Skip Navigation


BJA Advance Access originally published online on December 22, 2007
British Journal of Anaesthesia 2008 100(2):219-223; doi:10.1093/bja/aem372
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
100/2/219    most recent
aem372v1
Right arrow E-Letters: Submit a response to the article
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Conlon, N.
Right arrow Articles by Marsh, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Conlon, N.
Right arrow Articles by Marsh, B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?


© The Board of Management and Trustees of the British Journal of Anaesthesia 2007. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

Long-term functional outcome and performance status after intensive care unit re-admission: a prospective survey

N. Conlon1,*, B. O'Brien1, G. P. Herbison2 and B. Marsh1

1 Department of Anaesthesia and Intensive Care, Mater Misericordiae University Hospital, Eccles St, Dublin 7, Ireland
2 Department of Preventive and Social Medicine, Dunedin School of Medicine, PO Box 913, Dunedin 9054, New Zealand

* Corresponding author. E-mail: tallniamh{at}hotmail.com

Accepted for publication November 5, 2007.


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 References
 
Background: Intensive care unit (ICU) re-admission identifies a high-risk group in terms of hospital mortality, length of stay, and resource utilization. Only hospital and ICU mortality are well described in the literature on critically ill patients needing re-admission.

Methods: With ethical committee approval, from a prospectively collected database of all admissions to a combined medical and surgical ICU from January 1 to December 31, 2004, we identified all ICU re-admissions from within the hospital and analysed the factors associated with increased incidence of re-admission. At 2–3 yr after discharge, we evaluated the functional outcome of the surviving re-admitted patients as Glasgow Outcome Score (GOS) and Karnofsky index and identified determinants of both mortality and good functional outcome.

Results: Seventy-three (7.4%) of the 1061 patients who survived their first ICU stay were re-admitted during the study period. Of the 73 re-admitted patients, 14 died in ICU, 17 died later in the same hospital stay, and 10 died in the interim. Thus, 32 (43.8%) were alive 2–3 yr after discharge. The median [IQR] GOS of the survivors was 4 (see Mackle and colleagues in One year outcome of intensive care patients with decompensated alcoholic liver disease. Br J Anaesth 2006; 97: 496–8 and Jennett and Bond in Assessment of outcome after severe brain damage. A practical scale. Lancet 1975; 1: 480–4) and their median Karnofsky index was 70 [70–90] at follow-up. The only independent predictors of favourable functional outcome at follow-up were lower Apache 2 scores on both first ICU admission and on re-admission.

Conclusions: Although the ICU, hospital, and subsequent mortalities are high in patients after ICU re-admission, most survivors at 2–3 yr had by then made a good functional recovery and were independent.

Keywords: complications; intensive care; recovery


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 References
 
Typically, 4–14% of patients discharged from Intensive Care Units (ICUs) will be re-admitted.1 ICU re-admission identifies a high-risk group in terms of hospital mortality, length of stay, and resource utilization. Various studies have found hospital death rates in patients re-admitted to ICU to be increased by a factor ranging from 2 to almost 10.1 Indeed, an 11-fold increase in relative risk of mortality was described after adjusting for severity of illness in a prospective cohort study.2 Similarly, the length of hospital stay is at least twice as long as for those not re-admitted.1 2 It is even argued that the ensuing increase in costs imposes a ‘fiscal responsibility’ on health care practitioners to inform themselves of the outcomes of such patients, and indeed all who respond poorly to intensive therapy, so that we may use these expensive and limited resources more prudently.3 4

A consistent pattern in recent ICU outcome studies has been to record more meaningful measures than survival alone. However, only hospital and ICU mortality are well described in the literature on critically ill patients needing re-admission.

Many instruments have been devised to measure functional outcome and quality of life, depending on the patient population. One such instrument for measuring functional outcome is the Glasgow Outcome Score (GOS), which is well validated in trauma,5 cardiac arrest,6 and some medical7 8 and cardiothoracic patients,9 despite certain limitations.10 Another measurement tool, the Karnofsky index,11 is a similarly well established tool with proven validity and reliability for the assessment of independent functioning in the critical illness12 and cardiac surgery setting.13 14 The Karnofsky index is a descriptive, ordinal scale that ranges from 100 (good health) to 0 (dead) and emphasizes physical performance and dependency, whereas the GOS measures broader categories of functional outcome. A GOS of 4 or 5 or a Karnofsky index of 70–100 is generally considered a favourable functional outcome.15 16

