Obesity, dyslipidemia predict poor response to remdesivir for COVID-19

Unfavourable metabolic conditions, such as obesity and dyslipidemia, appear to have a negative impact on clinical outcomes following remdesivir treatment in COVID-19 patients who are hospitalized for moderate-to-severe disease, a study has found.

The analysis included 142 patients with SARS-CoV-2-related pneumonia admitted to a hospital and treated with remdesivir. Researchers used the Ordinal Scale for Clinical Improvement to assess clinical improvement within 28 days of hospitalization. They also evaluated short-term mortality.

During the observation period, 111 patients (78.2 percent) exhibited a clinical improvement. Compared with no-response controls, responders were younger (63.7 vs 74.3 years; p<0.01), less likely to be obese (20.7 percent vs 41.9 percent; p=0.03), and had lower body mass index (27.1 vs 31.1 kg/m²; p<0.01) and mean low-density lipoprotein cholesterol (LDL-C) levels (78 vs 103 mg/dl; p=0.03).

Multivariable logistic regression analysis revealed a negative association between obesity and clinical improvement after remdesivir treatment (odds ratio [OR], 0.48, 95 percent confidence interval [CI], 0.17–0.97; p=0.04). Furthermore, dyslipidemic patients had a slower clinical improvement than counterparts with normal lipid levels (p=0.04).

Finally, the odds of mortality risk were more than twofold high in the presence of obesity (OR, 2.82, 95 percent CI, 1.05–7.71; p=0.04) or dyslipidemia (OR, 2.78, 95 percent CI, 1.17–7.16; p=0.03).

There were no significant differences in the mean time for remdesivir prescription (from the onset of symptoms) between dyslipidemic vs no-dyslipidemia controls (4.9 vs 4.7 days, respectively; p=0.72) and between those with vs without obesity (5.2 vs 4.8 days, respectively; p=0.42).

Moreover, average PaO2/FiO2 ratio at the time of first remdesivir administration was similar in dyslipidemic vs no-dyslipidemia controls (305.6 vs 321.1, respectively; p=0.49) and in those with vs without obesity (315.2 vs 301.6, respectively; p=0.53).