INTRODUCTION

Centrally mediated abdominal syndrome (CAPS) is defined by constant, nearly constant, or repeated episodes of abdominal pain, which lasts for more than 6 months according to Rome IV criteria (1). Although CAPS is not the most prevalent disorder of gut-brain interaction (DGBI) with the prevalence 0.5%–2.1% (2), recurrent unknown abdominal pain still negatively affects patients' lives and generates a heavy burden of disease across the globe (3). A large number of patients sought medical care in outpatient department and unnecessary invasive examinations repeatedly, such as endoscopy and abdominal ultrasound, but the clear etiology remains poorly elucidated (4).

The pathogenesis of CAPS may be associated with gut-brain abnormalities. Dysfunction of gut-brain axis can cause gastrointestinal (GI) symptoms such as abdominal pain, dyspepsia, and constipation (5). The current theory holds that the central sensory processing plays a dominant role in the pathophysiology of CAPS, rather than peripheral stimulation. As long as central sensitization is established, abdominal pain persists even in the lack of peripheral stimulation. Some researchers surmised that dysregulation of descending pain pathway was the main cause of CAPS, and the mental disorder seemed to be a risk factor (1). For example, anxiety could affect pain coping and impaired the endogenous opioid release, which was involved in pain control (6). However previous studies merely showed that CAPS and mental disorders were commonly comorbid in children (7). CAPS is known to be one of the DGBI, which is also called functional gastrointestinal disorder (FGID). It has been suggested that the prevalence of mental disorders are higher among patients with FGIDs, particularly depression, anxiety, and somatization disorder (8). A retrospective study demonstrated that the diagnosis of anxiety or other mental disorders was 3 years before FGID diagnosis; the researchers of that study supposed that we could prevent FGID through intervening mental disorders (9).

A 1-year follow-up study found that if patients had anxious or depressive symptoms at baseline, they were more likely to experience irritable bowel syndrome (IBS) or other DGBI after 1 year (10). In addition, numerous studies indicated that psychological factors strongly linked to IBS, which was the most common DGBI (11). Thus, we assume that there seems to be a similar relationship between CAPS and psychological factors.

Sleep disturbance is likely to be another contributed factor for CAPS. Koh et al (12) found that sleep disturbance was important for the etiology of functional abdominal pain in nurses with shift work. Sleep disturbance causes circadian rhythm disruptions. Normal motility and immune barrier of gastrointestinal tract were dependent on normal circadian rhythms. Furthermore, abnormal circadian rhythms were critical for gut-brain signaling pathway (13). In addition, lifestyle and habits, gender, education level, marital status, and physical activity were also significantly related to functional symptoms (14).

To our knowledge, detailed prospective studies for clinical characteristics of CAPS are lacking. Clinicians have not comprehended adult CAPS deeply. Given this, the aim of our study was to explore the clinical characteristics and the related factors of CAPS.

METHODS Study design

This study was conducted between October 1, 2022, and January 31, 2023, which prospectively included 73 patients with CAPS and 132 age-matched and gender-matched healthy controls in Ruijin Hospital, Shanghai Jiao Tong University School of Medicine. All participants filled paper questionnaires. Ethical approval was gained from Ethics Committee of Ruijin Hospital, Shanghai Jiaotong University School of Medicine before the start of the study (Number: KY2022-150). All enrolled participants signed their informed consent before the inclusion. All information collected was kept confidential strictly.

Sample size

The sample size in this study was determined based on the reported proportions of anxiety and depressive symptoms in patients with CAPS (15) and in the Chinese general population (16). We assumed a proportion of 38.9% of anxiety in patients with CAPS and a proportion of 14.15% of anxiety in the control group, a statistical power of 95%, a 2-sided significance level of 0.05, and a 1:2 ratio for numbers of participants in the case and control groups. The calculation was conducted using the PASS software, and a sample size of 64 cases and 128 controls was needed for this study. We oversampled an additional 15% cases and recruited a total of 73 patients with CAPS. However, we were not able to match each case with 2 controls and recruited a total of 132 controls in this study.

