Fatigue in Inflammatory Bowel Diseases: Etiologies and Management

Fatigue is well known to occur in chronic inflammatory diseases, such as multiple sclerosis (MS) and systemic lupus erythematosus (SLE). Similarly, it occurs in patients with cancer [26, 27]. Fatigue is hypothesized to be mediated by inflammatory cytokines [28] and an activated Th1-immune system [29], primarily through the brain–gut axis. The brain–gut axis is the bidirectional pathway between the central and enteric nervous system, mediated by the HPA axis. Pro-inflammatory cytokines, in response to environmental or clinical stress, activate the HPA axis through secretions of corticotropin-releasing factor (CRF) from the hypothalamus. CRF then stimulates the adrenocorticotropic hormone (ACTH) from the pituitary gland, which then stimulates the release of cortisol, a stress hormone, from the adrenal gland [30]. Cortisol has systemic effects and can affect the brain, resulting in fatigue [30].

Multiple studies have shown that fatigue is associated with clinically active IBD. In a single-center cross-sectional study, it was discovered that 48.7% of the 187 patients with IBD suffered from fatigue. After adjustment for age and sex, fatigue was inversely associated with clinical and endoscopic but not histological remission [7]. In a larger study of the Inflammatory Bowel Disease in South-Eastern Norway (IBSEN) cohort, of the 440 patients with IBD who had disease for at least 20 years and completed the FQ, those with clinically active disease had higher fatigue scores than those with inactive disease (UC 17.1 versus 12.4, p < 0.001, and CD 17.5 versus 13.3, p < 0.001) [6]. In a study of patients with IBD in clinical remission, the number of memory T cells and neutrophils was significantly higher (p = 0.005 and 0.033, respectively), while that of monocytes was lower (p = 0.011) in those with fatigue (n = 55) compared with those without fatigue (n = 29), as measured by the CIS-fatigue scale. Additionally, the levels of inflammatory cytokines such as tumour necrosis factor alpha (TNFα), interferon-γ (IFNγ), interleukin-12 (IL-12), and IL-10 were significantly higher and IL-6 lower, suggesting differences in immunological activity between the two groups [29]. Furthermore, longitudinal data over 2 years from the Manitoba IBD cohort (n = 312) revealed that fatigue was strongly associated with inflammation [5]. In a study of 67 children with IBD, those in the lowest insulin-like growth factor I (IGF-1) z score quartile had significantly greater fatigue (p = 0.02), as well as higher levels of IL-10, IL-17A, IL-6, and IFNγ [31] implicating inflammatory pathways in fatigue pathogenesis. However, these studies did not determine endoscopic or biochemical remission, making it difficult to determine underlying IBD-related inflammatory burden accurately. One study (n = 288), in which fatigue was assessed during active disease and during deep remission, those in deep remission had no significant difference in immune markers, i.e., calprotectin, C-reactive protein (CRP), IL-17A, IL-6, IL-1β, IL-10, IL-8, TNFα, IL-13, IFN, IL-4, and IL-12, between those with fatigue and those without fatigue [9]. Thus, in quiescent disease, inflammation may not have a significant role in the pathogenesis of fatigue. However, active IBD with elevated cytokines, such as TNFα and IFNγ, as well as elevated fecal calprotectin, correlates with the severity of fatigue [5, 6, 9, 29, 32‐34].

Inflammation is associated with a greater catabolic state and increase in resting energy expenditure, which could contribute to fatigue. In a study of 75 adult patients with CD, those with active disease had higher resting energy expenditure (REE) than those in remission (28.8 ± 5.4 vs 25.9 ± 4.3 kcal/kg, p < 0.001) [35]. Furthermore, pro-inflammatory cytokines can lead to anorexia and a decrease in caloric intake [36], dysregulation of the HPA axis [37], and promote anxiety and depressive symptoms through the gut–brain axis in susceptible persons [38], which could be perceived as fatigue.

While most studies note an association of inflammation and fatigue, there is heterogeneity among published studies. In a study by Villoria et al., among 202 patients with IBD in clinical remission, there was no correlation between levels of CRP, IL-5, IL-8, and IL-12 and fatigue scores, although fatigue was prevalent in 54% of the cohort. However, in this study, endoscopic inflammation and calprotectin were not provided. The mean CRP in all three groups (none, mild, and severe fatigue) was normal and with a wide standard deviation [39]. A second prospective study of 544 patients with IBD found that fatigue was associated with anxiety, depression, and sleep disorders, but not associated with disease activity. This study, however, defined active disease in CD as a Harvey–Bradshaw index score greater than 4 points. This scoring system is based on factors including abdominal pain, abdominal mass, and complications such as arthralgia. It does not take into account objective measures of active disease, such as CRP, fecal calprotectin, endoscopic markers, of inflammation [40].

