Journal Information
Vol. 58. Issue S1.
CURRENT SITUATION AND LONG-TERM CONSEQUENCES OF COVID-19 INFECTION
Pages 69-72 (April 2022)
Share
Share
Download PDF
More article options
Visits
...
Vol. 58. Issue S1.
CURRENT SITUATION AND LONG-TERM CONSEQUENCES OF COVID-19 INFECTION
Pages 69-72 (April 2022)
Scientific Letter
Full text access
Chest CT as a Prognostic Tool in COVID-19
La TC de tórax como herramienta pronóstica en COVID-19
Visits
...
Mariane Gonçalves Martynychen Canana,
Corresponding author
mariane.canan@hc.ufpr.br

Corresponding author.
, Caroline Souza Sokoloskia, Vitor Loureiro Diasa, João Mario Clementin de Andradeb, Ana Carolina Nicoletti Bassob, Carolina Chomiskib, Dante Luiz Escuissatoc, Itsem Campos Andrade Juniord, Izabele Cristine Vazd, Rebecca Saray Marchesini Stivale, Karin Mueller Storrerf
a Cardio-Pulmonology Division, Federal University of Paraná, Curitiba, Brazil
b Federal University of Paraná, Curitiba, Brazil
c Radiology Unit, Federal University of Paraná, Curitiba, Brazil
d Medical School, Federal University of Paraná, Curitiba, Brazil
e Emergency Department, Federal University of Paraná, Curitiba, Brazil
f Department of Internal Medicine, Federal University of Paraná, Curitiba, Brazil
Article information
Full Text
Bibliography
Download PDF
Statistics
Tables (2)
Table 1. Chest computed tomography results.
Table 2. Factors associated with in-hospital death.
Show moreShow less
Additional material (1)
Full Text
To the Director,

Considering the high variability of clinical course in the COVID-19, prognostic tools remain necessary for strategic healthcare planning. In hospitalized patients, clinical and demographic characteristics are associated with greater complications and death due to the disease.1–4 Chest computed tomography (CT) has also been widely used in the evaluation of these patients due to its high sensitivity,5 speed and relative availability, not only as a complementary diagnostic tool but also as an auxiliary method in clinical management and risk stratification.6,7 Therefore, the objective of our study is to analyze tomographic findings at admission of hospitalized patients with COVID-19 and their relationship with disease severity, using the need for invasive mechanical ventilation (MV) and in-hospital mortality as main outcomes.

We conducted a cross-sectional analysis of all confirmed COVID-19 cases in hospitalized adult patients, from January through December 2020, at a tertiary care public university referral hospital for COVID-19 care in Curitiba, the largest city in southern Brazil. We excluded patients who did not undergo chest CT or whose chest CT was performed outside the period of 5 days after hospital admission, as well as patients with incomplete data.

Chest CT was performed with high-resolution technique, with patients in a supine position and in complete inspiration, on a Toshiba Aquilion 64 CT scanner. The images were evaluated by certified radiologists. Patients were evaluated regarding the predominant tomographic pattern, which was considered to affect more than 50% of involved lung parenchyma. Patterns considered were ground-glass opacities, crazy paving (ground-glass opacities with intralobular septal thickening), consolidation and mixed pattern (half of the affected area with consolidation and the remainder with another tomographic pattern). Pulmonary artery diameter (PAD) and the dimensions of the mediastinal lymph nodes were evaluated in axial plane. Lymph nodes were considered enlarged when the smallest diameter found was greater than 10mm. The extent of lung parenchyma involvement was estimated according to visual scale in the axial plane and categorized into none, less than 25%, 25–50%, 50–75% and more than 75%. Clinical, laboratory, and demographic data were also collected.

Statistical analysis was performed using IBM SPSS Statistics software version 27. Quantitative variables were described in means±standard deviation or median (interquartile range [IQR]), as appropriate. Categorical variables were described as frequencies and percentages. The relationship of demographic, clinical and tomographic variables with mortality and the need for MV was evaluated using univariate logistic regression. A multivariate logistic regression was further performed to identify independent predictors. Only variables with at least 690 valid entries and with p values <0.05 in the univariate logistic regression analysis were included. The stepwise backward approach was used. Odds ratio (OR) with a 95% confidence interval (CI) were reported for all the variables analyzed. P values <0.05 were considered statistically significant.