We hypothesized that we could identify patient variables during a first ICU admission that would predict an increased risk of subsequent ICU re-admission, and furthermore, that patient variables during ICU re-admission could determine both mortality and good functional outcome at follow-up. The aims of our study were to describe the incidence of re-admission to the ICU during a 1 yr period and to identify factors associated with an increased risk of re-admission. Having identified this cohort, we then aimed to evaluate their GOS and Karnofsky indices 2–3 yr after discharge and identify possible determinants of their functional outcome.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 References
 
This study was conducted in an 18-bed, university-affiliated, tertiary referral combined medical and surgical ICU. This unit is located in a 570-bed, inner-city hospital that serves as the national centre for cardio-thoracic surgery, including transplantation, and for spinal injuries.

After institutional ethics committee approval, a prospectively recorded database of ICU admissions was used to identify all cases of ICU re-admission from January 1 to December 31, 2004. ICU re-admission was defined for the purposes of this study as a second (or further) admission to the ICU during a single hospital stay. The prospectively collected data for all ICU admissions included socio-demographic data, such as age and sex, the reason for admission which was based on a pre-defined list of medical and surgical diagnoses, specialty category, nature of the admission (i.e. elective or emergency), severity of illness as measured by Apache 2 score, length of ICU stay, and survival status at ICU discharge.

The subset of re-admitted patients had additional data recorded. These included the source of re-admission, the timing of re-admission [classified as immediate (within 24 h), early (24–72 h), and late (after 72 h)],17 length of stay, Apache 2 score on re-admission, number of re-admissions, and number of organs supported.

In January 2007, a median of 2.5 yr after discharge, we analysed the charts of all re-admitted patients and identified the survivors. We used structured interviews by telephone18 to assess the functional outcome in the form of GOSs and Karnofsky indices. We analysed factors that predicted mortality at 2.5 yr and factors that predicted good functional outcome (defined as a GOS of 4–5 or Karnofsky index of 70–100).19–22

Statistical analysis was performed using Stata v9. For tests of statistical significance, Student’s t-test and {chi}2 test were used. Logistic regression was used to explore the influence of a series of variables on both survival status and good functional outcome at a median of 2.5 yr after re-admission.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 References
 
During the period from January 1 to December 31, 2004, 1160 patients were admitted to our ICU. Of these, 99 patients died during their first admission. Of the 1061 patients who survived their first ICU stay, 73 (7.4%) patients were discharged and then re-admitted to the ICU during the same hospital stay, the remaining were not re-admitted. Of the re-admitted patients, 16 patients were re-admitted twice, 4 patients were re-admitted three times, and 1 patient was re-admitted four times. The diagnoses leading to re-admission were categorized as respiratory in 46 patients, cardiovascular in 13, septic shock in 9, gastrointestinal in 3, and neurological in the remaining 2 patients. The commonest cause of re-admission was pulmonary infection, which accounted for 28 patients, or 38% of all re-admissions.

During their first ICU stay, patients who were re-admitted differed from other patients in several respects (comparisons based on ICU day 1 data during the first admission are presented in Table 1). They were older and more likely to have been admitted to the unit in emergency circumstances. They also had higher Apache 2 scores on their original admission. Patients with a general surgical or vascular surgical admission diagnosis were relatively over-represented in the re-admitted group, whereas the reduced likelihood of cardiothoracic patients requiring re-admission was just below the significance level with P=0.054. Finally, re-admitted patients had a two-fold increase in incidence of death in the ICU (19.2 vs 9.1%, P=0.005).


View this table:
[in this window]
[in a new window]

  		Table 1 Comparing re-admitted with non-re-admitted patients who survived their first ICU stay. Data are mean (SD) or absolute numbers and percentage

 
Of the 73 re-admitted patients, 45 (61.6%) were male and 28 (38.4%) were female, with a mean age of 66.9 yr. Their mean (SD) Apache 2 score on re-admission was 16.8 (6.39) and they had longer stays [6.4 (8.2) days] than on their first admission [4.7 (7.8) days]. All the re-admissions were emergent in nature. Of the 73 patients, 14 (19.2%) died in ICU and 17 (23.3%) died later during the same hospital stay. Thus, 42 patients (57.5%) of the re-admitted 73 were discharged alive from the hospital. A further 10 of these patients (13.7% of the original group or 23.8% of hospital survivors) had died in the intervening 2–3 yr period. The median [IQR] GOS of the group was 1 [1–4] and the median Karnofsky index was 0 [0–70] at follow-up (see Figs 1 and 2).


