Influence of Different Post-Processing Procedures on the Accuracy of 3D Printed Dental Models Using Vat Polymerization: A Systematic Review

Abstract

Introduction: Three-dimensional (3D) printing technology has rapidly evolved across various fields of medicine and dentistry, particularly in orthodontics. One key application in orthodontics is the fabrication of dental models. Numerous parameters throughout the dental cast fabrication workflow can influence the accuracy of 3D-printed models. This review aims to evaluate the influence of post-processing procedures, specifically post-curing and post-rinsing, on the dimensional accuracy of 3D-printed dental casts. Materials and Methods: An initial data search was conducted using specific keywords across four databases (PubMed, Scopus, Web of Science, and Google Scholar). A secondary search of references and citations was also performed. This systematic review ultimately identified five studies that met the inclusion criteria (in vitro studies and studies referred to post-processing only of 3D-printed models) for further evaluation and analysis, whereas reviews, opinion studies, and papers in languages other than English were excluded. Based on the QUIN tool, all studies were assessed for their risk of bias. Because of the studies’ heterogeneity, a qualitative descriptive synthesis was conducted. Results: All five included studies were in vitro investigations. One study examined the influence of the post-curing process on dimensional accuracy, while the remaining four explored the impact of post-rinsing procedures on both dimensional accuracy and other surface characteristics of 3D-printed dental casts. Conclusions: According to the findings of the included studies, both post-curing and post-rinsing procedures had statistically significant effects on the dimensional accuracy of 3D-printed dental models. Nevertheless, all five studies concluded that the observed deviations remained within clinically acceptable limits, rendering the casts suitable for diagnostic orthodontic purposes or device fabrication. However, further research is needed to reinforce current findings and to enhance our understanding of the optimal post-processing protocols of additively manufactured dental casts.

1. Introduction

Recently, advanced technologies such as artificial intelligence and three-dimensional (3D) printing have gained traction in dentistry, particularly in orthodontics [1,2,3,4]. The rapidly growing Computer-Aided Design/Computer-Aided Manufacturing (CAD/CAM) technology has transformed restorative dentistry over the past two decades [5]. There are two main approaches in CAD: subtractive manufacturing (SM), which is based on milling prefabricated blocks or disks, and additive manufacturing (AM) [5,6].

AM, also known as 3D printing, builds objects layer by layer to fabricate three-dimensional structures [7]. Due to its advantages, such as enhanced precision, faster production, customization, and reduced material waste, AM has become increasingly popular in dentistry compared to SM [7,8,9,10,11]. Additive manufacturing (AM) is a fabrication technology that enables the creation of complex structures beyond the limitations of traditional machining or molding techniques. Recent advancements have expanded AM into 4D printing. This technology produces bio-inspired, stimuli-responsive structures capable of self-assembly, self-repair, or adaptive behavior in response to environmental triggers such as temperature, light, or humidity [12]. According to the American Society for Testing and Materials (ASTM) standard classification, seven 3D printing techniques are recognized: material extrusion, powder bed fusion, vat photopolymerization, material jetting, binder jetting, directed energy deposition (DED), and sheet lamination [13]. Material Extrusion (FDM/FFF) uses thermoplastic filaments for prototyping and low-cost production. Powder Bed Fusion (PBF) fuses metal or polymer powders using lasers or electron beams to create functional components. Vat Photopolymerization (SLA/DLP) cures resins for high-precision parts. Material Jetting (MJ) deposits photopolymers or waxes for multi-material prototypes. Binder Jetting (BJ) binds powders with liquid for metal, ceramic, and sand parts. Directed Energy Deposition (DED) deposits and melts metals for repair or feature addition, while Sheet Lamination (SL) bonds sheets of metal, polymer, or paper for rapid prototyping and tooling [14]. Composite materials are fabricated using laser melting deposition (LMD), a type of Directed Energy Deposition (DED) additive manufacturing [15]. Three-dimensional printing’s dental applications include the fabrication of surgical guides for implants, orthodontic and occlusal splints, clear aligners, mouthguards, and prosthetic devices such as crowns, denture bases, and custom impression trays [8,16,17,18,19]. In contrast, one of the earliest applications of 3D printing in orthodontics was the fabrication of dental casts [4,20]. Three-dimensional printing can be used to directly print appliances in orthodontics [21,22].

In 1986, Charles Hull introduced the very first 3D printer [4]. Nowadays, the most used 3D printers in dentistry are those that employ AM with the vat photopolymerization technique [4,23]. The vat photopolymerization process can be divided into two main types: stereolithography (SLA) and digital light processing (DLP). It can also be further classified according to the position of the light source, either bottom–up or top–down [13]. More recently, liquid crystal display (LCD) 3D printers have also gained popularity in dental applications. All these resin-based 3D printers share similar components, including a tank of liquid photopolymer resin, a light source, and a build platform [7,8,9,10,11,24,25,26]. The primary difference lies in the type of light source used to polymerize the resin into solid layers: SLA uses an ultraviolet (UV) laser; DLP uses a digital micromirror device (DMD) to cure an entire layer simultaneously; LCD printing, which closely resembles DLP, uses light-emitting diodes (LEDs) to solidify each resin layer in one pass [4,27].

The AM workflow for vat-polymerization 3D printers consists of three major stages: data processing, printing, and post-processing [28]. The accuracy of the final printed object can be influenced by multiple factors, including the selected vat polymerization technology [4,29], printer calibration, ambient temperature, and light source intensity. Other factors include the support structure design [26], build platform positioning [29,30,31,32,33,34,35,36], resin properties [37], slicer software settings [34], resin color [36], model geometry [31,38,39], printing orientation [30,33,34,35,40,41], printing parameters [41,42,43,44,45,46,47], photoinitiators and especially the post-processing protocols [36,38,39,48].

Two key terms are used to describe 3D printing accuracy: “trueness” and “precision.” Precision refers to the degree of reproducibility when identical objects are printed under the same conditions, while trueness indicates how closely the printed model replicates the original virtual design [4,25,49]. A high trueness printer means that the printer can produce a result close to or equal to the digital 3D object’s actual dimensions [4,49,50,51,52,53,54].

