Granulocyte-colony stimulating factor, a growth factor, which can be added to usual therapy for treating foot infections in people with diabetes

People with diabetes can develop foot infections which can be difficult to treat, and treatment failure can result in lower extremity amputation. We found five trials which included a total of 167 people. The trials showed that adding G-CSF to usual therapy did not significantly affect the likelihood of the infection resolving or the improved healing of foot wounds, nor did it reduce the period of treatment with oral antibiotics. However, G-CSF does appear to reduce the need for surgical interventions, especially amputations, and the number of days spent in hospital. There are limitations to this analysis related to the variations in the people included in the studies (e.g. the severity of infection, the timing of the clinical assessment, the use of different G-CSF preparations, and for different lengths of time). Therefore caution is required in the interpretation of the findings.

Authors' conclusions: 

The available evidence is limited, but suggests that adjunctive G-CSF treatment in people with a diabetic foot infection, including infected ulcers, does not appear to increase the likelihood of resolution of infection or healing of the foot ulcer. However, it does appear to reduce the need for surgical interventions, especially amputations, and the duration of hospitalisation. Clinicians might consider adding G-CSF to the usual treatment of diabetic foot infections, especially in patients with a limb-threatening infection, but it is not clear which patients might benefit.

Read the full abstract...

Granulocyte-colony stimulating factor (G-CSF) increases the release of neutrophil endothelial progenitor cells from the bone marrow and improves neutrophil functions, which are often impaired in people with diabetes.


To examine the effects of adjunctive G-CSF compared with placebo or no growth factor added to usual care on rates of infection, cure and wound healing in people with diabetes who have a foot infection.

Search strategy: 

In March 2013, for this second update, we searched the Cochrane Wounds Group Specialised Register (searched 14 March 2013); the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2013, Issue 2); Ovid MEDLINE (1948 to March Week 1 2013); Ovid EMBASE (1974 to 2013 March 13); Ovid MEDLINE (In-Process march 13,2013); and EBSCO CINAHL (1982 to 28 February 2013).

Selection criteria: 

Randomised controlled trials (RCTs) that evaluated the effect of adding G-CSF to usual care in people with a diabetic foot infection.

Data collection and analysis: 

Three review authors independently assessed trial eligibility, methodological quality and extracted data. We reported risk ratio (RR) or, for continuous outcomes, mean differences (MD), with 95% confidence intervals (CI). In the case of low or no heterogeneity we pooled studies using a fixed-effect model.

Main results: 

We identified and included five eligible trials with a total of 167 patients. The investigators administered various G-CSF preparations, at different doses and for different durations of time. Adding G-CSF did not significantly affect the likelihood of resolution of infection or wound healing, but it was associated with a significantly reduced likelihood of lower extremity surgical interventions (RR 0.38; 95 % CI 0.21 to 0.70), including amputation (RR 0.41; 95 % CI 0.18 to 0.95). Moreover, providing G-CSF reduced the duration of hospital stay (MD -1.40 days; 95% CI -2.27 to -0.53 days), but did not significantly affect the duration of systemic antibiotic therapy (MD -0.27 days; 95% CI -1.30 to 0.77 days).