Stem Cell-Based Tissue Engineering for Cleft Defects: Systematic Review and Meta-Analysis

Abstract

Objectives

This study aimed to analyze the efficacy of stem cell-based tissue engineering for the treatment of alveolar cleft (AC) and cleft palate (CP) defects in animal models.

Design

Systematic review and meta-analysis.

Setting

Preclinical studies on alveolar cleft repair in maxillofacial practice.

Patients, Participants

Electronic search was performed using PubMed, Embase, and Cochrane databases. Pre-clinical studies, where stem cell-based tissue engineering was used in the reconstruction of AC and CP in animal models were included. Quality of the selected articles was evaluated using SYRCLE (SYstematic Review Centre for Laboratory animal Experimentation).

Interventions

Review of alveolar cleft bone augmentation interventions in preclinical models.

Main Outcome Measures

Outcome parameters registered were new bone formation (NBF) and/or bone mineral density (BMD).

Results

Thirteen large and twelve small animal studies on AC (21) and CP (4) reconstructions were included. Studies had an unclear-to-high risk of bias. Bone marrow mesenchymal stem cells were the most widely used cell source. Meta-analyses for AC indicated non-significant benefits in favor of: (1) scaffold + cells over scaffold-only (NBF P = .13); and (2) scaffold + cells over empty control (NBF P = .66; BMD P = .31). Interestingly, dog studies using regenerative grafts showed similar to superior bone formation compared to autografts. Meta analysis for the CP group was not possible.

Conclusions

AC and CP reconstructions are enhanced by addition of osteogenic cells to biomaterials. Directions and estimates of treatment effect are useful to predict therapeutic efficacy and guide future clinical trials of bone tissue engineering.

Introduction

Oral clefts consist of heterogeneous congenital malformations that are typically presented as incomplete formation of the upper lip (cleft lip) and/or the roof of the mouth (cleft palate). The malformations occur in about 1 in 700 live births. They can appear individually, or both defects may occur together (cleft lip and palate). ¹ The conditions may develop as a unilateral or bilateral malformations with a wide range of severity. ² The oral cleft may also occur with other congenital anomalies or be part of a genetic syndrome.^2,3 The malformations are usually associated with the following factors: heredity, genetics, nutritional disturbances, stress during developmental stages, inadequate vascular supply, mechanical disturbances, infections, and teratogens that inhibit the union of nasal process and palatal shelves between the fourth and tenth week of gestation age. ⁴

One of the crucial steps of oral cleft surgery is the reconstruction of the alveolar cleft and cleft palate by a multidisciplinary team with various approaches depending on the degree of the defect.^5,6 The gold standard for cleft palate surgery is primary palate repair, usually performed around 18 months. ⁶ However, this method is often associated with insufficient tissue to close the defect properly ⁷ or post-surgical results such as facial growth disturbance and oronasal fistula. ⁵ As for alveolar cleft surgery, the standard therapy uses autologous bone grafts to replace the lost bone. ⁵ The timing of alveolar cleft surgery, in general, is divided into three stages: early repair (<5 years old), secondary repair around the canine eruption (>10 years old), and late repair (>13 years old). ⁶ The therapy, however, has several side effects, such as growth disturbances, ⁶ specific to donor site morbidity such as infection, bleeding, loosening of splint, pain, or sensory deficiency.^8,9 Allograft and synthetic materials as alternatives to autologous bone grafts also have several side effects such as infection, immunologic reaction, ⁵ and reduced bone formation rates. ¹⁰ All of these standard approaches may become more complex due to the need for simultaneous repair (eg, cleft palate and alveolar cleft repair at the same time) in areas where health facilities are limited. ¹¹

These challenges prompted the search for better alternatives for the golden standard procedure. Preferred technologies are those that are feasible, adaptive, and cost-effective with minimal side effects, and can be implemented even in limited settings. One example is the use of stem cell-based tissue engineering. The technology combines stem cells, biomaterials or scaffolds, and/or biomolecules to generate new tissue.^12,13 The combination can be used to replace the harvesting process of autologous bone graft for alveolar cleft repair ¹² and to overcome the poor quality or quantity of mucosa for cleft palate repair. ¹³ The application of stem cell-based tissue engineering for the alveolar cleft is not new several clinical applications have been reported.^14,15 In contrast, the progress of stem cell-based tissue engineering application for palatal bone is still limited to animal studies.^16,17

Many article reviews have discussed the topic of tissue engineering for cleft palate or alveolar cleft. To name a few, Moreau et al wrote an article review about the general concept of tissue engineering as an alternative way of cleft palate reconstruction. ¹³ It was Zuk et al who first wrote an article review focused on possible applications of adipose stem cells for cleft-palate tissue engineering procedure. ¹⁸ In 2015, Gladyzs et al described stem cell-based tissue engineering for alveolar cleft in a narrative review, but only summarized the pre-clinical studies, early case reports, and ongoing trials. ¹⁹ Recently, Shanbhag et al conducted a large systematic review and meta-analysis of cell-based tissue engineering in clinical and pre-clinical studies in a broader manner in all oral and maxillofacial areas. ²⁰ However, none of these reviews focused on stem cell-based tissue engineering for the alveolar cleft and cleft palate.

Therefore, the present study aims to evaluate the efficacy of stem cell-based tissue engineering for cleft palate and alveolar cleft defects by conducting a systematic review and meta-analysis of pre-clinical studies.

Materials and Methods

Protocol and Eligibility Criteria

This review was conducted based on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement. ²¹ The protocol was registered on PROSPERO (ID: CRD42021259614). The inclusion criteria were:

English language studies

Randomized or non-randomized controlled animal experimental studies with two or more experimental groups

Transplantation of differentiated or undifferentiated mesenchymal stem cells seeded on biomaterial scaffolds in at least one experimental group

A control group receiving “cell-free” biomaterial scaffolds and/or autogenous bone

Reported results quantitative histomorphometric new bone formation/growth (%NBF/NBG), quantitative radiographic assessment of bone formation via computerized tomography (CT) or micro-CT (%NBF/NBG), quantitative histomorphometric assessment of remaining defect (RD), and/or quantitative radiographic assessment of RD or Bone Mineral Density (BMD) using CT or micro-CT.

The exclusion criteria were:

1. In vitro studies

2. Case reports

3. Absence of a control group

Information Sources and Search

The electronic literature search was performed using MEDLINE (via PubMed), Embase, and Cochrane for relevant English-language articles until 5 April 2022. Other literature was searched via the Google and Google Scholar search engines. A specific search strategy was developed for MEDLINE and adapted for other databases.

