Cardiopulmonary instability score and in-hospital mortality in critically ill patients with cardiovascular disease: a retrospective eICU cohort study.

In-hospital mortality increased progressively with higher CPI scores, ranging from 1.4% at CPI 0 to 28.0% at CPI 3-4.

• Cohort included 61,273 eligible ICU stays from 49,991 patients across 206 hospitals during 2014-2015.
• Overall in-hospital mortality was 13.4% (8,238 stays).
• Mortality by CPI score: 1.4% (CPI 0), 4.6% (CPI 1), 11.6% (CPI 2), and 28.0% (CPI 3-4).
• All four CPI components were available for 55,601 stays (90.7% of the cohort).

CPI ≥2 was independently associated with in-hospital mortality in adjusted analysis.

• Adjusted odds ratio for in-hospital mortality at CPI ≥2: OR 4.45 (95% CI 4.06 to 4.89).
• Relative risk for in-hospital mortality at CPI ≥2: RR 3.64 (95% CI 3.35 to 3.97).
• Findings were consistent in sensitivity analyses excluding mechanical ventilation, restricting to stays >24 hours, and accounting for repeated stays.

CPI alone demonstrated moderate apparent discrimination for in-hospital mortality.

• Area under the receiver operating characteristic curve (AUROC) for CPI alone was 0.732 (95% CI 0.726 to 0.738).
• At the CPI ≥2 threshold, sensitivity was 90.4% and specificity was 39.4%.
• Positive predictive value at CPI ≥2 was 18.8% and negative predictive value was 96.3%.
• The low positive predictive value and moderate discrimination were cited as reasons CPI should not be used alone for individual prognostication.

CPI ≥2 was also associated with secondary outcomes including ICU mortality and prolonged ICU and hospital stays.

• Secondary outcomes included ICU mortality, prolonged ICU stay, and prolonged hospital stay.
• The association between CPI ≥2 and secondary outcomes was consistent with the primary outcome findings.
• The CPI score assigned one point each for hypotension, hypoxaemia, tachycardia, and mechanical ventilation, yielding a total score of 0 to 4.

The study was conducted as a retrospective multicentre cohort using the eICU Collaborative Research Database, covering ICU admissions from US hospitals in 2014-2015.

• The eICU database comprised ICU admissions from hospitals in the USA during 2014-2015.
• Eligible participants were adults aged ≥18 years with cardiovascular disease and known hospital outcome status.
• The unit of analysis was the ICU stay, not the individual patient.
• 206 hospitals were represented in the cohort.

The CPI score is composed of four readily measurable clinical components assessed during the first 24 hours of ICU admission.

• The four components are: hypotension, hypoxaemia, tachycardia, and mechanical ventilation.
• Each component contributes one point, yielding a total score ranging from 0 to 4.
• The score is described as 'a simple descriptive summary of first-day cardiopulmonary instability.'