Sie sind auf Seite 1von 7

390 Delle Karth et al.

, Outcome and functional capacity after prolonged intensive care unit stay
Original Article WIENER KLINISCHE
WOCHENSCHRIFT
The Middle European Journal
of Medicine
Wien Klin Wochenschr (2006) 118/13–14: 390–396
Printed in Austria
DOI 10.1007/s00508-006-0616-z

Outcome and functional capacity after prolonged intensive care unit stay
Georg Delle Karth, Brigitte Meyer, Sabine Bauer, Mariam Nikfardjam, and Gottfried Heinz
Department of Cardiology, Intensive Care Unit, Medical University of Vienna, Vienna, Austria

Received January 17, 2006, accepted after revision April 18, 2006
© Springer-Verlag 2006

Klinischer Verlauf und Morbidität nach langem zeigte, dass die Notwendigkeit zu einer Nierenersatzthe-
Intensivstationsaufenthalt rapie während des Intensivstationsaufenthaltes der einzig
Zusammenfassung. Studienhintergrund: Viele kri- unabhängige Prädiktor für das Versterben im Kranken-
tisch kranke Patienten, die ihre akute Erkrankung über- haus und innerhalb eines Jahres nach Intensivstations-
leben, verbleiben in einem abhängigen Zustand und be- Entlassung war (OR = 2,88; 95%CI 1,12–7,41, p = 0,028
nötigen für Wochen bis Monate eine Behandlung an der und OR = 3,66, 95%CI 1,36–9,83, p = 0,01). In 28/31 der
Intensivstation. Die Datenlage betreffend die Spitalsleta- Langzeit-Überlebenden Patienten (90%) mit einem Inten-
lität und insbesondere betreffend die Langzeit-Letalität sivstationsaufenthalt ≥ 30 Tagen zeigte der Barthel Index
bzw. die -Morbidität bei überlebenden Patienten ist je- keine oder nur mäßige Einschränkungen bei Alltagstätig-
doch noch spärlich. Ziel dieser Studie war es, den klini- keiten.
schen Verlauf und Prognosefaktoren bei Langzeit-kritisch Schlussfolgerung: Die Hospitals-Letalität bei kritisch
kranken Patienten zu untersuchen. kranken Patienten mit einer Intensivstationsaufenthalts-
Methoden: Diese retrospektive Beobachtungs-Kohor- dauer < 30 oder ≥ 30 Tagen ist vergleichbar. Die Notwen-
ten-Studie wurde an unserer gemischten kardiologischen digkeit zu einer Nierenersatztherapie war der einzige un-
8-Betten-Intensivstation in einem 2200-Betten-Universi- abhängige Prädiktor für das Versterben im Krankenhaus
tätsspital durchgeführt. Patientendaten wurden zwischen und für die 1-Jahres-Letalität bei Langzeit-Patienten. Kri-
dem 1. März 1998 und dem 31. Dezember 2003 analy- tisch kranke Patienten mit einer Intensivstationsaufent-
siert. Patienten mit einer Stations-Aufenthaltsdauer von haltsdauer ≥ 30 Tagen haben ein hohes und anhaltendes
≥ 30 Tagen bildeten die Studiengruppe. Die Evaluation Sterberisiko nach Spitalsentlassung. Dennoch, eine be-