#1 “Mesenchymal Stromal Cells"[Mesh] OR “Mesenchymal Stem Cell Transplantation"[Mesh] OR Mesenchymal Stromal Cell*[tiab] OR Mesenchymal Stroma Cell*[tiab] OR Mesenchymal Stem Cell*[tiab] OR BMSC*[tiab] OR Mesenchymal Progenitor Cell*[tiab] OR Bone marrow stromal cell*[tiab] OR Bone marrow stroma cell*[tiab] OR Bone marrow stem cell*[tiab]

#2 “Adipose Tissue"[Mesh:NoExp] OR “Abdominal Fat"[Mesh] OR ADSC*[tiab] OR ASC[tiab] OR ASCs[tiab] OR Abdominal Adipose Tissue*[tiab] OR Abdominal fat pad*[tiab] OR Adipose Derived Stem Cell*[tiab] OR Adipose Stem Cell*[tiab] OR stromal vascular fraction*[tiab] OR SVF[tiab]

#3 “Cleft Palate"[Mesh] OR OR cleft palate*[tiab] OR palatal cleft*[tiab] OR alveolar cleft*[tiab]

#4 “Alveolar Bone Grafting"[Mesh] OR (Alveolar Bone[tiab] AND (graft*[tiab] OR repair*[tiab] OR transplant*[tiab]))

#5 ((#1 OR #2) AND (#3 OR #4))

Study Selection, Data Collection, and Data Items

Title and abstract screening were conducted by two independent reviewers (DSNK and SAA) to obtain full texts of all eligible articles. Disagreements in determining the eligible articles were resolved by discussion. A third reviewer (FNH) was consulted for statistical analysis and, if necessary, to evaluate the articles. Three authors (DSNK, SAA, and NEN) reviewed the full-text articles and decided on the final eligible studies based on the inclusion and exclusion criteria. A summary of the whole screening process is presented in Figure 1.

Figure 1.

Flowchart of the study selection process.

The extracted data from the eligible articles were the following: author, publication year, subjects/models, number of subjects, age, stem cell criteria (source, expanded/non-expanded, osteogenic medium usage, cell dose/density), scaffold criteria (type, size), growth factor criteria (type, dose/concentration), control group, observation time, method, and result (histomorphometry, CT, other). Descriptive data of the included studies were stored in tables. Quantitative data of histomorphometric new bone formation (%NBF), radiographic assessment of bone formation via CT or micro-CT (%NBF), histomorphometric assessment of remaining defect (RD), and/or radiographic assessment of RD or BMD using CT or micro-CT data were extracted for possible meta-analysis. If data were only expressed graphically, numerical values were requested from the authors, and if no response was received digital ruler software was used to measure graphical data (ImageJ; National Institutes of Health, Bethesda, MD, USA). When studies reported outcomes at multiple time points, outcomes from the latest time points were extracted. The outcome data from similar time-points of different studies were pooled for meta-analysis by DSNK and NEN. When studies reported outcomes of more than one experimental group, meta-analysis was performed by “including each pair-wise comparison separately, but with shared intervention groups divided out approximately evenly among the comparisons” (Cochrane Handbook Chapter 16.5). ²²

Risk of Bias

The risk of bias (RoB) of animal studies was assessed using SYRCLE (SYstematic Review Centre for Laboratory animal Experimentation).^23,24 The results were presented in the risk of bias graph and summary using RevMan 5.4 program (Review Manager. The Cochrane Collaboration, 2020).

Meta-Analysis

The data were analyzed using Review Manager Software version 5.4 (Review Manager. The Cochrane Collaboration, 2020). Meta-analysis was performed by comparing the standardized mean difference of outcome measures for new bone formation and remaining defects after using differentiated or undifferentiated mesenchymal stem cells for cleft palate and alveolar cleft defects. Subgroup analyses were performed at the level of animals. A P-value < .05 was considered statistically significant. Statistical analysis was performed for an evaluation period of at least 6 weeks (42 days). Statistical heterogeneity was analysed using Cochrane's Q test and the inconsistency I² test, in which value higher than 50% were considered indicative of substantial heterogeneity. Publication bias was not assessed using the symmetry of funnel plots because there were less than 10 studies thus the assessment methods are not very reliable).^25,26

Results

Initially, 365 articles were identified from MEDLINE (via PubMed), Embase, and Cochrane databases. No studies were identified from other sources. Of 365 articles, only 25 studies were included for qualitative analysis, and only 10 of the 25 studies were eligible for quantitative analysis. All articles were in vivo studies in an animal model that investigated the alveolar cleft (21 studies) or cleft palate (4 studies) using cell-based tissue engineering. The maximum follow-up time ranged from 6 weeks (42 days) to 6 months (180 days). The characteristics of the included studies were summarized in Tables 1 and 2.

Table 1.

Animal Studies of the Alveolar Cleft Defect.

