Secondary Logo

Journal Logo

Original Article

Establishment and Evaluation of a Prediction Model of BLR for Severity in Coronavirus Disease 2019

He, Zebao1,2; Rui, Fajuan3; Yang, Hongli4; Ge, Zhengming1,2; Huang, Rui5,6; Ying, Lingjun1,2; Zhao, Haihong1,2; Wu, Chao5,6,∗; Li, Jie4,5,6,∗

Editor(s): Wang, Haijuan

Author Information
Infectious Diseases & Immunity: April 2022 - Volume 2 - Issue 2 - p 100-108
doi: 10.1097/ID9.0000000000000048
  • Open

Abstract

Introduction

In December 2019, a series of pneumonia with clinical presentations closely similar to viral pneumonia, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), was reported in Wuhan, China.[1,2] The disease was confirmed to be caused by the novel coronavirus and designated coronavirus disease 2019 (COVID-19), which has aroused great concern worldwide. World Health Organization (WHO) has also announced COVID-19 is a public health emergency with international concern.[3–5] As of September 2021, the cumulative number of cases reported worldwide has exceeded 224 million, with more than 4.6 million cumulative deaths.[6]

Recent researches indicated that some cases of COVID-19 are similar in severity to SARS-CoV,[1,7–11] and the proportion of severe/critical case ratio of COVID-19 is about 7% to 10%.[12] The Chinese Center for Disease Control and Prevention (CDC) reported that the case-fatality rate in the largest series of 44,672 confirmed COVID-19 cases in mainland China was 2.3%, and 1023 cases of deaths all occurred in critical cases.[13] Given the numbers of both infected patients and deaths are still growing rapidly, there are no clinically approved effective drugs. Analysis of clinical characteristics of severe cases might help identifying the severity of the disease. Therefore, in this study, we investigate the clinical factors associated with the severe COVID-19 patients, establish risk prediction model, and we hope this may help physicians identifying the patients with poor prognosis early.

Methods

Ethical approval

This study was approved by ethics committee of Enze Hospital, Taizhou Enze Medical Center (Group) and Nanjing Drum Tower Hospital in Jiangsu Province (Protocol number: K20211206 and 2020-023-01). Written informed consent was waived by the Ethics Commission due to public health outbreak investigation.

Study participants and recruitment

For this observational, retrospective study, we recruited 402 patients with COVID-19 between January 24 and March 12, 2020, at Enze Hospital, Taizhou Enze Medical Center (Group) and Nanjing Drum Tower Hospital in Jiangsu Province. Among them, 98 patients from Taizhou were in the training set and 304 patients from Nanjing were in the validation set. According to WHO interim guidance, people who test positive result to real-time reverse transcriptase polymerase chain reaction assay of nasal and pharyngeal swab specimens are diagnosed with COVID-19. Only the laboratory-confirmed cases were recruited the analysis. Laboratory confirmation of SARS-CoV-2 infection was conducted by the Chinese CDC.

Data collection

Epidemiological, laboratory, clinical, and imaging data were recorded with data collection forms from the electronic medical record. Recorded information included demographic characteristics, underlying comorbidities, medical and exposure history, symptoms, laboratory findings (liver and renal function, complete blood count, electrolytes, coagulation test), and chest computed tomographic scans. The duration from the contact of the transmission source to the onset of symptoms was defined as the incubation period. The date of symptom onset was defined as the day of disease onset. The durations from disease onset to admission were recorded. The definition of severe COVID-19 was based on WHO's interim guidance criteria for severe pneumonia. The patients were divided into severe and non-severe groups according to this definition.

Statistical analysis

Statistical analyses of the data were performed using MedCalc®19.8 (MedCalc Software BVBA, Ostend, Belgium) and SPSS version 23.0 (IBM Corp., Armonk, NY, USA). If they are normally distributed, the continuous measurements are represented as mean (standard deviation), or if they are not, as the median (Quartile 1, Quartile 3), and categorical variables are represented as count (%). To assess differences between severe and non-severe COVID-19 patients, we use two-sample t test or Wilcoxon rank-sum test based on parametric or non-parametric data for continuous variables and Chi-square or Fisher's exact test for categorical variables. A multivariate logistic regression analysis was implemented to determine odds ratios (ORs) and 95% confidence intervals (CIs) between individual factors influencing disease severity. A two-sided a of less than 0.05 was considered statistically significant.

Results

Characteristics of COVID-19 patients

The average age of 402 COVID-19 patients was 45 (34–55) years (221 males and 181 females). Demographics, clinical features, laboratory parameters, and radiologic findings were summarized in Table 1. There were 54 cases (13.43%) in severe group and 348 cases (86.57%) in non-severe group; 145 cases of normal weight (36%), 105 cases of overweight (26%), and 152 cases of obesity (38%). The most frequent symptoms were fever (57%), followed by cough (52%), expectoration (38%), diarrhea (14%), shortness of breath (13%), myalgia (11%), and sore throat (9%). There were no significant differences in the distribution of hemoglobin, lymphocyte count, aspartate aminotransferase (AST), γ-glutamyl transpeptadase (GGT), albumin, creatine kinase, serum chloride, and activated partial thromboplastin time (APTT) between the training and validation groups (all P > 0.05).

