Towards standardizing cytotoxicity testing of 3D-printed orthodontic aligners and retainers: a scoping review

Abstract

Background

Three-dimensional (3D) printing enables customized orthodontic aligners and retainers, but concerns persist regarding the biocompatibility of printable resins. Although ISO 10993-5 and ISO 10993-12 provide guidance for in vitro cytotoxicity testing and sample extraction, substantial flexibility in reporting and execution may hinder comparability across studies.

Objective

To map how in vitro cytotoxicity testing is performed for materials used to 3D-print orthodontic aligners/retainers and to identify variation and reporting gaps in key ISO 10993-5/-12 testing domains.

Methods

Following PRISMA-ScR, Ovid MEDLINE, Embase, and Web of Science were searched from database inception to 9 April 2025. Peer-reviewed studies assessing in vitro cytotoxicity of 3D-printed aligners/retainers (and closely related intraoral appliances where relevant) were included. Data were charted across extraction protocols (ISO 10993-12-related), cell exposure parameters, and cytotoxicity assays (ISO 10993-5-related).

Results

Twenty-five studies published between 2020 and 2025 were included. Seventeen studies (17/25, 68%) explicitly stated compliance with ISO 10993-5 and/or ISO 10993-12. The remaining studies varied in ISO referencing (threshold-only, intentional deviation, or no ISO reference). Most studies used extract-only testing (15/25, 60%), followed by direct contact only (6/25, 24%) and combined extract-based and direct-contact approaches (3/25, 12%); one study used an alternative indirect monomer-in-ethanol approach (1/25, 4%). Among studies with an extraction component (19/25, 76%), extraction duration ranged from 24 h to 14 days, with 24 h being most common (9/19, 47%) and 14 days reported in 3/19 studies (16%). Extraction medium was reported in 18/19 studies (95%); among these, DMEM/DMEM-based media were most common (12/18, 67%). Numerical surface-area-to-volume reporting, a key ISO 10993-12 parameter, was provided in 11/19 studies (58%), while 8/19 (42%) explicitly did not report it. Most studies relied on a single metabolic activity assay (19/25, 76%), most frequently the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay, with limited use of complementary membrane integrity assays (5/25, 20%) or biomass-based assays (1/25, 4%).

Conclusions

In vitro cytotoxicity testing of 3D-printed orthodontic materials shows substantial heterogeneity and frequent reporting gaps across extraction conditions, exposure parameters, and assay strategies. Claims of ISO adherence are common but are variably supported by reporting of key ISO 10993-5/-12 domains. More consistent reporting and harmonization of extraction and testing parameters are needed to improve reproducibility and clinical interpretability.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40510-026-00633-2.

Background

Since its introduction in the 1980s, three-dimensional (3D) printing has become increasingly integrated into healthcare workflows, including dentistry and orthodontics [1]. In orthodontics, early digital approaches focused on virtual design and fabrication of removable appliances such as retainers [2]. Recent technological developments have accelerated the transition from printed models toward the direct manufacture of customized clear aligners and retainers [3, 4]. This shift is clinically relevant because directly printed appliances may streamline production, reduce material waste, and support fully digital treatment pathways [4–7].

However, the growing adoption of directly printed aligners and retainers is accompanied by concerns regarding their biocompatibility during prolonged intraoral use [5, 7–9]. Most printable materials for these appliances are photopolymer resins, commonly methacrylate- and urethane(dimeth)acrylate-based formulations, whose biological performance may be influenced by factors such as print parameters, post-processing, and the potential release of residual monomers or other leachable components [3–5, 8, 9]. Current evidence on biological safety remains limited and heterogeneous, and recent reviews have emphasized that differences in study design and reporting complicate clinical interpretation and cross-study comparison [5, 8].

In vitro cytotoxicity testing is typically the first step in the biological evaluation of medical devices intended for intraoral contact. Within the ISO 10993 framework, two parts are particularly relevant. ISO 10993-12 addresses sample preparation and extraction, including parameters such as specimen geometry, extraction medium, extraction duration, temperature, and surface area–to–volume (S/V) ratios. ISO 10993-5 provides guidance on in vitro cytotoxicity testing, including exposure approaches (extract-based and direct contact) and interpretation of viability endpoints [10]. Importantly, variability or incomplete reporting in ISO 10993-12 extraction domains can substantially influence the cytotoxicity outcomes subsequently measured under ISO 10993-5, thereby limiting comparability and clinical interpretability across studies.

