MINISTRY OF EDUCATION MINISTRY OF DEFENCEANDTRAINING 108 INSTITUTE OF CLINICAL MEDICINE AND PHARMACY -------------------------------------------------------- BUI DUC THANH STUDY OF THE CHANGES OF SERUM NT-proBNP LEVEL AND THE ASSOCIATION WITH LOW CARDIAC OUTPUT SYNDROME AFTER CORONARY ARTERY BYPASS GRAFTING SURGERY

Specialty: Anesthesiology Code: 62.72.01.46

SUMMARY OF MEDICAL DOCTORAL THESIS Ha Noi – 2020

The work was completed at: 108 INSTITUTE OF CLINICAL MEDICINE AND PHARMACY

Full name of supervisor:

1. Nguyen Hong Son. Assoc. Prof. 2. Nguyen Thi Quy. Assoc. Prof.

The Objections: 1. 2. 3.

The thesis will be protected at the Board of Insitute doctoral thesis evaluation at 108 Institue of Clinical Medicine and Pharmacy At the time: / 2020 Can be found the thesis in:

1. National Library. 2. The library of 108 Institue of Clinical Medicine and Pharmacy.

1 BACKGROUND

function, longer the

Low cardiac output syndrome (LCOS) is a clinical condition caused by a transient decrease in systemic perfusion due to cardiac dysfunction, resulting in an imbalance between supply and demand for cellular oxygen level that le led to metabolic acidosis. LCOS is common in elderly patients, patients with reduced systolic and diastolic left ventricular time of aortic cross-clamp orextracorporeal circulation, re-surgery, valve replacement surgery, and coronary artery bypass grafting (CABG) surgeries.

include

The causes of LCOS including reducing myocardial contraction, etiology of pre-load, and afterload. Factors leading to impair left inflammatory response, ventricular function after LCOS inadequate myocardial anemia, hypothermia, reperfusion damage, cardioprotection, and ventricular surgery. Reduced cardiac output in heart failure after surgery is a common condition accounting for 30% of CABG cases.

In recent years, the role of diuretic peptides (natriuretic peptide) was got attention. Many studieshave shown the role of NT-proBNP (N- Terminal pro-B-Type Natriuretic Peptide) in early diagnosis of heart failure, assessment of severity, evaluation of the efficacy of treatment efficacy prognosis of heart failure. NT-proBNP is also used to determine the factors associated with heart failure.

In Vietnam, the study of NT-pro BNP was mainly internal medicine. From a surgical perspective, there has not been any study about NT-pro BNP in patients with CABG surgery. For theabove reasons,wehavecarriedout the thesis:“Study of the changes of serum NT-proBNP level and the association with low cardiac output syndrome after coronary artery bypass grafting surgery”withthefollowing objectives:

1. Investigate the changes in serum NT-proBNP level in patients undergoing CABG with extracorporeal circulation. 2. Evaluate the association between serum NT-proBNP level and LCOS after CABG.

2 Chapter 1 OVERVIEW 1.1. Coronary artery bypass grafting surgery with extracorporeal circulation.

After coronary artery bypass grafting surgery with extracorporeal circulation, the patient underwent the recovery process of an important organ such as a cardiopulmonary organ with an average time was 2-7 days.This period was called the early stage after heart surgery. The common complications were seen in this period:

-Blood pressure: Hypotension during the first hours after surgery. - Arrhythmias: bradycardia, sinus tachycardia, atrial fibrillation. - Low cardiac output syndrome: common in about 6-8 hours after surgery.

- Right ventricular failure and pulmonary hypertension. - Diastolic dysfunction. - Distribution shock. - Myocardial anemia and myocardial infarction.

1.2. Low cardiac output syndrome after heart surgery caused by heart failure 1.2.1. Cardiac Output and Cardiac Index

The activity of the heart is reflected through cardiac output (CO). It is an average of blood volume that the heart pumps per minute to demand the metabolic needs. CO = Stroke volumex Heart rate

Mean of CO: 5 – 6l/min. The Cardiac Index (CI) is another presentation of cardiac output, defined as CO per skin area. This index is not depending on the height, weight, and feasibility to apply in clinical practice. CI = CO / S Mean ofCI: 2.5 – 3.5 l/min/m2. 1.2.2. Low cardiac output syndrome caused by heart failure

Low cardiac output syndrome caused by heart failure is a clinical condition resulted from decreasing in systemic perfusion pressure,which led to a decrease in myocardial function with imbalance cellular oxygen supply and consumption and formed metabolic acidosis.

There is no definition consensus of LCOS after cardiac surgery with extracorporeal circulation. According to many authors, LCOS after

to used post-operative

3 cardiac surgery with extracorporeal circulation is a condition that patient intra-aortic balloon pumpe or needs cardiovascular medication such as vasomotor (dopamine dose > 5 µg/kg/min or dobutamine, adrenalin, noradrenalin, milrinone at any doses) from ≥ 30 minutes after surgery to maintain systolic blood pressure> 90 mmHg and CI > 2.2 L/min/m2, after optimizing the preload, afterload, and hemostatic condition (electrolyte and blood gas). Recommendation of Vela aboutLCOS after cardiac surgery with extracorporeal circulation:

- Cardiac index <2.2 L/min/ m2 without reducing blood volume. The etiology maybe are the failure of right, left, or both ventricles with or without pulmonary congestion. Blood pressure maybe normal or decreased.

