No s o c o mi a l i n f e c t i o n s in an o n c o l o g y

i n t e n s i v e c a r e uni t
Eduardo Velasco, MD
Luiz Cl audi o Santos Thuler, MD, MSc
Carlos Alberto de S. Martins, MD
Leda Maria de Castro Dias, RN
Vania Maria da S. e C. Gon~alves, RN
Rio de Janeiro, Brazil
Introduction: Treatment of cancer has contributed to a growing number of
immunocompromised patients with life-threatening nosocomial infections (NI). High
mortality with considerable cost is observed when they are admitted to the intensive care
unit (ICU). Few studies on infection control and surveillance have been undertaken in this
population group.
Methods: All patients treated at a six-bed medical-surgical oncology ICU for >48 hours were
prospectively observed for the development of an NI and the influence of device utilization
on infection rates. The analysis used the standard definitions of the National Nosocomial
Infection Surveillance System Intensive Care Unit surveillance component.
Results: From September 1993 through November 1995, 370 infections occurred in 623
patients during 4034 patient-days, for an overall rate of 50.0 per 100 patients or 91.7 per
1000 patient-days. Pneumonia (28.9%), urinary tract infections (25.6%), and bloodstream
infections (24.1%) were the main types of infection. The most common microorganisms
isolated were Enterobacteriaceae (29.7%), fungi (22.2%), and Pseudomonas aeruginosa
(13.2%). The median device utilization ratios were 0.63, 0.83, and 0.86 for ventilator,
indwelling urinary catheter, and central venous catheter, respectively. The highest median
device-specific associated infection rate was 41.7 for ventilator. The median for the average
length of stay was 8.8 days, and the average severity of illness score was 4.0. There was a
strong positive correlation between the overall NI patient rate and device utilization (r =
0.56, p < 0.01), average severity of illness score (r = 0.54, p < 0.01), and average length of
stay (r = 0.67, p < 0.0l). No correlations were statistically significant when patient-days
were used in the denominator. Among the devices only the number of central venous
catheter days was significantly correlated with infections (r = 0.51, p = 0.01). The NI
patient-day rates were progressively higher the longer the patients stayed in the ICU.
Conclusions: The high rates reported in this study may reflect a combination of several
factors related to the underlying illness, neutrophil count, and exposure to invasive
procedures. The adjusted infection rates described here provide specific surveillance data
for further interhospital comparisons and also to assess the influence of invasive medical
interventions, allowing the implementation of preventable measures to control infections.
(AJIC Am J Infection Control 1997;26:458-462)
Pr evi ous s t udi es have e mpha s i z e d t he cons e-
que nc e of i nf e c t i ons as a r eal t hr e a t t o i mmuno-
s uppr e s s e d hos t s . 1~3 Aggr e s s i ve a nt i ne opl a s t i c
c he mot he r a py has r e nde r e d pa t i e nt s wi t h c a nc e r
From the Infectious Disease Service and Hospital Infection Con-
trol Committee, National Cancer Institute, Rio de Janeiro.
Reprint requests: Eduardo Velasco, MD, Rua General Glicerio,
486/1002, 22245-120, Rio de Janeiro, Brazil.
Copyright 1997 by the Association for Professionals in Infection
Control and Epidemiology, Inc.
0196-6553/97 $5.00 + 0 17/46/76926
i mmunoc ompr omi s e d a nd t hus mor e vul ne r a bl e
t o i nf ect i ons . Never t hel es s , i mpr ove me nt s i n an-
t i mi c r obi a l agent s a nd s uppor t i ve me a s ur e s have
l e ngt he ne d s ur vi val a nd e ve n a t t a i ne d cur e i n
s ome ne opl a s t i c di seases, des pi t e t he l i f e- t hr eat -
e ni ng i nf e c t i ous c ompl i c a t i ons t ha t e ve nt ua l l y re-
qui r e s ome ki nd of i nt e ns i ve car e s uppor t . St udi es
have r e por t e d hi gh mor t a l i t y a nd c ons i de r a bl e
cost f or cr i t i cal l y ill c a nc e r pa t i e nt s a dmi t t e d t o
t he i nt e ns i ve car e uni t (ICU), 4,5 ma i nl y be c a us e of
t he sever i t y of t he unde r l yi ng di s eas e a nd t he ex-
pos ur e t o i nvas i ve pr oc e dur e s .
