Sodium–glucose co-transporter-2 inhibitors for hospitalised patients with COVID-19: a prospective meta-analysis of randomised trials

Claire Vale; Peter Godolphin; David Fisher; Peter Horby; Mikhail Kosiborod; Judith Hochman; Katie E Webster; Julian P T Higgins; Andrew Althouse; Otavio Berwanger; Remo Furtado; Samvel Gasparyan; Richard Haynes; Gary Koch; Martin Landray; Eric Leifer; John Marshall; Srinivas Murthy; Matthew Neal; Natalie Staplin; Janet Diaz; Jonathan A C Sterne; Manu Shankar-Hari

doi:10.1016/S2213-8587(24)00219-5

Sodium–glucose co-transporter-2 inhibitors for hospitalised patients with COVID-19: a prospective meta-analysis of randomised trials

Claire Vale, Peter Godolphin, David Fisher, Peter Horby, Mikhail Kosiborod, Judith Hochman, Katie E Webster, Julian P T Higgins, Andrew Althouse, Otavio Berwanger, Remo Furtado, Samvel Gasparyan, Richard Haynes, Gary Koch, Martin Landray, Eric Leifer, John Marshall, Srinivas Murthy, Matthew Neal, Natalie StaplinJanet Diaz, Jonathan A C Sterne, Manu Shankar-Hari^*

^*Corresponding author for this work

Bristol Medical School (PHS)

Research output: Contribution to journal › Article (Academic Journal) › peer-review

2 Citations (Scopus)

Abstract

Background Sodium–glucose co-transporter-2 (SGLT2) inhibitors have been proposed as a potential treatment for
adults hospitalised with COVID-19, due to their potential anti-inflammatory and endothelial protective effects.
Published evidence from randomised control trials (RCTs) does not provide evidence of benefit. We aimed to estimate
the effect of oral administration of SGLT2 inhibitors compared with usual care or placebo in adults hospitalised with
COVID-19.
Methods Eligible RCTs that estimated the effect of oral administration of SGLT2 inhibitors compared with usual care
or placebo on 28-day all-cause mortality (primary outcome) were included in this prospective meta-analysis. The
primary safety outcome was ketoacidosis by 28 days. Trials were identified through systematic searches of
ClinicalTrials.gov, EudraCT, and the WHO ISRCTN registry between Nov 1, 2022 and Jan 31, 2023. The search terms
were “random*” AND “COVID” AND each SGLT2i, not restricted by trial status or language. Individual searches
were then combined. Prespecified summary outcome data, overall and within subgroups of interest, were provided by
each trial. The primary analyses were inverse variance weighted meta-analysis of odds ratios (ORs). Risk of bias was
assessed using the Cochrane Risk of Bias tool. This study was registered with PROSPERO, CRD42023406442.
Findings Three eligible trials randomly assigned 6096 participants (3025 to the SGLT2 inhibitor group and 3071 to the
usual care or placebo group). 2381 (39%) patients were women and 1547 (25%) had type 2 diabetes at randomisation.
By 28 days, there were 351 deaths in the SGLT2 inhibitor group and 382 deaths in the usual care or placebo group
(summary OR 0·93 [95% CI 0·79–1·08]; p=0·33, I² for inconsistency across trials 0%). The risk of bias was assessed
as being low. Ketoacidosis was observed in seven participants in the SGLT2 inhibitor group and two patients in the
usual care or placebo group.
Interpretation Although administration of SGLT2 inhibitor was safe, we found no clear evidence that adding SGLT2
inhibitor therapy improved outcomes in patients hospitalised with COVID-19 compared with usual care or placebo.
These data do not support the use of SGLT2 inhibitors as standard treatment in adults hospitalised for COVID-19

Original language	English
Pages (from-to)	735-747
Number of pages	13
Journal	The Lancet Diabetes and Endocrinology
Volume	12
Issue number	10
Early online date	6 Sept 2024
DOIs	https://doi.org/10.1016/S2213-8587(24)00219-5
Publication status	Published - 1 Oct 2024

Bibliographical note

Publisher Copyright:
© 2024 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/S2213-8587(24)00219-5Licence: CC BY

Handle.net

Persistent link

Cite this

Vale, C., Godolphin, P., Fisher, D., Horby, P., Kosiborod, M., Hochman, J., Webster, K. E., Higgins, J. P. T., Althouse, A., Berwanger, O., Furtado, R., Gasparyan, S., Haynes, R., Koch, G., Landray, M., Leifer, E., Marshall, J., Murthy, S., Neal, M., ... Shankar-Hari, M. (2024). Sodium–glucose co-transporter-2 inhibitors for hospitalised patients with COVID-19: a prospective meta-analysis of randomised trials. The Lancet Diabetes and Endocrinology, 12(10), 735-747. https://doi.org/10.1016/S2213-8587(24)00219-5

