Method for predicting covid-19 fatality in individuals with arterial hypertension

Abstract

FIELD: infectology.SUBSTANCE: method used to predict the fatal outcome of COVID-19 in individuals with arterial hypertension. Blood sampling is performed, the level of natriuretic peptide (NTproBNP), oxygen saturation (SpO2), C-reactive protein (CRP) is determined, taking into account the patient’s age. The probability level (P) is estimated using the binary logistic regression model, in which the exponent Logit is determined according to the stated formula. If P is greater than or equal to 0.2957, a fatal outcome is predicted.EFFECT: method provides the ability to obtain a numerical indicator that allows for assessing the individual outcome of COVID-19 in patients with arterial hypertension, identifying the potential risk of a negative outcome of coronavirus infection in patients with arterial hypertension in the early stages of the infectious process, which can be used to subsequently improve the routing and management of patients in order to increase patient survival, by determining the level of NTproBNP, SpO2, CRP in the blood and taking into account the patient's age.1 cl, 1 tbl, 2 ex

Description

[0001] The invention relates to medicine, in particular to infectious diseases, and provides the ability to assess the likelihood of an unfavorable outcome for patients with arterial hypertension (AH) in COVID-19.

[0002] The coronavirus disease 2019 (COVID-19) pandemic declared by the World Health Organization on March 11, 2020, has demonstrated the urgency of the threat of the global spread of new infectious diseases (see Coronavirus disease 2019 (COVID-19) Situation Report / / World Health Organization 2020; February 20, 2020; Huang C., Wang Y., Li X., Ren L., Zhao J., Hu Y., Zhang L., Fan G., Xu J., Gu X., Cheng Z., Yu T., Xia J., Wei Y., Wu W., Xie X., Yin W., Liu M., Xiao Y., Gao H., Guo L., Xie J., Wang G., Jiang L., Cao Z., Jin Q., Wang J., Cao B. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet. 2020; 395 (10223): 497-506; Rodriguez-Nava G., Yanez-Bello MA, Trelles-Garcia DP, Chung CW et al. Performance of the quick COVID-19 severity index and the Brescia-COVID respiratory severity scale in hospitalized patients with COVID-19 in a community hospital setting / / Int. J. Infect. Dis. 2021. Vol. 102. P. 571-576.https: / / doi.org / 10.1016 / j.ijid.2020.11.003).

[0003] The infectious disease caused by coronavirus (CVI) is mild, moderate, or severe (see 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. Vol. 323, No. 13. P. 1239-1242. DOI: https: / / doi.org / 10.1001 / jama.2020.2648).

[0004] COVID-19 affects people of all ages, regardless of gender, race, location, and living conditions. By the end of 2024, more than 700 million cases of COVID-19 and 7 million deaths have been registered worldwide, accounting for one-third of the total number of cases (see the State Report "On the State of Sanitary and Epidemiological Welfare of the Population in the Russian Federation" for the Republic of Sakha (Yakutia) for 2024).

[0005] Meanwhile, according to the World Health Organization (WHO), approximately 1.28 billion adults (aged 30 to 79) worldwide suffer from high blood pressure. Moreover, according to the Russian Ministry of Health and Rospotrebnadzor, the prevalence of hypertension among the adult population in Russia exceeds 40%. In the Republic of Sakha (Yakutia), the prevalence of hypertension increased over 20 years of observation from 30.3% in 2003 to 54.3% in 2022 (see Shalnova S.A., Deev A.D., Balanova Yu.A., Kapustina A.V., Imaeva A.E., Muromtseva G.A., Kiseleva N.V., Boytsov S.A. Twenty-year trends in obesity and arterial hypertension and their associations in Russia / / Cardiovascular therapy and prevention. - 2017. - Vol. 16, No. 4. - P. 4-10).

[0006] Numerous studies conducted during the pandemic have revealed a close link between the unfavorable prognosis of COVID-19 and the presence of concomitant circulatory diseases, especially arterial hypertension (AH). This is due to the high similarity of the SARS-CoV-2 S protein with the angiotensin-converting enzyme type 2 (ACE2), the expression of which is significantly reduced during infection with the virus, which can lead to a worsening of arterial hypertension (see Sofronova S.I., Romanova A.N. Arterial hypertension and some risk factors for its development in the indigenous and immigrant population of Yakutia / / Bulletin of the North-Eastern Federal University named after M.K. Ammosov. Series: Medical Sciences. 2023. No. 3 (32). P. 39-44; Shalnova S.A., Deev A.D., Balanova Yu.A., Kapustina A.V., Imaeva A.E., Muromtseva G.A., Kiseleva N.V., Boytsov S.A. Twenty-year trends in obesity and arterial hypertension and their associations in Russia / / Cardiovascular therapy and prevention. - 2017. – T.16, No. 4. – P. 4-10).

