Predictors of lethal outcomes in severe cases of a new coronavirus infection COVID-19

Full Text

Background

COVID-19, a novel coronavirus infection (NCI), primarily damages lung tissue. In some cases, this damage is attributed to the development of mixed viral–bacterial pneumonia, which worsens the disease and increases the risk of mortality [1, 2]. In this context, the risk of life-threatening NCI is particularly relevant for older patients who often have multiple chronic conditions, such as arterial hypertension (13%–17%), diabetes mellitus (5%–35%), cardiovascular diseases (3%–4%), chronic lung diseases (2%), and oncologic pathology (0.5%–3%) [3, 4].

Moreover, contracting pneumonitis or pneumonia at a young age can result in the loss of work capacity, which can have a detrimental effect on professions that require teamwork. Furthermore, having pneumonia can lead to rehabilitation measures, which can further prolong the period of disability coverage for specialists [1, 5–8].

Currently, the scientific community is closely examining the phenomenon of comorbidity in NCI. Garg et al. [9] analyzed the comorbid background of patients hospitalized for NCI in March 2020 and determined that approximately 90% had comorbidities. The most common comorbidities were arterial hypertension (49.7%), obesity (48.3%), chronic lung diseases (34.6%), diabetes mellitus (28.3%), and cardiovascular diseases (27.8%). These findings are consistent with those of Deng et al. [10], who reported that 42.3% of COVID-19 patients in China had been diagnosed with diabetes mellitus.

The literature analysis indicates that the identification of predictors of unfavorable outcomes in patients with severe NCI, their comparative assessment, and the prediction of clinical dynamics have not been adequately investigated. This study aims to address this knowledge gap.

The lung lesions observed in COVID-19 patients are characterized by significant blood clotting in the capillaries of interalveolar septa and the branches of pulmonary arteries and veins. This clotting is accompanied by erythrocyte sludge, fresh fibrin, organizing blood clots, and intrabronchial, intrabronchiolar, and intraalveolar hemorrhages, which can lead to hemoptysis. Most observations show an alveolar hemorrhage syndrome, which can progress to the formation of hemorrhagic infarcts (although true hemorrhagic infections are not uncommon), distinguishing the pulmonary changes in COVID-19 from those previously identified in influenza A/H1N1 and other coronavirus infections. Importantly, microcirculatory disorders that develop in NCI play a crucial role in increasing the severity of tissue hypoxia, which in turn increases the risk of lethal outcomes, particularly in patients with unfavorable comorbidities [1, 11–13].

The mechanism of action of succinic acid preparations, such as meglumine sodium succinate, is attributed to the changes in the transport of mediator amino acids, catalytic effect on oxygen utilization in the Krebs cycle, activation of aerobic metabolism, restoration of the redox state of mitochondria, and increased use of fatty acids, ketone bodies, glucose, and lactate.

Under hypoxic conditions, the tricarboxylic acid cycle slows down and anaerobic glycolysis is activated. The restoration of the energy exchange process becomes dependent on succinate, including its exogenous contribution, which in turn stimulates the oxidation of succinate oxidase, leading to the restoration of its consumption in the respiratory chain of mitochondria and the increase in the activity of the antioxidant function of glutathione. Succinic acid also enhances the activity of the protein component of the body’s antioxidant system.

These effects of succinic acid work together to prevent reperfusion injury when oxygen levels are imbalanced [14–17].

This study aims to identify predictors of unfavorable outcomes in severe cases of COVID-19 infection.

Materials and methods

Medical record data from the mono-patient hospital at the Municipal Clinical Hospital No. 1 of Chita, Ministry of Health of the Zabaikalsky Krai, from June 2020 to August 2020 were analyzed. This study was designed as a retrospective observational study.

This study received approval from the local ethical committee of the Chita State Medical Academy on October 20, 2020 (Protocol No. 103).

The inclusion criteria for patient records were as follows:

1. Patients with severe COVID-19 who were treated in the intensive care unit (ICU);
2. Presence of respiratory support (e.g., O2 insufflation and invasive/noninvasive artificial ventilation);
3. Patient’s age over 30 years.

The noninclusion criteria for patient records were as follows:

1. Infection caused by human immunodeficiency virus or immunodeficiency of another genesis;
2. Hemodialysis before the development of COVID-19;
3. Acute cerebral circulatory failure or acute myocardial infarction before the development of COVID-19;
4. Pregnancy;
5. Alcoholic delirium.

A total of 46 inpatient records containing information on the treatment of severe forms of NCI were selected from the archives of the medical organization, based on both groups of criteria.

All patients were treated according to the current version of the temporary guidelines of the Ministry of Health of the Russian Federation for the prevention, diagnosis, and treatment of new coronavirus (COVID-19) infection [1]. Moreover, clinical recommendations were followed for any other pathology requiring treatment during hospitalization.

