Letters to the Editor  |   August 2013
The Effect of OMT on Postoperative Medical and Functional Recovery of Coronary Artery Bypass Graft Patients
Author Affiliations
  • Donald R. Noll, DO
    New Jersey Institute for Successful Aging; Professor of Medicine, Rowan University School of Osteopathic Medicine, Stratford, New Jersey
Article Information
Cardiovascular Disorders / Osteopathic Manipulative Treatment
Letters to the Editor   |   August 2013
The Effect of OMT on Postoperative Medical and Functional Recovery of Coronary Artery Bypass Graft Patients
The Journal of the American Osteopathic Association, August 2013, Vol. 113, 595-596. doi:
The Journal of the American Osteopathic Association, August 2013, Vol. 113, 595-596. doi:
To the Editor: 
I congratulate Dr Wieting and colleagues1 for their original contribution from May 2013. The authors have successfully brought a well-designed study to publication, which is no easy task. The osteopathic medical profession needs many more such projects completed to the publication stage if osteopathic manipulative treatment (OMT) is to be realized to its full potential. I particularly appreciate that the authors published power analyses to provide guidance for future studies. 
Nevertheless, I am puzzled by the authors' assertion that OMT had a more positive effect on the postoperative recovery of patients who underwent coronary artery bypass grafting when none of the outcomes were statistically significant. In other words, I do not believe it is correct to characterize the intervention as “beneficial, though not statistically significant.” For example, the authors report that the mean length of stay for the OMT group was 0.6 days shorter than the conventional care control group. However, the P value for this datum is .49, which to my understanding means that they were 51% confident something other than random chance is occurring. This result does not inspire confidence, nor does it justify implying a beneficial effect. As is well known, it requires a P value of .05 or less to say with confidence that something other than random chance has influenced the outcome. Despite the authors' enthusiasm, these results are not persuasive for using OMT as a standard therapeutic treatment for patients with the condition studied. 
We should remember this was an exploratory study. If the treatment protocol was tweaked and the study size increased, then OMT might be shown to be efficacious after coronary artery bypass grafting in the future. Publishing the study's protocol, design, preliminary results, and power analysis is the real value of the study. The authors do not need to settle the question of benefit in 1 step. The results are encouraging enough for future projects to build upon the work. The ultimate goal is not to prove OMT, but to discover what helps people. This distinction may be subtle but is important to remember. 
The authors also wrote, “A 1976 study2 by Rogers and Rogers has shown potential changes in autonomic nervous system function in coronary heart disease after OMT is performed.” This statement is a classic example of how myths percolate through the osteopathic medical profession. The reference is impressive and I am certain was given in good faith. Enthusiastic proponents of osteopathic principles and practice will no doubt read this statement and repeat it to students, who in turn will take it at face value. Few will check the reference because the study by Rogers and Rogers2 is available online as a citation only. “The Role of Osteopathic Manipulative Therapy in the Treatment of Coronary Heart Disease,” the title of the Rogers and Rogers article,2 is also misleading. 
In fact, the Rogers and Rogers article2 is not technically a study. It is a well-written review of the literature and case report of 2 individuals who had definite symptoms of ischemic heart disease but no abnormalities at coronary angiography. Additionally, the patients had transient coronary artery spasms in coronary arteries that were free of atherosclerosis. In these 2 case reports, Rogers and Rogers did not report use of OMT or somatic dysfunction measurements; nor did they include information about the autonomic nervous system function of the patients. As enthusiastic advocates of OMT, however, the authors speculated, “It is logical to assume that manipulative treatment, by normalizing the action of the autonomic nervous system, might influence both cellular metabolism and the vasomotor dynamics of the coronary arteries.” Nevertheless, assuming is a long way from reporting hard data. 
Lastly, I believe the study by Wieting et al1 is a good illustration of the limits of the 3-arm study design for small, exploratory clinical trials of OMT. I can speak with more authority than most because I was one of the first to use a 3-arm design for OMT3 and gained much experience with it during the Multicenter Osteopathic Pneumonia Study in the Elderly.4 The principal problem with a 3-arm study design—which encompasses an OMT group, a sham or placebo treatment group, and a conventional care only control group—is that the statistical power of the study is invariably reduced. 
Hróbjartsson and Gøtzsche,5 who conducted a systematic review of the literature on the systemic effects of placebo in clinical trials, questioned the power of the placebo. Preliminary, exploratory, or pilot projects are often statistically underpowered, and the 3-arm study design exacerbates this problem. A 2-arm trial (ie, experimental treatment vs conventional care–only control group) is a reasonably good design for a preliminary project. This study design addresses the question I most frequently ask: Does OMT improve standard care? Such a design is also technically easier to implement because it saves the investigators the trouble of performing a sham or placebo treatment. The mystical effects of touch can be sorted out later. 
On the other hand, analyzing the effect of OMT vs sham or placebo may work well in some instances because the use of a non-treatment arm allows the investigator to offer some type of treatment to all the prospective participants, which can facilitate recruitment. A placebo treatment also controls for attention, touch, and other theoretically beneficial aspects of a physician treating a patient. All the same, a 3-arm study design should be reserved for large, well-funded, definitive projects where each group is adequately powered. 
Wieting JM, Beal C, Roth GLet al. The effect of osteopathic manipulative treatment on postoperative medical and functional recovery of coronary artery bypass graft patients. J Am Osteopath Assoc. 2013;113(5):384-393. [PubMed]
Rogers JT, Rogers JC. The role of osteopathic manipulative therapy in the treatment of coronary heart disease. J Am Osteopath Assoc. 1976;76(1):21-31. [PubMed]
Noll DR, Shores JH, Gamber RG, Herron KM, Swift JJr. Benefits of osteopathic manipulative treatment for hospitalized elderly patients with pneumonia. J Am Osteopath Assoc. 2000;100(12):776-782. [PubMed]
Noll DR, Degenhardt BF, Morley TFet al. Efficacy of osteopathic manipulation as an adjunctive treatment for hospitalized patients with pneumonia: a randomized controlled trial. Osteopath Med Prim Care. 2010;4:2. [CrossRef] [PubMed]
Hróbjartsson A, Gøtzsche PC. Is the placebo powerless? an analysis of clinical trials comparing placebo with no treatment [published correction appears in N Engl J Med. 2001;345(4):304]. N Engl J Med. 2001;344(21):1594-1602. [CrossRef] [PubMed]