RU2488347C1 - Method of estimating probability of mural thrombosis of right heart chambers - right atrium and right ventricle - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: invention relates to field of medicine, namely to cardiology. Collection of anamnesis, clinical and echo0cardiographic examination of 18-year old over 18 year old patients are carried out. In estimation of thrombus formation probability taken into account are: patient's height, body weight, presence/absence of inflammatory process in patient's organism, presence/absence of diabetes mellitus, presence of liquid in pericardium, right atrium dilation, thickness of right ventricle wall in millimetres, presence of signs of post-infarction cardiosclerosis of left ventricle apex. Probability of presence of mural thrombus formation/mural thrombosis in right parts of heart is estimated on the basis of obtained data.

EFFECT: method makes it possible to improve diagnostics of thrombus formation in right chambers of heart in adult people and determine group of patients with high risk of thrombi presence in right heart chambers, who need urgent preventive measures to prevent thromboembolism into pulmonary artery and lethal outcome.

3 tbl

Description

The invention relates to medicine, namely to cardiology. Pulmonary thromboembolism (pulmonary embolism) is the third leading cause of death after heart attack and stroke [1]. Despite the recommendations of the European Society of Cardiology on pulmonary embolism issued in 2008 and adopted in 2009, the Russian clinical guidelines for the diagnosis, treatment, and prevention of venous thromboembolic complications, this problem has not yet been studied [2, 3]. It is well known that the main source of thrombosis in pulmonary embolism is the veins of the lower extremities. However, a considerable part of the cases of thrombosis is associated with other branches of the inferior vena cava, the superior vena cava and the heart chambers, and most researchers put the heart chambers in second place [2], as in the study of V. Drozdov. in 25.6% of cases of sources of thrombosis were right heart cavities [4]. It should be noted that the presence of blood clots in the right cavities of the heart is extremely dangerous from the point of view of the immediate prognosis of the patient. According to the European Society of Cardiology, mortality in the presence of a mobile blood clot in the right heart in the absence of treatment is 80-100% [2, 5]. The prevalence of blood clots in the right sections under random sampling is approximately 4% [6]. In patients with pulmonary embolism, it increases to 7-18% and is associated with higher levels of early mortality. The presence of such blood clots requires immediate treatment [5, 7, 8]. Therefore, the assessment of the likelihood of parietal thrombi in the right chambers of the heart is very important for determining further tactics of patient management. The most accurate methods in this regard are magnetic resonance imaging and coronoventriculography, but they are not used in everyday clinical work, they need good evidence. In addition, the disadvantages of these methods include the following: 1) They are quite expensive; 2) They have limited availability (carried out only in large specialized medical centers); 3) Coronary ventriculography refers to invasive procedures and requires special training.

The research methods used in routine clinical practice do not accurately determine the presence of blood clots. The sensitivity of transthoracic echocardiography (echocardiography) for the detection of intraventricular thrombi is 70-98%. However, its sensitivity to detecting blood clots in the left atrium is 39-63% according to various sources, and in the right one even lower [9, 10]. The diagnostic value for detecting blood clots in the ears of the atria is minimal and does not reach 20% [9]. It should be emphasized that transthoracic echocardiography in patients with pulmonary embolism is often even less informative than in the general population, since they often have obesity, chronic lung diseases, severe tachycardia, which significantly impede visualization.

Transesophageal Echocardiography (sensitivity - 92-100%, specificity - 98-100%) is a high-precision method for detecting blood clots in heart cavities [9, 10]. However, it also has limitations:

1. The complex structure of the atrial ears, the horizontal location of the heart in the chest, severe obesity significantly impede the full visualization.

2. The moderate and severe condition of the patient in most cases makes conducting a transesophageal Echo-KG problematic or completely impossible. This applies, in particular, to patients with heart failure, tachypne, severe shortness of breath, as well as to patients with massive and submassive pulmonary embolism.

3. Transesophageal echocardiography is also performed only in large specialized medical centers. At the same time, the importance of determining the source of embolism for patients with pulmonary embolism is difficult to overestimate. It is knowledge of the source that allows you to choose the most effective way to treat thromboembolism and the most reliable way to prevent its recurrence.

