Psychological and hematological factors associated with fatigue in patients with Crohn’s disease receiving pharmacological treatment

Abstract

BACKGROUND

Fatigue is debilitating and costly for patients with Crohn’s disease (CD) and the healthcare system. Thus, as there are no well-established therapies for fatigue in patients with CD, it is essential to investigate its risk factors and collaborate to prevent or reduce its burden.

AIM

To evaluate the variables associated with fatigue in outpatients with CD receiving pharmacological treatment.

METHODS

It is an observational study. Data, including sociodemographic information, inflammatory bowel disease fatigue scores, visual analog scale scores, Depression Anxiety Stress Scale-21 scores, insomnia assessment results, and laboratory test results, were collected. Statistical analyses included Student’s t tests, logistic and linear regressions, and receiver operating characteristic curve analysis, with a P value of < 0.05 indicating statistical significance.

RESULTS

One hundred patients were included (77% presented with fatigue). In the linear regression analysis, symptoms of depression, anxiety, stress, and insomnia were simultaneously included as predictors of fatigue. Although the model was statistically significant (adjusted R² = 0.128; P = 0.002), no single symptom was significantly associated. A composite score (0-4) was developed by summing the scores of the 4 symptoms, which were coded dichotomously (odds ratio = 2.60; 95% confidence interval: 1.61-4.83; P < 0.001). The composite score showed good discriminative capacity (area under the curve = 0.775). Patients with fatigue had higher total leukocyte (P = 0.034) and segmented neutrophil (P = 0.017) counts and lower lymphocyte (P = 0.019) and eosinophil (P = 0.036) counts. Effect sizes ranged from moderate to high (Cohen’s d 0.39-0.61), indicating that fatigue may be associated with a leukocyte pattern consistent with relative neutrophilia and lymphopenia.

CONCLUSION

Fatigue is associated with psychological symptoms, a sedentary lifestyle, and alterations in leukocyte subpopulations. Assessments incorporating composite symptom scales and hematological parameters may be practical and cost-effective for patients with fatigue.

Key Words: Inflammatory bowel disease; Crohn’s disease; Fatigue; Anemia; Immune system; Insomnia; Leukocyte profile

Core Tip: Fatigue is one of the most common and challenging symptoms reported by adults with inflammatory bowel disease. We evaluated the association between fatigue and sociodemographic profile, depression, anxiety, stress, insomnia, iron deficiency, vitamin B1 and B12 deficiencies, thyroid function, anemia, leukocyte count, and sedentary lifestyle in patients with Crohn’s disease (CD). An association was found with the following variables: Sedentary lifestyle, shorter time since CD diagnosis, combination of depression, anxiety, stress and insomnia, total leukocyte and segmented neutrophil levels, and reduced lymphocyte-eosinophil-lymphocyte ratio. These data may contribute to improving decision-making in the management of fatigue in patients with CD.

INTRODUCTION

Inflammatory bowel disease (IBD) is a chronic inflammatory condition resulting from the interaction between genetic and environmental factors that influence immune responses. IBD affects seven million individuals worldwide and often manifests as abdominal pain, diarrhea, and rectal bleeding, leading to irreversible damage and bowel dysfunction. The main forms of IBD are ulcerative colitis (UC) and Crohn’s disease (CD)[1,2]. CD is a chronic inflammatory disease that can cause lesions in various locations from the mouth to the anus and may also present with several extraintestinal manifestations[3]. IBD is associated with a substantial impact on quality of life; impaired work productivity; and increased risks of surgeries, hospitalizations, fatigue, pain, stress, anxiety, and depression[4].

Fatigue is among the most common symptoms reported by adults with IBD, affecting approximately 80% of IBD patients with active inflammation and up to 50% of those in remission[2]. With an etiology that is likely multifactorial and related to inflammation, anemia, nutrient deficiencies, medications, and the intestinal microbiome, fatigue is more common in patients with CD than in those with UC[5,6]. Fatigue is debilitating and burdensome for patients and the healthcare system, with no well-established, effective therapies to date.