Cytotoxicity assessments in this field commonly rely on extract-based testing, direct-contact testing, or a combination of both [10]. Outcomes are frequently measured using metabolic activity assays, while other assay classes, targeting membrane integrity or cell number/biomass, provide complementary information on cellular damage [10]. At the same time, differences in cell models (e.g., murine fibroblasts versus human oral-derived cells) and exposure durations further contribute to heterogeneity, particularly when intended clinical wear periods extend beyond short in vitro exposures [5, 8, 9]. Taken together, heterogeneity in extraction conditions, exposure designs, and assay strategies hampers synthesis of the literature and limits the interpretability of findings that reference ISO standards.

Given this context, a scoping review is well suited to map the breadth of existing testing approaches, identify methodological heterogeneity and reporting gaps, and clarify how ISO-guided domains are operationalized in practice. In PCC terms, the population comprises 3D-printed orthodontic aligners and retainers (and closely related printed intraoral appliances where relevant), the concept is in vitro cytotoxicity testing methodology across extraction, exposure, and assay strategies with attention to key ISO 10993-5/-12 domains, and the context is laboratory-based in vitro testing of printable dental photopolymers. Therefore, this scoping review aims to chart and evaluate in vitro cytotoxicity testing protocols used for materials applied in the 3D printing of orthodontic aligners and retainers, focusing on three methodological domains: (1) material extraction procedures, (2) exposure conditions and cell models, and (3) analytical techniques used to assess cytotoxicity.

Materials and methods

This scoping review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines [11]. Ethical approval was deemed neither mandatory nor necessary by the Ethics Committee of the Academic Centre for Dentistry Amsterdam.

Protocol and registration

A review protocol was not registered prior to study initiation.

Eligibility criteria

We included peer-reviewed studies that assessed in vitro cytotoxicity of materials used to 3D-print orthodontic aligners, retainers and closely related intraoral printed appliances where relevant. Studies were excluded if they (1) did not evaluate cytotoxicity, (2) assessed only non-dental 3D-printed materials, (3) were case reports or non-peer-reviewed sources, or (4) lacked sufficient methodological detail to chart extraction, exposure, and assay parameters. When key methodological information was not reported, it was coded as ‘not reported’ and was not imputed.

Information sources and search strategy

A comprehensive literature search was performed in Ovid Medline, Embase, and Web of Science (Core Collection) from database inception (i.e., the earliest indexed record available in the database) to April 9, 2025. The search was designed in collaboration with a medical information specialist from the Vrije Universiteit (VU) Amsterdam Medical Library (Supplementary Table 1). The search strategy combined controlled vocabulary and free-text terms related to three-dimensional printing, orthodontic appliances, and cytotoxicity/biocompatibility. As part of search strategy development, a sensitivity check was performed in MEDLINE on 28 February 2025 by omitting the term “three-dimensional printing” to evaluate its impact on retrieval. Although this increased the number of records, dual screening of a sample of excluded records confirmed that no relevant studies were missed; therefore, the term was retained in the final search strategy to improve precision.

All database-specific search strings, including the date each search was executed, are provided in Supplementary Table 1 to ensure reproducibility.

Study selection

Records were de-duplicated in EndNote [12, 13], and imported into Rayyan for screening. Two reviewers (K.T. and N.B.) independently screened titles and abstracts against the eligibility criteria, followed by full-text assessment of potentially eligible articles. Disagreements were resolved through discussion; if consensus could not be reached, a third reviewer (A.B.) adjudicated. Abstract-only records were eligible at the screening stage to minimize the risk of missing relevant evidence. However, inclusion at the eligibility stage required sufficient methodological detail to chart extraction, exposure, and assay parameters, and therefore no abstract-only record was included in the final sample.

Data extraction and charting

A standardized data-charting form was developed based on the review objectives and key ISO 10993 domains relevant to in vitro cytotoxicity testing and extraction/sample preparation [10]. The form was pilot-tested on five included studies by two reviewers (K.T. and N.B.) and refined accordingly. Data were extracted independently by the two reviewers and compared for consistency. Extracted items included: material type; extraction parameters (e.g., specimen dimensions, extraction medium, duration, temperature/incubation conditions, and surface area–to–volume ratio or mass-to-volume ratio); exposure parameters (e.g., cell line, exposure duration, extract dilution strategy and/or direct-contact setup); and cytotoxicity analysis technique. When methodological parameters were unclear or not reported, they were coded as “not reported” and were not imputed. Studies lacking sufficient methodological detail to chart the core domains (extraction, exposure, assay strategy) were excluded at eligibility assessment.

Mapping ISO 10993-5/-12 reporting domains

Because ISO compliance is frequently stated but variably operationalized across studies, we mapped the reporting of key ISO-relevant domains rather than judging definitive compliance. At study level, we coded whether key extraction/sample preparation domains (e.g., extraction medium, duration, temperature/incubation conditions, and numerical surface area–to–volume ratio) and cytotoxicity testing domains (e.g., cell model, exposure duration, and assay strategy) were reported. Study-level coding is provided in Supplementary Table 2.

