Table 3.
Results of univariate analysis of potential prognostic factors. Data are presented as median (interquartile range) unless otherwise stated. NS not significant
| Factor | Outcome at 28 days | Significance (p) | |
|---|---|---|---|
| Alive and not ventilated (n=34) | Dead or ventilated (n=20) | ||
| Age (years; mean±SD) | 41±12 | 64±13 | <0.001 |
| Male (%) | 53 | 65 | NS |
| APACHE II | 9 (7–11) | 13 (11–21) | <0.001 |
| Delay between symptom onset and ICU admission (days) | 10 (8–12) | 6 (4–11) | <0.05 |
| Delay between hospital admission and ICU admission (days) | 6 (2–8) | 6 (2–8) | NS |
| Chronic disease or immunosuppression (%) | 15 | 40 | <0.05 |
| Independent in activities of daily life (%) | 91 | 95 | NS |
| PaO2/FIO2 ratio (mean±SD) | 161 (76) | 138 (62) | NS |
| Admission multiple-organ dysfunction score | 3 (2–3) | 3 (3–5) | <0.05 |
| Mean daily steroid dose (mg) | 173 (141–308) | 163 (94–207) | <0.01 |
| Admission lymphocyte count (×106/ml) (mean±SD) | 0.47±0.26 | 0.69±0.52 | <0.05 |
| Change in lymphocyte count (x106/ml; mean±SD) from admission to ICU discharge, death or 28 days (whichever was shortest) | 0.23±0.37 | −0.28±0.56 | <0.001 |
| No. of episodes of nosocomial sepsis | 0 (0–0) | 1 (0–1) | <0.001 |
| Patients with bacteraemia (%) | 2.9 | 25 | <0.05 |
| Mean daily fluid balance (mean±SD) | |||
| Overall | 74±316 | 700±677 | <0.001 |
| First half of ICU stay | 194±480 | 793±773 | <0.001a |
| Second half of ICU stay | −47±431 | 606±883 | <0.001a |
| Admission serum lactate dehydrogenase concentration (IU/l; mean±SD) | 472±299 | 598±245 | NS |
| Change in serum lactate dehydrogenase concentration (IU/l; mean±SD) | −91±302 | −145±266 | NS |
aAnalysis of fluid balance by phase of ICU stay was carried out after initial univariate analysis showing that fluid balance was significantly associated with outcome