All patients included in this study were diagnosed with comorbidities in addition to the severe course of COVID-19. The most common comorbidities were encephalopathy of mixed genesis, diabetes mellitus, ischemic heart disease, arterial hypertension, and alimentary–constitutional obesity. The Charlson comorbidity index was used to assess the degree of influence of somatic pathology on patient health status. This index was developed for the prognosis of patients with multiple medical conditions [18]. This index is a point system used to assess age and the presence of certain comorbidities, reflecting the severity of the patient’s somatic condition.

The detailed characteristics of the patients are shown in Table 1.

 

Table 1. Gender and age composition and comorbidity status of patients from the sample of records (n = 46)

Indicator	Meaning	Statistics
Hospitalization data
Days in the ICU	5.0 [4.0, 7.0]	Median [Q25, Q75]
Days of hospitalization	14.5 [9.2, 18.8]	Median [Q25, Q75]
Death	17.4% [8.8%, 31.0%]	% [95CI−, 95CI+]
Gender and age composition
Males	45.7% [32.2%, 59.8%]	% [95CI−, 95CI+]
Age	68.5 [57.0, 76.5]	Median [Q25, Q75]
Comorbidity
Charlson comorbidity index	3.0 [2.0, 5.0]	Median [Q25, Q75]
Cardiovascular disease	54.3% [40.2%, 67.8%]	% [95CI−, 95CI+]
Diabetes mellitus	21.7% [12.1%, 35.8%]	% [95CI−, 95CI+]
Obesity	17.4% [8.8%, 31.0%]	% [95CI−, 95CI+]
Advanced atherosclerosis	17.4% [8.8%, 31.0%]	% [95CI−, 95CI+]
Chronic CNS disease	15.2% [7.3%, 28.5%]	% [95CI−, 95CI+]
Chronic kidney disease	10.9% [4.3%, 23.5%]	% [95CI−, 95CI+]
Chronic RS disease	8.7% [2.9%, 20.9%]	% [95CI−, 95CI+]
Chronic GI disease	8.7% [2.9%, 20.9%]	% [95CI−, 95CI+]
Oncology	6.5% [1.6%, 18.2%]	% [95CI−, 95CI+]

Note: ICU, intensive care unit; 95CI, 95% confidence interval; RS, respiratory system; GI, gastrointestinal tract; CNS, central nervous system.

 

Information about the dynamics of the patient’s condition and the results of laboratory and instrumental studies for the period of the patient’s stay in the ICU were extracted from the medical records. This information included the general blood count with detailed leukocytic formula, standard biochemical parameters, coagulogram, and arterial blood gas analysis.

The receiver operating characteristic (ROC) analysis included the clinical blood count with leukocyte count, acid–base status, and various biochemical parameters, such as aminotransferase activity, glucose concentration, bilirubin and its fractions, inflammatory marker levels, protein metabolism, and renal function parameters. The predictors considered in this study included gender, age, and the Charlson comorbidity index [18], which was specifically developed to assess the prognosis of patients with long-term follow-up. The Charlson comorbidity index is a point system that considers age and the presence of certain comorbidities, reflecting the severity of the patient’s somatic condition.

Statistical processing of the results was conducted using IBM SPSS version 23.0. This study implemented the following statistical processing techniques:

First, parameters with an area under the curve significantly different from 0.5 at a significance level of p < 0.05 and an area under the ROC curve higher than 0.7 (indicating good classification quality) were selected using ROC analysis. The predicted category was a lethal outcome.

Second, the initial values of indicators were transformed into risk factors for unfavorable outcomes using the method of determining cutoff levels by the maximum value of Youden’s j index, which was calculated as the sum of sensitivity and specificity minus one [19].

This study calculated the odds ratio (OR) of the selected indicators to evaluate their impact on the outcome. To account for cells with zero outcomes, 0.5 was added to all cells [20]. OR was considered statistically significant if the 95% confidence interval excluded one.

Results and discussion

According to the results of the analysis, the main risk factors were identified and ranked.

Significant results are presented in Table 2.

 

Table 2. Statistically significant predictors of unfavorable outcomes in severe coronavirus infection

Indicator	AUC (95% CI)	Significance	Cutoff value
Charlson comorbidity index	0,73 (0,53; 0,93)	0,044	4.5 points
Arterial blood anion gap	0,93 (0,82; 1,00)	0,017	32.095 mmol/L*
Fibrinogen	0,84 (0,69; 0,98)	<0,01	4.85 g/L
Aspartate aminotransferase	0,75 (0,57; 0,94)	0,036	47.5 IU/L
Urea	0,76 (0,50; 1,00)	0,031	8.335 mmol/L

Note: AUC, area under the curve; OR, odds ratio; CI, confidence interval. The asterisk (*) denotes the “reverse” risk factor, which indicates that the risk of a lethal outcome increases as the index decreases.

