What was tested
A.T. Elbaram (Dijlah University, Baghdad), S.A. Sami (Al-Hikma University College, Baghdad) and Z.F. Al-Sultani (Al-Qadisiyah University, Al-Diwaniyah) ran a head-to-head in vivo study of the two best-characterized dental mesenchymal cell populations: dental pulp stem cells (DPSCs) and periodontal ligament stem cells (PDLSCs), both from extracted human premolars, seeded at 1 x 10^6 cells per scaffold onto collagen-hydroxyapatite scaffolds and implanted into experimental defects in rabbit teeth and supporting tissues. The paper appeared 30 September 2026 in Babcock University Medical Journal (9(3):379-388), open access, with no declared funding and ethics approval from Dijlah University (No. 17, 19 January 2025). Twenty-four white rabbits, 7 to 12 months old and 2.5 +/- 0.4 kg, were allocated by lottery to four groups of six: DPSC-seeded scaffold, PDLSC-seeded scaffold, cell-free scaffold, and empty defect. Outcome assessors were blinded; there was no a priori power calculation. The cells came from premolars extracted for orthodontic reasons, pooled across donors, digested with collagenase type I and dispase, and used only if they showed the mesenchymal profile (CD73, CD90, CD105 positive; CD34, CD45 negative). The DPSC cultures reported met that profile: CD73 98.2 +/- 1.5%, CD90 96.7 +/- 2.1%, CD105 94.3 +/- 3.2%, CD34 0.8 +/- 0.3%, CD45 0.5 +/- 0.2%. Animals were sacrificed at 8 weeks.
What each cell population did
By histology, the DPSC-seeded defects contained organized pulp-like connective tissue, vascular structures, and odontoblast-like cells lining dentinal walls, while the PDLSC-seeded defects showed periodontal ligament-like fiber organization, new cementum, and increased alveolar bone. The quantitative readouts split by marker, not by a shared endpoint. Dentin sialophosphoprotein (DSPP) immunostaining, scored as an H-score, was significantly higher in the DPSC group (245.6 +/- 32.1) than in the scaffold-only (78.3 +/- 18.5) and empty-defect (42.1 +/- 12.3) groups, p < 0.001 by one-way ANOVA with Tukey post-hoc testing; VEGF staining was more abundant around the new vessels in the DPSC group but was reported qualitatively. On micro-computed tomography, bone volume fraction was significantly higher in the PDLSC group (BV/TV 34.2 +/- 5.1%) than in the scaffold-only (18.7 +/- 4.3%) and empty-defect (12.1 +/- 3.8%) groups, p < 0.05. No BV/TV value is reported for the DPSC group, and no DSPP value is reported for the PDLSC group, so each cell type is quantified on the marker that favored it. Scaffold-only and empty defects regenerated least, which establishes that the cells, not the collagen-hydroxyapatite carrier, drove the visible repair.
What the study does not show
The authors list the limits themselves: a rabbit model carrying human cells, six animals per group without a power calculation, an 8-week follow-up, no functional testing (no pulp vitality, no mechanical loading), no neural markers, donor cells pooled so donor-to-donor variability is invisible, no characterization of the xenogeneic immune response to the human grafts, and a single cell density on a single scaffold composition. Two further flags a reviewer should raise: the published PDF still contains a leftover editing placeholder, “[cite 2-3 relevant papers],” in the sample-size paragraph, and the H-score quantitation is never described in the Methods. Treat the exact effect sizes as provisional. Nothing here is a third-dentition result: this is repair of existing pulp and periodontal tissues by grafted adult stem cells, not generation of a new tooth, and the endpoint is structural histology weeks after surgery, not a functioning dentition.
Where it sits in the field
The piece belongs to the pulp and dentin repair program (/programs/pulp-dentin-repair/), which holds tier 3 on small human data; this adds an animal datapoint under it and moves nothing on the ladder. The PDLSC half doubles as root-supporting tissue biology for the root-organoids program (/programs/root-organoids/), in the same vein as the recent PDLSC senescence result this pass covered. Its practical content is cell-selection logic: the two cell types were not interchangeable in this model, and the scaffold alone did little, which is a real if modest design input for anyone planning graft-based pulp or periodontal repair. The current field pass (/field/) notes that everything upstream of a person remains animal or in vitro evidence; this study is squarely in that band, and the authors themselves call for quantitative, longer-term work before any clinical application.
Where we differ from the coverage
We found no press or popular coverage of this paper to differ from. Against the paper’s own framing, though, “comparative” overstates the design: DPSCs and PDLSCs were never tested against each other on a single shared endpoint, so the tidy split, pulp cells for pulp, ligament cells for the periodontium, is a reasonable reading of the histology but is not established by any direct statistical comparison. The conclusion that cell choice should track the tissue to be regenerated is a hypothesis this study motivates, not one it proves.
Provenance: grounded in the full open-access text of Elbaram, Sami and Al-Sultani (2026), Babcock University Medical Journal 9(3):379-388, DOI 10.38029/babcockuniv.med.j…v9i3.1848, read in full from the journal’s PDF; every number above was checked against that text. Method and sourcing standard at /method/.