IBD therapy and control of inflammation is consistently associated with improvement in fatigue. In post hoc analyses of various randomized trials, treatment of CD with infliximab, adalimumab, certolizumab, or ustekimumab over placebo is associated with improvement in fatigue, as measured by the IBDQ, SF-36, EQ-5D, and FACIT [41‐44] and similarly in patients with UC, upon treatment with infliximab, adalimumab, golimumab, vedolizumab, or tofacitinib [45‐49]. Improvement in fatigue has also been seen in patients with UC who were treated with non-biologic therapies, such as immunomodulators and 5-aminosalicylates (ASA) [50]. Mucosal healing is associated with improvement in fatigue [51]. Therefore, patients with ongoing fatigue symptoms in clinical remission, but not endoscopic remission, may also benefit from therapy titrated to this target.

Recently, Cannabis sativa use amongst patients with IBD has also become more prevalent with statewide legalization or decriminalization [52]. There are at least 70 known cannabinoids, which are the active ingredients of the plant, and act upon the endocannabinoid system (ECS) [53]. The ECS has a role throughout the gastrointestinal tract, regulating food intake, nausea and emesis, gastric secretion, motility, intestinal inflammation, and cell proliferation [54]. Cannabis has improved IBD-related symptoms, such as abdominal pain, diarrhea, and anorexia [55, 56]. One study assessing cannabidiol in the symptomatic treatment of UC found that fatigue was a common side effect [57]. No studies have evaluated the effect of cannabis or cannabis derivatives on fatigue among patients with IBD.

Anemia occurs in up to 20% of ambulatory patients and up to 68% of hospitalized patients with IBD [34, 58], and is an important cause of fatigue. Iron deficiency anemia (IDA) is the most common anemia in IBD and can be due to chronic gastrointestinal bleeding and decreased nutritional intake [59]. Ongoing inflammation, mucosal ulcers, and mucosally fragile pseudopolyps can lead to IDA. Iron deficiency, without overt anemia, was not associated with fatigue in the Manitoba cohort of 280 patients [60].

Vitamin B₁₂ and folate deficiency can also be linked to weakness and fatigue [59]. Folate deficiency and resultant macrocytic anemia can occur in the context of methotrexate (MTX) use or malabsorption [61]. B₁₂ deficiency and resultant macrocytic anemia can occur in patients with ileitis, ileal resection [62], or small intestinal bacterial overgrowth (SIBO) [63]. Anemia of chronic inflammation can occur in those with ongoing inflammation and impaired iron absorption and metabolism. Hemolysis can occur as a rare complication of IBD, as well as in association with IBD therapies, such as sulfasalazine and 5-aminosalicylates [64, 65]. Additionally, anemia as a component of bone marrow suppression is associated with azathioprine (AZA) and 6-mercaptopurine (6-MP) use, especially in those with thiopurine S-methyltransferase (TPMT) deficiency or intermediate level [66].

Routine complete blood cell count (CBC), ferritin, and CRP measurement are recommended every 3 months in patients with active disease and every 6–12 months for those in remission. Folate and vitamin B₁₂ measurements are recommended annually or in the context of macrocytosis, according to the European Crohn’s and Colitis Organization [67]. TPMT activity assessment prior to and routine cell count measurement during immunomodulator therapy are also recommended [68]. Basic anemia workup includes red blood cell indices, reticulocyte count, serum ferritin, transferrin saturation, and CRP measurement [67]. Further workup is generally performed in consultation with a hematologist if there is any confusion regarding etiology of the anemia.

Early and aggressive treatment of iron deficiency with parenteral or oral iron is recommended to improve IBD outcomes. The former is recommended in those with active IBD to maximize systemic bioavailability and avoid the risk of gastrointestinal toxicity with oral iron [69]. In a study of 543 patients, it was discovered that of the different parenteral iron preparations available, ferric carboxymaltose is most effective and safe with a response rate of 79%, and with an adverse event rate of 12.0%, and one serious adverse event (pulmonary embolism) [70]. Oral iron is preferred in motivated patients with inactive disease. The treat-to-target approach can be useful for the long-term management of iron deficiency, meaning that iron treatment should be continued with the goal of normalization of predetermined markers such as ferritin and iron saturation [71].

