Log-transformed B-type natriuretic peptide as a prognostic predictor in patients undergoing cardiovascular surgery

Abstract

Objective

This study was performed to explore the association between circulating B-type natriuretic peptide (BNP) and other mortality-related factors in patients undergoing cardiovascular surgery.

Methods

In this observational study, multilevel linear regression analysis and multilevel survival analysis were performed to measure the log-transformed BNP (lnBNP) value at four time points in 197 patients with 788 repeated data measurements. Effects of the interaction between the time points and the two intervention groups (cardiac surgery and vascular surgery) were also investigated. Six models were evaluated to identify the best fit for the data. Stata/MP® version 14.2 (Stata Corp., College Station, TX, USA) was used to analyze the two-level variance component model fitting.

Results

There were significant differences in the fixed-effect parameters of lnBNP, such as the time point, age, body mass index, emergency operation, prognostic nutritional index, and estimated glomerular filtration rate. According to the multilevel survival analysis for all-cause death and vascular death, lnBNP significantly differed and was a common prognostic marker.

Conclusion

As lnBNP increased by 1 point, all-cause death increased 2.07 times and vascular death increased 3.10 times. lnBNP is an important prognostic predictor and quantitative biochemical marker in patients undergoing cardiovascular surgery.

Keywords

Log-transformed B-type natriuretic peptide cardiovascular surgery biomarker prognosis mortality linear regression analysis survival analysis

Introduction

B-type natriuretic peptide (BNP) is a cardiac biomarker that is secreted mainly from the left ventricle in response to volume expansion.^1,2 BNP is reportedly a reliable marker for the diagnosis and prognosis in patients with heart failure.^3–5

Some studies have shown that an elevated postoperative BNP concentration is independently associated with postoperative cardiovascular complications.⁶ The European Society of Cardiology and European Society of Anaesthesiology guidelines for preoperative cardiac risk assessment have recommended consideration of preoperative BNP measurement in patients undergoing high-risk noncardiac surgery.⁷ Cagini et al.⁸ reported that postoperative BNP elevation is the strongest independent predictor of cardiopulmonary complications after thoracic surgery. Mitchell and Webb⁹ reported that preoperative and postoperative measurements of BNP can help to predict postoperative cardiac dysfunction and adverse outcomes in patients undergoing cardiac surgery. Several reports have revealed that the BNP level is an important biomarker in surgery; however, a limitation of BNP is the skewness of its distribution.^10,11 The BNP value must be log-transformed before analysis to achieve a normal distribution. However, few reports have focused on the natural logarithmic transformation of BNP (lnBNP) values in patients undergoing cardiovascular surgery. Therefore, in this study, we examined postoperative survival according to the lnBNP values, physiological findings, nutritional status, and blood biochemistry parameters of patients undergoing cardiovascular surgery.

Materials and methods

Data source

Patients who underwent cardiac surgery or vascular surgery from 2009 to 2017 were included in this observational study at Kanazawa Medical University Himi Municipal Hospital. Measurements were recorded at four time points: admission, discharge, 1 month after discharge, and >1 month after discharge. Measurements soon after surgery were excluded to reduce the effect of the surgical intervention on the BNP level. The timing of all-cause death and vascular death was examined. The anonymity of all patients was maintained, and the study was conducted in accordance with the principles of the Declaration of Helsinki. This study protocol was approved by the Ethics Committee of Kanazawa Medical University Himi Municipal Hospital, Toyama Prefecture, and written informed consent was obtained from each patient (approval number 102).

Study variables

We examined the physiological findings, medical history, nutritional status, and blood biochemistry parameters of patients who underwent cardiac or vascular surgery. Each physiological finding, nutritional status, and blood biochemistry parameter was measured at the four above-mentioned time points. The patients were divided into two groups: those who underwent cardiac surgery that required cardiopulmonary bypass and those who underwent vascular surgery (abdominal aortic aneurysm open repair, endovascular aneurysm repair, or thoracic endovascular aortic repair). The nutritional status was assessed using the prognostic nutritional index (PNI), which was calculated using the serum albumin and total lymphocyte count as follows: PNI = (10 × albumin) + (0.005 × total lymphocyte count).¹² We used the lnBNP values to meet the demands of a normal distribution.