Figure 1
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig 1 Re-admitted patients: Glasgow Outcome Scores at follow-up.

 


Figure 2
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig 2 Re-admitted patients: Karnofsky indices at follow-up.

 
There were 32 surviving patients at the time of follow-up. The analysis of outcome data on this subgroup showed that they, the survivors, had a median GOS of 4 [4–5] and median Karnofsky index of 70 [70–90]. Of the 32 survivors, 29 were at home whereas 3 were discharged from hospital to nursing homes where they remained. We used a multivariate model to analyse prognostic factors for both mortality and good functional outcome in the 73 re-admitted patients at the time of follow-up.

Several subgroups of patients were identified as having a significantly higher risk of mortality. These included older patients and patients who had had higher Apache 2 scores on both first admission and re-admission. Mortality was also increased in patients who were re-admitted in the first 24 h after discharge (all of the 10 patients re-admitted within 24 h of discharge were dead at follow-up) and had three organ failures (respiratory, cardiovascular, and renal). This risk of mortality includes ICU mortality, mortality during that hospital stay, and mortality in the intervening 2–3 yr. The odds ratios are presented in Table 2.


View this table:
[in this window]
[in a new window]

  		Table 2 Risk factors during re-admission for death at a median of 2.5 yr after re-admission. For continuous variables (age and Apache scores), the odds ratio is for each increment of 1 of that variable. Odds ratios for two and three organs supported are compared with less than two and three organs, respectively. Data are derived from the 73 re-admitted patients

 
With regard to favourable functional outcome at the time of follow-up, the only independent predictors identified were lower Apache 2 scores on first ICU admission and on re-admission. Age, sex, specialty category, timing of re-admission, length of stay, number of re-admissions, or number of organs supported did not predict favourable functional outcome at follow-up (see Table 3).


View this table:
[in this window]
[in a new window]

  		Table 3 Predictive factors during re-admission for good functional outcome (Glasgow Outcome Score of 4–5 or Karnofsky index of 70–100) at a median of 2.5 yr after re-admission. For continuous variables (age and Apache scores), the odds ratio is for each increment of 1 of that variable. Odds ratios for two and three organs supported are compared with less than two and organs, respectively. There are no odds ratios or 95% CIs for immediate re-admission as none of these patients were alive at follow-up. Data are derived from the 73 re-admitted patients. There are no figures for ‘immediate’ timing of re-admission as all of these patients were dead at follow-up

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
References
 
We found that 7.4% of the patients who survived their first ICU stay were re-admitted during the study period. This re-admission rate is comparable with international standards.1 Univariate analysis showed that older patients, patients with a higher Apache 2 score on first ICU admission, and general and vascular surgical patients had an increased risk of being re-admitted. There was a higher risk of mortality in older patients, those with higher Apache 2 scores on first and subsequent admissions, those re-admitted within 24 h of discharge, and those with three organ failures (respiratory, cardiovascular, and renal). Lower Apache 2 scores on first ICU admission and on re-admission were the only identified predictors of favourable functional outcome at follow-up. The fact that certain patient variables, such as length of ICU stay or number of re-admissions, were not identified as predictors of favourable outcome in this cohort is equally valuable information in clinical decision-making.

Thirty-two (43.8%) of the 73 re-admitted patients were alive 2–3 yr after discharge. The median GOS of the survivors was 4 [4–5] and their median Karnofsky index was 70 [70–90] at follow-up. A GOS of 4 indicates moderate disability, describing a patient who is able to function independently but not return to work, whereas a Karnofsky index of 70 characterizes one who can self-care but is unable to carry out normal activity or do active work. A patient with a Karnofsky index of 80 can return to work and normal activity with some effort and some symptoms of disease. Thus, although less than half of the patients were alive at 2–3 yr, those who survived had made a reasonably good functional recovery.

