Steroids for COPD: Oral Gets More Support

Many of us wouldn’t dream of not using steroids for a significant COPD exacerbation. We put it on the same level as not giving antibiotics for pneumonia or not giving aspirin for a patient having a heart attack. Despite this deeply engrained belief, the data behind steroid use for COPD exacerbations is shockingly small. I suspect most of you would be surprised to know that I was a second year medical student when steroids were really shown to be beneficial for COPD exacerbations (and no, I wasn’t a second year medical student in the 1980’s)^1,2 . With this lack of data, there are many questions that have not been answered about steroid use for COPD exacerbations. Much of what we do, or actually ignore as you will see, is based upon consensus statements or small studies. The evidence for steroid use received a small boost in last months JAMA, although not as big of a boost as many of us would hope. In this blog I will review the evidence for steroid use, including the recent article. So read on as we look at a topic that should have overwhelming evidence, yet doesn’t!

1999 Finally Some Evidence for Steroid Use
Prior to 1999, using steroids for COPD exacerbations was the standard of care, mainly based upon Asthma studies showing benefit. There were small studies looking at COPD, but the primary end points measured were FEV1’s. While this maybe a marker for clinical improvement, it is not a truly significant patient important outcome. This changed in 1999 when two studies came out.
The first, and the best, was a study by Niewoehner from the VA¹. This was a randomized, double blinded study comparing a 2 or 8 week course of steroids vs placebo. Patients were followed for 6 months for the primary outcomes of death, intubation, re-hospitalizations, or need to intensify treatment. The steroid arms all received 125mg IV solu-medrol q6 hour for 72 hours followed by prednisone starting at 60mg daily. At that point the patients were either tapered over 8 or 2 weeks. Amazingly, the study consisted of only 271 participants (which turns out to be quite large for COPD trials...). This was mainly related to many patients getting excluded. The exclusion criteria included steroid use within the past 30 days, comorbidities making 1 year survival unlikely (not clearly defined), less than 30 pack year of smoking, age less than 50, asthma, FEV1 over 1.5, and inability to give consent. The randomization also had a few problems. The 8 week steroid and placebo arms had a higher percentage of smokers and past steroid use than the 2 week steroid arm. Also, the two steroid arms had a higher percentage of diabetics. Other than these three categories, the patients were similar.
The study was stopped one year early due to clear benefit in the steroid arms for COPD. The treatment arms had reduced hospital stay for COPD (1.2 days shorter) along with a reduction in the primary outcomes measured at 30 and 60 days. This benefit appeared to disappear by 6 months out, with the groups similar at that point. A couple of interesting notes from the study. First is that there did not appear to be a benefit between the 2 and 8 week groups. The second is that while there was reduced hospital stays and readmissions for COPD in the steroid arm, there were increased non-COPD related hospital stays and readmissions for the steroid arm. The authors stated that these readmissions were not felt to be related to the steroids, but did not offer any further information.
I must admit that upon reviewing this study, I was rather disappointed. It was small, it had randomization problems, and the results bugged me (increased hospitalizations for the “other” category). It answered the question on whether steroids are beneficial for COPD exacerbations, but it also raised questions and concerns. I was hopeful that the other “big” COPD study in 1999 would be better. 
Get ready for even more disappointment!
The second study was by Davies et al, and was also a double blind, placebo controlled, randomized study. In this British study, a whopping 56 patients were either randomized to placebo (27 patients) or a 14 day course of 30 mg of daily prednisolone. To make matters worse, 6 patients dropped out, leaving a total of 50 patients. The steroid arm did show benefit, but I’m not sure a study of 50 people is worth spending more time on.

Oral Versus IV Steroids
The bioavailability is about the same for IV versus PO steroids³. With this knowledge, it would be reasonable to believe that PO should be equivalent to IV steroids. This has been shown to be the case in the pediatric asthmatic population, which is why the major organizations have recommended to use PO steroids for COPD exacerbations⁴. 
In 2007, a study from the Netherlands attempted to look at this question⁵. In a double blind, randomized study, 435 patients admitted with COPD exacerbations were given a five day course of 60mg IV or PO prednisolone. The primary outcomes were death, ICU admission, readmission, or the intensification of treatment. The results did not show any difference between IV and PO steroids when looking at the primary outcomes. While this study, in just reading the abstract, looks great and seems to really answer the question, there is a problem.
The study had a very difficult time recruiting patients. When I looked at the study design, I think I found the reason why. The study attempted to make sure that the patients were all treated similar, with the only variable being the steroid. All patients were placed on nebulizer treatments and antibiotics, as would be standard. The problem is that the patients were placed on a q6 hour schedule for the nebulizers! On our service, once the patient is doing well on q4 hour they are d/c’d home. We wouldn’t even admit a patient who could go with q6 hour nebulizers. So, for patients who do not have a severe COPD exacerbation, PO and IV are equivalent. That unfortunately wasn’t the question. 
Ugh, just more disappointment. Is there a COPD study that will not leave me disappointed? 

