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

Inhaled Iloprost for Severe Pulmonary Hypertension.

Olschewski H, Simmoneau G, Galie N, et al.

N Engl J Med 2002;347(5):322-9. [abstract]

Reviewed by Ayse Akcan Arikan, MD, Baylor College of Medicine, Pediatric Critical Care Section

Review posted June 16, 2003

---

I. What is being studied?:

The study objective:

To evaluate the effects of inhaled iloprost-a stable prostacycline analogue in patients with pulmonary hypertension "using a rigorous end point of clinical efficacy" (i.e. indices of exercise tolerance and severity of heart failure).

The study design:

Prospective, multicenter, randomized, double blind controlled clinical trial.

The patients included:

203 patients with primary pulmonary hypertension and selected forms of non-primary pulmonary hypertension (appetite suppressant and scleroderma associated and inoperable chronic thromboembolism related) were included from 37 European specialist centers.

Inclusion criteria:

• Mean pulmonary artery pressure > 30 mm Hg
• Ability to walk between 50-500 meters without encouragement on the six-minute walk test
• NYHA functional class III or IV despite standard medical management (including anticoagulants, diuretics, digitalis, calcium channel blockers, oxygen)

The patients excluded:

Patients with significant pulmonary disease (described as forced expiratory volume/second value two standard deviations below normal, forced vital capacity < 50%), renal insufficiency (creatinine clearance < 30ml/min), hyperbilirubinemia (bilirubin > 3 mg/dl), bleeding disorders, patients who had a pulmonary wedge pressure > 15 mm Hg at rest, cardiac index < 1.5 or > 4 L/min/m2 BSA and clinically unstable patients were excluded from the study. Patients taking investigational drugs, prostanoids, and beta-blockers were not included. During the course of the study, patients developing refractory systolic arterial hypotension (defined as systolic arterial blood pressure < 85 mm Hg), worsening congestive heart failure (manifested as refractory edema or ascites), patients with rapidly progressing cardiogenic, hepatic or renal failure were excluded. Patients with a decrease in measures of hemodynamic function (CVP, mixed venous saturations) or decrease of at least 30% in the distance walked in the six-minute walk test were also excluded.

The interventions compared:

Repeated daily inhalation of iloprost (six to nine times a day, median inhaled dose 30 mcg/day) to inhalation of placebo

The outcomes evaluated:

The primary outcome measure is stated as an increase of at least 10% in the distance covered in the six minute walk test and an improvement in the NYHA functional class. (However, in the discussion of the results, the first figure is a graph of the mean change in the distance walked against time only and this finding is referred to as the primacy efficacy end point, different from what is stated in the methods). Secondary end-points were any change in the six-minute walk test results, NYHA class, Mahler Dyspnea Index scores, hemodynamic variables, quality of life and also clinical deterioration, death, and the need for transplantation.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. It is stated that the assignment of the patients to experimental vs. placebo group was done by an independent committee - that was unaware of the patients' identities - in a random fashion after stratification according to NYHA class and type of pulmonary hypertension. However, we are not provided the details of how the actual randomization was carried out. We are also not given information regarding the number of patients screened for enrolment or the number of patients enrolled by each different center, so it is not possible to ascertain any selection bias.

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

Was followup complete?

Yes. Patients were followed for a period of 12 weeks during which one patient in the experimental group and four patients in the control group died. Three patients in the iloprost group and ten patients in the placebo group did not complete the study due to discontinuation of the study medication and withdrawal of consent due to clinical deterioration, insufficient clinical benefit, or both. (Although clinical deterioration is defined explicitly in the text, it is not exactly clear what authors mean by "insufficient clinical benefit"). Six patients in the experimental group and 17 patients in the control group had missing data in the six-minute walk test, these patients were considered not to have had a positive response. For patients who discontinued the study, the authors used a last observation carried forward analysis, if we assume that the patients who left the study did so because of failure to improve or of clinical deterioration, then it is unlikely that this analysis introduced any bias into the results.

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

Yes. There was no crossover between groups.

Secondary questions:

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

Yes, this was a double blind study. To prevent unblinding, baseline hemodynamic measurements post-inhalation of the first dose of drug or placebo were not carried out.

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

Yes. The only major differences were that the control group had more patients with collagen vascular disease and the experimental group had more patients with chronic thromboembolic pulmonary hypertension. In addition, the placebo group tended to have a slightly higher pulmonary vascular resistance and the iloprost group had a higher six-minute walk distance. However, none of these difference reached statistical significance and the groups were very similar form all other aspects.

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

We are not provided with adequate information to make a decision about this point. It is reasonable to suggest that there may have been differences in the management between different centers since there was no standardization of management.