Participants

Between October 1, 2022, and January 31, 2023, a total of 82 outpatients were diagnosed with CAPS according to Rome IV criteria by professional gastroenterologists. Finally, 73 patients were successfully included in the CAPS group according to the inclusion and exclusion criteria. During the same period, 132 age-matched and gender-matched participants were included in the healthy control group. All the healthy controls were enrolled from the subjects who underwent screening gastroscopy and colonoscopy for routine checkups in the digestive endoscopy center and had negative findings. Furthermore, the controls had no history of digestive tract disorder and surgery.

Inclusion and exclusion criteria.

The inclusion criteria for patients with CAPS were as follows: (i) aged 18–85 years, (ii) met the Rome IV diagnostic criteria and diagnosed by professional gastroenterologists, (iii) a signed informed consent, and (iv) all the questionnaires were completed. Inclusion criteria for healthy controls: (i) aged 18–85 years, (ii) no abdominal pain and other gastrointestinal symptoms, (iii) no organic lesions were detected by gastroscopy and colonoscopy within the past 6 months, (iv) a signed informed consent, and (v) all the questionnaires were completed.

The exclusion criteria were as follows: (i) participants complicated with organic lesions, (ii) with a history of abdominal surgery, (iii) pregnant and lactating women, and (iv) participants who failed to complete all the questionnaires.

Data collection Baseline data.

Baseline data including age, gender, BMI, medication history, educational attainment, marital status, smoking status, coffee intake, and physical activity engagement were obtained from questionnaire.

Psychological questionnaires.

To generally evaluate the severity of depression and anxiety, 4 questionnaires were applied including the Hamilton Depression Rating Scale (HAMD), Hamilton Anxiety Rating Scale (HAMA), 9-item Patient Health Questionnaire (PHQ-9), and 7-item Generalized Anxiety Disorder (GAD-7) scale.

HAMA.

HAMA is a clinician-rated instrument, which consists of 14 items and designed to measure severity of anxiety (17). HAMA score criteria are as follows: 0–6 = no anxiety; 7–13 = mild to moderate anxiety; and 14 or over = severe to extreme anxiety (18).

HAMD.

HAMD is the most commonly used clinician-rated scale to categorized patients into different severity depression groups (19). The HAMA score criteria are as follows: 0–6 = no depression; 7–23 = mild to moderate depression; 24 or over = severe to extreme depression (18).

GAD-7.

The GAD-7 is a self-reported anxiety rating scale with satisfied reliability (Cronbach ɑ = 0.89) and validity. The GAD-7 asks how frequently patients have experienced symptoms that GAD-7 mentions, within the past 2 weeks. The total scores are 21 (0–4 score = no anxiety, 5 or over = anxiety present). Higher scores present more severe anxious symptoms (20).

PHQ-9.

This measurement is widely used to quantify the severity of depression within the past 2 weeks in both busy clinical settings and researches. This questionnaire has higher reliability (Cronbach ɑ = 0.86–0.89) in various clinical settings. The total scores are 27 (0–4 score = no depression, 5 or over = depression present). Higher scores present more severe depressive symptoms (21).

Sleep quality questionnaire.

The PSQI is a self-reported sleep quality within the previous 1 month. This scale assesses 7 dimensions: sleep latency, sleep quality, sleep duration, use of sleep medication, daytime dysfunction, habitual sleep efficiency, and sleep disturbances. The final score is calculated as the sum of 7 dimensions. 0–4 score = good sleepers, 5 or higher score = poor sleepers (22). The PSQI is a reliable tool for evaluating all subjects' sleep quality because of Cronbach ɑ = 0.82–0.83 (23).

Visual analog scale (VAS).

The VAS is a self-described scale made up of a 10-cm long horizontal line and widely used in the clinical settings to reflect pain intensity (24). The score of VAS ranged from 0 to 10. Patients were asked to describe abdominal pain intensity by circling the number. The scores were categorized as follows: 0 = no pain, 1–3 = mild pain, 4–6 = moderate pain, and 7–10 = severe pain.