-The clinical manifestations of LCOS: using when unable to monitor cardiac output: oliguria (urine <0.5 ml/kg/h), central venous oxygen saturation <60% (with normal arterial oxygen saturation), and/or lactate > 3 mmol/l, without the insufficiency blood volume. - In severe cases: cardiac index <2 L/min/m2, systolic blood

from an pressure <90 mmHg, oliguria, and sufficient blood volume. 1.2.3.Hemodynamic monitoring in cardiac surgery anesthesiologist.

- Invasive arterial blood pressure measures. - Swan-Ganz Catheter: measuring pulmonary pressure, cardiac output, and other values. -Echocardiography:Trans-thoracicortrans-esophageal echocardiography. - Monitor cardiac output using a PiCCO or Flotrac system.

1.3. N-Terminal pro-B-type natriuretic peptide (NT-proBNP) 1.3.1. Structure and formation of NT-proBNP

Formation of NT-proBNP: In cardiomyocytes, preproBNP divide into proBNP (108 amino acid) and signal peptide (26 amino acids). ProBNP secreted into the blood by ventricles of the heart in response to myocardial injury or excessive stretching of the heart muscle cell (pressure or volume). In blood, undergoing the catalysis process ofcorin/furin enzyme, proBNP divide into BNP (32 amino acid) and NT- proBNP (76 amino acid).

4 1.3.2. Serum NT-proBNP level

The value 125 pg/ml ofNT-proBNP is considered a baseline in patients at risk for heart failure with a very high negative predictive value. However, It is useful when dividing by age:Under 50 years old: 50 pg/ml; From 50 to 75 years: 75 - 125 pg/ml; Over 75 years: 125 pg/ml. The US Food and Drug Administration (FDA) certified a value of

250-300 pg/ml for people  75 years old. 1.3.3. Identify the serum NT-proBNP level NT-proBNPwas performed by

immunological the

luminescent electrochemical immunization according to the principle of sandwich onCobase601 (Roche Elecsys 2010) using ECLIA (Electro chemiluminescence test by MODULAR immunoassay).Analyzing luminescentelectrochemical ANALYTICS E170. The principleof immunization. Principle of sandwiches:

- The first incubation period: the antigen in the test specimen sandwiched between a biotinized NT-proBNP-specific monoclonal antibody and a ruthenium-marked NT-proBNP-specific monoclonal antibody forming a sandwich complex (sandwich).

the electrode produces chemical to

- The second incubation period: After adding the microparticles coated with Streptavidin, the sandwich complex becomes cohesive and converts to the solid phase by the reaction of Biotin and Streptavidin. This mixture is sucked into the measuring chamber, where the particles are magnetically attracted to the surface of the electrode. The unbound substances will then be rejected with the procell solution. Applying luminescence. The voltage luminescent signal is received and measured with a magneto-optical amplifier. The results are determined based on a standard machine curve.

*ReagentM: microparticles surrounded by streptavidin; R1: biotinized monoclonal anti-NT-proBNP (from mouse) antibody; R2: NT-proBNP (sheep) monoclonal antibody labeled with a ruthenium complex.

* Specimen tube and storage: Blood tubes that contain K2- or K3- EDTA plasma. The blood is centrifuged and serous. Blood samples were stable for 3 days at a temperature of 200C – 250C, 6 days at 20C – 80C, 24 months at –200C

5 CHAPTER 2 MATERIALS AND METHODS 2.1.Study subjects

We conducted this prospective study on 107 patients who underwent CABG surgery withextracorporeal circulationat Ho Chi Minh Heart Institute from October 2012 to June 2014. 2.1.1. Inclusion criteria

- Aged 18 and older regardless of gender. - Indication for CABG surgery with extracorporeal circulation. 2.1.2. Exclusion criteria:

- Aged <18 - Concomitant cardiac surgery. - Renal function insufficiency (Creatinin > 1.6 mg/dl). - Myocardial infarction. - Severe COPD. - Do not agree to participate in the study. 2.1.3. Discontinuation from study

- Death within 24h after surgery. - Do not agree to participate in the study.

2.2. Methodology 2.2.1. Study design:Prospective, longitudinal descriptivestudy with comparison. 2.2.2. Materials

- The arterial catheter (20G, Vygon, French). - Central venous catheter (7Fr B.Braun, Germany). - Colour Echocardiography(Philips Bothel.WA, USA). - Flotrac/Vigileo System USA). - Monitor (Philips MP40, USA). - Immunology test (Cobas e602, Roche). -Dobutamine (Bivid Co, Germany): 250 mg/20ml - Noradrenalin (Levonor, Poland): 1mg/ml. - Adrenalin (Minh Dan, Viet Nam): 1mg/ml.

2.2.3. Study process 2.2.3.1. Before surgery:

Recording the patient history and general characteristic, diagnosis of chronic heart failure according to Framingham, heart failure

image, consultation decision on CABG

6 classification according to NYHA, EuroSCORE scale, taking blood tests, X-ray, electrocardiography, Doppler echocardiography, coronary artery surgery with cardiopulmonary bypass. 2.2.3.2. In surgery:

- Setup monitor to follow the vital signs. - Arterial catheter to monitor arterial blood pressure. - Monitoring central venous pressure using catheter: Central-line catheter number 16 was placed through the internal jugular vein. The position of the catheter was the position of the aortic vein in the right atrium (depth of the catheter about 15 cm). Central venous pressure measured by Truwave sensor (mmHg), The zero point is the intersection of the medial axillary line and the 4th intercostal space.

