Studying C-reactive protein and D-dimer levels in blood may prevent severe complications in Bangladeshi COVID-19 patients

The ongoing COVID-19 pandemic has been a scientific, medical and social challenge. Since clinical course of this disease is largely unpredictable and can develop rapidly causing severe complications, it is important to identify laboratory biomarkers which may help to classify patient’s severity during initial stage. Previous studies have suggested C - reactive protein (inflammatory) and D-dimer (biochemical) as an effective biomarker. The differential severity in patients across the world and our limited understanding in the progression of the disease calls for a multi-country analysis for biomarkers. Therefore, we have analyzed these biomarkers among 228 Bangladeshi COVID-19 patients. We show significant association of COVID-19 severity with these two biomarkers. Thus, we suggest using these biomarkers for Bangladeshi COVID-19 patients for disease monitoring. Such validated preventive measures may decrease the case fatality ratio substantially.


Introduction
COVID-19 patients are categorized into mild, moderate, severe, and critical based on disease severity. In order to stratify high risk patients there is an urgent need for reliable biomarkers related to coronavirus disease 2019 (COVID- 19) progression. Novel biomarkers are also needed to understand viral pathogenetic mechanisms and to categorize patients into those who suffer rapid disease progression to severe complications like death. Association of severity with some hematological parameters like white blood cell (WBC), lymphopenia, and some biochemical parameters, such as LDH, creatine kinase (CK), IL-6 and troponin have already been reported to be associated with COVID-19 severity [1].
C-reactive protein (CRP) is a pentameric protein, synthesized primarily in liver under the action of cytokine interleukin 6 (IL-6) [2]. Levels of CRP > 50 mg/dL are reported to be associated with bacterial infections while more elevated levels are also seen in injuries, cardiovascular processes and other in ammatory states [2]. D-dimer is a brin degradation product (or FDP), a small protein fragment present in the blood after a blood clot is degraded by brinolysis [3] Higher D-dimer levels are indicative of the activation of coagulation and following brinolytic process [3] and, are reported in in uenza like infections by respiratory viruses even before the outbreak of coronavirus pandemic [4].
Although number of studies have already shown the association of COVID-19 severity with CRP and Ddimer levels in blood, there are limited studies done on Bangladeshi populations [5,6,7,8,9]. In a recent study done by Ullah, Waqas et al [2] both high D-Dimer (> 501 ng/ml) and high CRP (> 101 mg/dl) were associated with increased need for upgrade to the ICU and higher requirement for (intermittent mandatory ventilation) IMV on day-7 of hospitalization (2,25). Therefore, for the rst time in Bangladeshi individuals, plasma CRP and D dimer levels is demonstrated to assist for discerning patients with low to severe (live and dead) COVID-19 pneumonia belonging to different age groups. This suggest that these testing may be useful as an earlier indicator for severe illness and help physicians to stratify patients for intense care unit transfer [10]. Our aim is to focus on CRP and D-dimer levels, which are potentially predictive of patients with severe complications and death in COVID-19 infection. In order to screen out biochemical indicators that are meaningful for the diagnosis of disease progression, we consulted the laboratory test results of all the dead and recovered patients from cabin/Ward and ICU.

Data Collection
In this retrospective study, we include hematological data of all con rmed cases of COVID-19 from Anwar Khan Modern Medical College Hospital, Lab Aid Medical College Hospital, and in some cases communication with patient's family from June 2020 to November 2020. Ethical permission was taken from the Biological Science Faculty, Dhaka University, Bangladesh. The medical treatment consent was obtained by the physicians of the two Hospitals. We have called few patients (or their family) and explained this study to them to obtain their consent.
COVID-19 test were done using reverse transcriptase polymerase chain reaction (RT-PCR) technique. We have accessed to patients' data at the end of November 2020, only with serial number (fully anonymous).
We include those patients with criteria having CRP and D-dimer results and admitted to ICU and non-ICU.
We considered 144 cases from ICU, 84 cases from non-ICU and through telephone communication. The patients' lab test data also include age, gender, patients' status and hospital status.
All medical laboratory data including the concentrations of D-dimer, and high-sensitivity C-reactive protein (CRP), were generated by the Biochemistry laboratory of two Hospitals recorded electronically in their database. The samples for laboratory tests were collected on admission and during the hospital stay or visit. Peripheral venous blood was collected and D-dimer was measured with a latex particle-enhanced immune turbidimetric assay Sysmex 1000 (Siemens Healthcare, NY, USA)). It is worth mentioning that CRP tests were detected using Dimensional RxL Max integrated chemistry system (Siemens Healthcare, NY, USA) in both hospitals.