der Morbidität und funktionellen Kapazität wurde mittels deutsame Anzahl von diesen Patienten sind Langzeit-
Telefoninterview unter Verwendung des Barthel Mobili- Überlebende mit keinen oder nur mäßigen Einschränkun-
täts-Scores durchgeführt. gen bei Alltagstätigkeiten.
Ergebnisse: Die Anzahl der Patienten mit einer Sta-
tions-Aufenthaltsdauer ≥ 30 Tagen betrug 135 (10% der Summary. Background: An important proportion of
Gesamtpatienten). Diese Gruppe belegte 5962 Betten- critically ill patients who survives their acute illness re-
tage, welches 40,9% der gesamten Bettenkapazität ent- mains in a critical state requiring intensive care manage-
sprach. Im Vergleich zu Patienten mit einer Stations- ment for weeks to months. Nevertheless, data on risk
Aufenthaltsdauer < 30 Tagen hatten die Patienten in der factors for in-hospital mortality and especially for long-
Langzeitgruppe einen signifikant höheren SAPS II Score term mortality and functional capacity are scarce. This
innerhalb von 24 Stunden nach Aufnahme (54 [IQR 41– study investigated outcome and prognostic factors in
65] vs. 38 [IQR 27–56], p < 0,001). Trendmäßig überwo- long-term critically ill patients.
gen die Männer in der Langzeitgruppe (98/135 [82,6%] Methods: This retrospective observational cohort
vs. 782/1215 [64,4%], p = 0,05). Unterschiede in der In- study was performed at our mixed adult 8-bed cardiologic
tensivstations- und Hospitalsletalität waren nicht signifi- ICU at a 2200-bed University Hospital. Patient data from
kant (28/135 [20,7%] vs. 295/1215 [24,3%], p = 0,620, our local database connected to an Austrian multicenter
und 46/135 [34,1%] vs. 360/1215 [29,6%], p = 0,285). Die program for quality assurance in intensive care were
Sterblichkeit bei Patienten, die den Spitalsaufenthalt analyzed. Data were collected between March 1st, 1998
überlebten, betrug nach einem bzw. nach vier Jahren and December 31st, 2003. Patients with an ICU stay ≥ 30
14% und 26% in der Kurzzeitgruppe verglichen mit 31% days formed the long-term study group. Morbidity and
und 61% in der Langzeitgruppe. Ein log-rank-Test er- functional capacity were assessed using the Barthel mo-
brachte eine signifikant höhere Überlebenswahrschein- bility index in telephone interviews.
lichkeit in der Kurzzeitgruppe nach Spitalsentlassung (log Results: Patients spending ≥ 30 days in the ICU num-
rank = 34,3, p < 0,001). Eine multivariate Datenanalyse bered 135 (10%) and occupied 5962 bed-days, repre-
Delle Karth et al., Outcome and functional capacity after prolonged intensive care unit stay 391