These functional outcome results are comparable with the published data from studies that describe long-term functional outcome among ICU survivors who were not necessarily re-admitted. In a study of ICU patients, Mahul and colleagues23 found that the subgroup of 295 patients aged >70 yr had a 1 yr cumulative mortality of 49%.23 Of the survivors, 88% returned home and 72% had relatively good functional status allowing an independent life. In a follow-up study of patients who spent 7 days or more in a surgical ICU, the 1 yr survival rate was 45.3% whereas Sickness Impact Profile score was lower than baseline at 1 yr.24 As in our study, these studies show that while mortality rate and resource utilization is significant, functional outcome is consistent with a good quality of life.

Pulmonary infection was the reason for re-admission in 38% of our patients. This warrants some consideration. Additional resources in the form of increased intermediate care facilities, intensive care outreach teams,25 tracheostomy care teams, and increased physiotherapy and speech therapy input outside of the ICU could all potentially facilitate early identification of these failed discharges and prompt re-admission with improved outcomes.26 An attendant reduction in re-admissions might reasonably be anticipated.

While these are the results of a single centre study, our basic patient characteristics, re-admission rates, and ICU and hospital mortality rates are consistent with other authors, with the implication that our functional outcome data should also be consistent. We used the GOS and Karnofsky index as a measure of functional outcome. Black and colleagues,12 in a systematic review of the adult ICU literature, found 161 outcome measures described, of which only 38 had more than one reported use. Of these 38, they recommended only 11 should be used in further studies. These included the GOS that was recommended as a measure of recovery and the Karnofsky index as a measure of physical function. Ideally, we would have also measured health-related quality of life using one of the recognized scoring systems such as Sickness Impact Profile or Short Form-36, but these were not suitable for the structured telephone interview setting. They also deal with matters beyond the scope of ICU care (e.g. community care, follow-up, family support, and rehabilitation) and are probably unlikely to impact on ICU decision-making. It would also be difficult to get 100% follow-up with such a detailed measurement, which we achieved in our study.

There were other limitations to our study. Pre-existing diseases and co-morbidities are important in determining both ICU survival and longer term functional outcome. Our database recorded only those included in the Apache 2 scoring system.

Our case-mix was influenced by the fact that we are a centre for cardiothoracic and heart transplantation, and spinal injuries. We also have no data on the patients who were discharged alive from ICU and not subsequently considered for re-admission, whether by patient choice or for presumed medical futility.

In conclusion, ICU re-admission commonly represents a major, distressing setback in a patient’s illness or recovery. Our data show that although the ICU, hospital, and subsequent mortalities are high in patients after ICU re-admission, most survivors at 2–3 yr had by then made a good functional recovery and were independent. This may provide some evidence to inform discussions with patients, their relatives, and advocates at the time of ICU re-admission about their long-term health and outcome.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Rosenberg AL, Watts C. Patients readmitted to ICUs: a systematic review of risk factors and outcomes. Chest (2000) 118:492–502.[CrossRef][Web of Science][Medline]

2 Rosenberg AL, Hofer TP, Hayward RA, Strachan C, Watts AM. Who bounces back? Physiologic and other predictors of intensive care unit readmission. Crit Care Med (2001) 29:511–8.[CrossRef][Web of Science][Medline]

3 Peruzzi W. Translating outcomes research into fiscal responsibility and knowledge management in health care. Crit Care Med (2001) 29:679–80.[CrossRef][Web of Science][Medline]

4 Mackle IJ, Swan DG, Cook B. One year outcome of intensive care patients with decompensated alcoholic liver disease. Br J Anaesth (2006) 97:496–8.[Abstract/Free Full Text]

5 Jennett B, Bond M. Assessment of outcome after severe brain damage. A practical scale. Lancet (1975) 1:480–4.[Web of Science][Medline]

6 Sandroni C, Barelli A, Piazza O, Proietti R, Mastria D, Boninsegna R. What is the best test to predict outcome after prolonged cardiac arrest? Eur J Emerg Med (1995) 2:33–7.[CrossRef][Medline]

7 Logi F, Fischer C, Murri L, Mauguierre F. The prognostic value of evoked responses from primary somatosensory and auditory cortex in comatose patients. Clin Neurophysiol (2003) 114:1615–27.[CrossRef][Web of Science][Medline]

8 Kastenbauer S, Pfister HW. Pneumococcal meningitis in adults: spectrum of complications and prognostic factors in a series of 87 cases. Brain (2003) 126:1015–25.[Abstract/Free Full Text]