Table 1 - Personal and clinical characteristics of COVID-19 patients
All patients Patients in training Patients in validation
Variables (n = 402) group (n = 98) group (n = 304) P
Sex [male, n (%)] 221 (55) 54 (55) 167 (55) 0.977
Age (years) 45 (34, 55) 47 ± 15 44 (32, 55) 0.052
Hypertension [n (%)] 62 (15) 14 (14) 48 (16) 0.720
Diabetes [n (%)] 35 (9) 10 (10) 25 (8) 0.545
BMI (kg/m2, mean±SD) 24.04 ± 3.22 23.59 ± 2.65 24.19 + 3.38 0.069
 Normal [<23, n (%)] 145 (36) 40 (41) 105 (35) 0.026
 Overweight [23.0–24.9, n(%)] 105 (26) 32 (33) 73 (24)
 Obese [≥25, n (%)] 152 (38) 26 (26) 126 (41)
Symptoms [n (%)]
 Fever 231 (57) 72 (73) 159 (52) <0.001
 Cougha 200 (52) 71 (72) 129 (45) <0.001
 Expectorationb 153 (38) 42 (43) 111 (37) 0.290
 Shortness of breathc 50 (13) 13 (13) 37 (12) 0.792
 Myalgiac 43 (11) 19 (19) 24 (8) 0.001
 Sore throatd 37 (9) 18 (18) 19 (6) <0.001
 Diarrheadd 56 (14) 18 (18) 38 (13) 0.148
Laboratory parameters
 Hemoglobin (g/L, mean ± SD] 139.49 ± 16.76 140.56 ± 16.04 139.15 ± 17.00 0.469
 Leukocyte count [×109/L, M(Q1, Q3)] 4.98 (3.89, 6.22) 5.20 (4.20, 6.80) 4.90 (3.85, 6.15) 0.029
 Platelet count (×109/L) 187.00 (143.75, 233.25) 209.68 ± 62.17 176.50 (141.00, 227.00) 0.002
 Neutrophil count [×109/L, M(Q1, Q3)] 2.99 (2.27, 4.20) 3.40 (2.38, 4.85) 2.94 (2.19, 3.94) 0.037
 Lymphocyte count [×109/L, M(Q1, Q3)] 1.20 (0.88, 1.63) 1.20 (0.80, 1.60) 1.21 (0.89, 1.64) 0.387
 CRP [μg/mL, M (Q1, Q3)] 6.70 (1.47, 20.40) 8.11 (2.38, 25.34) 5.81 (1.00, 18.85) 0.017
 ALT [U/L, M (Q1, Q3)] 25.00 (18.00, 36.50) 21.00 (14.75, 34.25) 26.00 (19.00, 38.00) 0.005
 AST [U/L, M (Q1, Q3)] 25.00 (20.00, 31.50) 24.50 (19.75, 31.00) 25.00 (20.00, 32.00) 0.960
 ALP [U/L, M (Q1, Q3)] 64.00 (53.00, 79.00) 71.00 (61.00, 82.00) 62.00 (51.00, 76.00) <0.001
 GGT [U/L, M (Q1, Q3)] 25.30 (16.00, 40.50) 25.50 (17.00, 43.50) 25.30 (15.85, 40.00) 0.612
 TBIL [μmol/L, M (Q1, Q3)] 11.10 (7.40, 15.70) 12.75 (8.63, 17.23) 10.40 (7.40, 15.10) 0.007
 Albumin (g/L) 40.00 (37.20, 43.00) 39.56 ± 3.83 40.10 (37.10, 43.30) 0.204
 CK [U/L, M (Q1, Q3)] 67.00 (44.00, 97.75) 70.50 (46.00, 94.75) 63.50 (41.00, 100.25) 0.144
 BUN [mmol/L, M (Q1, Q3)] 4.19 (3.47, 5.22) 4.55 (3.71, 5.73) 4.10 (3.40, 4.96) 0.003
 K (mmol/L, mean±SD) 3.96 ± 0.47 3.82 ± 0.43 4.01 ± 0.48 <0.001
 NA (mmol/L) 138.49 ± 3.62 137.67 ± 2.92 139.10 (136.30–141.25) 0.003
 Cl (mmol/L, mean±SD) 102.50 ± 3.66 102.02 ± 3.21 102.66 ± 3.80 0.103
 PT [S, M (Q1, Q3)] 12.60 (11.80, 13.20) 11.80 (11.15, 12.40) 12.80 (12.10, 13.40) <0.001
 APTT [S, M (Q1, Q3)] 30.20 (26.80, 35.40) 29.60 (27.75, 32.25) 30.55 (26.13, 36.80) 0.240
 INR [M (Q1, Q3)] 1.03 (0.99, 1.10) 1.03 (0.97, 1.08) 1.04 (1.00, 1.11) 0.031
Imaging findings [n (%)]
 Pneumoniad 371 (93) 94 (96) 277 (91) 0.141
 Bilateral pulmonary inflammationd 312 (78) 75 (76) 237 (78) 0.727
P values are calculated by two-sample t test or Wilcoxon rank-sum test based on parametric or non-parametric data for continuous variables and Chi-square or Fisher's exact test for categorical variables.
Data are shown as mean±SD (standard deviation) or M (Q1, Q3) [Median (Quartile 1, Quartile 3)] depending on whether they are normally distributed.
aData available for 386 patients in all patients and 288 patients in validation group.
bData available for 399 patients in all patients and 301 patients in validation group.
cData available for 400 patients in all patients and 302 patients in validation group.
dData available for 401 patients in all patients and 303 patients in validation group.BMI: Body mass index; CRP: C-reactive protein; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; ALP: Alkaline phosphatase; GGT: γ-glutamyl transpeptadase; TBIL: Total bilirubin; CK: Creatine kinase; BUN: Blood urea nitrogen; K: Potassium; NA: Natrium; Cl: Chlorine; PT: Prothrombin time; APTT: Activated partial thromboplastin time; INR: International normalized ratio.

