Accuracy of single-unit ceramic crown fabrication after digital versus conventional impressions: A systematic review and meta-analysis

Abstract

In the present era when interest in digital dentistry is increasing, the published literature is still confusing about whether digital impression provides similar accuracy as provided by a conventional impression for the fabrication of a single-unit ceramic crown. The aim of the study was to systematically review the in vivo studies comparing marginal, axial, and occlusal fit of single-unit ceramic crowns fabricated after digital impressions with the ones fabricated after conventional impressions. The PubMed, Scopus, and Cochrane online databases were searched for studies comparing the digital impression technique with the conventional technique for single-unit ceramic crowns. Data extraction was done for the year of publication, type of study, country, number of patients, impression system (intraoral scanner [IOS] or conventional impression), marginal fit, axial fit, and occlusal fit. Ten studies were included for meta-analysis regarding the discrepancy in marginal fit, axial fit, and occlusal fit. The digital impression proved to be better than the conventional impression. The mean difference for marginal fit was 6.54 μm (heterogeneity P < 0.00001, I² = 93%), for axial fit 24.69 μm (heterogeneity P = 0.34, I² = 11%), and for occlusal fit 6.99 μm (heterogeneity P = 0.03, I² = 59%). The results of meta-analyses suggest that there is no significant difference between the impression systems (marginally favoring digital impression). The digital impression technique provided better marginal and internal fit of single-unit ceramic crowns than the conventional impression technique. The digital workflow using IOS provided a clinically acceptable marginal fit for single-unit crowns.

INTRODUCTION

An impression is the negative replica of the oral tissue. It is standard to record the oral tissues for various dental treatments, especially for indirect restorations. Hence, the accuracy of the impression is a significant determinant of the success and longevity of the restoration. The longstanding success of restoration is subject to marginal fit between restoration and prepared abutment tooth; any discrepancy may result in the development of secondary caries or pulpitis, plaque retention, and periodontal problems.[1,2] The documented factors which impact the marginal fit of dental restoration are the design of preparation, location of the marginal finish line (supra- or subgingival), restoration type, impression material, and technique.[3,4]

Conventional impression techniques involving elastomeric impression materials such as polyvinyl siloxane (PVS) and polyether are predominant. PVS impression materials are the best in this regard.[5] The digital impression technique came into the scene in the early 1980s, and digital scanners are predominantly used for three-dimensional capturing of the prepared tooth for a complete digital workflow for the fabrication of ceramic restoration. Intraoral scanners work on different principles, including active triangulation, parallel confocal imaging, active wavefront sampling, and stereophotogrammetry.[6]

The digital impression method has improved with time. It has improved patient acceptance, real-time impression display, reduced gag reflex, reduced chair time, lesser distortion compared to conventional counterpart, cost-effectiveness, minimal invasiveness, and easy communication with laboratories and patients due to the transfer of digital data and no need for any storage.[7] The studies usually mention the mean value of the marginal discrepancy.[8] The mean value of marginal discrepancy for ceramic crowns with the conventional impression technique was between 70 and 154 μm, whereas the same was between 49 and 149 μm with the digital impression technique in the included studies.[9,10,11,12,13] Mclean and von Fraunhofer,[14] however, concluded that the maximum tolerable marginal opening is 120 μm and this value has been accepted by most of the authors.[11,15,16]

The accuracy of digital impressions has been previously assessed mainly through in vitro studies.[17,18,19,20] There is a measurable discrepancy between the values provided by different studies with different restorations. Some studies indicate the advantage of the conventional impression method and others display that the complete digital workflow provides a better marginal fit. The lowest values of marginal discrepancy quoted by different studies need to be substantiated as to whether the digital impression technique provides similar results to the already established conventional impression technique.

A systematic review evaluated the marginal fit of the single-unit ceramic restoration considering both the in vitro and in vivo studies.[21] The present systematic review aims to compare the accuracy of the single-unit ceramic crowns fabricated with digital and conventional impression techniques considering only the in vivo studies and to assess whether the digital method provides an acceptable level of accuracy as provided by the conventional method.

