Post-COVID-19 syndrome, inflammation and insulin resistance: a retrospective cohort study

BACKGROUND

COVID-19-induced diabetes is a novel and enigmatic disease. Our aim was to evaluate a possible relationship between post-COVID-19 syndrome (PCS) and increased insulin resistance (IR) in non-diabetic outpatients after mild COVID-19.

METHODS

Repeated measures design. Three evaluations [1E (pre-COVID, baseline), 2E (3 months post-COVID) and 3E (21 months post-COVID)] were performed, directed to PCS+ and PCS- subjects. Triglyceride-glucose (TyG) index ≥8.74 was considered IR, and albumin-to-globulin ratio (AGR) <1.50, inflammation.

RESULTS

We analyzed 112 individuals (median [IQR] age=44 [20] years, 58% women, 36 PCS+, 76 PCS-). PCS+ with very low basal IR (TyG <7.78, lowest quartile) showed a reduced inflammatory burden (basal AGR=1.81 [0.4] vs. 1.68 [0.2] in 2E; P=0.23), and increased TyG across evaluations (from basal 7.62 [0.2] to 8.29 [0.5]; P=0.018]. Conversely, PCS+ subjects with high basal TyG (TyG ≥8.65, highest quartile) did not show significant variations in TyG, but a greater inflammatory load (basal AGR=1.69 [0.3] vs. 1.44 [0.3] in 2E; P=0.10). In multivariable models addressing groups with reduced basal IR (TyG <8.01), PCS has been a consistent predictor for TyG, after adjusting for confounders. Partial correlation and multivariable analyses showed similarities involving acute polysymptomatic COVID-19 and PCS regarding IR.

CONCLUSIONS

PCS was associated with increased IR, being more evident when the baseline degree of IR was very low. PCS and increased IR were separately associated with inflammation. Acute polysymptomatic COVID-19 and PCS could be clinical expressions of underlying inflammatory state, which in turn may also trigger IR.

Post-COVID-19 syndrome, low-grade inflammation and inflammatory markers: a cross-sectional study

OBJECTIVE

Post-COVID syndrome (PCS) is a poorly known entity. An underlying chronic, low-grade inflammation (LGI) has been theorized as a pathophysiological mechanism. Available data on biomarkers in PCS show conflicting results. Our aim was to know whether subjects with PCS present higher levels of inflammatory markers, after a mild COVID-19.

METHODS

Analytical cross-sectional study. Cases of mild COVID-19 in a community setting were included. We collected epidemiological data (age, sex, BMI, smoking, comorbidities), variables of the acute COVID-19 (duration, symptoms), and data at 3 months after the acute phase (symptoms and laboratory test). Serum C-reactive protein (CRP), neutrophil and lymphocyte counts, neutrophil/lymphocyte ratio (NLR), lactate dehydrogenase, ferritin, fibrinogen, and D-dimer levels were analysed. LGI was defined as CRP >0.3 and <1.0 mg/dL. A subject was classified as PCS + if presented signs and symptoms >12 weeks after an infection consistent with COVID-19. Five composite indices (C1-C5) were developed, combining the upper ranges of biomarkers distributions. Multivariate analyses were performed.

RESULTS

We analysed 121 mild COVID-19 cases (mean age = 45.7 years, 56.2% women). Among the acute symptoms, women presented a higher frequency of fatigue (54.4% vs 30.2%; p = .008). PCS affected 35.8% of women and 20.8% of men (p = .07), and the most reported symptoms were fatigue (42.8%), anosmia (40%), ageusia (22.8%), dyspnea (17.1%) and myalgia (11.4%). Neutrophil count, NLR, CRP and fibrinogen showed the best correlations with PCS and were selected to develop the indices. In women PCS+, C1, C3 and C4 indices were more frequently met, while in men PCS+, C2, C5 and CRP were in the range of LGI. Anosmia, ageusia and fatigue were related to higher neutrophil counts, with sex differences. Fibrinogen levels were higher in persistent myalgia (510 ± 82 mg/dL vs 394 ± 87; p = .013). In multivariable analysis, a woman with a neutrophil count above the median, or with fibrinogen level or NLR in the highest tertile, had a 4-5-fold increased risk of prevalent PCS. A man with CRP in the range of LGI, or fibrinogen level or a neutrophil count in the highest tertile, had a 10-17-fold increased risk of prevalent PCS.

CONCLUSIONS

The data obtained in the present cross-sectional study seems to demonstrate a consistent association between PCS and upper ranges of the neutrophil count, NLR, fibrinogen, and CRP in the LGI range. Furthermore, composite indices appear useful in detecting relationships between slight elevations of biomarkers and PCS, and our study identifies relevant sex differences in symptoms and markers regarding the PCS.