As shown in Table 2, there was no significant difference in body mass index (BMI) classification between the severe and non-severe groups in the training cohort (P = 0.082), however obese (71%) patients were dominant in severe COVID-19 in the validation set (P = 0.002). Severe group patients were more likely to have symptoms including fever, expectoration, and shortness of breath (all P < 0.05). Compared with non-severe COVID-19 patients, severe patients had higher levels of AST, GGT, creatine kinase, neutrophil count, C-reactive protein (CRP), and had lower lymphocyte count and albumin level (all P < 0.05). Furthermore, severe patients were more likely to have bilateral pulmonary inflammation; 96% of the training cohort patients had bilateral pulmonary inflammation compared with 100% in the validation cohort. Further details regarding the comparisons between baseline characteristics of training and validation cohort were shown in Table 2, respectively.

Table 2 - Personal and clinical characteristics of COVID-19 patients from training and validation set
Training set Validation set


Variables Non-severe Severe Non-severe Severe
group (n = 72) group (n = 26) P group (n = 276) group (n = 28) P
Sex [Male, n (%)] 37 (51) 17 (65) 0.219 147 (53) 20 (71) 0.066
Age (years) 45 ± 15 53 ± 13 0.012 43 (32, 54) 50 ± 12 0.024
Hypertension [n (%)] 10 (14) 4 (15) 1.000 44 (16) 4 (14) 1.000
Diabetes [n (%)] 7 (10) 3 (12) 1.000 17 (6) 8 (29) <0.001
BMI (kg/m2, mean±SD) 23.11 ± 2.54 24.90 ± 2.54 0.003 23.94 ± 3.32 26.63 ± 2.98 <0.001
 Normal [<23, n (%)] 34 (47) 6 (23) 0.082 102 (37) 3 (11) 0.002
 Overweight [23.0–24.9, n(%)] 22 (51) 10 (38) 68 (25) 5 (18)
 Obese [≥25, n (%)] 16 (22) 10 (38) 106 (38) 20 (71)
Symptoms [n (%)]
 Fever 47 (65) 25 (96) 0.002 137 (50) 22 (79) 0.003
 Cougha 46 (64) 25 (96) 0.003 117 (45) 12 (40) 0.970
 Expectorationb 26 (36) 16 (62) 0.032 91 (33) 20 (74) <0.001
 Shortness of breathc 3 (4) 10 (38) <0.001 22 (8) 15 (54) <0.001
 Myalgiac 9 (13) 10 (38) 0.004 20 (7) 4 (14) 0.350
 Sore throatd 12 (17) 6 (23) 0.669 14 (5) 5 (18) 0.025
 Diarrhea 12 (17) 6 (23) 0.828 35 (13) 3 (11) 1.000
Laboratory parameters
 Hemoglobin [g/L, mean±SD] 141.10 ± 15.53 139.08 ± 17.62 0.585 139.28 ± 17.20 137.82 ± 15.03 0.665
 Leukocyte count [×109/L, M(Q1, Q3)] 5 .00 (4.00, 6.38) 6.05 (4.48, 7.65) 0.070 4.82 (3.84, 6.08) 5.51 (4.03, 6.92) 0.047
 Platelet count (×109/L) 218.93 ± 60.17 184.08 ± 61.54 0.014 176.50 (141.00, 230.00) 174.96 ± 62.94 0.367
 Neutrophil count [×109/L, M(Q1, Q3)] 2.85 (2.30, 4.08) 4.75 (3.40, 7.03) 0.001 2.87 (2.12, 3.82) 4.37 (2.92, 5.75) <0.001
 Lymphocyte count (×109/L) 1.40 (1.00, 1.90) 0.71 ± 0.28 <0.001 1.29 (0.94,1.68) 0.86 ± 0.34 <0.001
 CRP [μg/mL, M (Q1, Q3)] 4.85 (1.70, 19.31) 20.20 (7.67, 50.86) <0.001 4.59 (0.70, 16.84) 25.85 (8.50, 50.07) <0.001
 ALT [U/L, M (Q1, Q3)] 20.00 (14.00, 30.75) 24.00 (17.75, 53.50) 0.081 26.00 (18.90, 37.00) 32.00 (23.00, 53.25) 0.013