Quality of life questionnaire.

The 36-Item Short-Form Survey questionnaire is the most commonly used generic instrument to measure patients' health-related quality of life by most of clinicians and researchers. The questionnaire mainly measures physical health (physical component scale [PCS]) and mental health (mental component scale [MCS]) by means of 8 domains including physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. Higher scores indicate better quality of life (25).

Physical activity.

We defined the duration of exercise or walking ≥3 hr/wk as physical activity/exercise and the duration of exercise or walking <3 hr/wk as physical inactivity/nonexercise regardless of the occupations. This standard was established based on the General Practice Physical Activity Questionnaire (GPPAQ) (26) and GPPAQ-WALK (27) and combined with the actual situation of the participants in our research. Considering that some participants of our study had retired, some worked part-time, some were freelancers, and the participants took part in various physical activities, such as walking, gym, cycling, and swimming, GPPAQ and GPPAQ-WALK were not fully applicable to these people. Thus, we finally defined the abovementioned criteria.

Statistical analysis

Normally distributed continuous data were described as the mean ± SD, and non-normally distributed continuous data were presented as median (P25, P75). Categorical data were expressed as frequency (percentage). The Student t test, Wilcoxon rank-sum tests, χ2 test, and Fisher's exact test were used to analyze differences between groups. Univariate and forward stepwise multivariate logistic regression (LR), Spearman correlation analysis, and multivariable linear regressions were used for exploring risk or protective factors. Two-sided P < 0.05 was considered to be statistically significant. All statistical analyses were conducted using IBM SPSS Statistics version 26.0, and figures were made with GaphPad Prism 9.

RESULTS

The demographic and clinical features of 2 groups are summarized in Table 1. In the CAPS group, 47 (64.4%) of the patients were women; the mean age was 49.11 ± 14.75 years, and the BMI was 21.38 (19.93–24.50) kg/m2. In the healthy control group, 93 (70.5%) of the participants were women; the mean age was 46.37 ± 5.58 years, and the BMI was 21.93 ± 3.05 kg/m2. There were no significant differences in age, gender, and BMI between 2 groups. All patients with CAPS had a history of medication to improve symptoms before enrollment. Of the 73 patients with CAPS, 37 took pinaverium bromide, 12 took trimebutine, 8 took PPI, 9 took tricyclic antidepressant, 5 took duloxetine, and 2 took flupentixol-melitracen. All healthy controls had no related medication history. Among the participants, nonexercise, noncoffee, and nontea intake were more common in the CAPS group than in the healthy control group (nonexercise: 65 [89%] vs 68 [51.5%] P < 0.001; noncoffee: 69 [94.4%] vs 78 [59.1%] P < 0.001; nontea 65 [89%] vs 91 [68.9%] P < 0.05). This suggested that lifestyle and habits may be related to the prevalence of CAPS. In addition, education attainment and marital status were significantly different between patients with CAPS and controls. There were more higher-degree participants in the control group than in the CAPS group (university: 60 [45.5%] vs 18 [24.7%]; postcollege: 34 [25.8%] vs 2 [2.7%]). And there were more married participants in the CAPS group than in the control group (married: 62 [84.9%] vs 62 [47%]). The difference in smoke condition was not statistically significant.

