Table 3.
Characteristics of patients included in the target trial emulation of ECMO versus no ECMO
| Unique patients | Final cohorta | |||
|---|---|---|---|---|
| ECMO (N = 130) | No ECMO (N = 1167) | ECMO (N = 130) | No ECMO (N = 3565) | |
| Demographic characteristics | ||||
| Age (years) | ||||
| Median (IQR) | 49 (41–58) | 58 (49–64) | 49 (41–58) | 58 (48–64) |
| 18–49—no. (%) | 66 (50.8) | 308 (26.4) | 66 (50.8) | 1015 (28.5) |
| 50–59—no. (%) | 41 (31.5) | 350 (30) | 41 (31.5) | 1060 (29.7) |
| 60–70—no. (%) | 23 (17.7) | 509 (43.6) | 23 (17.7) | 1490 (41.8) |
| Male sex—no. (%) | 95 (73.1) | 757 (64.9) | 95 (73.1) | 2337 (65.6) |
| White race—no. (%) | 51 (39.2) | 402 (34.4) | 51 (39.2) | 1252 (35.1) |
| Body mass index (kg/m2) | ||||
| Median (IQR) | 32.5 (29.5–37.9) | 32.5 (28.1–39.1) | 32.5 (29.5–37.9) | 32.8 (28.4–39.5) |
| < 40—no. (%) | 98 (75.4) | 867 (74.3) | 98 (75.4) | 2618 (73.4) |
| ≥ 40—no. (%) | 21 (16.2) | 261 (22.4) | 21 (16.2) | 831 (23.3) |
| Unknown—no. (%) | 11 (8.5) | 39 (3.3) | 11 (8.5) | 116 (3.3) |
| Coexisting conditions | ||||
| Hypertension | 62 (47.7) | 682 (58.4) | 62 (47.7) | 2067 (58) |
| Diabetes mellitus | 38 (29.2) | 516 (44.2) | 38 (29.2) | 1524 (42.7) |
| Coronary artery disease | 4 (3.1) | 120 (10.3) | 4 (3.1) | 350 (9.8) |
| Congestive heart failure | 2 (1.5) | 108 (9.3) | 2 (1.5) | 324 (9.1) |
| Chronic obstructive pulmonary disease | 4 (3.1) | 86 (7.4) | 4 (3.1) | 264 (7.4) |
| Severity of illnessb | ||||
| PaO2/FiO2, mmHg—median (IQR) | 80 (65–99) | 90 (70–128) | 69 (60–80) | 78 (66–89) |
| PEEP—median (IQR) | 15 (12–18) | 14 (10–16) | 15 (12–18) | 14 (12–18) |
| Shockc—no. (%) | 81 (62.3) | 622 (53.3) | 104 (80) | 2352 (66) |
| Lactate, mmol/l—median (IQR) | 1.7 (1.2–2.5) | 1.6 (1.1–2.3) | 2.0 (1.4–3.1) | 1.5 (1.1–2.1) |
| Arterial pH—median (IQR) | 7.35 (7.29–7.45) | 7.36 (7.28–7.42) | 7.33 (7.28–7.39) | 7.34 (7.27–7.39) |
| Secondary infection—no. (%) | 11 (8.5) | 83 (7.1) | 26 (20) | 595 (16.7) |
| Renal SOFA score—no. (%)d | ||||
| 0 (Cr < 1.2 mg/dl) | 84 (64.6) | 664 (56.9) | 67 (51.5) | 1595 (44.7) |
| 1 (Cr 1.2–1.9 mg/dl) | 29 (22.3) | 250 (21.4) | 34 (26.2) | 752 (21.1) |
| 2–4 (Cr > 2 mg/dl, UOP < 500 ml, RRTe) | 17 (13.1) | 253 (21.7) | 29 (22.3) | 1218 (34.2) |
| Liver SOFA score—no. (%)d | ||||
| 0 (Bilirubin < 1.2 mg/dl) | 114 (87.7) | 1058 (90.7) | 109 (83.8) | 3088 (86.6) |
| 1 (Bilirubin 1.2– 1.9 mg/dl) | 11 (8.5) | 80 (6.9) | 13 (10) | 292 (8.2) |
| 2–4 (Bilirubin ≥ 2 mg/dl) | 5 (3.8) | 29 (2.5) | 8 (6.2) | 185 (5.2) |
| Coagulation SOFA score—no. (%)d | ||||
| 0 (Platelet count ≥ 150 K/mm3) | 109 (83.8) | 995 (85.3) | 108 (83.1) | 3111 (87.3) |
| 1 (Platelet count 100–149 K/mm3) | 15 (11.5) | 135 (11.6) | 17 (13.1) | 335 (9.4) |
| 2–4 (Platelet count < 100 K/mm3) | 6 (4.6) | 37 (3.2) | 5 (3.8) | 119 (3.3) |
| Rescue therapies for hypoxemia—no. (%)f | ||||
| Prone position ventilation | 56 (43.1) | 249 (21.3) | 92 (70.8) | 1651 (46.3) |
| Neuromuscular blockade | 52 (40) | 234 (20.1) | 100 (76.9) | 1678 (47.1) |
| Inhaled vasodilators | 23 (17.7) | 49 (4.2) | 47 (36.2) | 479 (13.4) |
PaO2 partial pressure of arterial oxygen over the fraction of inspired oxygen, PEEP positive end expiratory pressure, RRT renal replacement therapy, SOFA Sequential Organ Failure Assessment, UOP urine output
aThe number of observations in the final cohort differs from the number of unique patients because more than one observation per patient was used, thereby creating a pseudo-cohort. This approach (described further in the supplemental methods) was used to eliminate the potential for immortal time bias
bSeverity of illness data are shown on the day of ICU admission for the unique patients and on the day of ECMO initiation or non-initiation for the final cohort
cShock is defined as the requirement for at least one vasopressor
dCategories 2, 3, and 4 of the renal, liver, and coagulation components of the SOFA score were binned due to low frequency of events
eIncludes both acute RRT as well as end-stage renal disease requiring RRT
fRescue therapies for hypoxemia were assessed on the day of ICU admission for the unique patients and up to and including the day of ECMO initiation or non-initiation for the final cohort