Observational Study

A Comparative Study of Clinical Manifestation and Severity of CoronavirusDisease Infection in Patients with and without Diabetes Mellitus

Sudhir Bhandari¹, Govind Rankawat², Ajeet Singh³, Vishal Gupta⁴, Amitabh Dube⁵

Senior Professor, Department of General Medicine, S.M.S Medical College and attached group of Hospitals, Jaipur,Rajasthan, India.

2Resident/Fellow student, Department of General Medicine, S.M.S Medical College and attached group of hospitals, Jaipur,Rajasthan, India.

Senior Specialist, Department of General Medicine, S.M.S Medical College and attached group of hospitals, Jaipur,Rajasthan, India.

Associate Professor, Department of General Medicine, S.M.S Medical College and attached group of hospitals, Jaipur,Rajasthan, India.

Professor, Department of Physiology and Ethical Committee, S.M.S Medical College and attached group of hospitals,Jaipur, Rajasthan, India.

Corresponding Author:Govind Rankawat, Resident/Fellow student, Department of General Medicine, S.M.S MedicalCollege and attached group of hospitals, Jaipur, Rajasthan, India.

Email: govindrankawat@gmail.com

Article information

Received date: 3/07/2020; Accepted date: 10/08/2020; Published date: 31/08/2020

---

Abstract

Background:: Diabetes mellitus is one of the well-known chronic illness characterized by inflammatory overdriveand vascular complication. Secondary infection is abundant in diabetes mellitus but the impact of coronavirus disease(COVID-19) on diabetes mellitus not well established. The present study was designed to comparatively assess theclinical presentation, laboratory parameters and radiological findings for COVID-19 manifestation in patients ofdiabetes mellitus (DM) as compared to that of COVID-19 non-diabetic patients. Beside diabetes mellitus, the influenceof other antecedent comorbid conditions on COVID-19 manifestation was also comparatively analysed in both groups.

Aim: To assess whether diabetes mellitus influences the clinical manifestation, progression and severity of COVID-19disease.

Methods:: A total of 1,680 admitted patients were enrolled and categorized into four groups namely, group 1 of alldiabetic patients, group 2 of all non-diabetic patients, group 3 had patients with isolated DM after exclusion of othercomorbidities and group 4 included non-diabetic patients without other comorbidities. The epidemiological data, medicalhistory, symptoms and signs, laboratory findings, digital radiographic of chest, ultrasonography chest and high-resolutioncomputed tomography scans of chest were extracted for evaluation, interpretation and comparison among groups.

Results: In the present study, COVID-19 patients with isolated diabetes mellitus without other comorbidities exhibiteda higher prevalence of symptomatic presentation with an exaggerated inflammatory response and hypercoagulablestate. Serum levels of IL-6, C-reactive protein, ferritin, FDP and D-dimer were significantly raised (p < 0.01) in DMpatients compared to those without DM, suggesting higher susceptibility to an inflammatory storm in COVID-19.Radiological findings available from chest radiograph, USG chest and HRCT chest suggested severe lung involvementin diabetes group as compared to non-diabetics (p<0.05).

Conclusion: The severity of COVID-19 in diabetics could be attributable to the dysfunctional immune system whichfurther aggravated by inflammatory factors, hypercoagulability, organ damage and leads to an increased symptomaticpresentation, laboratory parameters, and radiological pulmonary involvement which precipitate severe and fatalCOVID-19 infection in these patients compared to non-diabetic patients. Hence, diabetic patients with COVID-19required extra preventive care.