Statistical analysis

Continuous variables between the two groups (cardiac surgery and vascular surgery) were compared using Student’s t-test and the Wilcoxon rank sum test. Categorical variables were compared using Fisher’s exact two-tailed test (Table 1). Because the data were multilevel, linear modeling was used to assess the association between the response variable (lnBNP) and explanatory variables (fixed-effect parameters, random-effect parameters (patient-level variables)). The random effect was considered a random parameter of both slope and intercept. Multilevel linear regression analysis was used to identify the marker that might be associated with lnBNP, namely right cardiac failure.

Table 1.

Baseline characteristics of patients undergoing the two interventions in the present study

		Cardiac surgery			Vascular surgery
Variables		Patients (n)	Mean or %	SD	Patients (n)	Mean or %	SD	P-value^*1
Findings on admission
Baseline age	Years	127	71.0	10.7	70	77.8	9.1	<0.001
Sex	Male, n, %	78	61.4		53	75.7		0.042
Smoking	Yes, n, %	58	45.7		49	70.0		0.001
Emergency operation	Yes, n, %	23	18.1		6	8.6		0.07
Physiological findings
Baseline body mass index	kg/m²	126	22.8	3.5	68	22.4	3.8	0.43
Systolic blood pressure	mmHg	127	126.0	21.1	70	124.1	17.4	0.515
Diastolic blood pressure	mmHg	127	71.6	14.3	70	68.9	12.7	0.193
Pulse rate	bpm	127	77.6	14.5	70	71.4	11.3	0.002
Cardiothoracic ratio	%	127	57.9	8.0	69	52.8	5.5	<0.001
Medical history
Hypertension	Yes, n, %	86	67.7		63	90.0		0.002
Dyslipidemia	Yes, n, %	43	34.1		29	41.4		0.31
Diabetes mellitus	Yes, n, %	37	29.1		15	21.4		0.24
Cardiac disorder	Yes, n, %	57	44.9		26	37.1		0.29
Respiratory dysfunction	Yes, n, %	7	5.5		18	25.7		<0.001
Digestive disorder	Yes, n, %	18	14.3		8	11.4		0.57
Kidney disorder	Yes, n, %	35	27.6		21	30.0		0.72
Cerebrovascular disease	n, %	30	23.8		23	32.9		0.17
Nutritional status
Prognostic nutritional index		121	45.9	6.8	66	44.8	7.4	0.34
Blood biochemistry
Estimated glomerular filtration rate	mL/min/1.73m²	126	61.1	26.1	70	59.8	24.4	0.73
Aspartate aminotransferase	IU/L	127	27.8	25.9	70	20.9	8.3	0.032
Alanine aminotransferase	IU/L	127	23.6	29.3	70	17.9	11.7	0.12
Alkaline phosphatase	IU/L	117	256.1	88.8	67	262.9	91.6	0.62
γ-Guanosine triphosphate	IU/L	127	42.3	39.9	69	32.1	27.1	0.06
Total bilirubin	mg/dL	124	0.71	0.48	69	0.60	0.38	0.09
Total protein	g/dL	126	6.8	0.6	69	6.9	0.7	0.39
Median BNP (IQR)	pg/mL	127	124.3	45.8–379.4^†	70	71.5	39.3–189.4^†	<0.001^*2
Log-transformed BNP	1	127	4.9	1.5	70	4.4	1.1	0.025
Inverse-transformed BNP as above	pg/mL	127	128.0	99.5–164.7^‡	70	79.0	60.6–103.1^‡

SD, standard deviation; BNP, B-type natriuretic peptide; IQR, interquartile range.

Comparison of patients treated with cardiac surgery with those treated with vascular surgery was performed by *¹the paired t-test or Fisher’s exact two-tailed test or *²the Wilcoxon rank sum test. ^†25% to 75% percentile, ^‡95% confidence intervals

The following equation was used to assess the random effect on time:

lnBNP = β₀ + β₁time_ij + μ_0j +μ_1jtime_ij + ε_ij,

where i = one of the four time points (admission, discharge, 1 month after discharge, or >1 month after discharge); j = 1, 2, … 197 patients; and β₀ = a constant term.