Post-processing begins after the printing process is complete and typically involves removing the printed object from the build platform, cleaning off uncured resin, post-curing, and detaching the support structures [40,55]. In the literature, the terms post-washing and post-rinsing are sometimes used interchangeably. However, post-washing generally refers to the overall cleaning stage after printing, which may involve solvent immersion in combination with other techniques, whereas post-rinsing specifically denotes the solvent-based step used to remove residual uncured resin [56]. After removal from the resin bath, the object is rinsed to eliminate residual unpolymerized resin, thereby achieving a smoother and more stable result. This is followed by a post-curing step, which further enhances the material’s mechanical properties [37]. Determining the appropriate rinsing solution and duration, as well as the optimal post-curing time and temperature, is crucial to ensuring biocompatibility and improving dimensional accuracy [57]. These post-processing stages contribute to the degree of polymer conversion and strengthen the internal molecular network of the printed object [58]. Figure 1 shows a clear and concise flowchart of the basic 3D printing post-processing steps.

Although several studies have evaluated the performance of different resins, printer types, or design parameters, the specific impact of post-processing variables, particularly rinsing and curing protocols, on the dimensional accuracy of 3D-printed dental models remains underexplored [59]. Moreover, existing findings appear heterogeneous, with inconsistent methodologies and limited standardization across studies. Given the growing reliance on 3D-printed models in clinical workflows, a systematic synthesis of available evidence is warranted to clarify the influence of post-processing on model accuracy and guide evidence.

Therefore, this systematic review aims to evaluate the accuracy of post-processed 3D-printed dental models, with a specific focus on the influence of post-rinsing and post-curing procedures.

2. Materials and Methods

2.1. Reporting

This review format follows the Prisma 2020 (Preferred Reporting Items for Systematic reviews and Meta-Analyses) guidelines [60,61]. The protocol for the present systematic review was registered in the Open Science Forum Database following the PRISMA-P guidelines and received the Protocol: 10.17605/OSF.IO/BVU2M.

Eligibility Criteria

Eligibility criteria were established using the PICOS framework (Participants, Intervention, Comparison, Outcomes, Study design, and Method of outcomes), as outlined in Table 1.

2.2. Inclusion and Exclusion Criteria

2.2.1. Inclusion Criteria

For this systematic review, based on the PICO(S) framework [62], the following inclusion criteria were applied:

• Participants: dental models;
• Interventions: post-curing and post-rinsing procedures of 3D-printed dental models;
• Comparisons: comparisons were made between the different rinsing agents and the curing parameters during post-processing;
• Outcome measures: differences in the degree to which each post-processing procedure influences the accuracy of 3D-printed dental casts;
• Study design: randomized and non-randomized in vitro studies.

2.2.2. Exclusion Criteria

Review articles, opinion papers, and studies published in languages other than English were excluded from this review.

Table 1. Eligibility criteria.

Inclusion Criteria	Exclusion Criteria
Studies that refer to the post-processing only of 3D-printed dental models In vitro studies	Review Articles Opinion Papers Studies in languages other than English language

Table 1. Eligibility criteria.

Inclusion Criteria	Exclusion Criteria
Studies that refer to the post-processing only of 3D-printed dental models In vitro studies	Review Articles Opinion Papers Studies in languages other than English language

2.3. A Literature Search Strategy

The following keywords were used in the search strategy for this systematic review: “dental models,” “3D printing,” “post-processing,” “accuracy,” “curing time,” “polymerization time,” “washing time,” and “rinsing time.” The electronic search was conducted across four databases: PubMed, Scopus, Web of Science, and Google Scholar. In addition, a manual (hand) search was also performed. There were no restrictions regarding the year of publication. The specific search strategy adapted for each database is presented in

Table 2. The search strategy for each database.

1. PubMed n = 3
a. ((dental models[MeSH Terms]) AND (3d printing[MeSH Terms])) AND (postprocessing[Title/Abstract]) b. (((dental models[MeSH Terms]) AND (3d printing[MeSH Terms])) AND (accuracy[Title/Abstract])) AND (postprocessing[Title/Abstract])
c. (((“dental models”[MeSH Terms]) AND (“3d printing”[MeSH Terms]))) AND (postprocessing[Title/Abstract] OR “curing time”[Title/Abstract] OR “washing time”[Title/Abstract])
d. (((“dental models”[MeSH Terms]) AND (“3d printing”[MeSH Terms]))) AND (postprocessing[Title/Abstract] OR “polymerization time”[Title/Abstract] OR “rinsing time”[Title/Abstract])
2. Scopus n = 94
a. (ALL (“dental models”) AND ALL (“3d printing”) AND ALL (postprocessing))
b. (ALL (“dental models”) AND ALL (“3d printing”) AND ALL (postprocessing) AND ALL (accuracy)) c. (TITLE-ABS-KEY (“dental models”) AND TITLE-ABS-KEY (“3d printing”) AND TITLE-ABS-KEY (postprocessing OR “washing time” OR “curing time”))
d. (TITLE-ABS-KEY (“dental models”) AND TITLE-ABS-KEY (“3d printing”) AND TITLE-ABS-KEY (postprocessing OR “rinsing time” OR “polymerization time”))
3. Web of Science n = 2
a. “dental models” (All Fields) AND “3d printing” (All Fields) AND “postprocessing” (All Fields) b. “dental models” (All Fields) AND “3d printing” (All Fields) AND “postprocessing” (All Fields) AND “accuracy” (All Fields)
c. “dental models” (All Fields) AND “3d printing” (All Fields) AND “postprocessing” OR “curing time” OR “washing time” (All Fields)
d. “dental models” (All Fields) AND “3d printing” (All Fields) AND “postprocessing” OR “polymerization time” OR “rinsing time” (All Fields)
4. Google Scholar n = 27
a. “dental models AND “3d printing” AND “postprocessing”
b. “dental models” AND “3d printing” AND “postprocessing” AND “accuracy”
c. “dental models” (All Fields) AND “3d printing” (All Fields) AND “postprocessing” OR “curing time” OR “washing time” (All Fields) d. “dental models” AND “3d printing” AND “postprocessing” AND “polymerization time” AND “rinsing time”

.

