Multiunit fixed dental prostheses (FDPs) are integral to contemporary dental practice and are routinely provided as the part of fixed prosthodontic treatments. While conventional impressions have long been used to fabricate these prostheses with consistent clinical success, they are not without limitations.^[1] The challenges associated with the conventional technique include patient discomfort during impression-taking, difficulties in selecting appropriate trays, risk of dimensional changes due to delayed pouring, limited wettability of gypsum casts, and the potential for impression distortion during transport to the laboratory.^[2] Such issues can contribute to inaccuracies in the marginal and internal fit of the final prosthesis.^[3]

The integration of computer-aided design and computer-aided manufacturing technology has significantly enhanced the fabrication process of multi-unit FDPs. Digital impression techniques, employing laser or optical scanning, facilitate rapid and precise capture of both soft and hard tissues within the oral cavity, generating detailed digital impressions within minutes.^[4] Two primary approaches to digital impression techniques exist: direct and indirect. The indirect method involves creating a gypsum cast from a conventional elastomeric impression, which is then scanned using a desktop scanner. In contrast, the direct method utilizes intraoral scanners (IOS) to capture the necessary anatomical structures directly from the patient’s mouth. The direct digital approach offers several advantages, including reduced chairside time and the elimination of conventional steps such as tray selection, impression material use, and cast pouring, thereby increasing efficiency and patient comfort.^[5]

IOS systems employ various image acquisition technologies to optimize accuracy. For instance, the Sirona Cerec Bluecam system, introduced in 1987, uses structured blue light and requires a light dusting of powder to capture the images via principle of triangulation, where three straight beams of light intersect at one point. In contrast, newer powder-free systems like the Cerec Omnicam allow for full-arch or quadrant scanning using continuous image capture.^[6] The iTero scanner utilizes confocal imaging with a strobe light and a probe for focus, eliminating the need for powder. Similarly, the Trios system (3Shape) employs ultrafast optical sectioning and confocal microscopy for accurate, powder-free imaging.^[7] The Lava Chairside Oral Scanner (3M ESPE) differs by using active wavefront sampling with continuous video capture, although it requires a thin layer of titanium dioxide for optimal accuracy.^[8]

The internal and marginal fit of multi-unit FDPs is critical to their clinical success. Holmes et al. defined internal fit as the perpendicular distance between the intaglio surface of the restoration and the prepared tooth surface.^[9] Poor internal adaptation can hinder proper seating of the prostheses, while marginal discrepancies may lead to recurrent caries, gingival discoloration, and prosthesis failure.^[10,11] Both conventional and digital impression techniques have demonstrated the ability to reproduce sufficient detail for accurate fit. Studies suggest that a marginal discrepancy of ≤120 μm and an internal discrepancy of ≤200 μm at the occlusal surface are clinically acceptable.^[12,13] The efficacy of IOS has been previously proven in single-unit FDPs^[14] and completely edentulous arches.^[15]

A previous systematic review by Morsy et al.^[16] concluded that zirconia FDPs fabricated using digital impressions exhibited superior marginal and internal fit compared to those made with conventional techniques. However, with the continuous advancement of IOS technology, newer generations are overcoming the limitations of earlier models. Therefore, the present systematic review aims to incorporate the most recent evidence to reassess and update the comparative evaluation of conventional and digital impressions in the fabrication of multi-unit FDPs.

The null hypothesis stated that there is no significant difference in the accuracy of the marginal and internal fit of prostheses fabricated using conventional and digital impression techniques, nor is there a difference in the accuracy of the various IOS available on the market.

This systematic review was conducted following the Preferred Reporting Items of Systematic Reviews and Meta-analyses guidelines and has been registered on the International Prospective Register of Systematic Reviews (PROSPERO) CRD42023458713.

The research question was formulated using the PICO format: Patient (P): Patients requiring multi-unit FDPs, Intervention (I): Digital impressions, Comparison (C): Conventional impressions, and Outcome (O): Accuracy of marginal and internal fit. The primary question is digital impressions a feasible and accurate alternative to conventional impressions in achieving optimal marginal and internal fit in multi-unit FDPs? A secondary question was also explored: Do differences exist in scanning accuracy among the various IOSs currently available on the market?

A total of 236 studies were identified through electronic searches (PubMed: 126, Cochrane: 26, and Web of Science: 78) and hand searches (n = 6). After removing 108 duplicate studies, 128 remained for screening based on title and abstract, leading to the exclusion of 96 studies. Subsequently, 32 full-text articles were assessed for eligibility, with 18 being excluded due to lack of comparison with conventional impressions (n = 8), qualitative analysis only (n = 5), absence of internal fit discrepancy values (n = 3), and case report format (n = 2). Ultimately, 14 studies were included in the review, comprising 11 in vitro studies^[13,17-26] and 3 RCTs.^[27-29] [Figure 1].

