ARTICLE IN BRIEF
Two research groups tried to determine whether vagotomy was associated with a reduced risk for Parkinson's disease. Experts who were not involved with the studies questioned the conclusions of the first report and the methodology of the second.
If Parkinson's disease pathology begins in the gut and slowly progresses to the brain via the vagus nerve, as is suggested by the Braak hypothesis, does a vagotomy reduce disease risk? Two different research teams mined the same national database to answer that hypothetical question — and reached very different conclusions.
One group found tentative evidence suggesting that those who underwent vagotomy had a reduced risk for Parkinson's, while a second study published shortly thereafter found no association. Both studies were published in the Annals of Neurology.
Experts who were not involved with the studies questioned the conclusions of the first study and the methodology of the second. The questions raised by the studies — whether vagotomy reduces the risk of Parkinson's, and whether the disease's pathology begins outside the brain — remain unaswered, they said.
TENTATIVE EVIDENCE FOR AN ASSOCIATION
First proposed by Heiko Braak, MD, now a professor of clinical neuroanatomy at Goethe University in Frankfurt, the Braak hypothesis posits that Parkinson's pathology begins in the periphery — for instance, in the nerves of the gut — and spreads misshapen alpha-synuclein in a prion-like manner along the vagus nerve up to the brain.
Elisabeth Svensson, PhD, a postdoctoral researcher in the department of clinical epidemiology at Aarhus University in Denmark, and colleagues sought to determine whether severing the vagus nerve would block one of the major hypothesized pathways for the transmission of alpha-synuclein, preventing the pathology from reaching the brain by that route and thus lowering the rate of Parkinson's.
The researchers examined data from the Danish National Patient Registry, a database the country maintains on all citizens, to explore whether people who had undergone a vagotomy as a treatment for peptic ulcers between 1977 and 1995 were less likely, years later, to develop Parkinson's disease (PD).
Two types of surgery were common for the treatment of peptic ulcers prior to the widespread use of proton pump inhibitors, Dr. Svensson and colleagues noted: full truncal, involving the severing of both vagal trunks, and superselective, in which only a single trunk is severed. The researchers compared the rates of Parkinson's among those who underwent either the full truncal or the superselective vagotomy, and matched both against a general population cohort.
Compared to patients who underwent superselective surgery, those who underwent truncal surgery had a lower unadjusted risk of developing Parkinson's after more than 20 years, although the 95% confidence interval crossed 1.0 (HR=0.58; 95% CI: 0.28-1.20). A barely significant effect was found only when comparing truncal surgical patients to the general cohort after more than 20 years (HR=0.53; 95% CI: 0.6328-0.99). Among superselective vagotomy patients, by contrast, no risk reduction was seen.
Dr. Svensson and colleagues pointed to the effect size, emphasizing that according to the hazard ratio, a full truncal vagotomy cut the risk of developing Parkinson's nearly in half. Patients who had both branches of their vagus nerve severed “had a clear reduction in the risk of a PD diagnosis, although the statistical precision of the estimates was limited and did not meet a statistically significant threshold,” they wrote in the July 2015 online edition of the Annals of Neurology ahead of the October print issue.
Critics of the study, however, emphasized that the wide confidence interval shows that the apparent risk reduction may be due purely to chance.
NO STATISTICAL ASSOCIATION
The second study, led by Ole-Bjørn Tysnes, MD, a professor and chief of neurology at Haukeland University Hospital in Norway, also analyzed data from the the Danish National Patient Registry, but for 16 years longer than the first study — from 1977 to 2011. The study, which was published online September 29, found no statistically significant correlation between truncal vagotomy and subsequent risk of Parkinson's.
“We believe that the main conclusion from both our data and from Svensson et al. is that it remains to be shown that vagotomy reduces the risk of having PD,” Dr. Tysnes and colleagues concluded. “In our study we included more cases, but the results are very similar to the data from Svensson et al.”
In a letter responding to the Tysnes paper published online in the Annals of Neurology on September 15, Dr. Svensson and colleagues pointed out that the statistical methods used in the two studies were different.
