RU2742162C1 - Method of assessing the risk of contrast-induced nephropathy in patients with acute myocardical infarction who underwent coronary angiography and percutaneous coronary intervention - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: invention relates to medicine, namely to therapy, and can be used to predict the occurrence of contrast-induced nephropathy (CIN) in patients with acute myocardial infarction (AMI) who underwent coronary angiography (CAG) and percutaneous coronary intervention (PCI). One should identify predictors in patients that are necessary and sufficient to determine the risk of developing CIN: patient age, anemia, the volume of contrast substance per unit of body weight (ml/kg). Then the probability of CIN development is calculated according to the original calculation formula. If CIN development is 0.05 and less than 0.05, the risk of CIN development is low. If CIN development varies from 0.051-0.15, the risk of CIN development is medium. If CIN development is 0.51 and higher than 0.51, the risk of CIN development is high.

EFFECT: invention makes it possible to effectively assess the risk of contrast-induced nephropathy in AMI patients who underwent CAG and PCI.

1 cl, 1 dwg, 4 tbl, 5 ex

Description

The invention relates to medicine, namely to therapy, and can be used to predict the occurrence of contrast-induced nephropathy (CIN) in patients with acute myocardial infarction (AMI) who underwent coronary angiography (CAG) and percutaneous coronary intervention (PCI).

Contrast-induced nephropathy is iatrogenic acute kidney injury (CI-AKI) that occurs after intravascular administration of an iodine-containing X-ray contrast agent, with the exclusion of other alternative causes. For the diagnosis of CIN, it is advisable to use the international classification system for CI-AKI according to the recommendations of the International Working Group Kidney Disease Improving Global Outcomes (KDIGO) with a severity assessment. CIN is diagnosed if one of the criteria is present: an increase in serum creatinine (SCr) by ≥26.5 μmol / L from baseline within 48 hours or an increase in SCr by 1.5 times compared with a known baseline level within a week before the study [http: // nonr .ru / wp-content / uploads / 2013/11 / Clinical guidelines for the prevention, diagnosis and treatment of contrast-induced nephropathy.pdf].

Contrast-induced nephropathy (CIN) is a severe complication of procedures that require the use of X-ray contrast agents. Despite the fact that the risk of renal dysfunction during medical imaging procedures remains relatively small, there are categories of patients in which the risk of developing acute kidney damage is quite high, in particular in patients prone to CAG and PCI [Suma M. V., Anand G ., Vijaya KS et al. Risk scoring system to predict contrast induced nephropathy following percutaneous coronary intervention // Indian Heart. J. - 2014. - No. 66 (5). P. 517-524].

A number of scientific studies are known in which certain predictors have been identified, on the basis of which models for assessing the degree of risk of CIN development are formed, while the set of CIN predictors in various works is variable [Lin K.Y., Zheng W.P., Bei W.J. et al. A novel risk score model for prediction of contrast-induced nephropathy after emergent percutaneous coronary intervention // Int J Cardio. - 2016. - No. 230. - P. 402-412].

There is a way to predict the risk of unfavorable cardiovascular and cerebrovascular events within 6 years after planned endovascular myocardial revascularization [A method for predicting the risk of adverse cardiovascular and cerebrovascular events within 6 years after planned endovascular myocardial revascularization [Electronic resource] / E. Vershinina. O., Lavrov A.G., Repin A.N .; applicant and patentee Federal State Budgetary Scientific Institution "Tomsk National Research Medical Center of the Russian Academy of Sciences". - No. 2017112673; declared 04/12/2017; publ. 01/29/2018, Bul. No. 4]. The method calculates the risk of CIN indirectly and is intended to calculate the occurrence of unfavorable cardiovascular and cerebrovascular events in the long-term (6 years) period, in addition, CAG and PCI were performed routinely for stable coronary heart disease.

A known method for predicting contrast-induced nephropathy [Method for predicting contrast induced nephropathy [Electronic resource] / Mceneaney D., Lamont J .; Randox laboratories Limited PCT / EP 2016/07597; appl. 10/27/2016; publ. 09/05/2018]. This method is applicable in patients with pre-existing chronic kidney disease, that is, with initially compromised kidney function. In addition, the method requires the use of biomarkers.

In the analyzed patent and medical scientific literature there is an analogue [Mehran R, Aymong ED, Nikolsky E, et al. A simple risk score for prediction of contrast-induced nephropathy after percutaneous coronary intervention / J Am Coll Cardiol 2004; 44: 1393-9].