To predict thromboembolism from the left heart, a modified Birmingham scale (2009) is currently used - the CHA ₂ DS2-VAS _c scale. It was included in the official recommendations of the European Society of Cardiology for the management of patients with atrial fibrillation as the main scale for assessing the risk of thromboembolic complications [11, 12, 13]. Regarding the rules for the clinical prognosis of pulmonary embolism, it is recommended to use the Wells scale and the Geneva scale [2]. However, in none of them thrombosis in the right chambers of the heart is considered as a risk factor for pulmonary embolism, primarily due to the difficulty of its in vivo diagnosis.

In the analyzed medical and patent literature, an adequate prototype of a statistical model that allows us to assess the likelihood of patients with parietal thrombosis of the right heart chambers was not found.

The objective of the invention is the development of a method for assessing the likelihood of parietal thrombosis in the right cavities of the heart.

The problem is solved by taking into account the following parameters characterizing the patient’s condition: body weight, height, pericardial fluid, dilatation of the right atrium, wall thickness of the right ventricle, scar of the apex of the left or right ventricle, concomitant inflammatory process, concomitant diabetes mellitus.

Such indicators as fluid in the pericardium (yes / no), dilatation of the right atrium (yes / no), wall thickness of the right ventricle, scar of the apex of the left ventricle (yes / no), can be determined by ECHO-CG data even when visualization is far from optimal [10].

The assessment of the probability of parietal thrombosis in the right cavities of the heart (not including the right ear) is determined as follows.

$$P=\frac{e^Z}{\mathrm{one}+e^Z},$$

where P is the probability of belonging to a group with the presence of parietal thrombi in the right heart;

Z is the calculated regression function;

e is the base of the natural logarithm (e = 2.7183).

The calculated regression function is obtained by summing the individual coefficients, each of which corresponds to a certain parameter, the reliability of the influence of which on the development of parietal thrombosis is confirmed by the method of step-by-step logistic regression [14, 15]:

Z = a + β ₁ X ₁ + β ₂ X ₂ + β ₃ X ₃ + β ₄ X ₄ + β ₅ X ₅ + β ₆ X ₆ + β ₇ X ₇ + β ₈ X ₈ ,

where a is a free term of the equation, a constant.

a-21.5118;

X ₁ - body weight in kilograms,

X ₂ - growth in centimeters,

X ₃ - the presence / absence of fluid in the pericardium according to ECHO-KG:

1 - is; 2 - no;

X ₄ - the presence / absence of dilatation of the right atrium according to echocardiography:

1 - is; 2 - no;

X ₅ - wall thickness of the right ventricle in centimeters according to echocardiography,

X ₆ - the presence / absence of signs of postinfarction cardiosclerosis of the apex of the left ventricle according to echocardiography:

1 - is; 2 - no;

X ₇ - the presence / absence of a concomitant inflammatory process:

1 - is; 2 - no;

X ₈ - the presence / absence of concomitant diabetes mellitus:

1 - is; 2 - no;

β ₁ , β ₂ , β ₃ , β ₄ , β ₅ , β ₆ , β ₇ , β ₈ are the coefficients of the variables X ₁ , X ₂ , X ₃ , X ₄ , X ₅ , X ₆ , X ₇ , X ₈ , relevant:

β ₁ = 0.0801;

β ₂ = (- 0.1596);

β ₃ = 2.5418;

β ₄ = (- 3.9178);

β ₅ = 12.5558;

β ₆ = (- 4.58);

β ₇ = (- 3.9849);

β ₈ = 5.1543.

With a value of P≥0.56, the patient has a high probability of the presence of parietal thrombosis in the right chambers of the heart. The calculated coefficients of the logistic regression equation are presented in table 1.

Table 1 Regression Function Coefficients Indicator Coefficient Standard Error Coefficient χ ² Walda Significance Reached Free member of the equation 21,5118 10,6404 4.0873 0.0432 Body mass 0.0801 0,0642 6.1727 0.0130 Height -0.1596 0,0329 5,9231 0.0149 Pericardial fluid 2,5418 1,0402 5,9707 0.0145 Dilation of the right atrium -3.9178 1,6456 5,6681 0.0173 wall thickness of the right ventricle 12,5558 5,2195 5.7867 0.0161 scar of the apex of the left ventricle -4.69 1,6604 7,9784 0.0047 concomitant inflammatory process -3.9849 1,5711 6,4333 0.0112 concomitant diabetes 5.1543 2.1460 5,7687 0.0163 Note: Percentage of correct classification (Percent Concordant) - 93;

The value of the Hosmer-Lemeshov agreement test is 0.5497.