4 5 8
AJI C
V o l u m e 2 5 , N u m b e r 6 Ve l a s c o e t al . 4 5 9
Despite the importance of surveying high-risk
patients, few studies on infection control and sur-
veillance have been undertaken in this population
gr oup. 3,~-1 In developing countries the study of
nosocomial infection (NI) has not been as effec-
tive as possible because of the lack of a uniform
approach for measuring quality of care. The ob-
jective of this article is to report a 27-month study
of nosocomial infection surveillance in an oncol-
ogy ICU that is based on National Nosocomial
Infections Surveillance (NNIS) system methods
These data might serve for comparative studies
with other oncology ICUs and also help to develop
a more efficient surveillance system, allowing for
detailed infection control measures for these
i
high-risk patients.
i
MAT ERI AL AND METHODS
i
The Cancer Hospital in Rio de Janeiro, Brazil, is
a 206-bed refe+al teaching hospital for oncologic
patients with an average annual admission of ap-
proximately 5800 patients, 130,000 outpatient
consultations, and 5000 annual surgical proce-
dures. It is the main hospital of the National
Cancer Institute complex and has a very active
training and teaching program The study was
i
done in the siX-bed medical-surgical ICU. All pa-
tients treated ]in this unit for >48 hours were
prospectively Observed for the development of a
I
NI and the influence of device utilization (in-
i
dwelling urinatT catheter [IUC], central venous
catheter [CVC], and ventilator [VEN]) on infec-
tion rates
The data were collected from September 1993
through Noverpber 1995, and the analysis used
the protocol of the NNIS ICU component 1~ and
the standard definitions of NI according to the
I .
Centers for D~lsease Control and Prevention22
Patients were surveyed for NIs from the day of ad-
mission until 48 hours after ICU discharge, if hos-
pitalized, or ~ntil ICU discharge, if transferred
from the hospital. ICU-related infections were de-
fined as those [hat developed after a 48-hour stay
in the unit and ]through the first 48 hours after dis-
charge from th~ ICU. All infection rates, including
patient and patient-day and device- and non-de-
vice-associate d infections were calculated accord-
ing to the forI~ulas as described in the NNIS sys-
1 2 '
tern report ~The denominator data collected
I
were the total number of patients at risk, the total
number of patient-days, the days of IUC, CVC,
I
and VEN support per month.
The three strategies used for comparison of ICU
infection rates [considered the average severity-of-
illness score (ASIS), average length of stay (ALOS)
in the ICU, and the device utilization (DU) ratio.
The ASIS was defined according to the Hospital
Infections Program, Centers for Disease Control
and Prevention criteria 11(i.e., patients received a
score of A through E depending on the degree of
illness when admitted to the ICU). To calculate the
ASIS, points are ascribed to patients once a week
on the basis of the classification scheme. One
point is ascribed to postoperative patients requir-
ing routine postoperative observation, 2 points to
physiologically stable patients requiring prophy-
lactic overnight observation, 3 points for nursing
and monitoring, 4 points for physiologically un-
stable patients requiring intensive nursing and
medical care with the need for frequent reassess-
ment and adjustment of therapy, and 5 points for
physiologically unstable patients who are in a
coma or shock or who require cardiopulmonary
resuscitation or intensive medical and nursing
care with the need for frequent reassessment. The
monthly ASIS for the unit is obtained by dividing
the total number of points ascribed to patients by
the total number of patients in the ICU on the day
of weekly prevalence. The ALOS is a proxy for in-
fection risk and is calculated with the numerat or
as total patient-days added to the number of pre-
vious days spent in the ICU by patients present in
the ICU on the first day of the mont h and the
number of additional days patients present in the
ICU on the last day of this mont h will stay in the
ICU. The denominator is the number of patients
in the ICU on the first day of this mont h plus the
number of patients admitted to the ICU during
this month. 11The DU of an ICU is one measure of
the unit's invasive practices and may serve as a
marker for the severity of illness of the patients in
the unit. The DU is calculated by dividing the
number of device-days by the number of patient-
days. 11
To reduce the confounding influence of patient
ICU stay or device exposure on the infection rate,
we evaluated the correlation between overall NI
rate (patient or patient-days) and average length
of patient stay or device-days utilization In addi-
tion, we also compared the device-associated in-
fection rates for pneumonia, bloodstream infec-
tions, and urinary tract infections
RESULTS
During the 27-month period 370 infections oc-
curred in 623 new patient admissions over 4034
patient-days. The medians of the overall NI patient
and patient-day rates were 50.0 infections per 100
AJI C
4 6 0 Vel asco et al. December 1997
10
8
6
4
2
0
150
100
50
200
F i g . 1 A . Number of IUC-associated urinary tract infec-
tions and IUC-days correlation coefficient (r = 0.17,
p > 0.05).