The method is based on the summation of the scores of the risk factors of KIOPP.

The risk of developing contrast-induced acute kidney injury (CI-AKI), depending on the score, is as follows.

The overall probability of a CI-AKI event for the developed score was 13% (from 7.5% at a low [≤5]) to 57.3% at a high [≥16] degree of risk, respectively); the risk of CIOPP increased exponentially with an increase in the total risk score. In the validated set, an increase in the total risk score was closely correlated with the development of CI-AKI (ranging from 8.4% to 55.9% for low and high risk, respectively) [Mehran R, Aymong ED, Nikolsky E, et al. A simple risk score for prediction of contrast-induced nephropathy after percutaneous coronary intervention / J Am Coll Cardiol 2004; 44: 1393-9].

The method has several disadvantages. There are a large number of risk factors for the likelihood of developing contrast-induced nephropathy, some risk factors involve taking a history of the disease, which is not always possible due to the severity of the patient's condition or the presence of cognitive impairments. The criteria for the number of assigned points for each risk factor for the development of CI-AKI are unclear. The method is based on a simple sum of points.

The objective of the invention is to create a simple and effective model for assessing the risk of contrast-induced nephropathy (CIN) in AMI patients who underwent CAG and PCI.

The problem is solved by analyzing the available factors (Table 3) in AMI patients who underwent CAG and PCI, followed by multivariate logistic regression analysis, as a result of which regression coefficients were obtained, on the basis of which a formula for assessing the likelihood of CIN development in patients with AMI who performed CAG and PCI.

The essence of the method lies in the fact that in patients with acute myocardial infarction who need CAG and PCI in a specialized hospital, from among the possible risk factors for the development of CIN, predictors are determined that are necessary and sufficient to determine the risk of developing CIN: the patient's age, anemia, the volume of contrast substances per unit of body weight (ml / kg). Then, using regression analysis, regression coefficients were obtained, on the basis of which a formula for assessing the probability of CIN development in patients with CAG and PCI was derived.

Implementation of the method.

The sample for the presented model consisted of 502 patients with AMI who underwent CAG and PCI, who were treated in the cardiology department of the Regional State Budgetary Institution of Health of the Krasnoyarsk Interdistrict Clinical Hospital of Emergency Medical Care named after N.S. Karpovich (KGBUZ KMKBSMP named after NS Karpovich). CIN was established based on an increase in creatinine levels by more than 26 μmol / L within 48 hours after CAG and PCI with the use of X-ray contrast agents. Another basis for verifying CIN was an increase in creatinine levels of more than 50% within a week after CAG and PCI. The sample did not include patients with stage 5 chronic kidney disease, as well as patients who required X-ray contrast studies for reasons not related to CAG and PCI. Patients who were diagnosed with shock were also excluded from the sample.

Touching upon the issue of the minimum sample size in relation to this study, it should be noted that with regard to the logistic regression method, there is no consensus on the procedure for determining the size of the minimum sample and, accordingly, the so-called. sensitivity of the method. Nevertheless, based on the literature data devoted to this problem, it can be seen that almost all researchers agree on the adequacy of the 500-unit logistic model (Bujang MA, Sa`at N, Sidik T, et al. Sample Size Guidelines for Logistic Regression from Observational Studies with Large Population: Emphasis on the Accuracy Between Statistics and Parameters Based on Real Life Clinical Data, Malays J Med Sci. 2018; 25 (4): 122-130).

The research was carried out as follows.

In our work, a regression model of oil recovery factor was built. Before conducting the regression analysis, the most significant independent variables (predictors) were searched using parametric and nonparametric one-way methods of statistical analysis (Student's test, one-way ANOVA, Mann-Whitney U-test). At the same time, the selected variables were included in the training group, from which, in turn, random generation of 100 samples was carried out, followed by logistic regression. For each of the generated samples, the predictors were evaluated step by step in order to select them for the final model. The final sample included three variables that were statistically significant (p <0.001) in more than 90% of the generated samples.

One of the selected variables ("anemia") is presented as categorical, taking two possible values: "yes" and "no", the other two variables (age and volume of contrast) are displayed in the interval category. To calculate the likelihood of development of oil recovery factors, the obtained regression coefficients were used in combination with the corresponding values of the predictors. In the logistic regression procedure, the odds ratio is calculated with 95% confidence intervals for each predictor.