“Our approach included presenting effect sizes with appropriate indicators of uncertainty, in agreement with ICMJE [International Committee of Medical Journal Editors] recommendations. This avoids relying solely on statistical significance testing, which fails to convey important information about effect size. Since the effect sizes in the different analyses remained consistent, we stand by the conclusion that truncal vagotomy may be associated with a reduced risk of PD.”
Neurologists and biostatisticians who reviewed both papers told Neurology Today that they remain unconvinced that the study by Dr. Svensson provides compelling evidence one way or the other.
Dr. Svensson and colleagues “do have a point that effect sizes are important in the overall picture,” said Sue Leurgans, PhD, a professor of neurological sciences and preventive medicine at Rush University Medical Center in Chicago and a statistician on the Parkinson Study Group Scientific Review Committee.
But, she said, “the [ICMJE] guidelines are partly for the opposite situation, in which there is statistical significance even though the effect size is small. So they have evidence that there may be an effect, but the lack of precision indicates that the evidence is not solid. I think they are overstating the strength of their evidence when they say that full truncal vagotomy is associated with a decreased risk. The word ‘is’ rather than ‘may be’ is not something I would agree with.”
Dr. Svensson's group did not demonstrate statistical significance, she said. “At the very least, it is certainly appropriate to call the initial paper's claim controversial.”
Seoun Kim, PhD, an assistant professor in the department of biostatistics at the University of Texas Health Science Center in Houston, who also serves on the scientific review committee for the Parkinson Study Group, concurred.
“While effect sizes are considered important by Svensson and the study suggests the overall trend of lower risk,” he said, “the fact that this study is exploratory and underpowered weakens the strength of the positive findings.”
“This hypothesis that Parkinson's starts in the periphery, in the gut, and then moves through the vagus nerve to the brain is intriguing,” said Joseph Jankovic, MD, FAAN, a professor of neurology, distinguished chair of movement disorders, and director of the Parkinson's Disease Center and Movement Disorders Clinic at Baylor College of Medicine in Houston.
“Most Parkinson's neurologists believe there is something to it, and a number of people suggested that perhaps Parkinson's can be prevented by vagotomy. This first study suggested that possibility, but the second study, in my opinion, disproves it,” he said.
He added: “I hope the second study puts this whole idea to rest and reassures people that they don't have to have a vagotomy to prevent Parkinson's.”
Andrew Lees, MD, a professor and emeritus director of neurology at the Reta Lila Weston Institute of Neurological Studies at University College London, questioned whether the study by Dr. Svensson should have been published at all. “The Annals of Neurology has had a record of being attracted to epidemiological studies using large databases in Parkinson's,” he said. “While this is of course a valid approach, I have concerns about the reliability of the data, however scrupulous the attempts may have been to collect it.”
Taking the two studies together, Dr. Lees said, “My own feeling is that the right conclusion to be drawn for now is that there is no evidence that truncal vagotomy protects against the subsequent development of Parkinson's disease.”
While agreeing that no firm conclusions can be drawn from the initial study, Irene Litvan, MD, director of the movement disorders program and the Tasch Endowed Professor in Parkinson Disease Research at the University of California, San Diego, pointed out that the second study did not attempt to control for any of the factors that the Svensson study did, including comorbidities such as chronic pulmonary disease or rheumatological disease, which can indirectly relate to smoking or anti-inflammatories, two potential confounders.
“They did not even control for smoking,” said Dr. Litvan. “I wish they had. I don't think they have disproved the hypothesis, but I also don't think Svensson has proved it. Further research is needed.”
Despite the equivocal findings, the neurologists all agreed that the underlying Braak hypothesis remains alive and well.
“Braak drew attention to the fact that the initial pathology of Parkinson's does not occur in the substantia nigra,” said Dr. Jankovic. “It may start in the caudal brainstem and then spread into the midbrain, diencephalon, and cortical areas. That's still a hypothesis, but many people accept it as valid. The question is whether the alpha-synuclein pathology starts in the caudal brainstem or starts in the periphery and then gets into the brain.”
Some have suggested that the pathology begins in the epicardium, Dr. Jankovic noted. “Some think the epicardium might be a more important source of alpha-synuclein pathology than the gut,” he said. “This is hotly debated, and the jury is still out.”
EXPERTS: ON THE PROPOSED ASSOCIATION BETWEEN VAGOTOMY AND PARKINSON'S RISK