 AST [U/L, M (Q1, Q3)] 23.00 (19.00, 29.75) 29.50 (24.75, 57.25) 0.001 24.00 (20.00,31.00) 30.50 (26.00, 49.75) <0.001
 ALP (U/L) 71.50 (60.75, 87.00) 69.50 (59.75, 77.75) 0.243 63.00 (51.00, 76.00) 60.74 ± 15.68 0.345
 GGT [U/L, M (Q1, Q3)] 24.00 (15.75, 37.25) 32.50 (20.75, 58.00) 0.036 24.00 (15.00, 38.00) 46.00 (28.00, 75.10) <0.001
 TBIL (μmol/L) 13.08 ± 6.30 13.90 (11.28, 20.08) 0.110 10.50 (7.40, 15.10) 10.15 (7.25, 16.03) 0.886
 Albumin (g/L) 40.37 ± 3.44 37.36 ± 4.04 <0.001 40.40 (37.80, 43.60) 35.47 ± 5.02 <0.001
 CK [U/L, M (Q1, Q3)] 67.00 (43.25, 83.00) 119.00 (72.75, 186.75) <0.001 60.00 (40.00, 93.25) 79.50 (63.75, 204.25) 0.006
 BUN (mmol/L) 4.55 (3.59, 5.51) 4.55 (3.91, 6.94) 0.338 4.10 (3.40, 4.90) 4.83 ± 1.88 0.049
 K (mmol/L) 3.81 ± 0.39 3.83 ± 0.54 0.810 4.01 ± 0.47 3.76 (3.59, 4.26) 0.494
 NA (mmol/L) 137.99 ± 2.75 136.79 ± 3.22 0.072 139.20 (136.38, 141.40) 137.63 ± 3.79 0.109
 Cl (mmol/L, mean ± SD) 102.19 ± 3.21 101.54 ± 3.21 0.378 102.78 ± 3.74 101.50 ± 4.24 0.094
 PT (S) 11.80 ± 0.87 11.80 (11.20, 12.95) 0.388 12.90 (12.20, 13.50) 12.16 ± 0.98 0.002
 APTT (S) 29.56 ± 2.70 30.70 (28.15, 33.35) 0.075 31.15 (26.50, 37.43) 27.26 ± 4.87 0.001
 INR 1.03 (0.96, 1.07) 1.04 ± 0.08 0.722 1.04 (1.00, 1.11) 1.03 ± 0.08 0.252
Imaging findings [n (%)]
 Pneumoniac 68 (94) 26 (100) 0.516 249 (91) 28 (100) 0.178
 bilateral pulmonary inflammationc 50 (71) 25 (96) 0.006 209 (76) 28 (100) 0.003
P values are calculated by two-sample t test or Wilcoxon rank-sum test based on parametric or non-parametric data for continuous variables and Chi-square or Fisher's exact test for categorical variables.
Data are shown as mean±SD (standard deviation) or M (Q1, Q3) [Median (Quartile 1, Quartile 3)] depending on whether they are normally distributed.
aData available for 261 patients in non-severe group and 27 patients in severe group of validation cohort.
bData available for 274 patients in non-severe group and 27 patients in severe group of validation cohort.
cData available for 274 patients in non-severe group of validation cohort.
dData available for 275 patients in non-severe group of validation cohort.BMI: Body mass index; CRP: C-reactive protein; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; ALP: Alkaline phosphatase; GGT: γ-glutamyl transpeptadase; TBIL: Total bilirubin; CK: Creatine kinase; BUN: Blood urea nitrogen; K: Potassium; NA: Natrium; Cl: Chlorine; PT: Prothrombin time; APTT: Activated partial thromboplastin time; INR: International normalized ratio.

Factors associated with severe COVID-19 patients

Univariate analysis on clinical characteristics was shown in Table 3. Lymphocyte count, AST, albumin, creatine kinase, CRP, BMI, alanine aminotransferase, leukocyte count, age, platelet count, APTT, and GGT were associated with the severe COVID-19 patients (all P < 0.05). However, on multivariable analysis, only lymphocyte count (0.004, 95% CI 0–0.055) and BMI (1.519, 95% CI 1.126–2.049) were associated with the severity of patients (all P < 0.05).

