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

Late steroid therapy in primary acute lung injury.

Varpula T, Pettilä, Rintala E, Takkunen O, Valtonen.

Intensive Care Med. 2000;26(5):526-31. [abstract]

Reviewed by Umesh Narsinghani, MD and M.J. Heulitt, MD, University of Arkansas for Medical Sciences/Arkansas Children's Hospital

Review posted September 22, 2000

I. What is being studied?:

The study objective:

To investigate the effect of steroid treatment in the late phase of primary acute lung injury with special emphasis on pneumococcal pneumonia.

The study design:

Retrospective study, in a multidisciplinary ICU in a university hospital.

The patients included:

31 consecutive patients meeting the criteria described below, with primary acute lung injury requiring mechanical ventilation for more than 10 days; 16 patients received treatment with methlyprednisolone and 15 served as controls.

Inclusion criteria:

1. Mechanical ventilation for acute respiratory failure for more than more than 10 days.
2. Pulmonary, infectious etiologies of acute lung injury.

The patients excluded:

Exclusion Criteria:

1. Prior corticosteroid treatment.
2. Chronic immunosupression.
3. Respiratory failure due to neurologic problems.
4. Surgery during the first 10 days in the ICU.

The interventions compared:

16 patients received empiric late steroid therapy; intravenous methylprednisolone was given in a dose of 80 mg in the morning and 40 mg in the evening. Based on a clinical response the dose was tapered. 15 patients were managed without steroids and served as controls.

The outcomes evaluated:

The mean start of therapy was 9.7 days after establishment of respiratory failure. The following data were recorded when steroid therapy was started: Multiple Organ Dysfunction Score (MODS), PaO2/FiO2 ratio, Lung Injury Score (LIS) and serum C-reactive protein concentration (S-CRP), and on day 3 and 7 thereafter.The corresponding data were collected for control patients on days 10, 13, and 17 after admission. Other outcomes measured were length of mechanical ventilation, length of ICU stay and 30 day mortality.

II. Are the results of the study valid?

1. Is there a strong, independent, consistent association between the surrogate end point and the clinical end point?

Although, there is some data available stating that improvement in PaO2/FiO2 ratio, reduction in lung injury score and decreased MODS score is associated with reduced mortality, this concept needs to be approached with caution, as these studies were limited by their sample size (1, 2). Contrary to this several other similar studies have failed to demonstrate improved outcomes.

2. Is there evidence from randomized trials in other drug classes that improvement in the surrogate end point has consistently led to improvement in the target outcome?

No, there is no consistent evidence that improvement in the surrogate end point of oxygenation, by use of other drugs, e.g., inhaled nitric oxide, surfactant, partial liquid ventilation, etc. has led to improvement in mortality in severe acute lung injury.

3. Is there evidence from randomized trials in the same drug class that improvement in the surrogate end point has consistently led to improvement in the target outcome?

Yes, at least in two recent studies published, prolonged administration of methylprednisolone in patients with unresolving ARDS was associated with improvement in oxygenation, lung injury and MODS scores, and reduced mortality (1, 3). Again, this statement needs to be approached with care until further studies estimating the positive effects are established.

IIa. Validity Questions for Therapy Articles:

Primary questions:

1. Was the assignment of patients to treatments randomized?

No, it was a retrospective study wherein the medical records of all mechanically ventilated patients during 1994 - 1998 were evaluated. Although guidelines for corticosteroid use had been adopted empirically for patients with unresolved ALI at this institution, no consistent approach had been developed, and only those patients with ALI and consistently high levels of CRP were considered for steroid therapy. This limitation of a retrospective study perhaps made it likely that there were significant differences between the groups, some of which may be known but others unknown - it is the unknown variables especially that randomization is expected to distribute equally between the groups.

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

Was followup complete? and were patients analyzed in the groups to which they were randomized?

Yes, patients from both the study and control groups were accounted for. Of the 31 patients enrolled, there were 3/16 (study group) and 3/15 (control group) non-survivors respectively.

