OUP user menu

Journal of Crohn's and Colitis: 10 (8)


Laurence J. Egan, Ireland

Associate Editors

Shomron Ben-Horin, IsraelSilvio Danese, ItalyPeter Lakatos, HungaryMiles Parkes, UKJesús Rivera-Nieves, USABritta Siegmund, GermanyGijs van den Brink, NLSéverine Vermeire, Belgium


Published on behalf of

Correlation between serum serotonin and endoscopy inflammation scores in patients with ileal pouches

Yinghong Wang, Hua Gong, Rocio Lopez, Lei Lian, Ravi P. Kiran, Edy E. Soffer, Bo Shen
DOI: http://dx.doi.org/10.1016/j.crohns.2012.07.028 e133-e142 First published online: 1 May 2013


Background: Inflammatory and functional complications are common in patients with inflammatory bowel disease (IBD) after restorative proctocolectomy with ileal pouch-anal anastomosis (IPAA). The pathogenesis of these complications remains poorly understood, and there is discrepancy between the clinical symptoms and objective endoscopic inflammation scores in these patients. While serum serotonin level has been reported to be associated with symptoms of irritable bowel syndrome, its association with ileal pouch disorders has not been studied.

Aim: To investigate the association between serum serotonin level and symptoms and endoscopic inflammation in patients with IPAA.

Methods: A total of 185 consecutive eligible IPAA patients who presented to a specialized Pouchitis Clinic from Jan 2009 to May 2009 were prospectively recruited. Patients were divided into 4 groups: normal pouch, irritable pouch syndrome (IPS), inflammatory pouch disorders (Crohn's disease, acute and chronic pouchitis, and cuffitis), and surgical complications. Serum serotonin level was measured and analyzed for correlation with clinical and endoscopic inflammation scores.

Results: Demographic and clinical variables were evaluated, including age, gender, smoking history, duration of UC, duration of the pouch, and disease category of the pouch. The median fasting serum serotonin level was comparable among the 4 groups: 94.0 ng/ml (interquartile range [IQR], 70.0, 128.1), 89.2 ng/ml (IQR 54.2, 155.9), 90.3 ng/ml (IQR 49.7, 164.1), 77.9 ng/ml (IQR 54.7, 129.0), for normal pouch, irritable pouch, inflammatory pouch disorders, and surgical complication groups, respectively (p = 0.91). A significant association between serum serotonin level and the Pouchitis Disease Activity Index (PDAI) endoscopy subscore of the pouch (odds ratio [OR] = 1.9, 95% confidence interval [CI]: 1.2, 2.9, p < 0.05) and total PDAI endoscopy score (OR = 1.8; 95% CI: 1.2, 2.8, p < 0.05) in the inflammatory complication group were noted.

Conclusions: Serum serotonin level appears to correlate with the PDAI endoscopy subscores and total PDAI score in patients with inflammatory complications suggesting that the hormone may be involved in mechanisms of mucosal inflammation. These findings may promote future treatment strategies for patients with pouch inflammation.

  • Ileal pouch
  • Inflammatory bowel disease
  • Restorative proctocolectomy
  • Serotonin
  • Ulcerative colitis

1 Introduction

5-hydroxytryptamine (5-HT) is released through complicated signaling processes which include mechanical, chemical, neural factors, infection, and inflammation in the gastrointestinal (GI) tract. While 5-HT biosynthesis can be altered in disease settings, its availability is also regulated by the degradation process which involves serotonin reuptake transporters (SERT) located at nerve terminals, enterocytes and vascular endothelial cells.1 5-HT plays an important role in the pathophysiology of irritable bowel syndrome (IBS). Activation of the mucosal immune system with morphologic changes involving lymphocytes, mast cells, enterochromaffin (EC) cells and enteric nerves by quantitative histologic and ultra-structural analyses is believed to contribute to the pathophysiology of IBS.2,3 EC cell hyperplasia has been reported in rectal biopsies in patients with post infectious-IBS.4 Blood and tissue 5-HT levels have also been evaluated in IBS patients, and a high blood 5-HT level demonstrated in diarrhea-predominant IBS patients.58

There are scant published data on the role of 5-HT in inflammatory bowel disease (IBD), although IBS-like conditions often coexist with IBD.8,9 Animal studies have suggested EC cell hyperplasia and increased 5-HT content in the inflamed colon.10,11 An increased level of 5-HT in the colon tissue specimens of patients with Crohn's disease (CD) has also been reported.8 However, serum 5-HT has not been evaluated in patients with IBD.