Statistical Analysis
CRP and D-Dimer levels were collected from the patients admitted in Cabin/Ward and ICU. The Pearson Correlation test was applied to test the correlation among patient's age, hospital status, D-dimer and CRP value and patient's status (Survived or Died) using SPSS v 26.0.0.0. We recoded patient recovered and dead status as 0 and 1 respectively. Similarly, male and female sexes were coded as 1 and 0 respectively. Hospital Cabin/Ward patients and ICU patients were coded into 0 and 1. We ran Logistic Regression Test for the combined dataset keeping Severity as dependent variable and CRP and D-dimer levels as independent. A P value below 0.05 was considered statistically signi cant.

Results And Discussion
In ammatory responses play a crucial role in progression of COVID-19 [11,12,13]. Rapid viral replication of the SARS-CoV-2 virus involves recruitment of macrophages and monocytes, release of cytokines and chemokines, thus triggering in ammatory responses [14]. CRP (C-Reactive protein) has been reported in dengue patients [15]. Dengue virus and SARS-CoV-2 are RNA virus, share similarity in the course of infection [15]. CRP is rapidly synthesized in response to a variety of eukaryotic and prokaryotic pathogens, by hepatocytes when stimulated by in ammation, facilitating complement activation through classical pathway[16], indicating immune activation, lymphocyte in ltration, immune molecules consumption and in ammation outbreak. It is also inferred that increased CRP levels could be early indicators of nosocomial infections in COVID-19 patients who were slow to recover [10]. Higher levels of CRP, a potential in ammatory biomarker has been reported to be signi cantly associated with disease severity in COVID-19 infections [10,17,18]. CRP levels have also been reported to be increased in elderly or old age individuals [17,18].
A total of 228 patients with a con rmed diagnosis of COVID-19 were included in our study.
[ Supplementary Tables 1 and 2].Based on disease severity, patients were divided into two comparison groups (Cabin/Ward and ICU patients).We found signi cantly strong correlation between D-dimer and CRP levels with age and severity in the combined dataset of ICU and non-ICU COVID-19 patients [ Fig. 1 and Table 1].Increased age and D-dimer values were found to be signi cantly associated with the patients admitted to ICU. Signi cant correlation of CRP levels with Age (Correlation coe cient = 0.250, Pvalue < 0.05), and Severity (Correlation Coe cient = 0.445, P-value < 0.001) was found in non-ICU patients [Supplementary Table 3]. In patients admitted to ICU, we found strong correlation of D-dimer with age (Correlation Coe cient = 0.304, P-value < 0.01) and Severity (Correlation Coe cient = 0.465, P-value < 0.001);correlation of CRP levels with severity (Correlation Coe cient = 0.292 P-value < 0.01) and D-dimer (Correlation coe cient = 0.193, P-value < 0.05) [Supplementary Table 4].Both Higher D-dimer (odds ratio = 1.723, 95% con dence interval: 1.420-2.089, p < 0.001) and CRP-values (odds ratio = 1.011, 95% con dence interval: 1.004-1.019, p = 0.001) were associated with increased severity in patients ( Table 2).  It is interesting to note that D-dimer is commonly high in patients with COVID-19. D-dimer levels indicate disease severity and are reliable prognostic test to rule out the presence of a serious blood clot in patients who were admitted for COVID-19 treatment. While higher levels of C reactive protein (CRP) may be a predictive marker in determining systemic in ammation and can predict which patients with mild COVID-19 will progress to a severe case.
Processes that involve production and breakdown of brin cause an elevation in D-dimer levels [2,19].
Increased D dimer levels are reported to develop acute respiratory distress in COVID-19, with the more chances of micro pulmonary embolism especially in severe forms of COVID-19 [22]. D-dimer levels have been reported to vary among patients with con rmed venous thromboembolism (VTE) depending on clot burden, timing of measurement, and initiation of treatment [23]. D-dimer has also been shown to increase with age, which can cause more false positive tests in older patients [24]. Also, several potential risk factors during hospitalization like, disseminated intra vascular coagulation, infection, dehydration, prolonged immobilization, mechanical ventilation, and central venous catheter use may further increase D-dimer concentrations [23,24].
In conclusion, we con rm the association of two main, in ammatory and biochemical covariates with COVID-19 severity for the rst time in Bangladeshi patients. The study can help in detail understanding of the complications caused and predict the progression of the disease with much more con dence.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.