By Samuel Nadler, MD, PhD
Clinical Instructor, University of Washington, Seattle
SYNOPSIS: The presence of metastatic disease, respiratory failure, elevated lactate levels, and poor prior performance scores were independent predictors of mortality in patients with solid tumors admitted with septic shock.
SOURCE: Cuenca JA, Manjappachar NK, Ramírez CM, et al. Outcomes and predictors of 28-day mortality in patients with solid tumors and septic shock defined by Third International Consensus Definitions for Sepsis and Septic Shock Criteria. Chest 2022;162:1063-1073.
With advances in cancer treatment, more patients with solid tumors and advanced malignancies are admitted with septic shock. The authors of this study aimed to better understand the overall outcomes of patients with solid tumors meeting the Sepsis-3 definition for septic shock and factors that predict poorer outcomes. This was a retrospective study conducted at single, large cancer center between April 2016 and March 2019.
In total, 271 patients were identified, with a median age of 62 years, with 57.2% men, and 52.5% white. The most common malignancies were lung cancer (19.2%), breast cancer (7.7%), pancreatic cancer (7.7%), and colorectal cancer (7.4%). An infectious agent was identified in 65.7% of admissions. In addition to sepsis, 79% of patients also experienced respiratory failure, 70.1% had acute kidney injury, and 21.8% were in atrial fibrillation.
As expected, mortality rates for patients with solid tumors admitted with septic shock were high. The primary outcome of 28-day mortality occurred in 69.4%, with intensive care unit (ICU), hospital, and 90-day mortality rates of 58.7%, 68.6%, and 77.1%, respectively. A multivariate logistic regression identified four variables predictive of mortality. These included: presence of metastatic disease (odds ratio [OR], 3.17; P = 0.004), respiratory failure (OR, 2.34; P = 0.018), highest lactate (OR, 3.19 per unit increase; P < 0.001), and European Collective Oncology Group (ECOG) score 3-4 (OR, 2.72; P = 0.006).
Interestingly, neither neutropenia (P = 0.912) nor recent chemotherapy (P = 0.606) was statistically correlated with mortality. Furthermore, many comorbidities, such as heart failure, chronic obstructive pulmonary disease (COPD), diabetes, and liver disease, also were not predictive of mortality, although the number of events in each category was low. Although not reaching statistical significance, there was a higher percentage of non-survivors than survivors in patients admitted with viral and fungal infections, while the opposite was found in patients with either gram-positive or gram-negative infections.
Of the interventions catalogued, the use of vancomycin was associated with greater survival (86.7% vs. 64.4%, P < 0.001), while vasopressin and phenylephrine use was higher in non-survivors. The use of antivirals favored survival as well (12% vs. 5.4%, P = 0.051).
COMMENTARY
The current study aligns with others demonstrating a high 28-day mortality in patients with malignancies admitted with septic shock. Awad et al published a 12-year retrospective analysis of cancer patients admitted with septic shock, reporting a hospital mortality rate of 64.9%.1 Similar to the current study, mechanical ventilation for respiratory failure was associated with increased mortality (OR, 2.109; P < 0.001), as were increased lactic acid levels (OR, 1.355; P = 0.0436) and stage of malignancy. Again, neither neutropenia nor recent chemotherapy seemed to contribute to mortality, although there was an association with liver disease.
Manjappachar et al reported a 28-day mortality of 67.8% in patients with hematologic malignancies admitted with septic shock.2 Respiratory failure and highest lactate were associated with the greatest odds of mortality (OR, 3.12; P = 0.003, and OR. 1.16; P < 0.01, respectively). As all three studies are retrospective, it is difficult to ascribe causation rather than simple association to these factors.
Two of these studies reported ICU admission from the hospital ward rather than the emergency department (ED) was associated with increased mortality. A univariate analysis within the current study showed a greater percentage of survivors than non-survivors in patients admitted from the ED (55.4% vs. 36.2%) than admitted from the hospital ward (44.6% vs. 63.8%, P = 0.003). Manjappachar et al showed a similar pattern with a larger percentage of survivors than non-survivors in patients admitted from the ED (39.1% vs. 23.4%) rather than admitted from the hospital ward (60.8% vs. 76.5%, P = 0.005).2 This raises the question of appropriate triage of patients as hospital ward admission rather than ICU admission might result in delays in recognition and treatment of septic shock.
Predictors of mortality can be used to facilitate goals of care discussions with patients and families. In this cohort of patients, 28-day mortality was 69.4%. In patients with respiratory failure, that increased to 79% and up to 84.5% in patients with metastatic disease. Of the patients who died within 28 days of admission, 91% were do not resuscitate (DNR) and 60.1% were receiving comfort-oriented care. Data on the proportion of patients with DNR or comfort care orders who survived was not available. Mortality in patients with solid tumors admitted with septic shock is high. The current study reinforces the importance of goals of care discussions with these patients and their families.
REFERENCES
- Awad WB, Nazer L, Elfarr S, et al. A 12-year study evaluating the outcomes and predictors of mortality in critically ill cancer patients admitted with septic shock. BMC Cancer 2021;21:709.
- Manjappachar NK, Cuenca JA, Ramírez CM, et al. Outcomes and predictors of 28-day mortality in patients with hematologic malignancies and septic shock defined by sepsis-3 criteria. J Natl Compr Caner Netw 2022;20:45-53.