In 2020, 1246 patients were hospitalized due to COVID-19 in our hospital, 492 of which were excluded from our analysis (Supplementary Fig. 1); thus, we assessed 754 patients. The mean age was 56.3±14.7 years old. The median time of symptoms onset until admission was 8 days (IQR 6–11). Clinical and demographic characteristics of the study population are shown in Supplementary Table 1.

Chest CT was performed within the first two days of hospitalization (median 0, IQR 0–1) in 86.7% of patients, with a median time between onset of symptoms and chest CT of 9 days (IQR 6–11). Similarly to data already reported,7–9 the most frequent finding was ground-glass opacity (53.9%), related to pulmonary alveolar oedema and partial filling of alveoli by proteinaceous exudate in the early stages of the disease10–12 (Table 1). As the disease progresses, with lymphocyte activation and release of inflammatory cytokines, these changes may become more extensive and there may be intra and interlobular septal infiltration with the presence of crazy paving, indicating both alveolar and interstitial involvement due to COVID-19.12,13 This pathophysiological mechanism seems to be related to the clinical worsening, which is in line with our findings, where crazy paving was found in 10.7% of our cases and, although not statistically significant regarding mortality in this analysis, was associated with more severe disease when compared to ground-glass opacities. Table 2 and Supplementary Table 2 show the characteristics of patients who died or needed MV, in comparison with those who did not.

Table 1.

Chest computed tomography results.

Chest CT characteristics
Dominant pattern  N=736  Lung involvement  N=749  Other findings  N=754 
Ground-glass opacity  397 (53.9%)  None  12 (1.6%)  Pleural effusion  87 (11.5%) 
Consolidation  53 (7.2%)  <25%  122 (16.3%)  Lymph node enlargement  158 (21.0%) 
Crazy paving  81 (11%)  25–50%  239 (31.9%)  PADa (mm)  27.6±4.2 
Mixed  205 (27.9%)  50–75%  215 (28.7%)  PAD>30mm  167 (22.1%) 
    >75%  161 (21.5%)     
  Laboratory abnormalities according to the degree of lung involvement in chest CT
  <25%  25–50%  50–75%  >75%  p* 
Lymphocytesb  134, 1152 (738–1766)  238, 881 (565–1304)  215, 850 (531–1200)  161, 765 (533–1065)  <0.001 
D-Dimerb  121, 1.12 (0.52–4.05)  217, 0.77 (0.45–1.44)  207, 0.95 (0.54–1.79)  157, 0.96 (0.6–2.09)  0.009 
Albuminc  113, 3.63±0.59  193, 3.63±0.47  174, 3.57±0.44  137, 3.37±0.45  <0.001 
LDHb  110, 269 (7–334)  18, 333 (62–416)  181, 384 (96–492)  143, 420 (330–591)  <0.001 
Ferritinb  105, 453 (168–989)  193, 830 (26–1527)  189, 1144 (439–1675)  159, 1547 (836–1675)  <0.001 
C-reactive proteinb  128, 2.54 (0.88–7.27)  232, 6.06 (2.99–11.63)  212, 7.86 (4.9–13.92)  161, 12.36 (6.83–16)  <0.001 
Procalcitoninb  24, 0.13 (0.04–1.02)  45, 0.1 (0.05–0.55)  53, 0.11 (0.05–0.54)  55, 0.15 (0.08–0.45)  0.463 
  Pairwise comparison analysis of laboratory abnormalities according to the degree of lung involvement in chest CT
  Lymphocytesd  Albumind  C-reactive proteind  Ferritind  LDHd 
<25% vs. 25-50%  <0.001  <0.001  0.003  0.029 
<25% vs. 50-75%  <0.001  <0.001  <0.001  <0.001 
<25% vs. >75%  <0.001  <0.001  <0.001  <0.001  <0.001 
25–50% vs. 50–75%  0.547  0.045  0.305  0.003 
25––50% vs. >75%  0.036  <0.001  <0.001  <0.001  <0.001 
50–75% vs. >75%  0.001  0.001  0.003  0.054 
  Relationship between pulmonary artery diameter (PAD) and laboratory abnormalities according to reference values
  Value (n)  PAD (mm)a  Value (nPAD (mm)a  P# 
Lymphocytes (×103/μL)  ≥1000 (317)  27.2±4.1  <1000 (436)  27.8±4.3  0.065 
D-Dimer (mg/L FEU)  ≤0.55 (199)  27.1±4.1  >0.55 (507)  27.9±4.3  0.023 
Albumin (g/dL)  ≥3.5 (396)  27.2±4.0  <3.5 (225)  28.3±4.5  0.001 
Cardiac troponin (pg/ml)  <15.6 (435)  27.1±3.8  ≥15.6 (135)  29.7±4.7  <0.001 
LDH (U/L)  <220 (61)  27.2±3.1  ≥220 (567)  27.7±4.3  0.305 
Ferritin (ng/mL)  ≤204 (74)  27.9±4.5  >204 (567)  27.6±4.2  0.562 
C-reactive protein (mg/dL)  ≤0.5 (28)  27.6±3.6  >0.5 (710)  27.5±4.2  0.92 
Procalcitonin (ng/mL)  <0.5 (132)  27.4±4.3  ≥0.5 (45)  27.9±3.9  0.544 