senting 40.9% of the total bed-days. Compared with pa- tients is generally poor [4–13]: Spicher and White report-
tients with an ICU stay < 30 days, patients in the long- ed that among patients requiring mechanical ventilation
term group had a significantly higher SAPS II score dur- for ≥ 10 days, > 70% were dead at 1 year after hospital
ing the first 24 hours after ICU admission (54 [IQR 41–65] discharge [4]; Hughes and coworkers reported that hos-
vs. 38 [IQR 27–56], p < 0.001). There was a trend to- pital mortality was 42% in patients with an ICU stay
wards male preponderance in the long-term group (98/ ≥ 30 days [11]. Nevertheless, data on risk factors for in-
135 [82.6%] vs. 782/1215 [64.4%], p = 0.05). Differences hospital mortality and especially for long-term mortality
in ICU and in-hospital mortality were not significant (28/ and functional capacity are scarce. This study investigated
135 (20.7%) vs. 295/1215 (24.3%), p = 0.620 and 46/135 outcome and prognostic factors in long-term critically ill
[34.1%] vs. 360/1215 [29.6%], p = 0.285, respectively). patients in the ICU.
After 12 and 48 months, the overall cumulative rates of
death in hospital survivors were 14% and 26%, respec- Patients and methods
tively in the short-term ICU group and 31% and 61% in
the long-term group. A log-rank test revealed a signifi- Patient data from our 8-bed adult, mixed cardiologic ICU
at a University Hospital were retrospectively analyzed from our
cantly higher probability of survival in the short-term
local database, which is connected to the Austrian Center for
group after hospital discharge (log rank = 34.3, p < 0.001).
Documentation and Quality Assurance in Intensive Care Med-
Multivariate analysis of hospital survivors and non-survi-
icine, a non-profit organization that has established a national
vors in the long-term group showed that the need for intensive care database and external quality assurance program
renal replacement therapy during the ICU stay was the [14]. Data were collected from all consecutive patients between
sole independent predictor for in-hospital death and March 1st, 1998 and December 31st, 2003. Data collection was
death within 1 year after ICU discharge (OR = 2.88; prospective and included demographic data, severity of illness
95%CI 1.12–7.41, p = 0.028 and OR = 3.66, 95%CI 1.36– as measured by the simplified acute physiology score (SAPS II)
9.83, p = 0.01, respectively). In 28/31 long-term survivors during the first 24 hours after ICU admission, and level of
(90%) in the long-term ICU group, the Barthel index provided care measured by the simplified therapeutic interven-
indicated no or only moderate disability during daily activ- tion scoring system (TISS-28). Outcome data included the vital
ities. status at the date of discharge from the ICU and the hospital.
Conclusion: Hospital mortality rates in critically ill Data were collected by staff physicians. Our ICU has a closed
patients with a stay < 30 or ≥ 30 days were comparable. design with staff intensivists. Most of our patients have medical
The necessity for renal replacement therapy was the sole conditions but about 1/3 are admitted after cardiac surgery;
independent predictor for in-hospital and 1-year mortality among the medical patients most are admitted because of cardi-
in long-term ICU patients. Critically ill patients with a stay ologic diagnoses. The nurse-to-patient ratio is roughly 1:1 dur-
≥ 30 days have a high and ongoing risk of death after ing the day and 1:2 during the night. The usual bed-occupation
hospital discharge; however, a substantial number of rate is > 95%. Discharge decisions are made by staff intensiv-
these patients are long-term survivors with no or only ists according to institutional criteria, which include that pa-
moderate disability during daily activities. tients must be respiratory-stable, free of mechanical ventilation
for about 24 hours and free of vasopressors or inotropes except
Key words: Critically ill, long-term care, survival, for low-dose dopamine or dobutamine. There are several inter-
morbidity. mediate-care wards in our hospital, but without facilities for
prolonged invasive or non-invasive mechanical ventilation.
Introduction Long-term mortality data up to December 31st 2003 were ob-
tained from the mortality register of the Austrian Statistics
A large and growing number of patients who survive Institute.
their acute critical illness remains in a critical state requir- Patients admitted for ≥ 30 days in our ICU formed the
ing prolonged intensive care management for weeks to long-term study group. If a patient was admitted more than
months [1]. Official data indicate that in the USA this once to the unit only the first stay was analyzed.
“chronically” ill patient group (diagnosis-related Group For survivors in the long-term ICU group, details on mor-
483) exceeds 100,000 patients annually and consumes bidity and functional capacity were requested and scaled using
considerable ICU resources [2]. “Chronic” critical illness the Barthel mobility index via telephone interview (Appendix).
is not only regarded as a prolongation of acute critical The Barthel index was introduced in 1965 by Mahoney and
illness but as a syndrome involving derangements in neu- Barthel [15] and is used for estimation and determination of
rologic, immunologic, endocrinologic and metabolic func- ability to perform the basic activities of daily life; the index is
tion [3]. Dependency on mechanical ventilation, pro- used internationally, validated and has a good retest and inter-
longed circulatory failure with the need for vasopressors rater reliability. Use of the index is simple and time-efficient,
or inotropes and multisystem problems impede transfers and interviews can be satisfactorily conducted by telephone.
to step-down wards and predispose to nosocomial compli- This study was approved by the Ethics Committee of the Med-
cations. There is still some controversy over whether pro- ical University of Vienna.
longed ICU management is overall beneficial or repre-
sents futile and sometimes painful care in most cases [1]. Statistical methods
Moreover in an era of cost containment the allocation of Continuous data are given as median and interquartile
disproportionate ICU resources to this relatively small range (IQR, range from the 25th to the 75th percentile). Categor-
number of patients requires justification, and careful ex- ical data are given as counts and relative frequencies. Chi-
amination of this cohort of patients is warranted. Existing squared tests or, if appropriate, Fisher’s exact tests were used to
data indicate that outcome of long-term critically ill pa- compare groups of categorical data and to test for trends. The
392 Delle Karth et al., Outcome and functional capacity after prolonged intensive care unit stay

