Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

A Comparison of Sucralfate and Ranitidine for the Prevention of Upper Gastrointestinal Bleeding in Patients Requiring Mechanical Ventilation

Cook D, Guyatt G, Marshall J, et al.

N Engl J Med 1998; 338: 791-797. [abstract]

Reviewed by Scot T. Bateman, M.D., Children's Hospital, Boston

Review posted August 6, 1998

---

I. What is being studied?:

The study objective:

To assess the efficacy of prophylaxis against gastrointestinal bleeding with either ranitidine or sulcralfate and to evaluate any difference in associated ventilator induced pneumonia and mortality.

The study design:

Randomized, placebo-controlled, double-blind. 16 study sites participated.

The patients included:

Consecutive adult ICU admissions who were projected to require mechanical ventilation for at least 48 hours.

The patients excluded:

Diagnosis of gastrointestinal bleeding or pneumonia on admission, gastrectomy, a prognosis considered hopeless, or receipt of two or more previous dosed of open-label prophylactic therapy.

The interventions compared:

Intravenous ranitidine vs. nasogastric sulcralfate

The outcomes evaluated:

1. Clinically important gastrointestinal bleeding: defined as overt bleeding (hematemesis, nasogastric aspirate containing blood or coffee-ground material, melena, or hematochezia) plus one of the following features: decrease in systolic or diastolic blood pressure of 20 mm Hg within 24 hrs; heart rate increase of 20 and systolic BP decrease of 10 mm Hg in upright position; decrease in Hgb of 2 g/dl in 24hr; or failure of Hgb to increase sufficiently after transfusion.
2. Ventilator-associated pneumonia: defined as a new radiographic infiltrate present for at least 48 hrs, plus at least two of the following: temperature >38.5 or <35.0, WBC >10K, purulent sputum, or isolation of pathogenic bacteria from an endotracheal aspirate.
3. Length of ICU stay.
4. Mortality.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. The main study center randomized patients to study groups in blocks of six from a computer generated random-number table.

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Was followup complete?

Yes. All 1200 patients enrolled in the study were included in the analysis.

Were patients analyzed in the groups to which they were randomized?

Yes. No cross-overs occurred.

Secondary questions:

3. Were patients, health workers, and study personnel "blind" to treatment?

Yes. It was a double-blind study design. The patients received placebo IV solution in place of ranitidine or oral/ng placebo that looked like sulcralfate. Clinicians did not monitor gastric pH by report (by not checking pH, the caregivers were blind, but it is unclear whether the ranitidine was used properly, because the dose should be adjusted if the pH is too low).

4. Were the groups similar at the start of the trial?

Yes. The demographic and baseline physiologic characteristics were similar in the two groups as presented in Table 1. There is no mention in the baseline characteristics as to how many patients had underlying coagulopathy, which would have been very helpful because as it is a significant risk factor independently.

5. Aside from the experimental intervention, were the groups treated equally?

Unclear. They reported an equal percentage of patients who missed doses, who received other stress ulcer prophylaxis (2.3 % vs 2.6%), and who received enteral feeds (70.3% vs. 71.8%).

III. What were the results?

1. How large was the treatment effect?

Gastrointestinal bleeding: A statistically significant difference in the rates of GI bleeding was noted: 1.7% in the ranitidine group vs. a 3.8% in the sulcralfate group (p=0.02). This represents an absolute risk reduction (ARR) of 2.1% (relative risk reduction or RRR of 65%). The reported number needed to treat (NNT) to prevent one bleeding episode using ranitidine compared to sulcralfate was 48.

Ventilator-associated pneumonia: The study was designed to provide statistical power to detect a difference in rates of pneumonia. The assumption was a 25% incidence of pneumonia and the study size (1200 patients) was designed to detect a 25% reduction if sulcralfate or ranitidine was better. The actual rate of pneumonia was not significantly different between the groups; 19.1% in the ranitidine group vs. 16.2% in the sulcralfate group. This results in an ARR of 2.9% of sulcralfate over ranitidine (RRR 15%) (p=0.19).

Mortality: The rates were not different; 23.5% (ranitidine) vs. 22.8% (sulcralfate).

ICU stay: Median length of stay was 9 days in both groups and duration of intubation was 7 days in each group.

2. How precise was the estimate of the treatment effect?

Gastrointestinal bleeding: The ARR of bleeding was as little as 0.3% or as high as 3.97% comparing ranitidine to sulcralfate, i.e., the 95% confidence intervals for the ARR of 2.1% were 0.3 and 3.97.

Ventilator-associated pneumonia: The ARR of pneumonia when using sulcralfate vs. ranitidine was between a 1.4% higher incidence to a 7.2% lower incidence.

IV. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

No pediatric patients were included in the analysis. However, these results are important due to the higher reported incidence of bleeding in studies of infants and children. The incidence of bleeding has been reported to be 9.3% in treated vs. 20% in untreated pediatric patients (1), notably higher than the incidence of 1.7% vs.3.8% that is reported in the current study. This higher baseline risk implies pediatric patients have potentially "more to gain" by effective prophylaxis (assuming the relative benefits would be similar in infants and children). Neonatal data suggest even higher levels of abnormal gastric mucosa in untreated patients (2).

2. Were all clinically important outcomes considered?

Yes. The outcomes evaluated comprise the most important variables that prophylaxis is trying to prevent. Other outcomes, which may have been important, include:

1. The ease/difficulty of administration of sulcralfate may be clinically important. Sulcralfate is a rather thick slurry which can, at times, block the nasogastric tube and make it necessary to change it. Replacing nasogastric tubes more frequently may alter pneumonia rates.
2. Also, as mentioned previously, by not checking ranitidine's effect on gastric pH, the drug's effect may not have been optimized. If the pH had been properly raised consistently, the problems of ventilator-associated pneumonias may have become more apparent (presuming that the lack of gastric acidity and bacterial "reflux" is a contributing factor in the pathogenesis of this complication).
3. The timing of the bleeding episodes would also have been relevant. For instance, it would be helpful to know if the bleeding episode occurred before or after the onset of enteral feedings.
4. In addition, there is a reluctance to use ranitidine in immunosuppressed and/or neutropenic patients because of its effect on bone marrow suppression; a sub-group analysis of this patient population might be helpful.

3. Are the likely treatment benefits worth the potential harms and costs?

Yes. Although there was no effect on truly important outcomes (length of ICU stay, ventilator days, or mortality), the use of ranitidine to decrease the amount of GI bleeding without increasing the incidence of ventilator-associated pneumonias makes its use beneficial. The ARR of 2% for GI bleeding is not a large benefit and 48 patients needing to treat with ranitidine to prevent one GI bleeding episode is rather many. However, given the apparently low risks and costs of prophylaxis with ranitidine, these numbers are probably convincing enough.

References

1. Lopez-Herce J, et al. Crit Care Med 1992; 20: 1082-1089. [abstract]

2. Kuusela AL et. al, Crit Care Med 1997; 25: 346-351. [abstract]

Comments

Back to the EB Journal Club Index