[0007] To determine the severity of the condition of patients with COVID-19, specialized and general clinical scales are used, such as NEWS2, APACHE II, SAPS II, SOFA, 4C Mortality Score, COVID Home Safely Now (CHOSEN) Risk Score for COVID-19, COVID-YKT (see Vechorko V.I., Averkov O.V., Suponeva N.A., Piradov M.A., Zimin A.A., Yusupova D.G., Zaitsev A.B., Grishin D.V., Polekhina N.V., Naminov A.V., Ramchandani NM, Knight SR, Semple MG, Harrison EM Validation of the Russian version of the 4C Mortality Score).

[0008] The prior art includes a method for individually predicting the outcome of the new coronavirus infection COVID-19 according to patent RU No. 2795141 (cl. A61B 5 / 107, G01N 33 / 62, G01N 33 / 68, G01N 33 / 72, published on 04 / 28 / 2023), a method for predicting the outcome of viral pneumonia in COVID-19 (see RU No. 2764002, cl. A61B 5 / 1455, G01N 33 / 62, G01N 33 / 68, G01N 33 / 569, published on 01 / 12 / 2022), a method for predicting a fatal outcome in patients with a severe form of COVID-19 (see RU No. 2780748, cl. C12Q 1 / 68, published 30.09.2022) and others.

[0009] The known solutions examined patients with varying degrees of severity, but did not take into account circulatory diseases.

[0010] The problem that the claimed invention is aimed at solving is to identify the potential risk of a negative outcome of coronavirus infection in patients with hypertension in the early stages of the infectious process.

[0011] The technical result of the invention is to obtain a numerical indicator that makes it possible to evaluate the individual outcome of the course of COVID-19 in individuals with arterial hypertension, which can be used to subsequently improve the routing and management of patients in order to increase the survival rate of patients.

[0012] To solve the problem, a method for predicting the fatal outcome of COVID-19 in individuals with arterial hypertension, including blood sampling, determining the level of natriuretic peptide (NTproBNP), oxygen saturation (SpO2), C-reactive protein (CRP) taking into account the patient's age, is characterized by the fact that the probability level (P) is assessed using the binary logistic regression model P = 1 / (1 + e ∧(−Logit)), in which the Logit exponent is determined by the formula: Logit = 8.4491 + 0.0012 (NTproBNP) − 0.1386 (SpO2) + 0.0482 (Age) − 0.0096 (CRP), where P is the probability of a fatal outcome; NTproBNP is the natriuretic peptide level, pg / ml; SpO2 is the oxygen saturation level, %; Age is the patient's age; CRP is the C-reactive protein level, mg / l, and if P is greater than or equal to 0.2957, a fatal outcome is predicted. The sensitivity of the cutoff point of 0.2957 in relation to assessing a fatal outcome is 72.0%, the specificity is 85.3%.

[0013] Analysis of the features of the claimed solution indicates that the claimed solution meets the criterion of “novelty”.

[0014] Thus, the declared solution is based on assessing the severity of the condition of COVID-19 patients with arterial hypertension by the method of mathematical modeling and forecasting using routine laboratory and clinical data, standard indicators widely available in practical healthcare, and can be implemented to predict a fatal outcome.

[0015] The co-authors, given the high incidence of COVID-19 and the widespread prevalence of arterial hypertension among the population, conducted a study to assess the impact of hypertension on the severity of COVID-19 and the risk of fatal outcomes in hospitalized patients in the Republic of Sakha (Yakutia) (see Sofronova S.I., Romanova A.N. Arterial hypertension and some risk factors for its development in the indigenous and immigrant population of Yakutia / / Bulletin of the North-Eastern Federal University named after M.K. Ammosov. Series: Medical Sciences. 2023. No. 3 (32). Pp. 39-44).

[0016] The declared technical solution for predicting the outcomes of COVID-19 in patients with hypertension is based on a retrospective analysis of 200 cases of the disease hospitalized in the infectious diseases department of the Yakutsk Republican Clinical Hospital from October 2020 to June 2022.

[0017] Upon hospitalization, patients' body mass index, blood oxygen saturation, and a number of laboratory parameters were measured in accordance with medical care standards. A predictive model for the probability of a specific disease outcome was created using binary logistic regression in StatTech v. 4.8.11. The developed model reflects the relationship between disease outcome and several clinical and laboratory parameters in patients with COVID-19 (see table).