METHODS

This study was designed as a systematic review of in vivo publications and a meta-analysis of the results. The study was registered in the PROSPERO database with registration number CRD42022297047. The search strategy was based on the PICO format considering the Population, Intervention, Comparison, and Outcome. An electronic search was conducted till August 2022 in PubMed, Scopus, and Cochrane databases [Figure 1]. The following terms were used in the search (Single unit crown OR fixed dental prostheses) AND (optical scanner OR intraoral scanner OR digital impression) AND (conventional impression OR silicone impression OR polyvinylsiloxane impression) AND (marginal fit OR marginal accuracy OR internal fit OR internal accuracy). References of the selected articles were also searched for inclusion. In addition, a hand search was performed in the following dental journals: The Journal of Prosthetic Dentistry, The Journal of Indian Prosthodontic Society, and Clinical Oral Investigations.

Figure 1.

PRISMA flowchart for literature search

Criteria for study selection

The following inclusion criteria were applied. Only in vivo studies, randomized controlled trials (RCTs), and prospective studies were included. Studies examining tooth-supported single-unit crowns were taken into consideration. Only those articles were included which compared both conventional and digital impression techniques in the same study. The details of the impression material and method used for conventional impression techniques and the system used for digital impressions had to be mentioned in the study. Publications only in the English language were included. In vitro studies comparing the digital and conventional impression techniques were excluded, and similarly, studies including implant-supported and multiunit fixed dental prostheses were excluded.

Two independent reviewers (JM and GS) screened the titles and abstracts as per the inclusion criteria mentioned earlier. The full text of those articles was retrieved which were selected after abstract screening. Through independent screening, the full text of all included articles for this systematic review was scrutinized by both reviewers for final selection, and any disagreement was resolved with the third reviewer SSD.

A reviewer (JM) extracted the data from the final included articles in the following manner: year of publication, type of study, country, number of patients, impression system (intraoral scanner [IOS] or conventional impression), marginal fit, axial fit, and occlusal fit. In the impression system, the type of IOS used and the type of conventional impression used were also noted. All these data were reevaluated by another reviewer (GS) independently.

Risk of bias assessment

Two independent reviewers (JM and GS) conducted the quality assessment of the included studies. Cochrane collaboration tool for risk of bias with Review Manager 5.4 was used to assess the risk of bias in all included studies [Figures 2 and 3]. All the studies had a low-to-moderate risk of bias. Randomization was present in all the studies except for two.[10,22] Participant and personnel blinding was mentioned in two studies[11,23] whereas it was unclear in all other studies. Outcome blinding was done in five studies.[9,11,16,23,24] No attrition was there in the included studies except for two studies.[10,15] In one of the studies[15] out of 34 patients, one dropped out of the study as it developed irreversible pulpitis after tooth preparation. In another study,[10] 31 patients with 64 teeth were initially scanned, out of which 7 patients with 17 teeth did not turn up for the final analysis, hence resulting in attrition, and finally, 24 patients with 49 teeth were included in the final analysis. Reporting bias was present in one of the studies[10] and unclear bias was in another study.[9]

Figure 2.

Risk of bias assessment of included studies using the Cochrane collaboration tool for risk of bias

Figure 3.

Risk of bias graph

RESULTS

A total of 702 articles were identified from an electronic search (Cochrane n = 21, PubMed n = 383, and Scopus n = 298 articles). Two hundred and fifty-five articles were removed as duplicates. Three hundred and eighty-nine were excluded based on title and abstract screening. After examination of the full text of 58 articles, 3 of them were excluded as they described implant-supported fixed prostheses, 17 of them described multiunit fixed dental prostheses, and 23 were excluded as they were in vitro studies. Two studies had the same sample, so one was excluded.[25] Two studies Brawek et al.[26] and Scotti et al.[27] compared two different intraoral scanners rather than comparing with any conventional impression technique and hence were excluded. Likewise, another study[28] evaluated metal-ceramic crowns and hence was excluded. One of the studies[23] compared only patient preferences using intraoral scanners and conventional impressions. This study was included in the qualitative analysis but not in the quantitative analysis. Finally, 11 studies were included in our systematic review [Figure 1].

Eight of the included studies were RCTs[9,11,12,13,15,16,23,24] and the rest were in vivo studies.[10,22,29] Ten studies were included for the meta-analysis of discrepancy in marginal fit,[9,10,11,12,13,15,16,22,24,29] four studies each for the discrepancy in axial fit,[9,11,16,24] and occlusal fit.[11,12,22,24] The forest plot was obtained using Review Manager 5.4.