Vitamin B₁₂ and folate should be supplemented if stores are discovered to be low. Folate supplementation of 1 mg/day is the recommended dose. Folate is also recommended in patients taking methotrexate or sulfasalazine and a common component in prenatal vitamins, although the dose is less at 400 μg/day [72, 73]. The recommended dose of B₁₂ if deficiency is present is more than 1000 μg daily orally [72]. However, the optimal method for vitamin B₁₂ supplementation is parenteral injections in those who have had more than 60 cm of ileum resected or have CD-related ileal inflammation resulting in disruption of ileal B₁₂ absorption; B₁₂ 1000 μg IM monthly is recommended for prophylaxis [74].

Macro- and micronutrient deficiency can occur in patients with IBD as a result of several factors including malabsorption, diarrhea, self-imposed dietary restrictions, or a catabolic state and can lead to fatigue [75]. Restrictive diets carry the risk of nutritional deficiencies [76], e.g., gluten-free foods can be deficient in calcium, iron, niacin, and thiamine unless fortified [77]. Patients with IBD have inadequate intake of calcium, folate, and iron-containing foods [78]. Deficiencies of vitamin B₆, vitamin B₁₂, folate, ferritin, and zinc have all been linked to increased disease activity, while vitamin D deficiencies have been found in both disease-active patients and those in clinical remission [32, 78]. However, vitamin D deficiency has been shown to have no significant association with IBD in fatigue patients [79]. Testing for and replacing the deficient nutrient and ensuring a nutritionally replete diet, generally in collaboration with an experienced dietician, could improve symptoms, although data are limited [32].

Less commonly, fatigue can be a direct medication-related adverse effect, e.g., due to AZA, 6-MP, or MTX rather than as a secondary mechanism of bone marrow suppression [32]. In a case series of five patients with IBD, discontinuation of 6-MP was associated with improvement in fatigue [80]. In addition, vedolizumab and infliximab both include fatigue as a potential side effect, occurring in 6% and 9% of patients, respectively. While steroids rapidly control inflammation and are associated with increase in energy and even insomnia, long-term use can cause myopathy and fatigue. Contrarily, adrenal insufficiency, manifested by severe fatigue, can occur with rapid steroid taper [81] and would warrant confirmation with early morning cortisol and ACTH measurement. Minderhoud et al. found no relationship of fatigue to clinically detectable adrenal insufficiency among a group of 80 well-described patients with IBD [82]. It will require higher power studies to elucidate if there is a correlation between fatigue and adrenal insufficiency in this population, suggesting that this avenue of investigation should be limited to those with high dose and prolonged steroid exposure.

Other medications, such as antidepressants and narcotics, more commonly prescribed to patients with IBD than the general population, can be associated with lethargy and somnolence [83]. In fact, narcotic use was associated with a significantly higher burden of fatigue in patients with IBD [84]. Cannabis use, especially by young patients with IBD, is associated with depressive symptoms [85] as well as a motivational syndrome, which may be interpreted as fatigue.

It is important to routinely review medication lists and consider withdrawing medications that could be associated with fatigue in the absence of other more likely explanation or after correction of identifiable risk factors. Fatigue in the presence of deep remission should warrant and a re-evaluation of potential culprit medications, including those used to treat IBD.

IBD is associated with dysbiosis and a decrease in the gut microbiome diversity, compared with healthy controls. In patients with IBD, there is a functional decreased composition of bacteroides and Faecalibacterium prausnitzii and an increased abundance of enteroaggregative Escherichia coli (EAEC) among many other changes to α/β diversity. Increased intestinal permeability from inflammation allows contents such as bacteria and bacterial metabolites to cross through the epithelial barriers into the systemic circulation [86]. These can act to incite the host immune system, resulting in systemic inflammation. Inflammatory cytokines increase permeability of the blood–brain axis, which may play a role in increased perception of fatigue symptoms. Evidence to support this hypothesis comes from studies performed on patients with chronic fatigue syndrome (CFS). Those with CFS have decreased stool bacterial diversity, similarly to those with IBD [87]. More recently, Borren et al., correlated a reduced abundance of Faecalibacterium spp. to patients with IBD with high fatigue and postulated that this affects the bidirectional flow of signals between the gut and the central nervous system, resulting in fatigue [87]. Interestingly, a recent systematic review showed that there were consistent changes in the microbiome of patients who responded to biologic agents, including infliximab, adalimumab, vedolizumab, and ustekinumab. There was a decrease in abundance of Escherichia and Enterococcus spp. and an increase in genera of short chain fatty acid producing species [88].

