key: cord-1016206-vuuxthx2
authors: Deng, Ming; Qi, Yongjian; Deng, Liping; Wang, Huawei; Xu, Yancheng; Li, Zhen; Meng, Zhe; Tang, Jun; Dai, Zhe
title: Obesity as a Potential Predictor of Disease Severity in Young COVID‐19 Patients: A Retrospective Study
date: 2020-06-29
journal: Obesity (Silver Spring)
DOI: 10.1002/oby.22943
sha: 6c6e553d90d4291bdaa9329060b93d6733afa280
doc_id: 1016206
cord_uid: vuuxthx2

OBJECTIVE: To explore the indicators for severity in young COVID‐19 patients age between 18 to 40. METHODS: This retrospective cohort study includes 65 consecutively admitted COVID‐19 patients age between 18 to 40 in Zhongnan Hospital of Wuhan University. Among them, 53 were moderate cases, 12 were severe or critical cases. Epidemiological, clinical and laboratory characteristics and treatment data were collected. A multivariate logistic regression analysis was implemented to explore risk factors. RESULTS: The severe/critical cases have obviously higher BMI (average 29.23 vs. 22.79kg/m(2)) and lower liver CT value (average 50.00 vs. 65.00mU) than moderate cases group. The severe/critical cases have higher fasting glucose, alanine aminotransferase (ALT) , aspartate aminotransferase (AST) , and creatinine (Cr) compared with moderate cases (All P<0.01) . More severe/critical cases (58.33% vs. 1.92%) have positive urine protein. The severe/critical cases will experience a significant process of serum albumin decline. Logistic regression analysis showed that male, high body mass index (especially obesity), elevated fasting blood glucose and urinary protein positive are all risk factors for severe young COVID‐19 patients. CONCLUSION: Obesity is an important predictor of severity in young COVID‐19 patients. The main mechanism is related to the damage of liver and kidney.

the Institutional Ethics Board of Zhongnan Hospital of Wuhan University (No. 2020011). A total of 65 patients were included in the analysis. All cases were classified as moderate, severe, or critical according to the Guidance for Corona Virus Disease 2019 (6th edition) released by the National Health Commission of China [18] . Moderate cases have fever and respiratory tract-related symptoms as well as visible signs of pneumonia on imaging. Cases are classified as severe if they meet any of the following criteria: (1) shortness of breath, according to a respiratory rate (RR) ≥30 times/min; (2) in the resting state, an oxygen saturation ≤93%; (3) arterial blood oxygen partial pressure (PaO 2 )/oxygen concentration (FiO2) ≤300 mmHg. Critical case need to meet any of the following criteria: (1) respiratory failure occurs and mechanical ventilation is required; (2) shock occurs; (3) organ failure in addition to respiratory distress requires intensive care unit (ICU) monitoring and treatment. The need for written informed consent was waived owing to the rapid emergence of this infectious disease.

Clinical data were collected up to March 13, 2020 , and data were collected for one critically ill patient until April 12, 2020 after transfer to Leishenshan Hospital. We use a standard case report form for data collection. Past medical history, family history, anthropometric data, laboratory test results, treatment details, and oxygen support data during hospital admission were collected from the patients' medical records by two independent researchers. If any key data were

This article is protected by copyright. All rights reserved missing, we contacted the physician in charge of the patient's care to request further information.

Data for metabolic indexes (i.e., fasting blood glucose [ 

The presence of the SARS-Cov-2 virus in respiratory samples was confirmed with real-time reverse transcription polymerase chain reaction (rt-PCR) using a nucleic acid detection kit developed by DAAN Gene Co., Ltd. of Sun-Yatsen University. This kit is designed to qualitatively detect the ORF1ab and N genes of the SARS-Cov-2 virus.

In this study, the Intelligent Evaluation System (Hangzhou YITU Healthcare Technology Co., Ltd.) was employed as the CT image analysis tool for the detection of COVID-19-related pneumonia. The system was used for segmentation of the left and right lungs and detection of patchy shadows, and from the collected images, quantitative parameters including the inflammatory volume and proportion of the inflammatory volume among the total lung volume were computed.

This article is protected by copyright. All rights reserved Axial CT images were reconstructed with slice thicknesses of 5 mm and 1 mm. The thickness of subcutaneous fat was measured on the axial images under the midline of the anterior abdominal wall, and the thickness in the largest transverse view was taken as the final measurement.