JAMA June 2010 (Please don’t disappointment me JAMA!)
In last month’s JAMA, a study came out that got my hopes up pretty high⁶. The title seemed to scream “This is going to be great!”. I was pumped. How can an article titled “Association of Corticosteroid Dose and Route of Administration With Risk of Treatment Failure in Acute Exacerbation of Chronic Obstructive Pulmonary Disease” not rock? This had the answers, right?
This study looked at patients admitted in 414 hospitals with a principle diagnosis of COPD exacerbation. While this makes it a retrospective study, it was able to gather information on 79,985 patients! They gathered data on the patients, including steroid use, hospitalization stay, cost, ICU stay etc. The study then seperated the groups into a “high dose steroid” arm and a “low dose steroid” arm. Interestingly, despite all of the major organizations recommending oral steroids for COPD exacerbations, 92% of the patients were placed on high dose IV steroids at admission! The primary outcomes studied were the need for intubation (after day 2 of hospitalization), length of stay, readmission, cost, and death. High dose steroids included what is often done at Christ Hospital, solu-medrol 125mg IV once then 60mg IV q6 hour. Low dose was equivalent oral prednisone of 80mg or less over the first 2 days (40mg PO daily). Patients in the low dose category were more likely to be admitted to a university setting, which I don’t think is all that surprising. 
I think you all see the problem with this retrospective study. Are patients who are sicker more likely to be given IV steroids? This is a significant bias. To help negate this bias, the investigator did some fancy statistical work. If anyone understands what “we developed a model combining an ecologic, or grouped treatment variable, with individual-level covariates and outcomes.”, let me know. I was a little lost. Ok, I think it means that they attempted to match patients who received low dose oral steroids with equivalent patients who received high dose IV steroids to negate any bias.
The results did not show a benefit for the high dose, IV steroid arm. In fact, the low dose arm seemed to have shorter stays and lower cost ($500). When looking at the two groups, the low dose arm was older with more co-morbidities, which was surprising. They were also more likely to be underserved.

Time for us to Change Care

As evidence goes, a retrospective study is not as good as a large, double blind, randomized trial, but this is likely the best we will get for COPD. According to the editorial accompany the JAMA article, it would require a study of 30,000 patients to detect a 1% difference between oral and IV steroids for COPD exacerbations⁷. We can’t even get a study of over 500, so I doubt this is going to happen anytime soon.
I do believe though we need to change our management at The Christ Hospital. First, there isn’t any pathophysiological reason to believe that IV should be superior. Second, the evidence, though not perfect, appears to support that low dose oral is equivalent to high dose IV. Third, all of the guidelines recommend low dose oral. This is enough to say we will be moving to oral prednisone (40mg daily) for our patients admitted for COPD exacerbations. More does not mean better. 

References
1. Niewoehner DE, Erbland ML, Deupree RH, et al, for the Department of Veterans Affairs Cooperative Study Group. Effect of systemic glucocorticoids on exacerbations of chronic obstructive pulmonary disease. N Engl J Med. 1999;340:1941- 1947.
2. Davies L, Angus RM, Calverley PM. Oral cortico- steroids in patients admitted to hospital with ex- acerbations of chronic obstructive pulmonary dis- ease: a prospective randomised controlled trial. Lancet. 1999;354:456-460.
3. Al-Habet S, Rogers HJ. Pharmacokinetics of intravenous and oral prednisolone. Br J Clin Pharmocol 1980; 10:503–508
4. Global Strategy for the Diagnosis, Management and Prevention of COPD, Global Initiative for Chronic Obstructive Lung Disease (GOLD)
5. de Jong YP, Uil SM, Grotjohan HP, Postma DS, Kerstjens HAM, van den Berg JWK. Oral or IV pred- nisolone in the treatment of COPD exacerbations: a randomized, controlled, double-blind study. Chest. 2007;132(6):1741-1747.
6. Lindenauer PK, Pekow PS, Lahti MC, Lee Y, Benjamin EM, Rothberg MB. Association of corticosteroid dose and route of administration with risk of treatment failure in acute exacerbation of chronic obstructive pulmonary disease. JAMA. 2010; 303(23):2359-2367.
7. Jerry A. Krishnan, Richard A. Mularski. Acting on Comparative Effectiveness Research in COPD. JAMA. 2010; 303(23): 2409-2410