Transcript Infections in the dialysis population: A major communicable disease

Infections in the dialysis
population: A major
communicable disease
issue!
Allan J. Collins, MD, FACP
Professor of Medicine
University of Minnesota
Director, USRDS Coordinating Center
Disclosures
• Institutional research funding: NIH, CDC, National
Kidney Foundation, Amgen, Baxter, Sigma Tau,
NxStage, Genzyme, BMS, AMAG Pharmaceuticals
• Advisory boards in last year: WHO Advisory Panel
on NCDs, Kidney Disease Improving Global
Outcomes (KDIGO), International Federation of
Kidney Foundations (IFKF),
• Epidemiology consulting: NxStage, Amgen,
Affymax
USRDS 2008 ADR
Infectious complication in dialysis patients:
an old yet new communicable disease
challenge
• Trends in overall mortality show reductions in death
rates in years 2-5 after onset of ESRD
• Death rates in the first year have not changed in 12
years raising concerns about the underlying
morbidity contributing to the early death rates
• Cardiovascular and Infectious hospitalizations are the
leading cause of morbidity yet little is known about
the trends in the first months of dialysis
• Here we review the trends in hospitalization rate with
particular emphasis on infections as a contributor to
morbidity and mortality in the first year of treatment
USRDS 2008 ADR
Mortality rates,
by modality
Figure 6.1 (Volume 2)
Incident ESRD
patients; adjusted
for age, gender,
race, & primary
diagnosis. Incident
ESRD patients,
2005, used as
reference cohort.
USRDS 2008 ADR
Change in all-cause & cause-specific
hospitalization rates, by modality
Figure p.16 (Volume 2)
Period prevalent ESRD
patients; adjusted for
age, gender, race, &
primary diagnosis.
ESRD patients, 2005,
used as reference
cohort. Vascular access
hospitalizations are
“pure” inpatient
vascular access events.
New vascular access
codes for peritoneal
dialysis patients
appeared in late 1998;
therefore, peritoneal
dialysis vascular access
values are shown as
changing since 1999
rather than 1993.
USRDS 2008 ADR
Adjusted admissions for principal
diagnoses, by modality
Figure 6.5 (Volume 2)
Period prevalent ESRD patients; adjusted for age, gender, race, & primary diagnosis. ESRD patients, 2005, used as
reference cohort.
USRDS 2008 ADR
Trends in hospitalization surveillance
data in the dialysis population
• Overall, hospitalization rates have changed little over the
last 10-12 years
• However, there have been significant changes in the types
of hospitalizations with vascular access events
transitioning to the outpatient setting with increases in
cardiovascular disease and infectious hospitalizations
• Hospitalization secondary to pneumonia have increased
but have begun to plateau since 2003
• Vascular access infectious event have more than doubled
in the last 10 years an area of major concern!


By most standards this would be considered a major
communicable disease problem yet the CDC stopped its
survey in 2004!!!!
Who and what is responsible for this trend?
USRDS 2008 ADR
Infectious complications in dialysis
patients
• Infectious hospitalizations have increased in part
related to the increases in pneumonia but this
trend appears to be reversing with increased
evidence of pneumococcal pneumonia
vaccinations
• Preventive care in the form of influenza
vaccinations needs attention and is clearly within
the domain of the providers!
• Yet other sources of infectious events likely
contribute to the increased hospitalization rates
USRDS 2008 ADR
Vascular access use at initiation, 1st
outpatient access, 2006 (From Medical Evidence
Form 2728)
Figure 3.1 (Volume 2)
82%
Catheters
incident hemodialysis
patients, 2006.
USRDS 2008 ADR
First access at initiation,
by nephrologist care, 2006
Figure 3.4 (Volume 2)
incident hemodialysis
patients, 2006, with
new (revised edition)
Medical Evidence
forms.
