J Korean Med Sci.  2025 Mar;40(9):e25. 10.3346/jkms.2025.40.e25.

Temporal Radiographic Trajectory and Clinical Outcomes in COVID-19 Pneumonia: A Longitudinal Study

• Ahn DW
• Seo Y
• Goo T
• Jeong JB
• Park T
• Yoon SH

• KMID: 2565798
• DOI: http://doi.org/10.3346/jkms.2025.40.e25

Abstract

Background
Currently, little is known about the relationship between the temporal radiographic latent trajectories, which are based on the extent of coronavirus disease 2019 (COVID-19) pneumonia and clinical outcomes. This study aimed to elucidate the differences in the temporal trends of critical laboratory biomarkers, utilization of critical care support, and clinical outcomes according to temporal radiographic latent trajectories.
Methods
We enrolled 2,385 patients who were hospitalized with COVID-19 and underwent serial chest radiographs from December 2019 to March 2022. The extent of radiographic pneumonia was quantified as a percentage using a previously developed deep-learning algorithm. A latent class growth model was used to identify the trajectories of the longitudinal changes of COVID-19 pneumonia extents during hospitalization. We investigated the differences in the temporal trends of critical laboratory biomarkers among the temporal radiographic trajectory groups. Cox regression analyses were conducted to investigate differences in the utilization of critical care supports and clinical outcomes among the temporal radiographic trajectory groups.
Results
The mean age of the enrolled patients was 58.0 ± 16.9 years old, with 1,149 (48.2%) being male. Radiographic pneumonia trajectories were classified into three groups: The steady group (n = 1,925, 80.7%) exhibited stable minimal pneumonia, the downhill group (n = 135, 5.7%) exhibited initial worsening followed by improving pneumonia, and the uphill group (n = 325, 13.6%) exhibited progressive deterioration of pneumonia. There were distinct differences in the patterns of temporal blood urea nitrogen (BUN) and C-reactive protein (CRP) levels between the uphill group and the other two groups. Cox regression analyses revealed that the hazard ratios (HRs) for the need for critical care support and the risk of intensive care unit admission were significantly higher in both the downhill and uphill groups compared to the steady group. However, regarding in-hospital mortality, only the uphill group demonstrated a significantly higher risk than the steady group (HR, 8.2; 95% confidence interval, 3.08–21.98).
Conclusion
Stratified pneumonia trajectories, identified through serial chest radiographs, are linked to different patterns of temporal changes in BUN and CRP levels. These changes can predict the need for critical care support and clinical outcomes in COVID-19 pneumonia. Appropriate therapeutic strategies should be tailored based on these disease trajectories.

Keyword

COVID-19; Pneumonia Trajectory; Critical Care Support; Clinical Outcome

Figure

• 

  Fig. 1 Diagram of showing enrollment of study subjects.COVID-19 = coronavirus disease 2019.

• 

  Fig. 2 Regrouping of radiographic latent classes into simplified radiographic trajectory groups for enhanced clinical interpretation. This figure illustrates the consolidation of six radiographic latent classes into three distinct trajectory groups aimed at simplifying clinical analysis and interpretation. The regrouping criteria were primarily based on the trend of slope changes in the pneumonia extent over time, facilitating a clearer comparative analysis. The purpose of this regrouping is to enhance clinical clarity and simplify the analysis, providing a more intuitive understanding of the disease progression patterns among patients.

• 

  Fig. 3 Example of pneumonic extent course in uphill group. Chest radiograph progression in a patient from the uphill group: a 66-year-old male COVID-19 patient exhibited a 4.53% extent of pneumonia on an initial chest radiograph (A) taken on the first day after hospital admission. Subsequent chest radiographs taken 6 days (B), 15 days (C), and 51 days (D) after hospital admission showed a gradual increase in pneumonia extent to 15.63%, 24.49%, and 50.91%, respectively. The patient died from aggravated COVID-19 pneumonia 52 days after admission.COVID-19 = coronavirus disease 2019.

• 

  Fig. 4 Association between radiographic trajectory groups and temporal changing trends of critical laboratory biomarkers. Our findings revealed significant variances in BUN levels among the groups. Specifically, the uphill group exhibited abnormally high BUN levels for most of the hospital stay, aside from the initial phase of the clinical course. By contrast, both the steady and downhill groups exhibited BUN levels within the normal range throughout their hospitalization, albeit with some fluctuations in levels (A). Regarding CRP levels, all groups initially presented values at the upper limit. In the uphill group, the CRP levels initially decreased before rising again over time. This fluctuation indicates that, as the severity of pneumonia increased quantitatively in the uphill group, so did the CRP levels, thus reflecting the degree of inflammation associated with pneumonia. Meanwhile, the downhill and steady groups exhibited a gradual decline in CRP levels as time progressed (B). In the uphill group, the WBC counts increased sharply from the beginning of hospitalization, surpassing the upper normal limit around day 10, and continued to rise with an abruptly increased slope. It peaked around day 14 before gradually decreasing, showing an inverted U-shape pattern. In the downhill group, the WBC counts also displayed an inverted U-shape similar to the uphill group, but the peak did not exceed the upper normal limit. In the steady group, the WBC counts exhibited a slightly increased slope throughout the hospitalization period (C). For the lymphocyte level, there is no specific trend over the hospitalization period among the three groups (D). The critical laboratory tests included in this paper were measured at irregular intervals, and the mean value of the measurements taken on the same date among patients within each group was presented as the representative value for each group.BUN = blood urea nitrogen, CRP = C-reactive protein, WBC = white blood cell.aThe black dashed lines in each subfigure indicate the cutoff values for the lower or upper limits of each laboratory finding.