III. What were the results?

1. How large was the treatment effect?

If we take mean improvement in distance walked in the six-minute walk test as the primary efficacy end point, then the odds ratio was 3.97, i.e. the odds of having an increase in the distance walked in the experimental group was 3.97 compared to the control group. The authors have also performed a logistic regression analysis that reveals baseline characteristics and demographic data were not associated with response to treatment.

For the combined end point of improvement in NYHA class and increase in the distance walked, the absolute benefit increase (ABI) was 17% - 5% = 12% and the number needed to treat (NNT) was 1/0.12 = 9. In other words, nine patients needed to be treated with iloprost to achieve an improvement in the NYHA class or a change in the distance walked.

The combined rate of clinical deterioration or death was lower in the experimental group (4.9% vs. 11.8% in the placebo group, ARR 6.9%, NNT 15). Among other secondary endpoints, the mean Mahler Dyspnea Index score was better in the experimental group at the end of the trial irrespective of the type of pulmonary hypertension. Quality of life, as measured by the Euro-Qol health-state score, improved significantly in the treatment group. The hemodynamics of the placebo group deteriorated significantly at week 12 whereas remaining unchanged before and improving considerably after the administration of the study medication in the treatment arm. The acute hemodynamic response to inhaled iloprost was preserved even at week 12 in the experimental group.

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

Not very precise. The 95% confidence interval for the odds ratio reported above was 1.47 to 10.75. The 95% confidence interval for the ABI was 3.49% to 20.5% and for the NNT was 4.9 to 28.7.

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

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

This is not an easy question to answer. It is an obvious fact that this is an adult study done in a population of patients with a variety of underlying diagnoses (such as pulmonary hypertension secondary to appetite suppressant agents or chronic thromboembolic disease) quite rare in the pediatric age group. On the other hand, fifty percent of the patients enrolled in the study have primary pulmonary hypertension, which is the most common form of the disease seen in our patient group. Although there are quite a number of case reports and short case series of inhaled iloprost use in the pediatric population (1,5), most of them are confined to cardiac patients and secondary pulmonary hypertension (2,3,4). Moreover, the results of this particular study should be carefully interpreted mainly because it is quite difficult to extrapolate the outcome measures to our patient population.

It would be appropriate to say that this study points out the presence of an inhaled agent that is efficacious in adults with pulmonary hypertension that seems to halt clinical progression as well as improve exercise tolerance and decrease severity of heart failure without many adverse effects. The magnitude of the benefit is big enough to suggest that a pediatric study is in order to look at iloprost use in our patient group.

2. Were all clinically important outcomes considered?

The authors should be commended for using a wide variety of outcome measures (namely NYHA functional class, exercise tolerance, dyspnea index, quality of life questionnaire, hemodynamic parameters and survival) all of which are measures of important parameters in deciding to implement a new treatment regimen. The adverse event rate seemed to be similar between the two groups with the notable exception of syncope, which was described as "serious" in the experimental group. The authors note that syncope resulted in head trauma in one patient in the iloprost group. This brings forth the question of additional health care resource use resulting from this adverse event (e.g. emergency room visits, hospitalization, and so on), which is the only outcome that is not investigated in this study.

It is important to note that the rapid improvement in response to inhaled iloprost might have identified patients in the treatment arm prematurely thus introducing bias and influencing the results.

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

Since the only serious harmful effect of the experimental treatment in this study was syncope; the absolute risk increase for syncope in the experimental group was 3% and the number needed to harm was 34, i.e., for each 34 patient on treatment one would have a syncopal episode.

The financial aspect of instituting this treatment was not investigated in this study.

References

1. Wittwer T, Pethig K, Struber M, et al. Aerosolized iloprost for severe pulmonary hypertension as a bridge to heart transplantation. Ann Thorac Surg. 2001;71(3):1004-6. [abstract]
2. Rimensberger PC, Spahr-Schopfer I, Berner M, et al. Inhaled nitric oxide versus aerosolized iloprost in secondary pulmonary hypertension in children with congenital heart disease: vasodilator capacity and cellular mechanisms. Circulation. 2001;103(4):544-8.
3. Winberg P, Lundell BP, Gustafsson LE. Effect of inhaled nitric oxide on raised pulmonary vascular resistance in children with congenital heart disease. Br Heart J. 1994;71(3):282-6.
4. Curran RD, Mavroudis C, Backer CL, et al. Inhaled nitric oxide for children with congenital heart disease and pulmonary hypertension. Ann Thorac Surg. 1995;60(6):1765-71.
5. Gorenflo M, Bettendorf M, Brockmeier K, et al. Pulmonary vasoreactivity and vasoactive mediators in children with pulmonary hypertension. Z Kardiol. 2000;89(11):1000-8.

Comments

Back to the EB Journal Club Index