Notesale: Turn your study into money

Document Preview

Extracts from the notes are below, to see the PDF you'll receive please use the links above

---

Downloaded from http://heart
...
com/ on March 30, 2015 - Published by group
...
com
605

CARDIOVASCULAR MEDICINE

Predicting mortality in patients with heart failure:
a pragmatic approach
M L Bouvy, E R Heerdink, H G M Leufkens, A W Hoes

...

Correspondence to:
Dr Marcel L Bouvy,
Department of
Pharmacoepidemiology
and Pharmacotherapy,
Utrecht Institute for
Pharmaceutical Sciences
(UIPS), PO Box 80082,
3508 TB Utrecht,
Netherlands;
m
...
uu
...


Objective: To develop a comprehensive and easily applicable prognostic model predicting mortality
risk in patients with moderate to severe heart failure
...

Setting: Seven general hospitals in the Netherlands
...
Duration of follow up was at least 18 months
...
The area under receiver
operating characteristic curves (AUC) was used to estimate the predictive ability of the prognostic models
...
Independent predictors of mortality were diabetes mellitus, a history of renal dysfunction (or higher creatinine), New York Heart
Association (NYHA) functional class III or IV, lower weight or body mass index, lower blood pressure,
ankle oedema, and higher scores on a disease specific quality of life questionnaire
...
These factors were used to derive various prediction formulas
...
77
...
80
...
84
...
85)
...


M

ortality among patients with heart failure discharged
from hospital has repeatedly been reported to be
high
...
A wide variety of factors is reported to be associated
with an increased risk of hospital admission or death, including
demographic factors (for example, male sex and single marital
status), clinical characteristics (lower systolic blood pressure,
renal dysfunction), history of heart failure (previous hospital
admissions), and comorbidity (diabetes and depression)
...
5–7
Our aim in this study was to develop a comprehensive and
easily applicable prognostic model predicting the risk of death
in patients with heart failure, based on information that is
readily available in medical practice
...
7
years, range 37–91 years) enrolled in a randomised controlled
trial evaluating the effect of a pharmacist led intervention on
drug compliance in patients with heart failure
...
Eligible patients with heart failure were
included by their cardiologist
...
Patients with severe psychiatric
problems or dementia, those with a planned admission to a
nursing home, those who did not manage their own drug
treatment (for example, where the treatment was given by
relatives or district nurses), and those with a life expectancy of
less than three months were excluded from the study
...
Most of the patients (70%) were enrolled in two large
regional hospitals (containing more than 500 beds)
...

As potential prognostic determinants, information from the
patients’ medical history, physical examination, and laboratory tests was obtained from hospital records, while quality of
life was assessed using a generic questionnaire (COOP/
WONCA charts) and a disease specific questionnaire (Minnesota “living with heart failure” questionnaire)
...

Data analysis
Crude risk ratios for 18 months mortality were calculated for
all potential prognostic determinants
...

Abbreviations: AUC, area under receiver operating characteristic
curve; NYHA, New York Heart Association; ROC, receiver operating
characteristic curve

www
...
com

Downloaded from http://heart
...
com/ on March 30, 2015 - Published by group
...
com
606

Bouvy, Heerdink, Leufkens, et al

Table 1

Crude association of potential prognostic determinants with 18 months mortality in heart failure patients
Survivors (n=101)

Deaths (n=51)

Variable

n

%

n

%

95% CI

p Value

Age (years)
Female
NYHA class I or II
NYHA class III or IV

69
67
56
27

66
...
5%
32
...
7%
25
...
0%

1
...
0 to 1
...
1 (0
...
2)
Reference
6
...
8 to 13
...
07
0
...
8%
23
...
5%
50
...
3%
6
...
9%

11
19
32
30
35
12
7

21
...
3%
62
...
8%
68
...
5%
13
...
3
1
...
7
1
...
0
4
...
3

0
...
08
0
...
33
0
...
006
0
...
7%
19
18
...
8%
78 (15)
79 (15)
27 (5)
8
...
1)
120 (43)
61 (26)
140 (4)
4
...
5)