Restorative proctocolectomy with ileal pouch-anal anastomotosis (IPAA) has become the surgical treatment of choice for ulcerative colitis (UC) patients who require colectomy. Inflammatory (e.g. pouchitis) and functional (e.g. irritable pouch syndrome [IPS]) complications are common after IPAA surgery. The pathogenesis of these inflammatory and functional pouch disorders is largely unknown. We have previously demonstrated a poor correlation between subjective symptoms and objective endoscopic inflammation scores in patients with pouchitis.12 One potential explanation for such discrepancy is that some of these patients may have concurrent IPS.13 Subsequently, we have demonstrated evidence of visceral hypersensitivity14 and EC cell hyperplasia15 in IPS patients. The clinical and pathogenetic role of serum 5-HT in patients with IPAA has not been evaluated. We hypothesized that an alteration in serum 5-HT may contribute to symptoms related to the pouch and mucosal inflammation in patients with IPAA. The aims of this study were to compare serum 5-HT level among patients with healthy and diseased ileoanal pouch and to assess the correlation between serum 5-HT, clinical symptoms and endoscopic inflammation.

2 Patients and methods

2.1 Patients

This is a prospective study of patients who were seen in our subspecialty Pouchitis Clinic between 2008 and 2010. Demographic, clinical, and endoscopic data were collated in a prospectively maintained, Institutional Review Board (IRB)-approved Pouchitis Database. A total of 185 patients were recruited, and divided into 4 groups: 1) normal pouch; 2) IPS; 3) inflammatory pouch, consisting of CD of pouch, pouchitis, and cuffitis; and 4) surgical complications. The senior investigator B.S. performed all clinical interviews, physical examinations, and pouch endoscopies and biopsies. The protocol was approved by our IRB.

2.2 Inclusion and exclusion criteria

The inclusion criteria included patients who underwent IPAA for underlying IBD, were older than 18 and hence able to give consent, and undergoing diagnostic or surveillance pouchoscopy. Exclusion criteria were patients 1) with IPAA for familial adenomatous polyposis; 2) extraintestinal infection, such as pneumonia, upper respiratory infection, urinary tract infection; 3) systemic disorders, such as chronic renal insufficiency, poorly controlled diabetes, and malignancy; 4) overt gastrointestinal bleeding or extra-intestinal blood loss (e.g. meno- or metrorrhagia).

2.3 Demographic and clinical variables

Demographic and clinical variables were evaluated, including age, gender, smoking history, duration of UC, family history of IBD, extent of colitis, indication for proctocolectomy, extraintestinal manifestations, duration of the pouch, disease category of the pouch, and selective serotonin re-uptake inhibitor (SSR use as previously described.15 The 12-point modified Pouchitis Disease Activity Index (mPDAI) consisting of symptoms (range of 0-6 points), and endoscopy (range of 0-6 points) was used to quantify symptoms and pouch inflammation.16 The mPDAI scoring system is listed in Table 1. The diagnosis of normal pouch and IPS was based on the criteria previously published by our group.13,1720 IPS was diagnosed based on the presence of abdominal pain, perianal discomfort, and increased bowel movements, with the mPDAI score > 1 point from the baseline, for a total of > 12 wks in the prior 12-month period in the absence of mucosal inflammation on pouch endoscopy and histology.13 Patients were considered to have healthy pouches or normal pouches if they were asymptomatic with the mPDAI symptom score = 0, when entering the study, with no evidence of endoscopic and absent or minimal histological inflammation. The diagnosis of pouchitis was based on the total mPDAI score of 5 or more. Pouchitis was further classified into three categories based on their response to antibiotic therapy: 1) antibiotic-responsive; 2) antibiotic-dependent; and 3) antibiotic-refractory. Patients with antibiotic-responsive pouchitis normally had an acute course with infrequent flare-ups of symptoms, which quickly responded to 2-wk antibiotic therapy with ciprofloxacin or metronidazole. Patients with antibiotic-dependent pouchitis typically had frequent flare-ups (e.g. > 4 per year), especially when the antibiotics were tapered. These patients often required long-term, low-dose antibiotic therapy as maintenance therapy or frequent pulse administration of a full-dose course of antibiotics. Patients with chronic antibiotic-refractory pouchitis had a chronic course and did not respond to the routine, single-agent, 4-week antibiotic therapy course. Topical or oral mesalamine or corticosteroids, even immunomodulators, or infliximab were occasionally used.21 Together with the CD of the pouch, and cuffitis, these were grouped as inflammatory complications of the pouch.