9 Gaudino M, Martinelli L, DiLella G, Marano P, Schiavello R, Possati G. Superior extension of intraoperative brain damage in case of normothermic systemic perfusion during coronary artery bypass operations. J Thorac Cardiovasc Surg (1999) 118:432–7.[Abstract/Free Full Text]

10 Anderson SI, Housley AM, Jones PA, Slattery J, Miller JD. Glasgow outcome score: an inter-rater reliability study. Brain Inj (1993) 7:309–17.[Medline]

11 Yinnon A, Zimran A, Hershko C. Quality of life and survival following intensive medical care. Q J Med (1989) 71:347–57.[Web of Science][Medline]

12 Black N, Jenkinson C, Hayes J, et al. Review of outcome measures used in adult critical care. Crit Care Med (2001) 29:2119–24.[CrossRef][Web of Science][Medline]

13 Glower DD, Christopher TD, Milano CA, et al. Performance status and outcome after coronary artery bypass grafting in persons aged 80 to 93 years. Am J Cardiol (1992) 70:567–71.[CrossRef][Web of Science][Medline]

14 Khan JH, McElhinney DB, Hall TS, Merrick SH. Cardiac valve surgery in octogenarians: improving quality of life and functional status. Arch Surg (1998) 133:887–93.[Abstract/Free Full Text]

15 O’Brien BP, Butt W, Suhr H, et al. The functional outcome of patients requiring over 28 days of intensive care: a long-term follow-up study. Crit Care Resusc (2006) 8:200–4.[Medline]

16 Gersbach P, Tevaearai H, Revelly JP, Bize P, Chiolero R, Von Segesser LK. Are there accurate predictors of long-term vital and functional outcomes in cardiac surgical patients requiring prolonged intensive care? Eur J Cardiothoracic Surg (2006) 29:466–72.[Abstract/Free Full Text]

17 Chen LM, Martin CM, Keenan SP, Sibbald WJ. Patients readmitted to the intensive care unit during the same hospitalization: clinical features and outcomes. Crit Care Med (1998) 26:1834–41.[Web of Science][Medline]

18 Montuclard L, Garrouste-Orgeas, Timsit JF, Misset B, De Jonghe B, Carlet J. Outcome, functional autonomy, and quality of life of elderly patients with a long-term intensive care unit stay. Crit Care Med (2000) 28:3389–95.[CrossRef][Web of Science][Medline]

19 Lawton MT, Du R. Effect of the neurosurgeon’s surgical experience on outcomes from intraoperative aneurysmal rupture. Neurosurgery (2005) 57:9–15.[CrossRef][Web of Science][Medline]

20 Andrews PJ, Sleeman DH, Statham PF, et al. Predicting recovery in patients suffering from traumatic brain injury by using admission variables and physiological data: a comparison between decision tree analysis and logisitic regression. J Neurosurg (2002) 97:326–36.[Web of Science][Medline]

21 Boussat S, El’rini T, Dubiez A, Depierre A, Barale F, Capellier G. Predictive factors of death in primary lung cancer patients on admission to the intensive care unit. Intensive Care Med (2000) 26:1811–6.[CrossRef][Web of Science][Medline]

22 Martinez-Moragon E, Aparicio J, Sanchis J, Menendez R, Cruz Rogado M, Sanchis F. Malignant pleural effusion: prognostic factors for survival and response to chemical pleurodesis in a series of 120 cases. Respiration (1998) 65:108–13.[CrossRef][Web of Science][Medline]

23 Mahul Ph, Petrot D, Tempelhoff G, et al. Short and long-term prognosis, functional outcome following ICU for elderly. Intensive Care Med (1991) 17:7–10.[CrossRef][Web of Science][Medline]

24 Lipsett P, Swoboda S, Dickerson J, et al. Survival and functional outcome after prolonged intensive care unit stay. Ann Surg (2000) 231:262–8.[CrossRef][Web of Science][Medline]

25 Goldhill DR. Preventing surgical deaths: critical care and intensive care outreach services in the postoperative period. Br J Anaesth (2005) 95:88–94.[Free Full Text]

26 Besserman E, Teres D, Logan A, et al. Use of flexible intermediate and intensive care to reduce multiple transfers of patients. Am J Crit Care (1999) 8:170–9.[Abstract]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
100/2/219    most recent
aem372v1
Right arrow E-Letters: Submit a response to the article
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Conlon, N.
Right arrow Articles by Marsh, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Conlon, N.
Right arrow Articles by Marsh, B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?