The CT value of the liver was determined by selecting the left lateral lobe, the left medial lobe, and the right hepatic lobe on the transverse images of the 12th Thoracic Vertebra using a Digital Imaging and Communications in Medicine (DICOM) viewer. During the measurement, the area of interest (ROI) as positioned as far away from the blood vessels as possible. The averages of the three measurements was used as the CT value of the liver. In addition, the evaluation criteria for fatty liver was a liver density lower than that of the spleen (<60 Hounsfield Units

[HU]) [19] . CT density of epicardial fat and visceral fat were measured by standard methods.

We assessed the volume of epicardial fat volume using the original chest CT data as the basic data. The pixel from the upper edge of the outer pericardium to the heart fat was extracted layer by layer, and the volume of the fat was accumulated according to the chest CT image thickness ( Figure S1 ).

This small-sample study had unbalanced numbers of patients in the moderate and severe/critical groups (severe and critical cases were combined in one group for analysis). For descriptive statistical analysis, categorical variables were expressed as frequencies and percentages, and continuous variables were expressed as medians with quartiles (Q1, Q3).

Continuous variables were also transformed into categorical variables for further analysis.

Differences in categorical variables between the moderate and severe/critical groups were assessed by Fisher's exact test, and differences in continuous variables between the two groups were tested using the Wilcoxon rank-sum test (nonparametric test).

Many factors presented statistical significance in the initial difference analysis. However, the sample sizes for the different conditions limited their simultaneous inclusion in multiple logistical regression. To maintain the interpretability of the multiple regression, variable selection was

This article is protected by copyright. All rights reserved performed for a set of binary predictors using Lasso logistic regression. Multivariable logistical regression was performed by Exact logistic regression and Firth's logistical regression, which are standard approaches for the analysis of binary outcomes with a small sample size. Exact logistic regression was selected to explore the effect of the selected variables. Sex and BMI were considered as forced-in covariates, and selected variables from Lasso regression were included in turn. To assess the robustness of the results, these binary predictors were replaced by their continuous types, and Firth's logistical regression was performed instead of exact logistical regression.

The data were analyzed using SAS software (SAS Institute Inc, Cary, NC) and R version 3.4.2 software (The R Foundation for Statistical Computing, Vienna, Austria) with the glmnet package. Two-sided P values of less than 0.05 were considered statistically significant.

A total of 65 COVID-19 cases among young adults were included in the analysis, of which 53 were moderate cases and 12 were severe or critical cases. Because only three cases were classified as critical, we combined the severe cases (n=9) and critical cases (n=3) for comparison with moderate cases (Table 1 ). Notable findings included that all severe/critical cases were male, making the proportion of male patients in this group significantly higher than that in the moderate group (100.00% vs. 45.28%). The age distribution and proportions of patients who reported cigarette smoking and alcohol drinking were similar between the two groups. Few patients in either group had a history of chronic disease or a related family history before admission. The systolic pressure and diastolic pressure were comparable between the moderate cases (average 122/77 mmHg) and severe/critical cases (average 120/77 mmHg). Before hospital admission, moderate cases and severe/critical cases experienced a similar duration of symptoms.

More severe/critical cases had diarrhea (2%), but the frequencies of other symptoms were similar

This article is protected by copyright. All rights reserved between the two groups.

Evaluation of metabolic and inflammatory indicators showed that the severe/critical group had higher levels of FBG (6.77 vs. 5.15 mmol/L), Lp(a) (147.85 vs. 74.80 mg/L), CRP (80.04 vs. 13 .72 mg/L), and LDH (364.50 vs. 163.00 U/L) and a lower level of HDL-C (0.84 vs. 1.09 mmol/L) compared with moderate cases ( Table 1 ). The neutrophil count also was higher in the severe/critical group than in the moderate group (4.10×10 9 vs. 2.71×10 9 /L). Coagulation indexes including D-dimer and aPTT were comparable between the two groups at admission (Table 1 ).

Indexes for dysfunction of different organs also were measured and compared between the two groups. With regard to cardiovascular functioning, the CK-mB levels of the moderate and severe/critical groups were similar (9.40 vs. 12.40 ng/ml). Because cardiac TnI and BNP data were available for only a few patients, these two indicators could not be included in the analysis.