View this table:
Table 1

The modified pouchitis disease activity index.

Stool frequency
Usual postoperative stool frequency0
1–2 stools/day > postoperative usual1
3 or more stools/day > postoperative usual2
Rectal bleeding
None or rare0
Present daily1
Fecal urgency or abdominal cramps
Fever (temperature > 37.8 °C)
Endoscopic inflammation
Loss of vascular pattern1
Mucous exudates1

2.4 Blood sample collection and endoscopic evaluation

As part of the routine clinical practice, patients underwent pouchoscopy for surveillance or evaluation for active symptoms in the outpatient endoscopy suite during their clinic visit. Fasting was not a requirement for the routine pouchoscopy. Blood sample was collected right before the pouchoscopy after consent was obtained. Immediately upon collection, the samples were centrifuged and serum was separated and stored in -80 °C freezer for future processing. Pouchoscopy was subsequently performed with the GIF-160 or H-180 video scopes (Olympus Optical, Tokyo, Japan). Endoscopic findings of the pouch were scored based on the following parameters (edema, granularity, friability, loss of vascular pattern, mucous exudates, ulceration: these were equally weighted). Biopsies of the neo terminal ileum, pouch body, and cuff were routinely obtained. mPDAI scores were prospectively entered into the registry.

2.5 Serum 5-HT measurement

Sample acylation was performed by adding 25 μL of standard, 25 μL of control, and 25 μL of serum into respective reaction tubes. 500 μL of acylation buffer and 25 μL of acylation reagent was then added to all tubes, mixed thoroughly and incubated for 15 min at room temperature. 25 μL of the prepared standards, controls and samples were collected for the Serotonin ELISA (Rocky Mountain Diagnostics, USA and added into the appropriate wells of the microtiter plate. 100 μL of the serotonin antiserum was then pipetted into all the wells and the plates incubated for 30 min at room temperature on a shaker (approx. 600 rpm). The contents of the wells were aspirated, washed thoroughly 3 times with 300 μL wash buffer, and dried by tapping the inverted plate on absorbent material. 100 μL of the substrate was then added into all the wells and incubated for 15 ± 2 min at room temperature on a shaker (avoiding exposure to direct sun light). The reaction was stopped by adding 100 μL of the stop solution to each well, while shaking the microtiter plate to ensure a homogeneous distribution of the solution. The absorbance of the solution in the wells was read within 10 min using a microplate reader set to 450 nm and a reference wavelength between 620 nm and 650 nm. The calibration curve was obtained by plotting the absorbance readings of the standards against the corresponding standard concentrations. The serum concentrations of 5HT were read directly from the calibration curve.

2.6 Outcome measurement

The primary outcomes were the comparison of serum 5-HT level among 4 study groups and the assessment of association between serum 5-HT level and the PDAI symptom and endoscopy subscores.

2.7 Statistical analysis

The sample size calculation was done based on the serotonin study from other group. Dunlop et al.4 reported that subjects with normal pouch had a median level of plasma 5-HT of 46.4 nmol/mg (interquartile range [IQR]: 18.6, 113.6). Based on this, we estimated that the standard deviation was approximately 70.4 (IQR/1.35). Based on this assumption, with a total of 20 subjects per group we should have at least 85% power to detect a minimum difference of 82.1 in serotonin between any 2 groups at a significance level of 0.0125.

Descriptive statistics were computed for all factors. These included means, standard deviations and percentiles for continuous variables and frequencies for categorical factors. Analysis of variance (ANOVA) was used to compare age and Kruskal–Wallis tests were performed for other continuous and ordinal variables; Pearson's chi-square tests were used for categorical variables. Ad-hoc comparisons were done using Pearson's chi-square tests for categorical variables and the Stee–Dwass multiple comparisons procedure for continuous and ordinal variables; a significance level of 0.008 (0.05/6) was considered for these. In addition, proportional odds regression models were constructed to assess whether 5-HT and the PDAI endoscopy subscore associations varied by disease group. Proportional odds models use cumulative probabilities and assume proportional odds for each level. Each score was modeled as the outcome and group, 5-HT and the interaction between these 2 were included as independent variables. A P < 0.05 was considered statistically significant. SAS version 9.2 software (The SAS Institute, Cary, NC) and R version 2.10.1 software (The R Foundation for Statistical Computing, Vienna, Austria) were used to perform all analyses.