To prevent fatigue and reduce its burden on patients with CD, it is important to investigate its associated risk factors[7,8]. Previous studies have revealed that sleep disturbances, anxiety, depression, anemia, systemic steroids, disease activity, female sex, a shorter duration of IBD, and IBD-related surgeries may be risk factors associated with fatigue in patients with IBD[9-11]. Additionally, several other factors, such as iron deficiency, low serum levels of vitamin D and magnesium, sleep disturbances, alcohol abuse, and emotional stress, as well as immune system activation with the release of proinflammatory cytokines, have been identified as contributors to fatigue in patients with IBD. Owing to the underlying complexity of the factors contributing to fatigue, treatments are often empirical, and symptom resolution can be challenging. Although common, fatigue is often underreported and untreated, despite the need for systematic assessment in patients with IBD[12,13].

To date, few studies involving IBD patients have measured fatigue as a primary outcome. Studies focusing on variables associated with fatigue in CD patients are even scarcer, especially in Latin American countries[14]. Therefore, findings from studies investigating the factors contributing to fatigue in patients with CD can not only help prevent or reduce the burden of fatigue on patients and the healthcare system, but also improve the quality of life of these patients. Hence, the aim of this research was to evaluate the variables associated with fatigue among outpatients with CD receiving pharmacological treatment.

MATERIALS AND METHODS

Type of study

This was a cross-sectional observational study conducted at the IBD outpatient clinic of the Hospital Geral Roberto Santos. This unit is a reference facility for IBD in Bahia (a state located in Northeast Brazil). This study was approved by the Ethics and Research Committee of Hospital Geral Roberto Santos (Approval No. 61349222.0.0000.5028), and all procedures were conducted in accordance with the 1964 Declaration of Helsinki and its subsequent amendments or comparable ethical standards. The participants had a confirmed diagnosis of CD, according to the guidelines of the European Crohn’s and Colitis Organization and were receiving pharmacological treatment for CD[15].

Sample, inclusion, and exclusion criteria

These participants were invited to participate in the study while waiting for a medical appointment at the IBD outpatient clinic of the Hospital Geral Roberto Santos. This study included data collected between January 4, 2023, and May 14, 2024. The inclusion criteria were as follows: Full autonomy to decide to participate, a confirmed diagnosis of CD, aged 18 years or older, and receiving pharmacological treatment for CD. The exclusion criteria were as follows: Any condition that restricted the participants’ ability to respond to the questionnaires, such as intellectual disability, Alzheimer’s disease, dementia in Parkinson’s disease, dementia in Huntington’s disease, anterograde amnesia, retrograde amnesia, delirium or other cognitive alterations that prevent understanding of the research; serious underlying comorbidities; and pregnancy. One hundred consecutive participants were included. Participant selection was non-probabilistic. The sample consisted of individuals awaiting care at the gastroenterology outpatient clinic of Hospital Geral Roberto Santos who, after being invited, agreed to participate in the study.

Data collection

A questionnaire was used to describe the sociodemographic profile of the research participants. This questionnaire included questions concerning age, sex, self-identified skin color (based on the Brazilian Institute of Geography and Statistics), employment status, education level, marital status, family income, and current level of physical health activity.

The Montreal classification (age at diagnosis, location, and behavior) and the presence of extraintestinal manifestations were recorded[16]. The Havey-Bradshaw index was used to assess clinical disease activity at the time of enrollment. An Havey-Bradshaw index scores greater than 4 was considered indicative of active CD. Fatigue was assessed using the IBD Fatigue (IBD-F) scale and the Visual Analog Fatigue Scale[17,18]. Symptoms of depression, anxiety, and stress were assessed using the Depression, Anxiety and Stress Scale-21[19]. A standardized questionnaire was used to assess the presence of insomnia.

