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

Ventilator-associated pneumonia and upper airway colonisation with Gram negative bacilli: the role of stress ulcer prophylaxis in children.

Lopriore E, Markhorst DG, Gemke RJBJ.

Intensive Care Med 2002; 28: 763-767. [abstract]

Reviewed by Diane Godorov, DO, Children's Healthcare of Atlanta at Egleston

Review posted March 24, 2003

---

I. What is being studied?:

The study objective:

To assess the risk of ventilator-associated pneumonia (VAP) and the incidence of upper airway colonization related to the use of stress ulcer prophylaxis in critically ill children

The study design:

Retrospective cohort study

The patients included:

Patients were included if they were ventilated for more than 48 hours and had a nasogastric tube in place.

The patients excluded:

Patients were excluded if they were diagnosed with pneumonia or ARDS upon admission or within 24 hours of admission; had received H2 blockers, sucralfate, or omeprazole within one week of admission; were treated with omeprazole or had acute renal failure and therefore could not be treated with sucralfate due to its potential renal side effects. This left 155 of 214 patients included in the evaluation.

The interventions compared:

Three groups of patients were retrospectively compared for the incidence of VAP and its relationship to stress ulcer prophylaxis in the form of sucralfate, ranitidine, or nothing. Sucralfate was dosed orally by age: children 0-1 year of age received 50 mg qid, children 1-2 years of age received 100 mg qid, children 2-6 years of age received 500 mg qid, children 6-12 years of age received 500 mg qid, and children over 12 years of age received 1000 mg qid. Ranitidine was given IV in a dose of 2.5 - 6 mg/kg per day bid.

The outcomes evaluated:

Primary outcome variables:

1. The incidence rate of VAP as determined by: chest film showing a new infiltrate consistent with pneumonia and at least two of a) temp > 38.5 or < 35.0, b) WBC count > 10,000 or < 3,000, c) isolation of pathogenic bacteria from tracheal aspirate, or d) purulent sputum.
2. Early vs. late VAP, upper airway (endotracheal) colonization determined by endotracheal aspirates in mechanically ventilated patients performed once weekly, or more often if clinically indicated.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

There was no randomization of patients in this study. Over a defined period of time two hundred and fourteen consecutive patients were intubated for a period of greater than 48 hours. Their charts were analyzed retrospectively for inclusion data. One hundred and fifty five charts were included in this study. The treatment regimen the patients received, presumably decided by clinicians without explanation as to what criteria they may have used to decide treatment, defined to which group they were assigned.

Not addressed in this was how the clinicians chose the particular regimen. Did the practice preference of clinicians differ from training or was there a group consensus? How was the NG tube size determined? How did particular patient characteristics influence the clinicians?

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

Was followup complete?

Yes, all patients included were accounted for. Data were analyzed until twenty-four hours post extubation or twenty-four hours after stress ulcer prophylaxis was discontinued. Gram negative bacilli grew in 8 of 11 (72.2%) endotracheal aspirates of patients with VAP. There were 13 patients with VAP, no mention is made of what the other two cases grew.

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

Yes. Crossover does not apply in this retrospective study.

Secondary questions:

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

At the time of active treatment, the medical staff were not blinded to the medications given to each patient. Only after treatment and discharge, however, was a study performed. The authors do not mention whether the retrospective diagnosis of VAP was determined in a blinded fashion. Since this is their main outcome I believe this would be an important point in to include and would improve the results significantly.

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

No, the groups were heterogeneous. The ages of all patients ranged from less than one month to greater than 12 years, and the uppermost age was not clearly defined. The primary diagnoses were quite varied as well, and ranged from respiratory disease to abdominal surgery to CNS surgery to polytrauma. There were statistically significant differences between the groups - patient's ages, incidence of primary respiratory disease, duration of intubation and rates of enteral feeding. This is an important issue in the no prophylaxis group, where 54% had respiratory disease without pneumonia (because this was an exclusion criteria.) This high rate of respiratory disease could easily have masked a different rate of VAP.

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

Treatments that could influence the rate of VAP (i.e., cisapride, erythromycin and antibiotic use) were not different between the three groups. However, no comment was made about any other aspect of treatment.

III. What were the results?

1. How large was the treatment effect?

VAP was diagnosed in 13 (8.4%) of patients. Patients in the ranitidine group (n = 54) developed 6 (11.1%) cases, patients in the sucralfate group (n = 53) developed 4 (7.5%) cases, and 3 (6.2%) patients in the group not treated for stress ulcers (n = 48). No patients developed a second pneumonia, defined as pneumonia developing fourteen days after the clearing of the first pneumonia. There was no significant difference between all three groups or between the ranitidine and the sucralfate groups (p = 0.52). Overall the median number of days from initiation of mechanical ventilation and onset of VAP was 5 (range 2 to 44).

