What was done
A single-center randomized trial at Government Dental College and Hospital, Ahmedabad, registered in the Clinical Trials Registry-India as CTRI/2025/04/085645, compared leukocyte platelet-rich fibrin (L-PRF) against mineral trioxide aggregate (MTA Plus, Prevest DenPro) as direct pulp capping agents. Twenty permanent molars in adults aged 18 to 45 with deep caries, a diagnosis of reversible pulpitis, a carious exposure of 0.5 to 1.0 mm, and bleeding controlled within five minutes with 2.5% sodium hypochlorite were allocated 1:1 using sealed opaque envelopes. The PRF group received a compressed fibrin membrane spun by Choukroun’s protocol (3,000 rpm for 10 minutes from the patient’s own blood); the MTA group received roughly 2 mm of cement. Both groups were sealed identically with a resin-modified glass ionomer liner and composite. One calibrated operator treated every tooth, and the CBCT scans were read by two blinded assessors with an intraclass correlation above 0.85.
What six months showed
Nineteen of 20 teeth completed follow-up. On the primary endpoint, a single CBCT scan at six months, the PRF group showed a mean dentin bridge volume of 0.01032 cc (SD 0.00341) against 0.00778 cc (SD 0.00120) for MTA: a difference of 0.00254 cc, about 2.5 cubic millimeters, reported at p = 0.049. On the clinical endpoints the groups did not separate at all: every retained tooth in both groups (10 of 10 PRF, 9 of 9 MTA) responded normally to cold and electric pulp testing at three and six months. One MTA tooth developed irreversible pulpitis at three months and went to root canal treatment. One PRF tooth showed periodontal ligament widening at six months while remaining asymptomatic and responsive.
The primary result is thinner than it looks
Three things argue for holding the headline number loosely. First, p = 0.049 sits at the threshold. Second, the methods state that the volume data were non-normal by Shapiro-Wilk and were compared with a Mann-Whitney U test, but Table 1 of the paper reports a t value of 2.118, which is a parametric statistic; those two tests are not the same, and at 19 teeth they can land on opposite sides of 0.05. The paper does not say which test actually produced 0.049. Third, the analysis was complete-case: the one MTA tooth that failed was excluded, so the volume comparison is among survivors. The trial was also powered on an expected effect size of d = 1.38 taken from a systematic review of pulpotomy (Noor Mohamed et al. 2018), a different procedure from direct pulp capping, which is how a 20-tooth trial comes to count as fully powered.
Where we differ from the paper’s own summary
The conclusions state that PRF demonstrated “superior dentine bridge formation and improved clinical outcomes relative to MTA.” The trial’s own tables support the first half and not the second: clinical outcomes among retained teeth were identical (100% versus 100%, reported at p = 0.99), and the single MTA failure is 1 of 10, an event the discussion itself declines to attribute to the material. The defensible claim is narrower: in adult molars with carious exposures, PRF produced a borderline larger radiographic bridge volume at six months with clinical performance indistinguishable from MTA. No claim about failure rates, and nothing about durability, survives contact with the sample size.
What this adds to the pulp repair route
This is a patient-level trial in adults, and it lands on a now familiar pattern for platelet concentrates in this route: an early hard-tissue or maturation signal on imaging, with clinical endpoints that do not separate. The controlled animal record showed no advantage of platelet derivatives over a plain blood clot (/analysis/platelet-derivatives-versus-blood-clot/), and the one 80-month human case we have logged showed early radiographic advantage failing to predict long-term pulp status (/analysis/prf-versus-blood-clot-80-months/). A 2.5 cubic millimeter difference in bridge volume at six months is a biological observation worth having; whether it means anything for tooth survival is exactly what a six-month window cannot say. The pulp repair program page is at /programs/pulp-dentin-repair/ and the current state of the field is at /field/.
Provenance: every claim above traces to the published trial report at the DOI cited, per our method at /method/.