No	Study	Model	Number	Cell Source	Scaffold	Growth Factor	Experimental group(s)	Control Group	Results
1.	Ahlfeld 2021	Immunocompetent Lewis rats (alveolar cleft)	16 (4 dropouts, total 12)	Syngenic (rMSC)	CPC and cell-laden fibrin hydrogel	n.a	2)CPC/fibrin/rMSCs	1)Historical data obtained with scaffold alone	Histomorphometry (12 weeks) Group 2: NBF 13.7 ± 12.1%. Micro-CT (12 weeks) Lamellar and cancellous bones were isodense with the 3D plotted CPC structure.
2.	Bangun 2021	Goats (alveolar cleft)	24	Xenogenic (hUMSC)	HA/Chitosan/Gelatin	BMP-2 (Novosys)	2) HA/Chito-san/Gelatin + rhBMP-2 3) HA/Chito-san/Gelatin + hUMSC + BMP-2	1) ICABG	Histology using Mankani Score (12 weeks) Group 1: 3.2 (3;3-4) Group 2: 2 (1.5;1-4) Group 3: 3.2 (3.5;2-4) 3D CT-Scan (12 weeks) Group 1: NBF 22.53% (20.6; 13.9-35) Group 2: NBF 19.05% (16.2; 10.8-33.1) Group 3: NBF 38.3% (39.8; 0.9-72.8)
3.	Kandalam 2021	Athymic nude rats (bilateral alveolar bone defect)	30	Xenogenic (dGMSC)	Hydrogel PuraMatrix™ (PM)	BMP-2	2) PM/BMP-2 3) PM/dGMSC 4) PM/dGMSC/BMP-2	1a) Empty defect 1b) PM alone	Histology (8 weeks) Group 1b: supported new bone formation Group 2: showed lamellar bone formation followed by mature bone Group 3: showed viable osteocytes with lacunae, osteoblasts, and immature blood vessels within the new bone Group 4: more matured bone with reversal lines and neovascularization with osteocytes Micro-CT (8 weeks) Group 1a: BV 22.40 ± 0.59% Group 1b: BV 22.75 ± 3.08% Group 2: BV 27.10 ± 3.08% Group 3: BV 26.22 ± 1.22% Group 4: BV 39.98 ± 7.73%
4.	Toyota 2021	Immunocompetent Sprague-Dawley rats (alveolar cleft)	15	Xenogenic (UC-MACS)	ReFit (HA + Collagen 80:20)	n.a	2) Scaffold + UC-MACS	1a) Empty control 1b) Scaffold only	Histology (8 weeks) Group 1a: NBF 24.5% ± 7.6% Group 1b: NBF 44% ± 5.7% Group 2: NBF 60.5% ± 6.9% Micro-CT (8 weeks) Group 1a: small number of island-shaped opacities Group 1b: a thin bone bridge between the bone defect Group 2: a bone bridge between the defect with a higher CT value than immediately after implant
5.	Korn 2020	Immunocompetent Lewis rats (alveolar cleft)	80	Syngenic (donor adult Lewis rats) (rMSC)	CPC	n.a	2) Scaffold A + rMSC 3) Scaffold B + rMSC	1a) empty control 1b) Scaffold A (60° rotated layers) only 1c) Scaffold B (30° rotated layers) only	Histology (12 weeks) Group 1a: NBF 22.5 ± 1.8% Group 1b: NBF 19 ± 1.8% Group 2: NBF 8.7 ± 1.8% Group 1c: NBF 10.2 ± 2.0% Group 3: 10.8 ± 1.9%
6.	Shahnaseri 2020	Mongrel dogs (alveolar cleft)	4	Autologous (MSCs subcutaneous adipose stem cell)	HA/TCP	n.a	2) Tissue engineered MSCs with HA/TCP	1)Autograft group (Tibia)	Digital radiography (90 days) Group 1: bone density 100.32 ± 41.17 Group 2: bone density 93.77 ± 29.73
7.	Sun 2020	JW Rabbits (immune status unknown) (alveolar cleft)	24	Xenogenic (hUMSC)	Bone collagen particles (HA and collagen 1) + collagen membrane	n.a	2) Bone collagen particles (Collagen + HA) + hUMSC	1) Empty control 1b) Bone collagen particles (HA and collagen)	Histology (6 months) Group 1a: no NBF Group 1b: Small amount of NBF Group 2: Large portion of NBF Micro-CT (6 months) Group 1: BT 11.05 ± 1.23% Group 1b: BT 25.29 ± 2.53% Group 2: BT 31.18 ± 2.12%
8.	Wang 2019	Rhesus monkey (alveolar cleft)	4	Syngenic (rBMSC)	3D-BG microspheres	BMP/CS nanoparticles	2)3D-BG + BMP 3)3D-BG + BMP/CS + BMSCs	1a) SO group 1b) 3D-BG	Histology (12 weeks) Group 1a: no sign of osteogenensis Group 1b: has undegraded scaffolds, new bone tissue, and massive connective tissue in bone defect Group 2: has a small number of undegraded scaffolds, surrounded by bone-like tissue around the unabsorbed scaffold material Group 3: scaffolds were all absorbed, and a large number of new bones and new blood vessels were formed.
9.	Caballero 2017	Yorkshire pigs (Michael Fanning Farms, Howe, IN) (alveolar cleft)	22	Autologous (UC-MSCs)	PLGA	n.a	2) Undiff - MSCs 3) Diff - MSCs	1a) Empty control 1b) Autologous ICG	CT (30 days) Group 1a: NBG 1.94 ± 1.35 mm3/kg Group 1b: NBG 2.58 ± 2.83 mm3/kg Group 2: NBG 7.23 ± 4.99 mm3/kg Group 3: NBG 5.82 ± 4.48 mm3/kg
10.	Korn 2017	Immunocompetent Lewis rats (alveolar cleft)	84	Syngenic (BMSC donor adult Lewis rats)	bHA	n.a	2) bHA 3) Undiff -MSCS + bHA 4) Diff – MSCs + bHA	1)Empty control	Histomorphometry (12 weeks) Group 1: NBF 43 ± 13.3% Group 2: NBF 30.8 ± 12.1% Group 3: NBF 20.5 ± 10.9% Group 4: NBF 11.5 ± 7.3% CT (12 weeks) Group 1: RD 11.07 ± 2.32 mm3 Group 2: RD 12.57 ± 1.17 mm3 Group 3: RD 13.25 ± 1.48 mm3 Group 4: RD 14.08 ± 1.36 mm3
11.	Wen 2016	Immunocompetent Sprague-Dawley rats (alveolar defect)	4	Allogenic (MSC and EPC from femur and tibia of rats)	FG; Pasteurized FG	n.a	2) MSCs/EPCs (coMSCs)	1) MSCs (monoMSCs)	Histology (6 weeks) Group 2 presented better healing condition, with a large amount of BMSC and osteogenetic cells in the center of the defect Micro-CT (6 weeks) Group 1: 362.67 ± 27.65 HU Group 2: 527.78 ± 23.37 HU
12.	Liang 2016	Immunocompetent Sprague-Dawley rats (alveolar bone defect)	27	Allogenic (MSC and EPC from 2-week-old rats)	n.a	n.a	2)EPC + MSC 3) MSC	1) Empty control	Histology (8 weeks) Group 1: NBF 28.53 ± 2.81% Group 2: NBF 44.72 ± 5.96% Group 3: NBF 70.28 ± 8.3% Micro-CT (6 weeks) Group 1: TMD 503.66 ± 29.58 mg/cc Group 2: TMD 546.62 ± 34.67 mg/cc Group 3: TMD 609.88 ± 48.01 mg/cc
13.	Huang 2015	Beagle dogs (alveolar cleft)	14	Autologous (BMSC)	β-TCP	n.a	2) Autologous ICG / No RME 3) Autograft ICG /RME 4) BMSCs/ β-TCP/RME	1) Empty control /No RME	Histomorphometry (12 weeks) Group 1: NBF 13.11 ± 1.72% Group 2: NBF 55.74 ± 9.26% Group 3: NBF 79.51 ± 4.92% Group 4: NBF 78.69 ± 6.39% Radiography (12 weeks) Group 2: VH 4.85 ± 0.4 mm Group 3: VH 5.85 ± 0.48 mm Group 4: VH 5.97 ± 0.48 mm
14.	Yuanzheng 2015	Beagle dogs (alveolar cleft)	20	Autologous (BMSC)	Autologous bone	PRF	2) Autologous graft (ICG)/BMSC/PRF 3) Autologous graft (ICG)/BMSC 4) Autologous graft (ICG)/PRF	1) Autologous graft (ICG)	CT (6 months) Group 1: BMD 733.56 ± 69.31 mg/cc K2HPO4 Group 2: BMD 1233.56 ± 94.93 mg/cc K2HPO4 Group 3: BMD 1182.47 ± 83.97 mg/cc K2HPO4 Group 4: BMD 1142.33 ± 80.27 mg/cc K2HPO4
15.	Korn 2014	Immunocompetent Female Lewis rats (alveolar cleft)	72	Syngenic (BMSC donor female Lewis rat)	HA ceramics/β-TCP/silica matrix	n.a	2) Scaffold only 3) Scaffold/Undiff -MSCs 4) Scaffold/Diff - MSCs	1) Empty control	Histomorphometry (6 weeks) Group 1: RD width 2.63 ± 0.52 mm Group 2: RD width 2.70 ± 0.66 mm Group 3. BMSC group: RD width 2.39 ± 0.23 mm Group 4: RD width 2.53 ± 0.22 mm CT (6 weeks) Group 1: RD volume 6.86 ± 3.21 mm3 Group 2: RD volume 5.50 ± 1.05 mm3 Group 3: RD volume 4.08 ± 1.36 mm3 Group 4: RD volume 5.00 ± 0.84 mm3
16.	Raposo-Amaral 2014	Non-immunosuppressedWistar rats (alveolar cleft)	28	Xenogenic (human MSCs from orbicularis oris muscle of cleft patients)	1. Bio-Oss collagen 2. α-TCP-matrix	n.a	2)Bovine bone mineral free of cells 3)Bovine bone mineral loaded with MSCs 4)α-tricalcium phosphate free of cells 5) α-tricalcium phosphate loaded with MSCs	1)Autogenous bone grafts	Histomorphometry (8 weeks) Group 1: NBF: 60.27 ± 16.13%, Fibrosis Volume: 3.89 ± 10.24% Group 2: NBF: 23.02 ± 8.6%, Fibrosis Volume: 19.85 ± 7.04 Group 3: NBF: 38.35 ± 19.59%, Fibrosis Volume: 18.55 ± 12.41% Group 4: NBF: 51.48 ± 11.7%, Fibrosis Volume: 13.24% ± 12.07% Group 5: NBF: 61.80 ± 2.14%, Fibrosis Volume: 0.64 ± 1.56%
17.	Pourebrahim 2013	Mongrel dogs (alveolar cleft)	4	Autologous (ASC)	HA/β-TCP	n.a	2)Adipose tissue (MSCs)/Scaffold	1)Autologous graft (Tibia)	Histomorphometry (15 days) Group 1: NBF 45 ± 14.14% Group 2: NBF 5 ± 1.75% Histomorphometry (60 days) Group 1: NBF 96 ± 3.55% Group 2: NBF 70 ± 16.41%
18.	Chung 2012	Miniature swine (alveolar bony defects)	9	Autologous (BMSC)	PF127	advBMP-2	2)AdvBMP-2 infected MSC/PF127	1)MSC/PF127	Histology (3 months) Group 2: Considerably more bone was formed than in the group 1, extending from the apical aspect of the defect through the coronal extension Group 1, only small amounts of immature bone were formed. The new bone contained some fatty marrow space and extended from the apical aspect of the defect to the middle of the root 3D CT (3 months) Group 1: TBV 690.55 ± 119.84 mm3 Group 2: TBV 1824.84 ± 14.36 mm3
19.	Yoshioka 2012	Beagle dogs (jaw cleft)	3	Autologous (BMSC)	CAP particles	n.a	2) CAP/BMSC	1) Scaffold (CAP particles)	Histology (6 months) Group 1: NBF was present in transplanted area, but fibroblastic cells were still located around CAP particles Group 2: NBF was observed in almost the whole area and CAP particles had almost disappeared X-Ray (6 months) Group 1: Radio-opacity NB 0.35 ± 0.15 Grroup 2: Radio-opacity NB 0.75 ± 0.2
20.	Zhang 2012	Immunocompetent Sprague-Dwayley rats (alveolar defect)	15	Allogenic (BMSC from femora of 2-week-old rat	FG	n.a	2)FG only 3) BMSC + FG	1)Empty control	Histology (6 weeks) Group 1: poor NBF Group 2: poor NBF Group 3: good healing conditions Micro-CT (6 weeks) Group 1: BMD 669.04 ± 6.72 HU Group 2: BMD 668.80 ± 6.70 HU Group 3: BMD 682.96 ± 6.70 HU
21.	Zhang 2011	Beagle dogs (alveolar cleft)	7	Autologous (BMSC)	β-TCP	n.a	2) Autologous graft 3) β-TCP 4) BMSC/ β-TCP	1)Empty control	Histomorphometry (20 weeks after orthodontic treatment is completed) Group 1: no data Group 2: NBF 78.68 ± 4.91% Group 3: NBF 54.98 ± 9.22% Group 4: NBF 70.79 ± 7.02% Ratio Residual Alveolar Height (X-Ray 20 weeks) Group 1: no data Group 2: 72.42 ± 8.72% Group 3: 56.31 ± 7.72% Group 4: 73.6 ± 6.51%

