Drivers of early readmission and mortality after metastatic spinal cord compression in cervical cancer: Real-world evidence on disease progression, infection, and palliative gaps.

Abstract

e17523 Background: Metastatic spinal cord compression (MSCC) is a catastrophic complication of cervical cancer, yet real-world patterns of local therapy, short-term outcomes, and drivers of early readmission are poorly defined. We evaluated treatment selection, mortality, and causes of 30-day readmission in a nationally representative cohort. Methods: We performed a retrospective cohort study using the Nationwide Readmissions Database 2022. Adult index hospitalizations for cervical cancer with MSCC were identified using ICD-10 codes. Receipt of local therapy (surgery and/or radiation) and no therapy were the exposures. Survey-weighted analyses compared demographics, hospital characteristics, discharge disposition, comorbidities, metastatic burden, and goals of care. Multivariable Cox regression evaluated predictors of 30-day readmission and mortality. Primary readmission diagnoses were categorized to identify dominant drivers of early utilization. Results: Among 2,382 weighted index admissions, only 8.4% received local therapy. Treated patients were younger (50.5 vs 53.9 years) and were more often managed at large, urban hospitals. Neurologic deficit (OR 8.51, 95% CI 2.60–27.82) and visceral metastases (OR 11.71, 95% CI 2.65–51.79) strongly predicted receipt of surgery or radiation. Length of stay (16.0 vs 8.3 days, p=0.002) and total charges ($311,661 vs $107,270, p<0.001) were substantially higher in the local therapy group. The overall 30-day readmission rate did not differ by treatment (36.2% vs 33.3%; OR 1.14, 95% CI 0.42–3.08). The most common primary readmission diagnoses were recurrent or progressive cervical cancer, sepsis, cancer-related pain, brain metastases, and bone metastases, indicating that early readmissions were driven primarily by disease progression, infection, and symptom crises rather than procedural complications. In adjusted models, palliative or hospice care was associated with lower 30-day readmission (HR 0.66, p<0.001) but higher 30-day mortality (HR 4.25, p<0.001), reflecting advanced disease and treatment limitation. Acute kidney injury, respiratory failure, ICU admission, and sepsis independently predicted early mortality. Conclusions: In cervical cancer–associated MSCC, definitive local therapy is uncommon and concentrated among patients with severe neurologic compromise and extensive metastatic burden. Early readmissions are driven predominantly by tumor progression, infection, and uncontrolled symptoms rather than treatment toxicity. Short-term outcomes are strongly influenced by acute organ failure and goals of care. These findings highlight the need for rapid neurologic triage, early palliative care integration, and post-discharge pathways targeting infection prevention and symptom control to reduce avoidable readmissions.