Procalcitonin and Cellulitis: Cohort Study of the Correlation of Procalcitonin Blood Levels with the Severity and Outcome in Patients with Limb Cellulitis

Background: Procalcitonin is a biochemical marker which is raised in bacterial infections. It has been used to assist in determining whether or not a patient will benefit from antibiotic therapy. Cellulitis is a common infection of the skin and is treated with a variety of antibiotics; none of which have been shown to be superior to any other. Within the context of a clinical trial we wished to examine the correlation between clinical observations, blood tests and local measurements of skin damage over the first 14 days following antibiotic therapy and to relate baseline levels of procalcitonin to time to recovery.

Methods: Blood was drawn for procalcitonin levels at Baseline, Day 5 and Day 10 following consent to enter the trial. Patients were followed up by telephone consultation up to approximately 40 days. Procalcitonin levels (logarithm) were statistically correlated (Pearson correlation) with a wide array of clinical and laboratory measurements. We also correlated baseline levels of procalcitonin with pain scores (VAS) and the return of the patient to their normal activities.

Results: 120 patients with 122 baseline sets of data provided 314 procalcitonin values which were correlated with 20 variables (including pulse, blood pressure, core temperature, affected skin area, CRP, neutrophils, lymphocytes) and one derived variable (neutrophil:lymphocyte ratio).

The mean (STD) baseline procalcitonin value was 1.67μg/L (9.4) reducing to 0.42 (1.49) at Day 5 and to 0.09 (0.15) at Day 10. However, 66% of patients had levels below 0.25μg/L at baseline, which is generally regarded as a cut-off value for initiating antibiotic therapy.

Procalcitonin was strongly correlated with CRP (0.574), neutrophil:lymphocyte ratio (0.567), total affected skin area (0.537) but was poorly correlated with systemic observations, apart from temperature (0.321).

High values were seen in patients with poor renal function.

Conclusion: Procalcitonin has little value in cellulitis over and above more commonly used measurements of severity. The low baseline procalcitonin values suggest that cellulitis is predominantly a toxin-mediated infection and that procalcitonin should not be used as a decisive test in cellulitis. Procalcitonin levels are uninterpretable in patients with impaired renal function.