Our current strategies of microbial manipulation include antimicrobials, fecal bacteriotherapy, and probiotics. Antimicrobials have questions around their safety, specificity, and durability [89]. A systematic review of randomized control studies of probiotics found only support for a single formulation (VSL#3) in a narrow indication of preventing UC relapse. However, this formulation as originally studied no longer exists under that trade name, but has been rebranded as Visibiome [90]. At this point, there are microbial changes associated with disease activity and fatigue, but currently no narrowly targeted strategy exists to manipulate the microbiome to manage fatigue.

Sleep disturbances have been implicated as the cause of worsening symptoms including pain and fatigue in chronic inflammatory diseases. Poor sleep has become a keen area of research interest because it is thought to be an environmental factor that is closely linked to worsening IBD symptoms, specifically fatigue [91, 92]. Significant fatigue has been shown to be strongly correlated with poor sleep quality and daytime sleepiness [93]. Population-based sleep quality studies estimate that 32% of adults report poor sleep. Sleep deficiency is an even more common issue among patients with IBD with numerous authors including Graff et al. suggesting rates of problematic sleep in individuals with both active and inactive IBD to be greater than 50% [5, 32, 84, 93‐96].

Survey studies confirm the presence of sleep concerns in patients with IBD. A majority of these studies utilized validated questionnaires to assess subjective sleep quality including the Pittsburgh Sleep Quality Index (PQSI), the Patient-Reported Outcomes Measurement Information System (PROMIS), the Modified Fatigue Impact Scale (MFIS), or the Epworth Sleepiness Scale (ESS). Data from these studies was consistent and patients with IBD, regardless of whether disease was quiescent or active, reported poorer sleep than healthy controls [5, 91, 97‐104].

Subjective sleep assessments suggest that there is a strong association between IBD and sleep. Patients with IBD report worse sleep than healthy controls, and subjects with active IBD report poorer sleep than those in remission. Furthermore patients with IBD have inferior quality of life scores. In a case–control study, Ranjbaran et al. found that subjects with IBD identified prolonged sleep latency, increased fragmentation, use of sleeping aids, and decreased energy as factors associated with poorer overall sleep [91]. Another prospective cohort study corroborated the higher utilization of sleep medications in patients with IBD [100]. Factors limiting restorative sleep identified by study subjects in these surveys include pain from ongoing active disease and nocturnal symptoms. Other factors associated with poor sleep among patients with IBD include the use of steroids, narcotics, anti-TNF therapy, smoking, female sex, and depression [102, 105, 106]. Patients with IBD are vulnerable to the increasing obesity epidemic and are also at risk for obstructive sleep apnea (OSA) [107]. Interestingly a polysomnography study by Keefer et al. found that 13% of patients with IBD had OSA [108]. OSA prevalence is estimated to be around 20% in the overall population [109].

There is currently limited sleep data in patients with IBD through objective parameters. In the literature there are two studies using polysomnography and two studies utilizing wrist actigraphy. Actigraphy involves collecting data from a sensor worn on the non-dominant wrist during sleep. Computer-based algorithms help analyze sleep characteristics [110]. Although polysomnography remains the gold standard for sleep studies, wrist actigraphy has some advantages because it can be done in the patient’s home environment, it is cost-effective, noninvasive, and less burdensome thus data can be collected over a longer period of time [110].

Keefer and colleagues were the first to perform a small prospective study using both subjective (PSQI) and objective (PSG) measures of sleep disturbances in healthy adult controls, patients with IBS, and patients with IBD [108]. In comparing the PSQI to PSG, self-reported hours of sleep were significantly correlated with PSG total sleep time. Subjects with IBD had poorer sleep quality by PQSI survey as well as polysomnography, when compared to healthy controls. However interestingly patients with IBD did not seem to score differently from patients with IBS on objective or subjective sleep parameters. Interestingly, these patients all had quiescent disease, suggesting that some other factor may be responsible for the experience of poorer sleep.