Abbreviations: rMSC, rat mesenchymal stromal cells; CPC, calcium phosphate cement; NBF, new bone formation; hUMSC, human umbilical cord mesenchymal stromal cells; HA, hydroxyapatite; BMP-2, bone morphogenetic protein 2; ICABG, iliac crest bone graft; dGMSC, differentiated gingiva derived mesenchymal stem cells; BV, bone volume; UC-MACS, enzymatic digested human umbilical cord MSC using magnetic-activated cell sorting; n.a., not applicable; TCP, tricalcium phosphate; BT, bone trabeculae; rBMSC, rhesus marrow bone MSC; 3D-BG, 3D printed bioglass; BMP/CS, BMP-2 gene loaded nanoparticles; SO, sham-operated; UC-MSCs, umbilical cord mesenchymal stem cells; PLGA, poly(lactic-co-glycolic acid); NBG, new bone growth; Undiff, Undifferentiated; Diff, Differentiated; ICG, iliac crest cancellous bone graft; bHA, bovine hydroxyl apatite/collagen; RD, remaining defect; β-TCP, Beta tricalcium phosphate; RME, rapid maxillary expansion; VE, vertical height; PRF, platelet rich fibrin; BMD, bone mineral density, K₂HPO₄, Dipotassium hydrogen phosphate; EPC, endothelial progenitor cell; FG, fibrin glue; co-MSC, co-cultured MSC; monoMSC, mono-cultured MSC; TMD, tissue mineral density; PF127, pluronic F127; advBMP-2, Adenovirus BMP-2; TBV, total bone volume; NB, new bone; CAP, calcium phosphate.

