Table 2.
Predictors of diffuse alveolar damage in patients with clinical criteria for acute respiratory distress syndrome
| Bivariate test, OR (95% CI), p value | Multivariate logistic regression, OR (95% CI), p value | |
|---|---|---|
| Characteristics of the patients | ||
| Female sex, n (%) | 1.79 (1.15–2.78), p < 0.01 | Not significant |
| Pneumonia, n (%) | 1.54 (0.99–2.39), p = 0.05 | Not significant |
| Duration of evolution of ARDS | ||
| > 72 h | 3.92 (2.43–6.37), p < 0.001 | 3.29 (1.95–5.55), p < 0.001 |
| Severity of hypoxemia | <0.0001 | |
| Mild ARDS (n = 49) | Reference | |
| Moderate ARDS (n = 141) | 4.72 (1.88–11.8), p < 0.001 | 3.92 (1.48–10.3), p = 0.006 |
| Severe ARDS (n = 166) | 10.1 (4.06–24.9), p < 0.001 | 6.18 (2.34–16.3), p < 0.001 |
| Ancillary variables of severity | ||
| PEEP ≥ 10 cmH2O | 2.33 (1.51–3.59), p < 0.001 | Not significant |
| Dynamic compliance < 40 ml/cmH2O | 5.12 (1.73–15.1), p < 0.01 | Not significant |
| VE corrected PaCO2 ≥ 10 L/minute | 2.34 (1.40–3.93), p < 0.01 | Not significant |
| PaCO2 > 45 mmHg | 1.73 (1.14–2.65), p = 0.01 | Not significant |
| Dynamic driving pressure (peak − PEEP), cmH2O | 1.06 (1.04–1.09), p < 0.001 | 1.06 (1.04–1.09), p = 0.007 |
| Extent of infiltrates on chest radiograph | <0.0001 | |
| Two quadrants | Reference | |
| Three quadrants | 2.45 (1.26–4.76), p < 0.01 | 1.63 (0.79–3.39), p = 0.19 |
| Four quadrants | 5.43 (2.93–10.1), p < 0.0001 | 2.91 (1.47–5.75), p = 0.002 |
Abbreviations: ARDS Acute respiratory distress syndrome, DAD Diffuse alveolar damage, PEEP Positive end-expiratory pressure, VE Minute ventilation; PaCO2, Partial pressure of arterial carbon dioxide
All variables nonredundant associated with DAD with a p value < 0.15 were included in the maximal model, including severity score, sex, pneumonia as a risk factor for ARDS, duration of ARDS evolution > 72 h, severity of hypoxemia (mild, moderate, or severe using mild as the reference), PEEP ≥ 10 cmH2O, dynamic compliance of the respiratory system < 40 ml/cmH2O, expired minute volume corrected by PaCO2 ≥ 10 L/minute, hypercapnia defined as PaCO2 > 45 mmHg, dynamic driving pressure defined as peak pressure minus PEEP, and the extent of radiological infiltrates (two, three, or four quadrants using two quadrants as the reference)
Bold data are variables that remained significanltly associated with DAD after multivariate logistic regression