A summary of the characteristic of included studies is presented in Table 1. Six studies used the TRIOS intraoral scanner,[9,10,12,22,24,29] and four used the LAVA COS system.[10,11,13,15] One study used the ITERO system,[16] and two studies the CEREC OMNICAM system.[10,29] One study[23] did not mention the system’s name; they only discussed that it was a powdered system. Likewise, different conventional impression materials were also used in different studies. Most of the studies used vinyl poly siloxane (VPS), while two studies used polyether,[11,12] and one used vinyl polyether silicone.[10]

Table 1.

Characteristics of included studies

Author, year	Type of study	Country	Number of patients	Impression system	Marginal fit	Axial fit	Occlusal fit
Haddadi et al., 2019[24]	RCT (split-mouth design)	Germany	19	CI=VPS	CI=125±39	CI=115±95.81	CI=182±147.4
			19	IOS=Trios 3	IOS=104±50	IOS=95±76.38	IOS=156±132.66
Bosniac et al., 2019[29]	In vivo study	Germany	63	CI=VPS	CI=143.29±100.71	NR	NR
			63	IOS 1=Cerec AC Omnicam	IOS 1=86.09±64.46
				IOS 2=Cara Trios	IOS 2=88.95±54.46
Berrendero et al., 2016[9]	RCT	Germany	30	CI=VPS	CI=119.9±59.9	CI=105.2±49.8	NR
			30	IOS=Trios	IOS=106.6±69.6	IOS=82.8±40.3
Rödiger et al., 2017[22]	In vivo study	Germany	20	CI=VPS	CI=117.94±74.21	NR	CI=164.22±73.17
			20	IOS=Cara Trios	IOS=147.88±63.88		IOS=207.60±69.99
Sakornwimon and Leevailoj, 2017[23]	RCT	Thailand	8	CI=VPS	NR	NR	NR
			8	IOS=NR
Ahrberg et al., 2016[11]	RCT	Germany	17	CI=Polyether	CI=70.40±28.87	CI=92.13±49.87	CI=171.51±60.98
			17	IOS=Lava COS	IOS=61.08±24.77	IOS=88.27±41.49	IOS=155.57±49.85
Zarauz et al., 2016[16]	RCT	Germany	26	CI=VPS	CI=133.51±48.78	CI=120.93±53.22	NR
			26	IOS=Cadent iTero	IOS=80.29±26.24	IOS=81.62±26.91
Boeddinghaus et al., 2015[10]	In vivo study	Germany	49	CI=Vinyl polyether silicone	CI=113±19	NR	NR
			49 (patient 24)	IOS1=Cerec	IOS1=149±26
				Omnicam IOS2=Trios	IOS2=112±13.75
				IOS3=3M Lava	IOS3=88±17
Pradíes et al., 2015[15]	RCT	Spain	33	CI=silicon	CI=91.46±72.17	NR	NR
			33	IOS=Lava COS	IOS=76.33±65.32
Gjelvold et al., 2016[12]	RCT	Sweden	13	CI=polyether	CI=154±66	NR	CI=158±65
			12	IOS=Trios	IOS=125±53		IOS=121±41
Syrek et al., 2010[13]	RCT	Germany	9	CI=VPS	CI=71±26	NR	NR
			9	IOS=Lava COS	IOS=49±18

RCT: Randomized controlled trial, CI: Conventional impression, IOS: Intraoral scanner, VPS: Vinyl poly siloxane, NR: Not reported

A confidence interval of 95% was used for all three meta-analyses. The inverse variance statistical method was used. The analysis model was fixed type. The mean difference between the studies was compared. While comparing the discrepancy in the marginal fit, the total number of cases was 439 for digital impressions and 440 for conventional ones. Similarly, for the axial fit, the total number of samples was 92 for both groups, and for the occlusal fit, the sample size was 68 for digital impression and 69 for conventional impression.

In the obtained forest plot, the mean difference for marginal fit was 6.54 μm [heterogeneity P < 0.00001, I² = 93%, Figure 4], and for axial fit 24.69 μm [heterogeneity P = 0.34, I² = 11%, Figure 5], and for occlusal fit 6.99 μm [heterogeneity P = 0.03, I² = 59%, Figure 6]. Both digital and conventional impressions showed similar results. This suggests no significant difference between both impression systems (marginally favoring digital impression).