Mann-Whitney U-test was used to compare groups of continu- vs. female patients, respectively, p = 0.89). Female patients
ous data. Multivariate logistic regression analysis was used to were significantly older than male patients (67 [IQR 55–
assess the independent effects of different baseline and clinical 76] vs. 62 [IQR 54–72]) years, p < 0.001). There were no
variables on survival and to adjust for potential confounding significant sex differences in ICU, in-hospital or long-term
factors. Baseline and clinical variables were selected for the mortality (24.3 vs. 23.2%, 30.7 vs. 28.9% and 46.9 vs.
model if they were significant predictors for hospital mortality 47.1%, male vs. female patients, p = 0.68, p = 0.5 and
in univariate comparisons. Patients’ ages and the SAPS II p = 0.89, respectively). Overall, 67% of patients were med-
scores during the first 24 hours after ICU admission were also ical and 33% were surgical.
included. We used a hierarchical modeling strategy to assess
the effect of demographic variables and clinical variables sep-
Long-term ICU patients
arately and jointly. Interaction was assessed using multiplica-
tive interaction terms and log-likelihood ratio Chi-squared Long-term patients spending ≥ 30 days in ICU num-
tests, and an analysis of residuals was performed. Regression bered 135 (10%); they occupied 5962 bed-days, represent-
diagnostics and overall model-fit were used according to stan- ing 40.9% of the total bed-days, and their median ICU
dard procedures; the Hosmer-Lemeshow test was used for glo- stay was 39 (IQR 33–50) days. These patients were on
bal goodness-of-fit testing [16]. Survival was calculated from mechanical ventilation for a median of 33/39 (IQR 26–45)
the date of hospital discharge to December 31st 2003 using days and were on vasopressors or inotropes for 28/39
Kaplan-Meier plots. No minimum follow-up time was defined; (IQR 22–37) days. Fifty-six patients (42%) in the long-
median follow-up was 694 (IQR 169–1407) days. A log-rank term group received renal replacement therapy (RRT) be-
test was used to compare survival times between the short-term cause of acute renal failure during their ICU stay. Differ-
and long-term groups. A two-sided p-value < 0.05 was consid- ences in baseline and clinical variables between the long-
ered as statistically significant. and short-term groups are shown in Table 1. Patients in
the long-term group had significantly higher SAPS II
Results scores on admission and there was a trend towards male
Study population characteristics preponderance. There were no significant differences be-
A diagram of identification and enrolment of both tween the groups in age or in the type of admission, and
short- and long-term ICU patients is shown in Fig. 1. The there were no significant differences in ICU or in-hospital
median age of the patients was 64 (IQR 54–73.2) years and mortality.
the median length of ICU stay was 5 (IQR 3–13) days. The
median SAPS II score was 40 (IQR 29–58) and the mean Survival after hospital discharge
predicted hospital death rate was 25% (IQR 10–64%). The Among the hospital survivors, the overall cumulative
patient group comprised 470 women (34.8%) and 880 men rates of death were 14%, 20% and 26% in the short-term
(65.2%). There were no significant sex differences in the ICU group and 31%, 44% and 61% in the long-term
SAPS II scores (38 [IQR 27–57] vs. 38 [IQR 27–55] male group, after 12, 24 and 48 months, respectively. Two