Keywords: Clinical manifestation, COVID-19, diabetes mellitus, severity of disease

Introduction

A novel coronavirus, known as the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has been identified as theCOVID-19 pathogen, that triggered clinical manifestation ranging from asymptomatic to severe pneumonia and acute lungfailure.¹Coronavirus belongs to coronaviridae family that possesses an envelope with a positive-sense, extraordinarily largeRNA genome and a nucleocapsid of helical symmetry.² The average incubation period of SARS-CoV-2 is 5-6 days rangingfrom 2 to 14 days. The usual clinical presentation of COVID-19 is fever followed by cough, fatigue,shortness of breath, muscleand joint pains. Less common symptoms include headache, haemoptysis, diarrhoea, decreased sense of smell or disturbancesin taste.^3,4 Moreover, some COVID-19 positive patients do not exhibit noticeable symptoms along the course of the disease.The associated complications of COVID-19 include pneumonia, acute respiratory distress syndrome (ARDS), multi-organfailure, septic shock, disseminated intravascular coagulation and death.5,6 COVID-19 patients have also been reported to havecardiovascular complications viz. myocardial infarction, heart failure, arrhythmias, myocarditis and neurologic manifestationsin some cases that include seizure, stroke, encephalitis and Guillain–Barré syndrome.^5,6

Angiotensin-converting enzyme 2 (ACE2) is the identified surface receptor utilized by SARS coronavirus (SARS-CoV),with direct interaction with the spike glycoprotein (S protein).⁷ A recent study suggested 10-20 folds higher affinity betweenACE2 and the receptor-binding domain (RBD) of SARS-CoV-2 as compared to that of SARS-CoV, indicating a plausible roleof ACE2 as a receptor facilitating SARS-CoV-2 invasion.^8, Diabetes mellitus (DM) is one of the major causes of morbidity andmortality worldwide that is capable of affecting almost every bodily system. (9) Consequently, a deregulated immune systemin diabetics may increase the susceptibility to infectious diseases like COVID-19.¹⁰ Moreover, due to glycosylation in DM,ACE2 expression is reduced, with subsequent inhibition of anti-inflammatory action of ACE2. This exaggerated immuneresponse could explain the occurrence of severe lung injury and ARDS in COVID-19 patients. Hence, diabetics either withor without other comorbidities might be at higher risk of COVID-19 infection and a severe outcome. The present study inthe above context was designed to find out whether DM was associated with the occurrence, progression and prognosis ofCOVID-19 patients or not. Clinical presentation and severity of disease in diabetes mellitus and non-diabetic patients werealso compared.

Method

The present descriptive, observational study was conducted on 1,680 COVID-19 patients admitted to S.M.S. Medical Collegeand attached hospitals, Jaipur, Rajasthan, India as of 30^th June 2020. All the patients with a positive reverse transcriptase-polymerase chain reaction (RT-PCR) for SARS-CoV-2 were categorized into four groups. Data for analysis regardingepidemiology, medical history, symptoms and signs, laboratory findings, digital chest radiograph findings, ultrasonography(USG) chest and high-resolution computed tomography (HRCT) scans of the chest was extracted and tabulated accordingly.

Data Collection

The diagnosis of COVID-19 was based on the World Health Organization (WHO) interim guidance of RT-PCR responseto nasal and pharyngeal swab specimen for SARS-CoV-2.¹

The laboratory investigation included data of hemogram, C-reactive protein (CRP), ferritin, erythrocyte sedimentationrate (ESR) at 1^st hour, fibrin degradation product (FDP), D-dimer and interleukin-6 (IL-6). Radiological findings wereinferred using chest radiograph (CXR), ultrasonography (USG) and high-resolution computed tomography (HRCT) scansof the chest. Digital chest radiographs (CXR) was evaluated using the average visual score (rated 0 to 4 according to visualassessment of involved lung area) and proportional classic for COVID images (basal and peripheral predominant, multiple,bilateral & ground glass haziness).¹¹ USG chest was evaluated using average severity score (sliding scoring scale of severity

on basis of 14 zone severity scores ranging from 0 to 42).¹² HRCT chest was evaluated for CT severity score (CT SeverityScore was assigned out of 25 based on the percentage area involved in each of the 5 lobes) and proportion of patients whohad CT severity score >10/25.¹³ The aforementioned data was compiled, tabulated, compared and interpreted among groupsto establish differences of COVID-19 manifestation in DM.

