Annals of African Medicine
Home About AAM Editorial board Ahead of print Current Issue Archives Instructions Subscribe Contact us Search Login 

Table of Contents
Year : 2023  |  Volume : 22  |  Issue : 2  |  Page : 176-182  

Variation in biochemical parameters in COVID-19 patients admitted at a tertiary care dedicated COVID hospital: A prospective study

1 Department of Biochemistry, All India Institute of Medical Sciences, Patna, Bihar, India
2 Department of Community and Family Medicine, All India Institute of Medical Sciences, Patna, Bihar, India

Date of Submission15-Feb-2022
Date of Decision26-May-2022
Date of Acceptance03-Jun-2022
Date of Web Publication4-Apr-2023

Correspondence Address:
Ayan Banerjee
Department of Biochemistry, All India Institute of Medical Sciences, Phulwarisharif, Patna - 801 507, Bihar
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/aam.aam_37_22

Rights and Permissions

Introduction: COVID-19 infection has a myriad of presentation. Rural India and other developing nations are relatively resource poor, not having access to modern specialized investigations. In this study, we tried to evaluate only biochemical parameters in predicting the severity of the infection. The aim of this study was to find a cost-effective means to predict the clinical course at the time of admission and thereby to reduce mortality and, if possible, morbidity by timely intervention. Materials and Methods: All COVID-19-positive cases admitted at our hospital from March 21 to December 31, 2020, were recruited in this study. The same acted as sham control at recovery. Results: We observed a significant difference in biochemical parameters at the time of admission and discharge, between mild/moderate disease and severe disease. We found slightly deranged liver function tests at admission, which becomes normal at the time of discharge. Urea, C-reactive protein (CRP), procalcitonin, lactate dehydrogenase, and ferritin concentrations in severe/critical patients were significantly higher than that in the mild/moderate group. Receiver operating characteristic curves were plotted to predict the severity on the basis of biochemical parameters independently, of the patients based on these values. Conclusion: We proposed cutoff values of certain biochemical parameters, which will help in judging the severity of the infection at admission. We developed a predictive model with a significant predictive capability for CRP and ferritin values, using normal available biochemical parameters, routinely done in resource-poor centers. Clinicians working in resource-poor situations will be benefitted by having an idea of the severity of the disease. Timely intervention will reduce mortality and severe morbidity.

   Abstract in French 

Introduction: L'infection au COVID19 a une myriade de présentations. L'Inde rurale et d'autres pays en développement sont relativement pauvres en ressources, non avoir accès aux enquêtes spécialisées modernes. Dans cette étude, nous avons essayé d'évaluer uniquement les paramètres biochimiques pour prédire la gravité de l'infection. Le but de cette étude était de trouver un moyen rentable de prédire l'évolution clinique au moment de l'admission et ainsi de réduire la mortalité et, si possible, la morbidité par une intervention rapide. Matériels et méthodes: Tous les cas positifs au COVID19 admis à notre hospitalisés du 21 mars au 31 décembre 2020, ont été recrutés dans cette étude. La même chose a agi comme un contrôle factice lors de la récupération. Résultats: Nous avons observé une différence significative dans les paramètres biochimiques au moment de l'admission et de la sortie, entre une maladie légère/modérée et une maladie grave. Nous avons trouvé des tests de la fonction hépatique légèrement dérangés à l'admission, qui deviennent normaux au moment de la sortie. Urée, protéine Créactive (CRP, les concentrations de procalcitonine, de lactate déshydrogénase et de ferritine chez les patients sévères/critiques étaient significativement plus élevées que chez les patients légers/modérés groupe. Les courbes caractéristiques de fonctionnement du récepteur ont été tracées pour prédire la gravité sur la base de paramètres biochimiques indépendamment, deles patients en fonction de ces valeurs. Conclusion: Nous avons proposé des valeurs seuils de certains paramètres biochimiques, qui permettront de juger de la gravité de l'infection à l'admission. Nous avons développé un modèle prédictif avec une capacité prédictive significative pour les valeurs de CRP et de ferritine, en utilisant les paramètres biochimiques normaux disponibles, systématiquement effectués dans les centres pauvres en ressources. Les cliniciens travaillant dans des situations où les ressources sont limitées bénéficier d'avoir une idée de la gravité de la maladie. Une intervention rapide réduira la mortalité et la morbidité grave.
Mots-clés: COVID19, ferritine, lactate déshydrogénase, urée