Fig. 1. Diagram of patient identification and enrolment


Delle Karth et al., Outcome and functional capacity after prolonged intensive care unit stay 393

Table 1. Baseline and outcome data in patients with an ICU stay ≥ 30 days (long-term group) and < 30 days (short-term group)

Long-term group (n = 135) Short-term group (n = 1215) p Value

Age, years (IQR) 64 (58–71) 63 (54–74) 0.869


Sex, female (%) 37 (27.4) 433 (35.6) 0.057
SAPS II score (IQR) 54 (41–65) 38 (27–56) < 0.001
Length of ICU stay, days (IQR) 39 (33–50) 4 (2–10) < 0.001
Type of admission, no. (%) 0.329
medical 99 (73) 815 (67.1)
surgical 36 (27) 400 (32.9)
ICU mortality, no. (%) 28 (20.7) 295 (24.3) 0.620
Hospital mortality, no. (%) 46 (34.1) 360 (29.6) 0.285

SAPS simplified acute physiology score.

hundred and twenty-six patients died after hospital dis- the need for RRT during the ICU stay was the sole inde-
charge. Median survival was 7 (IQR 2–19) months. Medi- pendent predictor for in-hospital death (Table 3) and for
an follow-up was 33 (IQR 15–51) months for the 718 death within 1 year after ICU discharge (OR = 3.66,
patients still alive in December 2003. Kaplan–Meier anal- 95%CI 1.36–9.83, p = 0.01).
ysis revealed a high and ongoing risk of death in the long-
term group throughout the study period (Fig. 2). A log- Morbidity of hospital survivors in the long-term
rank test revealed a significantly higher probability of ICU group
survival after hospital discharge in the short-term group Of the 89 patients in the long-term group discharged
than in the long-term group (log rank = 34.3, p < 0.001). from hospital, 43 were alive at December 31, 2003. The
Barthel index was obtained for 31 patients (72%) at Feb-
Predictors for hospital and 1-year survival in ruary 2005. Direct interview was conducted by telephone
long-term ICU patients for 31 patients, and for two patients with a high degree of
Table 2 compares hospital survivors and non-survi- disability the interviews were conducted with the closest
vors in the long-term ICU group. Among patients who did family members. The Barthel index was assessed after a
not survive hospital, there were significantly more men median of 1321 (IQR 623–1902) days after hospital dis-
and patients with a history of chronic renal failure. In charge. The shortest follow-up period was 14 months. In
addition, non-survivors had a significantly higher median 13 patients (42%) the index was > 90, representing no
daily TISS score, were longer on mechanical ventilation, disability during activities of daily life; in 15 patients
vasopressors or inotropes, and had received parenteral (48%) the index was between 55 and 90, representing
nutrition. Significantly more non-survivors received RRT moderate disability; in 3 patients (9%) the index was < 55,
during their ICU stay. Multivariate analysis revealed that representing severe disability.

Discussion
Recent therapeutic and technological advances have
improved both the short- and long-term prognosis of crit-
ically ill patients [17, 18]. However, an important and
growing number of patients who survive their acute criti-
cal illness remains in a critical state requiring intensive
care for weeks to months [1]. The ability of medical staff
to distinguish between life-saving therapies and futile care
in intensive care medicine has become challenging [19].
Outcome data are important, especially in the current
situation of limited resources in all healthcare systems;
nevertheless, studies on the acute and long-term prognosis
of “chronically“ ill patients are rare. This study was de-
signed to evaluate the short- and long-term outcomes,
including functional capacity, in patients requiring ≥ 30
days of treatment in the ICU.
Our data indicate that, although patients requiring
≥ 30 days of ICU treatment were relatively few in number,
Fig. 2. Kaplan-Meier estimates of the cumulative probability a considerable proportion of our ICU resources was allo-
of survival as a function of the numbers of days after hospital cated to this group. Mortality after hospital discharge was
discharge for patients with an ICU stay ≥ 30 days (solid line) significantly higher in long-term patients than in those
and patients with an ICU stay < 30 days (dotted line) requiring shorter intensive care. Not surprisingly, there
394 Delle Karth et al., Outcome and functional capacity after prolonged intensive care unit stay

Table 2. Baseline and clinical data of hospital survivors and non-survivors in patients with an ICU stay ≥ 30 days

Hospital survivors (n = 89) Hospital non-survivors (n = 46) p Value

Age, years (IQR) 63 (58–71) 64.5 (58.8–71) 0.496


SAPS II score (IQR) 52 (41–63) 55.5 (43–70) 0.200
Admission Diagnoses 0.677
medical
cardiac 40 18
respiratory 11 7
sepsis 5 4
other 9 5
surgical
cardiac 23 12
other 1 0
Sex male/female, no (%) 60/29 (67/33) 39/7 (84/15) 0.031
Chronic renal failure, no (%) 18 (20.2) 19 (41.3) 0.009
Chronic heart failure, no (%) 27 (30.7) 16 (34.8) 0.629
Length of ICU stay, days (IQR) 37 (32.5–48.5) 40 (33–54) 0.232
Daily TISS score (IQR) 34.9 (32.8–36.3) 35.6 (33.2–38.5) 0.037
RRT during ICU stay, no (%) 26 (29.2) 30 (65.2) < 0.001
Days with parenteral nutrition (IQR) 9 (3–19) 16 (6.7–24) 0.034
Days on mechanical ventilation (IQR) 32 (23–40.5) 38.5 (30–48.5) 0.006
Days on vasopressors/inotropes (IQR) 27 (21.5–33.5) 32 (23.7–39) 0.019

RRT renal replacement therapy; SAPS simplified acute physiology score; TISS simplified therapeutic intervention scoring system.

