Background: Neutropenia and neutropenic fever are a major cause of in-hospital morbidity and mortality in cancer patients. In this study, we aim to analyze trends in hospital outcomes of cancer-related neutropenic fever admissions and investigate clinical predictors associated with outcomes.

Methods: We queried the National Inpatient Sample (NIS) database (2008-2017), using ICD-9 and ICD-10 codes accordingly, to identify cancer-related neutropenic fever admissions in the US hospitals. Cases younger than 18 years of age and cases that did not have malignancy as a primary or secondary diagnosis were excluded. Baseline characteristics of survivors were compared with non-survivors. Statistical analysis was performed using SPSS v26 (IBM Corp, Armonk, NY, USA). The adjusted odds ratio (aOR) and 95% confidence interval (CI) were calculated using the Cochran-Mantel-Haenszel test. A multivariate regression model was deployed to assess predictors of inpatient mortality. Complex weights were used throughout all calculations, enabling appropriate national projections.

Results: A total of 159,065 records were identified using our inclusion and exclusion criteria, corresponding to 778,427 admissions nationally. The in-hospital mortality rate for all patients was 5.9%. Most common documented sources of infections were respiratory (24.3%), followed by urinary tract infections (13.2%), intraabdominal infections (7.6%), and skin and soft tissue infections (7.1%). Overall, sepsis was present in about quarter of the admissions, while acute kidney injury and respiratory failure were also prevalent (13.2% and 6.9%, respectively). On regression analysis, older age (OR 1.026; 95% CI, 1.025 - 1.027; P <0.001), heart failure (OR 1.90; 95% CI, 1.84 - 1.96; P <0.001), liver disease (OR 1.64; 95% CI, 1.56 - 1.72; P <0.001), chronic kidney disease (OR 1.325 95% CI, 1.30 - 1.39; P <0.001), coagulopathy (OR 1.67; 95% CI, 1.64-1.71; P <0.001), and metastatic disease (OR 1.70; 95% CI, 1.65- 1.74; P <0.001), were independent predictors of increased in-hospital mortality among patients admitted with febrile neutropenia, whereas female sex was associated with relatively favorable outcome (OR 0.93; 0.92-0.95; P <0.001). Based on the source of infection that was documented during admission, we found respiratory infections (OR 1.27; 95% CI, 1.24-1.30; P <0.001), meningoencephalitis (OR 1.23; 95% CI, 1.12-1.36; P <0.001), and systemic mycosis (OR 1.28; 95% CI, 1.24-1.32; P <0.001) were associated with worse outcome, while patients who had urinary (OR 0.75; 95% CI, 0.72-0.78; P <0.001) or soft tissue source of infection (OR 0.75; 95% CI, 0.71-0.79; P <0.001), or neutropenic enterocolitis (OR 0.91; 95% CI, 0.84-0.98; P <0.001) had a relatively better outcome. In terms of complications during admission, we found the occurrence of acute kidney injury (OR 2.13; 95% CI, 2.07-2.18; P <0.001), sepsis (OR 2.39; 95% CI, 2.32-2.45; P <0.001), septic shock (OR 3.25; 95% CI, 3.14-3.35; P <0.001), acute encephalopathy (OR 1.71 95% CI, 1.63-1.79; P <0.001), and acute respiratory failure (OR 9.98; 95% CI, 9.71-10.24; P <0.001) were associated with higher mortality.

Conclusion: Despite the advances in risk stratification and frequent use of unified protocols that incorporate prophylactic growth factors in the therapeutic regimens, incidence of in-hospital mortality of febrile neutropenia appears to be unchanged over the decade between 2008 and 2017. We identified older age, metastatic disease, presence of chronic heart failure, liver disease, chronic kidney disease, coagulopathy, presence of acute kidney injury, sepsis, acute respiratory failure, encephalopathy, and septic shock as independent predictors of increased in-hospital mortality among patients admitted with febrile neutropenia. Further studies are needed to further explore predictors of poor outcomes in those patients.

Disclosures

Anwer:Incyte, Seattle Genetics, Acetylon Pharmaceuticals, AbbVie Pharma, Astellas Pharma, Celegene, Millennium Pharmaceuticals.: Honoraria, Research Funding, Speakers Bureau.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract