Table 1 Performance of the prediction of 3 biomarkers with the most evidence.

From: Precision prognostics for cardiovascular disease in Type 2 diabetes: a systematic review and meta-analysis

Biomarker

Study

Clinical factors / biomarkers to be compared

Improvement in C-statistics

NRI

IDI

Yes /No

Estimate (95% CI) and P-value

NRI (95% CI)

P-value

IDI (95% CI)

P-value

NT-proBNP

Sharma 2020

Age, sex, SBP, history of HF, duration of diabetes, prior MI, hypertension, hyperlipidemia, smoking, eGFR.

Yes

0.05 (NR)

0.39

(0.30–0.47)

NR

0.09

(0.08–0.10)

<0.05

Wolsk 2017

Prior MI, BMI, NSTEMI (index event), heart rate, HbA1c, percutaneous coronary intervention at the index event, cerebrovascular disease, AF, prior HF, sodium concentration, macroalbuminuria, PAD, age, and LDL concentration.

Yes

0.01 (CI NR),

P < 0.05

0.11

(5.7–16.6)

<0.05

0.08

(0.03–1.6)

<0.05

Wong 2019

Age ≥ 65 years, Male, T2D, Hypertension

Yes

0.03 (CI NR),

P = 0.001

0.35

(0.24–0.45)

<0.001

0.01

(0.01–0.02)

<0.001

Scirica

2016

Treatment arms, age, SBP, sex, history of HF-, duration of diabetes, prior MI, history of hypertension, history of hyperlipidemia, smoking, and eGFR

Yes

0.07 (CI NR),

P < 0.001

0.040

(0.03–0.04)

<0.05

0.48

(0.41–0.55)

<0.05

Van der Leeuw 2016

Female sex, age at diabetes diagnosis, duration of diabetes, HbA1c, square of HbA1c, SBP, square of SBP, TC/HDL ratio, urinary ACR, current smoking status, history of major macrovascular disease

Yes

0.02 (0.00–0.04), P-value NR

0.2

(0.10–0.44)

<0.05

NR

NR

Van der Leeuw 2016

Female sex, age at diabetes diagnosis, duration of diabetes, HbA1c, systolic blood pressure, TC/HDL ratio, eGFR, current smoking status, history of major macrovascular disease

Yes

0.02 (0.00–0.05), P-value NR

0.50

(0.26–0.73)

<0.05

NR

NR

TnT

Lepojarvi

2016

Age, sex, history of acute MI, BMI, Canadian Cardiovascular Society grading of angina pectoris, left ventricular ejection fraction and mass index, HDL cholesterol, ACR, HbA1c, and type of glucose metabolism disorder

Yes

0.10

(CI and P-value NR),

0.231

(0.067–0.394)

<0.01

0.05

(0.03–0.08)

<0.001

Scirica

2016

Treatment arms, age, SBP, sex, history of HF, duration of diabetes, prior MI, history of hypertension, history of hyperlipidemia, smoking, and eGFR

Yes

0.07 (CI NR),

P < 0.001

0.440

(0.380–0.510)

<0.05

0.03

(0.02–0.03)

<0.05

Rørth

2019

Natural logarithm of NT-proBNP, age, sex, treatment effect, ejection fraction, NYHA class, BMI, heart rate, SBP, creatinine, LDL, prior angina pectoris, AF and pacemaker implantation.

Yes

0.02 (CI NR),

P = 0.02

0.150

(0.051–0.261)

0.007

0.03

(0.01–0.06)

<0.001

CCTA

Lee

2017

Age, male sex, HTN, smokers, hyperlipidemia, eGFR, and HbA1c

Yes

0.07 (CI NR),

P = 0.03

0.550

(0.343–0.757)

<0.0001

0.05

(0.02–0.07)

0.0006

Halon

2016

UKPDS and log CAC Score

Yes

0.35 (CI NR),

P = 0.021

0.63

(CI NR)

NR

0.65

(CI NR)

NR

  1. Data on improvement in C-statistics was collected from the study, either as reported or derived by comparing the C-statistic from the reference model with the C-statistic obtained from the combination of the reference model and the novel biomarker. NR Not Reported, SBP systolic blood pressure, HF heart failure, MI myocardial infarction, BMI body mass index, AF atrial fibrillation, PAD peripheral artery disease, T2D Type 2 diabetes, eGFR estimated glomerular filtration rate, ACR albumin-creatinine ratio. P-values were extracted from studies as reported, therefore some p-values are given as <0.05 or <0.001 because exact p-values were unavailable.
Back to article page