Statistical Analysis

The descriptive statistics for quantitative data were expressed as mean and standard deviation and qualitative data wasexpressed as proportions. The parameters were compared among different groups using chi-square test and z-score forsignificant differences.^{14, 15} The level of significance was assigned at a p-value less than 0.05.¹⁶

Results

In the present study, a total of 1,680 patients of confirmed RT-PCR for COVID-19 were assessed and required data collected,evaluated, interpreted and correlated in all four groups (Figure 1).

Clinical Presentation, Laboratory Parameters and Radiological Findings Among Group 1 and 2

The overall mean age of SARS-CoV-2 infected patients was 55.98 year (55.98±11.64 year). Mean age in the diabetes group(56.73±9.57 year) was slightly higher (p=0.0512) as compared to that of the non-diabetes group (55.48±12.17 year) (Table1). The gender distribution in terms of sex ratio (M: F) among both groups did not differ significantly, where sex ratio (F: M)was 0.49 in the diabetes group, 0.47 in the non-diabetes group and 0.48 in the sample population.

Overall, 468 patients (27.86%) were symptomatic and percent symptomatic diabetic patients (36.36%) was significantly morethan that compared to non-diabetes group (24.84%) with p<0.001. The percent clinical presentation of COVID-19 positivepatients was fever (18.10%), cough (16.43%), sore throat (15.48%), shortness of breath (14.76%), headache (8.57%), chestpain (2.14%) and other non-respiratory symptoms like pain abdomen, vomiting, diarrhoea, altered sensorium made up 2.38%.Clinical presentation with fever (21.36% in Group 1 and 16.94% in Group 2), cough (19.77% in Group 1 and 15.24% inGroup 2) and shortness of breath (27.27% in Group 1 and 10.32% in Group 2) were significantly high (p<0.05) in diabetesgroup as compared to that observed in the non-diabetes group.

The observed underlying comorbidities were hypertension (15.00%), pulmonary diseases (12.62%), cardiovasculardiseases (8.10%), malignancy (3.57%), chronic kidney disease (4.52%), cerebrovascular disease (1.90%), chronic liverdiseases, (1.43%) and other chronic diseases (1.90%) like hypothyroidism, immunodeficiency disease, nutritional deficiencydiseases, etc. Hypertension (27.27% in Group 1 and 10.65% in Group 2, p<0.001), cardiovascular diseases (14.55% inGroup 1 and 5.81% in Group 2, p<0.001), chronic kidney disease (8.18% in Group 1 and 3.23% in Group 2, p<0.001) andmalignancy (5.45% in Group 1 and 2.90% in Group 2, p=0.0131) were found significantly high in diabetes group as comparedto that observed in the non-diabetes group.

The following parameters were observed in the present study: mean haemoglobin 12.12 g/dL (95% CI: 12.12±0.30, SD=1.61),total leukocyte count 7.24 x 10⁹/L (95% CI: 7.24±1.32, SD=3.62), platelet count 2.20 Lac/μl (95% CI: 2.20±0.20, SD=0.87),neutrophil/lymphocyte (N/L) ratio 2.96 (95% CI: 2.96±0.54, SD=1.24), C-reactive protein 5.59 mg/L (95% CI: 5.59±1.05,SD=3.14), ferritin 275.54 ng/mL (95% CI: 275.54±42.44, SD=317.26), ESR 41.08 mm at 1st hour (95% CI: 41.08±1.82,SD=19.62), fibrin degradation product 20.08 μg/L (95% CI: 20.08±4.09, SD=27.52), D-dimer 3.45 μg/L (95% CI: 3.45±0.74,SD=5.80) and interleukin-6, 52.04 pg/mL (95% CI: 52.04±29.42, SD=98.84).The mean value of different laboratory parameters in Group 1 and Group 2 were observed as follows: haemoglobin11.98 vs. 12.18 (p=0.0239), N/L ratio 3.16 vs. 2.84 (p<0.001), ferritin level 318.74 ng/mL vs. 262.60 ng/mL (p=0.0022),D-dimer 4.10 μg/L vs. 3.01 μg/L (p<0.001), IL-6 level 63.90 pm/mL v/s 48.54 pg/mL (p =0.0061). The aforementionedparameters were significantly high in diabetes group as compared to that observed in the non-diabetes group that suggeststhat COVID-19 patients with diabetes could presumably be prone to develop excessive uncontrolled inflammation responsesand hypercoagulable states that in effect could contribute to a guarded prognosis for COVID-19.

