Marginal Adaptation and Internal Fit of 3D-Printed Provisional Crowns and Fixed Dental Prosthesis Resins Compared to CAD/CAM-Milled and Conventional Provisional Resins: A Systematic Review and Meta-Analysis

Abstract

The aim of this systematic review was to evaluate the marginal fit and internal adaptation of provisional crowns and fixed dental prostheses (FDPs) fabricated using 3D-printing resins and compared them with those fabricated by CAD/CAM (computer-aided designing/computer-aided manufacturing) milling and conventional resins. The null hypotheses tested were that there would be no differences in the marginal fit and internal adaptation of 3D-printed provisional crowns and FDP resins when compared to CAD/CAM-milled and conventional provisional resins. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were used to construct this systematic review. The focused PICO/PECO (Population, Intervention/Exposure, Comparison, Outcome) question was “Do provisional crowns and FDPs (P) fabricated by 3D-printing (I) have similar marginal adaptation and internal fit (O) when compared to those fabricated by CAD/CAM milling and conventional techniques (C)?”. The protocol used for this systematic review was pre-registered in the International Prospective Register of Systematic Reviews (PROSPERO). Electronic databases (e.g., MEDLINE/PubMed and Web of Science (Core Collection)) were systematically searched for indexed English literature published up to June 2022. In the initial electronic search of the selected databases, 519 articles were identified. Duplicates were removed, and screening was performed to select the articles that met the preset inclusion criteria. Sixteen studies were selected for qualitative analysis, but only ten of them provided comparative data and were selected for quantitative analysis. The modified CONSORT scale was used for qualitative analysis, and most of the included studies were rated to be of moderate quality. Based on the findings, it could be concluded that provisional crowns and FDPs fabricated from 3D-printing resins have a superior marginal fit and internal adaptation when compared to CAD/CAM-milled and conventional provisional resins; thus, they can be used as a dependable alternative to other resins.

1. Introduction

Provisional crowns and fixed dental prostheses (FDPs) are an integral part of fixed prosthodontic treatment and play an important role in aesthetics, pulp protection, gingival tissue management, maintaining the position of the prepared teeth, and diagnosis and treatment planning [1,2,3,4,5].

While managing complex full-mouth rehabilitation cases and implant prosthodontics, provisional restorations might need to be in function for prolonged time periods. In order to serve adequately for longer durations, provisional resins should possess adequate physical and mechanical properties [6,7].

Conventional techniques for the fabrication of provisional crowns and FDPs involve using poly(methyl methacrylate) (PMMA), poly(ethyl methacrylate) (PEMA), bis-acrylic, and dimethacrylate resins [6], which can be used either chairside (direct technique) or in production labs (indirect technique). PMMA resins have some inherent shortcomings, including high polymerization shrinkage, poor stain resistance, and heat production during polymerization, which has been overcome to a great extent by using bis-acrylic resins [5,8,9].

Continuous advancements in the field of data collection and fabrication have made digital dentistry more popular [10] by making procedures simpler and more user-friendly. With the advent of computer-aided design and computer-aided manufacturing (CAD/CAM) technology, provisional restorations can be fabricated via CAD/CAM milling (a subtractive manufacturing technique). This involves the use of pre-polymerized resin blocks that are milled using computer-assisted machining tools [5,11]. When compared to conventional techniques, restorations manufactured by CAD/CAM milling have been reported to be more reliable and to require less manufacturing time [5,7,11,12,13]. Three-dimensional (3D) printing technology is recent compared to techniques and involves 3D printing of provisional resins using the layering technique [14,15,16,17,18,19,20]. The procedure is cost-effective and causes less wastage of materials [5,7,16,21,22,23]. Various generations of 3D-printing machines are available, with each having its own merits and demerits [10,14,15,16,17,18,19,20].

In general, it is reported that 3D-printed provisional crowns and FDP materials have superior mechanical properties, whereas CAD/CAM-milled provisional materials have better physical properties when compared to conventionally fabricated provisionals [24]. However, using CAD/CAM technology for provisional restorations has some disadvantages, which include expensive manufacturing devices [14], higher fabrication costs, and limitations in complex milling shapes, particularly on the intaglio surface [25].

Internal adaptation and marginal fit are crucial parameters to consider when selecting the material for the fabrication of provisional crowns and FDPs [5,25,26,27,28]. Good marginal fit prevents microleakage, cement dissolution, plaque accumulation, and secondary caries [25,26,27,28,29,30,31,32], whereas good internal adaptation improves the retention form, resistance form, and durability of the provisional restoration [25,29,33]. Multiple studies have been conducted to compare these two parameters, but they have reported varied results [5,7,10,14,25,34,35,36,37,38,39,40,41,42,43,44].

To the best of our knowledge, there is no published systematic review comparing the marginal fit and internal adaptation of 3D-printed provisional resins with those of resins used in CAD/CAM milling and conventional techniques. The question arises as to whether provisional crowns and FDPs fabricated using recently introduced 3D-printing technology have a comparable fit (internal and marginal) when compared to those fabricated using other techniques. Thus, the aim of this systematic review was to evaluate the marginal fit and internal adaptation of provisional crowns and FDPs fabricated using 3D-printing resins and compare them with those fabricated using CAD/CAM milling and conventional resins. The null hypotheses tested were that there would be no differences in marginal fit and internal adaptation of 3D-printed provisional crowns and FDP resins when compared to CAD/CAM-milled and conventional provisional resins.

2. Materials and Methods

2.1. Permission and Registration

The framework of this systematic review was constructed based on the guidelines set out in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [45]. The protocol used for this systematic review was pre-registered in the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42022348367).

2.2. Eligibility Criteria

The inclusion and exclusion criteria used in this review are listed in

Table 1. Inclusion and exclusion criteria.

Inclusion Criteria	Exclusion Criteria
In vitro studies	Incomplete trials, unpublished abstracts, reports, commentaries, letters to the editor, case reports, technical reports, dissertations, cadaver studies, and opinion-based reviews [24]
Human clinical studies	Animal studies
Literature in the English language	Literature in any language other than English
Studies comparing the internal fit of the 3D-printed provisional fixed dental prostheses (FDPs) and crown materials with other materials and techniques used for the fabrication of provisional crowns and FDPs [24]	Studies comparing properties other than internal fit and marginal adaptation
Studies comparing the marginal adaptation of the 3D-printed provisional fixed dental prostheses (FDPs) and crown materials with other materials and techniques used for the fabrication of provisional crowns and FDPs [24]	Studies comparing only trueness, precision, 3D deviation, or accuracy of 3D-printed provisional materials with other types of provisional materials
-	Studies evaluating materials under trial.
-	Studies evaluating only 3D-printed provisional materials (examining the effects of various 3D-printing parameters, such as printing orientation, resin color setting, layer thickness, degree of conversion, etc., on marginal adaptation and internal fit) without comparing them with other types of provisional materials [24]

.

2.3. Exposure and Outcome

The focused PICO/PECO (Population, Intervention/Exposure, Comparison, Outcome) question was “Do provisional crowns and FDPs (P) fabricated by 3D-printing (I) have similar marginal adaptation and internal fit (O) when compared to those fabricated by CAD/CAM milling and conventional techniques I?”:

• P—Provisional crowns and FDPs;
• I—3D-printing method;
• C—CAD/CAM milling or conventional methods;
• O—Marginal adaptation and internal fit.

2.4. Search Strategy

The abovementioned PICO question was used in the search strategy. The electronic databases searched for indexed English literature were MEDLINE/PubMed, Scopus, Cochrane Library, and Web of Science (Core Collection). The systematic search was carried out in July 2022 by two independent authors (S.J. and A.A.) using the keywords (MeSH and non-MeSH) and Boolean operators listed in Table S1. A manual citation search was conducted for additional relevant articles that were not discovered during the electronic database search. Minor modifications in the search strategy were implemented to meet the requirements of each electronic search database.

2.5. Study Selection and Data Extraction

After the initial electronic and manual search, duplicate articles were removed, and no difference was found in the list of articles. Two authors (M.E.S. and S.J.) used preset inclusion and exclusion criteria to screen the titles and abstracts of these selected articles. Subsequently, two authors (M.H.D.A.W. and N.M.A.) reviewed the full texts of titles that appeared to meet the inclusion criteria or in which the abstracts did not provide sufficient information to make a decision. Four authors (S.J, M.H.D.A.W., M.E.S. and N.M.A.) discussed the disputed articles to resolve any disagreements. Later, four authors (S.J., H.A., A.A.A. and A.H.A.) extracted the relevant data from the selected articles and tabulated them in two predesigned tables.

Table 2. Summary of the studies included in the systematic review.