Table 2.

Animal Studies of Cleft Palate Defect.

No	Study	Model	Number	Cell Source	Scaffold	Growth Factor	Experimental group	Control Group	Results
1.	Abe 2020	Beagle dogs (cleft lip and palate)	1	Autologous (BMSC)	CAP	n.a	2) CAP/BMSC	1) Scaffold (CAP)	X-Ray (3 months) Group 1: residual CAP was still detected Group 2: almost no granular opacity, bone bridge structure was present
2.	Naudot 2020	Sprague Dawley rats (Janvier Labs, Le Genest-Saint-Isle, France) (critical sized-cleft palate)	27	Allogenic (BMSC)	Alginate-based hydrogel scaffolds	n.a	2)Scaffold 3)Scaffold/ BMSC	1) Empty control	Histology (12 weeks) Group 1: nonmineralized healing connective tissue, a few blood vessels, and mature bone at defect margin. Group 2: Defect full of fibrous tissue. Group 3: New bone formation in the center of the defect Micro-CT (12 weeks) Group 1: NBF 36.91 ± 5.132%. Group 2: NBF 61.01 ± 5.288%. Group 3: NBF 17.24 ± 6.886%.
3.	Amalraj 2017	Wistar albino rat pups (cleft palate induced by Triamcinolone acetonide	12	Allogenic (BMSC donor Wister albino female rat)	PLGA	n.a	2) PLGA 3)PLGA/BMSC	1)No cell trasplant	Group 3: Complete reconstruction of the cleft palate in the group 3 of rat pups which received BMSCs along with PLGA scaffold. Bone growth in the cleft defect was faster
4.	Liceras-Liceras 2017	New Zealand white rabbits (cleft palate defect)	12	Autologous (ASC)	Fibrin-agarose	n.a	2)ASC/ Fibrin-agarose	1a)Positive control (no surgical procedure of the palate) 1b)Negative control/Blank control (cleft palate was left untreated)	CT Palate bone length (right side) Group 1a: 50.15 ± 0.41% Group 1b: 38.57 ± 1.74% Group 2 :(1 month): 48.21 ± 0.44% Group 2: (4 months): 47.21 ± 0.52% Palate bone width (right side) Group 1a: 51.22 ± 0.02% Group 1b: 46.15 ± 0.15% Group 2: (1 month): 50.55 ± 0.24% Group 2: (4 months): 52.51 ± 1.12% Right side = operated side

Abbreviations: BMSC, bone marrow stem cells; CAP, calcium phosphate; n.a, not applicable; PLGA, poly(lactic-co-glycolic acid); ASC, adipose stem cell.

In the sections below, we will discuss outcomes of individual studies and group meta-analyses sometimes in terms of “better, more, or higher levels.” These statements should be regarded as qualitative and indicative, but certainly not as being statistically relevant. Nevertheless, we thought it important in which direction the differences between cell-based and control reconstructions headed, even though we realize ourselves that this is maybe not scientifically correct, but rather “telling.”

Synthesis of Results

Alveolar Cleft

A total of 21 articles have provided information on cell-based tissue engineering in the alveolar cleft animal model. Six types of animals were used namely rat, rabbit, pig, dog, goat, and monkey. Genetically, cell transplantation was comprised of 4 types (autologous, allogenic, syngenic, and xenogenic). Cell sources were bone marrow mesenchymal stem cells from animals ((rat,^27–33 dog,^34–37 pig, ³⁸ monkey ³⁹ ) n = 13), umbilical cord mesenchymal stem cells from human^40–42 (n = 3) or animal (pig, ⁴³ n = 1), human differentiated gingiva derived mesenchymal stem cells ⁴⁴ (n = 1), dog adipose stem cells^45,46 (n = 2), and finally human mesenchymal stem cell from orbicularis oris muscle ⁴⁷ (n = 1).

Five types of scaffolds were applied, namely ceramics, synthetic polymers, natural polymers, autologous bone, or without any scaffold. Four articles used a single type of ceramics scaffold.^28,34,36,37 Four articles used a single type of synthetic polymer scaffold.^38,39,43,44 Two articles used a single-type natural polymer scaffold,^30,33 one article used autologous bone, ³⁵ and one article did not use any scaffold in its study. ³¹ Five articles used a combination of ceramics and natural polymers,^{27,29,40–42} 3 articles used a combination of at least two types of ceramic scaffolds,^32,45,46 and only one article used three types of ceramic scaffolds separately. ⁴⁷

Two types of growth factors were applied, namely BMP-2 and PRF. Two articles used BMP-2,^40,44 1 article used BMP-2 gene-loaded nanoparticles, ³⁹ 1 article used adenovirus BMP-2, ³⁸ and 1 article used PRF. ³⁵ The remaining 16 articles in this group did not use growth factor in their study.^{27–34,36,37,41–43,45–47}

All studies reported the osteogenic potential as an outcome parameter. Still, we only focused on the outcome results based on histology, histomorphometry, CBCT, and/or CT-Scan analysis. One study expressed a higher level of bone formation with the cell-only application for alveolar cleft reconstruction. ³¹ Nine studies showed a trend towards higher bone formation for alveolar cleft reconstruction with cell + scaffold combination.^{27,30,32,33,36,41–43,47} Five studies showed that the combination of cell + scaffold showed similar levels of alveolar cleft reconstruction compared to the control group.^{28,29,34,37,45} Five studies expressed more bone formation for alveolar cleft reconstruction with cell + scaffold + GF combination compared to control conditions.^{35,38–40,44}