PAD: pulmonary artery diameter.

Data reported in n (%), except when indicated otherwise.

a

Mean±standard deviation.

b

n, median (IQR).

c

n, mean±standard deviation.

d

One way ANOVA and Bonferroni Test, p<0.05.

*

One way ANOVA (leukocytes, lymphocytes, albumin, LDH, ferritin, C-reactive protein); Kruskal–Wallis (d-dimer, procalcitonin); p < 0.05.

#

Student's t-test for independent samples, p < 0.05.

Table 2.

Factors associated with in-hospital death.

Clinical characteristics  N  Hospital Discharge  Death  OR  CI95%  p* 
Age60 years old  318  224 (70.4%)  94 (29.6%)  4.66  3.07–7.08  <0.001 
Male sex  390  325 (83.3%)  65 (16.7%)  0.92  0.63–1.34  0.665 
Smoking history  177  134 (75.7%)  43 (24.3%)  2.15  1.36–3.39  0.001 
Obesity  224  194 (86.6%)  30 (13.4%)  0.66  0.43–1.03  0.07 
Hypertension  379  305 (80.5%)  74 (19.5%)  1.38  0.94–2.02  0.096 
Diabetes Mellitus  238  190 (79.8%)  48 (20.2%)  1.34  0.90–1.98  0.149 
Coronary artery disease  63  49 (77.8%)  12 (22.2%)  1.42  0.76–2.65  0.276 
COPD  48  34 (70.8%)  14 (29.2%)  2.09  1.09–4.03  0.027 
Chronic heart disease  132  101 (76.5%)  31 (23.5%)  1.62  1.03–2.56  0.038 
Chronic lung disease  110  85 (77.3%)  25 (22.7%)  1.51  0.92–2.47  0.101 
Chronic liver disease  3 (37.5%)  5 (62.5%)  8.28  1.95–35.1  0.004 
Chronic kidney disease  34  21 (61.8%)  13 (38.2%)  3.19  1.55–6.55  0.002 
Active cancer  33  21 (63.6%)  12 (36.4%)  2.92  1.40–6.10  0.004 
Laboratory tests  N  Hospital discharge  Death  OR  95% CI  p* 
Lymphocytesa,b  753  623, 910 (577–1327)  130, 760 (462–1013)  0.92  0.89–0.96  <0.001 
C-reactive proteinb  738  608, 6.58 (2.91–12.68)  130, 10.52 (5.54–15.62)  1.08  1.04–1.12  <0.001 
Ferritina,b  641  520, 900 (412–1675)  121, 1210 (478–1675)  1.04  1.003–1.07  0.031 
LDHa,b  628  511, 336 (263–428)  117, 436 (319–593)  1.4  1.26–1.55  <0.001 
d-Dimerb  706  581, 0.84 (0.48–1.71)  125, 1.44 (0.77–3.67)  1.03  1.01–1.05  0.009 
Albuminc  621  507, 3.61±0.45  114, 3.32±0.60  0.32  0.21–0.49  <0.001 
Procalcitoninb  177  104, 0.09 (0.05–0.24)  73, 0.26 (0.08–1.35)  1.27  1.01–1.59  0.039 
Cardiac troponina,b  570  455, 10 (10–10)  115, 14.3 (10–61.40)  1.01  0.99–1.03  0.394 
Chest CT  N  Hospital discharge  Death  OR  95% CI  p* 
Lung involvement area
≤75%d  588  505 (85.9%)  83 (14.1%)       
>75%  161  114 (70.8%)  47 (29.2%)  2.51  1.66–3.78  <0.001 
Dominant pattern
Ground-glass opacityd  397  323 (81.4%)  74 (18.6%)       
Mixed  205  181 (88.3%)  24 (11.7%)  0.58  0.35–0.95  0.03 
Crazy paving  81  59 (72.8%)  22 (27.2%)  1.63  0.94–2.82  0.083 
Consolidation  53  45 (84.9%)  8 (15.1%)  0.78  0.35–1.72  0.531 
Pleural effusion  87  56 (64.4%)  31 (35.6%)  3.18  1.95–5.17  <0.001 
Lymph node enlargement  158  120 (76.0%)  38 (24.1%)  1.73  1.13–2.66  0.011 
PAD (mm)c  754  624, 27.3±4.1  139, 28.9±4.6  1.09  1.04–1.13  <0.001 
  Independent predictors of in-hospital mortality (multivariate analysis)e
  OR  95% CI  p* 
Age60 years old  4.38  2.76–6.95  <0.001 
Active cancer  2.99  1.21–7.17  0.018 
Chronic liver disease  11.58  2.22–60.38  0.004 
Lymphocytesa  0.95  0.91–0.997  0.037 
C-reactive protein  1.05  1.002–1.09  0.038 
Lung involvement>75%  2.25  1.39–3.63  <0.001 
Pleural effusion  2.36  1.33–4.19  0.003 
*

Logistic regression model and Wald test, p<0.05.

a

OR corresponds to each increment of 100 units of the variable.

b

n, median (IQR).

c

n, mean±standard deviation.

d

Reference variable.

e

Variables initially included in the model: age60 years old, chronic heart disease, chronic obstructive pulmonary disease, active cancer, chronic kidney disease, chronic liver disease, C-reactive protein, lymphocytes, d-dimer, lung parenchyma involvement >75%, pleural effusion and pulmonary artery diameter.

When assessing the extent of lung disease on chest CT, half of the patients had parenchyma involvement greater than 50%. Greater lung parenchyma involvement was significantly associated with high values of CRP, ferritin and LDH and low lymphocytes and albumin, probably indicating that these findings are related to more extensive involvement on CT. In addition, lung parenchyma involvement greater than 75% was independently associated with a higher risk of clinical worsening (OR=2.25, 95% CI 1.39–3.63, p<0.001) and death (OR=2.6, 95% CI 1.37–3.09, p<0.001). The relationship between the degree of lung parenchyma involvement, assessed by different scores, and worse outcomes for COVID-19 has also been described in other studies,8,9,11,14 which reinforces the role of chest CT as a prognostic tool for COVID-19.