@article{6dbc655a7b304b8f87aaad96ea452363,

title = "Sodium–glucose co-transporter-2 inhibitors for hospitalised patients with COVID-19: a prospective meta-analysis of randomised trials",

abstract = "Background Sodium–glucose co-transporter-2 (SGLT2) inhibitors have been proposed as a potential treatment for adults hospitalised with COVID-19, due to their potential anti-inflammatory and endothelial protective effects. Published evidence from randomised control trials (RCTs) does not provide evidence of benefit. We aimed to estimate the effect of oral administration of SGLT2 inhibitors compared with usual care or placebo in adults hospitalised with COVID-19.Methods Eligible RCTs that estimated the effect of oral administration of SGLT2 inhibitors compared with usual care or placebo on 28-day all-cause mortality (primary outcome) were included in this prospective meta-analysis. The primary safety outcome was ketoacidosis by 28 days. Trials were identified through systematic searches of ClinicalTrials.gov, EudraCT, and the WHO ISRCTN registry between Nov 1, 2022 and Jan 31, 2023. The search terms were “random*” AND “COVID” AND each SGLT2i, not restricted by trial status or language. Individual searches were then combined. Prespecified summary outcome data, overall and within subgroups of interest, were provided by each trial. The primary analyses were inverse variance weighted meta-analysis of odds ratios (ORs). Risk of bias was assessed using the Cochrane Risk of Bias tool. This study was registered with PROSPERO, CRD42023406442.Findings Three eligible trials randomly assigned 6096 participants (3025 to the SGLT2 inhibitor group and 3071 to the usual care or placebo group). 2381 (39%) patients were women and 1547 (25%) had type 2 diabetes at randomisation. By 28 days, there were 351 deaths in the SGLT2 inhibitor group and 382 deaths in the usual care or placebo group (summary OR 0·93 [95% CI 0·79–1·08]; p=0·33, I² for inconsistency across trials 0%). The risk of bias was assessed as being low. Ketoacidosis was observed in seven participants in the SGLT2 inhibitor group and two patients in the usual care or placebo group.Interpretation Although administration of SGLT2 inhibitor was safe, we found no clear evidence that adding SGLT2 inhibitor therapy improved outcomes in patients hospitalised with COVID-19 compared with usual care or placebo. These data do not support the use of SGLT2 inhibitors as standard treatment in adults hospitalised for COVID-19",

author = "Claire Vale and Peter Godolphin and David Fisher and Peter Horby and Mikhail Kosiborod and Judith Hochman and Webster, {Katie E} and Higgins, {Julian P T} and Andrew Althouse and Otavio Berwanger and Remo Furtado and Samvel Gasparyan and Richard Haynes and Gary Koch and Martin Landray and Eric Leifer and John Marshall and Srinivas Murthy and Matthew Neal and Natalie Staplin and Janet Diaz and Sterne, {Jonathan A C} and Manu Shankar-Hari",

note = "Publisher Copyright: {\textcopyright} 2024 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license",

year = "2024",

month = oct,

day = "1",

doi = "10.1016/S2213-8587(24)00219-5",

language = "English",

volume = "12",

pages = "735--747",

journal = "The Lancet Diabetes and Endocrinology",

issn = "2213-8587",

publisher = "Lancet Publishing Group",

number = "10",

}

Vale, C, Godolphin, P, Fisher, D, Horby, P, Kosiborod, M, Hochman, J, Webster, KE , Higgins, JPT, Althouse, A, Berwanger, O, Furtado, R, Gasparyan, S, Haynes, R, Koch, G, Landray, M, Leifer, E, Marshall, J, Murthy, S, Neal, M, Staplin, N, Diaz, J, Sterne, JAC & Shankar-Hari, M 2024, 'Sodium–glucose co-transporter-2 inhibitors for hospitalised patients with COVID-19: a prospective meta-analysis of randomised trials', The Lancet Diabetes and Endocrinology, vol. 12, no. 10, pp. 735-747. https://doi.org/10.1016/S2213-8587(24)00219-5

TY - JOUR

T1 - Sodium–glucose co-transporter-2 inhibitors for hospitalised patients with COVID-19

