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

Acute and sustained effects of early administration of inhaled nitric oxide to children with acute respiratory distress syndrome.

Fioretto JR, De Moraes MA, Bonatto RC, Ricchetti SM, Carpi MF.

Pediatr Crit Care Med. 2004 Sep;5(5):469-74 [abstract; full-text (192K)]

Reviewed by Jennifer Matchey MD, Medical College of Wisconsin, Children's Hospital of Wisconsin, Milwaukee, WI

Review posted December 20, 2005

---

I. What is being studied?:

The study objective:

To determine the acute and sustained effects of early inhaled nitric oxide (iNO) on some oxygenation indexes and ventilator settings and to compare inhaled nitric oxide administration and conventional therapy on mortality rate, length of stay in intensive care, and duration of mechanical ventilation in children with acute respiratory distress syndrome (ARDS).

The study design:

Observational Study - including Prospective Cohort and Retrospective Historical Cohort

The patients included:

The prospective treatment group (iNOG) included eighteen children (ages 1 month to 12 years) with ARDS, admitted to the PICU from 2000-2002. ARDS was defined according to the American-European Consensus Conference published in 1994 (1). All patients had <88% arterial saturation on positive end-expiratory pressure (PEEP) ≥10cm H2O and FiO2 ≥ 0.6.

The historical control group (CTG) (n=21) were children admitted from 1998-2000, were part of a previous report of ARDS patients managed conventionally (2).

Location was at the University Hospital of Botucatu Medical School in Sao Paulo, Brazil.

The patients excluded:

Patients with chronic cardiac or pulmonary disease were excluded.

The interventions compared:

The two interventions compared were iNO versus conventional therapy for ARDS. For the treatment group (iNOG), iNO was first delivered at 20 ppm for 30 minutes, reduced to 10 ppm for 30 minutes, and then reduced to 5 ppm for 3 hours. If there was a positive response then iNO was continued at 5 ppm with a weaning protocol to decrease by 1 ppm until the lowest level was reached which maintained an improvement in oxygenation. If there was no response then the trial was repeated but beginning at 40 ppm. If there was still no response then there was a daily trial of 20 ppm for 4 hours. For the control group (CTG), there was no iNO delivered and only conventional therapy given.

The outcomes evaluated:

Outcomes included PaO2/FiO2 ratio, oxygenation index (OI), FiO2, peak inspiratory pressure (PIP), mortality rate, intensive care stay, and duration of mechanical ventilation.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

No, it was an observational study with the iNOG group a prospective cohort and the CTG group a retrospective cohort.

This makes drawing conclusions regarding important clinical outcomes highly suspect.

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

Was followup complete?

The original report of their control group had 24 patients (2) but 3 were excluded presumably due to chronic cardiac or pulmonary disease. Otherwise everyone was included in the results.

Yes. All measures studied were performed on each patient. It is not entirely clear how long patients were followed, but all subjects appear to have been followed to death or PICU discharge.

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

No randomization occurred. They were analyzed based on being in the treatment or historical control group.

Secondary questions:

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

No, it was a prospective treatment group and retrospective control group.

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

The groups were compared for patient age, gender, presence of multiple organ system failure, PRISM III score, inotropic support and ARDS etiology. In the comparison table (Table 4) it appears that both groups are similar based on those variables. However, the authors provide a detailed demographic table for the CTG (Table 1) but do not have a similar table for the iNOG. Only a comparison table for the two groups (Table 4) is provided. I feel to be complete they should include an identical demographic table for both groups or not provide this detail of information for either group.

When analyzing the ventilator settings at the start of the trial, the iNOG had higher FiO2 than the control group. The iNO group also had worse oxygenation indexes and PaO2/FiO2 ratios than the CTG. This may indicate that the iNOG had worse lung disease at the time of ARDS diagnosis. However, given the uncertainties in quality and timing of measurements between a prospective and historical cohort, these differences should be regarded with suspicion.