Table 1. - Baseline characteristics of the participants Variables Patients with CAPS
N = 73 Healthy controls
N = 132 P Gender, n (%) 0.23  Male 26 (35.6) 39 (29.5)  Female 47 (64.4) 93 (70.5) Age (y) 49.11 ± 14.75 46.37 ± 5.58 0.06 BMI (kg/m2) Medication history (%) 21.38 (19.93–24.50) 21.93 ± 3.05 0.929  Pinaverium bromide 37 (50.7) 0 NA  Trimebutine 12 (16.4) 0 NA  PPI 8 (11) 0 NA  TCA 9 (12.3) 0 NA  Duloxetine 5 (6.8) 0 NA  Flupentixol-melitracen 2 (2.8) 0 NA Education, n (%) <0.001***  Primary school or below 12 (16.4) 6 (4.5)  Junior school 29 (39.7) 11 (8.3)  High school 12 (16.4) 21 (15.9)  University studies 18 (24.7) 60 (45.5)  Postcollege 2 (2.7) 34 (25.8) Marital status, n (%) <0.001***  Married 62 (84.9) 62 (47)  Single 10 (13.7) 68 (51.5)  Divorced 1 (1.4) 1 (0.8)  Widowed 0 1 (0.8) Smoke, n (%) 0.464  Smoke 9 (12.3) 12 (9.1)  Nonsmoke 64 (87.7) 120 (90.9) Coffee, n (%) <0.001***  Coffee 4 (5.5) 54 (40.9)  Noncoffee 69 (94.4) 78 (59.1) Tea, n (%) 0.001**  Tea 8 (11) 41 (31.1)  Nontea 65 (89) 91 (68.9) Exercise, n (%) <0.001***  Exercise 8 (11) 64 (48.5)  Nonexercise 65 (89) 68 (51.5) GAD-7 score M (P25–P75) 5 (3–7.5) 2 (0–5) <0.001*** GAD-7 severity, n (%) <0.001***  GAD-7 <5 35 (47.9) 98 (74.2)  GAD-7 ≥5 38 (52.1) 34 (25.8) PHQ-9 score M (P25–P75) 4 (3–6) 2 (1–5.75) 0.001*** PHQ-9 severity, n (%) 0.006**  PHQ-9<5 38 (52.1) 94 (71.2)  PHQ-9≥5 35 (47.9) 38 (28.8) PSQI score M (P25–P75) 6 (5–9) 5 (3–7) <0.001*** PSQI severity, n (%) 0.005**  PSQI≤5 27 (37) 76 (57.6)  PSQI>5 46 (63) 56 (42.4) HAMA score M (P25–P75) 8 (5.5–10) 4 (2–7) <0.001*** HAMA severity, n (%) <0.001***  No anxiety (0–6) 27 (37) 97 (73.5)  Mild-to-moderate anxiety (7–13) 37 (50.7) 30 (22.7)  Severe-to-extreme anxiety (≥14) 9 (12.3) 5 (3.8) HAMD score M (P25–P75) 7 (5–10) 3 (1–5) <0.001*** HAMD severity, n (%) <0.001***  No depression (0–6) 28 (38.4) 115 (87.1)  Mild-to-moderate depression (7–23) 44 (60.3) 17 (12.9)  Severe-to-extreme depression (≥24) 1 (1.4) 0

BMI, body mass index; CPAS, centrally mediated abdominal pain syndrome; GAD-7, 7-item Generalized Anxiety Disorder; HAMA, Hamilton Anxiety Rating Scale; HAMD, Hamilton Depression Rating Scale; PHQ-9, 9-item Patient Health Questionnaire; PPIs, proton pump inhibitors; PSQI, Pittsburgh Sleep Quality Inventory; TCA, tricyclic antidepressants.

**P < 0.01; ***P < 0.001.

The questionnaire results of psychological status and sleep quality between 2 groups are summarized in Table 1. As expected, patients have higher GAD-7 score (5 [3–7.5] vs 2 [0–5] P < 0.001), PHQ-9 score (4 [3–6] vs 2 [1–5.75] P < 0.05), HAMA score (8 [5.5–10] vs 4 [2–7] P < 0.001), HAMD score (7 [5–10] vs 3 [1–5] P < 0.001), and PSQI score (6 [5–9] vs 5 [3–7] P < 0.001) than healthy controls. Furthermore, 50.7% of patients with CAPS showed mild-to-moderate anxiety, and 12.3% showed severe-to-extreme anxiety, and the corresponding proportions of healthy controls were 22.7% and 3.8%. Approximately 60.3% of patients with CAPS showed mild-to-moderate depression, and 1.4% showed severe-to-extreme depression. The different severity of anxiety (P < 0.001) and depression (P < 0.001) between the 2 groups were statistically significant. In summary, compared with healthy controls, more patients had complication of depression or anxiety. Meanwhile, poor sleep quality (PSQI>5) was more frequently noted in the CAPS group than in the healthy control group (63% vs 42.4%, P < 0.05).

