A retrospective study of association of CT severity with clinical profile and outcomes of patients with COVID-19 in the second wave

INTRODUCTION

Coronavirus disease 2019 (COVID-19) was originally identified in a group of 41 patients in Wuhan, China, in December 2019, who all had unexplained cases of pneumonia. COVID-19 was declared to be public health emergency by the World Health Organization on January 30, 2020. COVID-19 is an infection that has spread widely and quickly worldwide, resulting in a pandemic with serious consequences for the economy and healthcare delivery systems.^[1]

It was a new and unforeseen challenge because the clinical presentations ranged from asymptomatic carriers to patients requiring assisted ventilatory support, and ICU stays with higher mortality. The diagnostic test utilized as the standard of reference for confirmation of the disease has been the nasopharyngeal swab reverse transcription-polymerase chain reaction (RT-PCR) test.^[2] The test has proven to be an effective tool, but it is reported to be associated with a fewer but significant number of false-negative findings.^[3]

There is a significant role of the non-contrast high-resolution CT (HRCT) chest in detecting the disease in the early stage, specifically among patients who have false-negative reports of RT-PCR. It also helps in observing the disease progression.^[4] The typical findings of chest CT in the case of COVID-19 pneumonia include “bilateral, peripheral, and basal predominant ground-glass opacities with or without consolidation and broncho-vascular thickening.”^[5] In addition, atypical findings are “cavitations, central upper lobe predominance, nodules, masses, tree-inbud sign, lymphadenopathy, and pleural effusion.”

For the diagnosis of COVID-19 pneumonia after a mean duration of 5 days, the HRCT chest is reported to show a sensitivity of 97%.^[6] Chest computed tomography severity score was developed to assist evaluation of the effect of COVID-19 on the scan performed at the time of admission.^[7] This severity scoring system gives an approach for recognizing the requirement of hospital admission in patients. In addition, the severity of the disease can be determined through findings of CT scan, which supports the clinicians in clinical evaluation and taking decisions regarding appropriate management.^[8]

During the acute phase of the disease, a CT severity score system allows patients’ outcomes to be recognized. In critically-ill patients, the severity of the disease can also affect the prognosis, allowing for proper early involvement of intensive care.^[9,10]

Many previous researches used visual as well as software quantitative assessments for examining the involvement of lungs on the chest CT images.^[11,12] There is a scarcity of studies that evaluated the association of chest CT severity scores and the outcomes in COVID-19 patients, especially in India.^[13-19] Thus, we conducted this study to find out the association of CT severity score with demographic and clinical characteristics as well as mortality in the patients who were confirmed to have COVID-19 disease in the second wave.

MATERIALS AND METHODS

RESULTS

The mean age of the patients in the present study was 49.62 ± 15.3 years with 69.52% of the patients being males. The chief symptoms were fever (80.63%), cough (68.95%), and dyspnea (60.11%). Hypertension diabetes mellitus and obesity were present in 25.07%, 24.50%, and 12.25%, respectively [Table 3].

Regarding respiratory support parameters, the mean values of SpO2 on admission were 90.49 ± 6.07. The median duration of NRBM, NIV, HFNC, venture/face mask, and intubation were 0, 0, 0, 1, and 0 respectively [Table 4].

TLC, absolute neutrophils, absolute lymphocyte, Creatinine, ALT, CRP, LDH, Ferritin, IL-6, and D-dimer were deranged in 39.03%, 38.18%,85.19%,45.87%, 58.69%, 95.16%, 66.38%, 64.96%, 96.40%, and 82.72%, respectively [Table 5].

CT severity score was normal in 5.98% of patients, mild in 21.94%, moderate in 41.60%, and severe in 30.48% of patients [Figure 1]. Mortality rate was 5.70% in the study [Table 6]. The representative case images with varying severity is shown in Figures 2-4.

Close

Figure 1:

Distribution of CTSS of study subjects.

Export to PPT

Close

Figure 2:

A 54-year-old female admitted with complaints of shortness of breath and fever for 1 week. HRCT coronal section revealed ground glass density opacities (black arrow) diffusely scattered in bilateral lungs, along with subpleural fibrotic bands (white arrow) and traction bronchiectasis. There is more than 75% involvement of all five lobes of both the lungs, with severity score of 25/25, which is severe involvement associated with poor prognosis.

Export to PPT

Close

Figure 3:

A 50-year-old male who presented with dyspnea and cough. HRCT coronal section revealed patchy areas of consolidation (black arrow) and subpleural ground glass opacities (white arrow) scattered in bilateral lungs. The CT severity score is 13/25 suggestive of moderate lung involvement.