New in the present invention is the selection of the main parameters characterizing the likelihood of blood clots in the right chambers of the heart, the formula for assessing the likelihood of parietal clots in the right chambers of the heart and the specific value p≥0.56, which allows the patient to be assigned to the group with a high probability of the presence of parietal clots in right chambers of the heart, which allows you to choose the most effective way to treat the patient (in this case, thrombolysis or thrombectomy). The formula is a way of assessing the likelihood of the presence of just parietal thrombosis in the right cavities of the heart and is not intended to assess thrombosis in the right ear of the heart.

The task was to identify factors that significantly affect the development of parietal thrombosis of the right heart chambers and to develop a mathematical model to assess the likelihood of blood clots in the right heart chambers.

In this regard, we analyzed the data of autopsy reports of patients (653 cases) for whom a pulmonary embolism was detected during postmortem examination (we used the data of autopsy protocols of all prosectoral departments of the city of Tomsk from 2003 to 2010). The study group with parietal thrombosis of the right heart chambers amounted to 157 patients aged 26 to 85 years. Then, a selection of indicators was carried out that can be determined during the patient’s life in general therapeutic, general surgical hospitals.

The following factors were included in the analysis: age; body mass; heart mass fluid in the pleural cavities; ascites; fluid in the pericardium; dilatation of the left atrium, right atrium, left ventricle, right ventricle; wall thickness of the left ventricle, right ventricle; the presence of cicatricial changes in the left ventricle, right ventricle; volume and localization of the scar, if any; the presence of myocardial infarction of the left, right ventricles and its localization; the presence of aneurysm, pathology of the heart valves, cardiac arrhythmias, the location of blood clots in the cavities of the heart, if any, concomitant diseases - arterial hypertension, diabetes mellitus, coronary heart disease, the presence of an inflammatory process. A total of 69 indicators were analyzed. As a result of a step-by-step logistic regression, the most significant characteristics were selected, coefficients were determined, odds ratios were calculated, and a regression function was constructed. The model includes the following indicators:

1) body weight

2) growth

3) fluid in the pericardium (yes / no),

4) dilatation of the right atrium (yes / no),

5) the wall thickness of the right ventricle (yes / no),

6) scar of the apex of the left or right ventricle (yes / no),

7) concomitant inflammatory process (yes / no),

8) concomitant diabetes mellitus (yes / no).

The model was tested on an independent group of deceased patients (n = 21). To assess the quality of the model constructed using logistic regression, we used ROC analysis [16]. The area index under the AUC (Area Under Curve) curve was 0.93. As a result, the sensitivity was 0.87, the specificity was 0.94.

For practical use of the obtained function, we present clinical examples.

Example 1

Patient M., 53 years old, was hospitalized in the general medicine department as an ambulance with complaints of pressing pain behind the sternum of low intensity with radiation to the left shoulder lasting for hours, weakness, shortness of breath with minimal physical exertion, dry cough for 1-2 days.

Clinical diagnosis: CHD: angina pectoris 2 f.k. Atherosclerosis of the coronary arteries (stenosis of the middle third of the anterior descending 60%). Paroxysmal form of atrial fibrillation. Hypertension 3 tbsp. Risk 4.

Complications: Recurrent pulmonary embolism of segmental and lobar branches of the right and left pulmonary arteries, pulmonary bifurcation. Chronic pulmonary heart: right ventricular hypertrophy. Hydrothorax. Ascites. H 2B, f.k. IV by NYHA.

table 2 The indicators taken into account in the calculation indicator value The weight 83 Height 172 Pericardial fluid no (2) Dilation of the cavity of the right atrium of the right ventricle and / or is (1) Right ventricular wall thickness 0.6 Cicatricial changes in the apex of the heart no (2) The presence of an inflammatory process no (2) The presence of diabetes no (2)

We carry out calculations for patient M. using a model created on the basis of logistic regression. First we determine the value of the function Z:

Z = a + β ₁ X ₁ + β ₂ X ₂ + β ₃ X ₃ + β ₄ X ₄ + β ₅ X ₅ + β ₆ X ₆ + β ₇ X ₇ + β ₈ X ₈ .

