We extracted data for 276 patients who were admitted to the ICU with COVID-19 who were positive or became swab positive during their admission (Consort diagram, Fig. 1). Of these, five (1.8%) died within 6 hours of ICU admission and would not have been considered for beta blockade. A further 14 (5.1%) never reached the entry criterion of a heart rate greater than 90 bpm for 12 hours, while a further 22 (8.0%) were excluded because they were receiving beta blockers before admission to ICU. This left a cohort of 235 (85.1%) patients, 144 of whom received norepinephrine and were included in the analysis.
The characteristics of patients are summarized in Table 1. Average age was 56.1 years (SD 13.9), 102 (70.8%) were male, 54 (37.5%) were of white ethnicity and 90 (62.5%) were of non-white ethnicity. Patients had a mean APACHE II score of 14.3 (SD 4.0), a mean heart rate of 99.3 (SD 13.1) and mean CRP 183 (IRQ 111, 267). The predominance of males is consistent with national UK data (31). Males were also more likely to receive beta blocker treatment and this precluded a matched analysis. The only significant difference was a lower final lymphocyte count in patients receiving beta blockade (1.2 vs 1.6 x10− 9/ml, p = 0.042).
Table 1
Baseline demographic and clinical characteristics summarized by beta blockade status
Criteria | No Beta Blockade (N = 61) | Beta Blockade (N = 83) | Overall (N = 144) | P-value |
Age (years), Mean (SD) | 57.1 (12.8) | 55.3 (12.5) | 56.1 (13.9) | 0.399 |
Gender (male), N (%) | 35 (57.4) | 67 (80.7) | 102 (70.8) | 0.002 |
Ethnicity, N (%) | | | | |
White | 23 (37.7) | 31 (37.4) | 54 (37.5) | 0.965 |
Non-white | 38 (62.3) | 52 (62.6) | 90 (62.5) | |
Steroid use (Yes), N (%) | 32 (52.5) | 46 (55.4) | 78 (54.2) | 0.724 |
Apache II, Mean (SD) | 14.3 (4.4) | 14.3 (4.0) | 14.3 (4.0) | 0.970 |
Time from ICU Admission to Heart Rate > = 90 Median (IQR) Days | | | | |
| 0.58 (0.18, 1.48) | 0.38 (0.16, 1.23) | 0.50 (0.17, 1.32) | 0.422 |
Norepinephrine Dose Median (IQR) mcg/kg/min | | | | |
Start* | 0.06 (0.03, 0.10) | 0.07 (0.03, 1.04) | 0.07 (0.03, 0.10) | 0.547 |
Heart Rate Mean (SD) bpm | | | | |
Start | 99.2 (13.1) | 99.4 (13.3) | 99.3 (13.1) | 0.904 |
End | 95.8 (22.7) | 94.0 (18.1) | 94.8 (20.1) | 0.593 |
C-Reactive Protein Median (IQR) mg/L | | | | |
Start | 194 (141, 280) | 173 (101, 252) | 183 (111, 267) | 0.200 |
End | 97 (18, 215) | 61 (14, 174) | 63 (16, 197) | 0.167 |
Lymphocyte Count Median (IQR) x 10− 9 /ml | | | | |
Start | 1.0 (0.6, 1.3) | 1.0 (0.7, 1.5) | 1.0 (0.6, 1.4) | 0.382 |
End | 1.2 (0.9, 1.9) | 1.6 (1.1, 2.2) | 1.4 (1.0, 2.1) | 0.042 |
Neutrophil Count Median (IQR) x 10− 9 /ml | | | | |
Start | 8.8 (6.7, 12.0) | 8.3 (5.9, 13.1) | 8.45 (6.3, 12.4) | 0.805 |
End | 9.4 (6.3, 13.9) | 8.1 (5.7, 14.5) | 8.8 (5.9, 14.3) | 0.522 |
Neutrophil:Lymphocyte Ratio Median (IQR) | | | | |
Start | 8.27 (6.36, 13.45) | 8.30 (5.58, 15.23) | 8.28 (5.85 13.71) | 0.528 |
End | 7.84 (3.95, 13.94) | 4.77 (3.18, 12.33) | 6.01 (3.33, 12.94) | 0.079 |
IQR: 25% and 75% |
* All alive patients had been weaned from Norepinephrine at 28-days |
For the beta blocker group, the mean time from entry criteria to starting beta blockers was 7.91 days (IQR 3.89, 13.15) (Table 2). The median length of treatment was seven days with 25% of patients receiving fewer than four days treatment (IQR 4.0–14.0). Beta blockers were used for 396 days (17.4%) out of a possible follow up period of 28 days in 83 patients (Total possible 2324 days) and only 46 of the 83 (55.4%) patients demonstrated a reduction in heart rate of 10% or more of baseline in the 24 hours following initiation.
Table 2
Beta Blocker Compliance with Therapy
Time from Cohort Trial Entry to Beta Blocker Median (IQR) Days | 7.91 (3.89, 13.15) |
Duration of Beta Blocker Therapy Median (IQR) Days | 7.0 (4.0, 14.0) |
Number of Beta Blocker Treated patients with 10% decrease of Heart Rate at 24 hours | 46/83 (55.4%) |
Of the 83 beta blocker patients, 24 (28.9%) died within 28 days compared with 47.5% of those not treated with beta blockers. From the adjusted primary analyses, there was a statistically significant association with beta blocker treatment and 28-day mortality (adjusted odds ratio 0.43 95% (CI 0.20–0.95) p = 0.036) (Table 3 and Fig. 2). To compensate for the risk of immortal time bias, we performed Cox regression with time-dependent covariate analysis. This suggested that delay in exposure to beta blockade was associated with an increased risk of death, though this did not achieve statistical significance (Hazard Ratio 1.42 p = 0.264). Exposure to beta blockade was associated with an adjusted odds ratio for death of 0.76 (95% CI 0.64–0.91; P = 0.002) for each day treated with beta blockade.
Table 3
Mortality data summarised by beta blockade status
Analysis | No Beta Blockade (N = 61) | Beta Blockade (N = 83) | |
Cox Regression Analysis Died in ICU within 28 days N (%) | 29/61 (47.5%) | 24/83 (28.9%) | Adjusted OR (95% CI); P-value: 0.43 (0.20, 0.95); P = 0.036 |
Cox Regression with Time Dependent Analysis Died in ICU within 28 days N (%) | Exp(B); P-value: 1.42; P = 0.264 |
Logistic Regression of Beta Blocker Exposure No of Days out of total possible (%) | | 396/2324 (17.4%) | Adjusted OR (95% CI); P-value: 0.76 (0.64, 0.91); P = 0.002 |
Length of ICU stay (days), median (IQR) | 13.7 (6.6, 21.4) | 23.8 (15.8, 33.1) | < 0.001 |
Length of Hospital stay (days), median (IQR) | 19.8 (10.1, 34.2) | 34.2 (22.8, 47.5) | < 0.001 |
Adjusted for the variables Age, Gender, APACHE-II, Ethnic Status, Steroid use and Log Transformed Norepinephrine Dose at Trial Start |
In both Cox Regression analyses, Gender and Log Transformed Norepinephrine Dose were significant at P < 0.05 |
For the secondary outcomes (Table 3), there was a statistically significant difference in the median lengths of ICU stay (beta blockade group 23.8 days (IQR 15.8, 33.1), no beta-blockade group 13.7 days (IQR 6.6, 21.4) (p < 0.001), and hospital stay 34.2 days (IQR 22.8, 47.5) versus 19.8 days (IQR 10.1, 34.2) (p < 0.001) respectively.