Table 5.
In-hospital ventilator management and complications of the severe COVID-19 ARDS patients on VV-ECMO between survivors vs. non-survivors.
| Characteristics | Survivors (n = 21) | Non-survivors (n = 26) | P-valuea |
|---|---|---|---|
| ECMO cannulation since intubationb (days) | 4 (2, 5) | 4.5 (2, 6) | 0.57 |
| Tracheostomy, n (%) | 18 (85.7) | 25 (96.2) | 0.31c |
| Tracheostomy placement since ECMO cannulationb (days) | 14 (10.5, 16) | 11 (8, 14) | 0.17 |
| Total ventilator daysb (days) | 49 (30, 59) | 48.5 (23.3, 83.3) | 0.55 |
| Length of hospitalizationb (days) | 60 (43, 80) | 45 (21.5, 86.3) | 0.35 |
| Complications | |||
| RV dysfunction, n (%) | 3 (14.3) | 13 (50.0) | 0.01c |
| AKI, n (%) | 7 (33.3) | 10 (38.5) | 0.72 |
| Secondary infection, n (%) | 18 (85.7) | 20 (76.9) | 0.71c |
| Bleeding, n (%) | 15 (71.4) | 23 (88.5) | 0.26c |
| ICH, n (%) | 4 (19.0) | 7 (26.9) | 0.73c |
| Fatal ICHd, n (%) | 0 (0) | 5 (19.2) | 0.056c |
| Thrombosis, n (%) | 9 (42.9) | 5 (19.2) | 0.11c |
AKI, acute kidney injury; ECMO, extracorporeal membrane oxygenation; ICH, intracerebral hemorrhage; IQR, interquartile range; RV, right ventricle. ap < 0.05 was considered statistically significant. bValues were presented as median (IQR). Wilcoxon rank-sum test was used. cFisher's exact test was used. dFatal intracerebral hemorrhage was defined as those resulting in unsalvageable injury (i.e., brain herniation) and/or clinical diagnosis of brain death.