Author and Year	Study Type	Studied Characteristics	Sample Size (n)	Trade Name and Manufacturer of the Evaluated Materials	Main Chemical Composition	Specimen Fabrication Technique	Abutment Type (Tooth/Implant) and Region	Type of Framework (Single Crown/3-Unit FPD)	Scanning Software Program/Scanner Used
Park et al., 2016 [14]	In vitro	MF and IF	n = 120 (40 per group)	(A) Jet (Lang Dental Mfg Co Inc., Wheeling, IL, USA) (B) Pekkton Ivory (Cendres & Metaux, Biel, Switzerland) (C) E-Dent (Envision TEC, Gladbeck, Germany)	(A) PMMA (B) PEEK (C) PMMA	(A) Conventional self-curing (B) CAD/CAM milling (C) 3D printing	Implant transfer abutment: maxillary and mandibular 2nd premolar and maxillary 1st molar region	Crown	Noncontact blue-light scanner (Identica Blue; Medit, Seoul, South Korea)
Lee et al., 2017 [43]	In vitro	IF	n = 30 (10 per group)	(A) Vipi block (VIPI, Pirassununga, Brazil) (B) VeroGlaze MED620 (Stratasys, Eden Prairie, MI, USA) (C) ZMD-1000B (Dentis, Daegu, Korea)	(A) PMMA (B) Acrylic formulation (photopolymer) (C) Urethane methacrylate oligomer	(A) CAD/CAM milling (B) 3D printing (C) 3D Printing	Dental stone abutment tooth, maxillary 1st molar	Crown	Identica Hybrid, Medit, Seoul, Korea
Mai et al., 2017 [44]	In vitro	MF and IF	n = 36 (12 per group)	(A) Alike (GC Europe, Leuven, Belgium) (B) Ceramill TEMP (Amann Girrbach, Charlotte, NC, USA) (C) VeroGlaze MED 620 (Stratasys, Rehovot, Israel)	(A) PMMA (B) PMMA (C) Acrylic formulation (Photopolymer)	(A) Conventional self-curing (B) CAD/CAM milling (C) 3D printing	Dental stone study cast Crown: mandibular 1st molar	Crown	Desktop scanner (Ceramill Map 400; Amann Girrbach, Charlotte, NC, USA)
Alharbi et al., 2018 [10]	In vitro	MF and IF	n = 80 (40 per group, 10 each for 4 different types of finish lines: KE, C, RS, RSB)	(A) Polycon ae (Straumann, Basel, Switzerland) (B) Temporis: (DWS, Frankfurt am Main, Germany)	(A) PMMA (B) Hybrid composite resin	(A) CAD/CAM milling (B) 3D printing	Resin tooth, maxillary central incisor	Crown	Desktop optical scanner (Medit Identica BlueTM; Renishaw, Wharton, UK)
Al Deeb et al., 2019 [36]	In Vitro	Marginal integrity and internal adaptation	n = 30 (10 per group)	(A) TrimPlus, Bosworth, Keystone Dental, Gibbstown, NJ, USA (B) Ceramill TEMP (Amann Girrbach, Charlotte, NC, USA) (C) White Resin, Temporary CB (Formlabs, Somerville, MA, USA)	(A) PMMA (B) PMMA (C) PMMA	(A) Conventional self-curing (B) CAD/CAM milling (C) SLA, 3D printing	Implant analogues mounted on blocks of orthodontic resin	3-Unit fixed partial denture	Ceramill Map 400, Amann Girrbach, Charlotte, NC, USA
Peng et al., 2020 [5]	In vitro	IF and MD	n = 36 (12 per group)	(A) Protemp Plus (3M ESPE, Seefeld, Germany) (B) ZCAD Temp Fix (Harvest Dental, Brea, CA, USA) (C) NextDent C&B MFH (NextDent by 3D system, Soesterberg, the Netherlands)	(A) Bis-acrylic composite resin (B) PMMA (C) Methacrylic oligomers	(A) Conventional self-curing (B) CAD/CAM milling (C) 3D printing	Resin tooth, mandibular 1st molar	Crown	3D laboratory scanner (D2000; 3Shape, Copenhagen, Denmark).
Earar et al., 2020 [42]	In vitro	IF	n = 30 (10 per group)	N/M	(A) PMMA (B) PMMA	(A) CAD/CAM milling (B) 3D printing	Dental stone abutment tooth, mandibular 2nd premolar	Crown	Laboratory scanner AutoScan-DS200+ (Shining 3D)
Peng et al., 2020 [25]	In vitro	IF and MD	n = 48 (16 per group)	(A) Jet, (Lang Dental, Wheeling, IL, USA) (B) ZCAD Temp Fix 98, Harvest Dental, Brea, CA, USA) (C) (NextDent C&B MFH; NextDent by 3D system, Soesterberg, the Netherlands)	(A) Autopolymerized PMMA (B) PMMA (C) Methacrylic oligomers	(A) Conventional self-curing (B) CAD/CAM milling (C) 3D printing	Stereolithic resin dies, mandibular 1st molar	Crown	Laboratory scanner (D2000; 3Shape)
Sampaio et al., 2020 [7]	In vitro	IF (CT)	n = 48 (24 each for veneer and crown and 6 per group according to fabrication method)	(A) Acrílico Marche, Marche (B) Protemp3, M Espe (C) Vipiblock, Trilux (D) Raydent C&B for temporary crown and bridge, 3D materials	(A) Autopolymerized PMMA (B) Bis-acrylic resin (C) PMMA (D) Hybrid composite resin	(A and B) Conventional self-curing (C) CAD/CAM milling (D) 3D printing	3D-printed resin preparations Veneer: maxillary central incisor Crown: maxillary second molar	Veneer and crown	Intra-oral scanner (CEREC Omnicam, Charlotte, NC, USA)
Chaturvedi et al., 2020 [34]	In vitro	MF and IF	n = 90 (30 per group, 10 each for 3 different types of finish lines: C, RS, RSB)	(A) Protemp 4 (3M ESPE, Seefeld, Germany) (B) Ceramil TEMP (Amann Girrbach, Koblach, Austria) (C) Dental SG (Formlabs, Somerville, MA, USA)	(A) Bis-acrylic composite resin (B) PMMA and methacrylic acid ester-based crosslinked resin (C) Methacrylate oligomers	(A) Conventional self-curing (B) CAD/CAM milling (C) 3D printing	Dental stone abutment tooth Crown: maxillary second premolar	Crown	Desktop scanner (Ceramill Map 400; Amann Girrbach, Koblach, Austria)
Wu et al., 2021 [35]	In vitro	IF and MA	n = 48 (16 per group)	(A) Luxa crown (DMG America, Ridgefield Park, NJ, USA) (B) Lava Ultimate (3M ESPE, Seefeld, Germany) (C) Dima Print Denture Teeth (Kulzer North America, South Bend, IN, USA)	(A) Dimethacrylate resin (B) Resin nanoceramic material (C) PMMA	(A) Conventional self-curing (B) CAD/CAM milling (C) 3D printing	3D-printed resin dies, mandibular 1st molar	Crown	Model scanner (D2000, 3Shape)
Mohajeri et al., 2021 [41]	In vitro	MF	n = 42 (14 per group)	(A) GC Tempron (GC America Inc., Alsip, IL, USA) (B) PMMA blocks (Yamahachi Dental MFG. Co., Gamagori, Japan) (C) Freeprint Temp UV (Detax, Ettlingen, Germany)	(A) PMMA (B) PMMA (C) Methacrylate-based resins	(A) Conventional self-curing (B) CAD/CAM milling (C) 3D printing	Implant transfer abutment: maxillary canine region	Crown	3Shape Trios scanner (3Shape Trios; 3Shape, 3Shape Trios; 3Shape, Denmark, Denmark)
Aldahian et al., 2021 [40]	In vitro	IA and MI	n = 30 (10 per group)	(A) Jet Tooth Shade™ Self-Curing Acrylic Resin (Lang Dental Manufacturing Co Inc., Wheeling, IL, USA) (B) Cercon base PMMA blocks; DeguDent GmbH, Hanau, Germany (C) Freeprint Temp; DETAX GmbH & Co. KG, Ettlingen, Germany	(A) PMMA (B) PMMA (C) Methyl methacrylate	(A) Conventional self-curing (B) CAD/CAM milling (C) 3D printing	Dental stone dies Crown: mandibular 1st molar	Crown	S 50 Zenotec CAD; Wieland Dental, Pforzheim, Germany
Karasan et al., 2022 [38]	In vitro	IF	n = 60 (10 per group) (40 for 3D printing, divided into 4 groups based on the type of resin and printing technology)	(A) Protemp 4 (3M ESPE AG, Seefeld, Germany) (B) Telio-CAD (Ivoclar Vivadent, Schaan, Liechtenstein) (C: FL-ST) & (D: RS-ST) P-Pro-C&B (Institut Straumann AG, Basel, Switzerland) (E: FL-SH) & (F: RS-SH) SHERAprint-cb (SHERA Werkstoff-Technologie GmbH & Co., Lemförde, Germany)	(A) Bis-acrylic composite resin (B) PMMA (C and D) Flowable, light-curing acrylic-based composite (matrix: methacrylate) (E and F) Flowable, light-curing acrylic-based composite (matrix: methacrylate oligomers)	(A) Conventional self-curing (B) CAD/CAM milling (C–F) 3D printing	3D-printed resin master model: replacing mandibular 1st molar (tooth number 45, 46, 47)	3-Unit fixed partial denture	Iscan D104i, Imetric 3D SA, Courgenay, Switzerland
Sidhom et al., 2022 [37]	In vitro	MF	n = 8 (4 per group)	(A) Tempo-CAD PMMA discs, on dent, Bornova, Turkey (B) Temporary CB (Formlabs, Somerville, MA, USA)	(A) PMMA (B) Esterification products of 4,40-isopropylidenediphenol	(A) CAD/CAM milling (B) 3D printing	Dental stone dies, replacing mandibular 1st molar (tooth number 45, 46, 47)	3-Unit fixed partial denture	TRIOS 3 basic (3Shape, Copenhagen, Denmark)
Thakare et al., 2022 [39]	In vitro	IF and MF	n = 45 (15 per group) (30 for 3D printing, divided into 2 groups based on the type of printing technology)	(A) Ruthinium CAD/CAM PMMA blank, Badia Polesine, Italy (B) & (C) NextDent C&B MFH (3D Systems, Stone Mountain, SC, USA)	(A) PMMA (B and C) Methacrylic oligomers	(A) CAD/CAM milling (B and C) 3D printing	Metal master model, mandibular 1st molar	Crown	Imes-icore i3Dscan dental model scanner

MF: marginal fit, IF: internal fit, MA: marginal adaptation, MD: marginal discrepancy, MI: marginal integrity, IA: internal adaptation, N/M: not mentioned, CAD/CAM: computer-aided design/computer-aided manufacturing, SLA: stereolithography, KE: knife edge, C: chamfer, RS: rounded shoulder, RSB: rounded shoulder with bevel.

is a common table containing information related to authors’ names, years of publication, types of study, characteristics studied, sample size, brands of materials evaluated, their main composition, fabrication techniques, abutment types, types of framework, types of scanning program, and scanners used.