The adequacy of the resulting model was verified using the Hosmer-Lemeshev test. This test allows you to evaluate the hypothesis about the correspondence of the predicted and observed values of the dependent variable. It is especially relevant for logistic regression models that use interval variables and when there are relatively few observations in the samples. The predicted capabilities of the generated model are determined based on the ROC analysis, which implies the construction of a curve reflecting the dependence of the number of correctly classified positive examples on the number of incorrectly classified examples.

и стандартного отклонения (s). Переменные, не подчиняющиеся закону нормального распределения, отображались в виде медианы (Me) и квартилей (Q1;Q3). Категориальные переменные отображены в процентах с указанием 95% доверительных интервалов (ДИ).Statistical studies and graphical visualization were performed using the software package and the "R" programming language. Variables in which the distribution of data obeyed the normal distribution law are displayed as mean

and standard deviation (s). Variables that do not obey the law of normal distribution were displayed as median (Me) and quartiles (Q1; Q3). Categorical variables are displayed as percentages with 95% confidence intervals (CIs).

The main clinical data of patients with verified CIN, included in the formed sample, are presented in Table 1. The average age of patients in the group with established CIN was 72.2, with 35 patients (63%) over 70 years old. The distribution of male and female patients was approximately the same (Table 1). The average creatinine level on admission was 101.7 μmol / L.

In the study sample of 502 patients, 57 patients who developed CIN were verified. Variables such as age, anemia, and volume of contrast in milliliters per kilogram of body weight were identified as predictors. The variable "anemia" is presented as categorical, the rest of the variables were used as continuous numerical values (interval values). The Hosmer-Lemeshev test with the value χ ² = 3.3 (p = 0.2) confirmed the adequacy of the created model.

Another assessment of the quality of the binary classification - the construction of the ROC-curve - showed a high proportion of objects bearing the selected features (area under the ROC-curve = 0.83) (Fig. 1).

As a result of the multifactorial logistic regression, regression coefficients were obtained (see Table 2), on the basis of which the formula for assessing the likelihood of CIN development in patients with CAG and PCI was derived:

where pKIN is the probability of developing CIN, OC is the volume of contrast in ml / kg, B is the age of full years, A is anemia (if there is - 1, if not - 0), exp is the base of the natural logarithm (2.7183), "7.613" - constant of the equation. The constant is obtained from logistic regression analysis.

The resulting probability value is in the range from 0 to 1. To obtain a percentage, the obtained value should be multiplied by 100.

Thus, on the basis of the multifactorial study, we have created a scale for assessing the degree of CIN risk in patients with CAG and PCI, which includes 3 predictors: age (OR-0.06, p <0.01), anemia (OR-2.23, p <0.01) and volume contrast agent per unit of body weight (ml / kg) (OSh-0.43, p <0.01). Table 4 shows that the largest share in the CIN risk assessment scale and the widest value of the confidence interval belongs to anemia, which is explained by the fact that this variable has only 2 values: "yes" and "no". In the group of patients with established CIN of 52 patients, anemia was diagnosed in 23 patients (44%), while in the group of patients without CIN, the proportion of patients with anemia was 36% (163 patients), while the 95% confidence interval for the development of anemia was 32 % -41%. When assessing the distribution of frequency values in patients with and without CIN, according to the% criterion, no significant differences were found. The lower value of the regression coefficient for the variable “age” is explained by its higher values in the interval value with a mean value of 72.2 ± 9.1; in patients not included in the CIN group, the same indicator was 63.2 ± 11.9. The predictor “contrast volume” had the highest value - 4.6 ml / kg, with the median and quarterly interval being 1.4 (1; 2).

The following scale for assessing the risk of CIN development in patients with CAG and PCI is proposed. The presented gradations of the degrees of risk are displayed on the basis of other works devoted to assessing the degree of risk of developing CIN (Mehran R, Aymong ED, Nikolsky E, et al. A simple risk score for prediction of contrast-induced nephropathy after percutaneous coronary intervention / J Am Coll Cardiol 2004 ; 44: 1393-9).

≤5% - Низкая степень риска,

≤5% - Low risk,

5,1%-15% - Средняя степень риска,

5.1% -15% - Medium risk,

15,1%-50% - Высокая степень риска,

15.1% -50% - High degree of risk,

>50% - Очень высокая степень риска.

> 50% - Very high risk.

Clinical examples are given in retrospect.

