The attributable mortality and length of intensive care unit stay of clinically important gastrointestinal bleeding in critically ill patients
This review may be edited
Posted By: Gordon S. Doig
E-Mail: gdoig@med.usyd.edu.au
Posted Date: 4 July 2002
Title: The attributable mortality and length of intensive care unit stay of clinically important gastrointestinal bleeding in critically ill patients
Authors: Deborah J. Cook, Lauren E. Griffith, Stephen D. Walter, Gordon H. Guyatt, Maureen O. Meade, Daren K. Heyland, Ann Kirby, Michael Tryba, and for the Canadian
Reference: Crit Care. 2001; 5 (6): 368-375
Link: Click here for a direct link to the paper. A password may be required for access to fulltext.
Abstract: Objective: To estimate the mortality and length of stay in the intensive care unit (ICU) attributable to clinically important gastrointestinal bleeding in mechanically ventilated critically ill patients.
Design: Three strategies were used to estimate the mortality attributable to bleeding in two multicentre databases. The first method matched patients who bled with those who did not (matched cohort), using duration of ICU stay prior to the bleed, each of six domains of the Multiple Organ Dysfunction Score (MODS) measured 3 days prior to the bleed, APACHE II score, age, admitting diagnosis, and duration of mechanical ventilation. The second approach employed Cox proportional hazards regression to match bleeding and non-bleeding patients (model-based matched cohort). The third method, instead of matching, derived estimates based on regression modelling using the entire population (regression method). Three parallel analyses were conducted for the length of ICU stay attributable to clinically important bleeding.
Setting: Sixteen Canadian university-affiliated ICUs.
Patients: A total of 1666 critically ill patients receiving mechanical ventilation for at least 48 hours.
Measurements: We prospectively collected data on patient demographics, APACHE II score, admitting diagnosis, daily MODS, clinically important bleeding, length of ICU stay, and mortality. Independent adjudicators determined the occurrence of clinically important gastrointestinal bleeding, defined as overt bleeding in association with haemodynamic compromise or blood transfusion.
Results Of 1666 patients, 59 developed clinically important gastrointestinal bleeding. The mean APACHE II score was 22.9 8.6 among bleeding patients and 23.3 7.7 among non-bleeding patients. The risk of death was increased in patients with bleeding using all three analytic approaches (matched cohort method: relative risk [RR]= 2.9, 95% confidence interval (CI)= 1.6-5.5; model-based matched cohort method: RR = 1.8, 95% CI = 1.1 - 2.9; and the regression method: RR = 4.1, 95% CI = 2.6-6.5). However, this was not significant for the adjusted regression method (RR = 1.0, 95% CI = 0.6-1.7). The median length of ICU stay attributable to clinically important bleeding for these three methods, respectively, was 3.8 days (95% CI = -0.01 to 7.6 days), 6.7 days (95% CI = 2.7 - 10.7 days), and 7.9 days (95% CI = 1.4 - 14.4 days).
Conclusions Clinically important upper gastrointestinal bleeding has an important attributable morbidity and mortality, associated with a RR of death of 1 - 4 and an excess length of ICU stay of approximately 4 - 8 days.
Are the Results Valid?
1. Were there clearly identified comparison groups that were similar with respect to important determinants of outcome, other than the one of interest?
Yes. The investigators used three different methods to help make sure comparision groups were as similar as possible: 1) Matched cohort study: patients with GI bleeding were matched to controls based on known risk factors, 2) Model-based matched cohort: regression is performed on all patients in database to determine study-specific risk factors for outcome, patients with GI bleeds are matched to controls using risk factors identified in regression analysis and 3) Regression with No matching. All available controls (pts w/o GI bleeds) are included in a regression model.
2. Were the outcomes and exposures measured in the same way in the groups being compared?
Yes. Two seperate data sources were used for this study. Both sources (data from an RCT and data from a prospective observational study) were collected prospectively with the a priori specified intent of diagnosing GI bleeds, and eventual patient outcomes.
3. Was follow-up sufficiently long and complete?
It appears that hospital discharge information was available on all patients however, the authors report that their regression model censored patients at day 28 post-ICU admission. There is no justification for this censoring provided by the authors. If outcomes post-day 28 were available, they should have been used.
4. Is the temporal relationship correct?
Probably. Patients with pre-existing bleeds were excluded, along with patients who had bleeds within 48 hours of ICU admission. This would suggest that GI bleeds were new onset and thus preceeded increases in LoS and mortality.
5. Is there a dose response gradient?
No. The authors do not quatify degree of clinically important bleeds (Ex based on BP drop, Hb drop or blood products required) and do not look for a dose response gradient.
What are the Results?
1. How strong is the association between exposure and outcome?
Relative risk of death was increased for patients with bleeds in all methods:
1) Matched cohort study: RR=2.9
2) Model-based matched cohort: RR = 1.8
3) Regression: RR = 4.1 uncontrolled
3) Regression: RR = 1.0 controlling for confounders
Length of stay was also increased in patients with bleeds by:
1) Matched cohort study: 3.8 days
2) Model-based matched cohort: 6.7 days
3) Regression: 7.9 days uncontrolled
2. How precise is the estimate of risk?
The overall estimate of increased relative risk of death ranges from 1.1 to 6.5 and the overall estimate of increased length of stay ranges from -0.01 to 14.4 days.
Will the Results Help Me In Caring For My Patients?
1. Are the results applicable to my practice?
The patients studied in this paper are representative of the average patient seen in a Canadian tertiary care ICU.
2. What is the magnitude of the risk?
The authors report an absolute increase in risk of mortality of 24% attributable to GI bleeds (mortality in GI bleed pts 45.8% - mortality in controls 20.9%). This suggests that for every 4.1 GI bleeds prevented, one additional life is saved.
According to Cook et al, 47 ventilated patients need to be treated with ranitidine to prevent 1 additional GI bleed. This converts to approximately 193 patients treated with ranitidine to prevent one additional death.
3. Should I attempt to stop the exposure?
A large randomized controlled trial demonstrates that ranitidine is effective at preventing clinically important GI bleeds. Clinically important GI bleeds are probably associated with increased LoS and increased risk of mortality. At the very least, clinically important bleeds increase blood product use and overall costs. The benefits of prophylactic treatment with ranitidine probably outweigh any possible costs or harms.
An evidence-based recommendation, which is the building block for an evidence-based guideline, that summarizes this topic can be found in the EBR section.

What other people had to say about:
The attributable mortality and length of intensive care unit stay of clinically important gastrointestinal bleeding in critically ill patients

Add Your Comment
[ Back ]

Any questions or comments please contact Gordon S. Doig
Implemented and designed by John Soer and Gordon Doig
Page last modified on Friday August 10, 2001