In the spirit of last week’s April Fool’s Day, I’d like to point out a paper recently published in THYROID from a Japanese group by Oda et al that’s been getting a lot of buzz: http://www.ncbi.nlm.nih.gov/pubmed/26426735
This paper’s premise is to track the surgical side effects and complications (including scarring) of thyroid cancer patients with papillary microcarcinoma (PMC) who have surgery and then compare that to the treatment side effects in thyroid cancer patients with PMC offered “observation” instead. Guess what the results were? [Spoiler alert:] Yes, the PMC patients who have surgery have more surgical complications than patients who don’t.
It’s natural to ask whether there was any other point to this paper. For example, did the authors actually evaluate the long-term complications that can occur when a biopsy-proven thyroid cancer that is not surgically removed metastasizes? No. The patients were only followed for a median of 4 years in the context of a disease where recurrence may not be seen for decades. And they were only followed with ultrasound and thyroid function tests, which is not sufficient to conclude whether patients eventually developed distant metastases (radioactive iodine cannot be used as a therapy or diagnostic tool in patients who do not have surgery). Among the patients who did have surgery, many had merely a lobectomy, which also precludes any radioactive iodine follow-up. So in the end, the claims that the “oncological” outcomes were similar in the patients who had surgery and those who did not, cannot be known and cannot be stated. Since all aggressive, poorly differentiated thyroid cancers start out in each person as a “microcarcinoma” you can’t predict which PMCs will spread to distant sites.
So it’s reasonable to conclude that a number of those unfortunate patients who had not received surgery could develop lung or bone metastasis that are not recognized and don’t receive any treatment or follow-up.
This begs the question as to whether this paper makes any contribution to science. The answer is No.
Studies with questionable heuristic value like this one are certainly in good company. The British Medical Journal, for example, published a study on sword swallowing, which found there were more side effects when the swallower is distracted or swallowing multiple or odd-shaped swords (See: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1761150/). It also published a study stating that people with acute appendicitis may feel pain when traveling over speed bumps (See: http://www.bmj.com/content/345/bmj.e8012).
The main problem with the Oda et al paper is that it’s not a joke. Not only did it make it through the peer review process into a good journal, but this paper could become an “Emperor has No Clothes” vehicle for proponents of observation.
So the ethical issue here is one of publication ethics and epistemic integrity. It’s not that a “Duh Study” was published. It’s the fact that it’s being misrepresented as having heuristic value when it’s really Fool’s Gold.
Note: This blog was expanded from an initial Letter to the Editor I submitted to THYROID on 3/29/16 entitled “What Heuristic Value?” co-authored by Kenneth Ain, MD. Professor of Medicine and Director, University of Kentucky Thyroid Oncology Program, and Peter Angelos, MD, Ph.D., Professor of Surgery and Chief, Endocrine Surgery, and Associate Director, MacLean Center for Clinical Medical Ethics. The letter was not accepted for publication.