were no differences in ICU and in-hospital mortality be- ing is not surprising and indicates that the severity of
tween long- and short-term ICU patients, because all acute illness defined by the SAPS II score at the beginning of
deaths were counted among the short-term group. As the ICU stay is not only a good predictor for in-hospital
shown in the Kaplan–Meier curve, the higher risk of death mortality but also for the length of ICU stay and the time
in the long-term group continued throughout the observa- required to recover from critical illness. This observation
tion period, but nevertheless a substantial number of long- is in line with the findings of Schuster and co-workers
term ICU patients were long-time survivors with good or [20]. The more severe an acute illness, the higher the risk
only modestly reduced functional capacity. Our outcome of dying and also of remaining in a critical state for a
data compare well with other studies on “chronic” critical- prolonged period of time. It is noteworthy that long-term
ly ill patients, where in-hospital mortality has been report- patients were on mechanical ventilation for 85% of all
ed to be 37–49% and long-term mortality 55–67% [4–13]. days spent in the ICU and were on vasopressors or ino-
Our data show that SAPS II scores during the first 24 tropes for 72% of the time, indicating that long-term
hours after ICU admission were significantly higher in patients required not only ICU monitoring but frank ICU
long-term patients than in short-term patients. This find- treatment during their ICU stay [21, 22] and also suggest-

Table 3. Multivariate analysis of risk factor for hospital death in patients with an ICU stay ≥ 30 days

Odds ratio 95% CI p Value

Age* 1.13 0.83–1.54 0.428


SAPS II score* 1.07 0.80–1.43 0.625
Male sex 2.03 0.71–5.77 0.183
Chronic renal failure 1.55 0.62–3.86 0.344
Daily TISS score* 0.95 0.67–1.34 0.779
Renal replacement therapy 2.88 1.12–7.41 0.028
Days with parenteral nutrition* 1.12 0.82–1.52 0.132
Days on mechanical ventilation* 1.21 0.86–1.71 0.262
Days on vasopressors/inotropes* 0.98 0.68–1.39 0.910

* Quintiles. Model fit c = 12.6; df = 8; p = .124. SAPS simplified acute physiology score; TISS simplified therapeutic intervention
scoring system.
Delle Karth et al., Outcome and functional capacity after prolonged intensive care unit stay 395