Table 3 - Univariate and multivariate logistic regression models for severe COVID-19
Univariate analysis Multivariate analysis


Variables OR (95% CI) P OR (95% CI) P
Lymphocyte count 0.006 (0.001–0.061) <0.001 0.004 (0–0.055) <0.001
AST 1.076 (1.031–1.123) 0.001
Albumin 0.800 (0.696–0.918) 0.002
CK 1.009 (1.003–1.016) 0.002
CRP 1.031 (1.011–1.052) 0.003
BMI 1.308 (1.085–1.577) 0.005 1.519 (1.126–2.049) 0.006
ALT 1.032 (1.008–1.057) 0.008
Leukocyte count 1.251 (1.046–1.497) 0.014
Age 1.043 (1.008–1.080) 0.015
Platelet count 0.990 (0.982–0.998) 0.017
APTT 1.159 (1.007–1.333) 0.040
GGT 1.020 (1.001–1.040) 0.044
Normal (<23) Referant
Overweight (23–24.9) 2.576 (0.819–8.098) 0.105
Obese (≥25) 3.542 (1.095–11.453) 0.035
AST: Aspartate aminotransferase; CK: Creatine kinase; CRP: C-reactive protein; BMI: Body mass index; ALT: Alanine aminotransferase; APTT: Activated partial thromboplastin time; GGT: γ-glutamyl transpeptadase; –: Not applicable.

Using logistic regression analysis, we get a prognostic model, the BMI and lymphocyte count (BL) prognosis model, logit (BL) =–5.552–5.473 × L + 0.418 × BMI. The AUC in the training and validation cohorts were 0.928 and 0.848, respectively [Figure 1A and B]. The model was simplified to get a new model (BMI and lymphocyte count ratio, BLR), and the AUC in the training and validation groups were 0.926 and 0.828 [Figure 1C and D].

F1
Figure 1:
Area under the receiver operating characteristic curve (AUC) analysis for severe COVID-19. (A) BL model in training set; (B) BL model in validation set; (C) BLR model in training set; (D) BLR model in validation set. BL: Body mass index and lymphocyte count; BLR: Body mass index and lymphocyte count ratio.

Comparison of BLR and BL prognosis models

Comparison between BLR prognosis model and BL prognosis model was shown in Figure 2. BL prognosis model had a predictive accuracy for COVID-19 (AUC 0.928) with 88.46% sensitivity and 90.28% specificity; and the positive and negative predictive values of 76.7% and 95.6%, respectively. The AUC of BLR prognosis model was 0.926. At the cut-off value of 25.16, sensitivity, specificity, positive predictive values, and negative predictive values were 88.46%, 84.72%, 67.6%, and 95.3%, respectively. Besides, there was no significant difference in the effectiveness of the BLR and BL prediction models (P = 0.8523).

F2
Figure 2:
Evaluation of the predictive value of the scoring systems for severe COVID-19. BL: Body mass index and lymphocyte count; BLR: Body mass index and lymphocyte count ratio.

Subgroup analysis of BMI, age, and albumin

In different subgroups, BLR prediction model achieved a good efficacy. The AUC of BLR prediction model in obese patients was the best (AUC 0.950), followed by normal weight patients (AUC 0.919), and overweight patients (AUC 0.864) [Figure 3]. The AUC in patients older than 50years was 0.919, which was higher than patients aged younger than 50years (AUC 0.915); and predictive accuracy washigherinpatients withalbumin ≥40 g/L(AUC 0.944) compared with albumin <40 g/L (AUC 0.913) [Figure 4].

F3
Figure 3:
Subgroup analysis: area under the receiver operating characteristic curve (AUC) of BLR model under different BMI classifications. (A) Obese (BMI≥25 kg/m2); (B) Overweight (BMI: 23–24.9 kg/m2); (C) Normal weight (BMI<23 kg/m2). BLR: Body mass index (BMI) and lymphocyte count ratio.
F4
Figure 4:
Subgroup analysis: area under the receiver operating characteristic curve (AUC) of BLR model in different age and albumin groups. (A) Age <50 years; (B) Age ≥50years; (C) Albumin <40 g/L; (D) Albumin ≥40 g/L. BLR: body mass index (BMI) and lymphocyte count ratio.

Discussion

Similar to other outbreaks of newly identified viruses such as Middle Eastern respiratory syndrome-CoV and SARS-CoV, SARS-CoV-2 is a coronavirus that has no proven regimen from conventional medicine and can contribute to approximately 7% to 30% severe/critical case ratio.[12] For severe/critical patients, aggressive treatment and intensive care are required; however, the mortality is still high. Global cumulative number of deaths is more than 4 million.[6] Incidence (33%) and mortality (42%) recorded a sharp increase in the African region when compared to the previous week.[14] Therefore, it is urgent to identify the clinical characteristics of severe/critical case and investigate factors that related with the severity of COVID-19. Analyzing the epidemiology and clinical characteristics of 402 patients from two medical centers specializing in infectious diseases, we observed 13.43% cases were categorized as severe patients, which had different clinical characteristics and laboratory parameters compared with non-severe COVID-19 patients. Together, we identified clinical factors which associated with the severity of disease, established an efficient risk prediction model, and believed it can help to early identify individuals with risk of becoming severe.

