Skip to main content

Table 2 Diagnostic accuracy of NT-proBNP as rule-out test for LV dysfunction (types as specified)*

From: Utility of NT-proBNP as a rule-out test for left ventricular dysfunction in very old people with limiting dyspnoea: the Newcastle 85+ Study

 

Prevalence of specified LV dysfunction % (n)

AUC (95% CI)

NT-proBNP cut point (ng/l)

Sensitivity %

Specificity %

PPV %

NPV %

% of participants at or above cut point (echo warranted)

Number of cases picked up per 100 screened

Number of cases missed per 100 screened

Number of false positives per 100 screened

Systolic dysfunction, any grade (LVEF≤50%)

34.2 (53)

0.58 (0.49-0.65)

         

ESC guideline rule-out cut point, 125ng/l

  

125

98.1

14.7

37.4

93.8

87.7

33.5

0.6

56.1

NICE guideline rule-out cut point, 400ng/l

  

400

54.7

50.0

36.3

68.0

51.0

18.7

15.5

32.9

Data-derived 'stringent' rule-out cut point (closest to 95% sensitivity)

  

131

96.2

16.7

37.5

89.5

87.7

32.9

1.3

54.8

Data-derived 'optimised' rule-out cut point (highest sensitivity with specificity at least 50%)

  

380

56.6

50.0

37.0

68.9

52.3

19.3

14.8

32.9

Moderate/severe systolic dysfunction (LVEF≤40%)

8.4 (13)

0.80 (0.73-0.86)

         

ESC guideline rule-out cut point, 125ng/l

  

125

100.0

11.3

9.4

100.0

87.7

8.4

0.0

81.2

NICE guideline rule-out cut point, 400ng/l

  

400

84.6

51.4

13.8

97.3

51.0

7.1

1.3

44.5

Data-derived 'stringent' rule-out cut point (closest to 95% sensitivity)

  

197

100.0

26.8

11.1

100.0

75.5

8.4

0.0

67.1

Data-derived 'optimised' rule-out cut point (highest sensitivity with specificity at least 50%)

  

363

84.6

50.0

13.4

97.3

52.9

7.1

1.3

45.8

Systolic dysfunction (any grade) OR isolated moderate/severe diastolic dysfunction

53.5 (83)

0.64 (0.56-0.72)

         

ESC guideline rule-out cut point, 125ng/l

  

125

95.2

16.7

56.8

75.0

87.7

50.9

2.6

38.7

NICE guideline rule-out cut point, 400ng/l

  

400

60.2

58.3

62.5

56.0

51.0

32.3

21.3

19.4

Data-derived 'stringent' rule-out cut point (closest to 95% sensitivity)

  

120

95.2

16.7

56.8

75.0

89.7

50.9

2.6

38.7

Data-derived 'optimised' rule-out cut point (highest sensitivity with specificity at least 50%)

  

319

63.9

51.4

60.2

55.2

56.8

34.2

19.4

22.6

Moderate/severe systolic dysfunction OR isolated moderate/severe diastolic dysfunction

27.7 (43)

0.71 (0.63-0.78)

         

ESC guideline rule-out cut point, 125ng/l

  

125

93.0

11.6

28.8

81.3

87.7

25.8

1.9

63.9

NICE guideline rule-out cut point, 400ng/l

  

400

74.4

57.1

40.0

85.3

51.0

20.6

7.1

31.0

Data-derived 'stringent' rule-out cut point (closest to 95% sensitivity)

  

113

95.3

11.6

29.3

86.7

90.3

26.5

1.3

63.9

Data-derived 'optimised' rule-out cut point (highest sensitivity with specificity at least 50%)

  

298

79.1

50.0

37.8

86.2

58.1

21.9

5.8

36.1

  1. *Abbreviations: AUC area under curve, PPV positive predictive value, NPV negative predictive value, ESC European Society of Cardiology, NICE National Institute for Health and Care Excellence.
  2. Detailed legend: Diagnostic accuracy of NT-proBNP as rule-out test for LV dysfunction (types as specified) at range of guideline recommended and data-derived rule-out cut points.
Back to article page

BMC Cardiovascular Disorders

ISSN: 1471-2261

Contact us