25
49
...
4%
15
29
...
0 (1
...
3 (0
...
9 (1
...
0)]
1
...
8 to 3
...
1 (0
...
6)
1
...
99 to 1
...
97 (0
...
99)
0
...
79 to 0
...
83 (0
...
15)
1
...
0 to 1
...
97 (0
...
99)
0
...
83 to 1
...
33 (0
...
61)

0
...
14
0
...
30
0
...
007
0
...
01
0
...
06
0
...
3
1
...
9
0
...
0
0
...
4
0
...
5
1
...
2
0
...
12
0
...
75
0
...
94
0
...
39
0
...
23
0
...
59
0
...
31
0
...
0%
11
...
3%
20
...
7%
48
...
8%
67
...
6%
7
...
6%
28
...
9%
15
...
7%
9
...
3%
21
...
4%
54
...
9%
9
...
1%
17
...
5 to 2
...
9 to 4
...
9 to 3
...
7 to 2
...
5 to 2
...
5 to 11
...
5 to 3
...
6 to 61
...
5 to 3
...
4 to 1
...
1 to 1
...
5 to 2
...
1 to 0
...
7 to 2
...
3 to 1
...
8 to 3
...
4 to 4
...
6 to 2
...
2 to 1
...
04 (0
...
14)
1
...
00 to 1
...
000

All values are mean (SD) or proportion
...


were initially analysed without categorisation, but various cut
off values were evaluated as well
...
10, together with age and sex, were included in
multivariable logistic regression analyses
...

Hence, we first included all variables from the patient’s history
into an overall “history model
...
10
...
For each model, the reliability (goodness
of fit) was quantified using the Hosmer and Lemeshow test
...
The statistic can be used to determine if the model
provides a good fit for the data
...
9 The predicted
values from the logistic regression model were used to
construct receiver operating characteristic (ROC) curves and
to calculate the area under the ROC curves (AUC)
...
heartjnl
...
5 (no discrimination,
like a coin flip) to 1
...
Subsequently,
to obtain an easily applicable prediction rule, the adjusted
regression coefficients of the model were multiplied by a factor 10 and rounded to the nearest integer
...
11 12 To decrease bias and increase statistical efficiency, it is better to impute these missing values rather than
doing a complete case analyses
...
0) software
...


RESULTS
Within the 18 months follow up period, 51 (34%) of the 152
heart failure patients died (mortality at six and 12 months, 26
(17%) and 43 (28%), respectively)
...
bmj
...
bmj
...
00
0
...
35
4
...
82
0
...
16

1
...
52
2
...
22
2
...
96
2
...
77 (0
...
84)

(0
...
04)
(0
...
43)
(1
...
77)
(1
...
38)
(1
...
71)
(0
...
98)
(1
...
25)

(0
...
04)
(0
...
34)
(1
...
85)
(1
...
45)
(1
...
82)
(0
...
99)
(1
...
22)

Clinical model + drug
treatment at baseline +
NYHA class

Clinical model + drug
treatment at baseline +
NYHA class + quality of
life score

1
...
04
2
...
14
1
...
96
1
...
00
1
...
76
3
...
62
0
...
81

(0
...
04)
(0
...
47)
(1
...
38)
(1
...
48)
(0
...
42)
(0
...
99)
(0
...
04)

(0
...
04)
(0
...
09)
(1
...
07)
(1
...
63)
(0
...
65)
(0
...
99)
(0
...
83)

3
...
73 to 7
...
40 (1
...
59)
4
...
33 to 10
...
21 (1
...
23)
4
...
95 to 8
...
24 (1
...
62)

0
...
72 to 0
...
84 (0
...
90)*

0
...
79 to 0
...

*Without oedema and lower systolic or diastolic blood pressure, AUC = 0
...

