Published online 2017 Jan 21. doi:
10.1186/s12974-017-0796-7
Interleukin-1 as a mediator of fatigue in disease: a narrative review
Chronic fatigue syndrome (CFS) is a condition of unknown origin that is characterized by the presence of severe fatigue for a duration of at least 6 months, next to several accompanying symptoms such as headaches, sore throat, and muscle and joint pain [
116]. Over the past decades, CFS has been attributed to a range of different causes, but a unifying cause has not been found. Even if a distinct abnormality is found repeatedly, for example relative hypocortisolism [
117], it is difficult to determine whether this is a causative factor or rather an epiphenomenon as a consequence of inactivity, depressive symptoms, sleep problems, etc. Perhaps, more than any other chronic disease associated with fatigue, cytokines have been measured by several investigators. A relationship between IL-1 and fatigue severity has often been assessed. The studies reveal a large heterogeneity, not only with respect to patient characteristics but also with respect to selection of controls and sample handling. In addition, there is a large variation in questionnaires used to measure fatigue dimensions and fatigue-related symptoms. These issues make it difficult to draw reliable conclusions.
A systematic review focusing on circulating cytokines in CFS was published recently by Blundell et al. [
118], who reviewed all studies published on this subject between the publication of the first CFS case definition in 1988 [
119] to March 2015. All 38 studies measuring circulating cytokines in diagnosed CFS patients compared to controls were included. As mentioned earlier, there were large differences with respect to recruitment of controls, sample handling, and exclusion of concomitant diagnoses. IL-1α was measured in 11 of the described studies, 27% of studies found increased concentrations, and 73% found no significant differences. IL-1β was determined in 28 studies, with only 25% reporting increased concentrations; the other studies did not find any significant differences. One of the more recent studies included in the review also discriminated patients with a short duration of illness (≤3 years,
n = 52) from patients with a long illness duration (
n = 246) and controls (
n = 348) [
120]. It appeared that patients with a short duration of illness had significantly higher IL-1α and IL-1Ra concentrations than controls. This was also found when comparing IL-1β levels in patients with short versus long illness duration. IL-1β appeared to be elevated in patients with a short duration and decreased in patients with a long illness duration (when compared to controls). After this extensive review of the literature by Blundell, three more studies on circulating cytokines were published [
121–
123]. A study by Russell et al. also tried to discriminate between patients with different illness durations [
123]. Comparing IL-1 concentrations, no differences could be found, although it has to be noted that patients with a “short” illness duration had been fatigued for a mean of 7 years, which is longer than the study mentioned earlier. In the linear classification model, however, IL-1α appeared to have predictive value in recently ill adolescent patients. Hardcastle et al. compared severely ill, house-bound patients (
n = 19), to moderately ill patients (
n = 22) [
121]. Although groups are rather small, IL-1β was significantly elevated in the moderately ill patients (
p = 0.002). There were no differences for IL-1Ra. The third study could not find any differences between patients and controls for either IL-1β or IL-1α in a group of 100 patients and 79 controls [
122]. We conclude from the literature that there is limited evidence for increased circulating IL-1 in CFS patients, although there might be a more pro-inflammatory pattern in those with a short illness duration [
120].
The effect of physical exercise on circulating cytokines was discussed in a separate systematic review [
124], although some of the studies discussed were also included in the review by Blundell [
125–
129]. The conclusion of this review is that also after exercise of varying intensity, there are no consistent differences with respect to IL-1β.
Another approach is to compare cytokine production capacity of PBMCs after stimulation between CFS patients and controls. An early study reported increased IL-1β production after LPS stimulation in a small group of CFS patients (
n = 9) compared to controls (laboratory personal,
n = 7) [
130]. Swanink et al. recruited neighborhood controls and found the opposite: lower LPS-induced IL-1β concentrations in patients (
n = 76) than in controls (
n = 69), with a large overlap between concentrations of cytokines [
131]. Lower IL-1β and IL-1α production after PHA stimulation was also reported by Mawle et al. in patients with a gradual onset of symptoms; no differences were observed when those with a gradual and acute onset analyzed together [
132]. A fourth study by Cannon et al., published in the same period, investigated IL-1β production in women during different phases of the menstrual cycle [
133]. In controls, spontaneous IL-1β production by PBMCs increased during the luteal phase, which already has been observed in healthy subjects many years ago [
134]. However, this could not be found in CFS patients. One recent study reported no differences between CFS patients and controls [
135].
IL-1β production by PBMCs in relation to fatigue has also been studied during the acute phase of an infection [
136] and in the phase of persisting symptoms [
137]. During the acute phase, the IL-1β concentration correlated significantly with fatigue symptoms; however, this relationship disappeared in the persistent phase. The perpetuation of fatigue symptoms in the absence of peripherally increased cytokine concentrations suggest that other, most likely central mechanisms, may be involved in persistent fatigue after an acute infection.
With the brain as the suspected target organ for immunological dysregulation in CFS, a limited number of studies measured cytokine concentrations in cerebrospinal (CSF) fluid of patients.
The first study, performed in 1991 by Lloyd et al., found no differences in IL-1β concentrations between patients and controls [
138].
Others had similar findings, and both IL-1α and IL-1β tended to be below the detection limit [
139,
140]. A more recent study compared 32 CFS patients to 40 patients with multiple sclerosis (MS) and 19 controls [
141].
CFS patients had lower CSF concentrations of both IL-1β and IL-1Ra compared to the MS and control group. When CFS patients were compared with MS patients only, IL-1α levels were also significantly lower.
Instead of creating more insight into pathological mechanisms in CFS, the described studies tend to raise more questions with respect to the role of IL-1 in CFS. It could be that disturbances of IL-1 signaling are only present in certain groups, for example only in those patients with short illness duration or those who experience fatigue after an infection, instead of when all patients are considered together.
One possible way to elucidate the role of IL-1 in CFS is to investigate the effect of blocking IL-1 on fatigue severity in CFS patients
Michael B. VanElzakker*, 
Sydney A. Brumfield and 
Paula S. Lara Mejia
