Skip to main content

Table 5 Comparison of risk prediction models before and after the addition of hemodynamics for all-cause mortality

From: Risk prediction in pulmonary hypertension due to chronic heart failure: incremental prognostic value of pulmonary hemodynamics

 

C-statistic

AIC

LR test, p value

NRI (95% CI)

P value

IDI (95% CI)

P value

SHFM

SHFM only

0.593 (0.519–0.667)

557

–

–

–

–

–

SHFM + hemodynamicsa (continuous)c

0.715 (0.648–0.782)

535

< 0.001

0.468 (0.161–0.752)

< 0.01

0.092 (0.035–0.171)

< 0.01

SHFM + hemodynamicsa (dichotomousb)c

0.684 (0.610–0.758)

545

< 0.001

0.230 (0.029–0.642)

< 0.01

0.064 (0.018–0.141)

< 0.01

MAGGIC

MAGGIC only

0.620 (0.540–0.700)

549

–

–

–

–

–

MAGGIC + hemodynamicsa (continuous)d

0.716 (0.647–0.785)

529

< 0.001

0.298 (0.106–0.615)

0.02

0.084 (0.033–0.151)

< 0.01

MAGGIC + hemodynamicsa (dichotomousb)d

0.689 (0.613–0.765)

538

< 0.001

0.162 (0.016–0.510)

0.01

0.066 (0.028–0.155)

< 0.01

  1. AIC, Akaike information criteria; LR test, likelihood ratio test; NRI, net reclassification improvement; IDI, integrated discriminatory index; CI, confidence interval; SHFM, the Seattle heart failure model; MAGGIC, Meta-Analysis Global Group in Chronic Heart Failure
  2. aHemodynamics included: diastolic pressure gradient (DPG) and mixed venous oxygen saturation (SvO2)
  3. bCutoff values for DPG and SvO2: 7 mmHg and 65%, respectively
  4. cCompared to SHFM risk score
  5. dCompared to MAGGIC risk score