Web site designed and maintained by Chris Cairns  © SICS EBM Group 2004                                  

Up

 

What effect does obesity have on intensive care morbidity and mortality?

Text Box: Bottom Line: Obesity is not demonstrated to be associated with a higher ICU mortality. Obese patients have an increased morbidity on the ICU as evidenced by prolonged requirement for mechanical ventilation and intensive care admission length. Level of Evidence: 1+ (Meta-analysis with a low risk of bias.) + (Meta-analysis with a low risk of bias.)

 

 

 

 

 

 

 

 

 

Citation/s: Akinnusi ME, Pineda LA, El Soth AA. Effect of obesity on intensive care morbidity and mortality: A meta-analysis. Crit Care Med 2008; 36: 151-158.


Lead author: Dr Ali A. El Solh. Division of Pulmonary, Critical Care and Sleep Medicine, Department of Medicine, Western New York Respiratory Research Center, University of Buffalo School of Medicine and Biomedical Sciences, Buffalo, NY solh@buffalo.edu

 

Three-part Clinical Question:

Patient: Obese patients, medical and surgical, in the intensive care unit (BMI≥30kg/m2). Treatment: Standard intensive care management.

Outcomes: Primary - ICU mortality; Secondary - Duration of mechanical ventilation. Length of ICU admission.


Search Terms: intensive care, critically ill, obesity, body mass index, mortality.

Search date 10th March 2008: Medline (1996-current) 41 hits, Embrase (1996-current) 45 hits, CINAHL (1982-current) 28 hits.

 

The Review:
Data Sources: Cochrane Library, Medline, Embase, non-English sources, BIOSIS Previews, Pubmed, contact with expert informants.
Study Selection:

Inclusion: Studies comparing obese and non-obese critically ill patients admitted to the ICU. Obesity defined as BMI ≥30kg/m2. Severity of illness scores (APACHE II, SAPS II, ISS) were comparable for both groups in all selected studies.

Exclusion: Non-comparative studies. Those performed outside the ICU setting or utilising the same patient population as another included study.
Data Extraction:

29 studies initially identified (15 identified potential studies from the available literature were excluded on eligibility criteria established a priori).

Total number of obese and non-obese patients = 62,045 (25% [15,347] of which were obese).

Mortality rate (14 considered for analysis [7 retrospective, 7 prospective])

ICU length of stay (13 studies).

Duration of mechanical ventilation (6 studies).
2 reviewers independently rated studies for inclusion. Studies were tested for heterogeneity.

 

The Evidence:

 

 Primary Outcome:

 

Patients (Obese vs Non-Obese)

RR (95% CIs) ICU Mortality

P Value

BMI ≥ 30

1.0 (0.86-1.16)

0.97

BMI 30-39.9

0.86 (0.81- 0.91)

<0.001

BMI ≥40

0.97 (0.74-1.26)

0.8

 Secondary outcomes:           

Outcome

BMI ≥30 vs non-obese

Mean difference in days

(95% CI)

P Value

Duration of ventilation

1.48 (0.07-2.89)

0.04

Duration of ICU stay

1.08 (0.27-1.88)

0.009

 

 EBM questions:

 

1) Do the methods allow the adequate testing of the hypothesis? Yes. The association of chronic diseases such as HT, IHD, DM and OSA with obesity intuitively would seem to potentially predispose ICU patients to a higher mortality. This meta-analysis demonstrates no difference in the primary outcome of mortality between obese and non-obese patients. The individual studies contributing to the meta-analysis varied in their mortality findings. The more recent studies appear to show a trend to improving outcome in obese patients the authors postulate that this may reflect improvements in ICU management, especially closer consideration of glycaemic control. 

2) Do the statistical tests correctly test the results to allow differentiation of statistically significant result? Yes. 

3) Are conclusions valid in light of results? Yes. Heterogeneity of the studies must be considered, but it represents a valid attempted pooling and analysis of the best available current evidence. The authors discuss potential mechanisms of a counterintuitive survival advantage, shown in the subgroup analysis of obese patients (BMI 30-39.9) compared to non-obese (RR 0.86) or the morbidly obese. They suggest immunomodulation by adipose secreting hormones (leptin and IL-10) may have an influence. It was also postulated that inclusion of poorly nourished chronically ill patients may adversely influence mortality in the non-obese group. 

4) Did results get omitted, and why? No.  

5) Did they suggest areas of further research? Yes. Studies of interventions aimed at the obese population to reduce their greater resource utilisation in the ICU setting.  

6) Did they make any further recommendations based on the results and were they appropriate? No. 

7) Is the study relevant to my clinical practice ?Yes. Incidence of obesity in the UK and globally is increasing. The increase morbidity (length of stay and ventilation days on ICU) has resource implications and potential future increasing costs for the NHS.  

8) What level of evidence does this study represent? 1+ (meta-analysis with a low risk of bias). There is potential risk of publication bias, with reduced publication rates of negative studies and selection bias (7 of analysed studies for ICU mortality were retrospective). 

9) What grade of recommendation can I make on this result alone? A. 

10) What grade of recommendation can I make when this study is considered along with other available evidence? A. 

11) Should I change my practice because of the results? No. 

12) Should I audit my current practice because of these results? No. 

Appraised by: Dr James Brown

c/o Anaesthetics Department, Box 93

Addenbrooke’s Hospital,

Cambridge University Hospital NHS Foundation Trust,

Hills Road, Cambridge. CB2 2QQ; 08 March 2008


Email: jprb_brum@yahoo.com

 

©SICS EBMG 2009

 

Printer friendly version