ing that the main obstacle to transfer to a step-down ward of hospital resources is spent. Hospital mortality rates in
was dependency on mechanical ventilation. critically ill patients with a stay < 30 or ≥ 30 days were
Acute renal failure with the need for RRT was identi- comparable. The necessity for RRT was the sole indepen-
fied as the sole independent predictor for in-hospital mor- dent risk factor for in-hospital death and 1 year mortality
tality and death within 1 year after ICU discharge in the in long-term patients. Critically ill patients with an ICU
long-term group; only 14 patients (25%) who received stay ≥ 30 days have a higher and ongoing risk of death
RRT during their ICU stay were alive 1 year after ICU after hospital discharge; however, a substantial number of
discharge. Acute renal failure commonly develops as a these patients are long-term survivors with no or only
manifestation of multiorgan failure and its role as inde- moderate disability during activities of daily life, support-
pendent prognostic factor in critically ill patients has been ing the concept that prolonged ICU management is not
shown in many cohorts including long-term ICU patients futile per se. Identification of patients with a poor progno-
[11, 23, 24]. Nowadays, acute renal failure is considered sis for whom prolonged ICU treatment is no longer ben-
not only as a marker for the severity of the underlying eficial places ICU practitioners in a position of great
illness but also as a promoter of systemic inflammatory responsibility. Further research is warranted to better de-
processes [24]. The unique role of renal failure as a prog- scribe the “syndrome” of chronic critical illness and also
nostic factor is supported by our observation that the to focus on socio-economic aspects of this highly de-
severity of illness during the first 24 hours after ICU manding entity.
admission had no significant influence on in-hospital
mortality on univariate or multivariate analysis. Other Appendix
surrogate markers for pulmonary, cardiovascular and gut The Barthel Index [15]
dysfunction were not independently associated with in- FEEDING
hospital death on multivariate analysis. 0 = unable
It is noteworthy that 90% of long-term survivors had 5 = needs help cutting, spreading butter, etc., or requires
no or only moderate disability during daily activities, modified diet
which compares well with other studies on functional 10 = independent
capacity after prolonged intensive care [8–10, 25–28]. It
BATHING
must be emphasized that the majority of our patients were
0 = dependent
able to live an independent life; only one patient was 5 = independent (or in shower)
dependent on long-term care facilities. This finding is in
contrast to the widely held belief, shared by many ICU GROOMING
practitioners, that if long-term ICU patients survive for an 0 = needs to help with personal care
extended period after hospital discharge their functional 5 = independent face/hair/teeth/shaving (implements
status is generally poor. provided)
There are some limitations to the present study. First, DRESSING
there are varying definitions of what constitutes prolonged 0 = dependent
ICU-stay and therefore the cut-off value of 30 days is 5 = needs help but can do about half unaided
arbitrary. Despite this caveat, 30 days is a reasonable 10 = independent (including buttons, zips, laces, etc.)
length of time for distinguishing between short- and long- BOWELS
term morbidity and mortality after an acute illness. Never- 0 = incontinent (or needs to be given enemas)
theless, because of the ambiguity in defining long-term 5 = occasional accident
ICU stay it is difficult to compare results of studies. A 10 = continent
second limitation is the 28% drop-out rate in assessment
BLADDER
of functional capacity in long-term group. Although an 0 = incontinent, or catheterized and unable to manage alone
inclusion rate of 70% is considered acceptable [29], high- 5 = occasional accident
er inclusion rates would have enhanced the generalizabil- 10 = continent
ity of the “morbidity” results in long-term survivors. Also,
because of the nature of the Barthel index, no conclusions TOILET USE
on the psychosocial wellbeing of the patients can be 0 = dependent
drawn. Lastly, since this was a single center study, find- 5 = needs some help, but can do something alone
10 = independent (on and off, dressing, wiping)
ings could be markedly influenced by local strategies e.g.
by end-of-life decisions in the ICU. It should be acknowl- TRANSFERS (BED TO CHAIR AND BACK)
edged, however, that mortality rates in our ICU are consis- 0 = unable, no sitting balance
tent with the severity of illness as measured by the SAPS 5 = major help (one or two people, physical), can sit
II score, and end-of-life practice in our ICU is in line with 10 = minor help (verbal or physical)
recently published national guidelines [30]. Thus, cautious 15 = independent
generalization of our data to other cohorts of critically ill MOBILITY (ON LEVEL SURFACES)
patients may be allowed. 0 = immobile or < 50 yards
5 = wheelchair independent, including corners, > 50 yards
10 = walks with help of one person (verbal or physical)
Conclusion
> 50 yards
Long-term ICU patients are an important subgroup of 15 = independent (but may use any aid; for example, stick)
critically ill patients on whom a considerable proportion > 50 yards
396 Delle Karth et al., Outcome and functional capacity after prolonged intensive care unit stay