The random effects, μ_0j and μ_1j, indicate the random intercept and random slope, respectively, and ε_ij denotes the overall error term. Six models were used to determine the best fit for the data (Table 2). We calculated the proportional change in variance (PCV) compared with the change in the null model (model 0) as the benchmark. We used an identity matrix to model the within-patient error correlation structure because of the two-level variance component model fitting. Interpatient reliability was estimated using an intraclass correlation coefficient (ICC). The models were assessed for the accuracy of the PCV of the predictors, Akaike’s information criterion, and the Bayesian information criterion.

Table 2.

Estimated log-transformed BNP from multilevel linear regression

		Model 0		Model 1		Model 2		Model 3		Model 4		Model 5
		Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE
Fixed-effect parameters (2a)
Time point	Admission			Reference		Reference		Reference		Reference		Reference
	Discharge			−0.27	0.09**	−0.26	0.09**	−0.34	0.10**	−0.43	0.11**	−0.39	0.11**
	1 month after discharge			−0.29	0.10**	−0.26	0.10**	−0.24	0.11**	−0.24	0.11**	−0.22	0.11**
	>1 month after discharge			0.09	0.12	0.11	0.12	0.10	0.12	0.10	0.12	0.05	0.12
Intervention groups	Vascular surgery			−0.39	0.18**	−0.34	0.17*	−0.26	0.16	−0.23	0.17	−0.21	0.17
Interaction	Admission × vascular surgery			Reference		Reference		Reference		Reference		Reference
	Discharge × vascular surgery			−0.02	0.15	−0.03	0.15	−0.13	0.16	−0.17	0.16	−0.24	0.17
	1 month × vascular surgery			0.42	0.17**	0.40	0.17*	0.28	0.17	0.27	0.17	0.21	0.18
	>1 month × vascular surgery			0.03	0.19	0.004	0.19	−0.01	0.19	−0.06	0.19	−0.10	0.19
Findings on admission
Baseline age	1 year			0.16	0.04**	0.13	0.04**	0.11	0.03**	0.10	0.03**	0.09	0.03**
Sex	Female			0.30	0.21	0.33	0.19*	0.09	0.18	0.08	0.18	0.04	0.18
Smoking	Yes			0.14	0.20	−0.06	0.18	−0.08	0.17	−0.09	0.16	−0.11	0.16
Emergency operation	Yes			0.87	0.22**	0.78	0.20**	0.55	0.20**	0.51	0.20**	0.50	0.19**
Medical history
Hypertension	Yes					−0.13	0.17	−0.07	0.16	−0.08	0.15	−0.08	0.15
Dyslipidemia	Yes					−0.19	0.15	−0.10	0.14	−0.09	0.14	−0.07	0.14
Diabetes mellitus	Yes					0.21	0.17	0.14	0.16	0.17	0.15	0.18	0.15
Cardiac disorder	Yes					0.44	0.15**	0.36	0.14**	0.35	0.13**	0.33	0.13**
Respiratory dysfunction	Yes					0.48	0.22**	0.39	0.20**	0.42	0.19**	0.48	0.19**
Digestive disorder	Yes					0.17	0.20	0.17	0.18	0.17	0.18	0.15	0.18
Kidney disorder	Yes					0.59	0.16**	0.60	0.15**	0.55	0.14**	0.44	0.16**
Cerebrovascular disease	Yes					0.16	0.16	0.02	0.15	0.02	0.14	0.03	0.14
Physiological findings
Body mass index	1 kg/m²							−0.06	0.02**	−0.05	0.02**	−0.05	0.02**
Systolic blood pressure	5 mmHg							0.02	0.02	0.02	0.02	0.03	0.02
Pulse rate	10 bpm							−0.001	0.006	−0.001	0.006	−0.003	0.006
Cardiothoracic ratio	5%							0.30	0.04**	0.28	0.04**	0.28	0.04**
Nutritional status
Prognostic nutritional index										−0.08	0.03**	−0.07	0.03**
Blood biochemistry
Estimated glomerular filtration rate	10 mL/min/1.73 m²											−0.039	0.020**
Aspartate aminotransferase	10 IU/L											0.004	0.005
Alkaline phosphatase	10 IU/L											0.0001	0.0033
γ-Guanosine triphosphate	10 IU/L											0.008	0.008
Total bilirubin	1 mg/dl											−0.012	0.063
Intercept		4.57	0.08**	4.21	0.36**	3.87	0.36**	4.23	0.33**	4.28	0.33**	4.38	0.34**
Random-effect parameters (2b)
Individual-level variance	Slope variance	0.034	0.013	0.041	0.012	0.036	0.011	0.020	0.009	0.015	0.009	0.008	0.009
	Intercept variance	1.031	0.137	0.784	0.113	0.566	0.090	0.443	0.074	0.426	0.073	0.435	0.073
	Residual variance	0.548	0.042	0.495	0.038	0.507	0.039	0.487	0.037	0.501	0.039	0.505	0.040
Z-score		7.5		6.9		6.3		5.9		5.8		6.0
Intraclass correlation coefficient		65.3%		61.3%		52.7%		47.6%		46.0%		46.3%
Proportional changes in variance (%)		–		24.0%		45.2%		57.1%		58.7%		57.8%