Substituting the values in the formula, we get:

Z = 21.5118 + (0.0801 · 83) + (- 0.1596 · 172) + 25418 · 2 + (- 3.9178 · 1) + 12.5558 · 0.6 + (- 4.69 · 2) + (- 3.98492) + 5.15432 = 2.3670

Now we substitute the value of the function Z = 2.3670 in the formula:

$$\begin{array}{l}P=\frac{e^Z}{\mathrm{one}+e^Z}=\\ P=\frac{e^Z}{\mathrm{one}+e^Z}=\frac{e^{2.3670}}{\mathrm{one}+e^{2.3670}}=\frac{\mathrm{10,6651}}{\mathrm{one}+\mathrm{10,6651}}=\frac{\mathrm{10,6651}}{11.6651}=0.9143\end{array}$$

According to the results of the model, the value of P = 0.9143> 0.56, therefore, patient M. is very likely to have a parietal thrombus in the right cavities of the heart. This clinical case was fatal. Pathological diagnosis:

CHD: angina pectoris 2 f.k. Atherosclerosis of the coronary arteries (stenosis with / 3 PNA 60%). Paroxysmal form of atrial fibrillation. Hypertension 3 tbsp. Risk 4.

Complications: Parietal thrombosis of the right atrium. Recurrent pulmonary embolism of segmental and lobar branches of the right and left pulmonary arteries, pulmonary bifurcation. Chronic pulmonary heart: right ventricular hypertrophy. Hydrothorax. Ascites. H 2B, f.k. IV by NYHA.

Example 2

Patient N., 65 years old, was hospitalized in the emergency department of cardiology as an ambulance with complaints of a feeling of lack of air, weakness, sweating, severe dizziness, unmotivated decrease in blood pressure. After the examination, the diagnosis was made:

Combined underlying disease:

1. IHD: repeated lower Q myocardial infarction from 09/26/2010

Postinfarction cardiosclerosis 84 and 90 years.

Background: Stage 3 hypertension. Obesity III tbsp.

2. Urolithiasis. Stones of the right kidney. Cystolithotomy and cystostomy from 09/08/10, drainage of the prevesical space 13/09/10.

Complications for the first: acute aneurysm of the posterior wall of the left ventricle. Rupture of the free wall of the left ventricle, hemopericardium, tamponade of the heart, ventricles. Persistent atrial fibrillation. For the second: Recurrent submassive pulmonary embolism. Chronic pyelonephritis, exacerbation, hydronephrosis on the left. CRF stage of compensation. Chronic cystitis, exacerbation. Associated: BPH stage I.

Table 3 The indicators taken into account in the calculation indicator value The weight 140 Height 185 Pericardial fluid is (1) Dilation of the cavity of the right atrium and / or right ventricle no (2) Right ventricular wall thickness 0.5 Cicatricial changes in the apex of the heart no (2) The presence of an inflammatory process no (2) The presence of diabetes no (2)

We will carry out calculations for patient N. using a model created on the basis of logistic regression. First we determine the value of the function Z:

Z = a + β ₁ X ₁ + β ₂ X ₂ + β ₃ X ₃ + β ₄ X ₄ + β ₅ X ₅ + β ₆ X ₆ + β ₇ X ₇ + β ₈ X ₈ .

Substituting the values in the formula, we get:

Z = 21.5118 + (0.0801 · 140) + (- 0.1596 · 185) + 2.5418 · 1 + (- 3.9178 · 2) + 12.5558 · 0.5 + (- 4, 69.2) + (- 3.9849-2) + 5.1543.2 = -2.8573

Now we substitute the value of the function Z = - 2.8573 in the formula:

$$\begin{array}{l}P=\frac{e^Z}{\mathrm{one}+e^Z}=\\ P=\frac{e^Z}{\mathrm{one}+e^Z}=\frac{e^{-2.8573}}{\mathrm{one}+e^{-2.8573}}=\frac{\mathrm{0,05742}}{\mathrm{one}+\mathrm{0,05742}}=\frac{\mathrm{0,05742}}{\mathrm{1,05742}}=\mathrm{0,0543}\end{array}$$

According to the results of the model, the value of P = 0.0543 <0.56, therefore, patient N. has a low probability of a parietal thrombus in the right cavities of the heart. This clinical case was fatal.