No, there was no true randomization of patients as this was a retrospective review of medical records. Nevertheless, there did not appear to be any crossover of patients between groups.

Secondary questions:

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

No, since the decision to initiate steroid therapy was based on consultation of an Infectious Disease specialist and individual evaluation of the patient; investigators were not "blind" to treatment.

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

Yes, both the groups were comparable on admission and at start of therapy; differences in terms of age, gender, APACHE scores, MODS, PaO2/FiO2 ratio, and etiology of ALI were insignificant. There seemed to be more patients with pneumococcal pneumonia in the steroid group but perhaps due to the small numbers of patients, this difference (11/16 vs. 7/15) was not statistically significant.

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

Apparently, as general care of patients consisted of invasive hemodynamic monitoring, routine ulcer prophylaxis, thromboprophylaxis with low molecular weight heparin and early enteral nutrition. Pressure control mode was used (limiting static inspiratory airway pressure under 35cm of H2O), and PEEP level was determined on a clinical basis; active use of prone position was a part of the ventilator strategy. Infection surveillance was performed twice a week. Selective decontamination of GI tract, extracoporeal life support or nitric oxide were not used. During the study period no major changes occurred in treatment protocols. However, no data is provided to confirm that there was no difference in these aspects of care between the two groups.

III. What were the results?

1. How large, precise, and lasting was the treatment effect? (Effect should be large, precise, and lasting to consider a surrogate trial as possible basis for offering patients the intervention.)

The authors, mention a significant improvement in PaO2/FiO2 ratio, with decreases in MODS and CRP within 3 days after the start of therapy. There is, however, no mention about concomitant decreases in FiO2 and pertinent levels of PEEP. No significant differences appeared in the length of mechanical ventilation, ICU stay, and 30 day mortality (although the authors acknowledge the limited power of their study to assess this outcome).

Surrogate endpoint	Steroid therapy (n=16)	Control patients (n=15)	p value
MODS at inclusion Change in 3days after inclusion	6.6 (2.6) -2.0 (0.4)	6.3 (3.3) -0.4 (0.6)	0.71 0.15
PaO2/FiO2 at inclusion Change in 3days after inclusion	136.9 (48.7) 60.0 (12.9)	160.4 (55.0) -6.0 (7.6)	0.15 0.004*
CRP at inclusion, mg/ml Change in 3days after inclusion	170.3 (47.9) -130.7 (9.1)	131.0 (50) -19.2 (10.1)	0.03 <0.001*

* Statistically significant.

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

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

Although the study was performed in adults, the pathophysiology of ARDS is believed to be the same in our PICU population. However, given the limitations inherent in this small retrospective study, and the lack of a known relationship between the surrogate and clinical endpoints, we should be hesitant in applying these results to our patients.

2. Were all clinically important outcomes considered?

Yes, the authors measured all the pertinent physiologic variables, including important outcomes like mortality, morbidity, length of mechanical ventilation days, and length of ICU stay.

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

The authors did not look at direct harms or costs involved in the study. The infectious complications observed were not significantly higher in the study group. As there is no single effective pharmacological therapy for ALI, and based on recent encouraging data, prolonged steroid administration in patients with unresolving ARDS appears to be a promising therapeutic approach. A large prospective study to address this important question is being conducted now by the ARDS-net group.

References

1. Meduri GU, Headley AS, Golden E, et al. Effect of prolonged methylprednisolone therapy in unresolving acute respiratory distress syndrome: a randomized controlled trial. JAMA. 1998; 280:159-65. [abstract] [PedsCCM EB Journal Club review]
2. Wiswell TE, Smith RM, Katz LB, et al. Bronchopulmonary segmental lavage with Surfaxin (KL(4)-surfactant) for acute respiratory distress syndrome. Am J Respir Crit Care Med. 1999;160(4):1188-95. [abstract]
3. Keel JB, Hauser M, Stocker R, Baumann PC, Speich R. Established acute respiratory distress syndrome: benefit of corticosteroid rescue therapy. Respiration. 1998; 65:258-64. [abstract]

 

Comments

Back to the EB Journal Club Index