Keywords: COVID-19, ferritin, lactate dehydrogenase, urea

How to cite this article:
Kumar V, Mahto M, Kumar S, Bansal A, Ranjan A, Ahmad S, Banerjee A. Variation in biochemical parameters in COVID-19 patients admitted at a tertiary care dedicated COVID hospital: A prospective study. Ann Afr Med 2023;22:176-82

How to cite this URL:
Kumar V, Mahto M, Kumar S, Bansal A, Ranjan A, Ahmad S, Banerjee A. Variation in biochemical parameters in COVID-19 patients admitted at a tertiary care dedicated COVID hospital: A prospective study. Ann Afr Med [serial online] 2023 [cited 2023 Jun 6];22:176-82. Available from:

   Introduction Top

After the confirmation of the first case of COVID-19 from Wuhan, China, the disease has spread globally to almost all the countries in the world. It has been declared pandemic by the World Health Organization (WHO) dated April 6, 2020.[1] The virus which causes the coronavirus disease 2019 (COVID-19) was named SARS-CoV-2 by the International Committee of Taxonomy of Viruses.[2] This pandemic has created a major burden on economy and patients mortality and morbidity not only in India but also globally.

Coronaviruses are positive-sense single-stranded ribonucleic acid (RNA) viruses. The RNA genome of SARS-CoV-2 has about 30,000 nucleotides, encoding for 29 proteins.[3] Gold standard diagnosis of COVID-19 is done by molecular identification of the SARS-CoV-2 using nucleic acid amplification tests such as the reverse transcriptase-quantitative polymerase chain reaction (RT-qPCR).[4] Since India and many developing countries are not equipped with sufficient laboratory and human resource capacity to perform massive molecular identification, it leads to delays between testing and confirmation. Even after confirmation, clinical presentation is unpredictable. Till date, no single parameter is good enough to predict mortality caused by this virus. Patient history, hematological and biochemical laboratory parameters, and imaging are the mainstays for monitoring patient status.

Major concern about SARS-CoV-2 is that this disease is rapidly deteriorating the health condition of high-risk groups such as elderly and immune compromised people,[5] so timely detection and monitoring of disease progression is much needed. Rural India as well as many developing countries also does not have the access to various specialized investigations such as interleukin-6 (IL-6), ferritin, and C-reactive protein (CRP). Hence, in this study, we focused on the variation in various biochemical parameters in COVID-19 patients and correlated accordingly. In this, we tried to evaluate the variations of the biochemical parameters in COVID-19-infected patients at admission and during discharge. We also tried to create a model to predict the values of IL-6, ferritin, and CRP from routine biochemical investigations along with the age and gender of the patients.

There are few studies done worldwide analyzing only the biochemical parameters and very few have been reported from India. Hence, the present study aimed to evaluate variation in various biochemical parameters in COVID-19 patients at active infection and with when they have clinically recovered.

   Materials and Methods Top

All patients admitted at our hospital from March 21 to December 31, 2020, were recruited in this study. Severity of the disease was classified as diagnosed positive cases of COVID-19 through RT-PCR with clinical picture: mild (PaO2/FIO2 = 101–200 mmHg), moderate (PaO2/FIO2 = 101–200 mmHg with PEEP ≥5 cm of H2O), and severe cases (PaO2/FIO2 ≤100 mmHg with PEEP ≥5 cm of H2O).

LFT, KFT, CRP, and lactate dehydrogenase (LDH) were measured on a BECKMAN COULTER AU680 machine as per the manufacturer's protocol.

Ferritin and IL-6 were measured by chemiluminescent method on SIEMENS ADVIA CENTAUR as per the manufacturer's protocol. Patients who had expired were excluded from the study.

