N-Terminal Prohormone Brain Natriuretic Peptide as a Prognostic Biomarker for the Risk of Complications in Type 2 Diabetes

A Systematic Review and Meta-Analysis

Zhian Salah Ramzi, PhD

Lab Med. 2023;54(4):339-351.

Abstract and Introduction

Abstract

Objective: This systematic review and meta-analysis aimed at summarizing the existing clinical evidence to evaluate the prognostic performance of N-terminal prohormone brain natriuretic peptide (NT-proBNP) in predicting cardiovascular events, cardiovascular-related mortality, and all-cause mortality in patients with type 2 diabetes.

Methods: Searches were performed in Medline, Embase, Scopus, and Web of Science databases before August 1, 2021. The data were recorded as adjusted hazard ratio (HR).

Results: An increase in NT-proBNP increases the risk of cardiovascular events (HR = 1.63), cardiovascular mortality (HR = 1.86) and all-cause mortality (HR = 1.54). Seemingly, the best cutoffs for predicting cardiovascular events (HR = 2.30) and cardiovascular mortality (HR = 3.77) are levels greater than 100 pg/mL. The best cutoff of NT-proBNP in predicting all-cause mortality is levels greater than 225 pg/mL (HR = 4.72).

Conclusion: A moderate level of evidence demonstrated that NT-proBNP serum levels can predict future cardiovascular events, cardiovascular mortality, and all-cause mortality. Thus, it can be used as risk stratification for type 2 diabetes.

Introduction

The prevalence of type 2 diabetes has greatly increased in recent years, and the disease has been recognized as a major health problem both in developing and developed societies.^[1] Type 2 diabetes is one of the leading causes of cardiovascular disease, which is one of the main causes of mortality and disease burden in many countries.^[2] Statistics show that about 6.2% of the world's population have type 2 diabetes, resulting in the death of more than 1 million people annually. These numbers also show that type 2 diabetes is the ninth leading cause of death worldwide. Current estimates suggest that if the trend of type 2 diabetes prevalence continues at this rate, its prevalence will rise to 10% of the world's population in 2030.^[3] Cardiovascular events and mortality are the main complications of type 2 diabetes. All of the existing interventions, including lifestyle modifications and drug treatments, are therefore aimed at preventing cardiovascular complications and death in type 2 diabetes patients.^[4]

Although risk stratification of type 2 diabetes patients is based on the risk factors of cardiovascular diseases such as hypertension, there is still a need for other prognostic factors for better identifying at-risk persons.^[5] Hence, researchers are investigating diagnostic tests and different biomarkers for diagnosis and prediction of the outcome in diabetes.

Natriuretic peptides are a family of 3 similar hormones: atrial natriuretic peptide (ANP), brain-type natriuretic peptide (BNP), and C-type natriuretic peptide.^[6] N-terminal-prohormone BNP (NT-proBNP) is an inactive prohormone, secreted from cells along with BNP.^[7] The prognostic value of these hormones and prohormones in predicting the outcome of diabetes has been evaluated in a number of studies. Occasionally, the findings of these studies are in contrast to one another, resulting in a consensus yet to be achieved. For instance, Ponikowska et al^[8] report that NT-proBNP serum levels cannot predict the incidence of cardiovascular events, whereas Bidakosh et al^[9] present a considerable prognostic value for this biomarker.

In addition, the optimum cutoffs of natriuretic peptides, wherein lie the best prognostic value for type 2 diabetes and the most appropriate peptide in predicting mortality and cardiovascular events, are not clear. To resolve these inconsistencies and respond to the uncertainties, this systematic review and meta-analysis is designed to evaluate the prognostic performance of NT-proBNP in predicting cardiovascular events, mortality rate, and all-cause mortality in type 2 diabetes patients based on the existing clinical evidence.