• in the ranitidine group this was 5.5 (range 2-44)
• in the sucralfate group this was 5 (range 2-9)
• in the non-prophylaxis group this was 3 (range 2-9)

Gram negative bacilli grew in 8 of 11 (72.2%) endotracheal aspirates of patients with VAP.

• 5/6 (83.3%) of the ranitidine group
• 2/4 (50%) of the sucralfate group
• 1/3 (33.3%) in the no prophylaxis group

There was no significant difference (chi square p = 0.52) when comparing just the ranitidine and sucralfate groups or between all three groups in the rate of gram negative isolates in patients with VAP. One patient death was attributable to VAP (7.7%), this patient was in the ranitidine group.

Early-onset pneumonia developed in 6/13 (46.2%) of all patients. Of this group, 3/5 (60%) cultures grew out Gram positive cocci. With late onset pneumonia, (developing after the first four days of mechanical ventilation) only 1/8 cultures grew Gram positive cocci (12.5%) (p = 0.052).

Duration of ventilation was significantly longer in patients with VAP, and patients with VAP were more likely to have been treated with cisapride and erythromycin (p = <0.05). Use of antibiotics or enteral feeding made no difference (p = NS). Number of days on mechanical ventilation was the only significant risk factor (p= 0.04) for development of VAP when using a logistic regression model to adjust for differences in variables among the study groups.

Overall 44 patients were found to be colonized with Gram negative bacilli.

• in the ranitidine group 14 (25.9%)
• in the sucralfate group 12 (22.6%)
• in the non prophylaxis group 18 (37.5%)

p = 0.69

So 44 patients grew out Gram negative bacilli, and 8 (18%) of them developed VAP. Five out of 14 with colonization in ranitidine, 2 of 12 in the sucralfate group, and 1/18 in the no treatment group developed VAP.

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

Treatment effect was not precise. The 95% confidence intervals for the individual rate of VAP are 0.05 and 0.22 for the ranitidine group, 0.03 and 0.17 for the sucralfate group, and 0.02 and 0.17 for the no prophylaxis group. These confidence intervals of these proportions overlap, and therefore, the different rates of VAP are not likely to be different between the groups. The difference between ranitidine and sucralfate is 11.1% - 7.5% = 3.6% The 95% CI for this difference is -8% to +15 %, this is obviously not significant since the confidence intervals cross zero.

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

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

Yes. I can apply these results by not letting the use of stress ulcer prophylaxis be altered by my concern that the patient may develop VAP. The results were not significant between groups treated for stress ulcer prophylaxis or nothing at all. Stress ulcer prophylaxis and potential ventilator-assisted pneumonia precautions cannot be considered related from these data alone.

2. Were all clinically important outcomes considered?

Other clinical outcomes were considered. This study showed that almost half of the cases of VAP were early onset most likely due to the introduction of oropharyngeal (Gram positive cocci) bacteria into the airway. However, this early onset distinction is difficult to interpret from the data and clearly wasn't powered to make any true estimations of its impact.

Duration of intubation had a significant impact on the development of VAP. This has been demonstrated in many larger studies (1, 2, 3). The authors themselves questioned whether patients who were intubated longer developed VAP or whether the pneumonia itself prolonged the intubation.

Treatment with enteral antibiotics did not have any effect on patients in the study; similar numbers of patients developed pneumonia with and without antibiotic treatment.

Prokinetics were expected to have a positive effect against GE reflux and risk of aspiration. This study found more pneumonia in patients treated with prokinetics than patients not treated with these agents. However, this is based on an interpretation of only 24 patients treated with cisapride.

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

No significant decreases or increase in the incidences of VAP was shown by using stress ulcer prophylaxis versus no prophylaxis at all. Therefore, there is nothing to benefit either way. Alternatively, there was no apparent increased risk from choosing from any of the three treatment options.

References

1. Elward AM, Warren DK, Fraser VJ. Ventilator-associated pneumonia in pediatric intensive care unit patients: risk factors and outcomes. Pediatrics. 2002 May;109(5):758-64. [abstract]
2. Byers JF, Sole ML. Analysis of factors related to the development of ventilator-associated pneumonia: use of existing databases. Am J Crit Care. 2000 Sep;9(5):344-9; quiz 351. [abstract]
3. Papazian L, Bregeon F, Thirion X, et al. Effect of ventilator-associated pneumonia on mortality and morbidity. Am J Respir Crit Care Med. 1996 Jul;154(1):91-7. [abstract]

Comments

Back to the EB Journal Club Index