Table 3 Summary of findings related to impact of VOC on admission to ICU

Alpha

• In both the adjusted and unadjusted analysis, the primary care group had a higher risk of admission to critical care for Alpha patients compared with the non-Alpha patients (adjusted HR: 2.15; 95% CI 1.75 - 2.65). There was no significant interaction between Alpha and sex, ethnic group, or age group (Patone et al., England, Nov 2020-Jan 2021, high quality)[68]

• In the Alpha wave, almost half of total ICU patients were admitted (803 out of the total 1680 patients), suggesting greater ICU admission than pre-VOC (Area et al., Spain, March-April 2021, high quality)[62]

• The probability of ICU admission was twice as high among patients with Alpha compared to wild-type (OR 2.11, 95 CI% 1.55 − 2.87) (Martinez-Garcia, Spain, Jan-April 2021, medium quality)[50]

• ICU admission was higher with Alpha (6.3%) compared to non-Alpha (3.4%) (Pascall et al., Scotland, Nov 2020-Jan 2021, high quality)[67]

• Significantly more hospitalized Alpha patients were transferred to ICU (27.7%) compared to 8.6% of non-Alpha patients (Vassallo et al., France, Oct 2020-Apr 2021, medium quality)[38]

• Alpha was associated with a 1.8-fold increased risk of ICU admission compared to non-VOC (aRR 95% CI 1.2-2.8) (Veneti et al., Norway, Dec 2020-Jun 2021, high quality)[71]

• There was no significant difference between those admitted to the ICU before Alpha was dominant (23%) compared to after (26 and 35%), p = 0.374. For ICU patients, neither the severity score at admission (SAPSII) nor the depth of the respiratory distress seemed to increase with Alpha (Courjon et al., France, Dec 2020-Feb 2021, medium quality)[47]

• No difference was found for progressing to critical disease between wild-type and Alpha (Abu-Raddad, Qatar, Jan-May 2021, medium quality)[75]

• There was no difference in ICU admission between pre-Alpha and Alpha wave (Cusinato et al., UK, January 2020-March 2021, high quality)[55]

• No overall increase in ICU admission was associated Alpha (HR 1.01, (95% CI 0.75-1.37, p = 0.94); however, women with Alpha may be at an increased risk of admission to ICU (HR 1.82, 95% CI 1.15-2.90, p = 0.011) (Stirrup et al., UK, Nov 2020-Jan 2021, high quality )[69]

• There was no change in the proportion of children and young people admitted to critical care between the pre-Alpha (12.9%) and Alpha wave (12.7%) (Swann et al., UK, Jan 2020-Jan 2021, medium quality)[37]

• There was no difference between Alpha patients (16%) and non-VOC patients (18%) admitted to the ICU (Whittaker et al., Norway, Dec 2020-Apr 2021, high quality)[73]

Beta

• Compared to Alpha, the odds of progressing to critical disease were 1.49-fold (95% CI 1.13-1.97) higher for Beta (Abu-Raddad, Qatar, Jan-May 2021, medium quality)[75]

• Beta was associated with a 2.7-fold increased risk of ICU admission compared to non-VOC (aRR 95% CI 1.2-6.5) (Veneti et al., Norway, Dec 2020-Jun 2021, high quality)[71]

• The proportion of patients admitted to the ICU was lower in the Beta wave (35.0%) compared to the pre-Beta wave (48.5%), p < 0.001 (Maslo et al., South Africa, June-Dec 2020, medium quality)[44]

• No differences for patients admitted to ICU with Beta (7 days) or wild-type (8 days) in terms of days between onset of symptoms (Pascall et al., Scotland, Nov 2020-Jan 2021, high quality)[67]

Gamma

• There were more patients admitted to the ICU during the Gamma wave (943) than the pre-Gamma wave (672), particularly among those under 60 years of age and without comorbidities (Nonaka et al., Brazil, May 202-Feb 2021, high quality)[74]

• While there was variation in the age profile of hospitalized patients between Feb 2020-Feb 2021, there was no evidence of an increase in hospitalization in the last period (related to high Gamma) for adults between 18 and 50 years (de Andrade et al., Brazil, Feb 2020-Feb 2021, low quality)[24]

• There was no significant difference in the ICU admission risk for maternal patients between 2020 and 2021 (p = 0.769) and there was no difference in length of ICU stay (p = 0.269) (Takemoto et al., Brazil, Mar 2020-Apr 12th, 2021, medium quality)[52]

• There was no difference between ICU admission between Gamma and non-Gamma patients (Zavascki et al., Brazil, June 2020-May 2021, medium quality)[39]

Delta

• Over a quarter of pregnant patients diagnosed during high spread of Delta required admission for severe or critical illness, compared to 5.4% before Delta was prominent (Adhikari et al., United States, May-Sept 2021, medium quality)[45]

• Similar amounts of patients were admitted to the ICU in both waves (34.9% in Wave 1 vs. 33.4% in Wave 2 [Delta]), although more patients required oxygen (74.1% vs 63.4%, p < 0.001) and invasive ventilation during the Delta wave (10.1% in Wave 2 vs. 8.7% in Wave 1, p = 0.002) (Budhiraja et al., India, April 2020-June 2021, medium quality)[29]

Combined VOC

• There was an estimated increase of 1.29 (95% CI 0.5-2.1, p < 0.05) additional COVID-19 patients in intensive care per 100,000 population, which is a 42% rise in hospitalization in VOC regions compared to pre-VOC regions (3.08 patients in intensive care per 100,000 population) (Mitze and Rode, Germany, Jan-Feb 2021, no appraisal)[77]

• VOC cases were more likely to be admitted to the ICU than non-VOC cases (Alpha: 1.4%, p = .002; Beta: 2.3%, p = 0.001; Gamma: 2.1%, p = 0.005 vs. non-VOC: 0.6%). In an unmatched analysis, VOC were 2.2-3.3 times more likely to be admitted to ICU than non-VOC. ICU admission did not differ for Alpha but increased for Beta (adjusted OR 8; 95% CI 3.7–17.3) only for those aged 40–59 years. For individuals aged 40 or older, there was a 2.9 to 13.9 times higher odds of ICU admission with Gamma than non-VOC (Funk et al., 7 Europe, Sep 2020-Mar 2021, high quality)[63]

• VOC (Alpha, Beta, Gamma) were associated with higher rates of ICU admission (OR, 3.31; 95% CI 2.84-3.86) compared to previous strains (Erman et al., Canada, January-April 2021, medium quality)[32]

• Increased rates of ICU admission were seen in VOC infections (all four) relative to non-VOC. Adjusted risk was 105% (95% CI 82-134) higher for VOC (Alpha, Gamma, Beta) ICU admission than with non-VOC and 287% (95% CI 198-399) higher for ICU admission due to Delta. Increased ICU admission was seen between Delta and other VOC: 101% (95% CI 79-124) (Fisman et al., Canada, Feb-June 2021, medium quality)[33]

• There were no significant differences between VOC and non-VOC on mean LOS (11.51 vs. 9.56 days), mean critical care LOS (15.25 vs. 18.93 days), or proportion of patients admitted to critical care (0.111 vs. 0.19) (Garvey et al., England, Dec 15th-31st, 2020, high quality) [64]