Contributions
Abstract: EP810
Type: E-Poster Presentation
Session title: Infections in hematology (incl. supportive care/therapy)
Background
Febrile neutropenia (FN) is a potential life-threatening complication of the administration of chemotherapy in oncohematological patients. Only 20% of FN episodes have microbiologically-proven bacteremia. In most of cases, the therapeutic approach is based on the clinical picture and the laboratory evaluation of surrogate markers. More recently, soluble CD14 subtype, also known as presepsin (PS), has been proposed as a novel biomarker of microbial infection. Whether PS may have a role in predicting response to treatment is still unexplored.
Aims
We aimed to assess the kinetic of PS concentrations in a cohort of adult febrile neutropenia patients with haematological malignancies with particular interest to the time course in relation to the appropriateness of first-line empirical antimicrobial treatment.
Methods
From April 2018 to December 2019, consecutive patients admitted to our Unit for high dose chemotherapy alone or followed by autologous bone marrow stem cell transplantation were considered. Three categories were considered: Sepsis, Bacteremia, and fever of unknown origin (FUO), As index tests, we considered PS, which was measured as follow: at patient admission; on the first day of chemotherapy (CT); at the onset of neutropenia (onset FN-); between 90 and 120 minutes from fever onset (onset FN+, before any broad-spectrum empirical antibiotic therapy); between 48 and 72 hours from the start of antibiotic therapy; between 24 and 48 hours from neutropenia resolution. Changes in presepsin concentration ([72 hours from empirical antibiotic - fever onset]/ fever onset) were categorised according to tertiles. The response to antibiotic therapy and the time to fever resolution according to tertiles of PS concentration were analysed considering in the model the factor tertile. The statistical significance was declared for p<0.05.
Results
273 episodes (148 subjects, 80 males and 68 females) were eligible for this study. Reason for treatment was acute leukemia in 106 episodes, multiple myeloma in 68, lymphoma in 99. Autologous stem cell transplantation was performed in 68. In neutropenic patients with fever (FN+) or without fever (FN-), there was a statistically significant effect of the febrile status (p=0.0126), the time point considered (p<.0001) and their interaction (p=0.0006). At onset FN+ compared to FN- had higher levels (p<.0001, fold-change 1.82, 95% CI: 0.26 – 2.64; fig.1A). PS concentration was significantly higher in patients with documented bacterial infection (670 ng/L, [165–1267] ng/L than in those with negative cultures (358 ng/L, [105–1134] ng/L (p<0.0001, not shown). A statistically significant reduction was observed at 48-72 hours from antibiotic therapy (median 244 ng/L, range 89-721, p<0.0001; fig.1B). Patients with major reduction of presepsin concentrations from fever onset to antibiotic therapy, according to tertiles, tended to have received early appropriate antibiotic therapy (21/34, 61.7% in tertile 1, < -46%; 11/35, 31.4% in tertile 2, -45 to -17%; 5/35, 14.3% in tertile 3, -16% to +246%; p = 0.0009 for trend; fig.1C). Response to therapy probability was higher in tertiles 1 and 2 compared to tertile 3 (OR1 vs 3=3.5 and 95%CI: 1.0-12.4, p=0.0482; OR2 vs 3=9.7 95%CI: 2.2-42.5; fig.1D). A graded decrease in fever duration was observed across tertiles of changes for PS (trend p= 0.0004).
Conclusion
In conclusion, our study provides preliminary indications that PS has a role in monitoring appropriate antibiotic therapy in oncohematological patients with FN.
Keyword(s):
Abstract: EP810
Type: E-Poster Presentation
Session title: Infections in hematology (incl. supportive care/therapy)
Background
Febrile neutropenia (FN) is a potential life-threatening complication of the administration of chemotherapy in oncohematological patients. Only 20% of FN episodes have microbiologically-proven bacteremia. In most of cases, the therapeutic approach is based on the clinical picture and the laboratory evaluation of surrogate markers. More recently, soluble CD14 subtype, also known as presepsin (PS), has been proposed as a novel biomarker of microbial infection. Whether PS may have a role in predicting response to treatment is still unexplored.
Aims
We aimed to assess the kinetic of PS concentrations in a cohort of adult febrile neutropenia patients with haematological malignancies with particular interest to the time course in relation to the appropriateness of first-line empirical antimicrobial treatment.
Methods
From April 2018 to December 2019, consecutive patients admitted to our Unit for high dose chemotherapy alone or followed by autologous bone marrow stem cell transplantation were considered. Three categories were considered: Sepsis, Bacteremia, and fever of unknown origin (FUO), As index tests, we considered PS, which was measured as follow: at patient admission; on the first day of chemotherapy (CT); at the onset of neutropenia (onset FN-); between 90 and 120 minutes from fever onset (onset FN+, before any broad-spectrum empirical antibiotic therapy); between 48 and 72 hours from the start of antibiotic therapy; between 24 and 48 hours from neutropenia resolution. Changes in presepsin concentration ([72 hours from empirical antibiotic - fever onset]/ fever onset) were categorised according to tertiles. The response to antibiotic therapy and the time to fever resolution according to tertiles of PS concentration were analysed considering in the model the factor tertile. The statistical significance was declared for p<0.05.
Results
273 episodes (148 subjects, 80 males and 68 females) were eligible for this study. Reason for treatment was acute leukemia in 106 episodes, multiple myeloma in 68, lymphoma in 99. Autologous stem cell transplantation was performed in 68. In neutropenic patients with fever (FN+) or without fever (FN-), there was a statistically significant effect of the febrile status (p=0.0126), the time point considered (p<.0001) and their interaction (p=0.0006). At onset FN+ compared to FN- had higher levels (p<.0001, fold-change 1.82, 95% CI: 0.26 – 2.64; fig.1A). PS concentration was significantly higher in patients with documented bacterial infection (670 ng/L, [165–1267] ng/L than in those with negative cultures (358 ng/L, [105–1134] ng/L (p<0.0001, not shown). A statistically significant reduction was observed at 48-72 hours from antibiotic therapy (median 244 ng/L, range 89-721, p<0.0001; fig.1B). Patients with major reduction of presepsin concentrations from fever onset to antibiotic therapy, according to tertiles, tended to have received early appropriate antibiotic therapy (21/34, 61.7% in tertile 1, < -46%; 11/35, 31.4% in tertile 2, -45 to -17%; 5/35, 14.3% in tertile 3, -16% to +246%; p = 0.0009 for trend; fig.1C). Response to therapy probability was higher in tertiles 1 and 2 compared to tertile 3 (OR1 vs 3=3.5 and 95%CI: 1.0-12.4, p=0.0482; OR2 vs 3=9.7 95%CI: 2.2-42.5; fig.1D). A graded decrease in fever duration was observed across tertiles of changes for PS (trend p= 0.0004).
Conclusion
In conclusion, our study provides preliminary indications that PS has a role in monitoring appropriate antibiotic therapy in oncohematological patients with FN.
Keyword(s):