The statistical performance of the DISCERN-FN "appears to be better than that of all previously published CDRs," Dr. Francois Dubos of the University of Lille and colleagues note in The Lancet Child & Adolescent Health.
"The specificity is good enough to permit safe treatment de-escalation for at least one-third of the children with febrile neutropenia," they report.
The derivation cohort included 539 febrile neutropenia episodes (270 in patients with blood cancer and 269 in patients with solid tumors).
Significant variables introduced into the decision tree were cancer type (solid tumor vs. blood cancer), age, high-risk chemotherapy, level of fever, C-reactive protein concentration at 24 to 48 hours after admission, and leucocyte and platelet counts and procalcitonin (at admission and at 24 to 48 hours later).
In the derivation set, sensitivity of DISCERN-FN was 98%, specificity was 56% and the negative likelihood ratio 0.04.
In the validation set, 1,806 febrile neutropenia episodes were analyzed, of which 332 (18%) were linked with severe infection.
In the validation set, DISCERN-FN had a sensitivity of 95%, specificity of 38% and negative likelihood ratio of 0.13.
The CDR reduced the risk of severe infection to a post-test probability of 0.8% (95% CI, 0.2 to 2.9) in the derivation set and 2.4% (95% CI, 1.5 to 3.9) in the validation set.
"The DISCERN-FN was generated from a large number of patient episodes, included age and procalcitonin thresholds (to discriminate for bacterial infections), and was derived by a decision-tree method," the authors note in their paper.
"The CDR distinguished between febrile neutropenia in children with blood cancers or solid tumors, since the risk of severe infection is three times higher in children with blood cancer," they report.
"This is therefore the first CDR to distinguish the risk of severe infection at different times, according to the type of cancer. It takes place early, with data collected at admission for children with a solid tumor (lowest risk of severe infection), and at 24 to 48 hours after presentation in patients with blood cancer, allowing for retesting and for greater confidence in a recommendation," they note.
"Physicians could rapidly try to test our CDR. Nonetheless, an impact analysis of this CDR is necessary before use in current practice. It should assess the tool's ease of use in practice, its impact in terms of patient and family quality of life, or cost savings and the absence of any harmful consequences. Such an analysis is necessary to show its usefulness in clinical practice and is currently underway," the study team adds.
This research was funded by The National League Against Cancer.
SOURCE: https://bit.ly/3ynKUu4 The Lancet Child & Adolescent Health, online December 3, 2021.
By Reuters Staff
Posted on
site created by:
© 2024 Medicom Medical Publishers. All rights reserved. Terms and Conditions | Privacy Policy