The Visual Analog Fatigue Scale is a unidimensional scale that measures the intensity of fatigue. Higher scores represent greater severity/intensity of fatigue[17]. The IBD-F scale is composed of items generated specifically from issues reported by people with IBD who experience fatigue and has been found to be valid and reasonably reliable during initial testing[18]. The Brazilian Portuguese version of the IBD-F has excellent internal consistency and adequate measurement properties. Its use is recommended in clinical practice and in research involving patients with IBD-F[20]. In this study, the cutoff point for the IBD-F was 1. A calculator from Crohn’s and Colitis (United Kingdom), a registered charity in England and Wales (https://crohnsandcolitis.org.uk/), was used. Binary analysis was used to assess the results (presence or absence of fatigue symptoms).

The Depression Anxiety and Stress Scale consist of 3 subscales with questions that assess symptoms of depression, anxiety, and stress. Symptoms are characterized as follows: Depression, which is characterized mainly by the loss of self-esteem and motivation and is associated with the perception of a low probability of achieving life goals that are meaningful to the individual as a person; anxiety, which highlights the links between persistent states of anxiety and intense fear responses; and stress, which is characterized by states of persistent excitement and tension and is accompanied by a low level of resistance to frustration and disappointment[19]. This scale has a version in Brazilian Portuguese[21]. In the Depression, Anxiety and Stress Scale-21, symptoms of depression, anxiety, and stress are divided by intensity.

Blood was also collected for biochemical analysis of the following parameters: Complete blood count, C-reactive protein (CRP), erythrocyte sedimentation rate, serum iron, ferritin, transferrin saturation, serum levels of vitamin B1, free T4, thyroid-stimulating hormone (TSH) and vitamin B12 (cobalamin). These tests were performed within 15 days after the questionnaires were completed. All tests were performed by the same laboratory.

Statistical analysis

Statistical analyses were performed using IBM SPSS statistics (version 21) and R studio (version 2024.12.1) software. Initially, comparisons between groups with and without fatigue were performed using Student’s t test or nonparametric tests, with calculation of the effect size (Cohen’s d). A P value less than 0.05 was considered to indicate statistical significance. Bivariate and multivariate logistic regressions were conducted to explore the association between dichotomous psychological symptoms and fatigue. Because multicollinearity increases SEs and widens confidence intervals, thereby weakening the significance of individual predictors[22], the use of a combined metric was methodologically appropriate. Accordingly, we constructed a composite symptom score (0-4) by summing the dichotomized presence of each psychological symptom. This score was used in a logistic regression model, with performance evaluated using the receiver operating characteristic curve, area under the curve, sensitivity, specificity, and ideal cutoff point (Youden index). Finally, laboratory data were included, with comparisons of the means and their associations with fatigue.

RESULTS

Initially, 101 patients were selected for the study. Of these, one was excluded from the analysis because of her unwillingness to undergo laboratory tests, resulting in a total convenience sample of 100 outpatients. Most patients were female (62%), with a mean age of 41.2 years (SD 13.32) and a mean disease duration of 104 months (SD 83.75). Only 3% of the patients were smokers, 5% had a diagnosis of controlled asthma, and the other 5% had a diagnosis of mild diabetes mellitus. With respect to the Montreal classification, CD was diagnosed before age 16 in 8% of patients, between the ages of 17 and 40 in 68% of patients, and after age 40 in 24% of patients. With respect to the affected site, L2 (colon) was the most common site (41%), followed by L3 (ileum-colon) (26%), L1 (terminal ileum; 21%), L1 + L4 (5%), and L4 (upper gastrointestinal tract; 2%). The predominant disease behavior was B1 (nonstricturing, nonpenetrating) in 38% of patients, followed by B2 (stricturing) in 25%, B1p (nonstricturing, nonpenetrating and perianal) in 12%, B3 (penetrating) in 10%, B2p (structuring associated with perianal) in 7%, and B3p (penetrating and perianal) in 5%.

Prevalence and variables associated with fatigue

Overall, 77% of patients experienced fatigue, according to the IBD-F. The mean fatigue score on the visual analog scale was 5.60 (SD 1.91). Patients with and without fatigue did not differ in terms of mean age. On the other hand, among the participants with fatigue, a significant majority were female (Table 1). The most commonly used medication for the treatment of CD was infliximab (49%), followed by azathioprine (36%) and adalimumab (25%). The mean disease duration in the total sample was 104 months (SD 83.75). Notably, the mean duration of CD was 96.92 months (SD 74.92) for patients with fatigue and 128.04 months (SD 106.75) for those without fatigue (P = 0.018). Conversely, no association was found between fatigue and Montreal classification in the studied population.