1 2 3 4 5

Next Section

Table 1. Summary of Included Studies
Study	Design	Condition	Sample Size	Mean Age, y	N, Male	Duration of Diabetes	FU, y	Comorbidity (%)				Albuminuria
Study	Design	Condition	Sample Size	Mean Age, y	N, Male	Duration of Diabetes	FU, y	CVD	HTN	Smoking	HLP	Albuminuria
Bidadkosh, 2017, Netherlands	RCT	Diabetic nephropathy	861	64	654	NR	0.8	NR	NR	NR	NR	NR
Birukov, 2020, Germany	PCS	General	545	NR	NR	NR	8	NR	NR	NR	NR	NR
Bruno, 2013, Italy	PCS	Elderly	1825	67.6	894	10.7	5.5	22.1	89.4	28.7	NR	25.2
Busch, 2021, Denmark	PCS	General	1030	65.2	679	11	4.7	18.6	NR	57.2	NR	22.7
Clodi, 2011, Austria	PCS	General	1071	61.5	613	15	2.75	23.7	67.5	NR	NR	23.7
Colombo, 2018, UK	RCT	Patients without history of CVD	2105	59.8	1423	6.7	5	NR	NR	65.6	NR	28.3
Czucz, 2011, Hungary	PCS	Chronic HF	192	69.4	142	NR	1.3	100	NR	NR	NR	NR
Gaede, 2005, Denmark	RCT	Diabetic nephropathy	160	55	119	NR	7.8	NR	NR	37.5	NR	NR
Gori, 2016, USA	PCS	General	1510	63	725	NR	13.1	12	60	NR	NR	NR
Gustafsson, 2010, Europe	RCT	General	389	NR	NR	NR	2.2	NR	NR	NR	NR	NR
Hillis, 2014, Australia	RCT	General	3862	66.89	2356	7.86	5	49.5	NR	15	NR	NR
Huelsmann, 2008, Austria	PCS	General	631	58.7	349	9.28	1	22.8	81.3	56.6	NR	NR
Jarolim, 2018, USA	RCT	CVD patients	5224	61	3539	9.6	1.64	96.2	83.2	13.6	NR	NR
Jhund, 2014, USA	RCT	General	8415	NR	NR	NR	NR	NR	NR	NR	NR	NR
Liu, 2021, China	PCS	General	2952	59.1	2076	NR	5	55.7	71	71	NR	NR
Malachias, 2020, 36 countries	RCT	General	5509	64.4	3811	NR	2.6	87.7	NR	50.8	NR	NR
Neuhold, 2011, Austria	PCS	General	544	60.5	299	13	3.3	38.5	65.3	55.4	NR	NR
Nguyen, 2020, USA	PCS	General	659	64.8	365	NR	12.4	NR	64.9	49.4	NR	NR
O'Donoghue, 2007, USA	PCS	General	157	68	82	NR	1	73.9	66	NR	NR	NR
Pfister, 2007, Germany	PCS	General	156	69	110	6.5	3.2	61.5	75.6	NR	NR	NR
Ponikowska, 2013, Poland	PCS	General	287	65	225	5	5	NR	NR	NR	NR	NR
Prausmüller, 2021, Austria	PCS	General	1690	62	907	11.3	10	24	67	20	68	NR
Price, 2014, UK	PCS	Elderly	1066	67.9	547	6.7	4	56.3	82.3	61.2	NR	16.5
Riphagen, 2015, Netherlands	PCS	General	1068	67	483	5	15	35.8	NR	21.6	37.2	40
Rørth, 2019, Denmark	RCT	Chronic HF	754	67	618	NR	3	100	86	12	NR	NR
Saely, 2019, Switzerland	PCS	CVD patients	317	NR	NR	NR	6.3	100	NR	NR	NR	NR
Tarnow, 2006, Denmark	PCS	General	315	54.1	194	6.8	15.5	33.2	37.6	47	NR	40.3
Trutz, 2015, Romania	PCS	CVD patients	206	62.5	108	NR	1	100	NR	NR	47.1	NR
Winkler, 2008, Germany	RCT	Hemodialysis	1255	65.6	672	NR	3.96	81.9	NR	40.2	NR	NR
Zanolin, 2018, Switzerland	PCS	PAD	134	NR	NR	NR	6	NR	NR	NR	NR	NR

