Transcript Facility-Level Interpatient Hemoglobin Variability in

When Using DOPPS Slides
• We welcome the use of DOPPS slides as we
value the distribution of our research for the
benefit of patient care and renal research
• Because the DOPPS data and analyses on the
following slides are published in the public
domain, we ask that you honor our attached
guidelines when using DOPPS slides for your
own research purposes
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
DOPPS Slide Use Guidelines
• Modifying DOPPS data, analyses, tables, and
graphics in any form is not permitted without
prior approval from the DOPPS coordinating
center staff at Arbor Research
• Each DOPPS slide used must include the
citation of the associated publication and feature
the corresponding DOPPS logo
• Contact Arbor Research for permission to
reproduce slides for publication and obtain a
high-quality version suitable for print
[email protected]
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
D
PPS
Dialysis Outcomes and Practice Patterns Study
Modifiable Practices Associated with Sudden
Death among Hemodialysis Patients in the
Dialysis Outcomes and Practice Patterns Study
Michel Jadoul; Jyothi Thumma; Douglas S. Fuller; Francesca Tentori; Yun Li;
Hal Morgenstern; David Mendelssohn; Tadashi Tomo; Jean Ethier;
Friedrich Port; Bruce M. Robinson
Clin J Am Soc Nephrol 2012; (7)5: 765-774
Background
• Sudden death accounts for 20%–25% of all deaths in
hemodialysis (HD) patients
• Although sudden death is also common among patients
on peritoneal dialysis, there are reasons to speculate that
practices related to the HD treatment itself may play a
specific role:
– Well-established risk of arrhythmias associated with
electrolyte imbalances, especially potassium shifts
– Increased risk of sudden death in the 12-hour period that
begins with the start of the HD session or at the end of the
long HD-free interval
– Several patient characteristics—such as heart failure, left
ventricular hypertrophy, and coronary artery disease—have
been associated with a high risk of sudden death in
hemodialysis patient
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Hypothesis
• We hypothesized that the rapid changes in
extracellular fluid and electrolyte concentrations
that occur during shorter dialysis sessions,
using low dialysate potassium (KD) and at a
higher ultrafiltration rate, may trigger
arrhythmias and other cardiac events that may
lead to higher incidence of sudden death.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
DOPPS Background (1)
• International, prospective cohort study of hemodialysis patients
and HD unit practices
• Uniform international data collection
• Goal: Identify HD practice patterns associated with improved
patient outcomes (adjusted for patient mix)
• Major outcomes: mortality, hospitalization, vascular access,
quality of life
• DOPPS is supported by scientific research grants from Amgen
(since 1996), Kyowa Hakko Kirin (since 1999, in Japan),
Sanofi/Genzyme (since 2009), and Abbott Laboratories (since
2009) without restrictions on publications
• Coordinated by Arbor Research Collaborative for Health (Ann
Arbor, MI USA)
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
DOPPS Background (2)
Years
Countries Facilities
Patients
DOPPS 1
1996-2001
7
308
17,034
DOPPS 2
2002-2004
12
322
12,839
DOPPS 3
2005-2008
12
300
11,170
DOPPS 1: France, Germany, Italy, Japan, Spain, UK, US
DOPPS 2, 3: DOPPS 1 Countries + Australia/New Zealand, Belgium, Canada, Sweden
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Data Sources and Variables (1)
• 37,765 participants on 3x-weekly HD from 930
dialysis facilities during the first three phases of
DOPPS.
• Baseline patient characteristics including
demographics, detailed comorbidities,
laboratory values, and dialysis treatment
characteristics are obtained at the time of study
entry by medical record abstraction.
• Clinical variables and medical events, including
mortality and cause of death, are updated every
four months.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Data Sources and Variables (2)
• Primary exposures
– Short duration of dialysis sessions (prescribed treatment time
<210 minutes)
– Low dialysis dose (single pool Kt/V <1.2)
– Large ultrafiltration (UF) volume (>5.7% of the post-dialysis weight)
– Dialysate potassium <3 mEq/L
– Prescription of beta blockers and QTI prolonging drugs
• Outcomes
– All-cause mortality
– Sudden death, defined as primary cause of death reported as
death due to cardiac arrhythmia, cardiac arrest, and/or
hyperkalemia
– Other cardiovascular death, if due to cardiovascular disease
– Non-cardiovascular deaths, not classified as “sudden death” or
other cardiovascular death
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Statistical Methods (1)
• Cox regression models were used to assess the
associations of the four dialysis practices and QTprolonging medications with all-cause mortality,
sudden death, other CV death, and non-CV death.