Table 3. Marginal adaptation and internal fit results.

Author and Year	Cement Film Thickness	3D Printer/Milling Machine Used	Layer Thickness and Orientation of Printing	Cementation or Non-Cementation Technique	Examination Method	Mean Marginal Discrepancy/Gap (μm)	Mean Internal Discrepancy/Gap (μm)	Authors’ Suggestions/Conclusions
Park et al., 2016 [14]	30 μm	Milling: 4-axial milling; (Cendres & Metaux SA, Biel, Switzerland) 3D printer: DLP; (Perfactory PixCera; Envision TEC, Gladbeck, Germany).	N/M	N/M	Silicone replica method; digital microscope (KH-7700; Hirox) at ×160 magnification	(A) Conventional: 120.92 ± 1.12 (B) Milled: 58.02 ± 19.75 (C) 3D-printed: 56.85 ± 22.24	(A) Conventional: (i) Intermarginal discrepancy: 149.71 (±60.70) (ii) Axiogingival: 47.97 (±19.14) (iii) Axio-occlusal: 46.88 (±17.23) (iv) Occlusal: 182.44 (±55.61) (B) Milled: (i) Intermarginal discrepancy: 96.70 (±25.38) (ii) Axiogingival: 67.02 (±17.97) (iii) Axio-occlusal: 81.41 (±30.64) (iv) Occlusal: 197.87 (±42.18) (B) 3D-printed: (i) Intermarginal discrepancy: 108.50 (±35.21) (ii) Axiogingival: 67.54 (±20.29) (iii) Axio-occlusal: 79.57 (±28.35) (iv) Occlusal: 167.81 (±41.86) Mean discrepancy: (A) Conventional: 100.20 ± 58.06 (B) Milled: 109.59 ± 71.53 (C) 3D-printed: 96.05 ± 50.23	Internal fit and marginal fit ranking: 3D printing > milling > conventional method All three methods were suitable, as the marginal discrepancy was within the clinically acceptable range The mean occlusal discrepancy was larger than the mean intermarginal, axiogingival, and axio-occlusal discrepancies
Lee et al., 2017 [43]	60 μm	(A) 3D-milling system, Zirkonzahn (B) 3D-printing system, Stratasys (C) 3D printing system, Dentis	N/M	N/M	Silicon replica method; image microscope system (EZVM- 452M, SomeTech) at ×300 magnification	NA	(A) Milled: (i) Intermarginal discrepancy: 119.1 ± 54.8 (ii) Axiogingival: 53.9 ± 23.2 (iii) Axio-occlusal: 247.0 ± 42.1 (iv) Occlusal: 266.3 ± 36.0 3D-printed (Stratasys): (i) Intermarginal discrepancy: 115.6 ± 68.4 (ii) Axiogingival: 137.4 ± 57.7 (iii) Axio-occlusal: 171.0 ± 74.5 (iv) Occlusal: 172.4 ± 43.3 (C) 3D-printed (Dentis): (i) Intermarginal discrepancy: 64.3 ± 30.1 (ii) Axiogingival: 93.8 ± 36.6 (iii) Axio-occlusal: 98.7 ± 30.2 (iv) Occlusal: 107.5 ± 34.5 Mean discrepancy: (A) Milled: 171.6 ± 97.4 (B) 3D-printed (Stratasys): 149.1 ± 65.9 (C) 3D-printed (Dentis): 91.1 ± 36.4	Internal fit ranking: 3D printing > milling
Mai et al., 2017 [44]	60 μm	Milling: 5-axis milling machine (Ceramill Motion 2; Amann Girrbach) 3D printer: PolyJet 3D printer (Object Eden 260VS; Stratasys, Rehovot, Israel).	Layer thickness: 5 μm Orientation: N/M	Non-cementation technique	Silicone replica method; microscope (MM-40; Nikon Corp, Tokyo, Japan) at ×50 magnification	Absolute marginal discrepancy (A) Conventional: 163 ± 86 (B) Milled: ≈101 ## (C) 3D-printed: 99 ± 19	(A) Conventional: Axial: 87 ± 62 μm Occlusal region: ≈186 and 217 ## (B) Milled: Axial: 125 ±30 micro m Occlusal region: 301 ± 74 and 328 ± 108 (B) 3D-printed: Axial: 139 ± 23 mm Occlusal region: ≈151 and 180 ##	Marginal fit ranking: 3D printing > milling > conventional method Internal fit ranking: conventional method > 3D printing > milling 3D printing could be an alternative approach to fabricating interim crowns
Alharbi et al., 2018 [10]	30 μm (Additional vertical space of 80 μm and a horizontal space of 30 micro m)	Milling: 5-axis milling machine (Wissner Ltd.; Göttingen, Germany) 3D printer: stereo- lithography-based 3D printer (DW028D, DWS)	Layer thickness: 50 μm Orientation of printing: N/M	N/M	Micro-CT scanning and analysis (Skyscan 1072 Bruker micro-CT, Kontich, Belgium). at × 400 magnification	(A) Milling: Vertical gap: (i) KE: 39 ± 21 (ii) C: 40 ± 5 (iii) RS: 34 ± 13 (iv) RSB: 19 ± 4 Horizontal gap: (i) KE: 31 ± 8 (ii) C: 25 ± 3 (iii) RS: 41 ± 8 (iv) RSB: 33 ± 4 AMD (i) KE: 56 ± 25 (ii) C: 54 ± 5 (iii) RS: 52 ± 16 (iv) RSB: 38 ± 3 (B) 3D printing: Vertical gap: (i) KE: 16 ± 10 (ii) C: 25 ± 6 (iii) RS: 21 ± 4 (iv) RSB: 26 ± 10 Horizontal gap: (i) KE: 25 ± 9 (ii) C: 22 ± 4 (iii) RS: 20 ± 3 (iv) RSB: 25 ± 4 Absolute marginal discrepancy (AMD) (i) KE: 43 ± 23 (ii) C: 48 ± 8 (iii) RS: 41 ± 5 (iv) RSB: 33 ± 6	(A) Milling: Incisal gap: (i) KE: 97 ± 17 (ii) C: 271 ± 18 (iii) RS: 259 ± 19 (iv) RSB: 208 ± 11 Mid-axial gap: (i) KE: 80 ± 6 (ii) C: 84 ± 7 (iii) RS: 111 ± 9 (iv) RSB: 99 ± 7 (B) 3D printing: Incisal gap: (i) KE: 93 ± 14 (ii) C: 213 ± 31 (iii) RS: 210 ± 6 (iv) RSB: 161 ± 15 Mid-axial gap: (i) KE: 39 ± 5 (ii) C: 45 ± 3 (iii) RS: 51 ± 4 (iv) RSB: 29 ± 2 Mean discrepancy (A) Milling: (i) KE: 89 ± 8 (ii) C: 177 ± 10 (iii) RS: 185 ± 11 (iv) RSB: 154 ± 7 (B) 3D printing: (i) KE: 66 ± 8 (ii) C: 149 ± 15 (iii) RS: 130 ± 4 (iv) RSB: 95 ± 7	Internal fit and marginal fit ranking: 3D printing > milling IF and MF are more affected by fabrication methods compared to the type of finish line
Al Deeb et al., 2020 [36]	200 μm	Milling: (Ceramill Motion II, Amann Girrbach, NC, USA 3D Printer: SLA; Form 2, FORMLABS, Somerville, MA, USA	N/M	Non-cementation technique	Micro- CT scanner (Skyscan 1172 High Resolution, Brukker, Belgium).	(A) Conventional: 283.31 ± 98.67 (B) Milled: 68.24 ± 18.11 (C) 3D-printed: 84.70 ± 27.59	(A) Conventional: 106.73 ± 54.98 (B) Milled: 77.75 ± 8.53 (C) 3D-printed: 85.8 ± 23.64	Marginal fit ranking milling > 3D Printing > conventional method Internal fit ranking: milling > 3D printing > conventional method
