AUTHORS
Shiliang Ge, MD, Isidora Rovic, MD2, Meha Bhatt, MD, MSc3, Abi Vijenthira, MD, MS4, Matthew C. Cheung, MD, SM, FRCPC5, Gordon Rubenfeld, MD, MSc6, Laveena Munshi, MD, MS7
AFFILIATIONS
1 University of Toronto, Toronto, Ontario, Canada
2 McMaster University, Hamilton, Ontario, Canada
3 McMaster University, Hamilton, Ontario, Canada
4 Princess Margaret Cancer Centre, Toronto, Ontario, Canada
5 Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada
6 University of Toronto, Toronto, Ontario, Canada
7 University of Toronto, Toronto, Ontario, Canada
ABSTRACT
Introduction
One in seven patients with hematologic malignancies are admitted to the intensive care unit (ICU) in the first year of diagnosis. While ICU survival in hematologic populations has improved, the impact of critical care illness on subsequent oncologic treatment trajectories is poorly understood. Our study aimed to evaluate the impact of critical illness on return to oncologic treatment, functional outcomes, and survival.
Methods
We conducted a retrospective cohort study of patients ≥18 years old who had a diagnosis of a hematologic malignancy and survived an ICU admission at Mount Sinai Hospital and Princess Margaret Cancer Centre between January 2018 and October 2024. Eligible patients had a documented active or planned pre-ICU oncologic treatment plan; those who completed treatment were excluded. Data were collected from electronic medical records for the oncology treatment plan and a pre-established ICU database for details of the critical care admission. Patients were followed for 12 months. The primary outcome, change in the intended oncology treatment plan at first follow-up after ICU discharge, was categorized as resumed as intended, modified or not resumed. Secondary outcomes included 12-month survival. Factors associated with return to treatment were evaluated using multivariable logistic regression analysis, adjusting for covariates including age, comorbidities, disease status, functional status, severity of illness at admission, and line of therapy (selected by univariate analysis).
Results
Of 473 patients with hematologic malignancies admitted to the ICU, 202 patients survived and were included. Median age was 61 years (IQR 52-69) and 41% of patients were female. The cohort consisted predominantly of patients with acute leukemias, including acute myeloid leukemia (n=80, 40%), acute promyelocytic leukemia (n=12, 6%), and acute lymphoblastic leukemia (n=30, 15%). Other myeloid malignancies were myelodysplastic syndrome (MDS; n=12, 6%) and myeloproliferative neoplasms (MPN; n=9, 4%). Aggressive non-Hodgkin B-cell lymphomas accounted for 10% and multiple myeloma for 8% of included patients. The median Charlson Comorbidity Index was 4 (IQR 3-6). At ICU admission, 49 patients (24%) were newly diagnosed and planned for induction. 89 (44%) were already receiving first-line treatment, 32 (16%) receiving second-line, 28 (14%) receiving third-line or beyond. The worst 24 hours sequential organ failure assessment (SOFA) score was 8 (IQR 6-10). The most common admission reasons were acute respiratory failure (44%) and sepsis (44%).
Within 12 months of ICU discharge at the first timepoint of follow up, 55 patients (27.2%) resumed their intended regimen, whereas 147 (73%) had treatment modified or not resumed. Median time from ICU discharge to therapy was 6 days (IQR 3.5–16) among those resuming as intended versus 28.5 days (IQR 7–56) among those modified (p<0.001). The most frequent reasons for non-resumption were decline in performance status (71%), disease progression (22%), patient preference (21%), and post-ICU organ dysfunction (cardiac 12%, pulmonary 10%, renal 8%, liver 4%). On multivariable analysis, ICU admission for treatment-related complication (vs. disease-related or other causes; OR 0.14, 95% CI 0.02–0.77) and higher ECOG score at 7 days post-ICU, (per 1-point increase; OR 0.30, 95% CI 0.14–0.64), were independently associated with failure to resume treatment, whereas admission SOFA score was not. 12-month survival was significantly higher in patients who resumed intended treatment than those who had modified/not resumed treatment (72% vs 41%, p<0.001).
Conclusion
In this cohort of ICU survivors with hematologic malignancies, fewer than one-third resumed their intended oncologic regimen after critical illness, and failure to do so was associated with worse 12-month survival. Independent predictors of treatment modification were ICU admission for treatment-related complication and poor post-ICU functional status. These findings suggest that functional recovery, more than initial severity of illness, shapes whether patients return to cancer-directed therapy – highlighting functional optimization during and after ICU as a potential intervention target. As one of the largest studies to characterize how critical illness alters treatment plans in patients with hematologic malignancies, this work establishes a benchmark dataset to inform shared decision-making at the interface of critical care and oncology.
DISCLOSURES
Shiliang Ge, MD: No relevant conflicts of interest to declare.