Model fit statistics (2c)
Log likelihood		−1021.9		−982.0		−946.5		−828.9		−813.2		−784.3
AIC		2051.8		1994.1		1938.9		1711.8		1682.4		1634.6
BIC		2070.0		2062.5		2043.4		1832.1		1806.6		1780.0

BNP, B-type natriuretic peptide; Coeff., coefficient; SE, standard error; AIC, Akaike’s information criterion; BIC, Bayesian information criterion

**P < 0.05, *P < 0.10

Model 1 was the same as the null model with the addition of the time point, patient group, interaction, and findings on admission as covariates to account for compositional differences between patients. Model 2 was the same as Model 1 with the addition of the medical history to account for background differences between patients. Model 3 was the same as Model 2 with the addition of physiological findings to account for measurable status differences between patients.

Model 4 was the same as Model 3 with the addition of the PNI to account for nutritional status differences between patients. Model 5 was the same as Model 4 with the addition of blood biochemistry parameters to account for blood status differences between patients. A variable in the same dataset was centered by subtracting patient means that were designated “group mean centering.”

The end of follow-up was defined as either the day of patient death or the last day that the patient was known to be alive.

While it is reasonable to assume the independence of patients, we would not want to assume that the timing of death within each patient is independent. The models were created to assess the correlation by assuming that death was the result of a patient-level effect. We estimated the hazard ratio (HR) and standard error for all-cause death and vascular death by applying a multilevel parametric proportional hazard regression model with a Weibull distribution. A two-sided P-value of <0.05 was considered to indicate statistical significance. All statistical analyses were performed using Stata/MP® version 14.2 (Stata Corp., College Station, TX, USA).

Results

In total, 197 patients were included in this study. The patients were categorized into two groups based on the type of surgery: cardiac surgery requiring cardiopulmonary bypass (n = 127) and vascular surgery (abdominal aortic aneurysm open repair, n = 47; endovascular aneurysm repair, n = 19; and thoracic endovascular aortic repair, n = 4) (Table 1). We found no statistically significant differences at baseline in age, sex, smoking status, pulse rate, cardiothoracic ratio, history of hypertension and respiratory dysfunction, or aspartate aminotransferase and lnBNP levels. When considering Models 1 to 5 in greater detail, the following parameters were significantly associated with lnBNP: the time point at discharge, the time point at 1 month after discharge, baseline age, emergency operation, history of cardiac disorder and respiratory dysfunction, kidney disorder, physiological findings of body mass index and cardiothoracic ratio, PNI, and estimated glomerular filtration rate. Stable coefficients were established regardless of the model (Table 2a). Models 2 to 5 showed no significant correlation between the extent of lnBNP and the intervention groups.