Peng et al., 2020 [5]	75 μm	Milling: DWX-51D Milling Machine, Roland DGA Co. (Irvine, CA, USA) 3D printer: digital light processing (DLP), MiiCraft 125; Young Optics	N/M	Both non-cementation and cementation techniques	Non-cementation technique: silicone replica method Cementation technique: μCT (X5000 CT system, North Star Imaging)	(i) Silicone replica method: (A) Conventional: 260 ± 110 (B) CAD/CAM milling: 180 ± 60 (C) 3D printing: 160 ± 40 (ii) Micro-CT scan technique: (A) Conventional: 240 ± 80 (B) CAD/CAM milling: 80 ± 40 (C) 3D printing: 80 ± 30	(1) Micro-CT scan technique (2D) (A) Conventional: 210 ± 110 (B) CAD/CAM milling: 160 ± 60 (C) 3D printing: 170 ± 40 (2) Cement space volume in mm3 (i) Silicone replica method: (A) Conventional: 59.18 ± 12.61 (B) CAD/CAM milling: 39.67 ± 5.88 (C) 3D printing: 36.55 ± 4.22 (ii) Micro-CT scan technique(3D): (A) Conventional: 28.76 ± 4.70 (B) CAD/CAM milling: 23.21 ± 1.58 (C) 3D printing: 26.64 ±3.07	Internal fit and marginal fit ranking: 3D printing = milling > conventional method
Earar et al.. 2020 [42]	20 μm	Milling: 5-axis dental milling machine, Rolland DWX-50 3D printer: DLP: MoonRay S 100 (SprintRay, Los Angeles, CA, USA)	N/M	Non-cementation technique	Electronic digital caliper Powerfix Profi+ (OWIM G mbH-AG, Neckarsulm, Gaermany)	NA	Inner diameter of PMMA crowns obtained (in mm): (A) Milling: (i) Gingival: B-L: 5.81 ± 0.118 M-D: 4.74 ± 0.101 (ii) Occlusal B-L: 4.78 ± 0.150 M-D: 3.80 ± 0.093 (B) 3D Printing: (i) Gingival: B-L: 5.81 ± 0.110 M-D: 4.63 ± 0.157 (ii) Occlusal B-L: 4.69 ± 0.158 M-D: 3.74 ± 0.100	Internal fit ranking: 3D printing > milling
Peng et al., 2020 [25]	60 μm	Milling: DWX-51D Milling Machine; Roland DGA Co. 3D printer: DLP: MiiCraft 125; Young Optics	N/M	Both non-cementation and cementation techniques	Non-cementation technique: silicone replica method Cementation technique: μCT (X5000 CT system, North Star Imaging)	Micro-CT scan technique (2D) (A) Conventional: 240 ± 90 (B) CAD/CAM milling: ≈180 ## (C) 3D printing: 160 ± 50	(1) Micro-CT scan technique(2D) (A) Conventional: 240 ± 120 (B) CAD/CAM milling: 170 ± 30 (C) 3D printing: 170 ± 40 (2) Cement space volume in mm3 (i) Silicone replica method: (A) Conventional: 92.04 ± 30.21 (B) CAD/CAM milling: 39.46 ± 5.45 (C) 3D printing: ≈36 ## (ii) Micro-CT scan technique(3D): (A) Conventional: 33.67 ± 6.91 (B) CAD/CAM milling: 24.09 ± 4.28 (C) 3D printing: ≈26 ##	Internal fit ranking: 3D printing = milling > conventional method Marginal fit ranking: 3D printing ≥ milling > conventional method
Sampaio et al., 2020 [7]	N/M	Milling: CEREC MCXL; Dentsply Sirona 3D printer: SLA: Photon; Anycubic	Layer thickness: 100 μm Orientation of printing: N/M	Cementation technique	μCT apparatus (μCT 40; Scanco Medical AG, Wangen-Brüttisellen, Switzerland).	NA	Mean cement film thickness in μm (i) Veneer (A) Conventional PMMA: 140 ± 30 (B) Bis-acrylic: 90 ± 40 (C) Milled: 180 ± 30 (D) 3D-printed: 320 ± 30 (ii) Crown (A) Conventional PMMA: 340 ± 60 (B) Bis-acrylic: 280 ± 80 (C) Milled: 240 ± 40 (D) 3D-printed: 620 ± 80	Internal fit ranking for veneer: conventional bis-acrylic > conventional PMMA > milling > 3D printing Internal fit ranking for crown: milling > conventional bis-acrylic > conventional PMMA > 3D printing
Chaturvedi et al., 2020 [34]	No cementation space	Milling: 5 axis milling machine; Ceramill Motion 2; Amann Girrbach, Austria 3D printer: Form 2 3D printer Formlabs Inc., Somerville, MA, USA	Layer thickness: 50 μm Orientation of printing: N/M	Non-cementation technique	SEM at ×27 magnification	(A) Conventional (i) C: 104.4 ± 34.1 and 106.7 ± 34.1 (ii) RS: 96.2 ± 24.3 and 98.4 ± 24.3 (iii) RSB: 92.2 ± 21.1 and 94.4 ± 21.1 (B) Milled (i) C: 54.3 ± 11.3 and 56.6 ± 11.3 (ii) RS: 51.4 ± 13.7 and 53.6 ± 13.7 (iii) RSB: 39.3 ± 5.9 and 42.5 ± 5.9 (C) 3D-printed (i) C: 43.4 ± 9.2 and 45.6 ± 9.2 (ii) RS: 32.6 ± 7.5 and 34.8 ± 7.5 (iii) RSB: 30.6 ± 5.3 and 32.8 ± 5.3	(A) Conventional (i) C: (a) Axial: 45.1 ± 24.5 and 46.5 ± 24.5 (b) Occlusal: 223.2 ± 24.1, 263.2 ± 9.3, and 225.4 ± 24.1 (ii) RS: (a) Axial: 71.6 ± 23.5 and 73 ± 23.5 (b) Occlusal: 214.6 ± 10.7, 145.5 ± 7.2, and 216.8 ± 10.7 (iii) RSB: (a) Axial: 63.1 ± 13.4 and 64.4 ± 13.4 (b) Occlusal: 176.8 ± 28, 125.5 ± 7.7, and 179 ± 28 (B) Milled (i) C: (a) Axial: 83.4 ± 12.8 and 84.8 ± 12.8 (b) Occlusal: 265.6 ± 26, 172.2 ± 10.9, and 267.1 ± 26 (ii) RS: (a) Axial: 130.7 ± 14.1 and 112 ± 14.1 (b) Occlusal: 258.3 ± 13.2, 165.2 ± 7.9, and 260.6 ± 13.2 (iii) RSB: (a) Axial: 99.4 ± 8.4 and 110.7 ± 8.4 (b) Occlusal: 201.8 ± 22.2, 138.4 ± 10.9, and 206.1 ± 22.2 (C) 3D-printed (i) C: (a) Axial: 24.2 ± 11.5 and 25.5 ± 11.5 (b) Occlusal: 214.3 ± 21.3, 151.1 ± 14.1, and 216.5 ± 21.3 (ii) RS: (a) Axial: 83.5 ± 12.1 and 84.5 ± 12.1 (b) Occlusal: 212.8 ± 6.8, 132.3 ± 6.2, and 214.4 ± 6.8 (iii) RSB: (a) Axial: 82.2 ± 6.6 and 82.6 ± 6.6 (b) Occlusal: 162.9 ± 16.1, 151.1 ± 16, and 165.2 ± 16.1	Internal fit ranking: 3D printing > conventional method > milling Marginal fit ranking 3D printing > milling > conventional method Marginal gap: lowest for rounded shoulder with bevel finish line
Wu et al., 2021 [35]	60 μm	Milling: 5-axis dental milling machine (DWX-51D, Roland DGA, Frenchs Forest, NSW, Australia) 3D printer: (cara® Print 4.0, Kulzer North America, South Bend, IN, USA)	N/M	Cementation technique	Internal Adaptation: silicone-checked method (measured using a microscope (FMA050; AmScope, Irvine, CA, USA) at ×10 magnification) Marginal adaptation evaluated by (a) PVS replica method (measured using a stereoscope (Measurescope 20; Nikon, Tokyo, Japan) at ×75 magnification) (b) Swept-source optical coherence tomography (SS-OCT) scanning technique ((Yoshida Dental OCT, Yoshida Dental Mfg, Tokyo, Japan)	(A) Conventional (a) Vertical marginal discrepancy: (i) PVS replica method: 21.5 ± 30.3 (ii) OCT: 20.3 ± 41.5 (b) Horizontal marginal discrepancy: (i) PVS replica method: 63.3 ± 55.2 (ii) OCT: 72.7 ± 59.5 (c) AMD: (i) PVS replica method: 71.3 ± 64.9 (ii) OCT: 82.7 ± 65.8 (B) Milling: (a) Vertical marginal discrepancy (i) PVS replica method: 7.0 ± 20.1 (ii) OCT scanning technique: 4.5 ± 14.8 (b) Horizontal marginal discrepancy: (i) PVS replica method: 94.8 ± 59.8 (ii) OCT scanning technique: 97.0 ± 54.3 (c) AMD (i) PVS replica method: 96.9 ± 60.2 (ii) OCT: 99.6 ± 54.6 (C) 3D-printed (a) Vertical marginal discrepancy: (i) PVS replica method: 13.3 ± 24.2 (ii) OCT: 6.8 ± 18.4 (b) Horizontal marginal discrepancy: (i) PVS replica method: 118.7 ± 67.7 (ii) OCT: 146.4 ± 36.6 (c) AMD (i) PVS replica method: 120.8 ± 70.9 (ii) OCT: 143.1 ± 39.9	Absolute internal discrepancy (A) Conventional: ≈49 ## (B) Milling: ≈70 ## (C) 3D printing: ≈68 ##	Internal fit ranking: conventional method > 3D printing > milling Marginal fit ranking: conventional method > milling > 3D printing