Table 1 Association of fatigue with sociodemographic characteristics of Crohn’s disease patients treated at a referral center, n (%).

Variable	Total (n = 100)	With fatigue (n = 77)	No fatigue (n = 23)	P value
Gender				0.086
Male	38 (38.0)	25 (33.0)	13 (56.0)
Female	62 (62.0)	52 (67.0)	10 (44.0)
Marital status				0.638
Single/widowed/divorced	66 (66.0)	52 (67.0)	14 (61.0)
Married	34 (34.0)	25 (33.0)	9 (39.0)
Education				0.722
No education-completed secondary education	20 (20.0)	16 (21.0)	4 (18.0)
Incomplete higher-completed higher education	80 (80.0)	61 (79.0)	19 (82.0)
Occupation				0.389
Signed employment contract/public servant	62 (62.0)	46 (60.0)	16 (69.0)
Self-employed/retired unemployed/student	38 (38.0)	31 (40.0)	7 (31.0)
Family income				0.572
Up to 2 minimum wages	70 (70.0)	55 (70.0)	15 (65.0)
Above 2 minimum wages	30 (30.0)	22 (30.0)	8 (35.0)

Among the 100 patients, 45 (65%) engaged in physical activity regularly, of which 30 (66%) experienced fatigue and 15 (34%) experienced no symptoms of fatigue. Furthermore, among the 35 patients (35%) who did not engage in physical activity, 27 (78%) experienced fatigue and 8 (22%) experienced no symptoms of fatigue (P = 0.026). Among the 100 patients included, 79 (79%) were in clinical remission and 21 (21%) had active disease. Among the 77 patients with symptoms of fatigue, 60 (78%) were in clinical remission and 17 (22%) had active disease. Among the patients without fatigue, 19 (83%) were in clinical remission and 4 (17%) had active disease. There was no statistically significant difference in disease activity between the groups with and without fatigue. Among the 100 patients, 39% underwent CD-related surgeries, of whom 80% presented with fatigue and 20% did not present with the symptoms of fatigue (P = 0.637). Similarly, 59% of the individuals had a previous history of CD-related hospitalization, of whom 76% presented with fatigue and 24% did not present with fatigue (P = 0.835).

Predictors of fatigue

The regression model showed good overall fit and calibration, explaining 26.9% of the variance in fatigue [overall model quality: χ² = 19.53 (P = 0.001), Nagelkerke R² = 0.269, Hosmer-Lemeshow P = 0.927, accuracy 77%]. However, its ability to distinguish between patients with and without fatigue was limited, which is expected given the overlapping nature of the psychological variables (Table 2). The classification matrix revealed 100% sensitivity but no specificity, suggesting that the model tends to overestimate the presence of fatigue. These findings, combined with the wide intervals and the lack of individual significance, reinforce the hypothesis of collinearity between the symptoms.

Table 2 Comparison of symptoms of depression, anxiety, stress and insomnia with the presence of fatigue in Crohn’s disease patients treated at a reference center.

Variables	OR [Exp(β)]	P value	95%CI
Depression	2.37	0.333	0.41-13.54
Anxiety	3.43	0.158	0.62-19.00
Stress	2.29	0.358	0.39-13.37
Insomnia	2.44	0.146	0.73-8.12

OR: Odds ratio; CI: Confidence interval.

When depression, anxiety, stress, and insomnia were included simultaneously in the multivariable model, formal multicollinearity diagnostics were performed following recommended practices for regression modeling. Variance inflation factors, calculated from the initial model, ranged from 1.2 to 1.6, which are well below conventional thresholds for severe multicollinearity (e.g., variance inflation factor ≥ 5-10) but are nonetheless consistent with mild to moderate shared variance among predictors. Condition indices and variance decomposition proportions similarly suggested overlapping contributions, particularly between stress and insomnia, indicating that these symptoms were not fully independent and that their inclusion could inflate coefficient variances and reduce interpretability.