T2 - a prospective meta-analysis of randomised trials

AU - Vale, Claire

AU - Godolphin, Peter

AU - Fisher, David

AU - Horby, Peter

AU - Kosiborod, Mikhail

AU - Hochman, Judith

AU - Webster, Katie E

AU - Higgins, Julian P T

AU - Althouse, Andrew

AU - Berwanger, Otavio

AU - Furtado, Remo

AU - Gasparyan, Samvel

AU - Haynes, Richard

AU - Koch, Gary

AU - Landray, Martin

AU - Leifer, Eric

AU - Marshall, John

AU - Murthy, Srinivas

AU - Neal, Matthew

AU - Staplin, Natalie

AU - Diaz, Janet

AU - Sterne, Jonathan A C

AU - Shankar-Hari, Manu

PY - 2024/10/1

Y1 - 2024/10/1

N2 - Background Sodium–glucose co-transporter-2 (SGLT2) inhibitors have been proposed as a potential treatment for adults hospitalised with COVID-19, due to their potential anti-inflammatory and endothelial protective effects. Published evidence from randomised control trials (RCTs) does not provide evidence of benefit. We aimed to estimate the effect of oral administration of SGLT2 inhibitors compared with usual care or placebo in adults hospitalised with COVID-19.Methods Eligible RCTs that estimated the effect of oral administration of SGLT2 inhibitors compared with usual care or placebo on 28-day all-cause mortality (primary outcome) were included in this prospective meta-analysis. The primary safety outcome was ketoacidosis by 28 days. Trials were identified through systematic searches of ClinicalTrials.gov, EudraCT, and the WHO ISRCTN registry between Nov 1, 2022 and Jan 31, 2023. The search terms were “random*” AND “COVID” AND each SGLT2i, not restricted by trial status or language. Individual searches were then combined. Prespecified summary outcome data, overall and within subgroups of interest, were provided by each trial. The primary analyses were inverse variance weighted meta-analysis of odds ratios (ORs). Risk of bias was assessed using the Cochrane Risk of Bias tool. This study was registered with PROSPERO, CRD42023406442.Findings Three eligible trials randomly assigned 6096 participants (3025 to the SGLT2 inhibitor group and 3071 to the usual care or placebo group). 2381 (39%) patients were women and 1547 (25%) had type 2 diabetes at randomisation. By 28 days, there were 351 deaths in the SGLT2 inhibitor group and 382 deaths in the usual care or placebo group (summary OR 0·93 [95% CI 0·79–1·08]; p=0·33, I² for inconsistency across trials 0%). The risk of bias was assessed as being low. Ketoacidosis was observed in seven participants in the SGLT2 inhibitor group and two patients in the usual care or placebo group.Interpretation Although administration of SGLT2 inhibitor was safe, we found no clear evidence that adding SGLT2 inhibitor therapy improved outcomes in patients hospitalised with COVID-19 compared with usual care or placebo. These data do not support the use of SGLT2 inhibitors as standard treatment in adults hospitalised for COVID-19

AB - Background Sodium–glucose co-transporter-2 (SGLT2) inhibitors have been proposed as a potential treatment for adults hospitalised with COVID-19, due to their potential anti-inflammatory and endothelial protective effects. Published evidence from randomised control trials (RCTs) does not provide evidence of benefit. We aimed to estimate the effect of oral administration of SGLT2 inhibitors compared with usual care or placebo in adults hospitalised with COVID-19.Methods Eligible RCTs that estimated the effect of oral administration of SGLT2 inhibitors compared with usual care or placebo on 28-day all-cause mortality (primary outcome) were included in this prospective meta-analysis. The primary safety outcome was ketoacidosis by 28 days. Trials were identified through systematic searches of ClinicalTrials.gov, EudraCT, and the WHO ISRCTN registry between Nov 1, 2022 and Jan 31, 2023. The search terms were “random*” AND “COVID” AND each SGLT2i, not restricted by trial status or language. Individual searches were then combined. Prespecified summary outcome data, overall and within subgroups of interest, were provided by each trial. The primary analyses were inverse variance weighted meta-analysis of odds ratios (ORs). Risk of bias was assessed using the Cochrane Risk of Bias tool. This study was registered with PROSPERO, CRD42023406442.Findings Three eligible trials randomly assigned 6096 participants (3025 to the SGLT2 inhibitor group and 3071 to the usual care or placebo group). 2381 (39%) patients were women and 1547 (25%) had type 2 diabetes at randomisation. By 28 days, there were 351 deaths in the SGLT2 inhibitor group and 382 deaths in the usual care or placebo group (summary OR 0·93 [95% CI 0·79–1·08]; p=0·33, I² for inconsistency across trials 0%). The risk of bias was assessed as being low. Ketoacidosis was observed in seven participants in the SGLT2 inhibitor group and two patients in the usual care or placebo group.Interpretation Although administration of SGLT2 inhibitor was safe, we found no clear evidence that adding SGLT2 inhibitor therapy improved outcomes in patients hospitalised with COVID-19 compared with usual care or placebo. These data do not support the use of SGLT2 inhibitors as standard treatment in adults hospitalised for COVID-19

U2 - 10.1016/S2213-8587(24)00219-5

DO - 10.1016/S2213-8587(24)00219-5

M3 - Article (Academic Journal)

C2 - 39250923

SN - 2213-8587

VL - 12

SP - 735

EP - 747

JO - The Lancet Diabetes and Endocrinology

JF - The Lancet Diabetes and Endocrinology

IS - 10

ER -

Sodium–glucose co-transporter-2 inhibitors for hospitalised patients with COVID-19: a prospective meta-analysis of randomised trials

Abstract

Bibliographical note

UN SDGs

Access to Document

Handle.net

Fingerprint

Cite this