The initial search started on 24 June 2025. Two authors (A.M. and I.A.T.) screened the results based on titles. Relevant titles were further assessed through abstract screening, and non-eligible studies were excluded. Any disagreements were resolved by consulting a third author (I.L.). After the initial screening, a manual search of the reference lists of the included studies was conducted to identify any additional relevant articles not captured in the database search. Final inclusion was determined after reading the full text of each study and applying the predefined inclusion and exclusion criteria.

2.4. Data Collection, Extraction, and Management

Data were extracted independently and in duplicate by two authors (A.M. and I.A.T.) using a pre-tailored data extraction table. The extracted information included details on model characteristics, applied post-processing procedures (e.g., rinsing and polymerization), comparison groups, outcome measures and their assessment methods, as well as the main findings and conclusions.

2.5. Risk of Bias/Quality Assessment in Individual Studies

The QUIN tool was used to conduct a methodology quality assessment of the included studies. This tool has been suggested by Vidhi H. Sheth et al. [63] as a risk of bias assessment tool for in vitro studies in dentistry and examines twelve bias domains:

• Clearly stated aims/objectives;
• Detailed explanation of sample size calculation;
• Detailed explanation of the sampling technique;
• Details of the comparison group;
• Detailed explanation of methodology;
• Operator details;
• Randomization;
• Method of measurement of outcome;
• Outcome assessor details;
• Blinding;
• Statistical analysis;
• Presentation of results.

The abovementioned 12 criteria are scored as follows:

• Two points for adequately specified criteria;
• One point for inadequately specified criteria;
• Zero points for non-specified criteria;
• Not applicable criteria were excluded from the calculation.

The 12 criteria grades in this study are as follows:

• Low risk of bias (when final score is >70%);
• Medium risk of bias (when final score is 50–70%);
• High risk of bias (when final score is <50%).

The final score is calculated using the formula below:

Final score (%) = (Total score obtained/maximum possible points) × 100 [63].

2.6. Data Synthesis

Due to the substantial heterogeneity among the included studies, a meta-analysis was not conducted. This heterogeneity stemmed from several key differences across the studies. First, the types of post-processing procedures varied considerably, including the use of different rinsing agents, rinsing durations, and polymerization conditions. Additionally, the resin materials employed differed, with some studies investigating water-washable resins and others using non-water-washable alternatives. The methodological approaches to measurement also diverged. Outcomes were assessed through a range of techniques, such as surface roughness, microhardness, dimensional deviations, and measures of trueness and precision. Moreover, the studies reported outcomes in distinct units, including Ra values, Vickers hardness numbers (HV), deviation areas in mm², or linear dimensional measurements, further complicating direct comparisons. Time points of assessment also varied, with some studies evaluating immediate post-processing outcomes and others assessing stability over extended storage periods.

Given these differences, a narrative synthesis was deemed the most appropriate approach. This approach involved categorizing the studies based on the primary variable under investigation (e.g., rinsing solution, duration of rinsing, and the polymerization condition). The findings were then summarized and compared both within and across these categories. Particular attention was given to identifying patterns, contrasting results, and emerging evidence on how post-processing affects the dimensional accuracy and surface characteristics of 3D-printed dental models.

3. Results

3.1. Study Selection

An initial search across four databases (PubMed, Scopus, Web of Science, and Google Scholar) yielded 134 reports. After removing duplicates and non-English records, 115 studies remained. Following a title and abstract screening, excluding reviews and opinion pieces, eight studies were selected for full-text eligibility assessment.

A secondary search was then conducted by reviewing the reference lists of these eight studies. This search yielded 262 additional reports. After applying exclusion criteria and removing duplicates, nine of these were deemed eligible for assessment.

Combining the results from both the initial search and the reference search, while eliminating duplicates, a total of 14 studies were thoroughly reviewed and evaluated by two independent reviewers. Ultimately, five studies were selected for inclusion in this systematic review. The study selection process is illustrated in the PRISMA flow diagram (Figure 2).

3.2. Characteristics of Included Studies

All five of the final selected papers were in vitro studies. One study investigated the impact of post-curing on the dimensional accuracy of 3D-printed dental models, specifically examining polymerization parameters including post-curing time and conditions [59]. Three studies investigated the impact of post-rinsing on the dimensional accuracy of 3D-printed dental models. Cleaning solvents, resin types, and rinsing durations are some of the variables considered [27,64,65]. The final study evaluated the impact of post-rinsing on other characteristics of 3D-printed dental models, specifically surface roughness and microhardness [66]. Table 3 presents a summary of the data extracted from these five studies, while Table 4 provides an overview of the trueness and precision characteristics of each study.

3.3. Risk of Bias Within Studies

The risk of bias assessment is presented in Table 5. Three studies investigating post-rinsing methods [64,65,66], as well as the study evaluating the influence of post-curing on the dimensional accuracy of dental casts [59], were generally rated as low risk of bias. Only the report by Lammers et al. [27] was assessed as having a medium risk of bias based on the final score assigned using the QUIN tool. A common strength across all studies was the transparent reporting of outcomes, whereas the absence of a registered study protocol introduces a potential risk of unmeasured bias.

Figure 2. PRISMA flow diagram.

Figure 2. PRISMA flow diagram.

Table 3. Data extraction.