– Time at risk was defined as the period from study
enrollment until the earliest of death, transplant, change
of treatment modality, withdrawal from dialysis, transfer
to another facility, or study end.
– Models were adjusted for age, sex, black race, body
mass index, years with ESRD, 14 comorbidity classes,
catheter use, serum albumin, phosphorus, calcium, PTH,
hemoglobin, serum creatinine, ferritin, white blood cell
count, single pool Kt/V, and indicators of facility
achievement of clinical guidelines.
– Models were stratified by DOPPS phase and country;
robust variance (sandwich) estimates were used to
account for facility-level clustering.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Statistical Methods (2)
• To partially reduce the impact of unmeasured patient-level
confounders, we also applied a two-stage instrumental variable
approach, using the dialysis facilities as the instruments.
– In the first stage, patient-level predicted treatment was determined
using a linear regression model.
– In the second stage, Cox regression was used to estimate the
hazard of mortality associated with the patient-level predicted
treatment.
• We also used limited information maximum likelihood
regression (LIML)
– We used one-year mortality as the second stage outcome and
modeled it as a dichotomous variable using linear regression as an
approximation to logistic regression.
• IVEware was used to perform multiple imputations of missing
data, including cause of death.
– Estimates and their variances from the multiple imputation results
were combined according to the Rubin method.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Patient Characteristics
by Treatment Practice (1)
Mean (SD) or %
N patients
Age
Male
Black
Vintage
Body mass index
Catheter
Comorbid Conditions:
CAD
Other cardiovascular
Cerebrovascular disease
CHF
Diabetes
Labs:
Hgb (g/dl)
Albumin (g/dl)
Potassium (mEq/l)
Phosphorous (mg/dl)
PTH (pg/ml)†
Creatinine (mg/dl)
Ferritin (ng/ml)†
†Serum PTH
Tx time < 210 min
No
Yes
28,878
8,536
61.0 (14.6)
64.6 (15.3)a
60
49a
13
15a
4.2 (5.5)
2.0 (3.8)a
24.9 (5.8)
24.1 (5.2)a
23
35a
Kt/V < 1.2
No
18,824
62.0 (14.8)
53
14
4.9 (5.7)
24.2 (5.6)
18
Yes
7,273
61.2 (14.5)c
72a
13a
2.6 (4.3)a
25.4 (5.9)a
28a
45
35
16
35
38
46
32b
18a
37
40
46
36
17
35
35
42
31a
16
33
44a
10.9 (1.7)
3.7 (0.5)
5.0 (0.8)
5.6 (1.8)
177 (275)
9.2 (3.2)
285 (431)
10.5 (1.7)a
3.6 (0.6)a
4.8 (0.8)a
5.6 (1.9)a
187 (282)
8.0 (3.1)a
247 (403)b
11.1 (1.6)
3.7 (0.5)
5.0 (0.8)
5.5 (1.7)
172 (272)
9.2 (3.0)
314 (456)
10.5 (1.8)a
3.7 (0.6)a
4.9 (0.8)a
5.7 (1.9)a
184 (269)b
9.2 (3.5)a
203 (332)a
and ferritin values summarized as median
(interquartile range). ap<0.001, b0.001≤p<0.05, and c0.05≤p<0.10
for 'yes' vs. 'no' column.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Patient Characteristics
by Treatment Practice (2)
Mean (SD) or %
N patients
Age
Male
Black
Vintage
Body mass index
Catheter
Comorbid Conditions:
CAD
Other cardiovascular
Cerebrovascular disease
CHF
Diabetes
Labs:
Hgb (g/dl)
Albumin (g/dl)
Potassium (mEq/l)
Phosphorous (mg/dl)
PTH (pg/ml)†
Creatinine (mg/dl)
Ferritin (ng/ml)†
†Serum PTH
UF volume > 5.7%
No
Yes
26,880
3,970
62.5 (14.6)
57.2 (14.9)a
58
55a
13
13a
3.7 (5.2)
6.3 (6.2)a
25.0 (5.7)
21.5 (4.0)a
23
11a
Dialysate K < 3 mEq/l
No
Yes
6,606
29,629
64.4 (14.5)
61.3 (14.8)a
54
59a
13
14a
1.8 (3.5)
4.1 (5.5)a
25.8 (5.9)
24.5 (5.6)b
41
22a
46
34
17
34
38
40
35a
14a
33a
35b
54
39
20
42
46
43a
33a
16a