Data synthesis

Extracted data were summarized descriptively and organized into three methodological domains: (1) material extraction/sample preparation, (2) exposure of cells to test materials, and (3) analytical techniques used to assess cytotoxicity. For each domain, we report ranges and frequencies (counts and percentages) of key parameters to describe methodological variability. To enhance transparency, we also provide a study-level overview of whether selected ISO-relevant parameters were reported or not reported (Supplementary Table 2). This supplementary table is provided for illustrative and traceability purposes only; we did not calculate a composite score, rate study quality, or judge definitive ISO compliance.

Quality appraisal and risk of bias

In line with scoping review methodology [14], we did not perform a formal risk-of-bias assessment. We focused on descriptive mapping of methods and reporting, supported by tabulated summaries.

Results

The systematic search yielded a total of 1,050 records: 307 from Ovid MEDLINE, 294 from Embase, and 449 from Web of Science (Core Collection). After removing duplicates, 570 unique records remained. Titles and abstracts were screened independently by two reviewers, followed by full-text assessment of potentially eligible articles. The study selection process is summarized in Fig. 1 (PRISMA flow diagram). No abstract-only record met the inclusion criteria at eligibility assessment and therefore none were included.

Fig. 1.

PRISMA Flow Chart

A total of 25 studies, published between 2020 and 2025, met the inclusion criteria (Fig. 1). The distribution across publication years was as follows: eight studies from 2022, six from 2023, four from 2024, three each from 2020 to 2021, and one from 2025. No additional records were identified through manual reference list screening.

Of the 25 included studies, 17 (17/25, 68%) explicitly stated compliance with ISO 10993-5 and/or ISO 10993-12 for cytotoxicity testing of medical devices [15–31]. Of the remaining eight studies (8/25, 32%), two referenced only the ISO cytotoxicity threshold of 70% cell viability without describing ISO-based procedures [32, 33], one intentionally deviated from ISO to simulate a worst-case scenario [34], and five did not reference ISO 10993 or any alternative recognized standard [35–39]. Study-level reporting of key ISO-relevant parameters is shown in Supplementary Table 2.

To aid readability, we summarize the most critical reporting gaps across ISO-relevant domains below. Across studies with an extraction component (n = 19), the most critical reporting gaps concerned the numerical surface area-to-volume ratio (not reported in 8/19, 42%) and specimen geometry/dimensions (not reported in 3/19, 16%), whereas extraction medium and temperature were reported more consistently (medium: 18/19, 95%; temperature: 18/19, 95%). Across all studies (n = 25), most relied on a single assay (19/25, 76%), and complementary membrane integrity endpoints were used infrequently (5/25, 20%). Supplementary Table 2 provides the study-level breakdown of reported versus not reported items.

Given the scoping nature of this review, results are presented descriptively and structured around three methodological domains: (I) material extraction/sample preparation, (II) exposure of cells to test materials, and (III) analytical techniques used to assess cytotoxicity (Table 1; Supplementary Table 2).

Table 1.