Cleft Palate

A total of four articles have provided information on cell-based tissue engineering in cleft palate animal models. Three animal groups were used, namely rat,^17,48 dog, ⁴⁹ and rabbit. ¹⁶ Cell transplantation was comprised of only two types (autologous and allogenic). Bone marrow was the sole source of mesenchymal stem cells in dogs and rats (n = 3),^17,48,49 whereas the rabbit model applied MSCs from adipose tissue (n = 1). ¹⁶ Four types of scaffolds were applied, namely calcium phosphate (n = 1), ⁴⁹ alginate-based hydrogel scaffolds (n = 1), ⁴⁸ poly(lactic-co-glycolic acid) (n = 1), ¹⁷ and fibrin-agarose (n = 1). ¹⁶ In this group, there were no growth factors applied.

The osteogenic potential was assessed as the primary outcome parameter in all studies. One study expressed more bone formation clinically with cell + scaffold application for cleft palate reconstruction. ¹⁷ Three studies described higher bone formation levels for cleft palate reconstruction with cell + scaffold combination compared to scaffold only conditions.^16,48,49

Meta-Analysis

Figure 2A is the forest plot of the meta-analysis of the percentage of new bone volume formation as assessed with histomorphometry analysis of autograft vs. cells-loaded scaffold group in alveolar cleft dog and rat models. In the dogs’ group, one study reported higher new bone formation in the scaffold + cell group compared to the autograft group. ³⁴ Two studies reported higher new bone formation in the autograft group compared to the scaffold + cell group.^37,46 These studies showed a standard mean difference (SMD) of −3.14 [95%CI (−28.67, 2.39), P = .81, with heterogeneity I² = 93%]. In the rats’ group, one study reported higher new bone formation in the autograft group, ⁴⁷ and one study reported similar bone formation results of autograft and scaffold + cell group ⁴⁷ SMD of −8.11 [95%CI (−30.73,14.50), P = .48, with heterogeneity I² = 58%]. Although far from significant, autograft was favored over scaffold + cell combination with a SMD of −5.11 [95% CI (−22.57, 12.36), P = .57, with heterogeneity I = 88%]. There was no statistically significant difference after subgroup analysis, indicating that the subgroup did not contribute to heterogeneity.

Figure 2.

Forest plots for the new bone volume formation (%) histomorphometry analysis alveolar cleft in dog and rat models: (A) autograft vs cells-loaded scaffold group; (B) scaffold-only group vs cells-loaded scaffold group; (C) blank control group vs cells-loaded scaffold group.

Figure 2B depicts the forest plot of the meta-analysis comparison of the scaffold-only group vs. the cell-loaded scaffold group in alveolar cleft dog and rat models, again as histomorphometrically assessed with % new bone volume formation as the outcome parameter. In the rat subset, four studies reported higher new bone formation in the scaffold + cell group compared to the scaffold-only group,^28,41,47 whereas 1 study reported higher new bone formation in the scaffold-only group compared to the scaffold + cell group. ²⁸ These studies showed a SMD of 4.74 [95%CI (−4.10,13.59), P = .29, with heterogeneity I² = 96%]. In the dogs’ group, one study reported the higher new bone formation of scaffold + cell group compared to scaffold only group ³⁷ SMD of 15.81 [95%CI (4.45, 27.17), P = .006]. The overall result, although not significantly, favored scaffold + cell over scaffold-only with a SMD of 6.49 [95% CI (−1.91, 14.88), P = .13, with heterogeneity I = 96%]. There was no statistically significant difference after subgroup analysis, indicating that the subgroup did not contribute to heterogeneity.

In Figure 2C, the meta-analysis of the histomorphometry assessment of the new bone formation of a blank control group vs cells-loaded scaffold group in alveolar cleft dog and rat models is depicted. In the rat subset, two studies reported higher new bone formation of blank control compared to the scaffold + cell group.^28,29 One study reported higher new bone formation of the scaffold + cell group compared to the blank control group. ⁴¹ These studies showed a SMD of −7.17 [95%CI (−17.94, 3.59), P = .19, with heterogeneity I² = 97%]. In the dog's group, one study reported the higher new bone formation of scaffold + cell group compared to the blank control group ³⁴ with a SMD of 65.58 [95%CI (58.88, 72.28), P < .00001].

The overall result, although not significantly, favored scaffold + cell over blank control SMD of 4.38 [95% CI (−15.28, 24.04), P = .66, with heterogeneity I² = 99%]. After subgroup analysis for animal species, a statistically significant difference was discovered, indicating that species subgroups contributed to heterogeneity.

Figure 3A depicts the forest plot of the meta-analysis addressing the remaining defect volume CT scan analysis of the scaffold-only group vs. the cells-loaded scaffold group in the alveolar cleft rat model. One study reported less remaining defect volume of scaffold + cell compared to the scaffold group. ³² The other study reported the opposite. ²⁹ Overall, scaffold + cell and scaffold only showed similar remaining defect volumes with a SMD of 0.03 [95%CI (−1.19, 1.24), P = .97, with heterogeneity I² = 58%].

Figure 3.

Forest plots for CT scan analysis in the alveolar cleft rat model: (A) the remaining defect volume of scaffold-only vs cells-loaded scaffold group; (B) the remaining defect volume of blank control vs cells-loaded scaffold group; (C) bone mineral density of blank control vs cells-loaded scaffold group.

The meta-analysis of the remaining defect volume CT scan analysis of the blank control group vs. cells-loaded scaffold group in the alveolar cleft rat model is given in Figure 3B. One study reported less remaining defect volume of scaffold + cell compared to the blank control group. ³² The other study showed the reverse effect. ²⁹ Overall, the blank control showed a slightly lower remaining defect volume than the scaffold + cell group, with a SMD of 0.41 [95% CI (−2.31, 3.13), P = .77, with heterogeneity I² = 66%].

The meta-analysis evaluating the bone mineral density CT scan analysis of blank control group vs. cells-loaded scaffold group in the alveolar cleft rat model is shown in Figure 3C. One study reported higher bone mineral density in the scaffold + cell group compared to the blank control group, ⁴² whereas the other study reported similar bone mineral densities in both groups.