Mohajeri et al., 2021 [41]	N/M	Milling: rainbow TM mill, Dentium, South Korea 3D printer: digital light processing printer (Prodent Labx, Product Bonyan Mecatronic, Tabriz, Iran)	N/M	Non-cementation	Stereomicroscope (SZX16, Olympus, Tokyo, Japan) at ×100 magnification	Mean of marginal gap (A) Conventional: 51.23 ± 21.7 (B) CAD/CAM milling: 75.28 ± 22.14 (C) 3D printing: 91.4 ± 36.07	NA	Marginal fit ranking: conventional method > milling > 3D printing
Aldahian et al., 2021 [40]	N/M	Milling: CAD-CAM milling machine (Versamill; Axsys Dental solutions, Wixom, MI, USA). 3D Printer: stereolithography-based 3D printer (MiiCraft 125; MiiCraft, Jena, Germany)	Layer thickness: 50 μm Orientation of printing: 0 degrees	N/M	Bruker micro CT (Skyscan 1173 high-energy spiral scan micro-CT; Skyscan NV, Kontich, Belgium)	(A) Conventional: 395.89 ± 80.33 (B) Milled: 244.95 ± 19.65 (C) 3D-printed: 211.87 ± 17.8	(A) Conventional: 269.94 ± 64 (B) Milled: 269.52 ± 54.17 (C) 3D-printed: 197.82 ± 11.72	Marginal fit ranking 3D printing > milling > conventional method Internal fit ranking: 3D printing > milling = conventional method
Karasan et al., 2021 [38]	30 μm	Milling: 5-axis milling Machine; Zenotec Select Hybrid, Wieland Dental, Germany 3D printer: (C: FL-ST) and (E: FL-SH): SLA FormLabs 2, Form-Labs, Somerville, MA, USA (D: RS-ST) DLP: P20II, Rapid-Shape, Heimsheim, Germany (F: RS-SH) DLP: P30, Rapid-Shape, Heimsheim, Germany	(C and D) Layer thickness: 100 μm Orientation of printing: 45 degrees (E) Layer thickness: 40 μm Orientation of printing: 45 degrees (F) Layer thickness: 50 μm Orientation of printing: 45 degrees	N/M	Optical laboratory scanner (Iscan D104i; Imetric 3D SA). Triple-scan method	Marginal (A) Conventional self-curing (i) Premolar: 165.5 ± 38.89 (ii) Molar: 188.38 ± 73.76 (B) CAD/CAM milling: (i) Premolar: 117.60 ± 62.65 (ii) Molar: 83.31 ± 68.72 (C–F) 3D printing (C) FL-ST: (i) Premolar: 172.50 ± 30.70 (ii) Molar: 114.76 ± 17.13 (D) RS-ST: (i) Premolar: 160.27 ± 33.69 (ii) Molar: 110.52 ± 24.3 (E) FL-SH: (i) Premolar: 103.07 ± 18.01 (ii) Molar: 69.10 ± 23.25 (F) RS-SH (i) Premolar: 130.92 ± 27.58 (ii) Molar: 102.35 ± 31.84	Axial: (A) Conventional self-curing (i) Premolar: 90.49 ± 26.15 (ii) Molar: 101.7 ± 46.89 (B) CAD/CAM milling: (i) Premolar: 109.72 ± 39.77 (ii) Molar: 93.94 ± 50.45 (C–F) 3D printing (C) FL-ST: (i) Premolar: 108.08 ± 15.07 (ii) Molar: 98.28 ± 13.13 (D) RS-ST: (i) Premolar: 122.60 ± 25.14 (ii) Molar: 85.54 ± 11.85 (E) FL-SH: (i) Premolar: 101.18 ± 6.81 (ii) Molar: 67.92 ± 14 (F) RS-SH (i) Premolar: 100.13 ± 16.40 (ii) Molar: 79 ± 20.5 Occlusal: (A) Conventional self-curing (i) Premolar: 142.38 ± 44.28 (ii) Molar: 150.78 ± 73.74 (B) CAD/CAM milling: (i) Premolar: 153.74 ± 70.41 (ii) Molar: 113.11 ± 82.95 (C–F) 3D printing (C) FL-ST: (i) Premolar: 116.29 ± 28.23 (ii) Molar: 67.6 ± 8.84 (D) RS-ST: (i) Premolar: 168.23 ± 36.33 (ii) Molar: 131.27 ± 11.74 (E) FL-SH: (i) Premolar: 169.48 ± 52.14 (ii) Molar: 103.03 ± 23.79 (F) RS-SH (i) Premolar: 155.33 ± 23.17 (ii) Molar: 128.12 ± 37.17 Total: (A) Conventional self-curing (i) Premolar: 128.04 ± 31.04 (ii) Molar: 170.98 ± 105.47 (B) CAD/CAM milling: (i) Premolar: 143.55 ±48.68 (ii) Molar: 98.85 ± 63.95 (C–F) 3D printing (C) FL-ST: (i) Premolar: 130.83 ± 22.81 (ii) Molar: 97.27 ± 11.5 (D) RS-ST: (i) Premolar: 147.82 ± 29.96 (ii) Molar: 107.98 ± 15.53 (E) FL-SH: (i) Premolar: 117.07 ± 13.71 (ii) Molar: 79.48 ± 18.2 (F) RS-SH (i) Premolar: 125.25 ± 19.83 (ii) Molar: 101.91 ± 28.22	Marginal fit ranking (mean of premolar and molar): FL-SH 3D > milling > RS-SH 3D > RS-ST 3D > FL-ST 3D > conventional method Internal fit ranking (mean of premolar and molar): FL-SH 3D > FL-ST 3D > RS-SH 3D > milling > RS-ST 3D > conventional method
Sidhom et al., 2022 [37]	30–60 μm	Milling: five-axis milling machine (CAM 5-S1 impression milling machine software, 3Shape, Copenhagen, Denmark 3D Printer: Formlabs SLA 3D printer (Formlabs Inc., Somerville, MA, USA	Layer thickness: N/M Orientation of printing: 0 degrees	Non-cementation technique	Stereomicroscope (Euromex, Microscope BV, Arnhem, The Netherlands) ×10 magnification	Vertical marginal gap: (A) Milled: (i) Mesial retainer: 34.4 ± 12.4 (ii) Distal retainer: 36.9 ± 17.9 (A) 3D-printed: (i) Mesial retainer: 31.1 ± 4.3 (ii) Distal retainer: 29.9 ± 4.3	NA	Marginal fit ranking: 3D printing > milling
Thakare et al., 2022 [39]	60 micro m	Milling: Arum 5X-200 3D Printer: (B) SLA, Formlabs form 2 (C) DLP, NextDent 5100	N/M	Non-cementation technique	Silicon replica technique and stereomicroscope	Marginal discrepancy (A) Milled: 113.10 (B) 3D-printed SLA: 90.21 (C) 3D-printed DLP: 65.15	Internal discrepancy (A) Milled: (i) Mid-axial wall: 127.73 (ii) Axio-occlusal edge: 185.59 (iii) Center of occlusal region: 291.98 (B) 3D-printed SLA: (i) Mid-axial wall: 100.78 (ii) Axio-occlusal edge: 127.01 (iii) Center of occlusal region: 206.74 (C) 3D-printed DLP: (i) Mid-axial wall: 77.53 (ii) Axio-occlusal edge: 83.73 (iii) Center of occlusal region: 160.76 Overall internal discrepancy (A) Milled: 179.60 (B) 3D-printed SLA: 131.18 (C) 3D-Printed DLP: 96.79	Marginal fit ranking: 3D-printed DLP > 3D-printed SLA > milled Internal fit ranking: 3D-printed DLP > 3D-printed SLA > milled