Characteristics of included studies

First author	ISO reference reported (as stated)	I. Extraction			II. Exposure		III. Analysis technique
		Materials	Preparation type	Extraction procedure	Cell line used	Exposure procedure
FayyazAhamed (2020)	ISO 10993-5 and ISO 10993-12	Methacrylate- and PMMA-based	Extract	Full-size aligner trays; S/V specified as per ISO 10993-5/-12; immersed in DMEM; 37 °C, 5% CO₂; medium refreshed at 1, 3, 5 & 7 days	3T3	Percentage extract dilution not specified; 37 °C, 5% CO₂; 24 h	MTT assay
Lin (2020)	ISO 10993-5	Methacrylate-based blend (Bis-EMA)/UDMA/TEGDMA)	Extract	Ø 10 × 10 × 10 mm cubes; S/V not reported; immersed in phenol-red-free DMEM + 10% FBS, antibiotics; 37 °C; 24 h with stirring; extracts sterile-filtered (0.2 μm)	L-929	100% extract dilution; 24 h	MTT assay
Wedekind [39]	ISO not reported	Acrylate-based blend (THFMA/UA/TPGDA/TPO)	Indirect extract	No extraction of material specimen; THFMA and TPGDA monomers dissolved in ethanol (≤ 1% v/v final)	HGF	Monomer dilutions; 37 °C and 5% CO₂ and 100% humidity; 24 h	XTT assay
Xu [32]	ISO 10993-5 and ISO 10993-12	Methacrylate-based	Extract + Direct	Exract dilution: Ø 14 mm × 2 mm discs; S/V = 1.25 cm²/mL; immersed in DMEM; 37 °C, 5% CO₂; 72 h. Direct: Ø 14 mm × 2 mm discs; S/V not applicable	L-929	Extract: 100% extract dilution; 37 °C, 5% CO₂; CCK-8 and live/dead fluorescence staining: 24 h. Direct: 37 °C, 5% CO₂; CCK-8 and live/dead fluorescence staining: 24 h	CCK-8 assay (WST-8) and live/dead assay
Bürgers (2022)	ISO 10993-5	Acrylate- and methacrylate-based	Direct	Ø 10 mm × 2.5 mm discs; S/V not applicable	L-929, HGF-1	In DMEM; 37 °C, 5% CO₂; 24 h	WST-8 assay (CCK-8)
Guerrero-Gironés [18]	ISO 10993-5 and ISO 10993-12	Methacrylate- and UDMA-based	Extract	Ø 6 mm × 2 mm discs; S/V = 1.5 cm²/mL; immersed in DMEM + 10% FBS, 1% GlutaMAX™, 1% antibiotics; 37 °C, 5% CO₂; 24 h	HGF	100%, 50%, and 25% extract dilutions; 37 °C, 5% CO₂; 24 h, 48 h, and 72 h	MTT assay
Lambart [25]	ISO 10993-5 and ISO 10993-12	Methacrylate-based	Extract	Ø 10 mm × 1.5 mm discs; S/V = 1.25 cm²/mL; immersed in DMEM; 37 °C; 72 h; extracts sterile-filtered (0.2 μm)	L-929	100% extract dilution; CCK-8 and live/dead fluorescence staining: 24 h	CCK-8 assay and live/dead assay
Pratsinis [36]	ISO not reported	Urethane-acrylate-based	Extract	Aligner fragments (n = 6); dimensions not described; S/V not reported; immersed in 25 mL sterile deionized water; 37 °C; 14 days with twice-daily agitation	HGF	20%, 10%, and 5% extract dilutions; 72 h	MTT assay
Raszewski1 [37]	ISO not reported	Experimental methacrylate-based blend (E4GMA/UDMA/TEGDMA/PETA)	Direct	Ø ≈ 5 × 5 × 1–2 mm discs; S/V not applicable	HGF	37 °C, 5% CO₂; 24, 48, 72 h	PrestoBlue® resazurin assay
Raszewski2 [38]	ISO not reported	Methacrylate-based	Direct	Ø 5 mm × 2 mm discs; S/V not applicable	HGF	In culture medium; 37 °C, 5% CO₂; PrestoBlue® resazurin: 24 h, MTT: 96 h	PrestoBlue® resazurin assay and MTT assay
Wulff [34]	Intentional deviation from ISO	Methacrylate- and polyesterdimethacrylate-based	Extract	Ø 5 mm × 1 mm discs; S/V = 5 cm²/mL; immersed in 2.615 mL RPMI; 37 °C, 5% CO₂; 24 h	RAW264.7 mouse macrophages	100% extract dilution and 1:2 serial dilutions (50%, 25%, 12.5%, …); 37 °C, 5% CO₂; 24 h	Crystal violet assay
Campobasso [35]	ISO not reported	UDMA-based	Extract	Ø 2 × 2 mm squares; S/V not specified; immersed in DMEM + 10% FBS; 37 °C, 5% CO₂; 7 & 14 days	Mouse pre-osteoblasts (MC3T3-E1)	100% extract dilutions; 7 and 14 days	MTT assay