†Without oedema and lower systolic or diastolic blood pressure, AUC = 0
...

CI, confidence interval; MHFQ, Minnesota heart failure questionnaire; NYHA, New York Heart Association; ROC, receiver operating characteristic
...
006
0
...
86
1
...
03
−0
...
74
1
...
06
+4
+9
+17
+10
−0
...

†Diastolic blood pressure < 70 mm Hg or systolic blood pressure < 110 mm Hg
...
0
10
...
3
46
...
3

Death‡

Survival‡

3
3
2
12
13
18
51

22
26
22
14
12
5
101

Values represent absolute number of patients, except for incidence of mortality (%)
...

†Observed incidence of mortality per score category
...


known in 35 of these patients: they all died from a cardiovascular cause (for example, terminal heart failure, sudden
cardiac death)
...
Table 1 shows
the results of the crude association of potential prognostic
determinants with the 18 months mortality
...

The overall clinical model (that is, the first column of table
2) had an AUC of 0
...
The inclusion of the use of β blockers
in this model improved the AUC to 0
...
84
...
85
...
2–0
...

As the clinical model with information on drug treatment
combines data readily available for practising clinicians we
transformed this model to a scoring rule: age/17 + 4 for male
+ 9 for presence of diabetes + 17 for history of renal dysfunction + 10 for presence of ankle oedema + 7 for systolic blood
pressure < 110 or diastolic blood pressure < 70 − weight/3 +
13 for absence of use of β blockers (table 3)
...
The score was calculated for
each subject by assigning points for each predictor present and
adding these points
...
heartjnl
...
bmj
...
bmj
...

†Combined end point of death or transplantation
...


weighing 70 kg, with a history of renal insufficiency, diabetes,
ankle oedema, and a blood pressure of 130/80 and who does
not use a β blocker receives a score of (60/17 + 9 + 17 + 10 −
70/3 +13) = 29
...
In our data, the score ranged from −32
...
7 (mean −0
...
5) and the AUC of the rule was
0
...
72 to 0
...

Table 4 shows the incidence of mortality among patients
across different score categories
...
For example, of 23 subjects with a
score of > 11, 78% (n = 18) died, while only 12% (n = 3) of
the 25 subjects with a score of < −15 died
...
Reading the table vertically provides
estimates of the sensitivity and specificity at different thresholds
...
Of these, 43 (12 + 13 + 18) indeed died, correctly predicting 84% of all deaths (that is, the sensitivity or true positive rate)
...


DISCUSSION
This study shows that a combination of easily obtainable variables accurately predicts 18 months mortality in patients with
heart failure
...

Several studies have used multivariate logistic regression to
derive predictive models
...
Our study was undertaken in patients
included in a randomised controlled trial
...
heartjnl
...
The exclusion of patients with a short life
expectancy mainly led to the exclusion of those with
malignancies and other comorbidities, but not to the exclusion
of those with moderate or severe heart failure
...
Patients had to give their
informed consent and this is likely to have led to the selection
of a group of relatively motivated patients
...

Other studies have often included variables that are not
widely available in heart failure patients
...
6 7
Our predictive values were comparable with those in a recent
study in which age, the presence of pulmonary crepitations, a
lower systolic blood pressure, and higher creatinine concentrations were most predictive of mortality
...
1
Although results of echocardiography (95%), chest radiography (87%), and electrocardiography (99%) were available
in most of our cohort, more specific information on these data,
such as ejection fraction (54%) and diastolic function (35%),
were only available in a proportion of the participants and
were subject to large intrahospital differences
...

Finally, although other studies have composed logistic
regression models, they did not assess the prognostic
performance of a scoring rule combining the individual
predictors derived from the model
...
16 As we included
more variables in our models, the precision of some of our
estimates may be limited
...
bmj
...
bmj
...
80 (95% confidence interval 0
...
87)
...
Therefore we could not carry out an external
validation study
...
Ideally, such a validation should
take place before the model can be applied in practice
...
In their daily
practice physicians normally try to estimate the prognosis of
their individual patients
...
For example, patients with a very poor prognosis might be excluded from invasive treatments
...
Although quality of life scores are
independent predictors of mortality, their added prognostic
value is too small to warrant quality of life measurements for
such a purpose in routine clinical practice
...


