2024 Volume 71 Issue 7 Pages 705-711
At the beginning of 2020, coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) led to epidemics worldwide. Obesity and visceral fat accumulation have been reported to be independent risk factors for severe COVID-19. Several reports have focused on the levels of adipocytokines/adipokines, including adiponectin (APN), which is exclusively secreted from adipocytes, although the importance of these factors in acute disease conditions remains unclear. Therefore, we investigated the relationship between serum adiponectin levels and COVID-19 severity. Patients with COVID-19 who were admitted to Sumitomo Hospital (Osaka, Japan) from May through October 2021 were included. A total of 107 patients were enrolled in this study. We obtained the anthropometric and clinical laboratory data of the patients at the time of admission and examined the associations between various parameters and COVID-19 severity. The mean period from onset to admission was 6.5 ± 2.8 days. We divided the patients into “non-severe” (mild, moderate-I and moderate-II) (n = 80) and “severe” (n = 27) groups. The “severe” patients were significantly older than “non-severe” patients. Additionally, no significant differences were observed in BMI, sex, or the period from onset to admission. The serum adiponectin levels of “severe” patients at the time of admission were significantly greater than those of “non-severe” patients even after adjusting for age, sex, and BMI. These results suggest that the serum APN levels at the time of admission can predict COVID-19 severity. However, further investigations on the changes in APN levels in acute diseases are needed.
SINCE THE BEGINNING OF 2020, coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been endemic worldwide, and numerous infections and deaths have been reported in Japan.
The risk factors for severe COVID-19 include age 65 years or older, male sex, obesity, smoking, chronic respiratory disease, chronic kidney disease, diabetes mellitus, hypertension, dyslipidaemia, and malignant disease [1-9]. Among these risk factors, obesity and visceral fat accumulation have been reported to be independent risk factors for severe COVID-19 [10, 11], and several reports have focused on the association between adipocytokines and the pathogenesis of severe COVID-19 [12-16], although the detailed mechanism remains unclear.
Adiponectin (APN) is a unique molecule that is associated with the onset and progression of metabolic syndrome [17, 18], and the serum APN levels decrease in patients with obesity and/or visceral fat accumulation [17, 18]. Several researchers have shown that low serum APN levels are associated with type 2 diabetes [19, 20] and atherosclerotic diseases [20-22] and that serum APN levels decrease in patients with acute myocardial infarction [23] and sepsis [24] after their admission to the hospital. However, the physiological significance of changes in serum APN levels is still not fully understood [25].
In this study, we measured the serum adiponectin levels in patients with COVID-19 and examined its association with COVID-19 severity.
This prospective observational study initially included a total of 114 patients with COVID-19 admitted to Sumitomo Hospital (Osaka, Japan) from May through October 2021, when the COVID-19 Alpha and Delta strains were prevalent in Japan. The presence of SARS-CoV-2 was confirmed by RT-PCR testing of nasopharyngeal swabs. The patients were hospitalized because they required oxygen therapy for respiratory distress or were at increased risk for severe COVID-19 due to older age, severe obesity, and/or underlying health conditions. Seven patients were excluded (five did not consent to adiponectin measurement, and two had no height or weight data); finally, we enrolled 107 patients in this study (Fig. 1).
The protocol and time course of the study. A total of 114 patients with COVID-19 were initially included. Seven patients were excluded based on the exclusion criteria listed in the Methods section. Serum samples were collected at admission, and COVID-19 severity was classified during hospitalization. The severity of COVID-19 was defined as “mild” (no respiratory symptoms and no need for oxygen therapy), “moderate-I” (respiratory symptoms but no need for oxygen therapy), “moderate-II” (oxygen therapy), or “severe” (mechanical ventilation or high-flow nasal cannula therapy). The periods (number of days) from COVID-19 onset to admission and from admission to discharge are shown as the means ± SDs.
Written informed consent for this study was obtained from each patient. The study protocol was approved by the human ethics committees of Sumitomo Hospital (approved number #2021-19 and #2023-17).
Clinical parameters and data collectionWe obtained the anthropometric and clinical laboratory data of the patients at their admission. The onset of COVID-19 was defined as the day symptoms of fever, sore throat, cough, and/or shortness of breath started. The patient was diagnosed with diabetes mellitus, hypertension, or dyslipidaemia if the patient was receiving medication for the disease.