The composite score was strongly associated with fatigue (odds ratio = 2.60; 95% confidence interval: 1.61-4.83; P < 0.001), indicating that each additional symptom increased the odds of fatigue by 2.6-fold. As shown in Figure 1, this score also demonstrated good discriminative ability (area under the curve = 0.775). The optimal cutoff determined by the Youden index corresponded clinically to ≥ 2 symptoms, yielding a sensitivity of 54.6% and a specificity of 91.3%.

Figure 1 Receiver operating characteristic curve of the composite symptom score (0 to 4 - depression, anxiety, stress and insomnia and association with fatigue) in patients with Crohn disease treated at a referral center. Receiver operating characteristic curve analysis - symptomatic score, discriminative performance: Area under the curve: 0.775. Interpretation: (1) Between 0.7 and 0.8 to good discriminative capacity; (2) The model has a 77.5% chance of correctly classifying a random pair of individuals (one with fatigue and one without); and (3) This indicates that the score is useful for distinguishing individuals with and without fatigue with good accuracy. Previously determined at probability ≥ 0.827, which corresponds to approximately score ≥ 2. At this point, the performance was sensitivity: 54.6% and specificity: 91.3%. ROC: Receiver operating characteristic curve; AUC: Area under the curve.

The analysis of relative risk across symptom categories revealed an exponential gradient: Compared with individuals without psychological symptoms, those with two symptoms had a 6.7-fold greater risk of fatigue, those with three symptoms had a 17.5-fold greater risk, and those with four symptoms had a 45.4-fold greater risk.

Table 3 presents the association between the composite symptom score and fatigue, indicating that each additional symptom increases the odds of fatigue by 2.6-fold. Table 4 expands this analysis by providing the estimated logit and odds ratios for each symptom-count category (0 to 4), illustrating the exponential increase in fatigue risk as the number of psychological symptoms increases.

Table 3 Association of fatigue with compound psychological/mental symptoms (depression, anxiety, stress and insomnia) in patients with Crohn’s disease treated at a referral center.

Predictor variable	β (coefficient)	OR [Exp(β)]	95%CI	P value
Symptomatic score	0.9538	2.60	1.61-4.83	< 0.001

Logistic regression model (n = 100); OR: Odds ratio; CI: Confidence interval.

Table 4 Association of fatigue with compound psychological/mental symptoms (depression, anxiety, stress and insomnia) in patients with Crohn’s disease treated at a referral center.

Number of symptoms	Estimated logit (βx)	OR (vs score 0)
0	0.000	1.00
1	0.954	2.60
2	1.908	6.74
3	2.861	17.49
4	3.815	45.39

The chance of fatigue increases exponentially with the number of symptoms. Odds ratio by number of psychological symptoms (n = 100). OR: Odds ratio.

The Leukogram analysis revealed a significant association between the presence of fatigue and changes in different leukocyte subpopulations in patients with CD. Patients with fatigue presented higher levels of total leukocytes (P = 0.034) and segmented neutrophils (P = 0.017) and reduced levels of lymphocytes (P = 0.019) and eosinophils (P = 0.036). Effect sizes were calculated using Cohen’s d; negative values indicate higher means in the fatigue group. Patients with fatigue had higher total leukocyte and neutrophil percentages and lower lymphocyte and eosinophil percentages, findings consistent with a stress-related leukocyte pattern (Table 5).

Table 5 Association between assessment with leukogram in Crohn’s disease patients treated at a reference center.