 

Arterial blood anion gap indices (OR = 28.78; p < 0.017) and fibrinogen levels (OR = 22.20; p < 0.01) had the most significant influence on the probability of the lethal outcome, increasing it by more than 20 times in patients with these factors. Urea level (OR = 12.8; p < 0.031), aspartate aminotransferase activity (OR = 7.22; p < 0.036), and Charlson comorbidity index (OR = 5.37; p < 0.044) had a lesser impact on the prognosis.

The obtained results are consistent with modern medical ideas regarding the diagnosis and treatment of such conditions. Studies have shown that comorbidities have a negative impact on the body’s resistance and functional reserves [21]. Moreover, the number of diseases that exacerbate each other increases with age [22], indicating that elderly patients are at a higher risk. The increase in the Charlson comorbidity index among the identified predictors indirectly supports this hypothesis. However, the analysis conducted in this study determined no statistically significant association between patient age and death. Therefore, we conclude that the presence of mutually aggravating diseases has a more significant impact on the outcome than patient age.

An elevated blood fibrinogen level indicates the intensity of the inflammatory process, which in the case of NCI is associated with the development of a “cytokine storm.” This cytokine storm determines the severity of pulmonary damage and increases the risk of further deterioration of the patient’s condition [1, 23]. This cytokine storm is particularly relevant considering the currently discussed hypothesis that “immunothrombosis” is a significant factor in the pathogenesis of severe COVID-19 [24]. Furthermore, fibrinogen can be used as a predictor because its cost is lower than more specific markers of inflammation, such as interleukin-6.

In cases of severe infection, which can lead to intoxication and endothelial dysfunction, tissue hypoxia, cytolysis, and ultimately organ damage with dysfunction, early indicators of a potential complication, such as multiorgan failure, need to be monitored [25–27]. Aspartate aminotransferase, urea level, and anion gap are valuable predictors from both statistical and clinical perspectives [1].

The presence of signs indicating the dysfunction of organs and systems, as well as hypoxia in a patient, warrants the expansion of therapy with metabolic support drugs that enhance cellular energy metabolism, which in turn improves the state of organ and system functions [7, 12, 28–31]. The criteria for prescribing this type of support in patients with NCI are not clearly defined, and data on this topic with an appropriate level of evidence, particularly to assess the efficacy of succinate-containing drugs in patients with different risks of death, are lacking in the literature. However, the intermediate-risk group appears promising for evaluating the effectiveness of metabolic support because of the lower likelihood of irreversibility of the pathological processes inherent in NCI [1, 11–13].

Conclusions

Arterial blood anion gap indices and fibrinogen levels are significant predictors of unfavorable outcomes in severe cases of COVID-19 infection. Blood urea level, aspartate aminotransferase activity, and Charlson comorbidity index have a lesser impact on the probability of a lethal outcome.

 

Authors’ contribution. K.G.Sh., supervising the work and editing; G.A.C., E.V.L., and P.V.P., conducting the research and drafting the article; A.A.L., collecting and analyzing the results and drafting the primary reporting material.
Funding source. This study had no sponsorships.
Conflict of interest. The authors declare no conflict of interest for the presented article.

About the authors

Konstantin G. Shapovalov

Chita State Medical Academy; City Clinical Hospital No. 1

Email: shkg26@mail.ru
ORCID iD: 0000-0002-3485-5176

M.D., D. Sci. (Med.), Prof., Head of Depart., Depart. of Anesthesiology, Resuscitation and Intensive Care

Russian Federation, Chita, Russia; Chita, Russia

Galdan A. Tsydenpilov

Chita State Medical Academy; City Clinical Hospital No. 1

Email: galdan.tsydenpilov@mail.ru
ORCID iD: 0000-0002-9734-8016

Clinical Resident, Depart. of Anesthesiology, Resuscitation and Intensive Care

Russian Federation, Chita, Russia; Chita, Russia

Evgeny V. Lozovsky

Chita State Medical Academy; City Clinical Hospital No. 1

Email: lozjek@gmail.com
ORCID iD: 0000-0003-4474-9366

Clinical Resident, Depart. of Anesthesiology, Resuscitation and Intensive Care

Russian Federation, Chita, Russia; Chita, Russia

Alexander A. Latyshov

Chita State Medical Academy; City Clinical Hospital No. 1

Email: soad3233@gmail.com

Clinical Resident, Depart. of Anesthesiology, Resuscitation and Intensive Care

Russian Federation, Chita, Russia; Chita, Russia

Polina V. Petrova

First St. Petersburg State Medical University named after I.P. Pavlov

Author for correspondence.
Email: apolly2017@yandex.ru
ORCID iD: 0000-0002-2658-9920

student

Russian Federation, St. Petersburg, Russia

References

Supplementary files