- Using Flotrac System to monitor cardiovascular indexes: Cardiac Output (CO), Cardiac Index (CI), Stroke Volume (SV), Stroke Volume Index (SVI), Stroke Volume Variations (SVV), Stroke Volume Resistants (SVR). - Testing NT-proBNP before going to the operative room, before

the procedure. 2.2.3.3. After surgery (in ICU)

- Continous following vital signs on monitoring: heart rate, invasive blood pressure, respiratory rate, respiratory pattern, SpO2, central venous pressure CVP, and temperature.

- Following CI, CO, SVI through Flotrac System. - Following LCOS after cardiac surgery. - Transthoracic echocardiography: assessment of left ventricular function (EF), systolic pulmonary arterial pressure (PAPs), regional movement disorders, pleural effusion, and pericardium effusion (if any).

- Monitoring and treatment LCOS after heart surgery. - Serum NT-proBNP test at postoperative days: + Sample preparation: Taking 1ml of blood from the vein into a tube containing K2- or K3-EDTA plasma. Tubes are marked with the full name and age of the study patient and are barcoded. After taking the blood, the specimen tube was put in an icebox and bring it to the laboratory. The maximum time from taking blood to putting the tube in the machine is 30 minutes. The assay was performed on Roche Cobas e602 automated immunoassay.

7

+ Performing test: Using pipettes Roche CARDIAC to take blood from a sample tube with a rubber lid. Before removing the blood sample from the tube, press the piston completely and then pierce the needle through the rubber tube cap. Always ensure homogeneity of blood specimens before inserting the test strip (by gently shaking the tube several times before taking the sample). Taking exactly 150 ml of blood from the tube into the pipette (according to the mark on the pipette) and ensure that there are no bubbles. + The analyzer automatically calculates the analyte concentration

of each sample (either in pmol/ L or pg/mL). 2.2.3.4. Data collection

The results were collected at the time points: - No: the day before surgery(NT-proBNP and hemodynamic index using Flotrac system were evaluated before going to the operating room to operate).

- N1: Postoperative day 1 (2 hours after surgery). - N2: Postoperative day 2(8 a.m.). - N3: Postoperative day 3(8 a.m.). - N4: Postoperative day 4(8 a.m.). - N5: Postoperative day 5(8 a.m.). Besides,we also recorded data about clinical characteristics, indexes, NT-proBNP,and dosage of cardiovascular cardiovascular medication at a time when patients showed cardiac impairment or when LCOS occurred. 2.2.4. Criteria in our research 2.2.4.1. Criteria in general characteristics: - Age (years) divided into three groups : < 50 years, from 50 – 75 years and > 75 years.

Underweight

- Gender: Male, female and % male/ female ratio. - Height (cm), wieght (kg). - Body Mass Index (BMI): BMI (kg/m2 ) = Weight/(Height)2. According to WHO: BMI < 18,5 BMI = 18,5 – 24,99 Normal BMI ≥ 25 Overweight - Diagnosing chronic heart failure before surgery according to Framingham.

8 - Classification by NYHA: NYHAI; NYHAII; NYHAIII,

NYHAIV. 2.2.4.2.Criteria for the changes of NT-proBNP level with clinical and subclinical characteristics

- Maximum and minimum blood pressure (mmHg): Diagnosis of hypotension according to the Vietnam Heart Association: systolic blood pressure <90 mmHg and/or diastolic blood pressure<60 mmHg.

- Central venous pressure (CVP): Using Flotrac. + Normal CVP: CVP = 5-12 cmH2O + High CVP: CVP > 12 cmH2O - Evaluating left ventricular function: Based on Ejection fraction (EF) on Doppler Echocardiography.

+ Normal EF: EF = 55 - 70%. + Decreased EF: EF < 55%. In our study, we used the index of EF ≤ 50% as left ventricular dysfunction.

- Evaluating Systolic Pulmonary Artery Pressures (PAPs): + Normal PAPs: < 30 mmHg + High PAPs: ≥ 30 mmHg - Evaluating hemodynamics index: Using monitor to follow continuous cardiac output (CCO) through Flotrac system: + Cardiac index (CI):

Normal CI: 2.5 – 4.0 L/min/m2 Decreased CI: ≤ 2.4 L/min/m2 + Cardiac output (CO):

Normal CO: 4.0 – 8.0 L/min Decreased CO: < 4.0 L/min + Stroke volume index (SVI):

Normal SVI: 33 – 47 ml/m2 Decreased SVI: ≤ 32 ml/m2

2.2.4.3. Criteria in the association between NT-proBNP and LCOS after surgery: - The association between NT-proBNP and the prognosis ability of LCOS.

Criteria for diagnosing LCOS after surgery: + Cardiac index <2.2 L/min/ m2 without reducing blood volume. The etiology maybe are the failure of right, left, or both ventricles with

9 or without pulmonary congestion. Blood pressure maybe normal or decreased. + In severe cases: cardiac index <2 L/min/m2, systolic blood pressure <90 mmHg, oliguria, and sufficient blood volume. - Recommendation the cut-off value of NT-proBNP to diagnose acute heart failure:

Table 2.1. The cut-off value for diagnosing heart failure by age

Optimal cut-off value Age Diagnosing heart failure

Confirmed diagnosis < 50 50 – 75 > 75 (pg/ml)  450  900  1800

Excluding at any ages < 300

The cut-off value was determined as the value that the NT-proBNP level had the maximal sensitivity and specificity, calculated by maximal J index (Youden Index). J = max (Sensitivity + Specificity -1). - The association between NT-proBNP and the ability of profnosis of LCOS after cardiac surgery.