8
0
_z
4
D
0
o
m 2
0
150
100
50
160
F i g , 1 B , Number of CVC-associated bloodstream infec-
tions and CVC-days correlation coefficient (r =0.62,
p < 0.001).
140
120
100
80
6 0
40
200
20
F i g ' 1 . Number of cases of VEN-associated pneumo-
nia and VEN-days correlation coefficient (r = 0.36,
p > 0,05).
T a b l e 1 . Devi ce-speci fi c utilization median ratios and
devi ce-associ ated infection rates
Range Median
CVC ut i l i zat i on r at i o*
VEN ut i l i zat i on r at i o*
Ur i nar y c a t h e t e r ut i l i zat i on r at i o*
CVC- a s s o c i a t e d BSl r a t e r
VEN- as s oc i at ed PNEU r a t e r
Ur i nar y c a t h e t e r - a s s o c i a t e d UTI rate1-
0. 66- 1, 02 0. 86
0. 31- 0. 81 0. 63
0. 59- 0. 99 0. 83
7. 5- 58, 8 25, 6
15. 4- 82. 4 41. 7
7. 9- 80. 6 27. 5
BSI, Bloodstream infection; PNEU, pneumonia; UTI, urinary tract infection.
*No. of device-days/No, of patient-days.
?(No. of specific device-associated infections/No, of devi ce- days) x 1000,
T a b l e 2 , Correlation between NI and overall DU ratio,
ASIS and ALOS
Correlation
coefficient p Value
DU and NI pat i ent r at e* 0. 56
DU and NI p a t i e n t - d a y rate1- 0, 33
ASl S and NI pat i ent r at e* 0. 54
ASl S a n d NI p a t i e n t - d a y r a t e r 0. 39
ALOS and NI pat i ent r at e* 0. 67
ALOS a n d NI p a t i e n t - d a y r a t e r 0, 22
<0. 01
>0. 05
<0, 01
>0, 05
<0. 01
>0. 05
*NI patient rate = (No, of infections/No, of patients) x 100.
1-NI pati ent-day rate = (No. of infections/No, pat i ent - days) x 1000.
pat i ent s (range 13.2 t o 100) and 91.7 i nfect i ons per
1000 pat i ent - days (range 35.7 t o 143.8). Pneumo-
ni a (28.9%), ur i nar y t r act i nfect i ons (25.6%), and
bl oods t r eam i nfect i ons (24.1%) wer e t he maj or re-
por t ed i nfect i ons. I sol at ed mi cr oor gani s ms wer e
Ent er obact er i aceae (29.7%), fungi (22.2%), Pseu-
domonas aeruginosa (13.2%), ot her gr am- negat i ve
bacilli (13%), Staphylococcus aureus (5.3%), and
ot her gr am- posi t i ve cocci (16.6%).
The numbe r of IUC days, CVC days, and VEN
days obser ved duri ng t he st udy was 3312, 3507,
and 2496 days, respectively. Table 1 shows t hat t he
medi an DU rat i os r anged f r om 0.63 for VEN t o
0.86 for CVC, wher eas t he hi ghest medi an device-
specific associ at ed i nfect i on rat e was 41.7 for VEN.
The ALOS r anged f r om 4.8 t o 20.4 days ( medi an
8.8), wher eas t he ASIS of pat i ent s r anged f r om 3.3
t o 4.5 ( medi an 4.0). Table 2 demons t r at es t he
st r ong posi t i ve cor r el at i on bet ween t he overall NI
pat i ent rat e and DU (r -= 0.56, p < 0.01), ASIS (r =
0.54, p < 0.01), and ALOS (r = 0.67, p < 0.01), but
a r educed and nonsi gni f i cant cor r el at i on wer e no-
t i ced when pat i ent - days was us ed in t he denomi -
nat or (r = 0.39 and 0.22, respect i vel y). When we
anal yzed hi gh-ri sk devi ces separ at el y (Figs. 1A t o
1C), a s i g n i f i c a n t p o s i t i v e c o r r e l a t i o n wa s
AJI C
Vo l u me 25, N u mb e r 6 Ve l a s c o e t al. 461
achi eved be t we e n t he n u mb e r of CVC-days a n d
as s oci at ed i nf ect i ons (r = 0.51, p = 0.01). Fig. 2
s hows a gr a dua l a n d pr ogr es s i vel y i ncr eas i ng
t r e nd i n t he NIs despi t e t he hi gh r at es of occur -
r ence at all s t r a t a of pa t i e nt stay.
!