The overall result showed a somewhat higher bone mineral density in the scaffold + cell group with a SMD of −0.42 [95%CI (−0.38, 1.22), P = .31, with heterogeneity I² = 99%].

Risk of Bias Within and Individual Studies

Figure 4 shows the overall results of the risk of bias assessment of the 25 studies included in this systematic review. Regarding selection bias item “sequence generation”, 48% of the studies were scored as “unclear risk”, 48% of the studies were scored as “low risk of bias”, and only 4% of the studies were scored as “high risk of bias”. All studies described that intervention and control groups were similar at the start of the experiment. Regarding the selection bias item “allocation concealment”, 48% of the studies were scored as “unclear risk”, 48% of the studies were scored as “low risk of bias”, and only 4% of the studies were scored as “high risk of bias”. In addition, 96% and 92% of the included studies were scored as unclear risk of bias concerning performance bias items ‘random housing’ and ‘blinding’, respectively. For the detection bias item ‘random outcome assessment’, 88% of the studies were scored as “unclear risk”. Only 28% of the included studies were scored as “low risk of bias” by outcome assessor-blinded. For attrition bias, 88% of the included studies scored as low risk of bias, as they adequately addressed incomplete outcome data. Overall, only 44% of the included studies were achieved as “low risk of bias” because it was stated in the studies that the experiment was randomized at any level and only 28% of the included studies were scored as “low risk of bias” because it was stated in the studies that the experiment was blinded at any level.

Figure 4.

Risk of bias graph & summary: review authors’ judgments about each risk of bias item presented as percentages across all included studies and as item for each included study.

Publication Bias

Since each meta-analysis consisted of less than ten studies and therefore lacked sufficient power to distinguish chance from real asymmetry, an assessment of publication bias via statistical testing or funnel plots was not performed. ²⁵

Discussion

Cleft lip and/or palate is one of the most common congenital malformations in the maxillofacial area and occurs in the setting of genetic and environmental factors. ⁶ Standard management of oral clefts including cleft palate and alveolar cleft surgery, has side effects that are often associated with post-operative results on the defect site or donor site. ⁵⁰ Clinicians and researchers have been working together to search for applicable stem cell-based tissue engineering to overcome these challenges.^14,15,51 Unlike alveolar cleft, stem cell-based tissue engineering technology for cleft palate is still in process for future clinical human application. ⁵² In addition, the application of new technologies for oral cleft treatments is often hampered by limited healthcare settings where many patients are left untreated until they reach adult age. ¹¹

Recently, a systematic review on alveolar bone tissue engineering in pre-clinical studies by Shanbhag et al reported: (1) the addition of osteogenic cells (MSCs or OB) to biomaterial scaffolds can enhance alveolar bone regeneration in small and large animal models; (2) Ex vivo BMP gene-transfer to MSCs and OB can enhance their in vivo osteogenic potential based on small animal models; (3) Bone tissue engineering may result in comparable alveolar bone regeneration as induced by autograft (limited evidence); and (4) Large heterogeneity between studies resulting from biological and methodological variability. ⁵³ However, most of the included studies (83.3%) used critical size defects in the mandible, where alveolar clefts do not occur. Only three included studies reported the use of maxillary cleft models. Therefore, we decided to update the results and focused on alveolar cleft and cleft palate pre-clinical models. A review by Alkaabi et al found that regenerative therapies showed better alveolar bone regeneration, although not significantly, compared to autogenous bone grafting on clinical application. ⁵⁴ However, this review could not conclude which type of regenerative therapy is the most optimal for alveolar bone grafting on clinical application.

In the present study, we performed a systematic review and meta-analysis of pre-clinical studies to evaluate the efficacy of stem cell-based tissue engineering for cleft palate and alveolar cleft defects. Twenty-five studies using stem cell-based tissue engineering technology were included, comprising 21 alveolar cleft animal studies^27–47 and 4 cleft palate animal studies.^16,17,48,49 Of these, 10 studies met the criteria to be included in the meta-analyses.^{28,29,32–34,37,41,42,46,47} Although only a relatively small number of studies could be included, it still enabled us to perform the meta-analyses and explore the effect of several subgroup variables. Despite this, there are some potential limitations related to this approach. First, as also addressed above, all experiments should preferably be performed in a similar manner when their results are being combined in a meta-analysis. However, the publications display experimental variability for the utilized animal species, defect type and size, the used cell types, the number of cells per defect, the biomaterials applied as cell carrier, the growth factor, the healing time after cell transplantation, and the result assessment parameters. Not surprisingly, substantial statistical heterogeneity was found. We performed subgroup analyses (animal species) in an attempt to tackle this issue, but this did not notably reduce the heterogeneity. We also conducted direct comparison of meta-analysis between control group (blank control, autograft, or scaffold without cell) versus stem cell-based tissue engineering group. In addition, we reported applications of stem cell-based tissue engineering for cleft palate reconstruction besides alveolar cleft. In the next paragraphs, these results will be discussed in more detail.

As shown in this systematic review, mesenchymal stem cells from bone marrow are the main used cell type for preclinical trials for both the alveolar cleft and cleft palate model. Another frequently used source of MSCs is adipose tissue. There is still controversy on which cell source has better osteogenic potential. Some say that bone marrow is better (eg, Musina ⁵⁵ 2006; Mohamed-Ahmed ⁵⁶ 2021, Brennan ⁵⁷ 2017), others state that adipose-derived MSCs may have higher osteogenic potential (Huang ⁵⁸ 2022; Holmes ⁵⁹ 2022) and some found similar osteogenic activities (Humenik ⁶⁰ 2022). In this regard, it should be kept in mind that variations in the distinctive features of both cell sources may depend on the source and method of isolation and epigenetic changes during maintenance and growth (Brown ⁶¹ 2019). Nevertheless, it would be worthwhile and fascinating to evaluate adipose stem cells for their efficacy in pre-clinical cleft models and subsequent clinical implementation.

In both the alveolar cleft and cleft palate groups, small and large animals were used. Small animal models can provide “proof of principle” and large animal models can be used to represent the efficacy of pre-clinical testing. ⁵³ In one meta-analysis, greater but not significant bone formation was observed in the cell-loaded scaffold group vs scaffold-only group for the alveolar cleft reconstruction of rats and dogs. Strikingly, the dog studies showed not only more efficient better bone formation compared to scaffold only, but also similar ³⁷ to superior bone formation ³⁴ compared to autografts. These interesting results show, at least preclinically, that regenerative grafts have equal or higher bone regeneration efficacy in comparison with autografts, and imply that regenerative grafts may be full-blown, suitable alternatives for the golden standard, which is still autologous bone.