The univariate LR model for CPAS are summarized in Table 2. According to the results of model, exercise condition, educational attainment, marital status, coffee consumption, tea consumption, HAMD/HAMA/PSQI severity, GAD-7, and PHQ-9 were included in the forward stepwise regression model (LR). Ultimately, nonexercise (adjusted odds ratio [AOR] 4.53; CI [1.602–12.809]), mild-to-moderate depression (AOR 7.931; CI [3.236–19.438]), and married status (AOR 3.656; CI [1.317–10.418]) are risk factors of CAPS after adjusting for age and gender. By contrast, drinking coffee is a protective factors of CAPS (AOR 0.199; CI 0.051–0.775) (Figure 1).

Table 2. - Univariate logistic regression analysis models of CAPS β Odds ratio (95% CI) P Education  Primary school Reference  Junior school 0.276 1.318 (0.397–4.380) 0.652  High school −1.253 0.286 (0.085–0.958) 0.042*  University studies −1.897 0.150 (0.049–0.456) 0.001**  Postcollege −3.526 0.029 (0.005–0.166) <0.001*** Marital status  Single Reference  Married 1.917 6.80 (3.046–13.747) <0.001*** Coffee  Noncoffee intake Reference  Coffee intake −2.480 0.086 (0.029–0.243) <0.001*** Tea  Nontea intake Reference  Tea intake −1.298 0.273 (0.12–0.621) 0.002** Exercise engagement  Exercise Reference  Nonexercise 2.034 7.647 (3.402–17.187) <0.001*** GAD-7  GAD-7<5 Reference  GAD-7 ≥5 1.141 3.129 (1.713–5.717) <0.001*** PHQ-9  PHQ-9 <5 = 0 Reference  PHQ-9 ≥5 0.823 2.278 (1.258–4.127) 0.007** HAMD  No depression (0–6) Reference  Mild-to-moderate depression (7–23) 2.364 10.63 (5.302–21.313) <0.001*** HAMA  No anxiety (0–6) Reference  Mild-to-moderate anxiety (7–13) 1.489 4.431 (2.329–8.430) <0.001***  Severe-to-extreme anxiety (≥14) 1.867 6.467 (2.329–20.908) 0.002** PSQI  PSQI≤5 Reference  PSQI>5 0.838 2.312 (1.285–4.160) 0.005**

CAPS, centrally mediated abdominal pain syndrome; GAD-7, 7-item Generalized Anxiety Disorder; HAMA, Hamilton Anxiety Rating Scale; HAMD, Hamilton Depression Rating Scale; PHQ-9, 9-item Patient Health Questionnaire; PSQI, Pittsburgh Sleep Quality Inventory.

*P < 0.05; **P < 0.01; ***P < 0.001.

Figure 1.:

Forest plot showed independent risk and protective factors of CAPS according to forward stepwise regression model. Adjusted odds ratio (AOR) for age and gender; mild-to-moderate depression: HAMD score = 7–23; no depression: HAMD score = 0–6. CPAS, centrally mediated abdominal pain syndrome; HAMD, Hamilton Depression Rating Scale.

To further assess the effect of different pain severity on a patient's health-related quality of life (HRQOL) and clinical figures, patients were asked for completing the VAS and 36-Item Short-Form Survey questionnaires. Seventy-three patients with CAPS were divided into 2 subgroups regarding the VAS score, mild pain group (VAS score 0–3) and moderate-to-severe pain group (VAS score 4–10). BMI, age, gender, exercise condition, severity of HAMD/PSQI, and GAD-7/PHQ-9/HAMA/HAMD/PSQI score were not statistically significant in 2 subgroups except for HAMA severity, which showed statistically significant differences between subgroups. Mild-to-moderate anxious patients were more common in the moderate-to-severe pain group than in the mild pain group (34 [58.6%] vs 3 [20%], P < 0.05) (Table 3). Based on the result of bivariate LR after adjusting for age and gender, patients with CAPS with mild-to-moderate anxiety had a significantly higher risk of experiencing moderate-to-severe abdominal pain when compared with patients without anxiety (odds ratio = 7.043 CI [1.319–37.593]) (Figure 2).