In our study, we used EuroSCOREto prognose the risk of acute heart failure after CABG surgery. This is quite simple based on 17 index according to patients, heart or surgery. It has three level of EuroSCORE:

 EuroSCORE 0-2 score: low risk  EuroSCORE 2-5 score: moderate risk  EuroSCORE > 5 score: high risk - The association betweenNT-proBNP level with Vasoactive Inotropic Score (VIS) after surgery:

IS = dopamin dosage (µg/kg/min) + dobutamin dosage (µg/kg/min) + 100 x epinephrin dosage (µg/kg/min). VIS = IS + 10 x milrinon dosage (µg/kg/min) + 10000 x vasopressin dosage (UI/kg/min) + 100 x norepinephrin dosage (µg/kg/min). The medication did not use, its dosage was calculated as 0.We evaluated VIS at N1, N2, N3, N4, N5. VIS > 15 was considered as high value and VIS ≤ 15 was considered as low value.

10 2.3. Statistical analysisData collected using Epi Data 6 and using STATA14.0 to analyze.

Indication for CABG surgery

Excluded Included

N0: Clinical characteristics, Echocardiography, NT-proBNP Hemodynamic indexes

CABG

ICU

N1 N5 N2 N3 N4

Clinical characteristics, Echocardiography, NT-proBNP Hemodynamic indexes

No LCOS LCOS

Objective 1 Objective2

Figure 2.1. Study design

11 Chapter 3 RESULTS 3.1. Demographic characteristics

- The mean age in the research group is 60.7 ± 9.4. Patients aged 50-75 years were major with 87.8%. The percentage of males was higher than females (72 male versus 35 female).

- There were 36/107 cases (33.64%) diagnosed with chronic heart failure before surgery. NYHA II and NYHA III were dominant in classification accounting for 55.6% and 33.3% respectively. - Patients with EuroScore 3-5 were mainly with 58% before surgery.

- CABG with 3 and 4 bridges was mainly with 48.6% and 33.6%. The mean of ECMO and clipped aorta artery time were 126.8 ± 27.1 minutes and 87.7 ± 23.4 minutes, respectively. 3.2.Evaluating the changes in the level of serum NT-proBNP in patients with CABG.

Table 3.11. Level of serum NT-proBNP at time points

NT-proBNP (pg/ml) Time

N0 ± SD 491.2 ± 601.1 Median 364.7

N1 972.5 ± 1608.05 838.5

N2 1915.1 ± 2513 1792.4

N3 4057.26 ± 4458.12 3766.2

N4 3981.78 ± 4549.03 3157.1

N5 3457.81 ± 4110.98 2845.3

p < 0.05

Comment:After surgery, the mean of NT-proBNP level increased slightly at N1 with 972.5 pg/ml, then increased fastly and peaked at N3 with 4057.26 pg/ml, and then tending decreased gradually at N5 with 3457.81 pg/ml. The difference in the NT-proBNP level between pre- and post-operative time was statistically significant with a p-value <0.05.

12 Table 3.14. Level of serum NT-proBNP and EF after surgery

Time EF% r

NT-proBNP (pg/ml) ± SD 571.65 ± 2196.71 447 ± 609.45 Median 949.15 195.4 N1 -0.396; > 0.05 ≤ 50 (n=50) > 50 (n=57) p

N2 2964 1063 -0.49; > 0.05

N3 4819 1816 -0.489; > 0.05

N4 5066 1873 -0.408; > 0.05

N5 -0.392; > 0.05 4069 1743 < 0.001 ≤ 50 (n=49) 4450.07 ± 4417.48 1514.91 ± 1384.37 > 50 (n=58) < 0.001 p ≤ 50 (n=50) 6026.43 ± 5597.83 2301.03 ± 2081.84 > 50 (n=57) < 0.001 p ≤ 50 (n=44) 5958.9 ± 5865.69 2600.93 ± 2608.74 > 50 (n=63) < 0.001 p 5202.64 ± 5311.8 ≤ 50 (n=43) 2285.5 ± 2480.12 > 50 (n=64) < 0.001 p Comment:

At time points of N1, N2, N3, N4, and N5 after surgery, the mean of NT-proBNP level in the group with EF ≤ 50% was higher than in the group with EF> 50%. The difference was statistically significant with a p-value <0.001. However, there was a negative correlation between the NT-proBNP level and EF.

13 Table 3.16. Level of serum NT-proBNP and CI after surgery

Time r

N1 Median 2731.5 255 -0.35; > 0.05 NT-proBNP (pg/ml) ± SD 2589.95 ± 1535 729.1 ± 1539.95 < 0.001

N2 5370 1529 -0.516; > 0.05

N3 10758 2431 -0.436; > 0.05 8376 ± 8321.13 2472.86 ± 2525.27 < 0.01 10118 ± 3659.22 3881.8 ± 4428.38 < 0.05

N4 3981.78 ± 4549.03 2430 -0.309; > 0.05

N5 3457.81 ± 4110.98 2079 -0.289; > 0.05 CI (l/min/m2) < 2.4 (n=14) ≥ 2.4 (n=93) p < 2.4 (n=11) ≥ 2.4 (n=96) p < 2.4 (n=3) ≥ 2.4 (n=104) p < 2.4 (n=0) ≥ 2.4 (n=107) p < 2.4 (n=0) ≥ 2.4 (n=107) p

serum NT-proBNP with the Comment: At the time points of N1, N2, N3, N4, and N5 after surgery, the mean of NT-proBNP level in the group with CI <2.4l/min/m2 was higher than in the group with CI≥2.4l/min/m2 (p-value<0.05). There was a negative correlation between the NT-proBNP level and CI. 3.3. The relationship between serum NT-proBNP with low cardiac output syndrome after CABG. 3.3.1.The correlation between predictability about LCOS

p- value Time

Table 3.19. The rate of LCOS after heart surgery(n=107) Low cardiac output syndrome(Yes) n 13 9 3 0 0 25

