Correlation Between COVID-19 Severity, Body Mass Index and Hepatic Steatosis: A Retrograde Cohort Study

and fatty liver steatosis are already considered metabolic risk factors that may aggravate the severity of COVID-19. This study aims to investigate the correlation between COVID-19 severity, body mass index, and hepatic steatosis.


Background
Evidence from different studies have supported the notion that obesity increases the severity and mortality rate of COVID-19 patients. The co-morbidities associated with COVID-19 hospital admission are fairly similar to those associated with many other serious infectious illnesses that require hospital admission or ICU level care 1 . Nevertheless, it is particularly signi cant that obesity is a risk factor for death, ICU admission, tracheal intubation, and hospitalization related to COVID-19 2 .
The World Health Organization identi es non-communicable disorders such as obesity as a leading risk factor resulting in critical illness with COVID-19 3 . Obesity is a signi cant potential risk factor that was not illustrated in initial Chinese reports on COVID-19 patients circumstances which could probably explain why COVID-19 mortality rate is much higher in countries with higher obesity prevalence, such as Italy, relative to China and Japan 4;5 .
There are numerous biochemical pathways by which COVID-19 can affect people with obesity, which is well reported to be pro-in ammatory condition 6;7 . Some of these mechanisms are chronic in ammation and activation of ACE2-RAS system which are caused by excess adipose tissue in obese people.
By analogy with other respiratory infections, high body mass index (BMI) and obesity were predictors of poor results and deaths related to H1N1 8 . With regard to The pathophysiological mechanism connecting obesity and COVID-19, relevant evidence has been observed from other research carried out in subjects with H1N1 infection, showing that individuals with obesity have a more intense release of IL-8, compared to individuals with normal body weight, which is an important chemokine for the activation and migration of neutrophils to tissues, a mechanism involved in the response to infection 9;10 .
Fatty liver disease, which was formerly renamed Metabolic Associated Fatty Liver Disease (MAFLD), impacts 25% of the world's population 11 . Recent studies showed that MAFLD patients with SARS-CoV-2 infection increases the severity of COVID-19 respiratory illness by up to 4-6 fold 12-17 , while the severity of the respiratory illness in obese patients without (MAFLD) increases by up to 3 fold 13 . There is also an increased risk of MAFLD progression to Non-Alcoholic SteatoHepatitis (NASH) in the long-term 18 .
This study aims to examine the potential association between obesity, hepatic steatosis and the severity of COVID-19 illness.

Methods
Data was collected for consecutive laboratory-con rmed COVID-19 patients admitted at the hospital devoted completely by Jordanian Government to isolate and treat COVID-19 patients (Prince Hamza Hospital -PHH) between March and May 2020. Patients younger than 15 years were excluded. COVID-19 was diagnosed as a positive result by polymerase chain reaction (PCR) assay of oropharyngeal swab specimens.
Demographic information and past medical history were obtained and blood samples were tested on the rst day of hospital admission. A case was considered as a 'Severe COVID-19' case if the patient was admitted to the Intensive Care Unit (ICU). Hospital policy for admitting COVID-19 patients to the ICU included the following indications: 1. Respiratory or cardiac arrests 2. Respiratory rate ⩾40 or ⩽8 breaths/min 3. Oxygen saturation < 90% on ⩾50% oxygen 4. Respiratory acidosis 5. Pulse rate < 40 or > 140 beats/min 6. Systolic blood pressure < 90 mm Hg 7. Sudden fall in level of consciousness 8. Two or more organ failures Obesity was assessed by Body mass index (BMI). Patients with BMI ≥ 30 were considered obese. The presence of hepatic steatosis was con rmed by a liver CT scan. The Fibrosis-4 score was calculated to estimate the degree of brosis in the liver by a formula that incorporated values for age, Liver enzymes (AST, ALT) and platelet count. 19 The liver CT scan used in this study was non-enhanced, which has been found to be better at measuring the hepatic Houns eld unit and presence of steatosis 20 . Measurement of attenuation of liver only on unenhanced CT scans is best for prediction of pathologic fat content. 21 CT scan diagnostic criteria for steatosis are liver attenuation of at least 10 Houns eld Units (HU) less than that of the spleen or absolute liver attenuation of less than 40 HU. Unenhanced CT scans have a sensitivity for steatosis ranging from 43 to 95% and a speci city of 90-100%. 20 Continuous variables were expressed as mean (range). Differences between categorical variables were examined with the chi-squared test or the Fisher's exact test as appropriate.
The association between exposure factors such as obesity, liver steatosis and liver steatosis among obese patient vs. the outcome of COVID-19 severity was assessed by using the Spearman bivariate correlation coe cient with signi cance at 0.05% level (2 tailed).
The Odds Ratio (OR) between exposure factors such as obesity, liver steatosis and liver steatosis among obese patient vs. the outcome of our interest (COVID-19 severity) was assessed by using the binary logistic regression. The Odds Ratio adjusted for covariates such as age, gender, smoking, diabetes and hypertension was assessed by multinomial logistic regression analysis.
Statistical analyses were two-sided and signi cance was set at p < .05. All statistical tests were performed using SPSS version 23.0 (SPSS Inc., Chicago, USA).

Results
Data was collected for 302 patients. 72 patients younger than 15 years were excluded. 230 patients aged between 15 and 84 years were enrolled in this study. Table-1 and table-2 show their characteristic.
The presence of obesity was associated with an around 290 fold increased risk of severe COVID-19 illness. This association remained signi cant even after adjusting for age, gender, smoking, diabetes and hypertension. The presence of hepatic steatosis was associated with an around 17 fold increased risk of severe COVID-19 illness. This association remained signi cant even after adjusting for age, gender but not for smoking, diabetes and hypertension. The presence of both obesity and liver steatosis together was associated with an over 300 fold increased risk of severe COVID-19 illness. This association remained signi cant even after adjusting for age, gender, smoking, diabetes and hypertension. Table-3 summarizes the association between hepatic steatosis or obesity, or both and COVID-19 illness severity.

