Table 1.
Characteristics of the Patients Included in the Study
| Whole Cohort | No ACEI/ARB | ACEI/ARB | P Valuea | ||||
|---|---|---|---|---|---|---|---|
| Age—N, median (IQR) | 149 | 65 (54–77) | 105 | 63 (51–74) | 44 | 70 (63 to 82) | .005b |
| Male sex—n/N, %, (95% CI) | 91/149 | 61 (53–69) | 63/105 | 60 (51–70) | 28/44 | 64 (49–78) | .68 |
| Patients’ medical history—n/N, %, (95% CI) | |||||||
| Hypertensionc | 66/133 | 50 (41–58) | 29/90 | 32 (22–42) | 37/43 | 86 (75–97) | <.0001 |
| Cardiovascular disease | 38/133 | 29 (21–36) | 17/90 | 19 (11–27) | 21/43 | 49 (33–64) | .0004 |
| Type 2 diabetesc | 38/133 | 29 (21–36) | 13/90 | 14 (7–22) | 25/43 | 58 (43–74) | <.0001 |
| Vascular disease | 36/133 | 27 (19–35) | 20/90 | 22 (14–31) | 16/43 | 37 (22–52) | .07 |
| Dyslipidemia | 30/133 | 23 (15–30) | 18/90 | 20 (12–28) | 12/43 | 28 (14–42) | .31 |
| Obstructive sleep apnea syndrome | 17/133 | 13 (7–19) | 9/90 | 10 (4–16) | 8/43 | 19 (6–31) | .17 |
| COPD | 15/133 | 11 (6–18) | 7/90 | 8 (2–13) | 8/43 | 19 (6–31) | .07 |
| Asthma | 8/133 | 6 (2–10) | 5/90 | 6 (1–10) | 3/43 | 7 (0–15) | .75 |
| Cancer | 8/133 | 6 (2–10) | 6/90 | 7 (1–12) | 2/43 | 5 (0–11) | .65 |
| Chronic kidney disease | 8/133 | 6 (2–10) | 4/90 | 4 (0–9) | 4/43 | 9 (0–18) | .27 |
| Outcomes—n/N, %, (95% CI) | |||||||
| Acute respiratory failure | 76/146 | 52 (44–60) | 50/103 | 49 (39–58) | 26/43 | 61 (45–76) | .19 |
| Intubation and mechanical ventilation | 54/146 | 37 (29–45) | 34/103 | 33 (24–42) | 20/43 | 47 (31–62) | .12 |
| COVID-19 related death | 19/147 | 13 (7–18) | 9/104 | 9 (3–14) | 10/43 | 23 (10–36) | .02 |
| Pulmonary embolism | 2/146 | 1 (0–3) | 1/103 | 1 (0–3) | 1/43 | 2 (0–7) | .52 |
Abbreviations: ACEI/ARB, angiotensin-converting enzyme inhibitor/angiotensin receptor blockers; CI, confidence interval; COPD, chronic obstructive pulmonary disease; IQR, interquartile range, 25th–75th percentile; n, number of observations; N, number of patients.
aχ 2 test or Fisher exact test, as appropriate.
bMann-Whitney U test.
cHypertension and type 2 diabetes were significantly correlated (Spearman rank correlation coefficient = 0.378; P < .0001). To avoid the multicollinearity issue in the multivariable models, hypertension and type 2 diabetes were assessed separately in the logistic regression analysis (model 1: hypertension, cardiovascular disease; model 2: type 2 diabetes, cardiovascular disease) and the multivariable multilevel analysis (model 1: type 2 diabetes; model 2: hypertension).