Among the cases with test results, all had a normal BNP level, and only one critically ill patient who experienced cardiac arrest before hospitalization had an elevated TnI concentration (108 pg/ml). We also found from CT imaging that all patients had good coronary vascular conditions. From the CT images, we observed that even the patient with highest BMI had no coronary calcification similar to the normal-weight patient ( Figure S1 ). With regard to liver functioning, the severe/critical group had higher levels of ALT (42.50 vs. 20.50 U/L) and AST (46.00 vs. 20.00 U/L) than did the moderate group. Because the increases in ALT and AST were nearly parallel (r=0.83), we considered that the elevated ALT and AST levels were caused by impaired liver function. With regard to renal functioning, the severe/critical group have higher levels of Cr compared with the moderate group. In addition, the frequency of urine protein positivity was greater among severe/critical cases than among moderate cases (58.33% vs. 1.92%). Because immune cell counts after admission were available for only a few moderate cases and less than half of severe/critical cases, we did not include this indicator in the analysis. However, among those patients, we observed a decreased trend in T cell counts in severe/critical cases (Table S1 ).

From the evaluation of obesity among the study population, the severe/critical group had an

This article is protected by copyright. All rights reserved obviously higher mean BMI than the moderate group (average 29.23 vs. 22 .79 kg/m 2 ; Table 1 ). In addition, the liver CT value was less for the severe/critical group compared with the moderate group (50.00 vs. 65.00 mU), which indicates more fat deposition in the liver ( Table 2 ). The epicardial fat volume is significantly higher in severe/critical group (207469 vs. 112861mm 3 ) .

The subcutaneous fat thicknesses were similar between the two groups.CT density of epicardial fat and viseral fat were similar between the two groups. According to CT examination, severe/critical cases had a larger intrapulmonary lesion volume compared to moderate cases (5.76% vs. 0.72%) ( Table 2) .

As treatment for COVID-19, more cases in the severe/critical group received glucocorticoids (83.33% vs. 24.52%) and albumin supplementation (25% vs. 0.00%). More cases in the severe/critical group required oxygen support, and accordingly, the frequencies of nasal catheter use (83.33% vs. 24.52%) and extracorporeal membrane oxygenation (ECMO, 25% vs. 0.00%) were higher in the severe/critical group than in the moderate group (Table 3) .

We observed the distribution of different indicators of obesity among young adult COVID-19 according to the different in disease severity ( Figure 1 ). According to BMI measurements, both severely and critically ill patients were distributed in the overweight/obesity interval. Among them, all three critically ill patients were with obesity. According to the liver CT value, all the severely and critically ill cases were in the fatty liver interval. Consistent with the results of the comparative analysis, subcutaneous fat thickness did not have a detectable effect on the severity of disease.

The serum albumin level was examined twice during hospitalization for four patients with

This article is protected by copyright. All rights reserved severe disease, and in these patients, the serum albumin level dropped rapidly in a short period of time ( Figure 2 ). In addition to changes in serum albumin, the D-dimer level in critically ill patients also changed significantly. Interestingly, the decline in the serum albumin level in critically ill patients and the rise in the D-dimer level showed an essentially parallel inverse relationship, even when anticoagulant drugs were used (Figure 3 ). From the detailed data for the three critically patients (Tables S2-S4) , we observed that the changes in serum albumin levels were also parallel with the changes in the lung lesions. Notably, all three of the critically ill patients experienced acute respiratory distress syndrome (ARDS). However, cardiovascular, liver, and kidney function did not show significant changes during the course of the disease. In critical patient 2, the elevated levels of TnI and renal and liver function indexes at the time of admission may have been related to the occurrence of cardiac arrest before admission, and these indexes quickly recovered and stabilized after treatment. Obvious changes in coagulation indexes were observed in all of the critically ill patients during the disease process. Critically ill patient 1 eventually experienced disseminated intravascular coagulation (DIC).

Logistic regression analysis was applied to assess the potential risk factors for the severity of young COVID-19 patients (Tables 4 and 5 ). The analysis showed that a high body mass index (especially obesity), an elevated FBG level, an elevated LDH level, and urinary protein positivity were all risk factors for severe COVID-19 in these young patients. In particular, urinary protein positivity could increase the risk of severe disease by more than 20-fold.

In the analysis of relevant risk factors for severe COVID-19 published in China before April, little was mentioned about the correlation between obesity and severe illness in patients with COVID-19 [20] . However,more and more updated researchs indicate that the role of obesity is very important [14] [15] [16] [17] . A study from NYU School of Medicine showed that in the population

This article is protected by copyright. All rights reserved under 60 years old, the proportion of people with a BMI over 30 kg/m 2 who were hospitalized and admitted to the ICU was more than double that among patients with a lower BMI [21] . A recently published study from China also showed that in metabolic-associated fatty liver disease patients, obesity can increase the risk for severe COVID-19 by about 6-fold [22] . The present study provide more detailed information about the role of obesity in the severity of young COVID-19 patients.