From our risk of bias assessments, we had to conclude that the animal studies suffered from many unclarities and high risk of bias in their publications. Key measures to avoid bias, such as randomization and blinding, were infrequently reported. For example, only 44% of the studies provided sufficient details to judge the adequacy of the method of randomization, and only 28% of the studies reported that the outcome assessment was blinded. Moreover, the results of the meta-analyses may be subject to publication bias from non-publication of negative results, true study heterogeneity or differences in study quality, which unfortunately statistical assessment with funnel plots was not conducted in this study because meta-analysis was consisted of less than 10 studies to confirm this. Nevertheless, the combined analysis of the included studies still generated extra and valuable information that could not be derived from individual studies. ²⁴ To generate reliable and unbiased data, it is suggested that the standards of animal experiment reporting should be more like the standards routinely applied in human randomized controlled trials. ²⁴ Also, standardization of follow-up periods may help reduce the enormous spread in post-operative monitoring points and maximum follow-up date, which now ranges from 6 weeks (42 days) to 6 months (180 days).

Although histomorphometry is considered the “gold standard” for the evaluation of bone structure, ⁵³ our study assessed bone regeneration using histology, histomorphometry, CBCT, and/or CT-Scan analysis with new bone formation, remaining defect or bone mineral density as outcome parameters. Recently, micro-computed tomography (micro-CT) has been proposed as an alternative method for assessing three-dimensional bone microarchitecture with high resolution and accuracy, in a fast and nondestructive manner. ⁵³ However, care should be taken when interpreting outcomes of CT or micro-CT because of the difficulties in differentiating between mineralized scaffolds and newly formed bone. ⁵³ In this regard, Prins and coworkers ⁶² showed that by varying threshold values in CT evaluations, it may still be possible to distinguish between both mineralized entities. In addition, this publication showed that it may be very useful to combine both methods, since it offered a mutual confirmation of the one method by the other. ⁶²

Defect size also influences the clinical application of cell-based tissue engineering. Unlike calvarial critical-size defects, alveolar critical-size defects models have not been well characterized in the literature regarding defect location, size, and morphology. Defect dimensions varied between studies for the same animal model/species. In many cases, the selection of a particular model appeared to be based on one previously established by the same or related research group(s). ⁵³

It is tempting to compare data obtained from pre-clinical and clinical studies to conclude the validity and feasibility of extrapolation of pre-clinical outcomes for the prediction of efficacy in clinical models. However, clinical studies employing cellular therapies for alveolar cleft are scarce. This scarcity of pediatric cell-based studies is a more general phenomenon, which has been covered extensively by Nitkin et al ⁶³ The most important issue is, and should be, thorough consideration of the ethical aspects for this vulnerable population. As also indicated above, a recent review by Alkaabi and co-workers addressed the use of regenerative grafts for alveolar cleft repair, including cell-based therapies. ⁵⁴ Still, unfortunately, the studies listed there used different cell preparations than those addressed in this review. ⁵⁴ So, for cleft studies, extrapolation from pre-clinical results to clinical implementation remains an issue nowadays.

Despite the limitations mentioned above, the results of this systematic review and meta-analysis revealed that cell-based approaches are favorable for alveolar cleft and cleft palate reconstructions. These are displayed by the positive effect of cell-based approaches on new bone formation, remaining defect volume, and bone mineral density. The meta-analysis did not show a statistically significant difference in osteogenic potential between the control group (blank control, autograft, or scaffold without cell) versus the stem cell-based tissue engineering group for in vivo alveolar cleft reconstruction. As for cleft palate reconstruction, limited result data hampered the meta-analysis to be performed.

In perspective, meta-analyses of animal studies tend to be exploratory rather than confirmatory. Standardization of alveolar cleft and cleft palate models to better represent the clinical scenario and standardization of study reporting should be essential considerations in future studies of alveolar and palate bone tissue engineering. Another issue, although slightly beyond the scope of this review, is that in most of the included preclinical studies also osteogenic peptides and recombinant growth factors are being used in combination with the regenerative cell populations, whereas in particular in pediatric cleft repair these stimulatory compounds are still not clinically implemented except in clinical trials. For example, the application of BMP-2 is still debated: In recent reviews Fisher et al ⁶⁴ advocate the use of BMP-2 to decrease donor site morbidity or when alternatives are contraindicated, whereas Sales et al, ⁶⁵ in particular based on high risk of bias in studies, conclude that recommendations to use BMP-2 in pediatric populations should be treated with caution. In our view, given the data presented in the latter review showing equal bone formation in BMP-2 vs. autologous bone treatment, avoiding iliac crest surgeries may be an important factor in reducing pediatric patients risks, as long as the high dosages causing major adverse events like in spinal surgeries ⁶⁶ are not applied. An alternative from our own experience may be ex vivo stimulation of regenerative (stem) cells with physiological dosages of rhBMP-2, thus avoiding body exposure to BMP-2 at all. ⁶⁷ Nevertheless, we advocate well-designed studies with cell-growth factor combinations to be evaluated for alveolar cleft repair, to accelerate clinical implementation of these potent candidates.

Further more extensive and prospective studies with greater methodological aspects and rigor in data collection, analysis, and reporting, as well as long-term post-operative follow-up periods with information on complications, are needed. Most importantly, the animal models presented in this systematic review were all fresh acute models, except for one study was conducted in rabbit models by creating a pseudo-cleft palate defect ¹⁶ and one study was conducted by injecting Triamcinolone acetonide (TAC) in pregnant rats. ¹⁷ In our view, the latter model properly reflects the real situation appropriately by creating chronic alveolar cleft/cleft palate defects, proper for regenerative medicine.

Conclusion

Alveolar cleft and cleft palate reconstructions using regenerative grafts are currently still in its infancy, and have so far not resulted in clear data about efficacy, in contrast to other craniofacial bone defect areas. The models used seem inadequate to reflect the human situation due to their non-chronic induction of the clefts, and uncertainty about whether critical size defects are being created. The Triamcinolone acetonide model is very promising in that regard and should probably be used as the new standard model for pre-clinical studies on cleft defects.