STAIRS 16. Hosmer DW, Lemeshow S (1989) Applied logistic regres-


0 = unable sion, 1st edn. Wiley, New York
5 = needs help (verbal, physical, carrying aid) 17. Vincent JL (2004) Evidence-based medicine in the ICU:
10 = independent important advances and limitations. Chest 126: 592–600
18. Hochman JS, Sleeper LA, Webb JG, Sanborn TA, White
TOTAL (0–100)
HD, Talley JD, et al (1999) Early revascularization in
acute myocardial infarction complicated by cardiogenic
References shock. SHOCK Investigators. Should we emergently re-
1. Carson SS, Bach PB (2002) The epidemiology and costs of vascularize occluded coronaries for cardiogenic shock.
chronic critical illness. Crit Care Clin 18: 461–476 N Engl J Med 341: 625–634
2. Available at: http://hcup.ahrq.gov/HCUPnet.asp. Accessed 19. Danis M (2004) How will we respond to chronic critical
August 13th, 2005 illness? Crit Care Med 32: 1617–1618
3. Nierman DM, Nelson JE (2002) Chronic critical illness. 20. Schuster HP, Wilts S, Ritschel P, Schuster FP (1996) Pre-
Crit Care Clin 18: 461–715 dictive value of score parameters of the Simplified Acute
4. Spicher JE, White DP (1987) Outcome and function fol- Physiology Score (SAPS)-II for the duration of treatment
lowing mechanical ventilation. Arch Intern Med 147: 421– of intensive care patients. Wien Klin Wochenschr 108:
425 462–466
5. Nasraway SA, Button GJ, Rand WM, Hudson-Jinks T, 21. Keenan SP, Busche KD, Chen LM, Esmail R, Inman KJ,
Gustafson M (2000) Survivors of catastrophic illness: Out- Sibbald WJ (1998) Withdrawal and withholding of life
come after direct transfer from intensive care to extended support in the intensive care unit: a comparison of teach-
care facilities. Crit Care Med 28: 19–25 ing and community hospitals. The Southwestern Ontario
6. Carson SS, Bach PB, Brzozowski L, Leff A (1999) Out- Critical Care Research Network. Crit Care Med 26: 245–
comes after long-term acute care. An analysis of 133 me- 251
chanically ventilated patients. Am J Respir Crit Care Med 22. Graf J, Graf C, Janssens U (2002). Analysis of resource
159: 1568–1573 use and cost-generating factors in a German medical inten-
7. Venker J, Miedema M, Strack van Schijndel RJ, Girbes sive care unit employing the Therapeutic Intervention
AR, Groeneveld AB (2005) Long term outcome after 60 Scoring System (TISS-28). Intensive Care Med 28: 324–
days of intensive care. Anaesthesia 60: 541–546 331
8. Lipsett PA, Swoboda SM, Dickerson J, Ylitalo M, Gordon 23. Lameire N, Van Biesen W, Vanholder R (2005) Acute renal
T, Breslow M, et al (2000) Survival and functional out- failure. Lancet 365: 417–430
come after prolonged intensive care unit stay. Ann Surg 24. Druml W (2004) Epidemiology Acute renal failure is not
231: 262–268 a “cute” renal failure! Intensive Care Med 30: 1886–1890
9. Montuclard L, Garrouste-Orgeas M, Timsit J, Misset B, De 25. Niskanen M, Ruokonen E, Takala J, Rissanen P, Kari A
Jonghe B, Carlet J (2000) Outcome, functional autonomy, (1999) Quality of life after prolonged intensive care. Crit
and quality of life of elderly patients with a long-term Care Med 27: 1132–1139
intensive care unit stay. Crit Care Med 28: 3389–3395 26. Stricker KH, Cavegn R, Takala J, Rothen HU (2005) Does
10. Combes A, Costa MA, Trouillet JL, Baudot J, Mokhtari M, ICU length of stay influence quality of life? Acta Anaes-
Gibert C, et al (2003) Morbidity, mortality, and quality-of- thesiol Scand 49: 975–983
life outcomes of patients requiring 14 days of mechanical 27. Nielsen D, Sellgren J, Ricksten SE (1997) Quality of life
ventilation. Crit Care Med 31: 1373–1381 after cardiac surgery complicated by multiple organ fail-
11. Hughes M, MacKirdy FN, Norrie J, Grant IS (2001) Out- ure. Crit Care Med 25: 52–57
come of long-stay intensive care patients. Intensive Care 28. Douglas SL, Daly BJ, Gordon N, Brennan PF (2002)
Med 27: 779–782 Survival and quality of life: short-term versus long-term
12. Gracey DR, Naessens JM, Krishan I, Marsh HM (1992) ventilator patients. Crit Care Med 30: 2655–2662
Hospital and posthospital survival in patients mechanically 29. Dillman DA (1978) Mail and telephone surveys: The total
ventilated for more than 29 days. Chest 101: 211–214 design method. John Wiley&Sons, New York
13. Martin CM, Hill AD, Burns K, Chen LM (2005) Character- 30. Empfehlungen zum Thema Therapiebegrenzung und -be-
istics and outcomes for critically ill patients with prolonged endigung an Intensivstationen Konsensuspapier der Inten-
intensive care unit stays. Crit Care Med 33: 1922–1927 sivmedizinischen Gesellschaften Österreich. Wien Klin
14. Metnitz PG, Vesely H, Valentin A, Popow C, Hiesmayr M, Wochenschr 116: 763–767
Lenz K, et al (1999) Evaluation of an interdisciplinary data
set for national intensive care unit assessment. Crit Care Correspondence: Dr. Georg Delle Karth, Department of
Med 27: 1486–1489 Cardiology, Medical University of Vienna, Währinger Gürtel
15. Mahoney FI, Barthel DW (1965) Functional evaluation: 18–20, 1090 Vienna, Austria,
the Barthel Index. Md State Med J 14: 61–65 E-mail: georg.delle-karth@meduniwien.ac.at

Das könnte Ihnen auch gefallen