USRDS 2008 ADR
African American
46.0
33.8
20.2
Native American
39.0
35.8
25.2
Asian
41.9
37.5
20.6
Hispanic
48.9
33.3
17.8
Access at initiation
Catheter
50.3
32.4
17.3
10.0
44.2
45.8
Table 3.a (VolumeFistula
2)
Graft
21.6
43.3
35.1
Maturing Fistula
31.1
42.5
26.4
None
0-12
>12
ESA use
5.7
52.1
42.2
All
41.6
35.0
23.4
Dietary care
1.1
56.0
42.9
Mean
62.3
63.0
63.3
eGFR age
Female
41.5
35.4
23.2
<5
56.4
26.6
17.1
40.4
35.1
24.5
Race 5-<10
White
39.8
35.4
24.9
10-<15
37.0
37.5
25.5
African
46.0
33.8
20.2
>=15 American
44.4
34.8
20.8
Native
American
39.0
35.8
25.2
Primary
diagnosis
Asian
41.9
37.5
20.6
Diabetes
36.9
38.2
24.8
Hispanic
48.9
33.3
17.8
*Diabetes as a comorbidity
38.7
37.0
24.3
Access
at initiation
Hypertension
44.7
34.0
21.3
Glomerulonephritis
32.5
35.0
32.5
Catheter
50.3
32.4
17.3
Fistula
10.0
44.2
45.8
Cystic kidney
18.4
36.7
44.9
Graft
21.6
43.3
35.1
incident ESRD
patients, 2006, with new 31.1
(revised edition)
Maturing
Fistula
42.5Medical Evidence
26.4 forms.
ESA use
5.7
52.1
42.2
USRDS 2008 ADR
Dietary care
1.1
56.0
42.9
eGFR
Pre-ESRD nephrologist care, 2006
Cumulative probability
of catheter placement
Figure 1.8 (Volume 2)
Medicare: hemodialysis
patients who initiate
dialysis at age 67 or older
during the year specified.
Includes those with
Medicare as primary payor
during the two years prior
to initiation & through the
first six months of ESRD;
pre-ESRD claims used for
months prior to initiation
date. Medstat (EGHP):
patients with first date of
regular & continuous
dialysis in 2000 or 2005,
regardless of age. Only one
year of claims prior to the
start of dialysis was
available for the 2000
cohort.
USRDS 2008 ADR
Cumulative probability of multiple catheter
placements determined from service claims
Figure 1.9 (Volume 2)
Medicare: hemodialysis
patients who initiate
dialysis at age 67 or older
during the year specified.
Includes those with
Medicare as primary payor
during the two years prior
to initiation & through the
first six months of ESRD;
pre-ESRD claims used for
months prior to initiation
date. Medstat (EGHP):
patients with first date of
regular & continuous
dialysis in 2000 or 2005,
regardless of age. Only one
year of claims prior to the
start of dialysis was
available for the 2000
cohort.
USRDS 2008 ADR
Per person per year vascular access
costs, by physician specialty
Figure 11.24 (Volume 2)
period prevalent
hemodialysis patients.
Costs determined as the
payment amount for
each line-level access
procedure billed
through physician/
supplier claims.
USRDS 2008 ADR
Catheter events
& complications
Figure 5.27 (Volume 2)
Prevalent hemodialysis
patients age 20 & older,
ESRD CPM data; only
includes patients who
are also in the USRDS
database. Year
represents the
prevalent year & the
year the CPM data
were collected. Access
is that listed as
“current” on the CPM
data collection form.
USRDS 2008 ADR
Arteriovenous fistula events
& complications
Figure 5.28 (Volume 2)
Prevalent hemodialysis
patients age 20 & older,
ESRD CPM data; only
includes patients who
are also in the USRDS
database. Year
represents the
prevalent year & the
year the CPM data
were collected. Access
is that listed as
“current” on the CPM
data collection form.