ACKNOWLEDGEMENT
We would like to acknowledge Dr Karel G M Moons from the Julius
Centre for Health Sciences and Primary Care, University Medical
Centre Utrecht, the Netherlands, for his help in developing the statistical model
...


Authors’ affiliations
M L Bouvy, E R Heerdink, H G M Leufkens, Department of
Pharmacoepidemiology and Pharmacotherapy, Utrecht Institute for
Pharmaceutical Sciences (UIPS), Utrecht, Netherlands
A W Hoes, Julius Centre for Health Sciences and Primary Care,
University Medical Centre, Utrecht, Netherlands

REFERENCES
1 Chin MH, Goldman L
...
Am J Cardiol 1997;79:1640–4
...
Survival of patients with a new
diagnosis of heart failure: a population based study
...

3 Khand AU, Gemmell I, Rankin AC, et al
...
Eur Heart J 2001;22:153–64
...
Evidence of improving
prognosis in heart failure: trends in case fatality in 66 547 patients
hospitalized between 1986 and 1995
...

5 Scrutinio D, Lagioia R, Ricci A, et al
...
The role
of the New York Heart Association classification, cardiopulmonary
exercise testing, two-dimensional echocardiography and Holter
monitoring
...

6 Zugck C, Kruger C, Kell R, et al
...

Eur J Heart Fail 2001;3:577–85
...
Development and
prospective validation of a clinical index to predict survival in ambulatory
patients referred for cardiac transplant evaluation
...

8 Jiang W, Alexander J, Christopher E, et al
...
Arch Intern Med 2001;161:1849–56
...
In: Applied logistic regression
...

10 Hanley JA, McNeil BJ
...
Radiology
1982;143:29–36
...
Multivariable prognostic models: issues in
developing models, evaluating assumptions and adequacy, and
measuring and reducing errors
...

12 Greenland S, Finkle WD
...
Am J Epidemiol
1995;142:1255–64
...
Prognostic significance
of atrial fibrillation in advanced heart failure
...

Circulation 1991;84:40–8
...
Prognostic importance of serum sodium
concentration and its modification by converting-enzyme inhibition in
patients with severe chronic heart failure
...

15 Parameshwar J, Keegan J, Sparrow J, et al
...
Am Heart J 1992;123:421–6
...
A simulation study of the number
of events per variable in logistic regression analysis
...


Reference linking to full text
of more than 200 journals
Toll free links
You can access the FULL TEXT of articles cited in

Heart online if the citation is to one of the more than

200 journals hosted by HighWire (http://highwire
...
edu) without a subscription to that journal
...


www
...
com

www
...
com

Downloaded from http://heart
...
com/ on March 30, 2015 - Published by group
...
com

Predicting mortality in patients with heart
failure: a pragmatic approach
M L Bouvy, E R Heerdink, H G M Leufkens and A W Hoes
Heart 2003 89: 605-609

doi: 10
...
89
...
605
Updated information and services can be found at:
http://heart
...
com/content/89/6/605

These include:

References
Email alerting
service

Topic
Collections

This article cites 15 articles, 8 of which you can access for free at:
http://heart
...
com/content/89/6/605#BIBL
Receive free email alerts when new articles cite this article
...


Articles on similar topics can be found in the following collections
Drugs: cardiovascular system (8132)
Epidemiology (3361)
Hypertension (2760)
Diabetes (797)
Metabolic disorders (951)

Notes

To request permissions go to:
http://group
...
com/group/rights-licensing/permissions
To order reprints go to:
http://journals
...
com/cgi/reprintform
To subscribe to BMJ go to:
http://group
...
com/subscribe/



---