Bieger [32]	ISO 10993 threshold only (70% viability)	Methacrylate- and urethane-based	Direct	Ø 12 mm × 1.5 mm discs; S/V not applicable	HGF-1	In DMEM; 37 °C, 5% CO₂; 24 h	WST-1 assay
Taher [28]	ISO 10993-12	Methacrylate-based	Direct	Ø 5 mm × 1 mm discs; direct contact; S/V not applicable	L-929, 3T3	In culture medium; 37 °C, 5% CO₂, 95% O₂; 14 days	MTT assay
Haugli [19]	ISO 10993-5 and ISO 10993-12	Methacrylate-based	Extract	Ø 13 mm × 2 mm discs; S/V = 3 cm²/mL; immersed in F-12 K + 5% FBS + antibiotics; 37 °C; 24 h	Human oral carcinoma cells (PE/CA-PJ49)	100%, 50%, and 25% extract dilutions; 37 °C, 5% CO₂; 24 h	MTT assay
Iodice [20]	ISO 10993-5 and ISO 10993-12	UDMA-based	Extract	Ø 10 × 10 × 0.5 mm squares; S/V not reported; immersed in human saliva or DMEM + 10% FBS, 2 mM L-glutamine, 1% antibiotics; 37 °C, 5% CO₂; 14 days	HGF-1	100% extract dilution; 37 °C, 5% CO₂; 72 h	MTT assay
Janjić (2024)	ISO 10993-5	Methacrylate-based	Direct	Ø 5 mm × 2 mm discs; S/V not applicable	L-929, GSC	37 °C, 5% CO₂; Resazurin and live/dead fluorescence staining: 24 h	Resazurin assay, and live/dead assay
Jin [22]	ISO 10993-12	Experimental UA-based	Extract	Ø 10 mm × 2.5 mm discs; S/V = 3 cm²/mL; immersed in DMEM + 10% FBS, 1% antibiotics; 37 °C, 5% CO₂; 24 h	HGF	100% extract dilution; 37 °C, 5% CO₂; 24 h	WST-8 (CCK-8) assay
Kim [23]	ISO 10993-5	Polyurethane-based	Extract	Aligner fragments (thickness = 0.5 mm; other dimensions not reported); S/V = 0.2 g/mL; immersed in RPMI 1640 + 10% FBS, 1% antibiotics; 37 °C, 5% CO₂; 24 h	L-929	Percentage extract dilution not specified; 37 °C, 5% CO₂; 24 h	MTT assay
Kollmuss [24]	ISO 10993-5	Methacrylate-based	Extract	Ø 6 mm × 1 mm discs; S/V = 3 cm²/mL; immersed in DMEM/F-12 + 10% FBS, 1% antibiotics; 37 °C, 5% CO₂; 24 h	HGF-1, hOK	100% extract dilutions from each material; 37 °C, 5% CO₂; 24 h	ATP-based assay
López-García (2024)	ISO 10993-5	Acrylate-based	Extract	Ø 6 mm × 2 mm discs; S/V = 1.5 cm²/mL; immersed in complete DMEM; 37 °C; 24 h; extract sterile-filtered (0.22 μm)	HGF	100%, 50%, and 25% extract dilutions; 24 h, 48 h and 72 h	MTT assay
Weżgowiec (2024)	ISO 10993-5 and ISO 10993-12	Methacrylate-based	Direct + Extract	Extract dilution: Ø 10 mm × 4 mm discs; S/V = 3 cm²/mL; immersed in complete DMEM; 37 °C; 24–72 h with 100 rpm agitation; extracts filtered. Direct: Ø 10 mm × 4 mm discs; S/V not applicable	HGF	Extract: PrestoBlue® resazurin: percentage extract dilutions not specified; 37 °C; 24–72 h. Direct: MTT and LDH: 37 °C, 5% CO₂; 24 h & 72 h	PrestoBlue® resazurin assay, MTT assay, and LDH assay
Wu [30]	ISO 10993-5	Urethane-acrylate-based	Direct + Extract	Extract dilution: Ø 10 mm × 2 mm discs; S/V = 3 cm²/mL; immersed in HyClone culture medium; 37 °C; 7 days; extracts sterile-filtered. Direct: Ø 10 mm × 2 mm discs; S/V not applicable	HGF-1, L-929	Extract: Elution assay: 100% extract dilution; 37 °C, 5% CO₂; 7 days. Direct: Agar diffusion assay: in culture medium; 37 °C, 5% CO₂; exposure time not reported	Elution assay (end-point not specified) and agar diffusion assay
Adami [15]	ISO 10993-5 and ISO 10993-12	PMMA-based	Extract	Sample dimensions not specified; S/V not specified; immersed in DMEM + 10% FBS + 1% antibiotics; 37 °C, 5% CO₂; 24 h	L-929	100% extract dilutions; 1, 2, and 7 days	MTT assay
Bleilöb (2025)	ISO 10993 threshold only (70% viability)	Epoxy-based	Extract	Ø 10 mm × 0.5–6 mm discs; S/V not specified; culture medium not specified; 37 °C, 5% CO₂; 12 days with intermittent shaking	HGF	Percentage extract dilution not specified; 48–72 h	AlamarBlue assay