Statistical analysis

All analyses were carried out using SPSS, version 22 (SPSS Inc., Chicago, IL, USA). Categorical variables were presented as proportion. Continuous variables were presented as median and interquartile range values. Shapiro–Wilk test was used to verify the normal distribution of all continuous variables taking P < 0.05. Almost all the biochemical variables were not normally distributed; hence, nonparametric tests, such as Mann–Whitney U-test, were applied for the continuous variables and Chi-square test for the categorical variables. Spearman's correlation coefficient was used to measure association between severities of COVID-19 with biochemical variables with significance test for the correlation coefficient. Receiver operating characteristic (ROC) curve analysis was used to determine the optimum cutoff points of significantly correlated biochemical variables with severity of disease and area under curve with 95% confidence interval with significance value for each variable. The cutoff point was determined using the index of union method. A 2-tailed P < 0.05 was considered statistically significant.

   Results Top

The study population included 202 hospitalized patients with confirmed COVID-19. Out of 202 total cases, 43 cases were of severe condition and 159 cases were of mild/moderate level. The mean age was 51.22 years (95% confidence interval [CI]: 49.1–53.3 years); there was no statistically significant difference in age (P = 0.42) between severe cases (52.8 years; 95% CI: 48.37–57.30) and mild/moderate cases (50.78 years; 95% CI: 48.37–53.19 years). There was no significant gender distribution between the two groups of disease severity (Chi-square at 1, df = 1.49; P = 0.22). The proportion of male cases was 76.7%.

From [Table 1], it is obvious that a highly significant difference was observed at admission in the biochemical variables such as urea, sodium, CRP, ferritin, LDH, and procalcitonin (PCT) between severe cases as compared to mild/moderate cases of COVID-19. Inflammatory parameters as well as acute phase reactants such as CRP, ferritin, and PCT are more elevated in severe cases compared to mild–moderate cases. Elevated urea, sodium, serum glutamic oxaloacetic transaminase (SGOT), and serum glutamic pyruvic transaminase (SGPT), may be, and are indicative that the patients belonging to the severe category have mildly deranged kidney and liver parameters.
Table 1: Comparison of biochemical variables of mild/moderate and severe coronavirus disease 2019 cases at admission

Click here to view

From [Table 2], it is evident that there is a significant difference in the biochemical variables such as total protein, albumin, globulin, ferritin, and LDH at discharge between severe cases as compared to mild/moderate cases of COVID-19. All the patients were discharged based on the clinical recovery. At the time of discharge, values of some inflammatory parameters and acute phase reactants as well as well SGOT and SGPT have fallen as compared with the time of admission, which signifies better outcome. Patients belonging to the severe category were admitted to the hospital for a prolonged duration when compared to the mild/moderate category, which may explain the lower protein, albumin, and globulin levels in them, at the time of discharge. Even though these patients were discharged based on clinical recovery, elevated ferritin and LDH values may indicate that they have not recovered biochemically completely and these parameters may take longer time to attain its normal values respectively, more gradually.
Table 2: Comparison of biochemical variables of mild/moderate and severe coronavirus disease 2019 cases at discharge

Click here to view

Further, we wanted to evaluate whether there exists any significant correlation between the biochemical parameters and severity of the disease, at the time of admission. [Table 3] presents the correlation coefficient of all biochemical variables with severity status of the patients at admission. Biochemical variables such as urea, sodium, CRP, ferritin, LDH, and PCT were found to be positively and significantly correlated with the severity of COVID-19. The positive correlation establishes the fact that disease severity is reflected by increased values of these parameters. These parameters will further help the clinicians to categorize the patients as mild/moderate or severe independently in resource poor situations.
Table 3: Correlation coefficient and P value between items and disease severity