All Z-scores were >2, and significant differences were found between the individual data. Additionally, both the variance of intercept (1.031) and the variance of slope (0.034) were significant at an individual level, as shown in Table 2, indicating notable variance between individuals. Furthermore, the higher the ICC, the greater the influence on intraindividual variability. The ICC of Model 0 was relatively high at 65.3%, as shown in Table 2; the extent of lnBNP was dependent upon the patients. In Models 1-5, the ICC tended to be lower than that in Model 0 but retained a more than relatively high score of 45.0%. The PCV in Models 1 to 5 improved from 24.0% to 58.7%, and the data suggested that about 60% of the variance between patients (the cause of interobserver variability) could explain the explanatory markers of Model 5 (Table 2b). The PCV in Model 4 was the highest (58.7%) (Table 2b), and the fitting values of the variance error in Model 5 were the lowest (log likelihood, −784.3; Akaike’s information criterion, 1634.6; Bayesian information criterion, 1780.0) (Table 2c).

We performed a multilevel survival analysis including all-cause and vascular death to obtain a model that included significant markers of a multilevel univariate analysis for validation of Model 5 in Table 2. We further evaluated the association between the developmental events of death and the significant markers from the multilevel univariate survival analysis results (baseline age, smoking, emergency operation, kidney disorder, body mass index, diastolic blood pressure, cardiothoracic ratio, PNI, estimated glomerular filtration rate, alanine aminotransferase, alkaline phosphatase, γ-guanosine triphosphate, total bilirubin, and lnBNP). The multivariate multilevel survival analysis results revealed significant differences in the smoking status (HR = 4.87), emergency operation (HR = 8.26), PNI (HR = 1.46 per 1-point increase), and lnBNP (HR = 2.07 per 1-point increase) in the model of all-cause death and in only lnBNP (HR = 3.10 per 1-point increase) in the model of vascular death (Table 3).

Table 3.

Multilevel analysis of two outcomes related to survival data

		Univariate analysis				Multivariate analysis (n = 188)
		All-cause death				All-cause death			Vascular death
Variable	Reference or per increased	Patients (n)	HR	SE	P-value	HR	SE	P-value	HR	SE	P-value
Strategy for therapeutic intervention
Vascular surgery	Cardiac surgery	196	0.60	0.30	0.31	2.02	1.32	0.28	1.54	1.45	0.65
Findings on admission
Sex	Male	196	0.57	0.28	0.26
Baseline age	5 years	196	1.28	0.15	0.028	1.27	0.21	0.15	1.01	0.24	0.98
Smoking	Yes	196	2.48	1.14	0.047	4.87	2.98	0.009	5.77	5.20	0.052
Emergency operation	Yes	196	46.5	24.5	<0.001	8.26	6.34	0.006	7.33	7.99	0.07
Medical history
Hypertension	Yes	196	2.75	1.55	0.07
Dyslipidemia	Yes	195	0.75	0.35	0.54
Diabetes mellitus	Yes	196	1.51	0.74	0.40
Cardiac disorder	Yes	196	1.23	0.56	0.64
Respiratory dysfunction	Yes	196	1.10	0.81	0.90
Digestive disorder	Yes	195	0.53	0.39	0.39
Kidney disorder	Yes	196	7.13	3.32	<0.001	1.50	0.93	0.52	1.96	1.79	0.46
Cerebrovascular disease	Yes	195	0.88	0.47	0.81
Physiological findings
Body mass index	1 kg/m²	193	0.85	0.04	0.001	1.01	0.07	0.87	1.02	0.11	0.89
Systolic blood pressure	10 mmHg	193	0.98	0.12	0.84
Diastolic blood pressure	10 mmHg	193	0.57	0.10	0.001	1.16	0.28	0.55	0.99	0.36	0.99
Pulse rate	5 bpm	193	1.01	0.01	0.59
Cardiothoracic ratio	5%	195	1.43	0.15	0.001	1.00	0.18	0.98	1.16	0.27	0.52
Nutritional status
Prognostic nutritional index		193	2.02	0.37	<0.001	1.46	0.28	0.046	1.45	0.43	0.21
Blood biochemistry
Estimated glomerular filtration rate	10 mL/min/1.73m²	194	0.66	0.06	<0.001	1.03	0.13	0.83	1.14	0.18	0.41
Aspartate aminotransferase	10 IU/L	195	1.02	0.00	<0.001
Alanine aminotransferase	10 IU/L	195	1.03	0.01	0.002	0.99	0.01	0.35	0.94	0.07	0.41
Alkaline phosphatase	10 IU/L	193	1.02	0.00	<0.001	1.01	0.01	0.53	1.02	0.02	0.33
γ-Guanosine triphosphate	10 IU/L	195	1.04	0.01	<0.001	1.02	0.04	0.67	1.00	0.07	0.98
Total bilirubin	1 mg/dl	195	1.33	0.07	<0.001	0.73	0.12	0.058	0.59	0.17	0.073
Total protein	1 g/dl	195	0.21	0.04	<0.001
Log-transformed BNP	1	196	1.98	0.21	<0.001	2.07	0.52	0.004	3.10	1.20	0.004

