Discussion
In this study, we aimed to monitor the changes in clinical, laboratory features, and L/S of COVID-19 patients during the course of disease. Many patients who are hospitalized due to COVID-19 get multiple chest CT scans depending on the clinical progression. There are a few series in the literature that have demonstrated how evolution of thoracic imaging findings happen [2, 11, 20]. In spite of the commonly observed chest imaging findings associated with COVID-19, reports of abdominal imaging findings are limited. A study has demonstrated that bowel involvement and cholestasis are the most common abdominal imaging findings in COVID-19 patients [23]. It has been reported that 16–53% of patients with COVID-19 have abnormal liver function tests during disease progression [24–26]. However, how the changes in liver parenchyma will be seen on imaging is not clear. It is known that changes of hepatic attenuation may be observed in some diffuse liver diseases and L/S is commonly used to assess the presence of liver fat [14, 15, 17]. In our study, we hypothesized the use of L/S to evaluate the changes in hepatic attenuation between the two consecutive scans and investigated if it may correlate with abnormal liver function tests. We found that L/S at follow-up was significantly different between the progressive and non-progressive groups. As the lung CT scores increased, L/S tended to decrease on follow-up CTs. Similarly, in our study, liver function tests did not differ between the two groups of patients at hospital admission however, the increase in ALT values at follow-up was significant in patients with progressed lung CT scores (p = 0.02). Additionally, in patients with decreased L/S, ALT levels were elevated at follow-up (p = 0.004).
Although it is difficult to determine the underlying mechanism, our study demonstrated the changes in hepatic attenuation during the course of disease in COVID-19 patients. It has been proposed that liver injury in COVID-19 patients might be the direct result of the infection of liver cells [13]. The derangement of liver function is generally mild and other explanations for the liver damage such as immune-mediated inflammation, cytokine storm, and pneumonia-associated hypoxia are suggested [13, 27]. Abnormal liver function tests may also be observed in drug-induced hepatotoxicity with the use of hydroxychloroquine, azithromycin, and favipiravir in COVID-19 patients [28–30]. In our study, only in the group of patients who received favipiravir therapy had significantly elevated AST and ALT levels at follow-up, however, there was no change of L/S between the groups.
It has been suggested that hepatic steatosis might develop during the course of COVID-19 [31]. Histopathological examinations of the liver in COVID-19 patients have shown hepatic steatosis, mild lobular and portal inflammation [31, 32]. The decrease in L/S in the progressive group on follow-up that was identified in our study might be attributed to hepatic steatosis. In general, hepatic steatosis is most commonly associated with alcohol abuse and non-alcoholic fatty liver disease but it may be caused by various factors including viral hepatitis, ischemia, drug toxicity, and malnutrition [33]. Moreover, it has been demonstrated that SARS-CoV-2 infection may affect lipid metabolisms especially in patients with severe disease [34, 35]. Thus, the causes of decreased L/S in patients with progressive disease need to be further investigated.
Chest CT plays an important role in the diagnosis and evaluation of severity of COVID-19. In our study, patients with increased lung CT scores had a significantly longer hospital stay (p = 0.035). Similarly, Li et al. reported that the CT scores of the patients with critical disease were higher than those of the ordinary COVID-19 patients and lung CT scores can be used to accurately differentiate severe patients [12]. In our study, in patients with progressed lung CT scores, CRP levels were found to be significantly elevated at follow-up. It has been suggested that increased values of CRP may be positively correlated with lung involvement and disease severity [36].
Laboratory parameters such as WBC, absolute values of lymphocytes, platelets, albumin, total bilirubin, creatinine, blood urea nitrogen, CRP, interleukin-6, AST, LDH, myoglobin, troponin, procalcitonin, and d-dimer have been reported to be risk factors for critical or mortal COVID-19 cases [10, 37]. In our study, WBC, neutrophil, lymphocyte, monocyte, and platelet levels at hospital admission were significantly different between the progressive and non-progressive groups of patients. Among these parameters, platelet count < 230 × 103/µL predicted the progression of lung CT score with a sensitivity of 95% and specificity of 85.7%.
Our study has several limitations including its retrospective design and small number of patients. Another limitation is that we did not evaluate the thoracic imaging characteristics of COVID-19. It has been reported that consolidation, linear opacities, crazy-paving pattern, bronchial wall thickening, and extrapulmonary lesions may be the imaging findings of severe COVID-19 patients. In our study, we evaluated the changes of L/S during the course of disease and used the lung CT scores to assess the extension and severity of the lung involvement in COVID-19 patients. Another limitation is the lack of contrast-enhanced liver imaging. Since multiple unenhanced chest CT scans were performed for COVID-19 patients, we wanted to investigate the changes in upper abdomen and focused on L/S. As L/S can be affected by the pre-existing liver diseases, we only evaluated the changes on two consecutive scans and excluded COVID-19 patients who had single CT scans. However, prospective studies with larger number of patients, using different imaging modalities are required to confirm and radiologically detect the changes in liver parenchyma, which could be multifactorial, in patients with COVID-19.
In conclusion, decrease in L/S may be observed in COVID-19 patients with elevated lung CT scores at follow-up. Additionally, WBC, neutrophil, lymphocyte, monocyte, and platelet counts that are obtained at hospital admission may predict the progression of disease.