Click here to view

Next, we wanted to determine a specific cutoff value of the biochemical parameters which can be used to predict the severity of the disease category independent of the clinical categorization. For this, we plotted a ROC curve for all the parameters which were significantly different between the two groups. The ROC curves of significantly associated biochemical variables were determined and are listed in [Table 4]. The ROC curves for urea (area under the curve [AUC] = 0.645, 95% CI: 0.56–0.73; P = 0.004; [Figure 1]a) suggested that the best cutoff point was ≥38 units with a sensitivity of 60.9% and a specificity of 60.8%. The ROC curve for sodium (AUC = 0.610, 95% CI: 0.52–0.70; P = 0.032; [Figure 1]b) suggested the best cutoff points ≥135.5 mEq/L with a sensitivity of 60.5% and a specificity of 51.6%. The ROC curve for CRP (AUC = 0.60, 95% CI: 0.51–0.696; P = 0.047; [Figure 1]c) suggested the best cutoff point ≥69 mg/L with a sensitivity of 60% and a specificity of 54.4%. The ROC curve for ferritin (AUC = 0.645, 95% CI: 0.55–0.734; P = 0.004; [Figure 1]d) suggested the best cutoff point ≥556 ng/ml with a sensitivity of 65% and a specificity of 55.6%. The ROC curve for LDH (AUC = 0.655, 95% CI: 0.56–0.75, P = 0.002. [Figure 1]e) suggested the best cutoff point ≥816 U/L with a sensitivity of 65.1% and a specificity of 53.5%. The ROC curve of PCT (AUC = 0.63 95% CI: 0.53–0.73, P = 0.015; [Figure 1]f) suggested the best cutoff point ≥0.47 ng/ml with a sensitivity of 63.2% and a specificity of 60.6%. The ROC curve for Chloride (AUC = 0.596, 95% CI: 0.51–0.659, P = 0.051; [Figure 1]g) suggested the best cutoff point ≥100 mEq/L with a sensitivity of 58.1% and a specificity of 61.1%.
Table 4: Receiver operating characteristic curves for cutoff values of significant biochemical variables

Click here to view
Figure 1: (a) ROC curve of urea at baseline. (b) ROC curve of sodium at baseline. (c) ROC curve of CRP at baseline. (d) ROC curve of ferritin at baseline. (e) ROC curve of LDH at baseline. (f) ROC curve of PCT at baseline. (g) ROC curve of chloride at baseline. ROC: Receiver operating characteristic, CRP: C-reactive protein, LDH: Lactate dehydrogenase, PCT: Procalcitonin

Click here to view

There are various hospitals which do not have access to specialized investigations. Hence, we tried to develop a predictive model for CRP, ferritin, and IL-6 at baseline using the routine tested parameters, age, and gender. [Table 5] and [Table 6] show prediction of model of CRP and ferritin at baseline of COVID-19 patients.
Table 5: Prediction of C-reactive protein at baseline of coronavirus disease 2019 patients

Click here to view
Table 6: Prediction of Ferritin at baseline of coronavirus disease 2019 patients

Click here to view

Model: F (5, 180) = 7.42; P < 0.001: R2 = 0.1708 (SEX = 1 for Male and 0 for Female in the above formula).

CRP = 479 + 31.43(SEX) 131.57 (AG) +0.52 (UREA) 13.64 (PHOS) 2.45 (CHLORIDE)

(Sex = 1 for Male and 0 for Female in the above formula). R2 = 0.1708

FERRITIN = 1145.33 + 215.06 (SEVERITY) +1.948 (SGPT) +6.25 (UREA) 121.26 (TP) 66.48 (CREATININE)

(Severity value is 1 for severe cases and 0 for mild/moderate cases in the above formula). R2 = 0.1933

Both the models predictive value is highly significant at P < 0.001.

   Discussion Top

We have tested almost all biochemical parameters that are done in our department and analyzed them. Most of the patients were of mild to moderate severity and had better recovery.[6] However, it is important to identify severely ill patients earlier so that timely management can be initiated and mortality can be prevented in some cases. We observed almost 202 COVID-19 patients having clinical features such as fever, fatigue, and cough as observed in other studies.[7] In this study, we observed the difference in biochemical parameters at the time of admission between the patients who were admitted in normal wards (mild/ moderate category) and in intensive care unit (severe category), but largely there was no significant difference in those parameters during discharge, which were significantly different at the time of admission, further establishing the fact the criteria set for discharge were robust and well thought of. A study done by Zhang C et al. showed that 2%–11% of patients with COVID-19 had liver comorbidities and 14%–53% of cases had abnormal alanine transaminase and aspartate transaminase levels during progression of COVID-19 disease. This study also showed that liver damage in mild cases of COVID-19 is often transient.[8] Our study also corroborated their findings as there is mild elevation in SGOT/SGPT in mild/moderate and severe category, respectively, at admission and it decreases with time at the time of discharge. There was a difference in the values of SGPT ,though not statistically significant (P < 0.06) at admission between the mild/moderate and the severe category of the patients stating deranged liver function tests at admission, more in case of severe category which recovers fully at the time of discharge of the patients. Another study done by Xu Wei et al.[9] had shown that urea and creatinine are increased in COVID-19 patients. Our study revealed that urea was significantly increased in severe variety of cases at the time of admission, but at the time of discharge, both urea and creatinine were normal. Almost all biochemical parameters are in the same range at the time of discharge as there is no significant difference between mild/moderate and severe cases except for total protein, ferritin, and LDH.

