Predictors of early port infection in adult patients with acute myeloid leukemia.

person Hyndavi Kandala University of Alabama at Birmingham, Birmingham, AL info_outline Hyndavi Kandala, Kendall Diebold, Kimo Bachiashvili, Ravi Bhatia, Manuel Ricardo Espinoza Gutarra, Omer Hassan Jamy, Pankit Vachhani, Vikas Dudeja, Aliaksei Salei, Harpreet Kaur Arora, Myles D Prados, Peter G. Pappas, Sravanti Rangaraju

Authors person Hyndavi Kandala University of Alabama at Birmingham, Birmingham, AL info_outline Hyndavi Kandala, Kendall Diebold, Kimo Bachiashvili, Ravi Bhatia, Manuel Ricardo Espinoza Gutarra, Omer Hassan Jamy, Pankit Vachhani, Vikas Dudeja, Aliaksei Salei, Harpreet Kaur Arora, Myles D Prados, Peter G. Pappas, Sravanti Rangaraju Organizations University of Alabama at Birmingham, Birmingham, AL, University of Alabma at Birmingham, Birmingham, AL, Department of Medicine, Division of Hematology/Oncology, University of Alabama at Birmingham, Birmingham, AL, Surgical Oncology, The University of Alabama at Birmingham/O'Neal Comprehensive Cancer Center, Brimingham, AL Abstract Disclosures Research Funding No funding received None. Background: The most concerning complications following port placement are infection and thrombosis. Early infections in patients with hematological malignancies are reported at 2-5% post-placement. However, data specific to patients with Acute Myeloid Leukemia (AML) is sparse. Here, we report infection rates within 30 days of placement and factors associated with increased risk of infection in patients with AML. Methods: We retrospectively reviewed charts of patients ≥ 18 years with AML who had ports placed between January 2019 to September 2022 at our institution. Baseline and peri-placement characteristics were collected. Port infection was defined as at least one of the following: 1) site infection, 2) gram-positive bacteremia, or 3) bacteremia that was documented by Infectious Disease consultants as a port source. Infections were classified as very early ( < 7 days) and early (7-30 days). Variables were compared in each group by using Pearson’s chi-square test. Results: A total of 101 ports were placed in 98 patients with AML. The median age at placement was 64 years and 58% of patients were female. A total of 14 (13.9%) port infections were identified; 5 (35.7%) were very early and 9 (64.2%) were early infections. The median time to infection was 15 days (range 1-25 days). 5 (35.7%) patients had site-only infections, 6 (42.8%) had isolated bacteremia, and 3 (21.4%) had both. 13 (93%) patients were hospitalized, and 1 patient died from infectious complications. 8 (57%) of infected ports were removed. 57 patients were on regimen-specific prophylactic antibiotics at the time of placement, of whom 12 developed an infection (21%). The infection rate was 10% vs 23% (p = 0.1) in patients with ANC ≥ 500 /µL and ANC < 500/µL respectively at 1 week after port placement, with no difference based on ANC at the time of placement. The infection rate was 21.9% vs. 6.5% in patients with platelets < 100 x10 3 /µL and ≥ 100 x10 3 /µL (p = 0.01) at the time of placement; 28% vs. 8.6% in patients with albumin < 3.5 g/L and ≥ 3.5 g/L (p = 0.02). The infection rate was numerically higher in Venetoclax-based regimens (19%) compared with cytarabine-based regimens (8%) (p = 0.06). Rates of infection were similar among inpatient and outpatient placements (14%). 5/53 patients (9.4%) were in complete remission (CR), 3/29 (10.3%) of newly diagnosed, and 6/19 (31.5%) relapsed/refractory (R/R) AML patients developed an infection (p = 0.02). Conclusions: In our single-institution study, a higher-than-expected port infection rate of 14% was seen in patients with AML. Platelets ≤100 x10 3 /µL and albumin < 3.5 g/L at the time of placement, Venetoclax-based regimens, and R/R state were associated with significantly higher rates of infection. Interestingly ANC at the time of placement was not associated with an increased infection rate. Strategies to prevent early infections in these high-risk patients are much needed.