The medical history of diabetes mellitus, hypertension, dyslipidaemia, cardiovascular diseases, pulmonary diseases, smoking, and COVID-19 vaccination was retrieved via medical interview. The serum adiponectin level was measured by a latex particle-enhanced turbidimetric immunoassay with a human adiponectin latex kit (Otsuka Pharmaceutical Co., Tokyo, Japan). The residual serum at admission was frozen at –20°C until analysis.
Definition of COVID-19 severityThe patients were treated according to the COVID-19 Guidelines Version 5.0 issued on May 28, 2021, by the Ministry of Health, Labour and Welfare in Japan [26], which were the latest guidelines at the time this study was conducted. COVID-19 severity was classified as “mild” (no respiratory symptoms and no oxygen therapy required), “moderate-I” (with dyspnoea or pneumonia but no oxygen therapy required), “moderate-II” (oxygen therapy required), or “severe” (mechanical ventilation or high-flow nasal cannula therapy required). In this study, severe COVID-19 was defined as the most severe state during hospitalization.
Statistical analysisThe data are presented as the means ± standard deviations or numbers/percentages of participants. Differences between “non-severe” and “severe” patients were assessed by the Welch t test or Fisher’s exact test. Multiple logistic regression analysis was conducted to assess the association between the serum adiponectin level at admission and the severity of COVID-19 during hospitalization. P values <0.05 were considered to indicate statistical significance. All the statistical analyses were performed using JMP Pro 17.1.0 (JMP Statistical Discovery, LLC).
Fig. 1 shows the time course of the present study. A total of 107 patients were enrolled in this study. The mean period (number of days) from onset to admission was 6.5 ± 2.8 days, and the mean length of hospitalization was 15.1 ± 9.0 days. The clinical characteristics of the patients are shown in Table 1. The mean age was 51.5 ± 16.2 years, and the mean BMI was 25.0 ± 4.9. The numbers of “mild” (no respiratory symptoms and no need for oxygen therapy, n = 3) and “moderate-I” (with respiratory symptoms but no need for oxygen therapy, n = 16) patients were relatively low compared to those of “moderate-II” (oxygen therapy, n = 61) and “severe” (mechanical ventilation or high-flow nasal cannula therapy, n = 27) patients, reflecting that the main reason for the hospitalization was the need for oxygen therapy. The patients in the “moderate-II” and “severe” groups had greater BMIs than did the subjects in the “moderate-I” group. The “mild” patients had the greatest BMI, mainly because the physicians were concerned about the patient’s risk of severe disease and considered that hospitalization was required for the patient. There were no diabetic patients treated with pioglitazone (thiazolidinediones), which elevates serum adiponectin levels.
Clinical characteristics of the patients on admission.
| “non-severe” | severe | |||
|---|---|---|---|---|
| mild | moderate-Ⅰ | moderate-Ⅱ | ||
| n (Male/Female) | 3 (2/1) | 16 (12/4) | 61 (43/18) | 27 (18/9) |
| Age (years) | 44.0 ± 12.5 | 42.0 ± 14.7 | 50.1 ± 15.3 | 61.1 ± 15.4 |
| BMI | 28.4 ± 7.1 | 22.7 ± 4.6 | 25.1 ± 4.2 | 25.7 ± 5.7 |
| Diabetes Mellitus (%) | 1 (33) | 2 (13) | 6 (10) | 5 (19) |
| Dyslipidemia (%) | 1 (33) | 1 (6) | 11 (18) | 6 (22) |
| Hypertension (%) | 1 (33) | 2 (13) | 15 (25) | 8 (30) |
| Cardiovascular disease (%) | 0 (0) | 2 (13) | 3 (5) | 5 (19) |
| Pulmonary disease (%) | 0 (0) | 4 (25) | 13 (21) | 11 (41) |
| Smoking (%) | 2 (67) | 7 (44) | 28 (46) | 14 (52) |
| Adiponectin (μg/mL) | 6.5 ± 4.6 | 9.3 ± 5.2 | 8.4 ± 5.0 | 11.0 ± 4.7 |
| eGFR (mL/min/1.73m2) | 82.0 ± 26.3 | 87.9 ± 14.6 | 75.6 ± 20.0 | 75.0 ± 25.2 |
| BNP (pg/mL) (number of patients) | 9.4 ± 16.2 (3) | 4.3 ± 6.0 (15) | 11.3 ± 26.5 (56) | 13.5 ± 18.9 (23) |