Authors/ Publication Year	Study Design	Participants (Number of Dental Models)	Intervention	Outcomes	Method of Outcome Assessment	Results	Conclusion
D. Mostafavi [59] (2023)	In vitro study	160 specimens (divided in two groups, N = 80. Each group was divided in 4 subgroups of 20 specimens each)	1. Post polymerization conditions (dry or water submerged) 2. Post polymerization time (2, 10, 20, 40 min)	Dimensional accuracy of 3d printed dental models	- Dimensional measurement (Length, Width, Height) by a low force digital caliper - Volume calculation using formula V = L × w × h - Data analysis: Kruskal-Wallis- Mann and Whitney U tests	- Dry post-polymerization had higher dimensional accuracy compared to water submerged - D2, D4 subgroups had the smallest variance	- None of the workflows offered a perfect match - Post curing time and conditions affect the accuracy - Highest dimensional accuracy on dry post curing time of 10 and 40 min
CA Lammer [65] (2025)	In vitro study	72 maxillary dental casts (divided in 3 groups based on resin type. Each resin group divided in 3 subgroups based on cleaning solutions)	1. Resin type: a. WW (EPAX WATER- WW1, EPAX 3D -WW2) b. NWW (keymodel Ultra resin- ivory) 2. Cleaning solution (Water, MES, IPA) 3. Storage duration	Dimensional accuracy of 3d printed dental models	- Post fabricated dental casts were divided and stored for 5 different periods (T0–T4) - Digitization of casts after storing using a laboratory scanner - Evaluation of casts’ dimensional stability with analysis software program	- Entire cast deviations (p < 0.001) by resin type, cleaning solution, storage duration - NWW casts: the lowest deviations - WW1 casts: the highest deviations	- NWW: the highest dimensional stability - MES use increases WW casts’ dimensional stability - Despite deviations, all casts were acceptable
G Çakmak [66] (2025)	In vitro study	3 dental models (36 specimens from each model)	1. Resin type: a. WW (EPAX - WW1, Phrozen-WW2) b. NWW (keymodel Ultra resin- beige) 2. Cleaning solution (Water, MES, IPA)	Surface roughness and microhardness	- Profilometer used for surface roughness measurement - Vickers hardness tester used for microhardness measurement - Images tested in laser digital microscope	- Surface roughness: 1. IPA: lower Ra Values 2. NWW resin: lower Ra Values - Microhardness under microscope 1. WW2 with IPA: the highest HV 2. NWW regardless solution used: the lowest HV	- NWW resin had smoother surface with IPA or MES - Overall, higher microhardness obtained with IPA use - With clinically comparable surface roughness and microhardness, WW2 resin cleaned with water or MES is more ecologically friendly
Yoojin lim [27] (2022)	In vitro study	46 dental models (one master model and 3 groups of 15)	IPA alternative rinsing solvents: 1. Mean Green 2. Yellow Magic 77 3. Propylene glycol	Dimensional accuracy of 3d printed dental models	- Frasaco maxillary typodont scanning - One master model and 45 models fabrication - 3D comparison of master model and samples with Geomagic Control X software	- Mean Green: highest average error (0.0015 to 0.0141 mm2) - Yellow Magic 7: average error 0.0056–0.001 mm2) - PG: Mean error ranged from 0.0009 to 0.0137 mm2	- PG: the most effective residual resin removal and the lowest average errors - Mean Green: the least effective residual resin removal and the highest average errors - No statistical difference between M. Green-Y. Magic 7
D Mostafavi [64] (2021)	In vitro study	160 printed specimens (80 per each solvent group)	Resin type (E-model Light; Envisiontec, Deadborn, MI) Interventions: 1. Different solvents (IPA, TPM) 2. Rinsing Time (5, 7, 9, 11 min)	Dimensional accuracy of 3d printed dental models	- Dimensional measurement with a low force digital caliper - Volume calculation using formula V = L × w × h - Statistical analysis: Kruskal-Wallis- Mann and Mann- Whitney U tests	- IPA groups had lower trueness and precision with the same rinsing protocol (p < 0.0018) - TPM1-TPM2 subgroups: Highest trueness and precision values (p > 0.0018)	- None of the workflows ended to a perfect match - TPM solvent: higher trueness and precision - Rinsing ultrasonic bath with TPM solvent for 3–4 min, followed by another 2–3 min bath: the highest accuracy values

Table 3. Data extraction.

