Letter to the Editor: Low-density lipoprotein is a potential predictor of poor prognosis in patients with coronavirus disease 2019

Abstract

Background: The pandemic of coronavirus disease 2019 (COVID-19) has become a global threat to public health. The lipid pathophysiology in COVID-19 is unknown.

Methods: In this retrospective longitudinal study, we monitored the serum lipids in 17 surviving and 4 non-surviving COVID-19 cases prior to their viral infections and duration the entire disease courses.

Results: In surviving cases, the low-density lipoprotein (LDL) levels decreased significantly on admission as compared with the levels before infection; the LDL levels remained constantly low during the disease progression and resumed to the original levels when patients recovered (pre-infection: 3.5 (3.0-4.4); on admission: 2.8 (2.3-3.1), p < 0.01; progression: 2.5 (2.3-3.0); discharge: 3.6 (2.7-4.1); median (IQR), in mmol/L). In non-surviving patients, LDL levels showed an irreversible and continuous decrease until death (1.1 (0.9-1.2), p = 0.02 versus the levels on admission). The ratio changes of LDL levels inversely correlated with ratio changes of high-sensitivity C-reactive protein levels. Logistic regression analysis showed increasing odds of lowered LDL levels associated with disease progression (odds ratio: 4.48, 95% IC: 1.55-12.92, p = 0.006) and in-hospital death (odds ratio: 21.72, 95% IC: 1.40-337.54, p = 0.028).

Conclusions: LDL levels inversely correlated to disease severities, which could be a predictor for disease progress and poor prognosis.

Keywords: COVID-19; LDL; Lipid; Odds ratio.

Fig. 1

Ratio changes for LDL (A), HDL (B) and TC (C) in COVID-19 patients during the course of disease. For each data point, the ratio is normalized to the levels of LDL, HDL and TC prior to infection for the same patient. The days listed are the duration (median (IQR)) for each period during the disease course. The date on admission is set as “day 0”. Data is presented as “mean ± 95% confidence interval”. # indicates p < 0.05 as compared to the levels of pre-infection stage, and * indicates p < 0.05 as compared to the levels on admission, by a Mann-Whitney U test. The sample actual values at each stage are listed in Supplementary Table S1. The ratios of hsCRP at each time point are normalized to the levels prior to infection in the same patients. Pearson correlation coefficient analysis show the ratios of hsCRP significantly inversely correlated with the ratios of LDL (D), HDL (E) and TC (F) during the disease course; n = 33 data pairs for each analysis.