Pathological diagnosis:

Combined underlying disease: 1. Urolithiasis. Bladder stone. Cystolithotomy, cystostomy, drainage of the prevesical space from 09/13/2010 2. Coronary heart disease. Acute, repeated transmural anteroseptal, posterior subendocardial left ventricular myocardial infarction. PEAKS (84, 90 years). Stenosing atherosclerosis of the coronary arteries. Atrial fibrillation.

Complications: Pelvic vein thrombosis. Tela. Postinfarction scars of the kidneys, spleen, subshell cysts of the occipital lobe of the right hemisphere of the brain, right amygdala of the cerebellum. Relative insufficiency of TC, MK. Circulatory disorder. Cerebral edema. Pulmonary edema.

Background: Hypertension III Art. Obesity III tbsp. Concomitant diseases: Chronic bronchitis. Emphysema. Pneumofibrosis. Atherosclerosis of the aorta, iliac arteries, type 4-5. Gall bladder cholesterosis. Chronic pyelonephritis. Hydronephrosis of the left kidney. Chronic cystitis Chronic duodenal ulcer. BPH.

We propose using the formula we derived in clinical practice to assess the likelihood of a patient having parietal blood clots in the right chambers of the heart. Depending on the results obtained, decide which option for further examination and medical care will be optimal for him. Proposed as an invention, the method will allow:

1. To increase the accuracy of diagnosis of parietal thrombi in the right chambers of the heart. This will help to choose the optimal therapeutic tactics in such patients, and in some cases prevent new episodes of pulmonary embolism, that is, positively affect (increase) their survival.

2. Conduct more expensive studies aimed at a group of patients, where the likelihood of blood clots is highest, ie save material resources of both citizens and medical institutions.

List of literary sources

1. Ruksin V.V. thrombosis in cardiology practice. Ed. 2nd, rev. and add. St. Petersburg: Nevsky dialect, Moscow: BINOM, 2001-125 p.

2. Torbicki A., Perrier A., Konstantinides S. et al. Guidelines on the diagnosis and management of acute pulmonary embolism. The Task Force for the Diagnosis and Management of Acute Pulmonary Embolism of the European Society of Cardiology (ESC). European Heart Journal 2008; 29 (18): 2276-2315

3. Russian clinical guidelines for the diagnosis, treatment and prevention of venous thromboembolic complications. Phlebology 2010; 1: 1-37.

4. Drozdov V.F. Comparative characteristics of new cases of thromboembolism of the branches of the pulmonary artery in hospitals in Tomsk in 2003-2007 and the causes of diagnostic errors. Abstract. dis. Cand. honey. sciences. - Tomsk, 2008 .-- 35 p.

5. Chartier I., Bera J., Delomez M., Asseman P., Beregi JP., Bauchart JJ. Et al. Freefloating thrombi in the right heart: diagnosis, management, and prognostic indexes in 38 consecutive patients. Circulation 1999; 99: 2779-2783.

6. Torbicki A., Galie N. .. Covezzolli A., Rossi E., Rosa M., Goldboher SZ. Right heart thrombi in pulmonary embolism: results from the international Cooperative Pulmonary Embolism Registry. J Am Coil Cardiol 2003; 41: 2245-2251

7. Casazza F., Bongarzoni A., Centonze F., Morpurgo M. Prevalence and prognostic significance of right-sided cardiac mobilethrombi in acute massive pulmonary embolism. Am J Cardiol 1997; 7: 1433-1435.

8. Capoutot L, Naseyrollas P, Meez D, Maes D, Maillier B, Jenesseaux C, et al. Freefloating right heart thrombi and pulmonary embolism: diagnosis, outcome and therapeutic management. Cardiology 1996; 87: 169-174.