Our findings show that obesity is an important predictor of severe COVID-19 among young patients. Further analysis found that obesity mainly affects the severity of COVID-19 through the deposition of ectopic fat in multiple organs, which in turn damages the organ function. According to our data, in young patients with severe COVID-19, the organs most likely to be affected were the liver and kidneys. In our study, the liver CT value in people with overweight and obesity (average 53mU) is similar with previous study [33] . Only one critically ill patient in this study had myocardial damage, and it was most likely caused by cardiac arrest. Notably, in the present study, all of the severely or critically ill COVID-19 patients were males, an observation which may also be related to the distribution of obesity in China. Among people under the age of 40 years, the number of men with obesity is three times greater than the number of women [23] .

Although all patients in the current study had comparable albumin levels on admission, we observed a rapid decline in albumin concentration in severely and critically ill patients after admission. Because the albumin level was not tested a second time in moderate cases during hospitalization, we cannot be sure that the moderate patients did not also experience a decline in albumin concentration. A study evaluating patients with COVID-19 after 2 weeks of hospitalization found that patients with aggravated conditions had significantly reduced albumin levels, while those with stable or improved conditions had an average albumin level of 41.27 g/L [24] , suggesting that moderate cases are less likely to experience a significant decrease in albumin.

Additional studies have suggested that hypoalbuminemia on admission can help predict the progression of COVID-19 cases to severe or critical condtiion [25] [26] [27] . Chronic liver disease due to synthesis disorders and chronic kidney disease due to glomerular leakage are common causes of hypoalbuminemia. Even without liver and kidney dysfunction, obesity itself will be accompanied

This article is protected by copyright. All rights reserved by hypoalbuminemia [28] . But abnormal albumin function is more common reported in fatty liver disease [29] . This study suggests that young patients with proteinuria at admission are more than 20 times more likely to develop severe and critical illness. A renal histopathological analysis of deceased COVID-19 patients showed that SARS-CoV-2 can directly infect and damage proximal tubular epithelial cells and podocytes [30] . Podocyte damage is usually associated with the formation of proteinuria [31] . Hypoproteinemia may be very obvious in some critically ill patients.

We observed that in one critically ill patient who continued to be treated with endotracheal intubation and mechanical ventilation at this time of this reporting, normal albumin levels could only be maintained by continuous large infusions of albumin. Our understanding of the functions of albumin continues to increase. Albumin can not only effectively maintain plasma osmotic pressure, but also has functions in transport, anti-thrombosis, immune regulation and endothelial stability [32] . In severe and critically ill COVID-19 patients, low albumin levels may also be accompanied by impaired organ functioning. In this study, it was considered that the rapid onset of hypoalbuminemia may be related to the attack of the virus on the organs, because almost all young COVID-19 patients had normal serum albumin levels at admission. Pre-existing damage to liver and kidney function may cause further damage to organ functions under the attack of SARS-CoV-2 infection.

In the present study, we also observed that although the difference in D-dimer between moderate and severe/critical COVID-19 cases was not significant at admission, the level of D-dimer continued to increase in critically ill cases. Some studies have suggested that D-dimer may be a risk factor for severe COVID-19 [34, 35] , and the risk of thrombosis is high for severe COVID-19 patients [36,37]. Our follow-up of three critically ill patients showed that even when anticoagulant drugs were administered, the decrease in albumin and increase in D-dimer were nearly parallel. In a critically ill patient who later died, DIC eventually appeared.

Hypoalbuminemia has been considered a high risk factor for thrombosis [38, 39] , and albumin itself also has anti-thrombotic effects due to its capacity to bind nitric oxide (NO) and prolong the biologic activity of NO [32] . Of course, the increased D-dimer level in critically ill COVID-19

This article is protected by copyright. All rights reserved patients in this study could also be related to other factors, including the hypercoagulable state combined with obesity, viral infection and later combined bacterial infections, and multiple organ dysfunction.

Consistent with previous studies [40] , this study also suggests that elevated fasting blood glucose levels at admission are also risk factors for critical illness in young COVID-19 patients, although most people have no previous history of diabetes. Considering that previous studies have suggested the role of inflammation in severe COVID-19 cases [41] , and that fat density in people with obesity may be related to inflammation of adipose tissue [42,43], we evaluated epicardial fat and visceral fat CT density in the study population, but find the difference between moderate and severe/critical group were not significant.