Figure 4.

Forest plot of comparison of studies evaluating discrepancy in marginal fit. CI: Confidence interval, IV: Inverse variance, SD: Standard deviation

Figure 5.

Forest plot of comparison of studies evaluating discrepancy in axial fit. CI: Confidence interval, IV: Inverse variance, SD: Standard deviation

Figure 6.

Forest plot of comparison of studies evaluating discrepancy in occlusal fit. CI: Confidence interval, IV: Inverse variance, SD: Standard deviation

DISCUSSION

The results of this systematic review prefer the digital impression technique over the conventional technique, though marginally when considering single-unit ceramic crown fabrication. From the meta-analysis, it was evident that marginal fit values are better with digital impression-making. The mean values of marginal fit were below the 120 μm threshold value of clinical acceptability for all the studies except for Rödiger et al.[22] for which it was 147 μm, and for Boeddinghaus et al.,[10] it was 149 μm. There are several techniques to quantity the marginal or internal fit of the crown for in vitro studies like the triple scan protocol used by Anadioti et al.[3] Another method used by Abdel-Azim et al.[20] measured the marginal gap at four different points of each specimen under ×45 magnification with a stereomicroscope. Measurement of the marginal or internal fit under in vivo conditions is not possible. Hence, assessment of the clinical fit of the crowns is done by silicone replica technique; wherein the fabricated crown was cemented on its corresponding prepared tooth applying ultra-light body silicone, thereafter crown was embedded in light-polymerizing resin to stabilize the recorded interface. All the specimens were then sectioned in buccolingual cross-sections, and silicone replica film thickness was measured employing a stereomicroscope;[9] this is a validated and accepted measuring technique used by several studies in our review.[30]

In most of the included studies, the authors have used PVS for conventional technique. Various erstwhile studies have established that when conventional impression materials are compared, PVS gives the most accurate results.[31] When different conventional impression techniques such as the one-step putty wash impression technique, two-step putty wash technique with 2-mm relief space created with prefabricated copings, and two-step putty wash technique with a polyethylene spacer used in the first stage were compared, the putty/wash two-step impression technique with PVS has proven to be the most accurate.[32,33]

A previous systematic review evaluated the marginal fit of the single-unit ceramic restoration. It showed a moderate heterogeneity value (I²) of 50% for in vivo studies, as the same intraoral scanner Lava COS was used in all the included studies.[21] In our study, the calculation of heterogeneity (I²) for the discrepancy in marginal fit was 93%, for axial fit was 12%, and for occlusal fit 59%. These values are on the higher side except for the axial fit. These heterogeneities are probably due to the diverse generations of the IOS used by the included studies. The different IOS worked on different technologies such as active wavefront technology and confocal microscopy. With the advanced software used in newer generation IOS, the discrepancy values for marginal accuracy will be lower.

Various in vitro studies have concluded that the digital impression gives better accuracy for the values of marginal fit than the conventional technique.[20,34] Earlier, different systematic reviews and meta-analyses have been undertaken by different authors. However, they have included in vitro studies only[35] or both in vitro and in vivo studies[36] in their analysis. Different factors, such as saliva, sulcular fluid, blood, and movement of the patient, affect the accuracy of impression in in vivo conditions which are absent in the in vitro method. Hence, clinical relevance of in vivo studies is better than that of in vitro studies. As only in vivo studies have been included in this systematic review, it has better clinical relevance than the previous reviews which included in vitro studies only.

The precision of IOS can be influenced by various parameters such as the position of the finish line, control of moisture, and gingival retraction and scanning strategy. The operator’s experience plays a major role in the accuracy of the digital impression. Considering all these factors, future research should be done keeping the abovementioned factors as variables.

CONCLUSION

With the limitations of this systematic review and meta-analysis, it can be concluded that digital impression techniques can be used as a substitute for conventional techniques in the future, as they have comparable accuracy in terms of discrepancy in marginal fit, axial fit, and occlusal fit. Nonetheless, further improvements must be done in digital impressions. Moreover, more studies must be done to draw a concrete differentiation between digital impression and conventional impression techniques.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.