Another independent predictor of mortality identified in chest CT was the presence of pleural effusion, found in 11.5% of patients, with an adjusted OR of 2.36 (95% CI 1.33–4.19, p=0.003) compared to patients without it. A similar result was found in a meta-analysis which reported a 9.5% prevalence of pleural effusion, with an OR of 4.53 (95% CI 2.16–9.49) for mortality in this group.15 This fact seems to be associated with direct injury to the lung tissue by SARS-CoV-2 and with a higher systemic inflammatory response,16 analogous to that already described in other viral respiratory infections, such as MERS17 and H1N1.18

In our cohort, patients who needed MV or who died from COVID-19 also had larger PAD. The mean PAD was 27.6±4.2mm and a diameter greater than 30mm was found in 167 patients (22.1%). Esposito et al. reported a relationship between PAD enlargement and in-hospital mortality due to COVID-19, even after adjustment for demographic characteristics and comorbidities, suggesting that this is an acute complication of the disease.19 The main mechanism related to larger PAD is related to increased pulmonary vascular resistance, secondary to endothelial injury, tissue inflammation and prothrombotic state with occlusion of small vessels.20 Although cardiovascular and thromboembolic complications were not evaluated in this study, patients with higher d-dimer and cardiac troponin presented significantly larger PAD when compared to patients with normal values (27.1±4.1mm vs. 27.9±4.3, p=0.023; and 27.1±3.8 vs. 29.7±4.7, p<0.001, respectively).

This study has limitations. The collection of data from medical records led to the exclusion of patients due to incomplete data since care was provided without a standardized protocol of data collection. Another limitation is that all patients were admitted to the same public hospital, so data may not accurately represent the reality of other regions.

In conclusion, easily obtainable chest CT data, such as extent of lung involvement in visual scale, PAD and presence of pleural effusion, can be used as predictors of disease severity and in-hospital mortality in COVID-19.

Ethics approval

The study was previously approved by the local Research Ethics Committee, under the opinion 4.215.032, of August 16th, 2020, with waiver of informed consent.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability statement

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Conflict of interest

None declared.

Acknowledgements

The authors would like to thank all professionals involved in the care of patients with COVID-19, as well as patients themselves and their families.

Appendix A
Supplementary data

The following are the supplementary data to this article:

References
[1]
A. Borobia, A. Carcas, F. Arnalich, R. Álvarez-Sala, J. Monserrat-Villatoro, M. Quintana, et al.
A cohort of patients with COVID-19 in a major teaching hospital in Europe.
JCM, 9 (2020), pp. 1733
[2]
J. Berenguer, P. Ryan, J. Rodríguez-Baño, I. Jarrín, J. Carratalà, J. Pachón, et al.
Characteristics and predictors of death among 4035 consecutively hospitalized patients with COVID-19 in Spain.
Clin Microbiol Infect, 26 (2020), pp. 1525-1536
[3]
M.S. Marcolino, P.K. Ziegelmann, M.V.R. Souza-Silva, I.J.B. Nascimento, L.M. Oliveira, L.S. Monteiro, et al.
Clinical characteristics and outcomes of patients hospitalized with COVID-19 in Brazil: results from the Brazilian COVID-19 registry.
Int J Infect Dis, 107 (2021), pp. 300-310
[4]
G. Suleyman, R.A. Fadel, K.M. Malette, C. Hammond, H. Abdulla, A. Entz, et al.
Clinical characteristics and morbidity associated with coronavirus disease 2019 in a series of patients in Metropolitan Detroit.
JAMA Netw Open, 3 (2020), pp. e2012270
[5]
T. Ai, Z. Yang, H. Hou, C. Zhan, C. Chen, W. Lv, et al.
Correlation of chest CT and RT-PCR testing for coronavirus disease 2019 (COVID-19) in China: a report of 1014 cases.
Radiology, 296 (2020), pp. E32-E40
[6]
E.A. Akl, I. Blažić, S. Yaacoub, G. Frija, R. Chou, J.A. Appiah, et al.
Use of chest imaging in the diagnosis and management of COVID-19: a WHO rapid advice guide.
Radiology, 298 (2021), pp. E63-E69
[7]
M. Yuan, W. Yin, Z. Tao, W. Tan, Y. Hu.
Association of radiologic findings with mortality of patients infected with 2019 novel coronavirus in Wuhan, China.
PLOS ONE, 15 (2020), pp. e0230548
[8]
R. Mogami, A.J. Lopes, R.C. Araújo Filho, F.C.S. de Almeida, Messeder A.M.D.C., Koifman A.C.B., et al.
Chest computed tomography in COVID-19 pneumonia: a retrospective study of 155 patients at a university hospital in Rio de Janeiro, Brazil.
Radiol Bras, 54 (2021), pp. 1-8
[9]
M.I. Metwally, M.A.A. Basha, M.M.A. Zaitoun, H.M. Abdalla, H.A.E. Nofal, H. Hendawy, et al.
Clinical and radiological imaging as prognostic predictors in COVID-19 patients.
Egypt J Radiol Nucl Med, 52 (2021), pp. 100
[10]
F. Pan, T. Ye, P. Sun, S. Gui, B. Liang, L. Li, et al.
Time course of lung changes at chest CT during recovery from coronavirus disease 2019 (COVID-19).
Radiology, 295 (2020), pp. 715-721
[11]
F. Pan, C. Zheng, T. Ye, L. Li, D. Liu, L. Li, et al.