Discussion
Jordanian health authority policy regarding management of COVID-19 pandemic was to admit every patient with positive PCR COVID-19 regardless of symptoms or clinical condition. Many individuals were tested for COVID-19 only because they were in close contact with COVID-19 patients. Therefore most patients were asymptomatic. This also explains why the laboratory tests on admission for most patients were within normal range. Most patients did not require more than paracetamol as treatment. This is different from other countries such as China and Italy.
BMI positively and strongly correlates with ICU admission increasing the risk of severe COVID-19 illness to around 300 folds. This is consistent with other published results [23][24][25] . Additionally, the presence of hepatic steatosis increased the risk of severe COVID-19 illness to around 17 folds. The presence of hepatic steatosis among obese patient increased the risk to over 300 folds.
A large New York City cohort report revealed that obesity (BMI 30-40) is strongly correlated with a progression to serious illness with a relatively increased odds ratio than any cardiovascular or pulmonary condition 4 . Another study in the early days of the Italian epidemic demonstrated that the Case-fatality rate of patients hospitalized with COVID-19 was about 20%. Older age, obesity and disease severity upon admission were factors related with increased risk of death 26 .
Even after adjustment for other possibly confounding factors such as age, gender, race and troponin levels, obesity was reported to be associated with signi cantly higher levels of ICU admission and mortality in hospitals [27][28][29][30] . Higher demand for assisted ventilation beyond pure oxygen support (Invasive Mechanical Ventilation or Non-Invasive Ventilation) and increased admission to intensive or semiintensive care units has been identi ed in patients in overweight and obesity 31 .
While other literature does not provide data about steatosis and its correlation with severity of COVID-19, our study provides valuable evidence about this correlation. On the other hand, there was rather a weak correlation between FIB-4 and ICU admission. Additional studies are needed to con rm these ndings and to better understand the underlying mechanisms for why the association is greater in those with obesity.
The total number of COVID-19 positive patients in Jordan from March until July 2020 was about 1200 and only 10 of them died (0.8%). Our cohort (230 patients) represents about 19% of the total COVID-19 patients in Jordan. Our cohort contained 9 of the 10 patients who died and 90% of all patients admitted to ICU. Most mortalities occurred in the rst cohort of patients (300) that were diagnosed in Jordan. Only one mortality occurred in the following cohort of patients (900) that were not included in our study.
Several reports indicate that obesity may be a very signi cant consideration in younger people [32][33][34] . A report of 3,615 New York patients revealed that obesity was strongly correlated with hospital and ICU admittance for patients under 60 years of age 32 . Another study showed that hospitalized patients younger than 50 with morbid obesity are more likely to die from COVID-19. This is particularly relevant in the western world where obesity rates are high. In the same context, the ndings of another study found a greater prevalence of obesity in younger hospitalized patients 33;34 .
This is similar to what is suggested by literature. data suggests a longer viral shedding time + 5 days & abnormal liver function x7 folds in patients with (MAFLD) 12;16 . There is an increased risk of symptomatic infection in MALFD patients 35 .
No mortalities occurred among children in Jordan. The youngest patient who died was 44 years old and had hypertension, diabetes and cardiac disease.
The Hashemite kingdom of Jordan managed to keep the COVID-19 pandemic under control. Jordan followed an early and strict lockdown strategy and all infected cases were admitted to hospital regardless of symptoms. COVID-19 did not widely spread in Jordan, which has a population of about 10 million. COVID-19 was only found in 1200 patients over 4 months, which is (0.012%).
Most Jordanian patients did not have symptoms. Most patients did not require treatment or antiviral medication. Therefore this could be the reason for the generally unaltered liver function enzymes.
Some limitations of this study should be recognized. While the liver was assessed by CT scan for severe steatosis, patients included in our study did not undergo liver biopsy, thus COVID-19 severity in relation to liver histology could not be assessed.
Another common characteristic of obesity is vitamin D de ciency, which raises the risk of bacterial infections and impairs the immunity response 36 . Vitamin D has several mechanisms by which it eliminates the chance of microbial infection and death according to an analysis of the role of vitamin D in lowering the risk of common cold, which divides these mechanisms into three groups: physical barrier, cellular natural immunity and adaptive immunity 37 . Level of Vitamin D was not tested in patients in this study.
In addition, patients were mostly of Arab ethnicity and thus the applicability of the results to other ethnic groups is uncertain.

Conclusions
BMI remained the most noticeable factor that strongly correlated with COVID-19 severity. Obesity even in the absence of hepatic steatosis greatly increased the risk for severe COVID-19. This association remained signi cant after adjusting for likely confounders. The presence of liver steatosis even in nonobese patients increased the risk for severe COVID-19 but to a much lesser degree compared to obesity. The risk of steatosis to COVID-19 severity was greater in those with, than those without obesity. This association also remained signi cant after adjusting for likely confounders. Availability of data and materials Data available upon request from the authors.

Competing of Interest Declaration
The authors declare that they have no competing interests.

Funding State
None Authors' contributions -Haitham Qandeel supervised all other co-authors.
-Haitham Qandeel and Raed Tayyem were the head of this research.
-Haitham Qandeel, Raed Tayyem, and Rand Tayyem conceived, designed and perform the statistical analysis of the data.
-Jehad Fataftah performed and analyzed the liver CT scans.
-Rami Alqassieh participated in writing this manuscript and went through the manuscript for revision.
-Haneen Ashour and Subreen Hasanat read page proofs, conduct the literature review, managed the references and communicate with journals about submissions.