Factors associated with pressure ulcers in patients in a surgical intensive care unit.

Purpose: We sought to describe the occurrence of pressure ulcers in patients managed in a surgical intensive care unit (ICU) and report preliminary work toward development of a pressure ulcer risk assessment tool for use in this population that incorporates comorbidities and other factors not addressed in previous instruments.

Subjects And Settings: Three hundred sixty-nine patients managed in the surgical ICU at Yale-New Haven Hospital comprised the sample.

Methods: Demographic and clinical data, including Braden Scale scores, were collected in this prospective, 2-phase study. The principal investigator (G.C.S.) performed skin assessments on all patients. We used chi-square analysis and t tests to determine variables to include in a stepwise logistic regression analysis to determine factors independently associated with the development of pressure ulcers.

Instrument: We collected data, using a form we developed that contained demographic and clinical factors found in previous research and in our clinical practice to be associated with pressure ulcers.

Results: Eighty-eight out of 369 patients (23.9%) experienced a hospital-acquired pressure ulcer. Braden Scale scores ranged from 6 to 21, with a mean score of 11.9 ± 2.2. A lower Braden Scale score, the presence of diabetes mellitus, and patient age 70 years or older independently predicted the development of a pressure ulcer. These factors have been incorporated into the preliminary Surgical ICU Pressure Ulcer Risk Assessment scale.

Conclusions: Findings from this study suggest that, in addition to a low Braden Scale score, age >70 years and a diagnosis of diabetes may represent clinically relevant pressure ulcer risk factors in the surgical intensive care population and that patients with these factors may benefit from more aggressive preventive care. In addition, the Surgical ICU Pressure Ulcer Risk Assessment scale requires additional psychometric testing before its use can be recommended for the research or practice settings.