Serum Procalcitonin for Predicting Significant Infections and Mortality in Pediatric Oncology

Vinod Gunasekaran1, Nita Radhakrishnan, Veronique Dinand

  • 1Pediatric Hematology Oncology and Bone Marrow Transplantation unit, Sir Ganga Ram Hospital, New Delhi, India. Correspondence to: Dr Anupam Sachdeva, Head of Department, Institute of Child Health, Sir Ganga Ram Hospital, Rajinder Nagar, New Delhi - 110 060, India. anupamace@yahoo.co.in.

Indian Pediatrics
|November 28, 2016
PubMed

Insights

Serum procalcitonin (PCT) can predict serious infections and mortality in febrile children undergoing chemotherapy. However, its low sensitivity limits its use in guiding clinical decisions.

Area of Science:

  • Pediatric Oncology
  • Infectious Diseases
  • Biomarkers

Background:

  • Febrile neutropenia is a common concern in children undergoing chemotherapy.
  • Early identification of significant infections is crucial for timely treatment and improved outcomes.

Purpose of the Study:

  • To assess the predictive value of serum procalcitonin (PCT) levels upon admission for significant infections and mortality in febrile pediatric oncology patients.

Main Methods:

  • Retrospective analysis of 821 febrile episodes in 316 children on chemotherapy.
  • Defined clinically significant infections (CSI) and microbiologically documented infections (MDI) using standard criteria.
  • Analyzed the association of PCT levels with CSI, MDI, and mortality.

Main Results:

  • Serum procalcitonin (PCT) levels were significantly higher in episodes with CSI and MDI.
  • PCT ≥0.7 ng/mL demonstrated optimal prediction for CSI (AUC-0.740) and MDI (AUC-0.636).
  • A PCT level ≥5 ng/mL was associated with a 7.1-fold increased risk of mortality, though sensitivity was poor (45-55%).

Conclusions:

  • Serum procalcitonin (PCT) is a valuable predictor of significant infections and mortality in pediatric oncology patients with fever.
  • Despite its predictive value, the poor sensitivity of PCT necessitates caution when using it to guide clinical decisions.
Abstract