Export to PPT

Close

Figure 4:

A 69-year-old female who presented with complaints of fever and mild cough. Coronal section of HRCT thorax revealed patchy subpleural ground glass opacities (white arrow) and patchy consolidation (black arrow). The score of involvement is 6/25 which is mild involvement. The occasional subpleural fibrosis is suggestive of concomitant resolution of inflammation and is of prognostic value.

Export to PPT

Age of the patients with mild, moderate, and severe CT severity scores was significantly more than those with normal scores (50 vs. 50 vs. 50 vs. 31, P = 0.0009). Gender was not associated with CT severity score (P = 0.195).

Regarding symptoms, when compared to patients with a normal score, those with mild, moderate, and severe CT severity scores had significantly higher dyspnea (10.39% vs. 67.81% vs. 97.20% vs. 0%). Other symptoms were comparable. When compared to patients with normal score, those with mild, moderate, and severe CT severity score had significantly more cases with diabetes mellitus (16.88% vs. 25.34% vs. 31.78% vs. 9.52%, P = 0.044), hypertension (27.27% vs. 21.23% vs. 32.71% vs. 4.76%, P = 0.026), and obesity (6.49% vs. 8.90% vs. 23.36% vs. 0%, P = 0.0005).

Derangement of blood parameters increased with increasing severity of CT score, as TLC, absolute neutrophil counts, creatinine, ALT, LDH, ferritin, and D-dimer were deranged in significantly more patients with severe scores (53.27%, 62.62%, 60.75%, 85.05%, 90.65%, 97.20%, and 95.35%, respectively), IL-6 and CRP were deranged in significantly more patients with moderate disease (98.18% and 98.63%, respectively).

Increasing severity scores were associated with increased mortality (mild vs. moderate vs. severe: 1.30% vs. 1.37 vs. 15.89%, P < 0.0001). SpO2 was significantly lowest in severe scores followed by moderate, mild, and normal scores (87 vs. 90 vs. 96 vs. 97, P < 0.0001). Duration of NRBM, NIV, HFNC, Venture/face mask, and intubation was also associated with increasing severity scores (P < 0.0001) [Table 7].

On performing multivariate regression, age(years), SpO₂ on admission (room air), intubation, obesity were significant independent risk factors of mortality after adjusting for confounding factors. With the increase in SpO₂ on admission (room air), risk of mortality significantly decreases with an adjusted odds ratio of 0.848 (0.727–0.99). With the increase in age (years) risk of mortality significantly increases with an adjusted odds ratio of 1.075 (1.006–1.148). Patients with intubation, obesity had a higher risk of mortality with an adjusted odds ratio of 4.049 (1.654–9.914), 6.643 (1.052–41.955) respectively [Table 8].

DISCUSSION

COVID-19 is a highly contagious disease that has spread around the world. Disease diagnosis is crucial to disease containment and patient management strategies. However, restricted laboratory facilities and insufficient availability of nucleic acid kits have hampered COVID-19 testing. The WHO suggested that chest imaging is advised for diagnosis of COVID-19 disease on non-availability of the RT-PCR testing, delay in test findings, and in the clinically suspected cases of COVID-19 with negative RT-PCR findings. Clinicians are suggested to work in collaboration with the radiologists for taking appropriate decisions regarding the choice of imaging modality.^[16,17]

We found a significant association of CT severity score with age, dyspnea, diabetes mellitus, hypertension, obesity, TLC, absolute neutrophil count, creatinine, ALT, CRP, LDH, ferritin, IL-6, D-dimer, SpO2 on admission, and duration of NRBM, NIV, HFNC, venture/face mask, and intubation.

Since COVID-19 remains a serious disease with adverse outcomes, the prediction holds the key in terms of CT severity score and other indexes like age, gender, symptoms, and investigative findings.

In the present study, increased age was associated with more severity of disease as mild, moderate, and severe scores had more age than those with normal scores (50 vs. 31 years, P = 0.0009).

This is in trend with the findings by Yang et al.^[18] who found that those with severe scores had significantly more age than patients with mild scores (52.83 vs. 43.70, P = 0.004).

Saeed et al.^[19] also found that normal scans and mild severity were present mostly in patients of 30–39 years. Moderate severity was present in the majority of patients with 40–49 years of age (31.7%) and severe score in the 50–59 years of age group (34.4%).

The elderly patients are more affected by COVID-19 because of several factors such as the stage of the pandemic during the conduction of the study, presence of comorbidities, and availability of healthcare services.

