“…Given the already small pool of available patients, researchers often cannot construct the stringent exclusion criteria necessary to control for confounding factors. Of the 20 studies involving wound fluid or ulcer biopsies, 17 did not exclude patients with diabetes (Harris et al, 1995;Higley et al, 1995;Hasan et al, 1997;Li et al, 1998;Mendez et al, 1999;Trengove et al, 2000;Kim et al, 2003;Lal et al, 2003;Tian and Li et al, 1998;Wallace and Stacey, 1998;Trengove et al, 2000;Cowin et al, 2001Cowin et al, , 2006Murphy et al, 2002;Tian and Stacey, 2003;Gohel et al, 2008). This weakness is especially important in studies on TNF-a, given its known positive association with the severity of active infection (Ambrosch et al, 2008).…”