D I S C U S S I ON
NIs are a n i mp o r t a n t cause of mor bi di t y a nd
mor t al i t y i n cfi ncer pat i ent s a dmi t t e d t o t he ICU. 13
Ma ny f act or s i ncr ease t he ri sk of i nf ect i ous com-
pl i cat i ons a nd cons equent l y af f ect t he i nf ect i on
r at es a nd t he ~r ognos i s of t hi s gr oup of hi gh- r i sk
pat i ent s. 14,1~ Mor e recent l y, several sur vei l l ance
st udi es bas ed o n t he NNI S s ys t em me t hods have
r epor t ed NI r at es i n ICUs. 16-19 However, an oncol -
ogy ICU woul d be expect ed t o have a gr oup of crit-
icaUy ill i mmUnos uppr e s s e d pat i ent s at hi gher in-
t r i nsi c ~ " " a nd ext r i nsi c ri sks f or i nf ect i ons wi t h l onger
st ay t h a n woul d an ICU wi t hout s uch pat i ent s.
These r i sk f act or s cer t ai nl y af f ect t he val i di t y of
overal l i nf ect i on r at e c ompa r i s ons a mo n g di f f er ent
I
ICU t ypes. A pr ecedi ng s t udy f r om t he Cent er s f or
Di sease Cont r bl a nd Pr event i on a nd t he NNI S sys-
t e m has al r eady e mpha s i z e d t hat dat a f r om cer t ai n
t ypes of ICUg s houl d be anal yzed separ at el y be-
i
cause of di ffe{ent i nf ect i on r at e di s t r i but i ons J 7
I
Our r es ul t s i s howe d a mu c h hi ghe r pa t i e nt - da y
r at e (91.7 i nf ect i ons per 1000 pat i ent - days ) a n d
i nvasi ve devi+es expos ur e (Tabl e 2) t h a n di d t he
pr evi ous l y pfubl i shed r e por t of NI r at es i n a
n o n o n c o l o g y ]adul t a n d pedi at r i c ICU. 17 We al so
f ai l ed t o f i nd a s t r ong c or r e l a t i on be t we e n t he
overal l pa t i e m- r a t e s a n d DU r at i os (Fig. 1), except
f or i
bl oods t r e a m i nf ect i ons a nd CVC expos ur e (r =
0.51), unl i ke ~he above st udy.
The ALOS Was al so not s t r ongl y as s oci at ed wi t h
NIs wh e n wd cont r ol l ed f or t he c o n f o u n d i n g in-
f l uence of l engt h of s t ay by us i ng pa t i e nt - da ys as
t he d e n o mi n a t o r (r = 0.22). However , st r at i f i ca-
t i on of l engt ~ of pa t i e nt s t ay s howe d t ha t NI
pa-
t i ent - day rat ~s t e n d e d t o be pr ogr es s i vel y hi ghe r
/
t he l onger t h~ pa t i e nt s s t ayed i n t he ICU.
[
Our h] gh- r at e o u t h e r r esul t s ma y r ef l ect a com-
bi na t i on of s~veral f act or s r el at ed t o t he sever i t y
of t he unde r l yi ng i l l ness, t he ne ut r ophi l count , in-
vasi ve pr ocedur es , us e of devi ces, a nd i nf ect i ons
t ha t us ual l y Occur i n i mmu n o c o mp r o mi s e d can-
cer pat i ent s . Adj us t ed i nf ect i on r at es i n an oncol -
ogy ICU pr ovi de speci f i c sur vei l l ance da t a f or
mo r e pr eci s~ i nt e r hos pi t a l c o mp o n e n t c ompa r -
i sons. They al so hel p assess t he i nf l uence of i nva-
sive me di c a l l i nt er vent i ons on t hes e hi gh- r i s k pa-
t i ent s , a l l owi ng f or t he i mp l e me n t a t i o n of
pr event i ve m~easures t o cont r ol i nf ect i ons .
120
100
I Ll
Z
o
I - -
o
LIJ
LI.
Z
. - 1
o
o
o
t / )
o
z
80
60
40
20
, ~ . ~ " ~ , . ~ " " ~
Fi g . 2. ALOS at lOS and cor r espondi ng NI pat i ent and
pat i ent -days rates. Stratification accor di ng to quarti i es.
R E F E R E N C E S
1. Schimpff SC. Infections in the cancer patient-diagnosis,
prevention, and treatment. In: Mandell GL, Bennett JE,
Dolin R, editors. Principles and practice of infectious dis-
eases. 4th edition. New York: Churchill Livingstone; 1995.
p. 2666-75.
2. Young LS. Nosocomial infections in the immunocompro-
mised adult. Am J Med 1981;70:398-404.