Our study indicated that patients with severe COVID-19 have higher BMI than those without severe COVID-19, which consist with previous study.[15] At least 25% of COVID-19 patients who died from the disease have obesity according to regional epidemiological data in the United States.[16] Meta-analysis also demonstrated a significant increase in the risk of critical illness and in-hospital mortality of COVID-19 among obese patients (BMI≥30 kg/m2).[17] Obesity was also reported as a more serious risk factor for COVID-19 in metabolic associated fatty liver disease (MAFLD) patients.[18] The role of obesity cannot be ignored, even though there are few available data on BMI for COVID-19 infected patients. Obesity might play a crucial role in the pathogenesis of SARS-CoV-2 and other respiratory infections.[19–21] So far, there is limited evidence on the mechanisms of the link between obesity and COVID-19, a number of interesting information can be inferred from studies conducted of influenza patients. In the influenza A cases, obesity can increase duration of virus shedding. In addition, obesity is an independent risk factor for hospitalization and death in H1N1 influenza[22] when people were first exposed to the novel virus SARS-CoV-2, no one has an established adaptive immune response. Obesity affects innate immunity, which might play an essential role in the severe COVID-19.[23] It is well known that obesity represents a chronic inflammatory state that can alter innate and adaptive immune responses, leading to the immune system more susceptible to infections and less responsive to vaccinations, antivirals, and antimicrobial drugs. Dysfunctional mesenchymal stem cells/ adipose-derived mesenchymal stem cells may also play an important role in promoting pulmonary fibrosis leading to lung functional failure and promoting systemic inflammation leading to cytokine storm, characteristic of severe COVID-19.[24] The putative receptor for COVID-19 entry into host cells is angiotensin converting enzyme 2 (ACE2). More interesting, compared with the expression in lung tissue, ACE2 expression in adipose tissue is higher.[25] This is a crucial finding because adipose tissue might be susceptible to COVID-19 patients.[26] Obese individuals have more adipose tissue, therefore the number of ACE2-expressing cells increases, which in turn produces a larger amount of ACE.

Our study also found that the continuous decline in the peripheral blood lymphocytes count is one of the early indicators for severe COVID-19. A retrospective analysis of 224 COVID-19 patients confirmed that the proportion of patients with lymphopenia in severe COVID-19 was significantly higher than that without lymphopenia.[27] Not only were patients with lymphopenia at higher risk of death, but they also had longer hospital stay.[27] Meta-analysis also confirmed that lymphopenia was closely associated with poor prognosis of COVID-19 patients.[28] In addition, the average value of lymphocyte subsets in severe COVID-19 patients generally decreased. The mean T-cell and NK cells values were below normal levels, while B-cell counts were at the low end of the normal range.[29] The significant reduction in the number of T-cell is the major cause of lymphopenia in COVID-19 patients.[29,30] T-cell proliferative capacity depends on telomere-length (TL). In the face of COVID-19, short telomeres might impair the ability to offset T-cell loss, which would result in lymphopenia.[31] Viral attachment can cause the reduction of lymphocyte counts. In addition, immune injuries caused by inflammatory mediators and the exudation of circulating lymphocyte into inflammatory lung tissues can contribute to the reduction of lymphocyte count.[31,32]

Consistent with the existing results, severe COVID-19 patients have lower levels of albumin and are older [Table 2].[33,34] BLR model had good prediction efficiency in different age groups and albumin groups by subgroup analysis shown in Figure 4. Albumin can not only effectively maintain plasma osmotic pressure, but also has functions in transport, anti-thrombosis, immunomodulatory, anti-inflammatory, and endothelial stability.[33,35,36] Glomerular leakage of chronic kidney disease and synthesis disorder of chronic liver disease are common causes of hypoproteinemia. Obesity itself can be accompanied by hypoalbuminemia, even in the absence of liver and kidney dysfunction.[33] In addition, TL shortens with age, leading to lymphopenia, which may partially explain the increased severity and mortality in older COVID-19 patients.[30] Obesity is related to an increased risk of diabetes, kidney disease, and cardiovascular disease, comorbidities that are considered to lead to increased vulnerability to pneumonia-related organ failures.[37] In the validation set, we found significant difference between severe and non-severe patients on hypertension and diabetes. However, we did not find significant differences in the training set.

The limitations of this study include small sample size and retrospective method. Some patients had missing documentation of the laboratory testing and the COVID-19 patients included in our study were only from Zhejiang and Jiangsu province and not very critically ill. This may lead to some bias in our overall understanding of the disease. On this basis, broader and larger studies are needed.

Overall, by analyzing 402 confirmed cases of SARS-CoV-2 infection, we preliminarily identified obesity and lymphopenia are the risk factor for COVID-19 related severity although the underlying mechanisms are unclear. Subjects with obesity and lymphopenia need intensive attention to reduce the risk of death. However, there is still a large gap in our understanding of the epidemiology, clinical characteristics, and outcome of this disease. Therefore, larger scale studies are needed to gain an in-depth understanding of the COVID-19, thereby providing a better evidentiary basis for standardized diagnosis and treatment.