[0018] Thus, to develop the most accurate predictive model for mortality, a multivariate logistic regression analysis was conducted on the study results using the backward stepwise selection method. The target variable was a binary outcome (mortality = 1, discharge = 0). The final model included four independent predictive factors demonstrating statistical significance at a p<0.05 level:

[0019] - age (years);

[0020] - oxygen saturation level SpO2(%);

[0021] - NTproBNP (natriuretic peptide) level (pg / ml);

[0022] - CRP (C-reactive protein) level (mg / L).

[0023] The overall statistical significance of the model was confirmed by the high likelihood ratio test (p-value < 0.001).

[0024] Interpretation of odds ratios (OR) showed that age (OR=1.049) is an independent risk factor, increasing the odds of death by 4.9% for each year lived.

[0025] Similarly, NTproBNP (OR=1.001) is a risk factor. Meanwhile, SpO2 is a strong protective factor (OR=0.871), with a one-percent increase in saturation reducing the odds of death by 12.9%. CRP levels (OR=0.990) are also formally associated with a slight reduction in odds, but their clinical effect is insignificant.

[0026] The linear predictor (Logit) used to calculate the probability of death (P) is described by the following equation, where the β coefficients reflect the contribution of each factor:

[0027] Logit=8.4491+0.0012(NTproBNP)−0.1386(SpO2)+0.0482(Adult)−0.0096(SRP)

[0028] P=1 / (1+e ∧ (−Logit))

[0029] Optimal classification threshold (Threshold): 0.2957 (29.57%), above which the risk of death is considered high.

[0030] The discriminatory ability of the model was assessed using the area under the ROC curve (AUC), which was 0.7985, which characterizes the model as having good prognostic efficiency, since in 80% of cases the constructed model is able to correctly rank patients by risk.

[0031] To optimize the balance between sensitivity and specificity, an optimal classification threshold of 0.2957 (using Youden's statistic) was determined. Using this threshold, the model demonstrated high sensitivity of 72.0% and specificity of 85.3%. This demonstrates that the developed model is a reliable and accurate tool for determining the risk of mortality.

[0032] The following clinical cases can be cited as examples of the application of the developed model:

[0033] Example 1. Patient M., 78 years old. Primary clinical diagnosis: coronavirus infection caused by COVID-19, confirmed by laboratory tests. History: the patient has suffered from arterial hypertension since youth. According to examination data during hospitalization, SpO2 was 84%, CRP level was 20.1 mg / L, NTproBNP – 303 pg / mL. Calculation of the probability of a fatal outcome using the constructed model showed:

[0034] Logit = 8.4491 + (0.0012 × 303) - (0.1386 × 84) + (0.0482 × 78) - (0.0096 × 20.1) = 0.75694

[0035] P = 1 / (1 + e^(-0.75694)) = 0.6807 (or 68.07%), which is greater than the calculated threshold of 0.2957.

[0036] Therefore, we determine the probability of death during hospitalization as high. Analysis of the patient's medical history revealed that on the fifth day of hospitalization, she was transferred to the intensive care unit due to worsening respiratory failure and subsequently died. Therefore, the predicted fatal outcome occurred.

[0037] Example 2. Patient D., 63 years old. Primary clinical diagnosis: laboratory-confirmed COVID-19 coronavirus infection. The patient's medical history indicates a history of hypertension for approximately 5 years. According to examination data on the first day of hospitalization, SpO2 on admission was 92%, CrO2 level was 8.0 mg / L, and NTproBNP level was 80 pg / mL. Calculating the probability of a fatal outcome using the constructed model revealed:

[0038] Logit = 8.4491 + (0.0012 × 80) - (0.1386 × 92) + (0.0482 × 63) - (0.0096 × 8.0) = -1.2463

[0039] P = 1 / (1 + e ∧(-(-1.2463))) = 1 / (1 + e ∧ (1.2463))= 0.2233 (or 22.33%), which is less than the threshold value and indicates a low probability of developing a fatal outcome during hospitalization.

[0040] The predicted outcome of the disease was confirmed: the patient was discharged from the infectious diseases department after 17 days of hospitalization with recovery, and the predicted favorable outcome developed.

[0041] Thus, the proposed model makes it possible to determine with a high degree of certainty the likelihood of a fatal outcome from COVID-19 in individuals with arterial hypertension in the early stages of the disease and can be used to select optimal treatment tactics for patients.

[0042] Table

[0043] Characteristics of the relationship between predictors of the prognostic model and the development of a fatal outcome

[0044] Predictor Coefficient β (Log-Odds) Odds ratio (OR) 95% CI for OR R NTproBNP 0,0012 1,001 (1,000, 1,002) 0,003 SpO2 (%) −0,1386 0,871 (0,805, 0,942) 0,001 Age 0,0482 1,049 (1,011, 1,089) 0,011 SRB −0,0096 0,990 (0,982, 0,998) 0,020