Clinical characteristics and outcomes of cancer patients requiring intensive care unit admission: a prospective study.

PURPOSE

The study was aimed to describe the characteristics of cancer patients admitted to the oncological ICU and to identify clinical features associated with outcomes.

METHODS

This is a prospective study (January 2014 to December 2015) of 522 cancer patients consecutively admitted to the oncological ICU. Patients with a length of oncological ICU stay ≤ 1 day were excluded. Demographic and clinical variables were obtained at oncological ICU admission. The primary outcome of interest was hospital mortality. Logistic regression analysis was performed to identify independent risk factors for hospital mortality.

RESULTS

The study cohort consisted of 492 (94.3%) patients with solid tumours and 30 patients (5.7%) with haematological malignancies. Advanced cancer was observed in 53.3%. Unplanned admission accounted for 25.3%. Hospital mortality rate was 13.0% (n = 68), and it was higher for patients with unplanned admission than those for electively admitted patients (35.6% vs. 5.4; p < 0.0001). Stage IV of cancer (OR 5.28; 95% CI 2.71-10.28; p < 0.0001), patients from the emergency department (OR 3.33; 95% CI 1.68-6.61; p = 0.001), unplanned admission (OR 7.99; 95% CI 4.45-14.33; p < 0.0001), non-malignancy-related admission (OR 5.80; 95% CI 3.26-10.32; p < 0.0001), sepsis (OR 4.81; 95% CI 2.28-10.16; p < 0.0001), chemotherapy-induced adverse event (OR 5.64; 95% CI 2.33-13.66; p < 0.0001), and invasive mechanical ventilation (OR 18.70; 95% CI 9.93-35.21; p < 0.0001) were independently associated with increased hospital mortality in multivariate logistic regression analysis.

CONCLUSIONS

ICU admission of cancer patients should be based on potential chance of recovering from the acute problem. Clinical predictor for mortality could support this purpose (UIN: researchregistry3484).