9. Zateyshchikov D.A., Zotova I.V., Dankovtseva E.N., Sidorenko B.A. Thrombosis and antithrombotic therapy for arrhythmias. Moscow, Practice, 2011, S.118-125.

10. Schiller N., Osipov M.A. Clinical echocardiography. Moscow, Practice, 2005, 344C.

11. Camm AJ, Kirchhof P, Lip GYH et al. Guidelines for the management of atrial fibrillation. The Task Force for the Management Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010: 31 (19): 2369-2429.

12. Hart RG, Peartce LA, McBride R et al. Factors associated with ischemic stroke during aspirin therapy in atrial fibrillation: analisis of 2012 participents in the SPAF I-III clinical trails. The Stroke Prevention in Atrial Fibrillation (SPAF) Intervestigators. Stroke 1999, 30: 1223-1229.

13. Stroke Prevention in Atrial Fibrillation Intervestigators. Risk factors for thromboembolism during aspirin therapy in patients with atrial fibrillation: the Stroke Prevention in Atrial Fibrillation Study. J Stroke Cerebrovasc Dis: 19995: 5: 147-157.

14. Rebrova O.Yu. statistical analysis of medical data. Application of the STATISTICA application package. Moscow, Media Sphere, 2006, 305 S.

15. Petri A., Sabin K. Visual medical statistics. Per. from English under the editorship of V.P. Leonova, Vol. 2. Moscow, GOETAR-Media, 2010, 168 C.

16.Gerald van Belle, Lloyd D. Fisher, Patrick J. Heagerty, Thomas Lumley. Biostatistics. A Methodology for the Health Sciences. Wiley, 2004 .-- 888 p.

Claims (1)

A method for assessing the likelihood of parietal thrombosis of the right heart chambers - the right atrium and the right ventricle in an adult population, characterized by taking into account body weight, height, the presence / absence of an inflammatory process in the patient's body, the presence / absence of diabetes mellitus, according to ultrasound data - the presence / absence of fluid in the pericardium the presence / absence of dilatation of the right atrium, the wall thickness of the right ventricle, the presence / absence of signs of postinfarction cardiosclerosis of the apex of the left ventricle; the probability of parietal thrombosis in the right cavities of the heart is determined as follows; determine the function Z by the formula:
Z = a + β ₁ X ₁ + β ₂ X ₂ + β ₃ X ₃ + β ₄ X ₄ + β ₅ X ₅ + β ₆ X ₆ + β ₇ X ₇ + β ₈ X ₈ ,
where a is the free term of the equation,
a = 21.5118;
X ₁ - body weight, kg
X ₂ - height, cm
X ₃ - the presence / absence of fluid in the pericardium according to echocardiography:
1 - is; 2 - no;
X ₄ - the presence / absence of dilatation of the right atrium according to echocardiography:
1 - is; 2 - no;
X ₅ - wall thickness of the right ventricle, cm, according to echocardiography,
X ₆ - the presence / absence of signs of postinfarction cardiosclerosis
tops of the left ventricle according to echocardiography:
1 - is; 2 - no;
X ₇ - the presence / absence of a concomitant inflammatory process:
1 - is; 2 - no;
X ₈ - the presence / absence of concomitant diabetes mellitus:
1 - yes; 2 - no;
β ₁ , β ₂ , β ₃ , β ₄ , β ₅ , β ₆ , β ₇ , β ₈ are the coefficients of the variables X ₁ , X ₂ , X ₃ , X ₄ , X ₅ , X ₆ , X ₇ , X ₈ , relevant:
β ₁ = 0.0801;
β ₂ = (- 0.1596);
β ₃ = 2.5418;
β ₄ = (- 3.9178);
β ₅ = 12.5558;
β ₆ = (- 4.69);
β ₇ = (- 3.9849);
β ₈ -5.1543;
$$P=\frac{e^Z}{\mathrm{one}+e^Z},$$

where P is the probability of belonging to a group with the presence of parietal thrombi in the right heart;
Z is the calculated regression function;
e is the base of the natural logarithm (e = 2.7183),
and the estimated regression
and at a value of P≥0.56, the likelihood of parietal thrombosis of the right heart chambers is assessed as high.