The Epidemiology of Nosocomial Blood Stream Infection BSI In the Elderly

1
The Epidemiology of Nosocomial Blood Stream
Infection (BSI) In the Elderly
Contact Keith Kaye, MD MPH Phone
919-668-1720 Address Box 3152 Durham, NC
27710 Email kaye0001_at_mc.duke.edu
Keith S. Kaye, MD, MPH, Deverick J. Anderson, MD,
Yong Choi, RN, Katherine Link, RN, Jane Briggs,
RN, Richard Sloane, MS, John J. Engemann, MD,
Daniel J. Sexton, MD, Kenneth E. Schmader,
MD Duke University Medical Center and the Durham
VA Medical Center, Durham, NC
Results
Abstract (revised)
Purpose The mortality rate due to nosocomial BSI
among all hospitalized patients is approximately
20. Few data exist, however, pertaining to the
epidemiology of and outcomes associated with
nosocomial BSI in elderly hospitalized patient.
The objective of this study was to report the
descriptive epidemiology of nosocomial BSI in the
elderly and to describe the clinical outcomes of
elderly patients with nosocomial BSI. Methods
A cohort study was conducted at Duke University.
Patients gt 64 yrs old with BSI occurring gt48
hours after admission were prospectively
identified between 3/1994-12/2002. Variables
studied included demographics, comorbidities,
functional status and hospital risk factors. The
outcomes studied were in-hospital mortality,
mortality 1 year following BSI and duration of
hospitalization during the 90 day period
following BSI (including readmissions to the
hospital). Results 754 patients were included in
the cohort. The mean age of all patients was
73.2 years 57.4 were male and 75.7 were white.
Most patients were admitted from home (56.2),
36.0 were transferred from another hospital and
3.9 were admitted from an assisted living
facility. Common comorbid conditions included
malignancy (30.5), diabetes (24.3), congestive
heart failure (19.9), chronic lung disease
(16.1) and renal disease (13.6). At the time
of BSI, 86 of patients required assistance
with bathing, dressing, or ambulation 59.5
required assistance with feeding 61.0 were
incontinent of urine or had an indwelling
catheter and 32.2 were incontinent of stool or
had a colostomy. 84.7 of BSI were
catheter-related. The single most common pathogen
was Staphylococcus aureus (SA) (30.6) 67.5 of
these isolates were methicillin-resistant (MR).
21.6 of pathogens were enterobacteriaciae, 12.1
were enterococci and 8.1 were fungus. The
median duration of hospitalization after BSI was
10 days (IQR 5-17). The in-hospital mortality
rate was 38.6 and the 1-year mortality rate was
62.9. Conclusion Elderly patients with
nosocomial BSI frequently have underlying
comorbid conditions and often lack independence
with activities of daily living. MRSA was the
single most common cause of BSI in our
population. Elderly patients with nosocomial BSI
cohort had a high mortality rate in-hospital
mortality approached 40 and 1-year mortality was
greater than 60.
Table 3. Clinical Outcomes of Elderly Patients
with Nosocomial Bloodstream Infection
Table 1. Characteristics of Elderly patients
with Nosocomial Bloodstream Infection (n754)
Conclusions

• The in-hospital mortality rate in this cohort of
  elderly patients with nosocomial bloodstream
  infection was high (approaching 40).
• Duration of hospitalization after bloodstream
  infection was greater than 1 week.
• Central venous catheters were the source of
  bloodstream infection for more than 80 of
  patients.
• Functional impairments were quite common among
  patients in this cohort the majority were
  incontinent of urine and unable to eat, ambulate,
  dress, or bathe independently
• Preventive interventions should focus on optimal
  technique during central venous catether
  insertion and on appropriate catheter utilization
  and care.

Background
Nosocomial bloodstream infection in all ages is
associated with overall mortality rates between
20-40 and an increase in the duration of
hospitalization of approximately one week.
Previously identified risk factors in the general
population for complicated bacteremia and death
include infection due to methicillin-resistant S.
aureus (MRSA), increasing APACHE II scores,
underlying pneumonia, and persistently positive
blood cultures. While the elderly population in
the United States is growing and elderly patients
are frequently hospitalized, scant data exist
regarding the impact of nosocomial BSI on this
population. Understanding the impact of
nosocomial bloodstream infections in the elderly
will help to guide the utilization of resources
to develop effective interventions that might
reduce morbidity and mortality associated with
bloodstream infections in this population. The
objective of this study was to report the
descriptive epidemiology of nosocomial BSI in the
elderly and to describe the clinical outcomes of
elderly patients with nosocomial bloodstream
infection.
References
1. Friedman ND et al. Health care--associated
bloodstream infections in adults a reason to
change the accepted definition of
community-acquired infections. Annals of Internal
Medicine. 137(10)791-7, 2002 Nov 19 2. Cosgrove
SE et al. The impact of methicillin resistance in
Staphylococcus aureus bacteremia on patient
outcomes mortality, length of stay, and hospital
charges. Infection Control Hospital
Epidemiology. 26(2)166-74, 2005 Feb. 2. Gopal AK
et al. Prospective analysis of Staphylococcus
aureus bacteremia in nonneutropenic adults with
malignancy. Journal of Clinical Oncology
18(5)1110-5, 2000 Mar. 3. Conterno LO et al.Risk
factors for mortality in Staphylococcus aureus
bacteremia. Infection Control Hospital
Epidemiology.19(1)32-7, 1998 Jan. 4. Yzerman EP
et al. Delta APACHE II for predicting course and
outcome of nosocomial Staphylococcus aureus
bacteremia and its relation to host defense. J
Infect Dis 1996 Apr173(4)914, 1996 Apr. 5.
Fowler VG, Jr. et al. Clinical identifiers of
complicated Staphylococcus aureus
bacteremia.Archives of Internal Medicine,
22163(17)2066-72, 2003 Sept.
Methods
This cohort study was conducted at Duke
University Medical Center between 3/1994 and
12/2002. Patients gt64 years of age who had
positive blood cultures greater than 48 hours
after admission to the hospital and who did not
have signs or symptoms of infection at the time
of admission, were prospectively identified by
infection control practitioners and entered into
an infection control database. Bloodstream
infections and the source of infection were
defined and characterized using CDC criteria.
Additional data (demographics, co-morbid
conditions, baseline functional status,
hospitalization-related variables) were collected
retrospectively from patient charts and hospital
databases. Baseline functional status was defined
as functional status prior to admission. Patients
with polymicrobial infections were excluded from
the study. The primary outcomes studied were
in-hospital mortality and duration of
hospitalization during the 90 day period
following BSI (including readmissions to the
hospital). These outcomes were acertained by
chart review.
For variables with missing data, percentages
are calculated based upon patients without
missing data