| COVID-19 Vaccination (%) | 0 (0) | 0 (0) | 6 (9.8) | 5 (18) |
| Onset to admission (days) | 2.0 ± 1.0 | 5.4 ± 2.8 | 6.8 ± 2.7 | 7.1 ± 2.6 |
| Hospitalization period (days) | 12.3 ± 6.8 | 9.1 ± 5.0 | 14.6 ± 8.8 | 22.2 ± 7.1 |
| Medications, number of patients (%) | ||||
| Sulfonylureas/Glinides | 0 (0) | 0 (0) | 1 (2) | 1 (4) |
| Biguanides | 0 (0) | 1 (6) | 3 (5) | 1 (4) |
| Alpha-GIs | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| Thiazolidinediones | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| DPP-4 inhibitors / GLP-1 analogs | 0 (0) | 1 (6) | 6 (10) | 3 (11) |
| SGLT2 inhibitors | 0 (0) | 1 (6) | 3 (5) | 1 (4) |
| Insulin | 1 (33) | 1 (6) | 3 (5) | 1 (4) |
| ACEI/ARBs | 1 (33) | 1 (6) | 10 (16) | 1 (4) |
| Statins | 1 (33) | 1 (6) | 10 (16) | 5 (19) |
The data are presented as the means ± SDs or number of patients. As BNP was measured in not all the patients, the number of patients who had measurements for BNP is shown. COVID-19 severity (mild, moderate-I, moderate-II, or severe) was classified according to the COVID-19 Guidelines Version 5.0 by the Ministry of Health, Labor and Welfare in Japan issued on May 28, 2021 (see Fig. 1 for the severity definitions). We divided patients into “non-severe” (mild + moderate-I + moderate-II) and “severe” groups. BMI: body mass index, eGFR: estimated glomerular filtration rate, BNP: brain natriuretic peptide, Alpha-GIs: alpha-glucosidase inhibitors, DPP-4: dipeptidyl peptidase iv, GLP-1: glucagon-like peptide 1, SGLT2: sodium-glucose cotransporter 2, ACEI: angiotensin-converting enzyme inhibitors, ARB: angiotensin receptor blockers.
Next, we divided the patients into “non-severe” (mild, moderate-I and moderate-II) (n = 80) and “severe” (n = 27) groups, which would be important for predicting the clinical outcomes of the patients.
As shown in Fig. 2, “severe” patients were significantly older than “non-severe” patients were (Fig. 2A). However, no significant differences were observed in BMI (Fig. 2B), sex (Fig. 2C), or the periods from onset to admission (Supplemental Table 1). Interestingly, the serum adiponectin levels of “severe” patients at the time of admission were significantly greater than those of “non-severe” patients (Fig. 2D). As shown in Supplemental Table 1, the duration of hospitalization was significantly longer in the “severe” patients than in the “non-severe” patients. The prevalence of underlying health conditions (including diabetes mellitus, dyslipidaemia, hypertension, cardiovascular diseases, pulmonary diseases, and smoking), eGFR, BNP levels, and vaccination status did not differ between the two groups.
Comparison of age (A), BMI (B), sex (C), and serum adiponectin level at the time of admission (D) between “non-severe” and “severe” patients. Welch’s t test (A, B, and D) or Fisher’s exact test (C) was performed.
Table 2 shows the results of multiple logistic regression analysis of the association between serum adiponectin levels at the time of admission and COVID-19 severity. Even after adjusting for age, sex, and BMI, high adiponectin levels at the time of admission were significantly associated with severe COVID-19.
Odds ratios (ORs) and 95% confidence intervals (CIs) determined by multiple logistic regression analysis for the association between the serum adiponectin level at the time of admission and the severity of COVID-19 during hospitalization.
| Variable | severe COVID-19 (vs. “non-severe”) | |
|---|---|---|
| OR (95% CI) | p value | |
| Age | 1.059 (1.023–1.097) | 0.001 |
| Gender (Female) | 0.723 (0.239–2.182) | 0.561 |
| BMI | 1.232 (1.101–1.486) | 0.002 |
| Log (Adiponectin) | 54.23 (3.044–1805.9) | 0.007 |
The serum adiponectin levels were logarithmically transformed because the data were skewed. BMI; body mass index.