In conformity with Liu J's[10] and Wan S's[11] study, our study also found that levels of CRP were significantly associated with the severity of COVID-19. In addition, LDH, PCT, and ferritin were also related to disease severity in a significant way.

There are 5 isoenzymes of LDH in humans (LDH-1 in cardiomyocytes, LDH-2 in reticuloendothelial system, LDH-3 in pneumocytes, LDH-4 in kidneys and pancreas, and LDH-5 in liver and striated muscle). Multiple organ injury and decreased oxygenation with upregulation of the glycolytic pathway is one of the major causes of elevated LDH values apart from cardiac damage. There is cytokine-mediated tissue damage and LDH release in severe infections. Severe form of interstitial pneumonia with acute respiratory distress is the salient feature of COVID-19. As LDH isoenzymes 3 is present in lung tissue, greater release of this enzyme is natural in COVID-19 infection.[12] In our study, there was a significant difference of LDH at admission and discharge between the mild/moderate and severe group of patients in consonance with the above-mentioned study. Even though there was a significant difference at discharge between the groups, the values of both the groups had fallen significantly when compared with that of the admission. The higher value at discharge in the severe category of patients is probably due to prolonged recovery time taken by these.

In our study, the PCT, LDH, and ferritin concentration in severe/critical patients was significantly higher than that in the mild/moderate group. CRP and ferritin are nonspecific acute-phase proteins and are increased in cases of inflammation.[13] To the best of our knowledge, this study is one of the early to determine relationships between CRP, ferritin, LDH, PCT, and COVID-19. We have derived cutoff values for ferritin ≥556.4 ng/ml, LDH ≥816 U/l, PCT ≥0.47 ng/ml, CRP ≥69.9 mg/L, which can stratify patients category based on biochemical parameters only. These cutoff values can be applied to fresh infections and the severity of the disease can be predicted using these values. The values above this cutoff were more likely to develop severe disease. High ferritin levels may be due to secondary hemophagocytic lymphohistiocytosis (sHLH) and cytokine storm syndrome in a severe variety of COVID-19 patients.[14] sHLH comprises an underrecognized, hyperinflammatory syndrome having a fulminant and fatal hypercytokinemia with multiorgan failure. In adults, sHLH is most commonly provoked by viral infections,[15] and it occurs in 3·7%–4·3% of patients having sepsis.[16]

As India is a resource-poor country, markers such as CRP, ferritin, IL-6, and PCT are not tested readily in rural regions. We tried to develop models which could be used to predict these values using simple routine biochemical parameters which are widely available. We tried to predict CRP and ferritin with the normal available biochemical parameters which are readily done in the periphery. Prediction of CRP was done with gender, AG ratio, urea, phosphorous, and chloride, while that of ferritin was done with SGPT, urea, total protein, and creatinine. These models can predict with a significance value of <0.001. These models can be used in rural set up not only in India but also other developing nations to predict CRP and ferritin values, which are important biochemical parameters used in disease stratification. We could not predict a successful model for IL-6 as the sample size was small for IL-6 to develop a predictive model.