Digital chest radiograph (CXR) of all patients, USG chest of 248 patients (14.76%) and HRCT chest of 512 patients (30.48%)were available for analysis. In the study population, CXR represented a classic picture for COVID images in 244 patients(14.52%) with an average visual score of 0.69 (95% CI: 0.69±0.45, SD=0.83) out of 4. USG severity score was 14.96 (95%CI: 14.96±4.62, SD=8.92) out of 14 and an average CT severity score of 6.04 (95% CI: 6.04±1.33, SD=5.75) out of 25. TheCT severity scores were >10/25 in 100 patients (19.53%). Results of radiological findings in diabetes and non-diabetes groupwere as follows: CXR average visual score was 0.88 and 0.61 (p<0.001) while classic for COVID images in 21.82% and11.94% patients (p<0.001) among Group 1 and 2 respectively. The CT severity score was significantly high in the diabetesgroup (7.25) as compared to the non-diabetes group (5.67) (p=0.0054) (Figure 2 and 3). The CT severity score was >10/25in 30.56% and 15.22% patients in respective groups (p<0.001)

To avoid the influence of comorbidities, other than diabetes mellitus on COVID-19 manifestations, in this part we excludedpatients with other comorbidities. In this part study population (1260 patients) were classified into two groups: patients withisolated diabetes without other comorbidities belongs to Group 3 while patients without diabetes and other comorbiditiesbelong to Group 4.

The mean age of SARS-CoV-2 infected patients without other comorbidities was 53.19±9.1 year. The mean age of thediabetes group (54.08 years) was not significantly higher as compared to that observed in the non-diabetes group (52.95years) (Table 3). The gender distribution did not differ significantly among the groups, where the sex ratio (F: M) was 0.68in the diabetes group and 0.49 in the non-diabetes group without co-morbidities.

Among patients devoid of pre-existing comorbidities, 352 patients (27.94%) were symptomatic for COVID-19, thoughonly 22.86% non-diabetic patient were symptomatic as compared to 45.71% symptomatic diabetic patients, a statistically significant difference (p<0.001). COVID-19 infected patients without other comorbidities predominantly presented withfever (21.90%), cough (24.13%), sore throat (18.10%), shortness of breath (14.60%), headache (10.16%), chest pain (4.13%)and other symptoms (3.81%) like pain abdomen, vomiting, diarrhoea, altered sensorium. Clinical presentation with fever(28.57% in Group 3 vs. 20.00% in Group 4), cough (30.36% in Group 3 vs. 22.35% in Group 4), shortness of breath (27.14%in Group 3 vs. 11.02% in Group 4) and headache (13.57% in Group 3 vs. 9.18% in Group 4) was observed significantly higherin the diabetes group compared to the non-diabetes group.

The mean value of laboratory parameters observed was as follows: haemoglobin 12.80±1.64 g/dL, total leukocyte count(TLC) 6.74x10⁹±3.12/L, platelet count 2.28±0.85 Lac/μL, neutrophil/lymphocyte (N/L) ratio 2.85±1.18, C-reactive protein5.55±4.06 mg/L, ferritin 302.27±294.58 ng/mL, ESR 37.08±19.88 mm/h, fibrin degradation product 17.20±23.19 μg/L,D-dimer 3.14±3.98 μg/L and interleukin-6 (IL-6) 39.45±67.75 pg/mL. The mean value of different laboratory parametersin Group 3 and Group 4 were observed as follows: Haemoglobin 12.56g/dL vs. 12.98 g/dL (p=0.0005), TLC 7.52x10⁹/L vs.6.48x10⁹/L (p<0.001), N/L ratio 3.1 vs. 2.78 (p<0.001), CRP 6.24 mg/L vs. 5.26 mg/L with (p=0.0004), ferritin level 364.58ng/mL vs. 284.46 ng/mL (p<0.001), FDP 20.84 μg/L vs. 16.18 μg/L (p=0.0026), D-dimer 3.98 μg/L vs. 2.76 μg/L (p<0.001),IL-6 level 58.60 pm/mL vs. 34.44 pg/mL (p<0.001). The aforementioned parameters were significantly higher in diabetesgroup as compared to non-diabetes group without comorbidities.

