Fourteen trials (1,646 participants, range 17 to 382) were included. Randomisation and allocation concealment were unclear or absent in many trials. Only one trial had an adequate sample size calculation and none reported an adequate ITT analysis.
Most trials showed a statistically significant benefit of the growth factor product for complete healing, time to complete healing or both. There was no statistically significant benefit of becaplermin compared with the extracellular wound matrix OASIS or BFGF compared with placebo (results not shown). Rates of adverse events were similar between groups.
A meta-analysis of four trials indicated that Dermagraft significantly increased the odds of complete wound closure compared with standard wound care (OR 2.65, 95% CI 1.11 to 6.30). Statistical heterogeneity was significant (Ι²= 67%, p=0.03).