Low cardiac output syndrome(No) n 94 98 104 107 107 82 % 87.9 91.6 97.2 100.0 100.0 76.6 % 12.1 8.4 2.8 0.0 0.0 23.4 POD 1 POD 2 POD 3 POD 4 POD 5 POD(ingeneral) < 0.05 < 0.05 < 0.05 - - < 0.05

0 0 . 1

0 0 . 1

5 7 . 0

5 7 . 0

0 5 . 0

y it c i f i c e p S / y t i

0 5 . 0

y t i v i t i s n e S

v i t i s n e S

5 2 . 0

5 2 . 0

0 0 . 0

0.00

0.25

0.75

1.00

0 0 . 0

0.50 Probabil ity cutoff

0.00

0.25

0.75

1.00

0.50 1 - Sp ecifici ty

Sens itiv ity

Speci fici ty

Area u nder ROC c urve = 0.8650

14 Comment:The patients with LCOS at postoperative day 1 (POD 1), POD 2, POD 3 were 13 cases (12.1%), 9 cases (8.4%), 3 cases (2.8%). The patients with low cardiac output syndrome after surgery was 25 cases (23.4%). The difference between the group with low cardiac output syndrome and the group without low cardiac output syndrome was statistically significant with a p-value <0.05.

Figure 3.1. Cut-off NT-proBNP and the predictability about LCOSPOD 1

0 0 . 1

0 0 . 1

5 7 . 0

5 7 . 0

0 5 . 0

0 5 . 0

y t i v i t i s n e S

y t i c i f i c e p S / y t i v ti i s n e S

5 2 . 0

5 2 . 0

0 0 . 0

0 0 . 0

0.00

0.2 5

0 .75

1.00

0 .50 P ro bab ilit y cuto ff

0 .00

0 .25

0.7 5

1.0 0

0.5 0 1 - Spe cificity

S ensit ivity

S pecif icity

Area u nder R OC curve = 0 .8 549

- Cut-off value of serum NT-proBNP POD 1: 951.5 pg/ml. - Sensitivity (Se): 92.3%; Specificity (Sp): 78.7%. - J index (Youden Index): 0.71 - AUC (CI95%): 0.865 (0.72 – 1) with p<0.05.

Figure 3.2. Cut-off NT-proBNP and the predictability about LCOS POD 2

0 0 1.

0 0 1.

.

5 7 . 0

5 7 0

y t i

c i f i

.

0 5 0

0 5 . 0

y t i v i t i s n e S

c e p S / y t i v i t i s n e S

5 2 0.

5 2 . 0

.

0 0 0

0 0 . 0

0.00

0.25

0.75

1.00

0 .50 Probability cutoff

0.00

0. 25

0.75

1.0 0

0. 50 1 - Spec if icit y

Sensitivity

Specifi city

Area und er R OC cu rve = 0.8 479

- Cut-off value of serum NT-proBNP POD 2: 2018 pg/ml. - Sensitivity (Se): 100.0%;Specificity (Sp): 65.3%. - J index (Youden Index): 0.65 - AUC (CI95%): 0.855(0.75-0.95) with p<0.05.

Figure 3.3. Cut-off NT-proBNP and the predicility about LCOS POD 3

- Cut-off point of serum NT-proBNP POD 3: 4601 pg/ml. - Sensitivity (Se): 100.0%;Specificity (Sp): 74.8%. - J index (Youden Index): 0.748 - AUC (CI95%): 0.848(0.72-0.98) with p<0.05.

15

low cardiac output syndrome according the to

3.3.2. The correlation between serum NT-proBNP with predictability of EuroSCORE. Table 3.1. Analyzing the diagnosis accuracy in low cardiac output syndrome according to EuroSCORE and cut-off value NT-proBNP at POD 1 Low cardiac output syndromeat POD 1

Index Total p No (n=94)

EuroSCORE Yes (n=13) n % 30.8 4 69.2 9 % 77.7 22.3 n % 77 72.0 30 28.0 <0.01

71 66.4 36 33.6 74.5 25.5 7.7 92.3 1 12 <0.001 Cut-off value NT-proBNP ≤ 5 > 5 Se. Sp. Acc ≤871.8pg/ml >871.8pg/ml Se. Sp. Acc n 73 21 Se: 77.7%; Sp 69.2%; Acc: 76.64% 70 24 Se: 74.5%; Sp 92.3%; Acc: 76.64%

Table 3.24.Analyzing the diagnosis accuracy in low cardiac output syndrome according to EuroSCORE and cut-off value NT-proBNP at POD 2 Low cardiac output syndrome at POD 2 Index p

EuroSCORE No (n=98) % n 75.5 74 24.5 24 Yes (n=9) n % 33.3 3 66.7 6 Total n % 77 72.0 30 28.0 <0.05 Se: 75.5%; Sp 66.7%; Acc: 74.8%

68 30 69.4 30.6 2 7 22.2 77.8 70 65.4 37 34.6 <0.01 Cut-off value NT-proBNP ≤ 5 > 5 Se. Sp. Acc ≤2516pg/ml >2516pg/ml Se. Sp. Acc Se: 69.4%; Sp 77.8%; Acc: 70.1%