^##: Data retrieved from plot digitizer app. N/M: not mentioned, NA: not applicable, KE: knife edge, C: chamfer, RS: rounded shoulder, RSB: rounded shoulder with bevel, AMD: absolute marginal discrepancy, DLP, digital light processing, PVS: polyvinyl siloxane, OCT: optical coherence tomography ≈: approximately, FL-SH: Formlabs-SHERA-cb, FL-ST: Formlabs-P Pro Crown and Bridge, RS-SH: Rapid Shape-SHERA-cb, RS-ST: rapid Shape-P Pro Crown and Bridge.

provides detailed information related to the marginal adaptation and internal fit of provisional crowns and FDPs, including cement film thickness, type of 3D printer used, layer thickness and printing orientation, examination method used, mean marginal discrepancy and internal discrepancy, and authors’ suggestions and conclusions.

2.6. Quality Assessment of the Included Studies

Quality assessment of the selected studies was performed using the modified CONSORT scale for in vitro studies [46,47] (

Table 4. Quality analysis results of the included studies.

Item	1	2a	2b	3	4	5	6	7	8	9	10	11	12	13	14
Studies
Park et al., 2016 [14]	Y	Y	Y	Y	Y	Y	N	N	N	N	Y	Y	Y	N	N
Lee et al., 2017 [43]	Y	Y	Y	Y	Y	N	N	N	N	N	Y	Y	Y	Y	N
Mai et al., 2017 [44]	Y	Y	Y	Y	Y	N	N	N	N	N	Y	Y	Y	Y	N
Alharbi et al., 2018 [10]	Y	Y	Y	Y	Y	N	N	N	N	Y	Y	Y	Y	Y	N
Al Deeb et al., 2019 [36]	Y	Y	Y	Y	Y	Y	N	N	N	N	Y	Y	Y	Y	N
Peng et al., 2020 [5]	Y	Y	Y	Y	Y	Y	N	N	N	N	Y	Y	Y	Y	N
Earar et al., 2020 [42]	Y	Y	Y	Y	Y	N	N	N	N	N	Y	Y	N	N	N
Peng et al., 2020 [25]	Y	Y	Y	Y	Y	Y	N	N	N	N	Y	Y	Y	Y	N
Sampaio et al., 2020 [7]	Y	Y	Y	Y	Y	N	N	N	N	N	Y	Y	Y	Y	N
Chaturvedi et al., 2020 [34]	Y	Y	Y	Y	Y	N	N	N	N	N	Y	N	N	Y	N
Wu et al., 2021 [35]	Y	Y	Y	Y	Y	Y	N	N	N	N	Y	Y	Y	Y	N
Mohajeri et al., 2021 [41]	Y	Y	Y	Y	Y	Y	N	N	N	N	Y	Y	Y	N	N
Aldahian et al., 2021 [40]	Y	Y	Y	Y	Y	N	N	N	N	N	Y	Y	Y	Y	Y
Karasan et al., 2022 [38]	Y	Y	Y	Y	Y	N	N	N	N	N	Y	Y	Y	Y	N
Sidhom et al., 2022 [37]	Y	Y	Y	Y	Y	Y	N	N	N	N	Y	Y	Y	Y	Y
Thakare et al., 2022 [39]	Y	Y	N	Y	Y	N	N	N	N	N	Y	N	Y	N	N

Note: Y: Yes, N: No.

). The 14 items included in this scale were as follows: Item 1: Abstract containing structured summary of study design, methodology, results, and conclusions; Item 2a: Introduction should have scientific background and detailed explanation of rationale; Item 2b: Introduction should have study objectives with a defined hypothesis; Item 3: Methodology should contain approach used in the experiment with sufficient details to enable replication; Item 4: Precisely stated primary and secondary outcomes to enable comparison; Item 5: Details of how sample size was determined; Item 6: Details of how random allocation sequence was generated; Item 7: Method used for random allocation concealment; Item 8: Who implemented randomization? Item 9: If randomization is performed, how was blinding followed? Item 10: Statistical assessment; Item 11: Results outcome and estimation; Item 12: Study limitations; Item 13: Details related to funding; Item 14: Details related to the availability of study protocol, if available (Table 4).

2.7. Quantitative Assessment

The data from the studies were compiled in a tabular format, and a meta-analysis was performed using RevMan (Review Manager) Version 5.4.1. (The Cochrane Collaboration, London, UK, 2020) [48]. The 3 types of 3D-printed crowns—methacrylate oligomers, methacrylate-based provisional resins, and hybrid composites—were compared to conventional bis-acrylic, conventional PMMA resin, and CAD/CAM-milled PMMA. The parameters considered were marginal discrepancy and internal discrepancy, which were converted to micrometers, i.e., continuous scale. Thus, an inverse variance was used to calculate the standardized mean difference using a fixed-effects model. Subgroups were constructed based on studies conducted on single crowns or 3-unit FDPs. The forest plots created showed the pooled results for each subgroup as well as the overall pooled standardized mean difference with a 95% confidence interval; p-values were calculated for the effects, and if they were less than 0.05, the null hypothesis was rejected. The heterogeneity was measured using the chi-squared test and I2 statistics and reported for subgroups as well as overall results.

3. Results

3.1. Identification and Screening

The initial electronic search of the selected databases found 519 hits (PubMed: 169; Scopus: 182; Web of Science: 149; Cochrane: 19); 97 titles were found to be duplicates and were removed. Of the remaining 422 articles whose titles and abstracts were reviewed, 385 were rejected because they did not meet the inclusion criteria. The full text of the remaining 37 articles was reviewed based on the inclusion criteria. After a manual search of the references of these articles, one additional article was selected for full-text review. Of these 38 articles, 22 were rejected: eight articles evaluated the internal adaptation and marginal fit of only 3D-printed materials without comparing them with other types of materials; ten discussed only precision and/or trueness; two compared the effects of different curing methods on the internal adaptation/seating of 3D-printed crowns; one evaluated microleakage using different types of cement; and one evaluated the marginal accuracy of only resin-based ceramics with 3D-printed provisional materials. Thus, the final synthesis included 16 articles for qualitative analysis, of which only 10 provided comparative data and were selected for meta-analysis (Figure 1).

3.2. Quality Assessment of the Included Studies

Sixteen in vitro studies met the inclusion criteria and were selected for this systematic review; 62% of the entries were positively reported. Items 1–4 and 10 (abstract, introduction, intervention, outcome, and statistical method, respectively) were reported by all 15 studies. Fourteen studies discussed the limitations (Item 12) and reported the precision of results as confidence intervals (Item 11). Twelve studies mentioned the source of funding (Item 13). Seven studies reported the sample size calculation details (Item 5), two studies gave details related to the accessibility of the full trial protocol (Item 14), and only one study provided details about blinding to avoid operator-based bias (Item 9). None of the studies discussed any details related to randomization (Items 6–8) (Table 4).

3.3. Study Characteristics

All 16 studies included in this review were in vitro studies, and the majority of them (15 out of 16) were published in the last 5 years. Eleven studies analyzed and compared both marginal fit and internal adaptation, three analyzed only internal adaptation, and two studies analyzed only the marginal fit of provisional crowns and FDPs. Relevant data obtained from these 16 studies are discussed under two subheadings according to the aims of this research.

Thirteen studies compared the marginal fit, while fourteen studies compared the internal adaptation of 3D-printed provisional resins with that of other materials (Table 3 and Table 4). Of the 16 included studies, 6 used resin typodont teeth, 6 used dental stone abutment teeth, 3 used implant abutments, and 1 used a metal master model for the evaluation of the studied property. In order to evaluate and measure the discrepancies, seven studies used a micro-CT scanner, seven used the silicone replica technique and a digital microscope, and one each used a stereomicroscope, optical scanner, scanning electron microscope, and electronic digital caliper.

3.4. Results of Studies Analyzing the Marginal Fit

Of the 13 studies comparing the marginal fit of provisional restorations, 9 studies used a single crown, 3 studies used a three-unit FDP (fixed dental prosthesis), and 1 study used both a single crown and a veneer.

3.4.1. Comparing the Marginal Discrepancy Values of PMMA-Based 3D-Printed Provisional Resins

(i)

Comparing the marginal discrepancy values of PMMA-based 3D-printed provisional resins with conventional PMMA-based resins:

(A)

Evaluation performed on single crowns: One study reported lower marginal discrepancies for PMMA-based 3D-printed provisional resins when compared to conventional PMMA-based resins [14].

(B)

Evaluation performed on three-unit FDPs: Al Deeb et al. [36] reported lower marginal discrepancies for PMMA-based 3D-printed provisional resins when compared to conventional PMMA-based resins.

(ii)

Comparing the marginal discrepancy values of PMMA-based 3D-printed provisional resins with CAD/CAM-milled PEEK resins:

(A)

Evaluation performed on single crowns: Higher marginal discrepancies were reported when PMMA-based 3D-printed provisional resins were compared to CAD/CAM-milled PEEK resins [14].

(iii)

Comparing the marginal discrepancy values of PMMA-based 3D-printed provisional resins with conventional dimethacrylate resins:

(A)

Evaluation performed on single crowns: Higher marginal discrepancies were reported when PMMA-based 3D-printed provisional resins were compared to conventional dimethacrylate resins [35].

(iv)

Comparing the marginal discrepancy values of PMMA-based 3D-printed provisional resins with CAD/CAM-milled PMMA resins:

(A)

Evaluation performed on three-unit FDPs: Al Deeb et al. [36] reported higher marginal discrepancies for PMMA-based 3D-printed provisional resins when compared to CAD/CAM-milled PMMA resins.

3.4.2. Comparing the Marginal Discrepancy Values of 3D-Printed Methacrylate Oligomers

(i)

Comparing the marginal discrepancy values of 3D-printed methacrylate oligomers with conventional bis-acrylic provisional resins:

(A)

Evaluation performed on single crowns: Lower marginal discrepancies were reported for 3D-printed methacrylate oligomers when compared to conventional bis-acrylic provisional resins [5,34] (Figure 2). The pooled SMD was −2.89 (−3.17 to −2.60) in favor of 3D-printed methacrylate oligomers and was statistically significant (p < 0.05).

(B)

Evaluation performed on three-unit FDPs: Karasan et al. [38] compared two different brands of 3D-printed methacrylate oligomers and reported lower marginal discrepancies when compared to conventional bis-acrylic provisional resins (Figure 2). The pooled SMD was −1.59 (−2.11 to −1.08) in favor of 3D-printed methacrylate oligomers and was statistically significant (p < 0.05).

The overall SMD was −2.59 (−2.83 to −2.34) in favor of 3D-printed methacrylate oligomers over conventional bis-acrylic resin and was statistically significant (p < 0.05).

(ii)

Comparing the marginal discrepancy values of 3D-printed methacrylate oligomers with CAD/CAM-milled PMMA provisional resins:

(A)

Evaluation performed on single crowns: Lower marginal discrepancies were reported for 3D-printed methacrylate oligomers when compared to CAD/CAM-milled PMMA provisional resins [5,25,34,39,44] (Figure 3). The pooled SMD was −1.23 (−1.44 to −1.01) in favor of 3D-printed methacrylate oligomers and was statistically significant (p < 0.05).

(B)

Evaluation performed on three-unit FDPs: Lower marginal discrepancies were reported for CAD/CAM-milled PMMA provisional resins when compared to two different brands of 3D-printed methacrylate oligomers [38] (Figure 3). The pooled SMD was 0.01 (−0.43 to 0.45) and was statistically insignificant (p > 0.05).

The overall SMD was −0.99 (−1.18 to −0.80) in favor of 3D-printed methacrylate oligomers over CAD/CAM-milled PMMA resin and was statistically significant (p < 0.05).