A summary of the characteristics of the included studies is depicted in Tables 2 and 3. Ten studies employed the impression replica technique to evaluate discrepancies. In most of these studies, polyvinyl siloxane was used as the replica material; however, Moustapha et al.^[20] utilized polyether for the replica technique, which may have contributed to methodological variability due to differences in material properties. Svanborg et al.^[24] used the triple scan technique; likewise, a study by Bandiaky et al.^[17] used a combination of impression replica along with micro computed tomography, Kocaağaoğlu et al.^[19] used direct view technique and Shembesh et al.^[23] used optical comparator. The observed heterogeneity may be partly attributed to variations in fit measurement techniques across studies, which can affect the accuracy and comparability of reported outcomes and, consequently, influence the pooled estimates.

This systematic review unequivocally supports the superiority of digital over conventional impression techniques, clearly rejecting the null hypothesis. The meta-analysis results revealed a statistically significant difference in both marginal and internal fit when using digital impressions as compared to conventional impression. The results are contradictory to finding of Moustapha et al.^[20] wherein he reported better internal fit with conventional impressions (104 μ) than with digital impressions (115.6 μ) Similarly, in the study by Benic et al.,^[28] internal fit achieved through conventional impressions (127.5 μ) was better than that of Lava (140 μ) and Cerec (131.8 μ).

Although all the studies included in our review demonstrated an acceptable marginal fit (i.e., 120 μ or less) according to McLean and von Fraunhofer’s^[12] criteria in both the digital and conventional groups, however, our studies consistently showed acceptable internal discrepancies within the range of 100–200 μ.

No direct correlation could be established between the impression material used for the conventional impression or between the number of units in FDPs and the marginal or internal fit. Similarly, the studies included in the review did not establish a precise correlation between variables such as cement spacer, region of edentulousness, and the material of prostheses with the observed discrepancy.

Considering the substantial statistical heterogeneity, we opted for a subgroup analysis to compare the accuracy of various IOS used in the studies. Trios IOS demonstrated consistent results with lesser heterogeneity thus rejecting the second null hypothesis. Studies by Amornvit et al.^[32] and Renne et al.^[32,33] concluded that the Trios series depicted the best results consistent with our study’s results. The observed heterogeneities in our review are likely attributed to the diverse generations of IOS used in the included studies and the inclusion of studies which scanned different regions of the mouth. Despite this variability, the overall pooled estimates consistently favored digital impressions, suggesting that the digital workflow offers comparable or superior fit accuracy to conventional methods. Notably, newer generation IOS, equipped with advanced software, tends to yield lower discrepancy values for marginal fit.

Such differences in the accuracy of scanners could be attributed to their different scanning strategies. There is a considerable possibility of errors while scanning with IOS, as it captures approximately 1200 images during the scanning process.^[34,35] Errors may arise from the superimposition of these images during scanning and processing. These errors are more common in anterior teeth due to their steep inclines and smaller surface area. Additional errors can occur during computer processing, often due to filter algorithms and calibration issues.^[36,37]

Previous studies reported that different IOSs varied in scanning accuracy.^[5,38] No conclusion has been reached regarding the superiority of a certain IOS from the current subgroup analysis because few studies were included, and heterogeneity was high. However, the Trios scanner had the best performance in the digital group for marginal fit with 14% heterogeneity and SMD of −0.91, consistent with a previous review by Morsy et al.^[16]

The accuracy of IOS is also contingent on various factors, including blood and moisture control, gingival retraction, and the position of the marginal finish line on the tooth. Notably, the operator’s skill during scanning also plays a crucial role in determining accuracy.^[39,40]

The limitations of this study were inclusion of only a few clinical studies, with the majority being in vitro, where frameworks were fabricated and tested under controlled laboratory conditions. This may not fully represent in vivo scenarios, slightly reducing clinical relevance. In future more in vivo studies with a long-term follow-up are needed to assess the accuracy of the fit of the prostheses. Although statistically significant differences were observed, the clinical relevance of these findings remains uncertain, as the magnitude of the effect may fall within clinically acceptable limits.

As the certainty of evidence is limited, the conclusions should be interpreted with caution.

The Digital Impression technique has demonstrated better marginal and internal fit than the conventional impression technique