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

The authors used the same mechanical ventilation protocol (allowing for permissive hypercapnia and relative hypoxemia), sedation and hemodynamic support. The explicit details of the protocols are not described. Some patients varied on whether they were prone, paralyzed or had a cuffed endotracheal tube. It was mentioned that there were equal numbers of patients in each group having prone positioning and cuffed tubes. It is not at all clear, however, that an explicit management algorithm was followed, leaving open the high likelihood of significant management differences over the course of time.

III. What were the results?

1. How large was the treatment effect?

There were two parts of the study with different outcomes:

The FIRST aim of the study was to determine the acute and sustained effects of early iNO on some oxygenation indexes and ventilator settings. This part can be seen as a descriptive study since there was no control group.

Oxygenation indexes:

• PAO2/FiO2 - this was only looked at for the acute effects of iNO. After a 4hr trial of iNO it was shown that there was an 83% increase from baseline in this ratio for the iNOG. However, this value was not calculated for the sustained effects of iNO.
• OI - During the acute effects of iNO it was shown that the iNOG had a 46% decrease in this value from baseline. The study also showed a sustained effect after 4 days of iNO with a 67% decrease from baseline. To note there was no comment of the change in OI over time for the CTG.

Ventilator Settings:

• PIP - the study showed a change in PIP over 3 days of being on iNO (median 30 to 24). It appears the PIPs on day 2 and 3 were lower in the iNOG subjects compared to the CTG subjects.
• FiO2 - the study showed a change in FiO2 over 3 days on being on iNO (0.91 to 0.51). There does not appear to be a difference over time in FiO2 between the two groups (higher on day 0 in the iNOG subjects).

The SECOND aim of the study was to compare iNO and conventional therapy on mortality rate, length of stay in the PICU, and duration of mechanical ventilation.

• Mortality Rate - The rate was lower in the iNOG than CTG (p< 0.001). For mortality rate, the relative risk calculates to be 0.35 (CI 0.113 to 1.079), Absolute Risk Reduction is 0.31 (CI 0.035 to 0.584), and the number needed to treat is 3.21 (CI 2-29).
• Length of stay and duration of mechanical ventilation - no significant difference was shown. In fact both had a trend toward higher median days in the iNO group.

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

The only non-descriptive statistical value to evaluate was mortality rate. The part 1 aims were all descriptive with no control group data. The only significant value with dichotomous variables occurred in the mortality data for part 2.

For mortality rate:

• The relative risk is not significant as the confidence interval range involves 1.
• For the absolute risk reduction calculation, the confidence interval range indicates that the risk difference between the two groups is between 3.5% and 58%.
• The number needed to treat lies between 2 and 29 patients in order to achieve a favorable outcome.

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

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

Yes, the range of ages and types of patients are commonly seen in the pediatric ICU.

iNO is readily available in the PICU.

2. Were all clinically important outcomes considered?

Many of the clinically important outcomes were examined, i.e., mortality, length of stay, ventilation parameters.

However, this was a descriptive study and the important outcome of whether the iNOG had clinically more effective results can not be concluded. There needs to be a randomized controlled trial with iNO.

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

If there is a difference in mortality that can be repeated in future studies than YES...

However, for this study it is unclear if the mortality improved because of iNO or if it is because of clinical improvements over the 4 year span. We cannot and should not conclude that iNO in pediatric ARDS improves mortality from this study.

References:

1. Bernard, G.R., et al. Report of the American-European Consensus Conference on ARDS: Definitions, mechanisms, relevant outcomes and clinical trial coordination. Intensive Care Medicine, 1994. 20: pp. 225-232. [abstract]
2. Fioretto, J.R., et al. Sindrome do Desconforto Respiratorio Agudo em Criancas: Incidencia, Mortalidade e Trocas Gasosas. Rev Bras Terap Intens, 2001. 2: pp.58-62.

Comments

Back to the EB Journal Club Index