Index p

16 Table 3.26. Analyzing the diagnosis accuracy in low cardiac output syndrome according to EuroSCORE and cut-off value NT-proBNP at POD 3 Low cardiac output syndrome POD 3 Total No (n=104) Yes (n=3) n n % 33.3 77 72.0 76 66.7 30 28.0 28 n % 1 2 % 73.1 26.9 EuroSCORE >0.05 Se: 73.1%; Sp 66.7%; Acc: 72.9%

74 29 71.8 28.2 0 3 0 100 74 69.8 32 30.2 <0.05 Cut-off NT-proBNP ≤ 5 > 5 Se. Sp. Acc ≤.3556pg/ml >3556pg/ml Se. Sp. Acc Se: 71.8%; Sp: 100%; Acc: 72.6% Comment:

- Table 3.22 showed at POD 1, the sensitivity (Se) and specificity (Sp) of the cut-off value NT-proBNP was higher when compared to the EuroSCORE 5 score. However, accuracy was equivalent to 76.64%. - Table 3.24 showed at POD 2, the accuracy of the cut-off value NT-proBNP was higher when compared to the EuroSCORE 5 score. - Table 3.26 showed at POD 3, the accuracy of the cut-off value

0 0 1

0 8

0 6

0 4

0 2

0

0

500 0

10000

15000

NT-ProBNP - N1

VIS - N 1

Fitt ed val ues

NT-proBNP was higher when compared to EuroSCORE 5 score. 3.3.3. Correlation between serum NT-ProBNP and VIS

0 0 2

0 5 1

0 0 1

0 5

0

0

50 00

10000

15000

2000 0

25000

NT-ProBNP - N2

VIS - N 2

Fitt ed val ues

VIS POD 1 = 0.002 x NT-ProBNP POD 1 + 12.15; r =0.175; p>0.05 Figure 3.4. Correlation between serum NT-ProBNP and VIS in POD1.

VIS POD 2 = 0.001 x NT-ProBNP POD 2 + 9.44; r= 0.193; p>0.05 Figure 3.5. Correlation between serum NT-ProBNP and VIS in (POD 2).

0 0 1

0 5

0

0

100 00

200 00

300 00

NT-ProBNP - N3

VIS - N 3

Fitted val ues

17

0 0 1

0 8

0 6

0 4

0 2

0

0

1000 0

3000 0

40000

20000 NT-ProBNP - N4

VIS - N 4

Fitt ed val ues

VIS POD 3 = 0.001 x NT-ProBNP POD 3 + 5.58; r= 0.257; p>0.05 Figure 3.6. Correlation between serum NT-ProBNP and VIS inPOD3.

0 0 6

0 0 4

0 0 2

0

0

10000

2000 0

300 00

NT-ProBNP - N5

VIS - N 5

Fitt ed val ues

VIS POD 4 = 0.0008 x NT-ProBNP POD4 + 5.69; r=

0.203;p>0.05Figure 3.7. Correlation between serum NT-ProBNP and VIS in POD4. VIS POD 5 = 0.001 x NT-ProBNP POD 5 + 8.42; r= 0.09; p>0.05 Figure 3.8. Correlation between serum NT-ProBNP and VIS inPOD5. Comment:

The figure above showed the correlation between NT-proBNP and VIS was a weak positive correlation and not statistically significant (p> 0.05).

18 Chapter 4 DISCUSSION 4.1. General characteristics

In our study, the average age was 60.7 ± 9.4 years, mostly from 50- 75 years old with 94 patients. We categorized the age into three groups: under 50 years old, 50 to 75 years old, and over 75 years old. This division is in line with Januzzi JL in choosing the age cutoff for NT- proBNP evaluation.

Regarding the distribution of patients by gender, our results showed that male subjects made up the majority with 72 patients, accounting for 67.3%. As for Nguyen Thi Quy's study on 330 patients undergoing coronary artery bypass grafting, the ratio of men to women was 2.75: 1. Our research showed an almost similar result when the ratio of men to women was 2.1.

In our study, the average cardiopulmonary bypass time was 126.8 ± 27.1 minutes, the average aortic cross-clamp time was 87.7 ± 23.4 minutes. This result was similar to the study of Nguyen Thi Quy and Duong Ngoc Dinh. Our research, as well as some other author in our country, showed that the cardiopulmonary bypass time and the average aortic cross-clamp time were longer than those of some authors from other countries, such as Mustafa Cerrahoglu's study with the result: 101 ± 36.41 minutes for the cardiopulmonary bypass time, 57.23 ± 20.66 minutes for the aortic cross-clamp time; the study of Guillermo Reyes has the cardiopulmonary bypass time of 101.8 ± 35.3 minutes and the aortic cross-clamp time of 66.4 ± 27.8 minutes. 4.2. Investigation of changes in the serum NT-proBNP level in patients undergoing CABG. 4.2.1. Changes in the NT-proBNP level on postoperative days

After surgery, the average serum NT-proBNP level tended to increase slightly from 491.2 ± 601.1 pg/ml before surgery (N0) to 972.8 ± 1608.05 pg/ml on N1 (2 hours after treatment at ICU). On the second postoperative day, the serum NT-proBNP level increased markedly with an average of 1915.1 ± 2513 pg/ml. On the third postoperative day, the serum NT-proBNP level continued to increase and peaked at 4057.26 ± 4458.12 pg/ml. The fourth postoperative day witnessed a marginal decrease in the serum NT-proBNP level, standing at 3981.78 ± 4549.03 pg/ml. Then, it decreased rapidly to 3457.81 ± 4110.98 pg/ml on the