Material extraction procedures

Among the 25 included studies, methacrylate-based photopolymers were the most frequently investigated material type, followed by urethane dimethacrylate (UDMA)-based photopolymers. Less frequently investigated materials included epoxy resin, an acrylate photopolymer (Med610), polymethyl methacrylate (PMMA; Smart Print Bio–Bite), and experimental acrylic/methacrylate blends (Table 1).

Cytotoxicity testing methods generally followed one of two approaches, as defined in ISO 10993-5: (1) indirect extract-based testing, in which materials are incubated in a liquid medium and the resulting eluate is applied to cells; or (2) direct-contact testing, in which the material is placed in direct contact with the cultured cells [10]. Most studies used extract-only protocols (15/25, 60%), followed by direct-contact only (6/25, 24%). A smaller proportion combined extract-based and direct-contact approaches (3/25, 12%). In addition, one study (1/25, 4%) employed an alternative indirect approach based on dissolving leached monomers in ethanol rather than performing specimen extraction [39] (Table 1).

An extraction component was present in 19/25 studies (76%) (Table 1). Among these, extraction duration ranged from 24 h to 14 days, with 24 h being most common (9/19, 47%). Longer extraction protocols were also reported, including 72 h (2/19, 11%), 7 days (2/19, 11%), 12 days (1/19, 5%), and 14 days (3/19, 16%). Extraction medium was reported in 18/19 studies (95%); one study did not specify the culture medium [33]. Among studies reporting the extraction medium (18/19), DMEM/DMEM-based media were most common (12/18, 67%), followed by RPMI (2/18, 11%) and other media (4/18, 22%). Extraction temperature was reported in 18/19 studies (95%), and where reported, extraction was performed at 37 °C (Table 1; Supplementary Table 2).

Specimen geometry/dimensions were reported in 16/19 extraction-based protocols (84%) (Table 1; Supplementary Table 2), with most studies using disc-shaped specimens but also including square/rectangular specimens and, in one case, full appliance extraction [17]. Reporting of the surface area–to–volume (S/V) ratio, which is a key extraction parameter, was variable: a numerical S/V (or mass-to-volume) ratio was provided in 11/19 studies (58%), while 8/19 (42%) did not report this parameter (Table 1; Supplementary Table 2). Where specified, S/V ratios were most commonly reported as cm²/mL, although one study used g/mL rather than cm²/mL [23].

Exposure of cells to test materials

Across the 25 included studies, 22/25 (88%) used at least one fibroblast- or connective tissue–derived cell model (Table 1). The most common choice was human gingival fibroblasts (HGF/HGF-1) alone (11/25, 44%). L929 fibroblasts alone were used in 5/25 studies (20%), while 2/25 (8%) combined HGF and L929. Only one study explicitly justified the use of HGF based on the anatomical relevance of gingival connective tissue as the primary interface with orthodontic aligners. Among L929-only studies, two explicitly cited ISO 10993-5 to justify use of the recommended murine fibroblast line [15, 25].

Additional cell models were used less frequently. One study combined HGF with human oral keratinocytes (hOK) (1/25, 4%) to represent epithelial–connective tissue interactions [24]. Two studies paired L929 with other connective tissue–related models: one combined L929 with 3T3 fibroblasts (1/25, 4%) to verify cytotoxicity across two murine lines [28], and another combined L929 with human gingival stromal cells (GSC) (1/25, 4%) to broaden the connective tissue representation [21]. Three studies used non–connective tissue dominant models, specifically RAW 264.7 macrophages [34], MC3T3-E1 pre-osteoblasts [35], or PE/CA-PJ49 oral carcinoma cells [19] (each 1/25, 4%).

Across studies, the maximum reported exposure duration ranged from 24 h to 14 days. Short exposures were most common, with 24 h reported in 13/25 studies (52%), followed by 72 h in 7/25 (28%) and 96 h in 1/25 (4%). Longer exposure periods were less frequently applied, including 7 days (2/25, 8%) and 14 days (2/25, 8%) (Table 1). Among studies using extract-based exposure (including combined designs; 19/25), testing frequently included undiluted eluates (14/19, 74%), often alongside serial dilutions or multiple extract concentrations. In direct-contact designs, specimens were placed directly onto the cell layer, most commonly for short exposure windows (typically within 24–96 h), with only one study using a longer contact period [28]. Study-level reporting of key exposure parameters is provided in Supplementary Table 2.

Analytical techniques for assessing cell viability

All 25 included studies assessed cytotoxicity by quantifying cell viability after exposure to material extracts, direct-contact specimens, or both (Table 1). Assays were grouped into three commonly used categories, consistent with ISO 10993-5 recommendations [10]:

(1) metabolic activity assays, (2) membrane integrity assays, and (3) biomass or cell number assays (Table 1).

Metabolic activity assays were used in 23/25 studies (92%) and were the dominant approach across the literature. These assays detect enzymatic conversion of colorless substrates into a colored or fluorescent product as a proxy for cell viability [10]. The most frequently reported test was the MTT assay (16/25, 64%), followed by water-soluble tetrazolium salt assays (4/25, 16%) (e.g., WST/CCK-8), and resazurin-based assays (3/25, 12%) (e.g., AlamarBlue/PrestoBlue). Less frequently used metabolic endpoints included XTT (1/25, 4%), and ATP-based luminescence assays (1/25, 4%) (Table 1).