Table 4. Comparison of laboratory and radiological parameters between diabetic and non-diabetic COVID-19patients after exclusion of other comorbidities. Abbreviations: COVID-19- Coronavirus disease 2019; ESR- Erythro-cyte sedimentation rate; FDP- Fibrinogen; IL-6- Interleukin-6; CI- Confidence interval; USG- Ultrasonography, HRCT-High-resolution computed tomography; #The p-values indicate differences between diabetes and non-diabetes patients.A p<0.05 was considered statistically significant; z-value is a standardized score which measures the distance betweenthe mean and an observation.

Digital chest radiograph (CXR) of all patients, USG chest of 200 patients (15.87%) and HRCT chest of 424 patients (33.65%)was available for analysis. CXR depicted a classic for COVID images in 208 patients (16.51%) with an average visual score of0.63±0.81 out of 4. USG severity score was 14.88±9.18 out of 14 and average CT severity score was observed to be 5.81±5.44out of 25 with CT severity score >10/25 in 80 patients (18.87%) (Figure 4 and 5). Radiological findings observed amongdiabetes group and non-diabetes group without comorbidities were as follow: The CXR average visual score was 0.74 and 0.58(p=0.0035) while classic for COVID images was observed in 28.57% and 13.06% patients (p<0.001) in Group 3 and Group4 respectively. USG chest severity score was significantly higher in diabetes group (17.86) as compared to the non-diabetesgroup (13.48) (p=0.0041) (Figure 6 and 7). CT severity score was 7.24 in the diabetes group and 5.32 in the non-diabetesgroup (p=0.0020) with a CT severity score >10/25 in 29.41% and 15.53% patients in the respective groups (p=0.0018).

Diabetes mellitus predisposes patients for various infectious disease including COVID-19,¹⁷ but how diabetes mellitusinfluences COVID-19, needs further exploration. A large proportion of the diabetic population can be predisposed forCOVID-19 infection as the prevalence of DM in India is 7.3%.¹⁸ Type 2 diabetes mellitus leads to raised inflammatory factorsand chemokines as a consequence of the dysfunctional immune system.^19,20 As a consequence of glycosylation, expressionof ACE2 receptors reduced in patients of diabetes mellitus which enhances inflammatory storm and invasion of virus leadsto severe lung injury and ARDS.¹⁸ In COVID-19 patients, immunostaining of islet tissue for ACE2 has been enhanced,suggesting a plausible role of coronavirus in islet destruction.²¹ The present study indicated a more severe clinical picturein the diabetes group as compared to the non-diabetics, especially when other underlying comorbidities were excluded.Clinical presentation among diabetes and non-diabetes group differed significantly in terms of symptoms of fever, cough andshortness of breath. Such a presentation might result due to early and extensive lung involvement of COVID-19 infectionin diabetes patients. Neutrophil to lymphocyte ratio in patients of diabetes was also significantly high as compared to thatof the non-diabetes group, that could be attributable to neutrophilia or a relative lymphocytopenia as a consequence ofSARS-CoV-2 infection. Mean values of total leukocyte count were observed to be significantly high in patients with isolateddiabetes as compared to that observed in non-diabetes patients without other comorbidities. Furthermore, higher levels ofserum inflammation-related biomarkers such as IL-6, serum ferritin, ESR and CRP were also observed in diabetes group ascompared to that observed in patients without diabetes. IL-6 has a prolonged expression time as compared to others cytokines(TNF and IL-1) and can be utilized as a predictor of disease severity and prognosis.²² Huang et al. confirmed the findings ofelevated IL-6 levels beside a significantly low lymphocyte count, in patients with SARS-CoV-2 infection, especially thosepresenting with severe pneumonia.3Excessively raised ferritin level is an indicator of activation of the monocyte-macrophagesystem, that contributes significantly to the inflammatory storm associated with COVID-19.²¹ In the present study raisedferritin levels were observed in diabetics, suggesting a higher susceptibility of such patients for an inflammatory storm,responsible for the rapid deterioration of COVID-19. Inflammatory storm in COVID-19 is associated with a significant risein D-dimer levels. Inflammation associated hypoxia might induce thrombin activation with a consequent unfolding of theexogenous coagulation pathway.23 In this study, FDP and D-dimer levels were observed to be significantly high in patientswith diabetes as compared to that observed in non-diabetes patients that is suggestive of a hypercoagulable state inclusive ofdisseminated intravascular coagulation in such patients.