Clinical example 1. Patient G., 59 years old, weighing 90 kg, medical history 26146, was hospitalized in the cardiology department of the KGBUZ KMKBSMP them. N.S. Karpovich with a diagnosis of ischemic heart disease, acute anterior myocardial infarction from 19.10.15 to 29.10.15, blood pressure on admission - 210/128 mm Hg. Art. During the entire stay of the patient in the hospital, the lower blood pressure values corresponded to 115/70 mm Hg. Art., the upper indicators are 140/80 mm Hg. Art. Emergency coronary angiography was performed with the introduction of 50 ml of contrast agent "Ultravist 370", which in recalculation per unit of body weight was 0.5 ml / kg. In the general analysis of blood at admission, the patient's hemoglobin level was 127 g / l, erythrocytes - 4.6 × 1012 / l. In the biochemical blood test, the creatinine level was 83 μmol / l. In the future, the patient's serum creatinine levels remained unchanged. Retrospectively, on the basis of the obtained formula, we assessed the degree of risk of developing CIN of the patient upon admission:

The resulting value is multiplied by 100 and, accordingly, we get 3.2%, which corresponds to a low degree of risk.

Clinical example 2. Patient B., 76 years old, weighing 55 kg, medical history 78016, was hospitalized in the cardiology department of the KGBUZ KMKBSMP them. N.S. Karpovich with a diagnosis of ischemic heart disease, acute anterior myocardial infarction from September 24, 2017 to October 4, 2017. Blood pressure on admission was 140/72 mm Hg. Art. During the entire stay of the patient in the hospital, the lower blood pressure values corresponded to 110/60 mm Hg. Art., the upper indicators - 130/75 mm Hg. Art. Emergency coronary angiography was performed with the introduction of 50 ml of contrast agent "Ultravist 370", which in recalculation per unit of body weight was 0.9 ml / kg. In the general analysis of blood at admission, the patient has a hemoglobin level of 123 g / l, erythrocytes - 4.2 × 1012 / l. In the biochemical analysis of blood, the creatinine level was 84 μmol / L. In the future, the patient showed lower levels of serum creatinine. Retrospectively, on the basis of the obtained formula, we assessed the degree of risk of developing CIN of the patient upon admission:

The resulting value is multiplied by 100 and, accordingly, we get 11.4%, which corresponds to the average degree.

Clinical example 3. Patient T, 76 years old, weighing 70 kg, medical history 5460, was hospitalized in the cardiology department of the KGBUZ KMKBSMP them. N.S. Karpovich with a diagnosis of ischemic heart disease, acute anterior widespread myocardial infarction from 01/09/2015 to 01/21/2015. On admission, emergency coronary angiography was performed with the introduction of 250 ml of contrast agent, Ultravist 370, which in terms of body weight was 3.6 mg / kg, PCI of the anterior interventricular branch of the left coronary artery. In the general blood test, the hemoglobin level was 99 g / l, erythrocytes - 3.1 × 012 / l. Blood pressure on admission is 200/105 mm Hg. Art. During the entire stay of the patient in the hospital, the lower blood pressure values corresponded to 112/70 mm Hg. Art., the upper indicators are 143/72 mm Hg. Art. In the biochemical blood test, the creatinine level was 81 μmol / L. After 48 hours, the creatinine level was recorded - 118 μmol / L. Contrast-induced nephropathy was verified in the patient based on the available criteria. Retrospectively, on the basis of the obtained formula, we assessed the degree of CIN development of the patient upon admission.

The resulting value is multiplied by 100 and, accordingly, we get 48.5%, which corresponds to a high degree of risk.

Clinical example 4. Patient S., 88 years old, weighing 70 kg, medical history 28133, was hospitalized in the cardiology department of the KGBUZ KMKBSMP them. N.S. Karpovich with a diagnosis of ischemic heart disease, acute repeated anterior widespread myocardial infarction from 11/15/2015 to 12/03/2015. Upon admission, emergency coronary angiography was performed with the introduction of 150 ml of contrast agent, Ultravist 370, which in terms of body weight was 2.1 ml / kg, PCI of the anterior interventricular branch of the left coronary artery. In the general blood test, the hemoglobin level was 78 g / l, erythrocytes - 2.62 × 1012 / l. Arterial pressure on admission is 140/90 mm Hg. Art. During the entire stay of the patient in the hospital, the lower blood pressure values corresponded to 110/70 mm Hg. Art., the upper indicators are 145/80 mm Hg. Art. In the biochemical analysis of blood, the creatinine level was 89 μmol / L. After 48 hours, the creatinine level was recorded - 124 μmol / L. On November 17, 2015, the patient developed anginal pains, to exclude stent thrombosis, repeated coronary angiography was performed with the introduction of 50 ml of contrast agent, Ultravist 370, which in terms of body weight was 0.7 ml / kg. Subsequently, the serum creatinine values decreased slightly and by the 12th day after the administration of the iodine-containing X-ray contrast agent was 103 μmol / L. Contrast-induced nephropathy was verified in the patient based on the available criteria. Retrospectively, on the basis of the obtained formula, we assessed the degree of CIN development of the patient upon admission.