   Conclusion Top

This study will help in stratifying COVID-19 patients in limited resource centers. However, our study is with certain lacunae. The sample size should have been even much larger to predict the models. Moreover, the models could not be tested in patients. Another study with a much bigger sample size is required to further reinforce our findings and test the predicted models.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

   References Top

Coronavirus Disease 2019 (COVID 19): Situation Report (GLOBAL). Coronavirus disease (COVID-19) Weekly Epidemiological Update and Weekly Operational Update. Available from [Last accessed on 2020 Mar 11].  Back to cited text no. 1
Gorbalenya AE, Baker SC, Baric RS, de Groot RJ, Drosten C, Gulyaeva AA, et al. The species severe acute respiratory syndrome-related coronavirus: classifying 2019-nCoV and naming it SARS-CoV-2. Coronaviridae Study Group of the International Committee on Taxonomy of Viruses. Nature Microbiology 2020;5:536-44.  Back to cited text no. 2
Lu R, Zhao X, Li J, Niu P, Yang B, Wu H, et al. Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding. Lancet 2020;395:565-74.  Back to cited text no. 3
World Health Organization. (2020). Laboratory testing for coronavirus disease 2019 (COVID-19) in suspected human cases: interim guidance, 2 March 2020. World Health Organization. Available from: WHO/COVID-19/laboratory/ 2020.4. [Last accessed on 2020 Mar 02].  Back to cited text no. 4
Wu Z, McGoogan JM. Characteristics of and important lessons from the coronavirus disease 2019 (COVID 19) outbreak in China: Summary of a report of 72 314 cases from the Chinese Center for Disease Control and Prevention. JAMA 2020;323:1239-42.  Back to cited text no. 5
Team NCPERE. The epidemiological characteristics of an outbreak of 2019 novel coronavirus diseases (COVID-19) in China. Zhonghua Liu Xing Bing Xue Za Zhi 2020;41:145-51.  Back to cited text no. 6
Guan WJ, Ni ZY, Hu Y, Liang WH, Ou CQ, He JX, et al. Clinical characteristics of coronavirus disease 2019 in China. N Engl J Med 2020;382:1708-20.  Back to cited text no. 7
Zhang C, Shi L, Wang FS. Liver injury in COVID-19: Management and challenges. Lancet Gastroenterol Hepatol 2020;5:428-30.  Back to cited text no. 8
Hong XW, Chi ZP, Liu GY, Huang H, Guo SQ, Fan JR, et al. Characteristics of renal function in patients diagnosed with COVID-19: An observational study. Front Med (Lausanne) 2020;7:409.  Back to cited text no. 9
Liu J, Li S, Liu J, Liang B, Wang X, Wang H, et al. Longitudinal characteristics of lymphocyte responses and cytokine profiles in the peripheral blood of SARS-CoV-2 infected patients. EBioMedicine 2020;55:102763.  Back to cited text no. 10
Wan S, Yi Q, Fan S, Lv J, Zhang X, Guo L, et al. Characteristics of Lymphocyte Subsets and Cytokines in Peripheral Blood of 123 Hospitalized Patients with 2019 Novel Coronavirus Pneumonia (NCP); 2020. Available from: [Last accessed on 2020 Mar 24].  Back to cited text no. 11
Henry BM, Aggarwal G, Wong J, Benoit S, Vikse J, Plebani M, et al. Lactate dehydrogenase levels predict coronavirus disease 2019 (COVID-19) severity and mortality: A pooled analysis. Am J Emerg Med 2020;38:1722-6.  Back to cited text no. 12
Pepys MB, Hirschfield GM. C-reactive protein: A critical update. J Clin Invest 2013;111:1805-12.  Back to cited text no. 13
Mehta P, McAuley DF, Brown M, Sanchez E, Tattersall RS, Manson JJ, et al. COVID-19: Consider cytokine storm syndromes and immunosuppression. Lancet 2020;395:1033-4.  Back to cited text no. 14
Ramos-Casals M, Brito-Zerón P, López-Guillermo A, Khamashta MA, Bosch X. Adult haemophagocytic syndrome. Lancet 2014;383:1503-16.  Back to cited text no. 15
Karakike E, Giamarellos-Bourboulis EJ. Macrophage activation-like syndrome: A distinct entity leading to early death in sepsis. Front Immunol 2019;10:55.  Back to cited text no. 16


  [Figure 1]

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
    Materials and Me...
    Article Figures
    Article Tables

 Article Access Statistics
    PDF Downloaded2    
    Comments [Add]    

Recommend this journal