Authors/ Publication Year	Study Design	Participants (Number of Dental Models)	Intervention	Outcomes	Method of Outcome Assessment	Results	Conclusion
D. Mostafavi [59] (2023)	In vitro study	160 specimens (divided in two groups, N = 80. Each group was divided in 4 subgroups of 20 specimens each)	1. Post polymerization conditions (dry or water submerged) 2. Post polymerization time (2, 10, 20, 40 min)	Dimensional accuracy of 3d printed dental models	- Dimensional measurement (Length, Width, Height) by a low force digital caliper - Volume calculation using formula V = L × w × h - Data analysis: Kruskal-Wallis- Mann and Whitney U tests	- Dry post-polymerization had higher dimensional accuracy compared to water submerged - D2, D4 subgroups had the smallest variance	- None of the workflows offered a perfect match - Post curing time and conditions affect the accuracy - Highest dimensional accuracy on dry post curing time of 10 and 40 min
CA Lammer [65] (2025)	In vitro study	72 maxillary dental casts (divided in 3 groups based on resin type. Each resin group divided in 3 subgroups based on cleaning solutions)	1. Resin type: a. WW (EPAX WATER- WW1, EPAX 3D -WW2) b. NWW (keymodel Ultra resin- ivory) 2. Cleaning solution (Water, MES, IPA) 3. Storage duration	Dimensional accuracy of 3d printed dental models	- Post fabricated dental casts were divided and stored for 5 different periods (T0–T4) - Digitization of casts after storing using a laboratory scanner - Evaluation of casts’ dimensional stability with analysis software program	- Entire cast deviations (p < 0.001) by resin type, cleaning solution, storage duration - NWW casts: the lowest deviations - WW1 casts: the highest deviations	- NWW: the highest dimensional stability - MES use increases WW casts’ dimensional stability - Despite deviations, all casts were acceptable
G Çakmak [66] (2025)	In vitro study	3 dental models (36 specimens from each model)	1. Resin type: a. WW (EPAX - WW1, Phrozen-WW2) b. NWW (keymodel Ultra resin- beige) 2. Cleaning solution (Water, MES, IPA)	Surface roughness and microhardness	- Profilometer used for surface roughness measurement - Vickers hardness tester used for microhardness measurement - Images tested in laser digital microscope	- Surface roughness: 1. IPA: lower Ra Values 2. NWW resin: lower Ra Values - Microhardness under microscope 1. WW2 with IPA: the highest HV 2. NWW regardless solution used: the lowest HV	- NWW resin had smoother surface with IPA or MES - Overall, higher microhardness obtained with IPA use - With clinically comparable surface roughness and microhardness, WW2 resin cleaned with water or MES is more ecologically friendly
Yoojin lim [27] (2022)	In vitro study	46 dental models (one master model and 3 groups of 15)	IPA alternative rinsing solvents: 1. Mean Green 2. Yellow Magic 77 3. Propylene glycol	Dimensional accuracy of 3d printed dental models	- Frasaco maxillary typodont scanning - One master model and 45 models fabrication - 3D comparison of master model and samples with Geomagic Control X software	- Mean Green: highest average error (0.0015 to 0.0141 mm2) - Yellow Magic 7: average error 0.0056–0.001 mm2) - PG: Mean error ranged from 0.0009 to 0.0137 mm2	- PG: the most effective residual resin removal and the lowest average errors - Mean Green: the least effective residual resin removal and the highest average errors - No statistical difference between M. Green-Y. Magic 7
D Mostafavi [64] (2021)	In vitro study	160 printed specimens (80 per each solvent group)	Resin type (E-model Light; Envisiontec, Deadborn, MI) Interventions: 1. Different solvents (IPA, TPM) 2. Rinsing Time (5, 7, 9, 11 min)	Dimensional accuracy of 3d printed dental models	- Dimensional measurement with a low force digital caliper - Volume calculation using formula V = L × w × h - Statistical analysis: Kruskal-Wallis- Mann and Mann- Whitney U tests	- IPA groups had lower trueness and precision with the same rinsing protocol (p < 0.0018) - TPM1-TPM2 subgroups: Highest trueness and precision values (p > 0.0018)	- None of the workflows ended to a perfect match - TPM solvent: higher trueness and precision - Rinsing ultrasonic bath with TPM solvent for 3–4 min, followed by another 2–3 min bath: the highest accuracy values

Table 4. Description of trueness and precision characteristics across the studies.

Authors/ Publication Year	Trueness and Precision Overview
Trueness: Assessment of dimensional deviations relative to a reference model Precision: Examination of variability in dimensional deviations across repeated prints
D. Mostafavi [59] (2023)	Assessment of dimensional accuracy across different post-curing protocols
	Results concerning trueness: Post-curing protocol, insufficient curing, over-curing or inappropriate curing conditions resulted in dimensional deviations in 3d printed models compared to the reference model (lower trueness).	Results concerning precision: Inconsistent post-curing procedure or suboptimal curing parameters led to higher variability between prints (lower precision).
CA Lammer [65] (2025)	Assessment of dimensional accuracy across different resin types, rinsing solutions and storage durations
	Results concerning trueness: NWW resin with IPA or MES had the lowest deviations (higher trueness).	Results concerning precision: Concerning reproducibility among repeated prints, dimensional deviations were within clinically acceptable thresholds (high precision even if trueness varied).
G Çakmak [66] (2025)	Assessment of surface properties (roughness and hardness) accuracy across different resin types and rinsing solutions
	Results concerning trueness: WW2-water and WW2-MES achieved the lowest deviations in both surface roughness and hardness (highest trueness).	Results concerning precision: Relatively low deviations across groups regardless of resin or rinsing solution (good precision in surface properties). Less stable solvents or inconsistent rinsing techniques resulted in variability between prints (low precision).
Yoojin lim [27] (2022)	Assessment of dimensional accuracy across different rinsing solvents
	Results concerning trueness: Different rinsing agents led to variations in the dimensional accuracy of the models affecting the printed model trueness. Propylene glycol had the lowest average errors (highest trueness).	Results concerning precision: Certain rinsing agents contributed to more consistent dimensions across prints, enhancing the precision of the manufacturing process.
D Mostafavi [64] (2021)	Assessment of dimensional accuracy across different rinsing solutions and times
	Results concerning trueness: The rinsing solution significantly affected trueness. Isopropyl alcohol (IPA) led to lower dimensional deviations (higher trueness).	Results concerning precision: Variability of deviations among repeated specimens across rinsing conditions was relatively low (high precision).

Table 4. Description of trueness and precision characteristics across the studies.

Authors/ Publication Year	Trueness and Precision Overview
Trueness: Assessment of dimensional deviations relative to a reference model Precision: Examination of variability in dimensional deviations across repeated prints
D. Mostafavi [59] (2023)	Assessment of dimensional accuracy across different post-curing protocols
	Results concerning trueness: Post-curing protocol, insufficient curing, over-curing or inappropriate curing conditions resulted in dimensional deviations in 3d printed models compared to the reference model (lower trueness).	Results concerning precision: Inconsistent post-curing procedure or suboptimal curing parameters led to higher variability between prints (lower precision).
CA Lammer [65] (2025)	Assessment of dimensional accuracy across different resin types, rinsing solutions and storage durations
	Results concerning trueness: NWW resin with IPA or MES had the lowest deviations (higher trueness).	Results concerning precision: Concerning reproducibility among repeated prints, dimensional deviations were within clinically acceptable thresholds (high precision even if trueness varied).
G Çakmak [66] (2025)	Assessment of surface properties (roughness and hardness) accuracy across different resin types and rinsing solutions
	Results concerning trueness: WW2-water and WW2-MES achieved the lowest deviations in both surface roughness and hardness (highest trueness).	Results concerning precision: Relatively low deviations across groups regardless of resin or rinsing solution (good precision in surface properties). Less stable solvents or inconsistent rinsing techniques resulted in variability between prints (low precision).
Yoojin lim [27] (2022)	Assessment of dimensional accuracy across different rinsing solvents
	Results concerning trueness: Different rinsing agents led to variations in the dimensional accuracy of the models affecting the printed model trueness. Propylene glycol had the lowest average errors (highest trueness).	Results concerning precision: Certain rinsing agents contributed to more consistent dimensions across prints, enhancing the precision of the manufacturing process.
D Mostafavi [64] (2021)	Assessment of dimensional accuracy across different rinsing solutions and times
	Results concerning trueness: The rinsing solution significantly affected trueness. Isopropyl alcohol (IPA) led to lower dimensional deviations (higher trueness).	Results concerning precision: Variability of deviations among repeated specimens across rinsing conditions was relatively low (high precision).