We found that CT severity scores were not associated with the gender of the patients. Similar findings were reported by Yang et al.^[18] who found no significant association between CT severity and gender of the patients. However, Saeed et al.^[19] found that males were more affected by the severe disease (93.4%). Factors behind this were the protective effect of estrogen in females.

In the present study, dyspnea increased with increasing severity of CT scores; however, other symptoms such as fever, cough, and diarrhea were not significantly different in patients with mild, moderate, and severe disease.

This is in accordance with the findings by Yang et al.^[18] as they also found fever (84.52% vs. 83.33%, P = 0.817), cough (67.86% vs. 83.33%, P = 0.191), and expectoration (5.95% vs. 16.67%, P = 0.293) to be similar in patients with mild and severe disease. Xiao et al.^[20] found that chest tightness was the only symptom present in significantly more patients with severe scores than moderate scores (42.5% vs. 23.5, P = 0.002). Other symptoms of COVID-19 were similar.

We found that hypertension, diabetes mellitus, and obesity were significantly associated with severity scores.

This is supported by previous studies that reported risk factors, such as hypertension, diabetes, lung diseases, and coronary artery diseases, is associated with poor prognosis, and worse outcome in the presence of multiple risk factors.^[21,22] However, Saeed et al.^[19] reported that there was no significant correlation of risk factors such as hypertension and diabetes with CT severity scores.

In the present study, blood investigations such as TLC, absolute neutrophil counts, creatinine, ALT, CRP, LDH, ferritin, IL-6, and D-dimer were more deranged in severe cases followed by moderate, mild, and normal severity scores.

As reported by Saeed et al.,^[19] there was a correlation between lymphopenia and increasing CT severity scores. Lymphopenia indicates an inflammatory cytokine storm. Reduced T-cell counts were also found more among patients with severe scores. The serum CRP level was also significantly correlated with CT severity scores.

Yang et al.^[18] reported that the severe cases had significantly lower mean lymphocytes (10⁹/L) (0.68 vs. 1.21) and significantly higher HSCRP (107.91 vs. 11.29) and PCT (0.09 vs. 0.4) (P < 0.001).

Treatment of COVID-19 at an early stage can be considered using CRP as a predictive marker for the likelihood of disease progression.^[23] Similarly, serum ferritin is a vital mediator of immune dysregulation, and its level was closely linked to the severity of the disease. D-dimer likewise can be used as a prognostic indicator, where higher levels are seen in more critical conditions. However, there is a lack of evidence regarding the causal effect. It is not yet clear whether this increase is related to the direct effect of the virus or the systemic inflammatory response

The mortality rate in the present study was 5.70%. Increasing severity scores were associated with increased mortality, with the highest mortality seen in patients with severe scores, followed by moderate and mild scores (P < 0.0001). Those with severe scores had the lowest SpO₂ followed by moderate, mild, and normal scores (87 vs. 90 vs. 96 vs. 97, P < 0.0001). Duration of NRBM, NIV, HFNC, venture/face mask, and intubation was also associated with increasing severity scores (P < 0.0001).

In the study by Saeed et al.,^[19] mortality rate was 3.7%. The death rate was increased in patients with severe severity scores. Other outcomes were similar to the present study as among severe patients, only two patients were not on oxygen support, the nasal cannula was needed in 13.1% patients, face mask in 13.1%, nonrebreather mask in 8.2%, bi-level positive airway pressure (BiPAP)/HFNC in 6.6%, and intubation in 55.7% patients. Oxygen requirement was significantly correlated with CT severity scores (P < 0.0001, r = 0.529). Overall, the need of oxygen increases with the increasing CT severity scores.^[24,25]

The present study found age, SpO₂ on admission (room air), intubation, obesity as the significant independent risk factors of mortality. Among other studies, Iftimie et al.^[26] found that in the first wave, the independent risk factors of mortality were old age and the presence of fever, dyspnea, acute respiratory distress syndrome, diabetes, and cancer; whereas, the risk factors associated with second wave were age, gender, and the presence of acute respiratory distress syndrome and chronic neurological diseases.

In another study, Carbonell et al.^[27] reported that there was no significant difference in ICU mortality between first and second waves according to age subsets except for the subgroup of 61–75 years of age, as decreased unadjusted ICU mortality was observed in the second/third waves (first vs. second/third: 38.7% vs. 34%, P = 0.048).

CONCLUSION

To conclude, CT scans can play an important role in guiding physicians with their management plans and can serve as a predictor of disease severity and outcomes. In patients with COVID-19 infection, the CT severity score is linked to age, diabetes, hypertension, obesity, inflammatory laboratory markers, and oxygen consumption. Higher CT scores showed a significant association with increased mortality, oxygen requirement, and duration of intubation.