Most studies relied on a single assay to characterize cytotoxicity (19/25, 76%), whereas 6/25 (24%) used two or more assays (Table 1). When multiple assays were applied, this typically combined a metabolic assay with a complementary method targeting membrane integrity. Membrane integrity assays were used in 5/25 studies (20%). These included Live/Dead fluorescence staining (3/25, 12%), an LDH release assay (1/25, 4%), and an agar diffusion approach (1/25, 4%) (Table 1). A biomass/cell number endpoint was rare, with one study (1/25, 4%) using the crystal violet assay.

Study-level reporting of assay selection is summarized in Table 1, with additional traceability provided in Supplementary Table 2.

Discussion

This scoping review provides an overview of how in vitro cytotoxicity testing is currently performed for materials used to 3D-print orthodontic aligners and retainers. Although the primary focus was aligners and retainers, we also included closely related 3D-printed intraoral appliances such as occlusal splints, as they frequently use overlapping printable resin chemistries and post-processing workflows that inform cytotoxicity testing methodology. Twenty-five studies were mapped across three methodological domains: (I) extraction/sample preparation, (II) exposure of cells to test materials, and (III) analytical techniques used to assess cytotoxicity. While ISO adherence was frequently claimed (17/25), substantial heterogeneity and incomplete reporting were observed across extraction conditions, exposure designs, and assay strategies (Table 1; Supplementary Table 2). In studies with an extraction component (19/25), extraction durations ranged from 24 h to 14 days, and reporting of key extraction parameters (e.g., numerical surface area–to–volume ratio) was inconsistent.

Where extraction temperature was reported, it was typically 37 °C, but this parameter was not universally documented. Most studies used fibroblast-based cell models and short exposure windows, and viability assessment relied predominantly on metabolic activity assays (most commonly MTT), with limited use of complementary endpoints (Table 1).

The variability observed across studies has important implications for both researchers and clinicians. Although many studies state adherence to ISO guidance, the practical implementation and reporting of key testing domains can differ substantially, which may undermine comparability across studies and complicate clinical interpretation. Gruber and Nickel (2023) demonstrated that even when laboratories follow ISO 10993-5, outcomes can vary unpredictably, reflecting flexibility in parameters such as extraction duration, extraction medium composition, and surface area–to–volume ratios [40, 41]. In this context, ISO claims alone may be insufficient to support clinical relevance when extraction and exposure conditions vary or are incompletely reported.

Several extraction parameters can influence test sensitivity and the measured magnitude of cytotoxicity. For example, longer extraction durations and smaller extraction volumes can increase the concentration of leachables in the eluate, potentially amplifying cytotoxic effects [40, 42]. The composition of the extraction medium may also affect the interaction between leachables and cells, thereby influencing measured viability [40]. A further challenge concerns the reporting of surface area–to–volume ratios. ISO 10993-12 recommends surface area–based reporting (cm²/mL) when feasible, as it most directly reflects exposure conditions, whereas mass-based ratios (g/mL) are appropriate only when surface area cannot be determined [10]. Overall, these considerations support more consistent reporting and harmonization of extraction parameters to improve reproducibility and cross-study comparability.

Only a small proportion of studies combined extract-based and direct-contact cytotoxicity testing (Table 1). Such combined designs may help capture both eluate-mediated effects and surface/contact-related effects, which could be relevant for intraoral appliances that simultaneously contact mucosa and release substances into the surrounding fluid phase. However, these approaches were applied inconsistently and were rarely accompanied by standardized reporting of extraction and exposure parameters, limiting direct comparisons. Future studies could more systematically evaluate combined designs using clearly reported extraction conditions, clinically motivated exposure durations, and complementary endpoints.

Several studies acknowledged limitations related to the use of a single cell type and short exposure durations. In this review, fibroblast-based models predominated, but the selection of cell lines varied and was often not explicitly justified. Exposure durations also differed widely, with many studies using short time points (e.g., 24–96 h) and fewer studies extending exposures to longer periods that may better reflect appliance wear (Table 1). More explicit justification of cell model choice and exposure duration would strengthen the interpretability and translational value of future studies.

Given that aligners and retainers are in prolonged contact with oral soft tissues, greater use of human oral–relevant cell models may improve biological relevance compared with exclusive reliance on murine fibroblast lines. While ISO 10993-5 recommends L929 as a standardized reference cell line, it does not prescribe a single optimal model for all applications. A pragmatic approach is to include human gingival fibroblasts and, where feasible, epithelial models or co-cultures, and to clearly justify the chosen cell model and experimental conditions. Importantly, increased biological relevance should be paired with standardized and transparently reported extraction and exposure conditions to support cross-study comparability.

