We appreciate the comments, but wonder why this kind of debate is raised in Tidsskriftet (The Journal of the Norwegian Medical Association) rather than the journal which published the original

article. Bliksrud gives an inaccurate image of the findings, with partially uncritical use of sources and an incomplete description of a complex research field.

We have suggested the hypothesis that impaired pyruvate dehydrogenase function may play a role in chronic fatigue syndrome. We have not made any statements “without reservations”. The hypothesis is based on scientific arguments and will be tested thoroughly through further research. We have stressed that there is no indication of a structural defect in, nor a lack of, the pyruvate dehydrogenase enzyme in patients with chronic fatigue syndrome. We argue that our results may imply a dysregulation of the enzyme function. We have repeatedly emphasized that these findings at present do not justify new methods for treatment or diagnostics. How the media chose to report on the findings was evidently difficult to control.

We have made every effort to describe our results in a clear and precise manner in order to avoid misunderstandings. Terms like “pyruvate dehydrogenase deficiency” are misleading in this context, and it is unfortunate that Bliksrud applies this particular term in his comment, and even in the title. There is no indication of an enzyme deficiency. Rather, we have explained why we believe the results to be consistent with a partial impairment of the enzyme function, as a result of changes in the underlying regulatory mechanisms.

Pyruvate dehydrogenase is an essential enzyme in energy metabolism, and is carefully regulated both allosterically, post-translationally and through gene transcription, partly via energy-sensitive signaling pathways (1, 2). When Bliksrud narrows the enzyme’s role in disease mechanisms to what is seen in genetic primary pyruvate dehydrogenase deficiency, the basis for comparison is limited. It is not obvious that that such a systemic, permanent condition should be comparable to a more fluctuating, perhaps local and context dependent impairment of the enzyme function. In a number of conditions, pyruvate dehydrogenase is associated with other potentially pathogenic mechanisms. One example is primary biliary cirrhosis, which involves autoantibodies against components of the pyruvate dehydrogenase complex, and where fatigue is a characteristic symptom (3). It is also relevant to look at other conditions involving glucose metabolism and regulation of pyruvate dehydrogenase, such as diabetes and cancer (2), and also normal adaptation to variable metabolic and physiological circumstances (4). In the article, we have elaborated on how these mechanisms may play a part in chronic fatigue syndrome.

There are no published systematic studies suggesting increased blood lactate at rest in chronic fatigue syndrome, but many patients describe a sensation of rapid accumulation of lactate upon activity. The accumulation of lactate may occur at a considerably lower muscular effort than among healthy. Alanine levels normally increase in plasma as a result of muscular work, in parallel with lactate (1), but we anticipate such an increase to be considerably more permanent in genetic primary pyruvate dehydrogenase deficiency than in chronic fatigue syndrome, where the effects are context dependent and triggered by activity. Since alanine plays a specific role in transporting amino groups to the liver, from amino acid catabolism in the muscles, the serum level is expected to vary independently of changes in glucose catabolism. In order to limit possible misinterpretations, we therefore chose not to include alanine in the statistical analyses of the amino acid categories. Studies of both lactate and alanine in chronic fatigue syndrome should preferentially be performed using standardized protocols for physical exercise testing. We would also like to point out that measurements of citric acid cycle substrate consumption are used in laboratory investigations to detect mitochondrial pyruvate oxidation defects (5).

Direct measurement of pyruvate dehydrogenase activity would of course be relevant, but is complicated. A method developed to detect genetic primary dehydrogenase deficiency would not necessarily be suitable for measurement of reversible and contextual enzyme inhibition, which may be caused by altered phosphorylation state. Therefore, as shown in the article, we are working with living cell cultures exposed to serum from either patients with chronic fatigue syndrome or healthy controls, to assess central parameters of energy metabolism.

Although individual patients describe avoidance of certain foods, there are no systematic studies which show that patients with chronic fatigue syndrome, with a body mass index equivalent to that of the general population, have a significantly different diet compared to healthy controls. Surprisingly, Bliksrud refers to a case series describing four patients with known eating disorder who developed chronic fatigue syndrome. Needless to say, this article does not constitute relevant evidence of the dietary habits of this patient population in general, nor of our study sample. Nevertheless, as we have not performed a detailed dietary anamnesis, we cannot categorically rule out that dietary factors could play a role. However, as there are no known or expected systematic differences in dietary habits, the number of included patients (153) and controls (102) will counteract significant bias in the analyses.

We observed that overnight fasting led to reduced levels of several amino acids, and therefore included only non-fasting patients in the comparison (see supplementary tables in the article). The altered amino acid profile was primarily observed in non-fasting women with chronic fatigue syndrome as compared to healthy women. Some changes were highly significant, of moderate effect sizes, with differences in mean values of approximately 15%. Yet the values were within the normal ranges. Thus, there are no amino acid deficiencies, but the data rather reflect compensation mechanisms for an altered metabolism. In the same samples, we observed no differences in triglyceride levels which increase after meals, or free fatty acids which increase upon fasting (1). We therefore find it unlikely that differences in fasting state should explain the findings. Also, our results are consistent with other reports, including studies where only fasting individuals were included (see references in article).

Our study included more patients than previous metabolism reports. The changes in amino acid profiles could not be explained by disease severity or duration, age, BMI or physical activity level (see supplementary tables in article). Bliksrud mentions, without further specification, that immobilization leads to metabolic changes, and refers to a study of mRNA analysis of muscle biopsies after limb immobilization. There are more relevant and comparable studies of serum metabolites after immobilization, where the results do not correspond to our findings in patients with chronic fatigue syndrome (6, 7).

We have not found reports showing an amino acid pattern equivalent to what we observed in patients with chronic fatigue syndrome, in healthy subjects after exercise, inactivity or dietary changes, nor in other patient groups. That does not signify that the observed changes are exclusive for patients with chronic fatigue syndrome. We have not maintained that these changes are specific to chronic fatigue syndrome, and we entirely agree with Bliksrud that it would be interesting to investigate whether similar changes take place in other groups suffering substantial fatigue.

We hope we have conveyed a more nuanced image than reflected in Bliksrud’s comments, and we confirm that our hypothesis still stands.

Litterature
1. Berg JM, Tymoczko JL, Gatto GJ, Stryer L. Biochemistry. 8. utgave. New York: W.H. Freeman & Company, 2015.
2. Gray LR, Tompkins SC, Taylor EB. Regulation of pyruvate metabolism and human disease. Cell Mol Life Sci. 2014;71(14):2577-604.
3. Bjorkland A, Loof L, Mendel-Hartvig I, Totterman TH. Primary biliary cirrhosis. High proportions of B cells in blood and liver tissue produce anti-mitochondrial antibodies of several Ig classes. J Immunol. 1994;153(6):2750-7.
4. Zhang S, Hulver MW, McMillan RP et al. The pivotal role of pyruvate dehydrogenase kinases in metabolic flexibility. Nutr Metab (Lond). 2014;11(1):10.
5. Sperl W, Fleuren L, Freisinger P et al. The spectrum of pyruvate oxidation defects in the diagnosis of mitochondrial disorders. J Inherit Metab Dis. 2015;38(3):391-403.
6. Glover EI, Yasuda N, Tarnopolsky MA et al. Little change in markers of protein breakdown and oxidative stress in humans in immobilization-induced skeletal muscle atrophy. Appl Physiol Nutr Metab. 2010;35(2):125-33.
7. Kujala UM, Makinen VP, Heinonen I et al. Long-term leisure-time physical activity and serum metabolome. Circulation. 2013;127(3):340-8.