BNP, B-type natriuretic peptide; CVD, cardiovascular disease; FU, follow up duration; General, all diabetic patients regardless comorbidities; HF, heart failure; HTN, hypertension; MR-ANP, midregional fragment of pro-A–type natriuretic peptide; NT-proBNP, N-terminal pro-B–type natriuretic peptide; NP, natriuretic peptide; NR, not reported; PAD, peripheral artery diseases; PCS, prospective cohort studies; RCT, randomized controlled trial.

Table 2. NT-proBNP Cutoffs, Assessed Outcomes, and Prevalence of Events among Included Studies
Study	Cutoff (pg/mL)	Outcome	No Event (n)	Event (n)	Prevalence
Bidadkosh, 2017, Netherlands	Continuous	MACE	754	107	12.43
Birukov, 2020, Germany	Continuous	MACE	495	50	9.17
Bruno, 2013, Italy	Continuous 41–90 91–200 >200	CVD mortality All-cause mortality	1650 1435	175 390	9.59 21.37
Busch, 2021, Denmark	>150	MACE	856	174	16.89
Clodi, 2011, Austria	Continuous	MACE	968	103	9.62
Colombo, 2018, UK	Continuous	MACE	1961	144	6.84
Czucz, 2011, Hungary	Continuous	All-cause mortality	146	46	23.96
Gaede, 2005, Denmark	>22.5	MACE	53	66	41.25
Gori, 2016, USA	>125	MACE	970	540	35.76
Gustafsson, 2010, Europe	Continuous	All-cause mortality	293	96	24.68
Hillis, 2014, Australia	Continuous	MACE All-cause mortality	3153 3156	709 706	18.36 18.28
Huelsmann, 2008, Austria	Continuous	MACE	587	44	6.97
Jarolim, 2018, USA	154.1–420.4 420.4–1084 ≥1084	MACE	4647	577	12.42
Jhund, 2014, USA	Continuous	MACE	7286	1129	13.42
Liu, 2021, China	Continuous 110.6–518.8 >518.8	MACE	2695	257	8.71
Malachias, 2020, 36 countries	Continuous	MACE All-cause mortality	4741 5040	768 469	13.94 8.51
Neuhold, 2011, Austria	Continuous	MACE All-cause mortality	481 517	63 27	11.58 4.96
Nguyen, 2020, USA	Continuous	MACE	488	171	25.95
O'Donoghue, 2007, USA	986	All-cause mortality	129	28	17.83
Pfister, 2007, Germany	518	MACE All-cause mortality	108 122	48 34	30.77 21.79
Ponikowska, 2013, Poland	Continuous	MACE All-cause mortality	227 228	60 59	20.91 20.56
Prausmüller, 2021, Austria	Continuous >122 >125 >376	MACE CVD mortality All-cause mortality	1322 1489 1242	368 201 448	21.78 11.89 26.51
Price, 2014, UK	Continuous	MACE	952	114	10.69
Riphagen, 2015, Netherlands	Continuous	CVD mortality All-cause mortality	843 549	225 519	21.07 48.60
Rørth, 2019, Denmark	Continuous 628–1342 >1342	CVD mortality All-cause mortality	NR NR	NR NR	NR NR
Saely, 2019, Switzerland	Continuous	MACE	157	160	50.47
Tarnow, 2006, Denmark	41–103 >103	CVD mortality	196 153	119 162	37.78 51.43
Trutz, 2015, Romania	400	MACE	138	68	33.01
Winkler, 2008, Germany	Continuous 1434–3361 3362–9251 9252	MACE CVD mortality All-cause mortality	790 1095 643	465 160 612	37.05 12.75 48.76
Zanolin, 2018, Switzerland	Continuous	CVD mortality All-cause mortality	110 72	24 62	17.91 46.27

CVD, cardiovascular disease; MACE, major adverse cardiovascular events.