Radiological imaging of chest provides an important clue regarding lung involvement in COVID-19 that is a prognosticindicator of disease severity. Digital chest X-ray imaging suggested a higher proportion of sample population exhibiting lung involvement in diabetics as compared to that observed in non-diabetics. A similar picture was portrayed by CT severityscore that was high in diabetic patients as compared to the findings of non-diabetics. However, USG chest severity score wassignificantly increased in the isolated diabetic group as compared to the non-diabetes group without other comorbidities. Theaforementioned findings suggest a severe form of pneumonia in diabetic patients as compared to the non-diabetic patients.Moreover, COVID-19 manifestation and its severity are adversely affected by the associated comorbid disease. In the presentstudy, clinical presentation and laboratory parameters indicated a significant difference among isolated diabetes and the non-diabetes group without comorbidities, compared to groups, with comorbidities.

This study included patients of a single centre, so geographic variation could not be appreciated. This is a retrospectiveobservational study, so definitive postulates could not be formed.

Our study concluded that diabetes mellitus could predispose an individual to a severe clinical and laboratory presentation forCOVID-19. The severity of COVID-19 infection in diabetics might be due to exaggerated immune response like cytokinestorm which leading to hypercoagulability and organ damage. Diabetic patients more prone to capture the severe form ofCOVID-19 disease, especially in the form of symptomatic presentation, increased inflammatory markers and radiologicalpulmonary involvement compared to non-diabetic patients. Diabetes mellitus highly prevalent in India so preventive controlmeasure for COVID-19 in diabetic patients must be initiated.

We would like to acknowledge Dr. Abhishek Agrawal, Dr. C. L. Nawal, Dr. S. Banerjee, Dr. Prakash Keswani, Dr. SunilMahavar, Dr. R. S. Chejara, Dr. Vidyadhar Singh, Dr. Shalaj Jain, Dr. Shivankan Kakkar, Dr. Bhupendra Patel, Dr. MeenuBagarhatta and the team of the Department of General Medicine and Department of Radiodiagnosis, S.M.S Medical Collegeand attached group of hospitals, Jaipur, Rajasthan, India for their valuable support and for providing radiological informationof COVID-19 patients.

Dr. S. Bhandari, Dr. G. Rankawat and Dr. A. Singh formulated the research questions, designed the study, developed thepreliminary search strategy, and drafted the manuscript; Dr. G. Rankawat and Dr. A. Singh collected and analysed data forstudy. Dr. G. Rankawat wrote the manuscript. Dr. A. Dube and Dr. V. Gupta conducted the quality assessment. All authorscritically reviewed the manuscript for relevant intellectual content. All authors have read and approved the final version ofthe manuscript.

All authors report no potential conflicts.

The authors did not receive any financial support for the research, authorship and/or publication of this article.

This study was approved by the ethical and research committee of S.M.S Medical College and Hospital, Jaipur, India (ethicalapproval number: MC/EC/2020/417).

All the figures/images are original and have never be used for publication of article(s)/chapter(s) in any other journal or book.

Available from the corresponding author upon reasonable request.