Abstract 1496P
Background
Determining when to initiate renal replacement therapy (RRT) in cancer patients with acute kidney injury (AKI) can pose a considerable challenge, particularly for those with limited treatment options for cancer and a short life expectancy. This study aims to explore the predictors of short-term mortality among cancer patients undergoing hemodialysis following AKI.
Methods
Between January 2014 and April 2023, this retrospective study included 139 hospitalized patients with various cancers who underwent hemodialysis for AKI in Hacettepe University Oncology Hospital. Clinical and laboratory features, comorbidities, and survival outcomes were collected. Predictors of 30-day mortality after initiation of hemodialysis were evaluated.
Results
Median age was 64 and 39% were females. ECOG performance score was 3-4 in 48%. Median Charlson comorbidity index was 8. Among the patients, 26% had gastrointestinal cancers and 23% had lung cancer. Notably, the etiology of AKI was prerenal-renal in 93% of the cases, with 45% of AKI instances occurring during hospitalization. 111 patients (80%) died within 30 days after the first hemodialysis session. In multivariate analysis, serum albumin less than 2.5 g/dl (OR: 3.14, 95% CI; 1.23-8.04, p=0.017), presence of sepsis (OR: 3.50, 95% CI; 1.34-9.10, p=0.010), prerenal-renal AKI (compared to postrenal AKI) (OR: 6.06, 95% CI 1.38-26.56, p=0.017) were associated with higher 30-day mortality. Patients with none of these risk factors had 33% 30-day mortality while those with all the identified risk factors had significantly higher rate of 93% (p<0.001).
Conclusions
In this study, patients with sepsis, pre-renal or renal AKI, and low albumin levels had higher early mortality rates after hemodialysis. Identifying AKI patients unlikely to benefit from RRT may assist clinicians in making informed decisions.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
1199P - Developing and systematically validating homologous recombination repair gene detection method based on next-generation sequencing
Presenter: Yi Sun
Session: Poster session 10
1200P - Investigation of multiphoton microscopy as an innovative tool for intraoperative section-free histologic investigations in just a few minutes
Presenter: Martí Homs Soler
Session: Poster session 10
1201P - Novel deep learning model and validation of whole slide images in lung cancer diagnosis
Presenter: Alhassan Ahmed
Session: Poster session 10
Resources:
Abstract
1202P - A deep learning approach using routine pathology images to guide precision medicine in metastatic CRC
Presenter: Chaitanya Parmar
Session: Poster session 10
1203P - Analytical evaluation of whole genome sequencing for acute myeloid leukemia
Presenter: Guidantonio Malagoli Tagliazucchi
Session: Poster session 10
1204P - Real-world utility of whole genome sequencing for patients with cancer: Evaluation of a regional implementation of the 100,000 genomes project
Presenter: Helen Robbins
Session: Poster session 10
1205P - A retrospective machine learning-based analysis of nationwide cancer CGP data across cancer types to identify features associated with recommendation of mutation-based therapy
Presenter: Hiroaki Ikushima
Session: Poster session 10
1478P - Dual single-nucleotide polymorphism biomarker combination to select opioid for cancer pain management
Presenter: Yoshihiko Fujita
Session: Poster session 10
1479P - Use of rescue opioids and pain control after ketamine initiation in refractory cancer pain: A multicentric observational study
Presenter: Pablo Gallardo Melo
Session: Poster session 10
1480P - Long term therapy with denosumab and zoledronic acid: A comparative real-world retrospective observational study on skeletal-related events and pain in patients with metastatic breast cancer
Presenter: Giacomo Massa
Session: Poster session 10