HR, hazard ratio; SE, standard error; BNP, B-type natriuretic peptide

Discussion

In this study, we observationally examined postoperative survival according to the lnBNP values, physiological findings, nutritional status, and blood biochemistry parameters of patients undergoing cardiovascular surgery and found that lnBNP is a prognostic predictor in these patients. Moreover, the multilevel survival analysis revealed that the lnBNP value was a significant prognostic factor for all-cause death and vascular death. Our findings suggest that the lnBNP value could be a very important prognostic predictor in patients undergoing cardiovascular surgery. This study was a multilevel analysis in which four iterative measurements were considered and five models were examined. Compared with the null model, Models 3 to 5 were able to improve the error between individuals by approximately 60%. This allowed for highly accurate prediction, which is considered clinically beneficial. This study is the first to involve conduction of a multilevel analysis using the lnBNP value and surgical operation.

BNP is the gold standard marker in the diagnosis and prognosis of heart failure.^3,5,13 However, some studies have showed that the BNP level might be limited by its skewed distribution.^10,11 Therefore, we used the natural lnBNP to meet the demands of a normal distribution.

BNP is a strong independent predictor of perioperative and long-term cardiovascular and cardiopulmonary complications following noncardiac surgical procedures.^6,8,14 The postoperative serum concentration of BNP is an independent predictor of mortality in patients undergoing cardiac surgery.¹⁵ In the present study, lnBNP was similarly predictive of the patients’ prognosis. As lnBNP increased by 1 point, all-cause death increased by 2.07 times and vascular death increased by 3.10 times; lnBNP was especially associated with a high risk of all-cause death and vascular death. Additionally, in patients undergoing emergency surgery, all-cause death and vascular death represent a risk of onset of 8 to 9 times. This suggests that postoperative follow-up is very important in patients who have undergone an emergency operation.

The prognostic factors in cardiac surgery are troponin and BNP.^16,17 The survival analysis in the present study showed that lnBNP, smoking, emergency operation, and PNI were prognostic factors in patients undergoing cardiovascular surgery. Smoking and an emergency operation have been reported as risk factors for surgery^18–20 and regulated prognostic factors; these results are consistent with the findings of our study, in which they were also prognostic factors. The PNI was also recently shown to be a predictive marker for postoperative complications in patients with some malignant tumours.^21–23 Our findings suggest that the PNI is a significant prognostic factor for all-cause death in patients undergoing cardiovascular surgery. The present study is likely to be the first to report this finding in patients undergoing cardiovascular surgery.