Funding

This study was supported by the National Natural Science Fund (No. 81970545, 82170609), Natural Science Foundation of Shandong Province (Major Project) (No. ZR2020KH006), Ji’nan Science and Technology Development Project (No. 202019079), the Zhejiang Provincial Medical and Health Technology Program (2021KY1223), and the Taizhou Science and Technology Program (20ywa21).

Author Contributions

Zebao He and Jie Li conceived the manuscript and supervised the study. Zebao He, Jie Li, and Chao Wu designed the study. Zhengming Ge and Rui Huang collected the data. Jie Li, Zebao He, Fajuan Rui, Hongli Yang, Lingjun Ying, and Haihong Zhao analyzed the data. Jie Li, Zebao He, and Fajuan Rui drafted the manuscript. All authors edited and approved the final manuscript.

Conflicts of Interest

None.

References

[1]. Huang C, Wang Y, Li X, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 2020;395(10223):497–506. doi: 10.1016/S0140-6736(20)30183-5.
[2]. Lu R, Zhao X, Li J, et al. Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding. Lancet 2020;395(10224):565–574. doi: 10.1016/S0140-6736 (20)30251-8.
[3]. Li J, Huang DQ, Zou B, et al. Epidemiology of COVID-19: A systematic review and meta-analysis of clinical characteristics, risk factors, and outcomes. J Med Virol 2021;93:1449–1458. doi: 10.1002/jmv.26424.
[4]. Bernsmeier C, Cavazza A, Fatourou EM, et al. Leucocyte ratios are biomarkers of mortality in patients with acute decompensation of cirrhosis and acute-on-chronic liver failure. Aliment Pharmacol Ther 2020;52(5):855–865. doi: 10.1111/apt.15932.
[5]. Sun K, Chen J, Viboud C. Early epidemiological analysis of the coronavirus disease 2019 outbreak based on crowdsourced data: a population-level observational study. Lancet Digit Health 2020;2(4):e201–e208. doi: 10.1016/S2589-7500(20)30026-1.
[6]. World Health Organization. Weekly epidemiological update on COVID-19-14 September 2021. Available from: https://www.who.int/publications/m/item/weekly-epidemiological-update-on-covid-19—14-september-2021. Accessed September 14, 2021.
[7]. Guan WJ, Ni ZY, Hu Y, et al. Clinical characteristics of coronavirus disease 2019 in China. N Engl J Med 2020;382(18):1708–1720. doi: 10.1056/NEJMoa2002032.
[8]. Chen N, Zhou M, Dong X, et al. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study. Lancet 2020;395(10223):507–513. doi: 10.1016/S0140-6736(20)30211-7.
[9]. Liu C, Jiang ZC, Shao CX, et al. Preliminary study of the relationship between novel coronavirus pneumonia and liver function damage: a multicenter study. Zhonghua Gan Zang Bing Za Zhi 2020;28(2):107–111. doi: 10.3760/cma.j.issn.1007-3418.2020.02.003.
[10]. Xu XW, Wu XX, Jiang XG, et al. Clinical findings in a group of patients infected with the 2019 novel coronavirus (SARS-Cov-2) outside of Wuhan, China: retrospective case series. BMJ 2020;368:m606. doi: 10.1136/bmj.m606.
[11]. Yang X, Yu Y, Xu J, et al. Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study. Lancet Respir Med 2020;8(5):475–481. doi: 10.1016/S2213-2600(20)30079-5.
[12]. Chan KW, Wong VT, Tang SCW. COVID-19: an update on the epidemiological, clinical, preventive and therapeutic evidence and guidelines of integrative Chinese-Western Medicine for the management of 2019 novel coronavirus disease. Am J Chin Med 2020;48(3):737–762. doi: 10.1142/S0192415X20500378.
[13]. Epidemiology Working Group for NCIP Epidemic Response, Chinese Center for Disease Control and Prevention. The epidemiological characteristics of an outbreak of 2019 novel coronavirus diseases (COVID-19) in China. Zhonghua Liu Xing Bing Xue Za Zhi 2020;41(2):145–151. doi: 10.3760/cma.j.issn.0254-6450.2020.02.003.
[14]. World Health Organization. Weekly epidemiological update on COVID-19-29 June 2021. Available from: https://www.who.int/publications/m/item/weekly-epidemiological-update-on-covid-19—29-june-2021. Accessed June 29, 2021.
[15]. Peng YD, Meng K, Guan HQ, et al. Clinical characteristics and outcomes of 112 cardiovascular disease patients infected by 2019-nCoV. Zhonghua xin xue guan bing za zhi 2020;48(6):450–455. doi: 10.3760/cma.j.cn112148-20200220-00105.