Parameter	Without fatigue	With fatigue	P value	Cohen’s d	Interpretation
Total leukocytes	6.140	7.052	0.034	-0.39	Increased leukocytes with fatigue
Segmented (%)	48.3	55.2	0.017	-0.53	Increase in segmented with fatigue
Eosinophils (%)	3.83	2.58	0.036	-0.61	Eosinophil reduction with fatigue
Lymphocytes (%)	38.1	32.2	0.019	-0.55	Lymphocyte reduction with fatigue

No associations were detected between fatigue and other laboratory test results [transferrin saturation index, ferritin concentration (ng/mL), serum iron concentration (μg/dL), vitamin B12 concentration (pg/mL), vitamin B1 concentration (ng/mL), free T4 concentration (ng/dL), ultrasensitive TSH concentration (μUI/mL), CRP concentration (mg/dL), or erythrocyte sedimentation rate (mm/hour)]. Effect sizes were calculated using Cohen’s d; positive values indicate higher means in the non-fatigue group, and negative values indicate higher means in the fatigue group. No other laboratory parameter was significantly associated with fatigue, suggesting that fatigue is not strongly associated with routine biochemical markers in this cohort (Table 6).

Table 6 Assessment of fatigue with different laboratory parameters in patients with Crohn’s disease treated at a reference center.

Laboratory parameter	Average (no fatigue)	Average (with fatigue)	P value	Cohen’s d
Erythrocytes (million/mm3)	4.7	4.61	0.432	0.23
Hemoglobin (g/dL)	13.59	13.22	0.199	0.38
Hematocrit (%)	40.29	39.19	0.252	0.32
RDW (%)	13.89	14.13	0.362	0.26
Total leukocytes (thousand/mm3)	6.140	7.052	0.034	0.39
Platelets (million/mm3)	278.12	283.53	0.726	0.07
Transferrin saturation index (%)	23.14	18.76	0.208	0.40
Ferritin (ng/mL)	130.70	144.35	0.742	0.09
Serum iron (μg/dL)	74.88	67.09	0.396	0.27
Vitamin B12 (pg/mL)	426.77	415.76	0.830	0.06
Vitamin B1 (ng/mL)	6.94	6.42	0.436	0.23
Free T4 (ng/dL)	1.13	1.10	0.525	0.18
Ultrasensitive TSH (μIU/mL)	2.16	1.79	0.186	0.39
CRP (mg/dL)	0.60	0.73	0.442	0.22
Erythrocyte sedimentation rate (mm/hour)	11.06	13.26	0.392	0.27

RDW: Red cell distribution width; TSH: Thyroid-stimulating hormone; CRP: C-reactive protein.

DISCUSSION

This study investigated the relationships between fatigue and sociodemographic, clinical, psychological, and laboratory test data in patients with CD. Ananthakrishnan et al[23] reported that IBD patients with fatigue incur significantly higher healthcare costs, as evidenced by more outpatient visits, more hospitalizations, and higher pharmacy expenses. This finding reinforces the importance of assessing variables associated with fatigue in patients with CD and their management in routine care.

Although depression, anxiety, stress, and insomnia were not individually associated, their combined effect was strongly associated with fatigue, with each additional symptom increasing the probability by 2.6 times, suggesting that the score can be used as a simple and effective clinical tool for stratifying the risk of fatigue in patients with IBD. These findings are consistent with those from previous research that identified psychological distress and insomnia as important contributors to fatigue in patients with IBD[24-26].

Patients with IBD report physical inactivity throughout the lifespan. This is important because reduced physical activity is associated with reduced mitochondrial function and other metabolic deficits in skeletal muscle[27,28]. In this study, a sedentary lifestyle was identified as an important determinant, corroborating findings from prior research demonstrating that regular physical activity is associated with lower fatigue severity[29,30]. Exercise is hypothesized to reduce fatigue through both psychological benefits and anti-inflammatory mechanisms[31]. Encouraging structured physical activity as part of routine care could represent an effective, nonpharmacological strategy for mitigating fatigue in patients with CD.

Interestingly, no associations were observed between fatigue and clinical disease activity, erythrocyte sedimentation rate, or CRP level, which is consistent with the findings of previous reports[31,32]. This suggests that fatigue can occur independently of measurable inflammatory activity, challenging the assumption that remission guarantees symptom resolution and indicating the need for studies exploring mechanisms beyond overt inflammation.