(iii)

Comparing the marginal discrepancy values of 3D-printed methacrylate oligomers with conventional PMMA resins:

(A)

Evaluation performed on single crowns: Lower marginal discrepancies were reported for 3D-printed methacrylate oligomers when compared to conventional PMMA resin [25,44] (Figure 4). The overall SMD was −0.82 (−1.37 to −0.28) in favor of 3D-printed methacrylate oligomers over conventional PMMA resin and was statistically significant (p < 0.05).

3.4.3. Comparing the Marginal Discrepancy Values of Methacrylate-Based 3D-Printed Provisional Resins

(i)

Comparing the marginal discrepancy values of methacrylate-based 3D-printed provisional resins with conventional PMMA resins:

(A)

Evaluation performed on single crowns: Aldahian et al. [40] reported lower marginal discrepancies, whereas Mohajeri et al. [41] reported higher marginal discrepancies for methacrylate-based 3D-printed provisional resins when compared to conventional PMMA resins (Figure 5). The overall SMD was 0.14 (−1.18 to −0.80) in favor of 3D-printed methacrylate resin over CAD/CAM-milled PMMA resin and was statistically significant (p < 0.05).

(ii)

Comparing the marginal discrepancy values of methacrylate-based 3D-printed provisional resins with CAD/CAM-milled PMMA resins:

(A)

Evaluation performed on single crowns: Aldahian et al. [40] reported lower marginal discrepancies, whereas Mohajeri et al. [41] reported higher marginal discrepancies for methacrylate-based 3D-printed provisional resins when compared to CAD/CAM-milled PMMA resins (Figure 6). The pooled SMD was −0.23 (−0.84 to 0.39) and was statistically insignificant (p > 0.05).

(B)

Evaluation performed on three-unit FDPs: Karasan et al. [38] compared two different brands of 3D-printed methacrylate resins and reported higher marginal discrepancies when compared to CAD/CAM-milled PMMA resins (Figure 6). The pooled SMD was 0.76 (0.30 to 1.22) in favor of CAD/CAM-milled PMMA resins and was statistically significant (p < 0.05).

The overall SMD was 0.41 (0.04 to 0.77) in favor of CAD/CAM-milled PMMA resins over 3D-printed methacrylate resins and was statistically significant (p < 0.05).

(iii)

Comparing the marginal discrepancy values of methacrylate-based 3D-printed provisional resins with conventional bis-acrylic provisional resins:

(A)

Evaluation performed on three-unit FDPs: Karasan et al. [38] compared two different brands of 3D-printed methacrylate resins and reported lower marginal discrepancies when compared to conventional bis-acrylic provisional resins (Figure 7). The overall SMD was −0.80 (−1.25 to −0.34) in favor of 3D-printed methacrylate resins over conventional bis-acrylic resins and was statistically significant (p < 0.05).

3.4.4. Comparing the Marginal Discrepancy Values of 3D-Printed Hybrid-Composite-Based and Isopropyl-Diphenol-Based Provisional Resins

(i)

Comparing the marginal discrepancy values of 3D-printed hybrid-composite-based and isopropyl-diphenol-based provisional resins with CAD/CAM-milled PMMA resins:

(A)

Evaluation performed on single crowns: Alharabi et al. [10] reported lower marginal discrepancies for hybrid-composite-based 3D-printed provisional resins when compared to CAD/CAM-milled PMMA resins.

(B)

Evaluation performed on three-unit FDPs: Sidhom et al. [37] reported lower marginal discrepancies for isopropyl-diphenol-based 3D-printed provisional resins when compared to CAD/CAM-milled PMMA resins.

3.5. Results of Studies Analyzing the Internal Adaptation

Nine out of thirteen studies comparing the internal adaptation used single crowns, three used three-unit FDPs, and one study used both a single crown and a veneer.

3.5.1. Comparing the Internal Discrepancy Values of PMMA-Based 3D-Printed Provisional Resins

(i)

Comparing the internal discrepancy values of PMMA-based 3D-printed provisional resins with conventional PMMA resin:

(A)

Evaluation performed on single crowns: Lower internal discrepancies were reported when PMMA-based 3D-printed provisional resins were compared to conventional PMMA resins [14].

(B)

Evaluation performed on three-unit FDPs: Al Deeb et al. [36] reported lower internal discrepancies for PMMA-based 3D-printed provisional resins when compared to conventional PMMA-based resins.

(ii)

Comparing the internal discrepancy values of PMMA-based 3D-printed provisional resins with conventional dimethacrylate resin:

(A)

Evaluation performed on single crowns: Lower internal discrepancies were reported when PMMA-based 3D-printed provisional resins were compared to conventional dimethacrylate resin [35].

(iii)

Comparing the internal discrepancy values of PMMA-based 3D-printed provisional resins with CAD/CAM-milled PEEK resin:

(A)

Evaluation performed on single crowns: Lower internal discrepancies were reported when PMMA-based 3D-printed provisional resins were compared to CAD/CAM-milled PEEK resin [14].

(iv)

Comparing the internal discrepancy values of PMMA-based 3D-printed provisional resins with CAD/CAM-milled PMMA provisional resins:

(A)

Evaluation performed on single crowns: Lower internal discrepancies were reported when PMMA-based 3D-printed provisional resins were compared to CAD/CAM-milled PMMA provisional resins [36].

(B)

Evaluation performed on three-unit FDPs: Al Deeb et al. reported higher internal discrepancies for PMMA-based 3D-printed provisional resins when compared to CAD/CAM-milled PMMA resin [36].

3.5.2. Comparing the Internal Discrepancy Values of 3D-Printed Methacrylate Oligomers

(i)

Comparing the internal discrepancy values of 3D-printed methacrylate oligomers with conventional bis-acrylic provisional resin:

(A)

Evaluation performed on single crowns: Lower internal discrepancies were reported for 3D-printed methacrylate oligomers when compared to conventional bis-acrylic provisional resins [5,34] (Figure 8). The pooled SMD was 0.52 (−0.65 to −0.38) in favor of 3D-printed methacrylate oligomers and was statistically significant (p < 0.05).

(B)

Evaluation performed on three-unit FDPs: Karasan et al. [38] compared two different brands of 3D-printed methacrylate oligomers and reported lower internal discrepancies when compared to conventional bis-acrylic provisional resins (Figure 8). The pooled SMD was −0.51 (−0.96 to −0.06) in favor of 3D-printed methacrylate oligomers and was statistically significant (p < 0.05).

The overall SMD was −0.52 (−0.64 to −0.39) in favor of 3D-printed methacrylate oligomers over conventional bis-acrylic resin and was statistically significant (p < 0.05).

(ii)

Comparing the internal discrepancy values of 3D-printed methacrylate oligomers with conventional PMMA resin:

(A)

Evaluation performed on single crowns: One study reported lower internal discrepancies for 3D-printed methacrylate oligomers when compared to conventional PMMA resin [25], whereas another study reported higher internal discrepancies [44].

(iii)

Comparing the internal discrepancy values of 3D-printed methacrylate oligomers with CAD/CAM-milled PMMA provisional resins:

(A)

Evaluation performed on single crowns: When 3D-printed methacrylate oligomers were compared to CAD/CAM-milled PMMA provisional resins, three studies reported lower internal discrepancies [34,39,43], one study reported a higher internal discrepancy [5], and another reported the same internal discrepancy for both materials [25] (Figure 9).

(B)

Evaluation performed on three-unit FDPs: Karasan et al. [38] compared two different brands of 3D-printed methacrylate oligomers and reported lower internal discrepancies when compared to CAD/CAM-milled PMMA resins (Figure 9).

The overall SMD was −2.25 (−2.41 to −2.08) in favor of 3D-printed methacrylate oligomers over CAD/CAM-milled PMMA resin and was statistically significant (p < 0.05).

3.5.3. Comparing the Internal Discrepancy Values of Methacrylate-Based 3D-Printed Provisional Resins

(i)

Comparing the internal discrepancy values of methacrylate-based 3D-printed provisional resins with conventional PMMA resins:

(A)

Evaluation performed on single crowns: Aldahian et al. [40] reported lower internal discrepancies for methacrylate-based 3D-printed provisional resins when compared to conventional PMMA resins.

(ii)

Comparing the internal discrepancy values of methacrylate-based 3D-printed provisional resins with CAD/CAM-milled PMMA resins:

(A)

Evaluation performed on single crowns: Aldahian et al. [40] reported lower internal discrepancies for methacrylate-based 3D-printed provisional resins when compared to CAD/CAM-milled PMMA resins.

(B)

Evaluation performed on three-unit FDPs: Karasan et al. [38] compared two different brands of 3D-printed methacrylate resins and reported that one of them had lower internal discrepancies, whereas the other had higher internal discrepancies when compared to CAD/CAM-milled PMMA resins (Figure 10).

The overall SMD was −0.01 (−0.45 to 0.43) between 3D-printed methacrylate resin and CAD/CAM-milled PMMA resin and was statistically insignificant (p > 0.05).

(iii)

Comparing the internal discrepancy values of methacrylate-based 3D-printed provisional resins with conventional bis-acrylic provisional resins:

(A)

Evaluation performed on three-unit FDPs: Karasan et al. [38] compared two different brands of 3D-printed methacrylate resins and reported that both had lower internal discrepancies when compared to conventional bis-acrylic provisional resins (Figure 11). The overall SMD was −0.49 (−0.94 to −0.04) in favor of 3D-printed methacrylate resin over conventional bis-acrylic resin and was statistically significant (p < 0.05).