Table 5. Risk of bias assessment.

Author (Year)	Outcomes	Clearly Stated Objectives	Details of Sample Size Calculation	Details of Sampling Technique	Details of Comparison Group	Detailed Methodology Explanation	Operator Details	Randomization	Method of Outcome Measurement	Outcome Assessor Details	Blinding	Statistical Analysis	Presentation of results	Final Score	Final Score (%)	Risk of Bias
D. Mostafavi [59] (2023)	Dimensional accuracy of 3d printed dental models	2	2	2	2	2	0	1	2	0	1	2	2	18	75%	Low
CA Lammer [65] (2025)	Dimensional stability of 3d printed dental models	2	1	2	2	2	1	1	2	0	1	2	2	17	70.8%	Low
G Çakmak [66] (2025)	Surface roughness and microhardness of 3d printed dental models	2	2	2	2	2	0	1	2	0	1	2	2	17	70.8%	Low
Yoojin lim [27] (2022)	Dimensional accuracy of 3d printed dental models	2	1	2	2	2	0	1	2	0	1	2	2	16	66.7%	Medium
D Mostafavi [64] (2021)	Dimensional accuracy of 3d printed dental models	2	1	2	2	2	1	1	2	0	1	2	2	18	75%	Low

Table 5. Risk of bias assessment.

Author (Year)	Outcomes	Clearly Stated Objectives	Details of Sample Size Calculation	Details of Sampling Technique	Details of Comparison Group	Detailed Methodology Explanation	Operator Details	Randomization	Method of Outcome Measurement	Outcome Assessor Details	Blinding	Statistical Analysis	Presentation of results	Final Score	Final Score (%)	Risk of Bias
D. Mostafavi [59] (2023)	Dimensional accuracy of 3d printed dental models	2	2	2	2	2	0	1	2	0	1	2	2	18	75%	Low
CA Lammer [65] (2025)	Dimensional stability of 3d printed dental models	2	1	2	2	2	1	1	2	0	1	2	2	17	70.8%	Low
G Çakmak [66] (2025)	Surface roughness and microhardness of 3d printed dental models	2	2	2	2	2	0	1	2	0	1	2	2	17	70.8%	Low
Yoojin lim [27] (2022)	Dimensional accuracy of 3d printed dental models	2	1	2	2	2	0	1	2	0	1	2	2	16	66.7%	Medium
D Mostafavi [64] (2021)	Dimensional accuracy of 3d printed dental models	2	1	2	2	2	1	1	2	0	1	2	2	18	75%	Low

4. Discussion

The continuous advancement of digital technology is significantly transforming the field of orthodontics. As a result, the new generation of orthodontists is increasingly integrating these technological innovations into their clinical workflows [67,68,69,70]. Among the various applications of 3D printing in orthodontics, dental casts were among the first to be implemented, and later, direct printed applications also emerged [4,20,21,22,23,68]. The dimensional accuracy of dental models can be affected both during and after the printing process, particularly through post-processing procedures.

This systematic review aimed to evaluate the dimensional accuracy of dental models following post-curing and post-rinsing procedures, based on a comparative assessment of five different in vitro studies.

Regarding post-rinsing procedures, it is essential for optimizing surface properties, dimensional accuracy, mechanical performance, and biocompatibility of 3D-printed dental resins [71,72,73]. Removal of residual uncured resin, typically achieved through immersion in an organic solvent, ensures accurate final dimensions and reduces cytotoxicity [72,74,75]. Key variables include rinsing duration, cleaning method (e.g., ultrasonic or centrifugal), and the choice of solvent [74,76]. Isopropyl alcohol (IPA) is widely regarded as the gold standard, particularly in clinical applications, though it is highly volatile, flammable, and may cause irritation or dizziness. Ethanol, with comparable risks, is also commonly used [35].

Three out of five studies included in the present review [27,64,65] have investigated variables such as resin type, rinsing time, and rinsing solutions to assess the dimensional accuracy of 3D-printed dental models. In contrast, G. Çakmak et al. [66] focused on changes in surface roughness and microhardness of printed dental models following different washing protocols.

Mostafavi et al. [64] compared the influence on dimensional accuracy of IPA and tripropylene glycol monomethyl ether (TPM) as rinsing agents, along with varying durations of rinsing. While IPA remains the most commonly used rinsing agent for removing residual resin, TPM demonstrated superior trueness and precision compared to IPA groups. The highest accuracy was achieved using TPM for 3–4 min, followed by an additional 2–3 min rinsing. Both insufficient and excessive rinsing times were associated with increased dimensional deviations. At the same time, all deviations were within the 100–300 μm clinical tolerance [77,78,79,80,81]. The findings suggest that some deviations may not be suitable for working casts intended for definitive restorations. The authors highlighted the need for further research due to the study’s limitations, including the use of only one resin, limited rinsing durations, and a lack of variability in the tested factors. Further research will help better understand the post-rinsing methods and their impact on dental cast accuracy [64].

Y. Lim et al. [27] evaluated three FDA-approved alternatives to IPA, Mean Green, Yellow Magic 7, and propylene glycol (PG), for cleaning LCD-printed dental models. PG demonstrated the lowest mean error, indicating superior dimensional accuracy among the tested agents. Propylene glycol demonstrated the lowest mean error and, therefore, the highest dimensional accuracy among the tested agents. Despite reported limitations (small sample size, absence of an ultrasonic bath, and use of a single LCD printer), all printed models exhibited discrepancies within the clinically acceptable range of 100–300 μm. Such results make them suitable for diagnostic purposes. The authors concluded that PG may be a viable alternative to IPA. However, this study highlights the need for further research and the development of standardized protocols [27].