By contrast, the leukogram analysis revealed significant associations between fatigue and alterations in leukocyte subpopulations, including leukocytosis, neutrophilia, relative lymphopenia, and eosinopenia. These findings suggest stress-induced immune activation, a pattern consistent with that of other chronic conditions characterized by fatigue, such as cancer, long coronavirus disease 2019, and aging[33-36]. The immunological profile observed in patients with fatigue may reflect the activation of hypothalamic-pituitary-adrenal axis pathways and subtle systemic inflammation. This axis constitutes the main neuroendocrine pathway through which stress alters immune responses, primarily through the secretion of glucocorticoids. Under conditions of chronic stress, the adaptive function of the hypothalamic-pituitary-adrenal axis may be lost. Prolonged exposure to cortisol can lead to impaired feedback and resistance to glucocorticoid receptors, resulting in diminished immunosuppressive effects and a proinflammatory state[37]. Fatigue assessments incorporating simple hematological markers may be a more accessible and low-cost tool for clinical practice.

Nutritional and endocrine variables, including anemia, iron deficiency, and vitamin B1 and vitamin B12 levels, were not significantly associated with fatigue. Although anemia is a common complication of IBD[38-42], its contribution to fatigue remains controversial, and supplementation studies have revealed inconsistent results[43-45]. Similarly, thyroid function parameters (TSH and free T4) were unrelated to fatigue, which is consistent with the findings of previous reviews[46]. Moreover, the findings suggest that correcting only these factors may not have a substantial impact on fatigue in CD patients. Patients using vitamin complexes and thyroid hormone replacement therapy were not excluded from the study, which may have affected the results.

The strong association between fatigue and composite psychological symptoms, as well as leukocyte alterations, may serve as a potential basis for the development of practical screening strategies. A multidimensional approach that integrates psychological and immunological domains could enhance clinical decision-making. The composite psychological score proposed in this study is promising but requires validation in independent cohort studies.

This study has several limitations that should be considered. First, the sample size was small. Studies with larger samples are needed to better assess anemia among patients with CD and to investigate the relationship between anemia and fatigue. The second limitation was the selection of patients, as only those who were followed up in a specialized or high-complexity hospital were included. Notably, 77% of the patients were using biologics, which may indicate selection bias toward patients with more severe disease and a higher expected prevalence of fatigue. The study was also conducted in a single specialized center, exclusively with outpatients, and did not include patients who were hospitalized; thus, the general condition of the patients with CD included in the study is not representative. Another limitation was the cross-sectional design of the study (it does not allow for causal inferences).

Despite the recognized limitations, some relevant points should be acknowledged. One of the strengths of this study is the choice of questionnaire to measure fatigue, as the IBD-F scale is specific to the IBD population and has been validated for use in Brazil. The patients received personalized care, which was provided by a well-trained team. Notably, laboratory tests were performed by a single laboratory and were conducted no more than 15 days after the forms were filled out (with an average of 8 days). Examinations were conducted exclusively for this research. Prospective studies with standardized, objective tools for fatigue evaluation are needed in the future, considering the increased use of fatigue as a primary or secondary endpoint in clinical trials.

CONCLUSION

Fatigue remains a highly prevalent and burdensome symptom in patients with CD, especially those receiving pharmacological treatment and in clinical remission. In this study, fatigue was independently associated with a shorter duration of disease, a sedentary lifestyle, and a greater burden of psychological symptoms, including depression, anxiety, stress, and insomnia. Additionally, specific hematologic patterns, namely, elevated total leukocyte and segmented neutrophil counts and reduced lymphocyte and eosinophil counts suggest stress-related immune activation that deserves further exploration. These findings reinforce the notion that fatigue in patients with CD is a multidimensional phenomenon that has been insufficiently supported by traditional inflammatory markers or disease activity scores. As such, clinical assessments incorporating composite symptom scales and basic hematologic parameters may be a practical, low-cost strategy that can better identify and support patients experiencing fatigue. Future longitudinal studies are warranted to validate these associations and guide targeted, multidisciplinary interventions aimed at reducing fatigue and improving quality of life in this population.

ACKNOWLEDGEMENTS

Department of Life Sciences, Postgraduate Program in Pharmaceutical Sciences, State University of Bahia, the institutional support, the infrastructure provided, and the academic environment made the development of this research possible.