[16]. Ryan PM, Caplice NM. Is adipose tissue a reservoir for viral spread, immune activation, and cytokine amplification in coronavirus disease 2019? Obesity 2020;28(7):1191–1194. doi: 10.1002/oby.22843.
[17]. Du Y, Lv Y, Zha W, et al. Association of body mass index (BMI) with critical COVID-19 and in-hospital mortality: a dose-response meta-analysis. Metabolism 2021;117:154373. doi: 10.1016/j.metabol.2020.154373.
[18]. Hussain A, Vasas P, El-Hasani S. Letter to the Editor: Obesity as a risk factor for greater severity of COVID-19 in patients with metabolic associated fatty liver disease. Metabolism 2020;108:154256. doi: 10.1016/j.metabol.2020.154256.
[19]. Muscogiuri G, Pugliese G, Barrea L, et al. Commentary: Obesity: The “Achilles heel” for COVID-19? Metabolism 2020;108:154251. doi: 10.1016/j.metabol.2020.154251.
[20]. Finer N, Garnett SP, Bruun JM. COVID-19 obesity. Clin Obes 2020;10(3):e12365. doi: 10.1111/cob.12365.
[21]. Onder G, Rezza G, Brusaferro S. Case-fatality rate and characteristics of patients dying in relation to COVID-19 in Italy. JAMA 2020;323(18):1775–1776. doi: 10.1001/jama.2020.4683.
[22]. Maier HE, Lopez R, Sanchez L, et al. Obesity increases the duration of influenza A virus shedding in adults. J Infect Dis 2018;218(9):1378–1382. doi: 10.1093/infdis/jiy370.
[23]. Gleeson LE, Roche HM, Sheedy FJ. Obesity, COVID-19 and innate immunometabolism. Br J Nutr 2021;125(6):628–632. doi: 10.1017/S0007114520003529.
[24]. Ritter A, Kreis NN, Louwen F, et al. Obesity and COVID-19: molecular mechanisms linking both pandemics. Int J Mol Sci 2020;21(16):5793. doi: 10.3390/ijms21165793.
[25]. Mohammad S, Aziz R, Al Mahri S, et al. Obesity and COVID-19: what makes obese host so vulnerable? Immun Ageing 2021;18(1):1. doi: 10.1186/s12979-020-00212-x.
[26]. Kassir R. Risk of COVID-19 for patients with obesity. Obes Rev 2020;21(6):e13034. doi: 10.1111/obr.13034.
[27]. Nie L, Liu Y, Weng Y, et al. Lymphocytes screening on admission is essential for predicting in-hospital clinical outcome in COVID-19 patients: a retrospective cohort study. Int J Lab Hematol 2021;43(6):1302–1308. doi: 10.1111/ijlh.13640.
[28]. Huang I, Pranata R. Lymphopenia in severe coronavirus disease-2019 (COVID-19): systematic review and meta-analysis. J Intensive Care 2020;8:36. doi: 10.1186/s40560-020-00453-4.
[29]. Azkur AK, Akdis M, Azkur D, et al. Immune response to SARS-CoV-2 and mechanisms of immunopathological changes in COVID-19. Allergy 2020;75(7):1564–1581. doi: 10.1111/all.14364.
[30]. Benetos A, Lai TP, Toupance S, et al. A mechanism for severity of disease in older patients with COVID-19: the nexus between telomere length and lymphopenia. medRxiv 2020;doi: 10.1101/2020.10.01. 20205393.
[31]. Gao YD, Ding M, Dong X, et al. Risk factors for severe and critically ill COVID-19 patients: a review. Allergy 2021;76(2):428–455. doi: 10.1111/all.14657.
[32]. Wang F, Nie J, Wang H, et al. Characteristics of peripheral lymphocyte subset alteration in COVID-19 pneumonia. J Infect Dis 2020;221(11):1762–1769. doi: 10.1093/infdis/jiaa150.
[33]. Deng M, Qi Y, Deng L, et al. Obesity as a potential predictor of disease severity in young COVID-19 patients: a retrospective study. Obesity 2020;28(10):1815–1825. doi: 10.1002/oby.22943.
[34]. Wang K, Qu M, Ding L, et al. Liver and kidney function biomarkers, blood cell traits and risk of severe COVID-19: a Mendelian randomization study. Front Genet 2021;12:647303. doi: 10.3389/fgene.2021.647303.
[35]. Casulleras M, Zhang IW, López-Vicario C, et al. Leukocytes, systemic inflammation and immunopathology in acute-on-chronic liver failure. Cells 2020;9(12):2632. doi: 10.3390/cells9122632.
[36]. Fernandez J, Clària J, Amorós A, et al. Effects of albumin treatment on systemic and portal hemodynamics and systemic inflammation in patients with decompensated cirrhosis. Gastroenterology 2019;157(1):149–162. doi: 10.1053/j.gastro.2019.03.021.
[37]. Stefan N, Birkenfeld AL, Schulze MB, et al. Obesity and impaired metabolic health in patients with COVID-19. Nat Rev Endocrinol 2020;16(7):341–342. doi: 10.1038/s41574-020-0364-6.
Keywords:

COVID-19; BMI; Lymphocyte count; Prediction model; SARS-CoV-2; Severity

Copyright © 2022 The Chinese Medical Association, published by Wolters Kluwer Health, Inc.