Full text at
http://annals.org.sci-hub.cc/aim/ar...nts-chronic-fatigue-syndrome-randomized-trial
Protocol is at
https://www.ncbi.nlm.nih.gov/pubmed/26438161
This seems a bit unusual, since it wasn't crowd-funded at all:
Primary Funding Source: Interleukin Foundation and an independent donor who wishes to remain anonymous.
CBT and GET get plugged, of course:
The only effective treatments are cognitive behavioral therapy and graded exercise therapy (3–5), although a substantial proportion of patients do not improve with these treatments (6).
It was double-blinded, including blinding the sponsor. This seems to suggest that the sponsor/donor funding the trial was involved in it in some capacity
Participants, all study personnel, and the sponsor were blinded to treatment allocation throughout the study.
Some mid-trial changes to the methodology, such as the exclusion of long-term CFS patients and allowing unspecified meds:
Recruitment started in July 2014 and ended in November 2015; follow-up was completed in May 2016. After approval of the original protocol but before the start of the trial, the number of patients was increased to 50 and a decision was made to exclude patients with illness lasting more than 10 years. During the trial, an amendment was made to allow medication use for less than 14 consecutive days for treatment of adverse events if the medication had no known interactions with anakinra.
Sounds like they changed their mind about an outcome measurement:
The published protocol mistakenly mentioned the number of accompanying CDC symptoms as a secondary outcome measure, whereas the original protocol did not specify those symptoms as a secondary outcome.
Patients were recruited using Fukuda. The abstract may have merely been unclear on that point. But the exception for the 10 year limit stinks a bit of doing contortions to find ways to exclude and include specific patients:
Patients were included if they met the CDC criteria (1, 2), were aged 18 to 59 years, had a maximum fatigue duration of 10 years or recent progression of symptoms, had a score of at least 40 on the fatigue severity subscale of the Checklist Individual Strength (CIS-fatigue), and had a score of at least 700 on the Sickness Impact Profile.
Only subjective outcome measurements were used:
Primary and secondary outcome measures were assessed using Web-based questionnaires, which were completed at baseline, 4 weeks, and 24 weeks. Questionnaires on fatigue and pain severity were completed weekly during the intervention and monthly during follow-up.
A diagram shows the exclusions. Indeed, the "10 years of fatigue unless it got worse recently" excluded exactly 1 patient. 4 were excluded for meds, and 12 for co-morbidity. The Beck Depression Inventory was used, which equates disability with depression - if you can't do something, it's taken as a sign of depression. So more disabled patients were likely excluded as a result.
They're saving the objective data for another day, a tactic also used by Dutch psychobabblers when they needed to bury null actometer outcomes:
Additional outcome measures (body temperature, heart rate, cytokine and cortisol concentrations, and microbiome) were collected and will be reported elsewhere.
This is blatant bullshit, and they cite to the study that proves it. The difference in inflammatory markers was happening at the 3-year point, not after 10 years:
We excluded patients with illness lasting more than 10 years because recent studies suggest an inflammatory pattern early in the course of CFS (15).
One of the few intelligent things they've said:
A possible explanation for our findings is that peripherally administered anakinra does not reach the brain in sufficient concentrations to have a biological effect.
The Anakinra group did have a larger improvement in SF36-PF of about 10 points compared to 5 points for the controls. It's also bizarre that every single endpoint score (Table 2) has a range for the standard deviations which is identical for both groups. Unless I'm missing something very basic, there seems to be a major problem with those figures.
The conclusion isn't as bad as I was expecting:
In conclusion, we found no clinically meaningful effect of anakinra on fatigue severity in patients with CFS. We believe that if IL-1 plays a role in CFS, blockade of IL-1 in the peripheral tissues, such as the neuromuscular compartment, has no effect. Future studies should focus on intracerebral processes in patients with CFS and on reducing inflammation with agents that reduce IL-1 activity in the brain (55, 56).