USRDS 2008 ADR
Arteriovenous graft events
& complications
Figure 5.29 (Volume 2)
Prevalent hemodialysis
patients age 20 & older,
ESRD CPM data; only
includes patients who
are also in the USRDS
database. Year
represents the
prevalent year & the
year the CPM data
were collected. Access
is that listed as
“current” on the CPM
data collection form.
USRDS 2008 ADR
Access procedures in prevalent
hemodialysis patients, by diabetic status
Figure hp.13 (Volume 2)
Change to cuffed catheters
illi
illi
lla
lla
Period prevalent hemodialysis patients with or without simple fistulas. Data from physician/supplier claims. Some
patients may have more than one access at a given point in time.
USRDS 2008 ADR
Vascular access utilization and
infectious complications
• Arteriovenous fistula utilization rates have
increased during the same time as the Fistula
First initiative has been stressed
• The Fistula First effort is also associated with
reductions in the use of grafts and changes in
catheter utilization rates
• Infectious complication are highest with catheters
and lowest with fistulas but there are similar rates
with grafts.
• What are the trends in infectious hospitalizations
that may relate to vascular access complications
USRDS 2008 ADR
Change in all-cause & cause-specific
hospitalization rates, by modality: prevalent
Figure p.22
Period prevalent ESRD
patients; adjusted for
age, gender, race, &
primary diagnosis.
ESRD patients, 2005,
used as reference
cohort. Vascular access
hospitalizations are
“pure” inpatient
vascular access events,
as described in
Appendix A. New
vascular access codes
for peritoneal dialysis
patients appeared in
late 1998; therefore,
peritoneal dialysis
vascular access values
are shown as changing
since 1999 rather than
1993.
USRDS 2008 ADR
Trends in cardiovascular Hospitalizations
in the First Year by months
Adjusted for age, gender, race and cause of ESRD
700 Increased cardiovascular hospitalizations
680
Rate per 1,000 Pt Yrs
660
640
1993
1997
2001
2005
620
600
580
560
540
520
500
All 9<12
Incident Cohort Year
All 6<9
All 3<6
All 2<3
All 1<2
All 0<1
USRDS 2008 ADR
Trends in Infectious Hospitalization rates in
the 1st year by month
Rate per 1,000 Pt Yrs
Adjusted for age, gender, race and cause of ESRD
700
650
600
550
500
450
400
350
300
Increased infectious hospitalizations
1993
1997
2001
2005
All 9<12
All 6<9
All 3<6
All 2<3
All 1<2
All 0<1
Incident Cohort Year
USRDS 2008 ADR
Trends in cardiovascular Hospitalizations
in the First Year by months
Adjusted for age, gender, race and cause of ESRD
Cardiovascular
Infections
Rate per 1,000 Pt Yrs
800
750
700
650
All
All
All
All
All
All
600
550
500
450
400
350
300
All 0
All 1
All 2
All 3
All 6
All 9
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
Incident Cohort Year
1994
1993
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
USRDS 2008 ADR
0<1
1<2
2<3
3<6
6<9
9<12
800
750
700
650
600
550
500
450
400
350
300
Trends in CVD and Infectious Hospitalization
rates in the first month
Rate per 1,000 Pt Yrs
Adjusted for age, gender, race and cause of ESRD
Infectious hospitalizations now
approach CVD for the 1st time!