34a
37a
10.9 (1.7)
3.7 (0.5)
4.9 (0.8)
5.5 (1.8)
178 (271)
8.9 (3.1)
286 (429)
10.6 (1.6)
3.8 (0.5)a
5.2 (0.8)a
5.9 (1.9)a
175 (297)a
10.1 (3.2)a
222 (439)c
10.8 (1.7)
3.5 (0.6)
4.7 (0.8)
5.4 (1.9)
190 (269)
7.3 (3.0)
278 (410)
10.8 (1.7)a
3.7 (0.5)a
5.0 (0.8)a
5.6 (1.8)a
178 (280)
9.2 (3.2)a
273 (426)a
and ferritin values summarized as median
(interquartile range). ap<0.001, b0.001≤p<0.05, and c0.05≤p<0.10
for 'yes' vs. 'no' column.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Association of Dialysis Treatment
Practices with Mortality
Patient level
HR
95% CI
IV approach
p
HR
95% CI
p
Treatment time <210 min (vs. ≥210 min):
All-cause mortality
1.06 (1.00, 1.13) 0.04 1.09 (0.94, 1.27) 0.26
Sudden death
1.13 (1.00, 1.27) 0.04 1.38 (1.04, 1.83) 0.03
Other cardiovascular death
1.00 (0.89, 1.12) 0.96 0.77 (0.58, 1.03) 0.08
Non-cardiovascular death
1.06 (0.97, 1.17) 0.20 1.16 (0.93, 1.44) 0.18
Kt/V <1.2 (vs. ≥1.2):
All-cause mortality
1.06 (1.00, 1.12) 0.07 1.39 (1.07, 1.81) 0.01
Sudden death
1.10 (0.97, 1.24) 0.15 1.71 (1.02, 2.87) 0.04
Other cardiovascular death
0.96 (0.85, 1.08) 0.48 0.95 (0.60, 1.50) 0.83
Non-cardiovascular death
1.10 (1.00, 1.21) 0.04 1.59 (1.11, 2.29) 0.01
UF volume >5.7% of post-weight (vs. ≤5.7%):
All-cause mortality
1.09 (1.01, 1.19) 0.03 1.32 (0.84, 2.08) 0.23
Sudden death
1.15 (1.00, 1.32) 0.04 1.56 (0.60, 4.03) 0.35
Other cardiovascular death
1.17 (1.01, 1.36) 0.04 2.36 (1.02, 5.46) 0.05
Non-cardiovascular death
1.00 (0.88, 1.14) 0.98 0.78 (0.43, 1.43) 0.42
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Association of Dialysate K*
with Mortality (1)
Patient-level
Dialysate K ≤1.5 (vs. ≥3)
Dialysate K 2-2.5 (vs. ≥3)
HR
95% CI
p
HR
95% CI
p
All-cause mortality
1.13
(1.03, 1.25)
0.01
1.08
(1.01, 1.16)
0.03
Sudden death
1.39
(1.12, 1.74)
0.004
1.17
(1.01, 1.37)
0.04
Other cardiovascular death
1.14
(0.95, 1.36)
0.16
1.04
(0.91, 1.19)
0.54
Non-cardiovascular death
0.99
(0.84, 1.17)
0.93
1.05
(0.94, 1.16)
0.38
All patients (N=37,741)
Among patients with serum K ≥5 (N=17,327)
All-cause mortality
1.09
(0.95, 1.26)
0.23
1.08
(0.97, 1.20)
0.17
Sudden death
1.21
(0.91, 1.61)
0.18
1.11
(0.90, 1.38)
0.33
Other cardiovascular death
1.16
(0.88, 1.52)
0.29
1.00
(0.82, 1.21)
0.97
Non-cardiovascular death
0.97
(0.77, 1.22)
0.81
1.10
(0.93, 1.31)
0.27
Among patients with serum K <5 (N=20,414)
All-cause mortality
1.15
(1.00, 1.33)
0.04
1.06
(0.98, 1.15)
0.15
Sudden death
1.53
(1.10, 2.13)
0.01
1.18
(0.98, 1.42)
0.08
Other cardiovascular death
1.05
(0.85, 1.31)
0.64
1.05
(0.88, 1.24)
0.58
Non-cardiovascular death
1.03
(0.77, 1.38)
0.83
1.00
(0.88, 1.15)
0.95
*N
KD ≤ 1.5=5,493;
NKD = 2-2.5=25,552; NKD ≥ 3=6,696
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Association of Dialysate K*
with Mortality (2)
IV approach
Dialysate K ≤1.5 (vs. ≥3)
Dialysate K 2-2.5 (vs. ≥3)
HR
95% CI
p
HR
95% CI
p
All-cause mortality
1.09
(0.88, 1.35)
0.43
1.23
(1.04, 1.45)
0.01
Sudden death
1.67
(0.99, 2.81)
0.05
1.61
(1.12, 2.30)
0.01
Other cardiovascular death
1.11
(0.75, 1.64)
0.62
1.23
(0.90, 1.68)
0.19
Non-cardiovascular death
0.84
(0.61, 1.16)
0.29
1.03
(0.83, 1.29)
0.76
All patients (N=37,741)
Among patients with serum K ≥5 (N=17,327)
All-cause mortality
1.13
(0.87, 1.47)
0.37
1.23
(0.99, 1.52)
0.06
Sudden death
1.27
(0.73, 2.22)
0.40
1.30
(0.81, 2.08)
0.28
Other cardiovascular death
1.18
(0.74, 1.88)
0.49
1.17
(0.79, 1.72)
0.44
Non-cardiovascular death
1.02
(0.70, 1.47)
0.93
1.21