Table 3. Subgroup Analysis According to Outcome
Subgroups	Effect Size		Heterogeneity		Meta-Regression
Subgroups	Adjusted HR (95% CI)	P Value	I²	P Value	Coefficient (95% CI)	P Value
Cardiovascular events
Duration of diabetes, y
5–10	1.71 (1.33, 2.20)	<.0001	92.29	<.0001	Reference	—
≥10	1.99 (1.18, 3.37)	.010	97.86	<.0001	0.13 (−0.39, 0.66)	.630
Follow up duration, y
0.8–1	2.22 (1.59, 3.10)	<.0001	61.76	.095	Reference	—
1.1–5	1.69 (1.39, 2.05)	<.0001	97.76	<.0001	−0.28 (−0.75, 0.18)	.232
≥5	1.39 (1.26, 1.53)	<.0001	18.09	.049	−0.38 (−0.91, 0.16)	.168
Condition
General diabetic	1.69 (1.43, 2.00)	<.0001	97.72	<.0001	Reference	—
Cardiovascular diabetic	1.74 (1.31, 2.31)	<.0001	91.57	<.0001	0.03 (−0.30, 0.36)	.845
Sample size
Less than 1000	1.69 (1.23, 2.32)	.001	98.27	<.0001	Reference	—
More than 1000	1.77 (1.54, 2.04)	<.0001	91.82	<.0001	0.10 (−0.18, 0.39)	.485
Cardiovascular mortality
Duration of diabetes, y
5–10	1.64 (1.10, 2.45)	.015	56.58	.129	Reference	—
≥10	2.31 (1.02, 5.22)	.044	97.13	<.0001	0.27 (−0.86, 1.40)	.638
Follow-up duration, y
0.8–1	No data		No data		No data
1.1–5	1.53 (1.07, 2.20)	.021	89.88	.002	Reference	—
≥5	2.08 (1.26, 3.43)	<.0001	96.47	<.0001	0.30 (−0.54, 1.14)	.479
Condition
General diabetic	2.31 (1.28, 4.16)	.005	97.12	<.0001	Reference	—
Cardiovascular diabetic	1.49 (1.17, 1.88)	.001	77.01	.007	−0.44 (−1.16. 0.28)	.231
Sample size
Less than 1000	1.72 (1.36, 2.18)	<.0001	42.21	.174	Reference	—
More than 1000	2.00 (1.06, 3.76)	.031	98.66	<.0001	0.12 (−0.69, 0.930)	.771
All-cause mortality
Duration of diabetes, y
5–10	1.77 (1.21, 2.61)	.004	95.31	<.0001	Reference	—
≥10	1.63 (0.88, 3.00)	.118	99.20	<.0001	−011 (−0.82, 0.59)	.750
Follow up duration, y
0.8–1	3.42 (1.09, 10.72)	.035	0.00	>.999	Reference	—
1.1–5	1.58 (1.23, 2.02)	<.0001	98.12	<.0001	−0.77 (−2.13. 0.59)	.267
≥5	1.78 (1.28, 2.48)	.001	96.28	<.0001	−0.65 (−2.03, 0.72)	.351
Patient condition
General diabetic	1.69 (1.34, 2.13)	<.0001	98.48	<.0001	Reference	—
Cardiovascular diabetic	1.84 (1.01, 3.35)	.047	96.87	.001	−0.001 (−0.48, 0.48)	.999
Sample size
Less than 1000	1.87 (1.27, 2.75)	.001	95.98	<.0001	Reference	—-
More than 1000	1.65 (1.30, 2.10)	<.0001	97.59	<.0001	−0.05 (−0.48, 0.38)	.815