750
700
650
600
550
500
450
400
350
300
All CV 0<1
All Infect 0<1
All CV 1<2
All Infect 1<2
2005
2004
Incident Cohort Year
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
USRDS 2008 ADR
Percent change in hospital
admissions from Day 90: 1993 to 2005
percent change in admission rates from
1993 to 2005
Incident dialysis patients age 20 and older
110
100
90
all-cause
cardiovascular
80
70
60
50
40
30
20
10
0
-10
infection
0-<1
1-<2
2-<3
3-<6
6-<9
9-<12
months after day 90 of dialysis
USRDS 2008 ADR
*Model based adjustment for age, sex, race, cause of ESRD:
Interval Poisson regression (ASN 2008 poster)
Percent change in hospital
admissions from day 1: 1993 to 2005
percent change in admission rates from
1993 to 2005
Incident hemodialysis patients age 65 and older
all-cause
230
210
190
cardiovascular
infection
vascular access infection
170
150
130
110
90
70
50
30
10
-10
0-<1
1-<2
2-<3
3-<6
6-<9
9-<12
months after dialysis initiation
USRDS 2008 ADR
*Model based adjustment for age, sex, race, cause of ESRD:
Interval Poisson regression (ASN 2008 poster)
Adjusted admissions for principal
diagnoses, by modality
Figure 6.5 (Volume 2)
*early data noted decline
but must be confirmed
Period prevalent ESRD patients; adjusted for age, gender, race, & primary diagnosis. ESRD patients, 2005, used as
reference cohort.
USRDS 2008 ADR
Adjusted relative risk of
mortality after pneumonia: dialysis
Figure 6.29
Incident dialysis patients
with 90-day rule,
Medicare as primary
payor, & Part A & B
claims, 1992–2000
combined; adjusted for
age, gender, race, &
primary diagnosis.
Reference cohort:
patients without
corresponding diseases in
the first year after ESRD
initiation + 90 days.
2004 ADR
Foley RN, Guo H, Snyder JJ, Gilbertson DT, Collins AJ:
Septicemia in the United States Dialysis Population, 1991 to 1999.
Journal of the American Society of Nephrology 15 (4): 1038-1045, 2004
Adjusted relative risk of
mortality after VA event: dialysis
Figure 6.30
Incident dialysis patients
with 90-day rule,
Medicare as primary
payor, & Part A & B
claims, 1992–2000
combined; adjusted for
age, gender, race, &
primary diagnosis.
Reference cohort:
patients without
corresponding diseases in
the first year after ESRD
initiation + 90 days.
2004 ADR
Foley RN, Guo H, Snyder JJ, Gilbertson DT, Collins AJ:
Septicemia in the United States Dialysis Population, 1991 to 1999.
Journal of the American Society of Nephrology 15 (4): 1038-1045, 2004
Trends in Infectious hospitalizations
• Infectious hospitalizations have substantially increased over
the last 10-13 years
• The increase in infectious hospitalizations is more than 4
times greater compared to cardiovascular events in the first
months
• The increase in infectious hospitalization rates is
approaching that of cardiovascular events which is a new
finding and they appear to be interrelated
• The risk of death after an infectious hospitalization event is
significant and is associated with CVD events (The
infection/inflammation multiplier hypothesis)!
USRDS 2008 ADR
Conclusions
•
•
•
•
•
•
•
Death rates in the first months of dialysis are high and need to be
addressed
Infectious hospitalizations in the 1st months of dialysis have
increased nearly 2 fold
Vascular access infectious hospitalizations are up nearly 3 fold
The consequences of late referral and delayed vascular access
planning result in the use of catheters with the accompanying
complications and risk of infection and premature death!
The high rates of catheter utilization and associated infectious
complications suggest a major effort is needed to reduce catheter
utilization (Catheter Last!).
CMS under the QIO scope of work should act to increase early
referral and a “Planned Transition” to dialysis consistent with the
MIPPA CKD Stage 4 education benefit!
CMS should implement a similar effort for planned transition as
was done for Fistula First by engaging health plans, physicians,
hospitals and public health infection control officials to reduce the
use of catheters!
USRDS 2008 ADR
Acknowledgements
• The USRDS Coordinating Center produced the material in
this talk based on contributions from the entire staff of







SAS programmers
Biostatistians
Epidemiologist
Physcians
Pharm Ds
Co-investigators
Graphics designers and editors
• We would like to thank CMS and the NIH project officers for
their support and oversight of the USRDS and the entire
kidney disease community for the hard work they do and
providing the data!
USRDS 2008 ADR