(0.87, 1.69)
0.25
Among patients with serum K <5 (N=20,414)
All-cause mortality
1.04
(0.80, 1.36)
0.76
1.23
(1.03, 1.46)
0.02
Sudden death
2.01
(0.96, 4.24)
0.06
1.86
(1.31, 2.63)
<0.001
Other cardiovascular death
0.94
(0.56, 1.56)
0.80
1.23
(0.86, 1.76)
0.26
Non-cardiovascular death
0.77
(0.51, 1.15)
0.20
0.95
(0.75, 1.22)
0.71
*N
KD ≤ 1.5=5,493;
NKD = 2-2.5=25,552; NKD ≥ 3=6,696
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Observed Cause of Death by Country –
As Reported
100
14 18
14
21
21 19 19
30
5
6
7 4
9 4
12 11
11
42
13
37
43
27
42
42 38
32 37 39
30
28
21
21
15
17 22
23 19 20 37
18
19
24 19 21
17 15 14
10 12 12 11
19
19 21
% of patients
80
60
40
20
0
4
4
Missing
8
Unknown
32
Non-CV death
Other CV death
22
Sudden Death
5
US IT JP FR BE ANZ CA GE UK SP SW
N deaths: 4687 445 596 431 419 286 375 522 381 534 370
All
9046
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Cause of Death, by Country – Excluding
Missing/Unknown Causes of Death
100
80
37
% of patients
55
46
47
53 57 51 52 48 55 56
44
Non-CV death
60
29
Other CV death
40
20
20
0
33
31
29
26
34
23 29 31
28 28 46
25 23 22 20
19 18 17 17 16
26
Sudden Death
7
US IT JP FR BE ANZ CA GE UK SP SW
N deaths: 3364 334 544 349 311 214 217 350 260 373 294
All
6610
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Causes of Death, by Country – Including
Imputed Causes
100
80
38
46
% of patients
57
45
46
52 54
54 56 50 50
44
Non-CV death
60
30
40
31
19
20
32
34
27 27 44
24 23 30 29
23 23 22 21 20 20 20 20 19
29
Other CV death
27
Sudden Death
12
0
US IT JP FR BE ANZ CA GE UK SP SW
N deaths: 4687 445 596 431 419 286 375 522 381 534 370
All
9046
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Facility % of Patients with
Short Treatment Time (<210 Minutes)
Facility % of Patients
100%
80%
Facility
Percentile
60%
95th
40%
75th
20%
50th
25th
5th
0%
ANZ BE
19
Facility N: 20
CA
19
FR
17
GE
20
IT
19
JP
62
SP
22
SW
19
UK
16
US
65
All
298
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Facility % of Patients with
Low Kt/V (<1.2)
Facility % of Patients
100%
80%
60%
Facility
Percentile
40%
95th
20%
75th
50th
25th
5th
0%
ANZ BE
19
Facility N: 17
CA
17
FR
17
GE
15
IT
17
JP
59
SP
21
SW
19
UK
15
US
61
All
277
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Facility % of Patients with
Large UF Volume (>5.7% of post-weight)
Facility % of Patients
100%
80%
60%
40%
Facility
Percentile
95th
20%
75th
50th
25th
5th
0%
ANZ BE
19
Facility N: 20
CA
19
FR
17
GE
18
IT
19
JP
62
SP
22
SW
19
UK
15
US
64
All
294
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Facility % of Patients with
Low Dialysate K (<3 mEq/L)
Facility
Percentile
Facility % of Patients
100%
95th
75th
80%
50th
25th
60%
40%
5th
20%
0%
ANZ BE
19
Facility N: 20
CA
19
FR
17
GE
20
IT
19
JP
62
SP
22
SW
19
UK
16
US
65
All
298
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Association of Treatment Practices
with Sudden Death
Patient model
Treatment time
(per 30 min lower)
IV approach*
All-cause mortality
Sudden death
Sp Kt/V
(per 0.1 unit lower)
UF volume
(per 1% higher)
Dialysate K
(per 1 mEq/l lower)
0.85
1.00
1.15
1.30
0.85 1.00 1.15 1.30 1.45 1.60 1.75
Hazard Ratio (95% CI)
*used predicted treatment from 1st stage
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Medications
• Higher risk of sudden death was evident for
amiodarone prescription (HR=1.44, [1.16-1.81],
p=0.001), but not for prescription of other QTI
prolonging drugs (HR=1.10, [0.94-1.28], p=0.22).