3.5.4. Comparing the Internal Discrepancy Values of 3D-Printed Hybrid-Composite-Based and Acrylic-Photopolymer-Based Provisional Resins

(i)

Comparing the internal discrepancy values of 3D-printed hybrid-composite-based provisional resins with CAD/CAM-milled PMMA resin:

(A)

Evaluation performed on single crowns: Two studies compared the internal discrepancies of 3D-printed hybrid-composite-based and CAD/CAM-milled PMMA-based provisional resins. One study reported lower [10], while another reported higher [7] internal discrepancies for 3D-printed hybrid composites (Figure 12). The overall SMD was −3.68 (−4.47 to −2.88) in favor of 3D-printed hybrid composites over CAD/CAM-milled PMMA resin and was statistically significant (p < 0.05).

(ii)

Comparing the internal discrepancy values of 3D-printed composite-based provisional resins with conventional PMMA and bis-acrylic provisional resins:

(A)

Sampaio et al. [7] reported higher internal discrepancies for provisional crowns fabricated with 3D-printed composite resins when compared to those fabricated using conventional PMMA and bis-acrylic provisional resins.

(iii)

Comparing the internal discrepancy values of 3D-printed acrylic photopolymer resins with CAD/CAM-milled PMMA resins:

(A)

Lee et al. [43] reported lower internal discrepancies for provisional crowns fabricated with 3D-printed acrylic photopolymer resins when compared to those fabricated using CAD/CAM-milled PMMA resins.

4. Discussion

This systematic review included all of the published English literature comparing the internal adaptation and/or marginal fit of 3D-printed provisional crowns and FDPs with those fabricated using other provisional materials and techniques. To the best of our knowledge, this is the first review of its kind. A total of 16 in vitro studies were included in this research, of which 11 studies analyzed both marginal fit and internal adaptation, 3 studies analyzed only internal adaptation, and 2 studies analyzed only marginal fit of provisional crowns and FDPs. The outcomes of this review suggest that the internal adaptation and marginal fit of provisional restorations are affected by the types of provisional materials used and the techniques of fabrication. Therefore, the tested null hypothesis was rejected.

A well-fabricated provisional prosthesis should have a precise marginal fit in order to minimize the microleakage, protecting the pulp of the prepared tooth and minimizing bacterial accumulation at the margins of the restoration, thus preventing the inflammation of the soft tissues around the tooth and the implant-supported restorations [14,25,26,27,28,29,30,31,32,49]. Good internal adaptation is also vital for provisional restoration to be successful. Structural durability is affected by adaptation at the occlusal surface, whereas maintenance requirements are affected by the axial fit of the provisional crowns and FDPs [14,49].

In general, the majority (i.e., 8 out of 13) of the studies included in this review reported that marginal discrepancies were least with the provisional restorations fabricated by 3D printing when compared to those fabricated by CAD/CAM milling and conventional techniques [10,14,34,37,38,39,40,44]. In two studies, the marginal discrepancy reported for 3D-printed provisional crowns was equal to that of those fabricated by CAD/CAM milling [5,25], whereas in two studies, the 3D-printed provisional crowns displayed the greatest marginal discrepancy [35,41]. Overall, most of the studies reported that marginal discrepancies were highest in provisional restorations fabricated using conventional techniques and materials [10,14,34,36,37,38,39,40,44]. Even though there were differences in marginal discrepancies in provisional restorations fabricated using different materials and techniques, they were within the clinically acceptable limit of less than 120 μm for most of the studies [50,51] (in some studies, a maximum misfit of up to 200 μm was considered to be clinically permitted [36,43,52,53]).

Two studies [10,34] evaluated the effect of the type of finish line on the absolute marginal discrepancy of provisional crowns and reported the least marginal discrepancy with a rounded shoulder with a bevel finish line [10,34] and the highest marginal discrepancy with chamfer [10,34] and knife edge [10] finish lines. The high marginal discrepancy for the chamfer finish line could be due to the topography of the chamfer finish line having a curved axiogingival line angle, which can increase the chances of stair-stepping errors during the incremental layer pattern of buildup in 3D printing [10]. It was reported that the fabrication method of provisional restoration has more effect on the marginal fit as compared to the type of finish line [10,34].

When the internal adaptation of provisional crowns and FDPs was analyzed, 7 out of 13 studies reported minimal internal discrepancies for 3D-printed restorations [10,14,34,38,39,40,43], 4 for CAD/CAM-milled restorations [5,7,36,42], and 2 for conventionally fabricated restorations [35,44]. One study [25] reported similar internal discrepancies for 3D-printed and CAD/CAM-milled provisional crowns and FDPs. The mean occlusal discrepancy was reported to be greater than the mean intermarginal, axiogingival, and axio-occlusal discrepancies [14,34,39,43]. The higher occlusal discrepancy with CAD/CAM-milled provisional could be attributed to the limited size and angle of cutting tools and uneven surfaces on occlusal areas, leading to problems in milling the intaglio surface [23,35,43,44,54].

Conventional provisional resins exhibit high volumetric polymerization shrinkage (higher with PMMA as compared to acrylic-based composite resins) [55,56,57,58]. They also involve manual trimming of excess material and removal during the setting time, leading to distortion [36,40] and, thus, exhibiting poor marginal and internal adaptation when compared to CAD/CAM milling procedures, where the restoration is milled from a dense, pre-polymerized block by an automated machine so that there is no polymerization shrinkage [59,60,61]. When compared to 3D-printed provisional crowns and FDPs, CAD/CAM-milled provisional restorations had poor marginal and internal adaptation. In CAD/CAM milling, the manufacturing process is affected by the size of the milling bur and its range of cutting movement [10,14,43,62,63,64], whereas, in 3D printing, there is an incremental layering process, which reproduces details accurately and compensates for polymerization shrinkage [25,37,44,65]. Few studies have reported better marginal and internal fit for CAD/CAM-milled as compared to 3D-printed provisional restorations. This could be attributed to the different types of 3D printers and different printing settings used in these studies. The accuracy of 3D printed materials is influenced by the type of printer, layer thickness, number of layers, layer intensity, printer wavelength, total thickness, die-spacer thickness, arrangement of crowns, UV intensity, post-processing method, build angle, and number and placement of support structures [16,18,36,38,66,67,68,69,70,71]. Irrespective of the materials and method of fabrication, overall discrepancies were reported to be higher in three-unit FDPs when compared to single crowns. For conventional resins, this could be due to the magnified volumetric shrinkage and deformation in FDPs when compared to crowns due to the geometric shape of the prosthesis [14,36,57].

Two studies [7,35] measured the discrepancies after cementing the provisional restoration with luting cement; seven studies [34,36,37,39,41,42,44] used a non-cementation approach, and two studies [5,25] used both cementation and non-cementation approaches. Studies have reported that the process of placement and cementation of the prosthesis can lead to variations and may cause bias in the outcome [36,37,72]. The method of measuring the fit also affects the outcomes of the studies. Included studies have used different methods to measure the fit of the provisional restoration. These include silicone replica techniques, micro-CT, and cross-section techniques [5,6,7,10,14,25,34,35,42,43,44]. The cross-section technique is reported to cause errors in measurements due to some deformation during cross-sectioning [73] or due to operator dependency, resulting in sectioning in inconsistent planes [74]. In contrast, using micro-CT allows multiple and repeated measurements of critical spaces by producing 2D and 3D images [75,76,77,78].

The strengths of this systematic review are its detailed search strategy and systematic methodology to avoid bias during the selection of the studies. In order to avoid missing out on any relevant studies, all articles analyzing the fit and adaptation of the selected materials were subjected to the selection criteria.

To reduce the variability in the included studies and to make the systematic review more informative, the authors would like to add some suggestions:

• Whenever possible, the researchers should keep measurement units constant.
• Studies should use the same standardized protocol for specimen fabrication and testing (set by ISO or ADA).
• Researchers should provide data in both graphical and tabular form as it is easier to extract data from tables. If this is not possible, then data should be provided as supplementary files, which can be accessed easily [79].

Limitations

The studies included in this systematic review showed high data variability, which could be due to the different types of resins and techniques used in the fabrication of the provisional restorations. Due to this high data variability, most of the meta-analyses included only two studies. Qualitatively, a moderate-to-high level of methodology was employed by most of the included studies, but the risk of bias was high. Recommendations include using standard protocols for in vitro studies and, wherever possible, providing data in tabular forms (along with graphs) while reporting the results so that they can be easily utilized. Additionally, only marginal fit and internal adaptation were evaluated and compared in this systematic review and meta-analysis. Further systematic reviews could be planned to determine the effects of 3D-printing parameters on the properties of 3D-printed resins.

5. Conclusions

Based on our findings, the following conclusions can be drawn:

• Digitally fabricated provisional crowns and FDPs have superior marginal fit and internal adaptation when compared to manually fabricated ones.
• Provisional crowns and FDPs fabricated from 3D-printing resins have a superior marginal fit and internal adaptation when compared to CAD/CAM-milled and conventional provisional resins. Thus, they can be used as a dependable alternative to other resins.
• For all three fabrication techniques, the marginal and internal discrepancy values were within the clinically acceptable ranges.
• Various factors affect the marginal fit and internal adaptation of 3D-printed provisional restorations, including the type of 3D-printing technology, layer thickness, printing orientation, type of provisional resin, etc.
• To improve the quality of future studies, in vitro studies should focus on reducing bias by following the recommended blinding protocols.