The lnBNP value was significantly related to laboratory variables reflective of disease severity and to systolic and diastolic myocardial dysfunction, suggesting that the lnBNP concentration may be a quantitative biomarker of myocarditis in patients with Kawasaki disease.¹¹ Yang and Bao²⁴ reported that lnBNP was an independent determinant of pulmonary systolic artery pressure. Mentias et al.²⁵ reported that a higher lnBNP level was significantly associated with increased mortality in asymptomatic patients with mitral regurgitation and a preserved left ventricular ejection fraction. Oikawa et al.²⁶ reported that multiplying lnBNP and the ratio of the mitral inflow early- and late-diastolic filling velocities is a useful parameter for detecting elevated pulmonary capillary wedge pressure regardless of the left ventricular ejection fraction. Zhou et al.²⁷ reported that lnBNP is an independent risk factor for contrast-induced acute kidney injury in patients with acute myocardial infarction. Therefore, we suggest that the lnBNP level is a useful marker that is simple and easy to implement in the clinical setting, not only in patients undergoing surgery but also in those with heart disease. The clinical value of lnBNP for various disease is expected to be revealed in future research.

This study has some limitations. First, this was an observational study. Second, the observation period was short and the number of patients was small. However, this study is the first multilevel analysis among longitudinal observational studies in cardiovascular surgery. Third, this study had biases due to the patients’ backgrounds. Therefore, further studies with longer observation periods and larger sample sizes but without bias due to the patients’ backgrounds are necessary. Multicenter research is needed to validate these findings. Fourth, a preoperative serum BNP assay was not performed in this study because of the biases resulting from the patients’ backgrounds. Finally, we did not analyze cardiac surgery and vascular surgery separately and did not perform non-discrimination between valve surgery and myocardial revascularization in cardiac surgery. Redfors et al.²⁸ reported that in the EXCEL trial, an elevated baseline BNP level in patients with left main coronary artery disease undergoing revascularization was independently associated with long-term mortality. Therefore, each type of surgery should be separately analyzed in future studies.

In conclusion, as the lnBNP level increased by 1 point, all-cause death increased 2.07 times and vascular death increased 3.10 times. The lnBNP value was correlated with survival. Therefore, lnBNP is an important prognostic predictor and quantitative biochemical marker in patients undergoing cardiovascular surgery.

Footnotes

Acknowledgment

The authors are grateful to Shinji Maeda (board-certified supervisory physician in public health and social medicine) for providing assistance with the statistical analysis.

Declaration of conflicting of interest

The authors declare that there is no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Levin

Gardner

Samson

WK.

Natriuretic peptides.

N Engl J Med 1998; 339: 321–328.

Tamura

Intervention for varicose veins of lower extremities lowers the brain natriuretic peptide values in varicose vein patients.

Ann Vasc Dis 2017; 10: 115–118.

Magnussen

Blankenberg

Biomarkers for heart failure: small molecules with high clinical relevance.

J Intern Med 2018; 283: 530–543.

Oremus

Don-Wauchope

McKelvie

et al . BNP and NT-proBNP as prognostic markers in persons with chronic stable heart failure. Heart Fail Rev 2014; 19: 471–505.

Sakamoto

Kanda

Validation of circulating BNP level >1000 pg/mL in all-cause mortality: a retrospective study.

J Int Med Res 2015; 43: 583–591.

Rodseth

Biccard

Chu

et al . Postoperative B-type natriuretic peptide for prediction of major cardiac events in patients undergoing noncardiac surgery: systematic review and individual patient meta-analysis. Anesthesiology 2013; 119: 270–283.

Kristensen

Knuuti

Saraste

et al.

2014 ESC/ESA Guidelines on non-cardiac surgery: cardiovascular assessment and management: the joint task force on non-cardiac surgery: cardiovascular assessment and management of the European Society of Cardiology (ESC) and the European Society of Anaesthesiology (ESA). Eur Heart J 2014; 35: 2383–2431.

Cagini

Andolfi

Leli

et al.

B-type natriuretic peptide following thoracic surgery: a predictor of postoperative cardiopulmonary complications. Eur J Cardiothorac Surg 2014; 46: 74–80.

Mitchell

Webb

ST.

Is brain natriuretic peptide a marker for adverse postoperative outcomes in patients undergoing cardiac surgery?

Interact Cardiovasc Thorac Surg 2011; 12: 467–472.

10.

Izumiya

Hanatani

Kimura

et al . Growth differentiation factor-15 is a useful prognostic marker in patients with heart failure with preserved ejection fraction. Can J Cardiol 2014; 30: 338–344.