Different computed tomography patterns of Coronavirus Disease 2019 (COVID-19) between survivors and non-survivors.
[12]
S. Tian, W. Hu, L. Niu, H. Liu, H. Xu, S.Y. Xiao.
Pulmonary pathology of early-phase 2019 novel coronavirus (COVID-19) pneumonia in two patients with lung cancer.
J Thorac Oncol, 15 (2020), pp. 700-704
[13]
D.M. Hansell, A.A. Bankier, H. MacMahon, T.C. McLoud, N.L. Müller, J. Remy.
Fleischner Society: glossary of terms for thoracic imaging.
Radiology, 246 (2008), pp. 697-722
[14]
C. Salvatore, F. Roberta, L. Angela, P. Cesare, C. Alfredo, G. Giuliano, et al.
Clinical and laboratory data, radiological structured report findings and quantitative evaluation of lung involvement on baseline chest CT in COVID-19 patients to predict prognosis.
Radiol Med, 126 (2021), pp. 29-39
[15]
S.S. Rathore, N. Hussain, A.H. Manju, Q. Wen, S. Tousif, C.A. Avendaño-Capriles, et al.
Prevalence and clinical outcomes of pleural effusion in COVID-19 patients: a systematic review and meta-analysis.
J Med Virol, 94 (2022), pp. 229-239
[16]
X. Wei, X. Wang, L. Ye, Y. Niu, W. Peng, Z. Wang, et al.
Pleural effusion as an indicator for the poor prognosis of COVID-19 patients.
Int J Clin Pract, 75 (2021),
[17]
K.M. Das, E.Y. Lee, M.A. Enani, S.E. AlJawder, R. Singh, S. Bashir, et al.
CT correlation with outcomes in 15 patients with acute middle east respiratory syndrome coronavirus.
Am J Roentgenol, 204 (2015), pp. 736-742
[18]
K. Schoen, N. Horvat, N.F.C. Guerreiro, I. de Castro, K.S. de Giassi.
Spectrum of clinical and radiographic findings in patients with diagnosis of H1N1 and correlation with clinical severity.
BMC Infect Dis, 19 (2019), pp. 964
[19]
A. Esposito, A. Palmisano, M. Toselli, D. Vignale, A. Cereda, P.M.V. Rancoita, et al.
Chest CT-derived pulmonary artery enlargement at the admission predicts overall survival in COVID-19 patients: insight from 1461 consecutive patients in Italy.
Eur Radiol, 31 (2021), pp. 4031-4041
[20]
Z. Varga, A.J. Flammer, P. Steiger, M. Haberecker, R. Andermatt, A.S. Zinkernagel, et al.
Endothelial cell infection and endotheliitis in COVID-19.
Lancet, 395 (2020), pp. 1417-1418
Copyright © 2022. SEPAR
Archivos de Bronconeumología

Subscribe to our newsletter

Article options
Tools
Supplemental materials

Are you a health professional able to prescribe or dispense drugs?