19 fifth postoperative day. The study of Guillermo Reyes et al. on 83 heart surgery patients also showed similar variation in serum NT-proBNP in 7 days after surgery. 4.2.2 Changes in the NT-proBNP level by blood pressure after surgery The study showed that: on N1, N2, and N3, the NT-proBNP level in the group of patients with MAP <90 mmHg was higher than that in the group with MAP ≥ 90 mmHg, the difference was statistically significant with p <0.05. Correlation analysis showed that there was a weak inverse correlation between NT-proBNP level and MAP, while blood pressure decreased, the NT-proBNP level increased. On N4, N5, there was no patient with low blood pressure. The study of Dang Duc Hoan on 123 patients with acute myocardial infarction receiving PCI also showed a moderate inverse correlation between blood pressure and NT-proBNP level (r = -0,381 and -0,338; p <0.0001). And the study of Nguyen Thi Tuong Van et al. also showed that serum BNP and systolic and diastolic blood pressure in patients with MI had a moderate inverse correlation (r = -0,41 and -0.46; p <0.01). 4.2.3 Changes in the NT-proBNP level by EF on postoperative days

The results showed that: on N1, N2, N3, N4, and N5, the NT- proBNP level in the group of patients with EF ≤ 50% higher than that in the group with EF> 50%, the difference is statistically significant with p<0.05. Correlation analysis showed that there is a moderate inverse correlation between the NT-proBNP level and EF. Research by Omland et al. showed an increase in NT-proBNP level (>545 pmol /L) associated with decreased left ventricular function. Serum NT-ProBNP level is a useful marker to assess the recovery of left ventricular ejection function (LVEF), especially in patients with a high-risk factor after CABG; And it had an inverse correlation with LVEF. If the NT-proBNP level increased, LVEF would decrease. 4.2.4Changes in the NT-proBNP level by cardiac index (CI) after surgery

The result showed that there were 28 patients with decreased cardiac index (CI <2.4 L / min / m2) after surgery. These cases all happened at the times N1, N2, and N3 after surgery. No patient had a decrease in CI on the fourth and fifth postoperative days. In patients with decreased CI, the NT-proBNP level increased twofold to threefold higher than that of the group with normal CI. The difference is

20 statistically significant with p <0.05. There is a moderate inverse correlation between the NT-proBNP level and CI. A study by Mustafa on 52 patients undergoing CABG showed that the variables, including left ventricular ejection fraction (LVEF), cardiac output (CO), and cardiac index (CI), in the group with increased levels of NT-proBNP, were lower than those in the group with normal NT-proBNP level (NT- proBNP <220 pg /ml). 4.3. The association between serum NT-proBNP and LCOS after CABG 4.3.1 The role of serum NT-proBNP in the diagnosis of LCOS after CABG

The NT-proBNP assay was highly valuable in detecting LCOS with AUC (CI95%): 0.865 (0.72-1) with p <0.05. At the level of 951.5 pg /ml, the serum NT-proBNP had a sensitivity of 92.3% and a specificity of 78.7%. The AUC of serum NT-proBNP level of the group with LCOS on N2 was AUC (CI95%): 0.855 (0.75-0.95) with p <0.05, indicating that NT-proBNP assay was also valuable in detecting LCOS. At the level of 2018 pg /ml, the serum NT-proBNP had 100.0% sensitivity and 65.3% specificity. The AUC of serum NT-proBNP level of the group with LCOS on N3 was AUC (CI95%): 0.848 (0.72-0.98) with p <0.05, indicating that NT- proBNP was also valuable in detecting LCOS. At the level of 4601 pg /ml, serum NT-proBNP had a sensitivity of 100.0% and a specificity of 74.8%.

To diagnose LCOS after cardiac surgery, we found the cut-off with the highest sensitivity and specificity, and the highest J (Youden Index). Therefore, the cut-off was determined to be the level of NT-proBNP on N1, standing at 951.5 pg /ml with a sensitivity of 92.3% and a specificity of 78.7%, the J (Youden Index) of 0.71, which were highest among postoperative days.

In the PRIDE study, the 300 pg/ml cutoff gave excellent application in excluding the diagnosis of acute heart failure, the negative predictive value (99%) was significantly higher than that of 100 pg/ml individual BNP cut-off (89%). The 900 pg /ml cutoff for the diagnosis of acute heart failure had a positive predictive value similar to that of the 100 pg/ml BNP cutoff (76% versus 79%).

21

the to low cardiac output syndrome according

4.3.2. The correlation between serum NT-proBNP with predictability of EuroSCORE. The results show that the threshold of NT-proBNP on the first day after surgery is 871.8 pg/ml and in combination with EuroSCORE before surgery, the accuracy of diagnosing LCOS with EuroSCORE low- moderate risk (mostly medium risk) is 80.5%, with the EuroSCORE high risk 70%.With a threshold of NT-proBNP on the second day after surgery is 2516 pg/ml and and in combination with EuroSCORE before surgery, the accuracy of diagnosing LCOS with EuroSCORE low- moderate risk (mostly medium risk) is 77.9%, with the EuroSCORE high risk 66.7%. With a threshold of NT-proBNP on the third day after surgery is 3556 pg/ml and and in combination with EuroSCORE before surgery, the accuracy of diagnosing LCOS with EuroSCORE low- moderate risk (mostly medium risk) is 81.6 %, with the EuroSCORE high risk 60%.Holm's retrospective study showed that the combination of predictive evaluation of postoperative heart failure between the EuroSCORE point and the preoperative NT-proBNP showed efficacy. 4.3.4. The association between NT-proBNP level and vasoactive inotropic score (VIS) after surgery The result showed that: basically, the correlation between NT-proBNP and VIS at times N1, N2, N3, N4, and N5 was positive, weak, and not statistically significant. This result is also in line with the study of Yanqin Cui and the study of Ibarra-Sarlat M. The study of Yanqin Cui et al. showed that there was a correlation between NT-proBNP and MIS (Maximum Inotropic Score) at 1, 12, and 36 hours after surgery (p <0.01). The study of Ibarra Sarlat M et al. on 40 patients after heart surgery showed a weak positive correlation (r = 0.26) and there was no statistical significance (p = 0.09) between NT-proBNP and VIS at 24 hours after surgery.