11.

Bang

Han

et al . Log-transformed plasma level of brain natriuretic peptide during the acute phase of Kawasaki disease is quantitatively associated with myocardial dysfunction. Korean J Pediatr 2011; 54: 340–344.

12.

Sze

Pellicori

Kazmi

et al.

Prevalence and prognostic significance of malnutrition using 3 scoring systems among outpatients with heart failure: a comparison with body mass index.

JACC: Heart Fail 2018; 6: 487–488.

13.

Piek

de Boer

et al . Novel heart failure biomarkers: why do we fail to exploit their potential? Crit Rev Clin Lab Sci 2018; 55: 246–263.

14.

Rodseth

Biccard

Le Manach

et al . The prognostic value of pre-operative and post-operative B-type natriuretic peptides in patients undergoing noncardiac surgery: B-type natriuretic peptide and N-terminal fragment of pro-B-type natriuretic peptide: a systematic review and individual patient data meta-analysis. J Am Coll Cardiol 2014; 63: 170–180.

15.

Murad Junior

Nakazone

Machado Mde

et al . Predictors of mortality in cardiac surgery: brain natriuretic peptide type B. Rev Bras Cir Cardiovasc 2015; 30: 182–187.

16.

Lurati Buse

Koller

Grapow

et al . The prognostic value of troponin release after adult cardiac surgery - a meta-analysis. Eur J Cardiothorac Surg 2010; 37: 399–406.

17.

Lurati Buse

Bolliger

Seeberger

et al . Troponin T and B-type natriuretic peptide after on-pump cardiac surgery: prognostic impact on 12-month mortality and major cardiac events after adjustment for postoperative complications. Circulation 2014; 130: 948–957.

18.

Grønkjær

Eliasen

Skov-Ettrup

et al . Preoperative smoking status and postoperative complications: a systematic review and meta-analysis. Ann Surg 2014; 259: 52–71.

19.

Seok

Hong

Lee

Impact of smoking history on postoperative pulmonary complications: a review of recent lung cancer patients.

Ann Thorac Cardiovasc Surg 2014; 20: 123–128.

20.

Guevara

Rubio-Romero

Ruiz-Parra

AI.

Unplanned reoperations: is emergency surgery a risk factor? A cohort study.

J Surg Res 2013; 182: 11–16.

21.

Feng

Chen

QX.

Significance of the prognostic nutritional index in patients with esophageal squamous cell carcinoma.

Ther Clin Risk Manag 2014; 10: 1–7.

22.

Tokunaga

Sakamoto

Nakagawa

et al . Prognostic nutritional index predicts severe complications, recurrence, and poor prognosis in patients with colorectal cancer undergoing primary tumor resection. Dis Colon Rectum 2015; 58: 1048–1057.

23.

Qiu

Shen

et al . Evaluation of prognostic nutritional index in patients undergoing radical surgery with nonsmall cell lung cancer. Nutr Cancer 2015; 67: 741–747.

24.

Yang

Bao

XR.

Pulmonary hypertension in patients with stage 1-3 chronic kidney disease.

Genet Mol Res 2014; 13: 5695–5703.

25.

Mentias

Patel

et al . Prognostic utility of brain natriuretic peptide in asymptomatic patients with significant mitral regurgitation and preserved left ventricular ejection fraction. Am J Cardiol 2016; 117: 258–263.

26.

Oikawa

Kobayashi

Wada

et al . Novel parameter to predict left ventricular filling pressure: product of log B-type natriuretic peptide and ratio of mitral inflow early and late diastolic filling velocities. J Card Fail 2018; 24: 29–32.

27.

Zhou

Sun

Zhuang

et al.

Development and validation of nomogram to predict acute kidney injury in patients with acute myocardial infarction treated invasively. Sci Rep 2018; 8: 9769.

28.

Redfors

Chen

Crowley

et al . B-type natriuretic peptide assessment in patients undergoing revascularization for left main coronary artery disease: analysis from the EXCEL trial. Circulation 2018; 138: 469–478.