Bar-Gil Shitrit et al. also found similar differences between patients with IBD with inactive disease and healthy controls in a polysomnography study [111]. They performed overnight polysomnography on 36 patients with inactive IBD and 27 healthy controls. The results showed a lower percentage of rapid eye movement (REM) sleep in subjects with IBD compared to healthy controls and more oxygen desaturation events not related to OSA.

Recently there was a small study that used the Fitbit Charge HR, a commercially available sleep monitor [106]. Data from wrist actigraphy studies in patients with IBD and polysomnography studies among patients with IBD were found to have longer sleep latency and lower sleep efficiency [97]. Additionally patients with IBD had greater numbers of nighttime awakenings [95]. In a study using the Fitbit charge HR, Sofia et al. found that sleep fragmentation was associated with increased chances of having clinically active disease [104].

Sleep and the intricate relationship between sleep and IBD are a keen area of research interest because they are now thought to be an important environmental factor that may contribute to disease flares. There is good evidence about the link between sleep deprivation and the activation of the immune system, and the ability to mount an immune response [112]. In a meta-analysis of experimental sleep deprivation studies, sleep disturbances were associated with increases in CRP and IL-6 [113]. The largest studies that show a relationship between altered sleep patterns and inflammatory disorders are in shift workers. In one study, male electricians (OR 1.7) and male food and beverage industry workers (OR 1.6) had a statistically significant increased propensity for developing UC and CD, which was thought to be related to irregular hours and interrupted sleep patterns [114]. Patients with IBD with poor sleep have increased rates of disease relapse, and lower sleep quality scores are predictive of subclinical histological inflammation and disease relapse at 6 months [101]. In another study, patients with CD in remission had a twofold increased risk of disease recurrence at 6 months; however, this was not seen in patients with UC [102]. Ananthakrishnan et al. also found a higher incidence of UC in individuals with less than 6 h of sleep or more than 9 h of sleep [105].

There is basic science data that supports the observed relationship between sleep deprivation and increased inflammatory states. In experimental models of rodents with colitis and sleep deprivation, sleep deprivation worsened the degree of dextran sulfate (DSS)-induced colitis in animals [92]. Furthermore sleep elevations in pro-inflammatory markers (IL-6, TNFα, and CRP) in response to sleep deprivation have been seen in animal and human models [113, 115‐118]. Interestingly administering IL-6 and TNFα to animals resulted in suppression of REM [119]. In humans, administration of endotoxins increased TNFα, IL-6 levels, and non-REM sleep and wakefulness [120].

Poor sleep quality and resulting fatigue in the IBD population have implications for practitioners. Sleep deprivation is a modifiable risk factor and thus we need to incorporate sleep assessments more routinely for the care of our patients with IBD. Here we describe a data-supported strategy to address this issue in our patients.

Through induction of remission, gastroenterologists can significantly improve quality of sleep. Considering the effect of inflammation on sleep, it is intuitive that by decreasing disease activity, significant improvement in sleep can be achieved. Patients treated with vedolizumab and anti-TNF medications showed significant improvement in sleep survey scores with reduced nocturnal diarrhea and less abdominal pain [121]. It is also important for clinicians to identify iatrogenic risk factors for sleep impairment and modify treatment regimens. Effective interventions include cessation of steroid and narcotic use. Additionally given the association between sleep quality and psychiatric comorbidities, there is a need to screen for these conditions and refer for treatment [91].

When sleep quality is of concern, cognitive behavioral therapy (CBT) for sleep has been shown to be effective as a first-line intervention, and may have an important role in the care of patients with IBD [98]. Education about appropriate sleep hygiene is also important [122].

Recently cannabis has generated a lot of interest as an option for IBD symptom control. Furthermore many patients with IBD use cannabis as a complementary therapy for symptom control [52]. Patients with IBD report improved sleep with cannabis use, although this conclusion is based on a few small studies. In a study of young adults with IBD, disease activity scores for cannabis users were similar to non-users; however, there was a perceived positive impact on sleep quality reported by patients with IBD [123]. In another placebo-controlled study, IBD cannabis users reported significant improvement in sleep [56].

Melatonin, a natural hormone released by the pineal gland in response to the dark, shows some promise. Animal models show some decrease in inflammatory markers [124]. Rats who had induced colitis subsequently treated with melatonin showed improvement in the colitis [125, 126]. However, given limited human trial data, we do not routinely recommend this for our patients.