Evaluation of Infection Risk Factors in Adult Hematologic Malignancy Patients

INTRODUCTION: Hematologic malignancies (HM) are often highly aggressive processes requiring multi-chemotherapy regimens and potentially hematopoietic stem cell transplant (HSCT) for treatment. Totally implantable venous access ports (TIVAP) are the standard access for administration of these regimens. The regimens usually involve periods of neutropenia for which infection can increase morbidity and require removal of the port. Although deemed safe for use in cancer populations, many studies have only looked at patients with TIVAP in solid tumor malignancies. Herein, we report our single institution experience identifying modulators for infection related to TIVAP in this high-risk population. OBJECTIVES: To identify risk factors for port infection in HM patients. METHODS: We performed an IRB-approved single institution retrospective review of HM patients who had a TIVAP removed for port infection versus completion of therapy. Medical records were evaluated for patient demographics and tumor type, details of previous therapy regimens such as chemotherapy and HSCT status, laboratory values including albumin, white blood cell, neutrophil and platelet counts, as well as characteristics of the TIVAP. Univariate and multivariate regression analyses were performed between the two cohorts to identify clinical predictors of port infection necessitating TIVAP removal. RESULTS: Between March 2015 and July 2018, 104 TIVAPs were removed from HM patients, 34 (33%) for infection and 70 (67%) for completion of treatment. Median patient age was 60 years (range 22-86) and there was an even number of men and women (n=52). Most TIVAPs were double lumen (n=98, 94%) and placed in the right IJV (n=100, 96%). The median number of port days was 49 in the infection group and 414 in the completion group, with removal of 12 (35%) of the infected ports within 30 days. Multiple factors were significant predictors of port infection on univariate analysis, including diagnosis of AML or ALL (p=0.0019), no prior HSCT (p=0.0148), neutropenia within 30 days of removal (p<0.0001), ANC <1500 (p <0.0001), leukopenia the day prior to removal (p<0.0001), thrombocytopenia (p <0.0001), and hypoalbuminemia (p<0.0001). Steroid use, leukopenia or leukocytosis the day of placement, and leukocytosis the day prior to removal were not significantly different between the two groups. On multivariate analysis, no prior HSCT (OR=48.08, 95% CI 2.64-875.63, p=0.009), neutropenia within 30 days of removal (OR=67.60, 95% CI 1.68-2726.66, p=0.026), and hypoalbuminemia (OR=405.52, 95% CI 15.64-10517.89, p<0.0001) remained significant. CONCLUSION: In patients with HM, significant predictors of port infection requiring TIVAP removal included neutropenia within 30 days of removal and hypoalbuminemia. Interestingly, patients who underwent HSCT were less likely to a develop a port infection requiring TIVAP removal.