3. Velasco ED, Martins CA, Vidal E, et al. Infec96es em pa-
cientes neutrop~nicos. Rev Bras Cancer 1986;32:195-204.
4. Jonhnson MH, Gordon PW, Fitzgerald E Stratification of
prognosis in granulocytopenic patients with hematologic
malignancies using the APACHE II severity of illness score.
Crit Care Med 1986;14:693-7.
5. Schapira DV, Studnicki J, Bradhum DD, Wolff E Jarret A.
Intensive care, survival, and expense of treating critically
ill cancer patients. JAMA 1993;269:783-6.
6. Robinson GN, Tegtmeier BR, Zaia JA, et al. Brief report:
nosocomial infections rates in a cancer treatment center.
Infect Control 1986;5:289-94.
7. Rotstein C, Cummings M, Nicolau AL, et al. Nosocomial
infection rates at an oncology center. Infect Control Hosp
Epidemiol .1989;9:13 -9.
8. Aguiar N, Bermudez LE, Vidal E, et al. Controle das in-
fec~6es hospitalares no Instituto Nacional de Cfincer, M.
Saude. Rev Bras Cancer 1982;29:7-20.
A J I C
4 6 2 Vel as c o et al . December 1997
9. Bermudez LE, Vidal E, Velasco E, et al. Controle de in-
fec~go hospitalar: experi~ncia de dois anos. Rev Bras
Cancer 1984;30:6-13.
10. Velasco EV, Martins CAS, Vidal E, Carvalho AD,
Gaglianone TC. Infec96es nosocomiais em um hospital
oncoldgico. Rev Paul Med 1990;108:61-70.
11. Emori TG, Culver DH, Horan TC, Jarvis WR, et al.
National nosocomial infections surveillance system
(NNIS): description of surveillance methods. AJIC Am J
Infect Control 1991;19:19-35.
12. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM.
CDC definitions for nosocomial infections, 1988. AJIC Am
J Infect Control 1988;16:128-40.
13. Bone RC. Concepts in emergency and critical care: inten-
sive care, survival, and expense of treating critically ill can-
cer patients. JAMA 1993;269:783-6.
14. Schuster DE Marion JM. Precedents for meaningful recovery
during treatment in a medical intensive care unit: outcome in pa-
tients with hematologic malignancy. Am J Med 1983;75:402-8.
15. Lloyd-Thomas AR, Wright I, Lister TA, Hinds CJ.
Prognosis of patients receiving intensive care for life
threatening medical complications of haematological ma-
lignancy. BMJ 1988;296:1025-9.
16. Hospital Infections Program, National Center for In-
fectious Disease, Centers for Disease Control and Pre-
vention, Public Health Service, U.S. Department of
Health and Human Services. National Nosocomial In-
fections Surveillance (NNIS) semiannual report, May
1995: a report from the National Nosocomial Infec-
tions Surveillance (NNIS) System. Am J Infect Control
1995;23:377-85.
17. Jarvis WR, Edwards JR, Culver DH, Hughes JM, Horan T,
Emori TG, et al. Nosocomial infection rates in adult and
pediatric intensive care units in the United States. Am J
Med 1991;91(suppl 3B):185S-91S.
18. Gaynes RP, Martone WJ, Culver DH, et al. Comparison of
rates of nosocomial infections in neonatal intensive care
units in the United States. Am J Med 1991;91(suppl 3B):
192S-6S.
19. Josephson A, Karanfil L, Alonso H, Watson A, Blight J.
Risk-specific nosocomial infection rates. Am J Med
1991;91(suppl 3B):131S-7S.
B o u n d v o l u m e s a v a i l a b l e t o s u b s c r i b e r s
Bound volumes of AJIC: Ameri can Journal o f I nf ect i on Control are available to subscribers
(only) for the 1997 issues from the Publisher, at a cost of $39.00 for domestic, $49.22 for
Canadian, and $46.00 for international subscribers for Vol. 25 (February-December).
Shipping charges are included. Each bound volume contains a subject and an author index.
The binding is durable buckram with the journal name, volume number, and year stamped
in gold on the spine. Payment mu s t accompany all orders. Contact Mosby, Subscription
Services, 11830 Wesfline Industrial Dr., St. Louis, Missouri 63146-3318, USA; phone
800-453-4351 or 314-453-4351.
S u b s c r i p t i o n s m u s t b e i n f o r c e t o qua l i f y . B o u n d v o l u m e s a r e n o t a v a i l a b l e i n p l a c e
o f a r e g u l a r j o u r n a l s u b s c r i p t i o n .