C.A. Lammer et al. [65] conducted an innovative study assessing the dimensional stability of additively manufactured dental models stored over three months. Their protocol combined different resins with various cleaning solutions, revealing that non-water-washable (NWW) resins exhibited the best dimensional stability. Among the water-washable materials, Phrozen resin outperformed EPAX resin across most time points and conditions. Among the rinsing agents, methyl ether solvent (MES) resulted in the lowest dimensional deviations compared to water and isopropyl alcohol (IPA), with water alone proving to be the least effective [64]. Despite statistically significant differences, all groups remained within clinically acceptable thresholds 200 μm [82]—250 μm [83]. Post-processed dental models, according to the study’s workflow protocol, were clinically accepted for diagnostic and orthodontic applications, with a storage period of up to three months. The authors emphasized the potential of alcohol-free solvents, such as MES, as an eco-friendly alternative to IPA. A limitation reported in this study was the use of different polymerization units, which may have introduced variability [65].

The final study included in this review, by G. Çakmak et al. [66], explored the influence of rinsing agents (water, IPA, and MES) on surface roughness and microhardness in both water-washable and non-water-washable resins. Their findings revealed significant differences in surface properties, depending on the resin and rinsing agent. The only exception was that surface roughness in water-washable groups remained unaffected by solvent type [66]. Notably, IPA was described as highly volatile, requiring safety precautions, despite being the most commonly used post-processing solvent [82]. WW2 (Phrozen) resin, when cleaned with MES or water, showed comparable surface characteristics to NWW resins, which were cleaned with IPA. The authors support the use of alcohol-free agents, such as MES or water, as sustainable alternatives. This study reports a limitation in its inclusion of only one NWW resin, and the authors recommend additional research into various resin types and their mechanical properties [66].

Post-polymerization is a crucial step in producing high-quality 3D-printed dental appliances, which is performed after the initial printing and rinsing. It enhances mechanical properties, biocompatibility, and overall print quality by increasing the degree of conversion (DC), the extent to which resin monomers are converted into a solid polymer [84,85]. Higher DC values reduce residual monomers and improve mechanical performance [46,48,55,86,87]. These ovens achieve this through UV light exposure, controlled heating, and adjustable settings, including turntables, curing time, temperature, and UV intensity. These settings allow for the optimization of different resins and desired outcomes [55,88].

Variations in post-curing equipment, such as UV ovens, wavelength, irradiance, and curing protocols, can significantly influence the degree of polymerization, including the dimensional stability of 3D-printed models. Models processed under different conditions (differences in UV intensity, exposure time, and curing conditions (e.g., dry versus submerged environments)) may undergo varying degrees of shrinkage or expansion. This inconsistency makes it difficult to directly compare outcomes across studies, as differences in measured dimensional accuracy may reflect variations in equipment or protocol rather than inherent material performance. To improve reproducibility and maintain consistent dimensional accuracy, standardization of post-curing equipment and protocols is essential. Clear guidelines on curing devices, UV parameters, and exposure durations would help reduce variability and enable more reliable interpretation of 3D printing outcomes in dental research [89,90,91].

D. Mostafavi et al. [59] found that varying curing times and conditions significantly affect dimensional accuracy. More specifically, dry post-curing conditions yielded greater dimensional accuracy in the dental casts compared to water-submerged conditions. Additionally, post-curing times of 10 and 40 min under dry conditions produced better dimensional accuracy than curing for 2 or 20 min [59]. The clinically acceptable dimensional accuracy for 3D-printed dental models ranges from 100 to 300 μm [77,78,79,80,81]. In terms of clinical implications, the post-curing dimensional deformation among the groups ranged from 10 to 90 μm. While such deviations are generally acceptable for diagnostic casts, they may not meet the precision requirements of more demanding restorative procedures. As mentioned in this study, limitations such as the use of only one UV-curing device, a limited number of tested polymerization times, manual measurement methods, and the number of dental models tested, the author emphasizes the need for further research to better understand the impact of the post-curing procedure on the accuracy of 3D-printed dental casts [59]. Table 6 presents a summary of the collected data and key findings from the five included studies.

All five studies demonstrate that post-curing and post-rinsing conditions have a measurable effect on the dimensional accuracy of 3D-printed dental models. While the observed deviations fall within clinically acceptable thresholds (100–300 μm), the selected studies suggest that 3D-printed dental casts are suitable for diagnostic purposes but may not be ideal for high-precision applications, such as definitive prosthetics. It is worth noting, however, that this hypothesis warrants further investigation. Methodological variations among the studies, as already mentioned, included differences in resin types, solvents, polymerization protocols, measurement methods, and sample sizes. Although these differences exist, they are clearly acknowledged by all authors, which adds credibility to the findings. A more thorough review of the existing literature reveals a strong correlation between resin type and dimensional stability. Lammer et al. [65] reported that non-water-washable resin materials provide the highest dimensional stability in 3D-printed dental models. Similarly, an in vitro study by Vanessa Knode et al. [92] confirmed that the choice of resin material significantly affects the dimensional stability of 3D-printed models. Furthermore, Long Ling et al. [93] compared the shrinkage, accuracy, and dimensional stability of a novel 3D-printing model resin with eight commercially available resins, demonstrating that certain materials exhibit superior stability. Regarding rinsing solutions, it is essential to note that while alternatives to isopropyl alcohol (IPA), such as MES and PG, show promise, their long-term effects on mechanical and biological properties have not been thoroughly investigated in the available studies.

