Retrospective evaluation of the simplified Therapeutic Intevention Scoring System (TISS-28) in a surgical intensive care unit

Objective: To compare the simplified Therapeutic Intervention Scoring System (TISS-28) with its original version, to provide reference values of daily TISS-28 assessment and to describe its association with severity of illness in surgical patients. Design: Retrospective evaluation of prospectively collected audit data; four documentation periods. Setting: Ten-bed intensive care unit (ICU) in a surgical university hospital. Patients: One thousand nine hundred eighty-six consecutive admissions (1,808 patients; 10,448 observation days) who stayed on ICU for at least 6 h. Patients were in hospital for abdominal, vascular or trauma surgery. The average age was 61.5 years, the mean APACHE II score on admission 10.3 points. Interventions: None. Measurements: Raw data for APACHE Lf score and TISS were recorded daily. TISS-28 was calculated retrospectively from the original TISS data. Results: Average TISS-28 values (28.7 points; SD = 9.7) do not differ substantially from the original TISS values (28.2 points, SD = 10.9) and overall correlation is high (r = 0.935). Of the patients, 57.3 % left the ICU after 1-2 days as survivors with a mean daily TISS-28 of 20.0 points. Variability between documentation periods was higher with the original TISS. On average, patients with increasing severity of disease require an increasing amount of care. Survivors have lower TISS-28 values than non-survivors (27.6 vs 34.9). Conclusions: In a surgical ICU the simplified version of TISS with 28 items (TISS-28) sufficiently reflects the amount of intensive care provided and may provide useful additional information on severity of disease and prognosis. It should replace the original index, at least in these cases.