The resulting value is multiplied by 100 and, accordingly, we get 51.7%, which corresponds to a very high degree of risk.

Clinical example 5. Patient R., 77 years old, weighing 75 kg, medical history 32382 / C2020, was hospitalized in the cardiology department of the KGBUZ KMKBSMP them. N.S. Karpovich with a diagnosis of ischemic heart disease, acute Q wave myocardial infarction with the capture of the anterior-lateral wall of the left ventricle, cop. chronic kidney disease stage C3a from 05/27/2020 to 06/04/2020. Blood pressure on admission is 160/80 mm Hg. Art. During the entire stay of the patient in the hospital, the lower blood pressure values corresponded to 115/62 mm Hg. Art., the upper indicators are 135/75 mm Hg. Art. Was performed emergency coronary angiography with the introduction of 100 ml of contrast agent "Iomeron 400", which in terms of per unit of body weight was 1.3 ml / kg. In the general analysis of blood at admission, the patient has a hemoglobin level of 147 g / l, erythrocytes - 4.8 × 1012 / l. In the biochemical blood test, the creatinine level was 105.9 μmol / l. Subsequently, the patient's serum creatinine levels decreased to 89.8 μmol / L. Based on the obtained formula, we assessed the degree of risk of developing CIN of the patient upon admission:

The resulting value is multiplied by 100 and, accordingly, we get 13.7%, which corresponds to an average degree of risk. Additionally, taking into account the concomitant pathology in the form of chronic kidney disease stage C3a, it was decided to carry out preventive measures for CIN. Before CAG, a physiological solution of sodium chloride was injected in a volume of 150 ml intravenously, during CAG, 200 ml intravenously, and 400 ml intravenously, after CAG. After CAG and PCI, the patient did not develop CIN, despite the average risk of developing CIN and concomitant pathology, due to timely preventive measures. At discharge, the serum creatinine levels decreased to 89.8 μmol / L.

The presented scale for assessing the likelihood of CIN development in patients with acute myocardial infarction who underwent CAG and PCI will help to choose adequate methods for the prevention and treatment of CIN.

The proposed method allows in patients with acute myocardial infarction after coronary angiography and percutaneous coronary intervention in a medical institution with a high sensitivity equal to 83%, which corresponds to the ROC curve (Fig. 1), to determine the likelihood and degree of risk of developing CIN, which will increase the possibilities for reducing the likelihood of developing CIN through the use of appropriate medical procedures.

The result of the work corresponds to the goal and objectives. The method is simple, since it contains only three predictors necessary and sufficient to determine the risk of CIN development: patient age, anemia, volume of contrast medium per unit of body weight (ml / kg). Effective, as a result of the use of a developed simple and understandable formula that can be integrated into any calculator and highly sensitive (up to 83%) in comparison with analogues. Thus, our method can be recommended for use in clinical practice.

Claims (9)

A method for predicting the risk of developing contrast-induced nephropathy in patients with acute myocardial infarction who underwent coronary angiography and percutaneous coronary intervention, which consists in identifying predictors of CIN development: patient's age in years, volume of contrast medium per unit of body weight ml / kg and presence of or the absence of anemia, determine their values in the patient, and then calculate the likelihood of developing CIN using the formula

, Where rKIN is the likelihood of developing CIN; OK - volume of contrast in ml / kg; B - age of full years; A - anemia, if there is - 1, if not - 0; exp - base of natural logarithm 2.7183; with a pKIN value of 0.05 and below, a low degree of CIN development is predicted, with a pKIN value of 0.051-0.15, an average degree of CIN development is predicted, with a pKIN value of 0.151-0.5, a high degree of CIN development is predicted, pKIN values of 0.51 and above indicate a very high risk of CIN development.

Images