The Association Between Pressure Ulcer/Injury Development and Short-term Mortality in Critically Ill Patients: A Retrospective Cohort Study.

Abstract

The aim of this retrospective study was to examine the association between PU/I development and mortality in a large cohort of consecutively admitted critically ill patients. Data from adult patients ( >18 years of age) admitted to an intensive care unit (ICU) between 2010 and 2013 were extracted from the ICU electronic clinical information system. Patients were excluded if they had a PU/I present on admission, no recorded admission modified Jackson/Cubbin (mJ/C) or Sequential Organ Failure Assessment (SOFA) score, or their ICU outcome was undetermined. The mJ/C risk scale (score range 9-48) was used to assess the PU/I risk (the lower the score, the higher the PU/I risk), the SOFA score (score range 0-24; the higher the SOFA score, the sicker the patient, with a higher risk of death) was used to assess the severity of the condition and outcome. ICU outcome was defined as 1) moved from the ICU to a ward/recovering or 2) no response to ICU treatment/deceased. All data were transferred to statistical software for analysis. Logistic regression analysis was used to examine the outcome related to PU/I development, SOFA, and mJ/C scores. Descriptive contingency tables of different scenarios were used to further evaluate relationships among different risk factors related to mortality; the Wald χ2 test was used to assess the statistical significance of the contingency tables. Of the 6582 patients admitted, 6089 were included for analysis. Two hundred, one (201) had a PU/In on admission, 212 had missing mJ/C or SOFA scores, and ICU outcome was undetermined in 80 patients. Patient mean age was 61.1 ± 15.8 (range 18-94) years; 3891 (63.9%) were male, average length of stay (LOS) was 3.6 days, denoted by quartile (Q) (median 1.58 days; Q1: 0.9, Q3: 3.9 days), and 1589 (26.1%) stayed 3 days or more in the ICU. The incidence of PU/I was 6.9% (423 patients), and ICU mortality rate was 9.1% (n=553). The mean LOS of patients with PU/I was 13.35 ± 15.56 days (median 8.95, Q1: 4.88, Q3: 16.2) and 2.84 ± 3.87 days for patients with no PU/I (median: 1.20, Q1: 0.90, Q3: 3.17; P <.0001). Mean LOS was 3.42 ± 5.95 days (median: 1.30, Q1: 0.90, Q3: 3.70) among recovering and 5.00 ± 7.17 days among deceased patients (median 2.56, Q1: 1.26, Q3: 6.40; P <.0001). The proportion of patients with an admission mJ/C score of 29 or less ranged from 48.8% to 51.5%, and the mean SOFA score was 7.0 ± 3.2. PU/I development and SOFA or mJ/C scores were independent predictors of mortality. The probability of a negative outcome was higher in persons with PU/Is compared to persons with no PU/Is. Persons with lower SOFA scores (ie, less severely ill patients) and higher mJ/C scores for each factor separately (ie, at low risk of PU/I development) each factor separately had a lower mortality risk. PU/I development in critically ill patients treated at an ICU is an independent predictor of mortality, even though the PU/I incidence and hospital mortality were relatively low. The ICU admission SOFA and mJ/C score also were independent prognosticators of ICU mortality. Future research could focus on the role of different steps in the cascade of PU/I development, especially to the role of inflammation.