– The association of a higher risk of SD with
amiodarone likely reflects a prescription by
indication bias.
• Beta-blockers were associated with a lower risk
of sudden death (HR=0.88, [0.78-0.99], p=0.03).
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Sensitivity Analyses
• Our findings were confirmed in a series of
sensitivity analyses.
– In models restricted to prevalent patients (time on
dialysis > 6 months) and to “healthier” patients
(non-diabetic participants aged < 65).
– When observed causes of death were used in place
of imputed causes of death.
– When all four practices were included in the same
Cox model, indicating that the association of each
practice with sudden death was independent of the
other three.
– When deaths attributed to hyperkalemia were
excluded.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Strengths
• The strengths of the present study are multiple:
–
–
–
–
Large sample size
Nationally representative patient samples
Prospective data collection
Extensive adjustment for patient characteristics,
comorbidities, and lab results
– The detailed medication database available in the
DOPPS made it possible for the first time to study the
impact of drugs that prolong the QTI on the risk of SD
– Sensitivity analyses show consistent associations
between the modifiable practices and SD
– Facility-level IV analyses tend to lessen the risk of
patient-level confounding by indication
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Limitations
• Several limitations have to be acknowledged:
– As an observational study, we are not able to claim
causality.
– The number of patients who had received an implantable
cardiac defibrillator was too small for studying the
impact of their use on the risk of SD.
– As in any large clinical database, some errors may be
present in coding but these should have favored the null
hypothesis.
– We did not have information on the time interval between
the end of the last HD session and SD.
– Finally, we did not have ECG data, regarding the
presence of left ventricular hypertrophy, or prolonged
QTI, nor potentially relevant biomarker data.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Conclusions (1)
• Our DOPPS results demonstrate for the first time
substantial differences between countries in the
proportion of deaths that are reported as sudden.
– 33.4% in the US, 17.5% in Canada, 17.8% in Europe,
19.2% in ANZ, and 23.2% in Japan
• The present paper is the first, to our knowledge, to
assess in HD patients the association between SD
and the prescription of drugs prolonging the QTI on
the ECG.
– Beta-blockers were associated with a lower risk of SD, in
line with their capacity to counteract the increased
sympathetic activity common in HD patients
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Conclusions (2)
• We have identified several readily modifiable facility
practices associated with an independent risk of SD,
both in standard patient-level analyses and in IV
analyses based on facility variation in practice:
– Specifically, short treatment time, low dialysis dose, high UF
volume, and low dialysate K are associated with sudden death,
as well as overall mortality.
– Since low (<3) dialysate K is commonly used and it is the easiest
practice to modify, our study suggests that more careful
adaptation of dialysate K level may help minimize the impact of
sudden death observed in HD patients.
– Clinical trials are needed to test this possibility, and may be
feasible for practices such as dialysate potassium where
substantial variation in care exists and clinical uncertainty
remains significant.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774
Acknowledgements
• Our thanks to DOPPS study coordinators, medical directors, and
participating patients for their dedicated work.
• The DOPPS would not be possible without the generous financial
support of the following companies who have demonstrated their
strong commitment to independent scientific research to improve
patient care:
–
–
–
–
–
–
Amgen (since 1996)
Kyowa Hakko Kirin (since 1999, in Japan)
Abbott Laboratories (since 2009)
Sanofi/Genzyme (since 2009)
Baxter Healthcare (since 2011)
Vifor Fresenius Medical Care Renal Pharma Ltd (since
2011)
• Support from DOPPS sponsors is provided without restrictions on
publications.
Jadoul, et al. Clin J Am Soc Nephrol 2012; 7(5): 765-774