional groups, in all studies. Based on these findings, we can deduce that either digital or conventional impression technique has no significant effect on the clinical outcomes of implant-supported prosthesis. Despite the inconsistency of type of intraoral scanners used in these studies, it has minimal effect on the clinical outcomes.

Four studies (Gjelvold et al.11, Zeltner et al.12, Benic et al.13, and Ahberg et al.14) on digitally fabricated tooth-supported fixed prosthesis had focused on other clinical outcomes which were marginal fit, internal fit, interproximal contact points, and occlusal contacts. The types of tooth-supported fixed prostheses fabricated in these studies vary, from single crowns to fixed dental prostheses (bridge) up to six units. No data on survival or success rates were reported by the respective authors.

There was no significant difference in the marginal fit in two studies by Gjelvold et al.11 and Zeltner et al.12. These were consistent with the findings in a systematic review by Bandiaki et al.20 and Nagarkar et al.21. Meanwhile, Ahberg et al.14 found that the restorations from digital group had significantly better marginal fit than those fabricated from conventional impressions. Benic et al.13 on the other side has specifically mentioned that digitally fabricated zirconia frameworks presented a similar or better fit than conventionally fabricated metal frameworks in the shoulder region.

Gjelvold et al.11 found that the digital group reported significantly better occlusal contacts than the conventional group. This might be attributed to the difference in the interocclusal registration between the two impression techniques, where the conventional impression technique is more prone to distortion on removal from the mouth.

Regarding internal fit, there were two studies by Zeltner et al.12 and Benic et al.13 which reported on how conventionally fabricated restorations achieved a more favorable fit in the occlusal region than digitally fabricated restorations. Despite that, Benic et al.13 has provided an argument that the poor occlusal fit of the zirconia FDP may not be clinically relevant due to the high intrinsic stability of zirconia. This might require further research into knowing the reason and implications of occlusal discrepancy of digitally fabricated restorations.

Patzelt et al.22 in a systematic review published in 2015 has given an interesting insight where he mentioned CAD/CAM restorations showed promising results in the short term. However, the current scientific evidence is limited due to the restricted quality of the available studies and the paucity of data on long-term clinical outcomes of 5 years or more. The author of this review agreed with Patzelt et al.22 because it is evident that data of 5 years or more is very limited. There was only a study by Joda et al.17 in this review which had a follow-up period of up to 3 years. Hence, future clinical trials could consider longer follow-ups (5 years or more), as this will be able to provide more valuable insight into the long-term outcomes of the treatment.

Several limitations have been reported by all eight studies such as small sample size, a limited number of restorations placed and short follow-up period.

Gjelvold et al.11 has reported that different impression techniques were not performed on the same patient and the types of crowns fabricated for each patient varied. Having the same types of crowns for each patient will be able to help create a more fair and valuable comparison in terms of marginal fit because each crown type has different fit values.

Mühlemann et al.15 has reported a 15% deviation in color ratings and anatomical form for monolithic zirconia implant crowns. However, this clinical parameter is very dependent on the examiner’s skills, and this could lead to the potential of operator bias. Future studies on the color rating assessment of fixed restoration by different operators will be useful.

Besides that, when it comes to digital impressions, clinicians need to consider several aspects that may influence the flow of the impression. Ahberg et al.14 in 2016 discussed these two points: the accuracy of intraoral scanners (IOS) is highly dependent on practitioners’ skills and patient-related factors can affect the process. The patient-related factors include patient movement, the presence of saliva, and the intraoral space. Therefore, an operating learning curve and sufficient learning time will be required to develop the appropriate skills for clinicians. In addition to this, it might be helpful to provide exposure to the digital impression to the dental students in their undergraduate dental curriculum, so that they will be equipped with adequate knowledge of digital impressions before practicing. A study conducted by a group of dental students at the University of Otago in 2020 has shown that dental undergraduate students are ready and confident to uptake the new technology23. For the meta-analysis, the clinical outcome (marginal bone loss at one-year follow-up) was selected for forest plot generation as the data was evident in all three studies. I2 test results were 0% which indicated low heterogeneity of data. P value was >0.05, signifying that there is no statistically significant difference in the marginal bone loss between the conventional (control) and digital (test) groups. This finding is consistent with several findings from similar studies on the implant-supported prosthesis which evaluated marginal bone loss in the digital group and the conventional group (Peñarrocha-Diago et al.24, Ferrini et al.25,26,27. From this meta-analysis, we can conclude that the type of impression technique used does not have any significant effect on the marginal bone loss in the implant-supported prosthesis. However, since this data is only exclusive to implant-supported prosthesis, future meta-analysis will be needed to assess the clinical outcomes of tooth-supported fixed prosthesis. One limitation of this systematic review and meta-analysis is the low number of included studies. Besides that, other factors too may influence the clinical outcomes of implant-supported prosthesis and tooth-supported fixed prosthesis. For example, patient-related factors such as oral hygiene, and operator-related factors such as operators’ experiences and skills in digital and conventional impression technique.

Conclusion

For implant-supported prostheses, there is no significant differences in the clinical outcomes when fabricated with digital or conventional impression technique. For tooth-supported fixed prosthesis, the clinical outcomes in the digital impression group have shown favorable findings in terms of the marginal fit, despite showing discrepancy in the occlusal region. More clinical trials with longer follow-up periods (5 years or more) will be required. The type of impression technique has no effects on the marginal bone loss in implant-supported prosthesis.