22 CONCLUSION 1. Changes in serum NT-proBNP level in patients undergoing CABGwith extracorporeal circulation.

- After CABG, the serum NT-proBNP level increased gradually from N1(972,5 ± 1608,05 pg/ml), increased andpeaked on N3 (4057,26 ± 4458,12 pg/ml), and then tended to decrease gradually at N5(3457,81 ±4110,98 pg/ml), (p<0,05).

- On N1, N2 and N3, there was weak inverse correlation ((r1 = - 0.207; r2 = -0.033; r3 = -0.027) between the serum NT-proBNP level and low blood pressure ( MAP < 90 mmHg).

- At pre- and post-operative times, the average NT-proBNP level in patients with EF ≤ 50was higher than that in patients with EF> 50% (p<0.05). NT-proBNP levels were in moderate inverse correlation with the average with LVEF at postoperative times (r1 = -0.396; r2 = -0.49; r3 = -0.489; r4 = -0.408; r5 = -0.392).

- At the times N1, N2, and N3, the average NT-proBNP level in the group of patients with low cardiac output (CI <2.4 l /min /m2; CO <4.0 l /min) was higher than that in the group with normal cardiac output (CI ≥ 2.4 l /min /m2; CO ≥ 4.0 l /min), the difference was statistically significant with p <0.05. There is a moderate inverse correlation between the NT-proBNP level and CI (r1 = -0.35; r2 = - 0.516; r3 = -0.436), and with CO (r1 = -0.32; r2 = -0.451; r3 = -0.434). 2. The association between serum NT-proBNP level and LCOS after CABG.

- To diagnose LCOS after CABG, the cut-off value was determined based on the level of NT-proBNP with the highest sensitivity and specificity, the highest J index, on N1 with 951,5 pg/ml: sensitivity of 92.3%, the specificity of 78.7%; J index (Youden Index) of 0.71; AUC (CI95%) of 0.865 (0.72-1) with p <0.05.

-The combination of preoperative EuroSCORE and postoperative serum NT-proBNP showed prognostic effects of LCOS: the cut-off of NT-proBNP level for detecting LCOS on N1, N2, N3 by the EuroSCORE was: 871.8 pg/ml, 2516 pg/ml, and 3556 pg/ml, level and respectively. The correlation between NT-proBNP EuroSCORE was a weak positive correlation (p> 0.05). - Serum NT-proBNP level has prognostic significance in LCOS by VIS:

23

The cut-off of NT-proBNP levelmay predict acute heart failure in the group with high VIS (VIS ≥ 15) on the first postoperative day was 1682 pg/ml (with a sensitivity of 93.7% and a specificity of 72.7 %; J index of 0.664); on the second postoperative day was 8844 pg/ml(with 90% sensitivity and 60% specificity; J index of 0.5);on the third postoperative day was 7152 pg/ml (with 63.2% sensitivity and 100% specificity; J index of 0.63). The correlation between serum NT-proBNP level and VIS was a weak positive correlation(p> 0.05).

24 RECOMMENDATION Based on the above results and conclusions, we would like to have the following recommendations:

NT-proBNP is a very necessary and useful test in the prediction and prognosis of LCOS after CABG. Therefore, Serum NT-proBNP should be early quantified in patients with CABG to diagnose quickly and accurately the LCOS that is caused by acute heart failure in order to devise treatment strategies and suitable prevention to limit risk factors after surgery. The serum NT-proBNP concentration can be combined with EuroSCORE index and the VIS to increase the predictability of LCOS after coronary bypass surgery.

LIST OF WORKS OF RESEARCH HAS PUBLISHED

AUTHOR RELATED TO THE THESIS 1. Bui Duc Thanh, Nguyen Hong Son, Nguyen Thi Quy, Nguyen Phu Khanh (2014), “ Stuyd the value of NT-proBNP tin prognosis and treatment heart failure after coronary artery bypass grafting surgery”, Journal of Military Pharmaco- Medicine, (6), p. 137-143.

2. Bui Duc Thanh, Nguyen Hong Son, Nguyen Thi Quy, Pham Ngoc Hung(2019), “Study the role of serum NT-proBNP in diagnosis of acute heart failure after coronary artery bypass grafting surgery”, Journal of 108- Clinical medicine and Pharmacy, (5), p.58-63.

3. Bui Duc Thanh, Nguyen Hong Son, Pham Ngoc Hung, et al (2019) “The role of serial NT-proBNP level in prognosis and follow-up treatment of acute heart failure after coronary artery bypass graft surgery”, Open Access Macedonian Journal of Medical Sciences (OAMJMS), 7 (24), pp. 4411-4415.