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 randomized controlled study of prophylactic ranitidine in preventing stress-induced gastric mucosal lesions in neonatal intensive care unit patients.

Kuusela AL, Ruuska T, Karikoski R, Laippala P, Ikonen S, Janas M, Maki M.

Crit Care Med 1997; 25:346-351. [abstract]

Reviewed by Girish Deshpande, MD, Kathleen Meert, MD, Children's Hospital of Michigan

Review posted June 12, 1998

---

I. What is being studied?:

The study objective:

To assess endoscopically the effect of prophylactic short-term ranitidine in the prevention of stress-induced gastric mucosal lesions in neonatal intensive care unit (NICU) patients.

The study design:

Prospective, randomized and controlled study.

The patients included:

All preterm and full term infants in whom mechanical ventilation was initiated within the first 2 hours of life were eligible for inclusion.

The patients excluded:

All other infants treated in the NICU were excluded. No specific exclusion criteria were used.

The interventions compared:

The treatment group consisted of 23 patients who received IV ranitidine 5mg/kg/day in three divided doses for 4 days. The control group consisted of 25 patients who did not receive ranitidine. No placebo was used.

The outcomes evaluated:

1. The visual description of the gastric mucosa- evaluated by one gastroenterologist using a gastroscope.
2. Histopathologic appearance of gastric mucosal biopsy specimen- evaluated by one pathologist.
3. Bacterial cultures were also obtained from the biopsy specimens.
4. After discharge, the patient's medical records were reviewed for GI problems including bleeding, vomiting, delayed emptying and proven or suspected bacteremia during the first 4 weeks of life.
5. Mortality.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. The randomization was performed with stratification for gestational age (< or > 33 wks) using a block design. The randomization procedure was accomplished using sealed envelopes.

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

Was followup complete?

No. 53 patients were randomized. 48 completed the study and were included in the results. Five neonates dropped out of the study after randomization but before gastroscopic evaluation. These 5 were not included in the results: 3 due to early death (2 patients < 33 weeks and 1 patient > 33 weeks) and 2 which were diagnosed with esophageal atresia (one patient in each group). These five drop-out patients could have been included in the mortality data. The authors do state that assuming that the three dropout neonates randomized to the treatment group had mucosal lesions and that the two dropout neonates randomized to the control group had normal looking mucosa, the results still show a beneficial effect of ranitidine prophylaxis.

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

Yes. There were no crossovers between the treatment groups.

Secondary questions:

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

No. The authors state that no placebo treatment was available. However an IV placebo could have been prepared and incorporated into this protocol. Because no placebo treatment was used, the nurse caring for the patient knew which group the patient was randomized to. However, the attending physician, endoscopist and pathologist were reportedly unaware of the study group. This is important because the results of the study are dependent on the endoscopy findings and the interpretation of biopsy specimens.

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

Patient characteristics such as birth weight, gestational age, Apgar score, cord blood pH and duration of intubation were similar in both the treatment and control groups.

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

Yes. The data describing the number of infants receiving other therapies (e.g., surfactant treatment for RDS, indomethacin for closure of PDA, dopamine infusion for hypotension, red cell transfusion) were similar in both groups.

III. What were the results?

1. How large was the treatment effect?

a) Visual appearance of mucosa: Of 23 infants in the treatment group, 14 had normal appearing mucosa and 9 had abnormal mucosa. Of 25 infants in the control group, 5 had normal and 20 had abnormal mucosa. Using the 2 x 2 table given below, we calculated a relative risk of 0.48. Relative risk between 0 and 1 means a beneficial effect of the treatment.

	Visual Appearance of Mucosa
	Abnormal	Normal	Totals
Ranitidine	9	14	23
Control	20	5	25
Totals	29	19	48

Using this data we also calculated an absolute risk reduction (ARR) of 0.41 [the proportion with abnormal mucosa in the control group (X) minus the proportion with abnormal mucosa in the ranitidine group (Y); 0.80 - 0.39]. The relative risk reduction (RRR) is 52% [RRR = (1 - (Y/X)) x 100%]. RRR of 52% means that the ranitidine treatment reduced the risk of mucosal lesions by 52% relative to that occurring among the control patients. The number needed to treat (NNT = 1/ARR) is 2.4. This represents the number of patients we would need to treat to prevent mucosal lesions in one patient.

