C-reactive protein in the diagnosis of postoperative infection in pediatric patients: a prospective observational study of 103 patients

Abstract Background Surgical stress elicits a systemic inflammatory response syndrome that contributes to the development of septic complications. C-reactive protein (CRP) is an acute phase protein released in inflammatory states. To analyze the usefulness of CRP as a marker of infection in surgical pediatric patients, we analyzed its kinetics in these patients in comparison with usual markers, such temperature, leukocytes, or fibrinogen. Methods One hundred three pediatric patients undergoing major surgery were enrolled consecutively in this observational study. C-reactive protein, leukocytes, neutrophils, platelets, fibrinogen, glycemia, and temperature were determined daily after surgery. Patients were classified as infected or not infected. Sensitivity, specificity, positive predictive value, negative predictive value, efficiency, precocity, positive likelihood ratio, and number of subjects to be treated were calculated. Results Peak in CRP was detected at 48 ± 24 hours. C-reactive protein was higher in the infected group from the first day, with significant differences between groups from the second day. Best cutoff for detecting infection was increases in CRP of 11 mg/dL in 48 hours, with a sensitivity of 87%, specificity of 89%, precocity of 1.7 days (0.82-2.54), number of subjects to be treated of 1.7 (1.4-2.6), and positive likelihood ratio of 7.9. Application of other markers alone or in combination did not surpass the sensitivity, specificity, or precocity for increases in CRP of 11 mg/dL in 48 hours. Conclusions Increases of CRP constitute an easy and cheap prognostic alert system and may be used to establish strategies aimed to detect infection in surgical pediatric patients.