From a clinical perspective, aligners are typically worn continuously for extended periods and replaced at regular intervals, meaning that cumulative exposure to leachables over days to weeks may be clinically relevant. In this review, however, longer exposure windows were applied in only a minority of studies (Table 1). Future in vitro work would benefit from aligning exposure durations more closely with clinically plausible wear intervals, alongside designs that better approximate intraoral conditions (e.g., dynamic fluid exposure, aging, and surface wear), while maintaining standardized and transparent reporting.

Consistent with prior reviews, most included studies relied on a single metabolic activity assay, which limits insight into the mechanism and nature of cellular damage and reduces robustness of the findings [41–43]. In our sample, single-assay designs were common, while complementary membrane integrity or biomass-based endpoints were used infrequently (Table 1). ISO 10993-5 recommends using more than one appropriate assay, but leaves the specific combination to the investigator [10]. Notably, Gruber and Nickel emphasized that variability in outcomes may be driven more by extraction and exposure conditions than by the specific metabolic assay selected [40]. This reinforces the importance of harmonizing extraction and exposure parameters, while also encouraging the use of complementary endpoints to strengthen interpretation.

This review highlights several opportunities to improve consistency and clinical interpretability of cytotoxicity testing in this rapidly evolving field. At minimum, future studies should report a core set of methodological items to enable reproducibility and cross-study comparison: (1) specimen geometry/dimensions; (2) numerical surface area–to–volume ratio (cm²/mL) or justified mass-to-volume ratio; (3) extraction medium composition (including serum content where applicable), duration, temperature/incubation conditions, and agitation; (4) cell model selection with a clear rationale and clinically motivated exposure duration; and (5) assay strategy including at least one metabolic endpoint and one complementary endpoint (e.g., membrane integrity or biomass/cell number). More consistent reporting of these items would substantially strengthen the evidence base and support meaningful standardization efforts.

For practicing orthodontists, the key takeaway is that claims of ISO adherence in the literature do not necessarily imply that materials were tested under uniform or clinically comparable conditions. Across studies, extraction parameters, exposure designs, and assay strategies varied widely, and key methodological details were not consistently reported (Table 1; Supplementary Table 2). Until greater harmonization is achieved, cytotoxicity findings should be interpreted in the context of the underlying test conditions rather than as direct proxies for clinical safety.

Clinicians and researchers evaluating new 3D-printed aligner materials should examine whether supporting studies clearly report: (1) extraction conditions (medium, duration, temperature/incubation conditions, and numerical S/V ratio), (2) the cell model and exposure duration, ideally aligned with clinically plausible wear intervals, and (3) whether conclusions are supported by more than a single metabolic assay. Transparent reporting and method harmonization will be essential for improving reproducibility, strengthening regulatory confidence, and facilitating safe adoption of emerging 3D-printed orthodontic materials.

This review has limitations. As a scoping review, we mapped methodologies and reporting but did not perform a formal risk-of-bias assessment. Interpretation is constrained by heterogeneity and incomplete reporting; comparisons are limited to explicitly reported parameters. In addition, the evidence base reflects peer-reviewed studies indexed in three databases with searches completed on 9 April 2025, and newer studies may not be captured in this rapidly evolving area. Finally, in vitro models cannot fully replicate intraoral conditions, and cytotoxicity findings should be interpreted accordingly.

Conclusions

This scoping review mapped how in vitro cytotoxicity testing is currently performed for materials used to 3D-print orthodontic aligners and retainers and examined alignment with key ISO 10993 testing domains. Despite frequent claims of ISO adherence, substantial heterogeneity and reporting gaps were identified across extraction/sample preparation, exposure conditions, and assay strategies, limiting cross-study comparability. In particular, inconsistent reporting of key extraction parameters (e.g., numerical surface area–to–volume ratio and extraction conditions) reduces the clinical interpretability of “ISO-based” cytotoxicity findings. More consistent reporting and harmonization of extraction and testing parameters, together with complementary assay strategies and clinically motivated exposure designs, are needed to strengthen reproducibility and support safer and more evidence-informed adoption of 3D-printed orthodontic materials.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

KT and NB carried out the data collection and wrote the manuscript. RV aided in data collection and helped draft the manuscript. AB helped with design and manuscript editing. LV conceived the study and participated in its design, coordination, and manuscript editing. The authors read and approved the final manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Data availability

All data generated or analysed during this study are included in this published article and its supplementary information files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.