b) Histologic findings: Of 23 infants in the treatment group, 13 had normal histology and 10 had abnormal histologic findings. Of 25 infants in the control group, 4 had normal and 21 had abnormal histology. By using a similar analysis as above, we calculated a relative risk of 0.51, an ARR of 0.42, a RRR of 50% and a NNT of 2.4 for histologic findings.

c) GI problems at 4 week follow-up: Bleeding, vomiting and delayed gastic emptying occurred less often in treated infants than controls at the 4 week follow-up examination (3/23 treated vs 10/25 controls, p = .036). We calculated a relative risk of 0.33, an ARR of 0.27, a RRR of 67.5% and a NNT of 3.7.

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

An estimate of the treatment effect is reflected in the confidence intervals for RR, ARR, RRR and NNT.

Visual Appearance of Mucosa

	95% Confidence Intervals
Relative Risk	28 - 82%
Absolute Risk Reduction	16 - 66%
Relative Risk Reduction	18 - 72%
Number Needed to Treat	1.5 - 6.3

Histologic Findings

	95% Confidence Intervals
Relative Risk	31 - 83%
Absolute Risk Reduction	18 - 66%
Relative Risk Reduction	17 - 69%
Number Needed to Treat	1.5 - 5.5

GI Problems at 4 Week Follow-up

	95% Confidence Intervals
Relative Risk	10% -104%
Absolute Risk Reduction	4% - 51%
Relative Risk Reduction	0 - 90%
Number Needed to Treat	2 - 25

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

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

Yes. The difficulty in applying the results of this study to our patients is in the uncertainty of the incidence of endoscopically visualized mucosal lesions in pediatric ICU patients. The definition of gastric bleeding varies between studies. Few studies have examined the incidence of gastric mucosal lesions by endoscopy and histology. Eddleston et al. (1) studied 60 adult ICU patients who were mechanically ventilated. On admission the frequency rate of erosions/ulcerations (assessed with an endoscope) was 13.5%. After 4 days this rate increased to 18% in sucralfate treated adults and 36% in ranitidine treated adults. Although untreated patients were not included in this study, the data reveals that the incidence of gastric mucosal lesions is high in adults, even when treated. Lopez-Herce et al. (2) found an incidence of slight upper GI hemorrhage in 76.4% of pediatric ICU patients and important upper GI hemorrhage in 9.3%. In this study, the untreated control group had a 20% occurrence rate of important upper GI hemorrhage. Because upper GI hemorrhage is a documented problem in ICU patients of all ages, we feel that the results of the Kuusela's study are applicable to our patients. Furthermore, the basic gastric acid output in preterm infants is lower than in term infants.

2. Were all clinically important outcomes considered?

Yes. The study demonstrates the efficacy of ranitidine prophylaxis in preventing gastric mucosal lesions in neonates, as well as preventing bleeding, vomiting and delayed gastric emptying within 4 weeks of age. The decrease in the rate of clinical GI problems lends weight to the endoscopic and histologic findings. The method for studying gastric emptying is not stated in the article. Long term effects of ranitidine, if any, were not evaluated.

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

Yes. In order to evaluate potential harms the investigators collected data on liver function tests (data not shown), gastric mucosal bacterial colonization and mortality. Liver function tests and incidence of bacterial colonization were not different between the treatment and control groups. There were seven deaths in total, 5 in the treatment group and 2 in the control group (p=0.175). The deaths were probably unrelated to the use of ranitidine. Although the authors did not evaluate the cost of prophylactic ranitidine, since the NNT was only 2.4 we expect that its use is cost-effective.

References:

1. Eddleston JM, Vohra A, Scott P, et al: A comparison of the frequency of stress ulceration and secondary pneumonia in sucralfate- or ranitidine-treated intensive care unit patients. Crit Care Med 1991; 19:1491-1496.
2. Lopez-Herce J, Dorao P, Elola P, et al: Frequency and prophylaxis of upper gastrointestinal hemorrhage in critically ill children: A prospective study comparing the efficacy of amalgate, ranitidine, and sucralfate. Crit Care Med 1992; 20:1082-1089.

Comments

Back to the EB Journal Club Index