Because the sample size of this study was small, and the distribution of the two groups was uneven. This has been considered as much as possible in the analysis. Two methods were used in the logistic analysis for identification of risk factors. In moderate COVID-19 cases, the absence of multiple albumin test results during hospitalization made it impossible to accurately determine the changes in albumin levels among moderate cases. In addition, immune cell numbers were not tested for many cases, making it impossible for us to thoroughly assess the role of the immune response in young COVID-19 patients. In addition, we need to note that Chinese definition of overweight and obesity is different from the United States and Europe [44] . Therefore, when applying the conclusions of this study, it is necessary to consider the possible impact of this difference.Our research suggests that the liver CT value is helpful for predicting severe cases of young COVID-19. However, we should note that liver CT value is an imaging index, which can not reflect pathophysiological changes. Therefore, we also need further research to help clarify low liver CT value associate with severe COVID-19 cases.

This article is protected by copyright. All rights reserved

The present study investigated predictive factors for severe COVID-19 among young patients and has several major findings: (1) 

This article is protected by copyright. All rights reserved This article is protected by copyright. All rights reserved Data are presented as n(%) or median(Q1, Q3).

Abbreviation: CT, computed tomography. 

This article is protected by copyright. All rights reserved 

This article is protected by copyright. All rights reserved 

This article is protected by copyright. All rights reserved OR, odds ratio. CI, confidence interval. BMI, body mass index. FBS, fast blood sugar. LDH, lactate dehydrogenase. Cr, creatinine. +∞, positive infinity. NE, not estimated. 

This article is protected by copyright. All rights reserved OR, odds ratio. CI, confidence interval. BMI, body mass index. FBS, fast blood sugar. LDH, lactate dehydrogenase. Cr, creatinine. +∞, positive infinity. 

WHO Coronavirus disease (COVID-19) Situation dashboard

A Pathological Report of Three COVID-19 Cases by Minimally Invasive Autopsies

COVID-19 Autopsies

Review and Prospect of Pathological Features of Corona Accepted Article This article is protected by copyright. All rights reserved Virus Disease

Analysis of heart injury laboratory parameters in 273 COVID-19 patients in one hospital in Wuhan, China

SARS-CoV-2 inflames the heart. The importance of awareness of myocardial injury in COVID-19 patients

Liver injury in COVID-19: management and challenges

COVID-19 and Kidney Failure in the Acute Care Setting: Our Experience From Seattle

Pathological evidence of pulmonary thrombotic phenomena in severe COVID-19

Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study

The Clinical and Chest CT Features Associated with Severe and Critical COVID-19 Pneumonia

Characteristics of and Important Lessons From the Coronavirus Accepted Article This article is protected by copyright. All rights reserved Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention

Clinical Characteristics of 30 Medical Workers Infected With New Coronavirus Pneumonia Zhonghua Jie He He Hu Xi Za Zhi

Covid-19 in Critically Ill Patients in the Seattle Region -Case Series

Factors associated with hospitalization and critical illness among 4,103 patients with COVID-19 disease in New York City

High prevalence of obesity in severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) requiring invasive mechanical ventilation

Obesity in patients younger than 60 years is a risk factor for Covid-19 hospital admission

New coronavirus pneumonia prevention and control program

Liver fat accumulation assessed by computed tomography is an independent risk factor for diabetes mellitus in a population-based study: SESSA (Shiga Epidemiological Study of Subclinical Atherosclerosis)

Obesity and COVID-19 severity in a

Obesity in patients younger than 60 years is a risk factor for Covid-19 hospital admission

Obesity as a risk factor for greater severity of COVID-19 in patients with metabolic associated fatty liver disease

Sex differences in the prevalence and adverse outcomes of sarcopenia and sarcopenic obesity in community dwelling elderly in East China using the AWGS criteria

Analysis of factors associated with disease outcomes in hospitalized patients with 2019 novel coronavirus disease

Risk factors for disease severity, unimprovement, and mortality of COVID-19 patients in Wuhan, China

Clinical and biochemical indexes from 2019-nCoV infected patients linked to viral loads and lung injury

Accepted Article This article is protected by copyright. All rights reserved

A Tool to Early Predict Severe Corona Virus Disease 2019 (COVID-19) : A Multicenter Study using the Risk Nomogram in Wuhan and Guangdong

Obesity and morbid obesity associated with higher odds of hypoalbuminemia in adults without liver disease or renal failure

Impaired albumin function: a novel potential indicator for liver function damage?

Renal histopathological analysis of 26 postmortem findings of patients with COVID-19 in China

The role of podocytes in proteinuria. Nephrology (Carlton)

Albumin in chronic liver disease: structure, functions and therapeutic implications

Clinical characteristics in urolithiasis formation according to body mass index

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Accepted Article