Reference

1. Zhang Y, Zheng L, Liu L, Zhao M, Xiao J, Zhao Q. Liver impairment in COVID-19 patients: a retrospective analysis of 115 cases from a single centre in Wuhan city, China. Liver Int. 2020; 40(9):2095–2103. PMID: 32239796.

2. Park JM, Koo HY, Lee JR, Lee H, Lee JY. COVID-19 mortality and severity in cancer patients and cancer survivors. J Korean Med Sci. 2024; 39(2):e6. PMID: 38225782.

3. Park YJ, Song KJ, Hong KJ, Park JH, Kim TH, Kim YS, et al. The impact of the COVID-19 outbreak on emergency medical service: an analysis of patient transportations and time intervals. J Korean Med Sci. 2023; 38(42):e317. PMID: 37904654.

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(13):1239–1242. PMID: 32091533.

5. Chen N, Zhou M, Dong X, Qu J, Gong F, Han Y, 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. PMID: 32007143.

6. Wang D, Hu B, Hu C, Zhu F, Liu X, Zhang J, et al. Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in Wuhan, China. JAMA. 2020; 323(11):1061–1069. PMID: 32031570.

7. Feng Z, Yu Q, Yao S, Luo L, Zhou W, Mao X, et al. Early prediction of disease progression in COVID-19 pneumonia patients with chest CT and clinical characteristics. Nat Commun. 2020; 11(1):4968. PMID: 33009413.

8. Zhou K, Sun Y, Li L, Zang Z, Wang J, Li J, et al. Eleven routine clinical features predict COVID-19 severity uncovered by machine learning of longitudinal measurements. Comput Struct Biotechnol J. 2021; 19:3640–3649. PMID: 34188785.

9. Pan F, Ye T, Sun P, Gui S, Liang B, Li L, et al. Time course of lung changes at chest CT during recovery from coronavirus disease 2019 (COVID-19). Radiology. 2020; 295(3):715–721. PMID: 32053470.

10. Yoon SH, Lee KH, Kim JY, Lee YK, Ko H, Kim KH, et al. Chest radiographic and CT findings of the 2019 novel coronavirus disease (COVID-19): analysis of nine patients treated in Korea. Korean J Radiol. 2020; 21(4):494–500. PMID: 32100485.

11. Kim JY, Jung KJ, Yoo SJ, Yoon SH. Stratifying the early radiologic trajectory in dyspneic patients with COVID-19 pneumonia. PLoS One. 2021; 16(10):e0259010. PMID: 34679127.

12. Gourdeau D, Potvin O, Archambault P, Chartrand-Lefebvre C, Dieumegarde L, Forghani R, et al. Tracking and predicting COVID-19 radiological trajectory on chest X-rays using deep learning. Sci Rep. 2022; 12(1):5616. PMID: 35379856.

13. Fang X, Li S, Yu H, Wang P, Zhang Y, Chen Z, et al. Epidemiological, comorbidity factors with severity and prognosis of COVID-19: a systematic review and meta-analysis. Aging (Albany NY). 2020; 12(13):12493–12503. PMID: 32658868.

14. Jang SY, Seon JY, Eun BL, Koh SB, Yoo JH, Lee WY, et al. Risk factors of outcomes of COVID-19 patients in Korea: focus on early symptoms. J Korean Med Sci. 2021; 36(18):e132. PMID: 33975399.

15. Yoo SJ, Kim H, Witanto JN, Inui S, Yoon JH, Lee KD, et al. Generative adversarial network for automatic quantification of coronavirus disease 2019 pneumonia on chest radiographs. Eur J Radiol. 2023; 164:110858. PMID: 37209462.

16. Berlin KS, Parra GR, Williams NA. An introduction to latent variable mixture modeling (part 2): longitudinal latent class growth analysis and growth mixture models. J Pediatr Psychol. 2014; 39(2):188–203. PMID: 24277770.

17. Wolff D, Nee S, Hickey NS, Marschollek M. Risk factors for COVID-19 severity and fatality: a structured literature review. Infection. 2021; 49(1):15–28. PMID: 32860214.

18. Liu YM, Xie J, Chen MM, Zhang X, Cheng X, Li H, et al. Kidney function indicators predict adverse outcomes of COVID-19. Med. 2021; 2(1):38–48.e2. PMID: 33043313.

19. Cozzi D, Albanesi M, Cavigli E, Moroni C, Bindi A, Luvarà S, et al. Chest X-ray in new coronavirus disease 2019 (COVID-19) infection: findings and correlation with clinical outcome. Radiol Med. 2020; 125(8):730–737. PMID: 32519256.