Insulin Therapy and Outcome of Coronavirus Disease 2019 (COVID-19): A Systematic Review, Meta-Analysis, and Meta-Regression

Background: Currently, the relationship between insulin therapy and COVID-19 outcome is not yet established. Our study aims to evaluate the possible association between insulin and the composite poor outcome of COVID-19. Methods: We systematically searched the PubMed and Europe PMC database using specific keywords related to our aims until December 12th, 2020. All articles published on COVID-19 and insulin were retrieved. Statistical analysis was done using Review Manager 5.4 and Comprehensive Meta-Analysis version 3 software. Results: Our pooled analysis showed that insulin use was associated with composite poor outcomes of COVID-19 [OR 2.06 (95% CI 1.70 - 2.48), p < 0.00001, I2 = 83%, random- effect modelling], and its subgroup which comprised of risk of COVID-19 [OR 1.70 (95% CI 1.40 - 2.08), p < 0.00001, I2 = 34%, random-effect modelling], severe COVID-19 [OR 2.30 (95% CI 1.60 - 3.30), p < 0.00001, I2 = 88%, random-effect modelling], and mortality [OR 2.14 (95% CI 1.47 - 3.10), p < 0.0001, I2 = 85%, random-effect modelling]. Meta-regression showed that the association was influenced by age (p = 0.008), but not by diabetes p = 0.423) and cardiovascular disease (p = 0.086). Conclusion: Physicians should be more aware and take extra precautions with diabetes patients who use insulin therapy.

Automated summary

This study aims to evaluate the association between insulin therapy and poor outcomes of COVID-19. Through a systematic search and statistical analysis, it was found that insulin use is significantly associated with composite poor outcomes of COVID-19, including COVID-19 risk, severe COVID-19, and mortality. The association is influenced by age, but not by diabetes and cardiovascular disease.