In this study, we compared the characteristics of “severe” and “non-severe” (mild, moderate-I, and moderate-II; i.e., no need for mechanical ventilation or high-flow nasal cannula therapy) COVID-19 patients (Table 1 and Fig. 2A–C). The results showed that serum APN levels at the time of admission were significantly greater in “severe” patients than in “non-severe” patients (Fig. 2D), even after adjusting for age, sex, and BMI (Table 2).
It has been reported that APN has various organ-protective functions in chronic diseases such as diabetes and obesity [27-32]. The serum APN levels decrease in patients with obesity and visceral fat accumulation [17, 18]. In the present study, a significant negative correlation between APN and BMI was also observed (r = –0.468, p < 0.001; data not shown). Simultaneously, the present study showed a significant association between relatively high serum APN levels at the time of admission and severe COVID-19.
There are several reports on changes in serum APN levels and their importance in acute disease conditions. We previously reported that the serum APN level in patients with acute myocardial infarction decreased the most 24 hours after the onset of infarction and that a smaller change in the serum APN level was associated with a larger CKMB AUC, which reflects the size of the myocardial infarction [23]. Ebihara et al. examined the serum adipocytokine profiles of sepsis patients and showed that the serum APN level decreased the most on the second day after admission and that patients whose serum APN level did not decrease had a worse prognosis [24].
As the periods from onset to admission tended to increase according to the severity of COVID-19 in this study (Table 1), the periods from onset to admission might affect the serum APN levels at the time of admission. However, there was no significant difference in the periods from onset to admission between “non-severe” and “severe” patients (Supplemental Table 1). In addition, the serum APN level at the time of admission was still significantly associated with COVID-19 severity after adjustment for periods from onset to admission (odds ratio (95% confidence interval) 74.14 (3.044–1805.9), p = 0.008; data not shown). Taken together, the higher serum APN levels in “severe” patients than in “non-severe” patients might suggest that the serum APN levels did not decrease between the onset of COVID-19 and admission to the hospital. Elucidating the importance of changes in serum APN levels in acute disease may help to understand the organ-protective effects of APN and its relationship with disease prognosis.
APN specifically and strongly binds to T-cadherin (T-cad), a glycosylphosphatidylinositol-anchored membrane protein expressed on the cell surface of the heart, muscle, endothelial cells, and mesenchymal stem/stromal cells [33-35]. We have demonstrated that the APN/T-cad system enhances exosome biogenesis [32] and is essential for protection of various organs [27, 36, 37]. Furthermore, we have reported that both mRNA and protein levels of T-cad are downregulated by endoplasmic reticulum (ER) stress [38]. It has been reported that the ER stress-induced chaperone GRP78/Bip is upregulated during SARS-CoV-2 infection [39]. Moreover, genome-wide association studies have revealed that genetic variations in CDH13 (the gene name of T-cad) are strongly associated with serum APN levels [40, 41]. Taken together, the disproportionally high APN level in the patients with severe COVID-19 in this study may indicate that APN might not be able to bind to T-cad-expressing organs efficiently because of T-cad downregulation by ER stress; hence, organ protection by APN might be impaired.
The risk factors for severe COVID-19 include advanced age, male sex, obesity, smoking, chronic respiratory diseases, chronic kidney diseases, diabetes mellitus, hypertension, dyslipidaemia, and malignant diseases [1-9]. On the other hand, there have been several reports [12-16] on the association between APN and COVID-19 severity, although the results are controversial. Possible reasons for such controversy may include differences in the control groups in each report, definitions of COVID-19 severity, race, and medical circumstances.
This study has several limitations. The present study is a cross-sectional and single-centre study of hospitalized patients. It is difficult to discuss the causality. The number of enrolled subjects was relatively small, and the serum APN levels before the onset of COVID-19 were unknown. The visceral fat area, which correlates with the serum APN levels could not be measured due to infection control problems.
In conclusion, the relatively high serum APN level in COVID-19 patients at the time of admission was associated with severe COVID-19.
We thank Dr. Yuya Yamada for valuable advice, Otsuka Pharmaceutical Co. for their cooperation on adiponectin measurements, and all the medical staff of Sumitomo Hospital for the treatment of COVID-19 patients and support of this study.
None of the authors have any potential conflicts of interest associated with this study.