Table 3. - Clinical figures and questionnaire results of CAPS subgroups Mild pain group n = 15 Moderate-to-severe pain group n = 58 P BMI (kg/cm2) 22.26 ± 2.79 21.88 (20.31–24.51) 0.544 Age(y) 45.46 ± 12.85 50.64 ± 15.04 0.255 Gender, n (%) 0.527  Female 8 (53.3) 39 (67.2)  Male 7 (46.7) 19 (32.8) Exercise status, n (%) 0.853  Exercise 2 (13.4) 6 (10.3)  Nonexercise 13 (86.6) 52 (89.7) PSQI score 6.23 ± 2.92 6.5 (5–9) 0.418 PSQI 0.565  PSQI≤5 6 (46.2) 21 (36.2)  PSQI>5 8 (53.8) 38 (63.8) GAD-7 score 4 (3–8) 5 (2–7.75) 0.787 GAD-7, n (%) 0.714  GAD-7<5 7 (46.7) 28 (48.2)  GAD-7≥5 8 (53.3) 30 (51.8) PHQ-9 score 4 (2.5–7.5) 4 (2.5–6) 0.757 PHQ-9, n (%) 0.921  PHQ-9<5 7 (46.7) 31 (53.4)  PHQ-9≥5 8 (53.3) 27 (46.6) HAMD score 6 (3.5–8) 7 (5–10) 0.238 HAMD severity, n (%) 0.469  No depression (0–6) 7 (46.7) 21 (36.2)  Mild-to-moderate depression (7–23) 8 (53.3) 36 (62.1)  Severe-to-extreme depression (≥24) 0 1 (1.7) HAMA, n (%) 6 (4.5–13.5) 8 (5.25–10) 0.507 HAMA severity 0.012*  No anxiety (0–6) 8 (53.3) 19 (32.8)  Mild-to-moderate anxiety (7–13) 3 (20) 34 (58.6)  Severe-to-extreme anxiety (≥14) 4 (26.7) 5 (8.6) SF-36  PF 100 (95–100) 100 (91.25–100) 0.843  RP 100 (75–100) 100 (75–100) 0.320  BP 62 (62–74) 62 (52–62) 0.016  GH 51.08 ± 17.50 49.50 ± 16.00 0.754  VT 80 (65–92.5) 80 (70–85) 0.902  SF 87.5 (81.25–100) 75 (75–87.5) 0.078  RE 100 (100–100) 100 (66.67–100) 0.135  MH 69.54 ± 15.01 80 (68–92) 0.137  PCS 78 (60.25–82) 72.38 (64.06–78.94) 0.282  MCS 80.38 (68.96–91.07) 79.09 (70.14–87.38) 0.825

Mild pain group: the visual analog scale score 0–3; moderate-to-severe pain group: the visual analog scale score 4–10.

BMI, body mass index; BP, bodily pain; CAPS, centrally mediated abdominal pain syndrome; GAD-7, 7-item Generalized Anxiety Disorder; GH, general health; HAMA, Hamilton Anxiety Rating Scale; HAMD, Hamilton Depression Rating Scale; MCS, mental component scale; MH, mental health; PCS, physical component scale; PF, physical functioning; PHQ-9, 9-item Patient Health Questionnaire; PSQI, Pittsburgh Sleep Quality Inventory; RE, role-emotional; RP, role-physical; SF, social functioning; SF-36, 36-Item Short-Form Survey; VT, vitality.

*P < 0.05.

Figure 2.:

Forest plot showed independent risk factors of abdominal pain intensity according to bivariate logistic regression model. Severe-to-extreme anxiety: HAMA score ≥14; mild-to-moderate anxiety: HAMA score 7–13; no anxiety: HAMA score = 0–6. HAMA, Hamilton Anxiety Rating Scale.