Diagnosis of infection in patients undergoing extracorporeal membrane oxygenation: a case-control study.

OBJECTIVE

Diagnosis of infection in patients receiving extracorporeal membrane oxygenation is challenging in clinical practice but represents a crucial aspect of the upgrading of therapeutic options. The aim of this study was to analyze the role of C-reactive protein and procalcitonin in the diagnosis of infection in patients requiring extracorporeal membrane oxygenation and to assess the difference between venovenous and venoarterial extracorporeal membrane oxygenation settings.

METHODS

A case-control study was performed on 27 patients. Serum values of procalcitonin and C-reactive protein were analyzed according to the presence of infection.

RESULTS

Forty-eight percent of patients had infection. Gram-negative bacteria were the predominant pathogens, and Candida albicans was the most frequent isolated microorganism. Procalcitonin had an area under the curve of 0.681 (P = .0062) for the diagnosis of infection in the venoarterial extracorporeal membrane oxygenation group but failed to discriminate infection in the venovenous extracorporeal membrane oxygenation group (P = .14). The area under the curve of C-reactive protein was 0.707 (P < .001) in all patients receiving extracorporeal membrane oxygenation. In patients receiving venoarterial extracorporeal membrane oxygenation, procalcitonin had good accuracy with 1.89 ng/mL as the cutoff (sensitivity = 87.8%, specificity = 50%) and C-reactive protein with 97.70 mg/L as the cutoff (sensitivity = 85.3%, specificity = 41.6%). The procalcitonin and C-reactive protein combined assay had a sensitivity of 87.2% and specificity of 25.9%. Four variables were identified as statistically significant predictors of infection: procalcitonin and C-reactive protein combined assay (odds ratio, 1.184; P < .001), age (odds ratio, 0.980; P < .001), presence of infection before extracorporeal membrane oxygenation implantation (odds ratio, 1.782; P < .001), and duration of extracorporeal membrane oxygenation support (odds ratio, 1.056; P < .001).

CONCLUSIONS

Traditional and emerging inflammatory biomarkers, especially if compounded in the procalcitonin and C-reactive protein combined assay, can aid in the diagnosis of infection in patients undergoing venoarterial extracorporeal membrane oxygenation.