The SARS-CoV-2 B.1.1.7 variant and increased clinical severity—the jury is out

Dan Frampton and colleagues¹ demonstrated increased viral load but not severity of disease or 28-day mortality in hospitalised patients infected by the B.1.1.7 variant of SARS- CoV-2. By contrast, we found slightly different results when assessing the risk of morbidity and mortality in a matched case-control study in 60 hospitalised patients—30 with the B.1.1.7 variant and 30 with non-B.1.1.7 variants. Cases were matched for admission period and age band. Clinical severity scores, requirement for ventilation, treatments received, and 28-day mortality were compared between groups using anonymised, retrospectively collected data and Wilcoxon rank-sum and χ² or Fisher's exact tests.

Our findings (table ) show consistent and rational evidence that patients infected with the B.1.1.7 variant developed more serious disease: they had greater clinical severity (eg, higher National Early Warning Score value, lower respiratory rate oxygenation index) and greater requirement for supplemental oxygen and mechanical ventilation, they more often received approved treatments for SARS-CoV-2 infection (eg, dexamethasone, remdesivir, and tocilizumab), and they had more serious clinical outcomes (ie, higher 28-day mortality, WHO clinical progression scale score). Although our results show a tendency towards severe disease with B.1.1.7 infection, it is likely that our study was underpowered as statistical significance was seen only for patients requiring dexamethasone. Nevertheless, our data echo the findings of other, larger studies. For example, Challen and colleagues,² who studied a younger population than described here or studied by Frampton and colleagues, with likely less comorbidity, found a 64% increase in 28-day mortality following community infection with the B.1.1.7 variant (control group, 0·26%; B.1.1.7 variant group, 0·41%). Similarly, Davies and colleagues³ concluded that infections with the B.1.1.7 variant were associated with a hazard of death of 61% (95% CI 42–82) higher than with pre-existing variants.

Table.

Demographics and outcomes of cases of SARS-CoV-2 infection with B.1.1.7 variant compared with non-B.1.1.7 variants

		n	Infection with B.1.1.7 variant	n	Infection with a non-B.1.1.7 variant	p value
Date of first positive PCR swab		30	Dec 3–20, 2020	30	Oct 10–Dec 20, 2020	..
Age (years)		30	77 (59–88)	30	79 (59–87)	0·976
Sex		30	..	30	..	0·436
	Male	..	15 (50%)	..	18 (60%)	..
	Female	..	15 (50%)	..	12 (40%)	..
Number of comorbidities		30	2 (1–3)	30	2 (1–3)	0·845
White ethnicity (%)		30	26 (87%)	30	30 (100%)	0·112
NEWS2*		30	..	30	..	..
	At presentation	..	4 (2–7)	..	2·5 (1–6)	0·135
	Maximum value	..	6 (4–8)	..	5 (3–9)	0·345
Respiratory rate oxygenation index		..	..	..	..	..
	At presentation	30	20 (15–26)	30	24 (15–27)	0·371
	At maximum FiO2	25	15 (11–21)	26	18 (13–26)	0.341
Sequential Organ Failure Assessment score		30	..	30	..	..
	At presentation	..	3·0 (2–7)	..	3·5 (2–6)	0·858
	At maximum FiO2	..	5·5 (2–7)	..	5·0 (2–7)	0·566
	4C Mortality Score	30	..	30	..	..
	At presentation	..	12·0 (9·0–14·8)	..	10·5 (9·0–14·0)	0·568
	At maximum FiO2	..	12·5 (8·3–14·0)	..	11·5 (9·0–13·0)	0·463
Maximum ventilatory support received		30	..	30	..	0·265
	Mechanical ventilation	..	3 (10%)	..	1 (3%)	..
	Non-invasive ventilation	..	0 (0%)	..	1 (3%)	..
	Standard oxygen therapy	..	18 (60%)	..	14 (47%)	..
	No supplemental oxygen required	..	9 (30%)	..	14 (47%)	..
Treatment		..	..	..	..	..
	Dexamethasone	18	13 (72%)	24	10 (42%)	0·049
	Remdesivir	14	2 (14%)	22	1 (5%)	0·547
	Anticoagulation	24	4 (17%)	30	8 (27%)	0·380
	Tocilizumab	28	1 (4%)	28	0 (0%)	1·000
28-day mortality (%, 95% CI)		28	9 (32·1%, 17·9–50·7)	29	6 (20·7%, 9·8–38·4)	0·326
Patients with a severe clinical outcome†		30	11 (37%)	30	8 (27%)	0·405

Data are n (%) or median (IQR), unless otherwise stated. F_iO₂=fraction of inspired oxygen.

^*NEWS=National Early Warning Score. *The NEWS2 score was not calculated at maximum F_iO₂.

^†A severe clinical outcome was defined as a WHO scale score by day 14 after symptom onset or the first positive SARS-CoV-2 PCR of at least 6 or death within 28 days.

We believe that the identification of any increased morbidity and mortality risks in patients infected by SARS-CoV-2 variants of concern requires adequately powered studies that use a combination of community and hospital PCR swabs, examine disease severity using more detailed clinical severity scores (such as the WHO clinical progression ordinal scale used by Frampton and colleagues) with physiological measures, and assess the impact of viral load, novel treatments, and vaccinations. For the purposes of future case-control studies, we estimate post-hoc that a sample size of 234 patients in each group is required to detect an effect size of 11·4% in 28-day mortality for a baseline mortality in the control group of 20·7% at 80% power with 5% significance. We believe the jury is still out on whether B.1.1.7 infections are associated with increased mortality, with more time and data required.

AC reports grants from NIHR, Asthma UK, Boehringer-Ingelheim Charity; consulting fees and honoraria from Sanofi; and support for meeting attendance from GSK. All other authors declare no competing interests.