Across the five studies, both trueness and precision of 3D-printed dental models were affected by post-processing parameters. Among these, post-curing conditions had the most significant impact. Variations in post-curing conditions, particularly curing time and environment, resulted in dimensional deviations ranging from 10 to 90 µm, underscoring their critical role in achieving high-precision outcomes. Rinsing protocols, solvent choice, and resin type also influenced precision and surface quality, with IPA, propylene glycol, and MES performing best, depending on the resin type. Non-water-washable resins consistently showed the lowest deviations, while water-washable resins benefited from appropriate post-rinsing. Most deviations observed in the included studies were within clinically acceptable ranges (100–300 µm) for diagnostic and orthodontic applications. While both post-processing steps play an essential role, post-curing protocols demonstrated the most significant influence on dimensional accuracy. Their optimization is, therefore, particularly critical when precision requirements extend to definitive restorative or orthodontic procedures.

This systematic review is limited by the heterogeneity of the included studies, which varied in their experimental designs, post-processing protocols, and measured outcomes. The absence of registered protocols in the included studies introduces a potential risk of unmeasured bias that reduces confidence in the generalizability of the results. Overall, the certainty of evidence on how post-processing affects the dimensional accuracy of 3D-printed dental models can be considered moderate. Each included study reported well-controlled conditions with clearly defined interventions and consistent specimen management. However, the available literature on this specific topic remains limited, making it challenging to identify studies that met the inclusion criteria. Finally, a further limitation of this review is the exclusion of studies published in languages other than English. Nevertheless, a common conclusion across all studies is the urgent need for further research to optimize and standardize post-processing procedures for additively manufactured dental models. Future research should examine a broader range of resin types and rinsing solvents, test different curing and rinsing durations, and evaluate additional model characteristics beyond dimensional accuracy.

Table 6. Summary of the collected data, findings, and limitations from the five selected studies.

Author/Publication Year	Outcomes	Variables Tested	Main Findings	Limitations
Mostafavi et al. (2021) [58]	Effect of rinsing agents and duration on dimensional accuracy	IPA vs TPM; rinsing time	TPM showed superior trueness & precision compared to IPA; optimal rinsing = 3–4 min + extra 2–3 min; all deviations within 100–300 µm	Only one resin tested; limited rinsing times; lack of variable control
Lim (2022) [26]	Alternative rinsing solutions for LCD models affecting their dimensional accuracy	IPA vs Mean Green, Yellow Magic 7, PG	PG showed lowest mean error and highest dimensional accuracy; all deviations within 100–300 µm (diagnostic use acceptable)	Small sample; no ultrasonic bath; single printer; limited scope
Lammer et al. (2025) [64]	Dimensional stability over 3-month storage	NWW vs WW resins; solvents (MES, IPA, water)	NWW resins most stable; Phrozen WW > EPAX WW; MES showed lowest deviations; water least effective; all results within 200–250 µm	Different polymerization units used; potential variability
Çakmak et al. (2025) [65]	Surface roughness & microhardness	Resins (WW vs NWW); solvents (IPA, MES, water)	Resin and solvent both influenced results; exception: surface roughness in WW resins not affected by solvent; MES & water provided sustainable alternatives	Only one NWW resin tested; limited scope on mechanical properties
Mostafavi (2025) [63]	Post-curing conditions on dimensional accuracy	Dry vs submerged curing; curing time (2–40 min)	Dry curing gave higher accuracy; best times: 10 and 40 min; deviations 10–90 µm (acceptable for diagnostics, borderline for restorations)	One UV device only; limited curing times; manual measurements; small sample size

Table 6. Summary of the collected data, findings, and limitations from the five selected studies.

Author/Publication Year	Outcomes	Variables Tested	Main Findings	Limitations
Mostafavi et al. (2021) [58]	Effect of rinsing agents and duration on dimensional accuracy	IPA vs TPM; rinsing time	TPM showed superior trueness & precision compared to IPA; optimal rinsing = 3–4 min + extra 2–3 min; all deviations within 100–300 µm	Only one resin tested; limited rinsing times; lack of variable control
Lim (2022) [26]	Alternative rinsing solutions for LCD models affecting their dimensional accuracy	IPA vs Mean Green, Yellow Magic 7, PG	PG showed lowest mean error and highest dimensional accuracy; all deviations within 100–300 µm (diagnostic use acceptable)	Small sample; no ultrasonic bath; single printer; limited scope
Lammer et al. (2025) [64]	Dimensional stability over 3-month storage	NWW vs WW resins; solvents (MES, IPA, water)	NWW resins most stable; Phrozen WW > EPAX WW; MES showed lowest deviations; water least effective; all results within 200–250 µm	Different polymerization units used; potential variability
Çakmak et al. (2025) [65]	Surface roughness & microhardness	Resins (WW vs NWW); solvents (IPA, MES, water)	Resin and solvent both influenced results; exception: surface roughness in WW resins not affected by solvent; MES & water provided sustainable alternatives	Only one NWW resin tested; limited scope on mechanical properties
Mostafavi (2025) [63]	Post-curing conditions on dimensional accuracy	Dry vs submerged curing; curing time (2–40 min)	Dry curing gave higher accuracy; best times: 10 and 40 min; deviations 10–90 µm (acceptable for diagnostics, borderline for restorations)	One UV device only; limited curing times; manual measurements; small sample size

5. Conclusions

Numerous variables can influence the accuracy of additively manufactured objects across the different stages of the 3D printing workflow (namely, before, during, and after printing). The included studies showed that both post-processing procedures significantly affect the dimensional accuracy of 3D-printed dental models. Despite these effects, the deviations consistently remained within clinically acceptable thresholds. This suggests that variations, while statistically significant, are unlikely to compromise clinical outcomes. The models can still be reliably used for diagnosis or for fabricating orthodontic devices, even after storage for up to three months.

Recent research has increasingly focused on identifying more ecologically sustainable post-processing methods, particularly regarding rinsing solvents. There is growing support for the use of non-alcohol-based cleaning agents as alternatives to isopropyl alcohol (IPA), which has traditionally been the most commonly used solvent.

Given the wide range of resin materials, curing devices, and durations, solvent types, and rinsing times, further investigation is required to determine the optimal combinations of post-processing conditions that yield the highest trueness and precision in 3D-printed dental models.