3 Results

3.1 Demographic and clinical characteristics

A total of 185 patients who met the inclusion criteria were recruited from our Pouchitis Clinic. Demographic and clinical data are summarized in Table 2. There were significant differences in the median duration of the pouch since IPAA creation (p = 0.006), indication for colectomy (p = 0.001), the PDAI symptom subscore (p < 0.001), and PDAI endoscopy subscores of the neo terminal ileum (p < 0.001), pouch (p < 0.001), and cuff (p = 0.002), among the 4 groups.

View this table:
Table 2

Demographic and clinical variables by diagnostic group.

FactorNormal pouchInflammatory pouchFunctional pouchSurgical complicationsP-value
(N = 9)(N = 125)(N = 24)(N = 29)
Mean age ± SD46.1 ± 13.246.8 ± 13.045.7 ± 16.246.3 ± 15.60.98
Male gender4(44.4)71(58.2)16(66.7)17(58.6)0.71
Current smoker0(0.00)5(4.1)3(12.5)2(6.9).
Duration of UC from diagnosis to pouch (year)11.0[1.00,19.0]5.0[2.0,11.0]6.0[2.5,15.5]8.0[2.0,16.0]0.43
Duration of pouch (year)3.0[2.0,4.0]9.0[5.0,15.0]†10.5[5.0,12.5]†7.0[3.0,10.0]0.006
Chronic NSAID use0(0.00)10(8.2)0(0.00)4(13.8)0.23
Family history of IBD0.74
Indication for colectomy0.001
Refractory disease5(55.6)110(90.2)16(66.7)‡26(89.7).
Pre-op diagnosis0.86
Indeterminate colitis1(11.1)13(10.7)1(4.2)3(10.3).
Extent of IBD0.068
Left-side colitis/Proctitis2(22.2)5(4.1)3(12.5)1(3.4).
Toxic megacolon1(11.1)13(10.7)2(8.3)2(6.9)0.93
Pre-op biological therapy0(0.00)16(13.1)2(8.3)7(24.1)0.19
Pouch type0.16
Stage of pouch surgery0.78
4 or redo0(0.00)7(5.7)0(0.00)3(10.3).
Post-op immunomodulators1(11.1)19(15.6)1(4.2)2(6.9)0.33
Post-op biologics0(0.00)16(13.1)2(8.3)6(20.7)0.35
Any autoimmune diseases3(33.3)15(12.3)1(4.2)5(17.2)0.14
Any extraintestinal manifestations*4(44.4)46(38.0)6(25.0)9(31.0)0.56
Primary sclerosing cholangitis1(11.1)8(6.6)0(0.00)0(0.00)0.24
Median 5-HT (ng/ml)94.0[70.0,128.1]90.3[49.7,164.1]89.2[54.2,155.9]77.9[54.7,129.0]0.91
Selective serotonin reuptake inhibitor/tricyclic antidepressant022(17.6)4(16.7)7(24.1)0.43
Median mPDAI symptom scores0.0 [0.0, 0.0]2.0 [0.0, 3.0]†2.0 [0.0, 4.0]†2.0 [2.0, 4.0]†< 0.001
Median mPDAI pouch endoscopy score0.0 [0.0, 0.0]0.5 [0.0, 2.0]0.0 [0.0, 0.0]‡0.0 [0.0, 0.0]< 0.001
Median neo terminal ileum mPDAI endoscopy score0.0 [0.0, 0.0]0.0 [0.0, 1.0]0.0 [0.0, 0.0]‡0.0 [0.0, 0.0]‡< 0.001
Median cuff endoscopy score0.0 [0.0, 0.0]0.0 [0.0, 2.0]0.0 [0.0, 0.0]‡0.0 [0.0, 0.0]0.002
Total mPDAI endoscopy score0.0 [0.0, 0.0]2.0 [0.0, 4.0]†0.0 [0.0, 0.0]‡0.0 [0.0, 0.0]‡< 0.001
  • * Data not available for all subjects. Values presented as Mean ± SD, Median [P25, P75] or N (%). † Significantly different from Normal Group; ‡ Significant difference from Inflammatory Group. Ad-hoc comparisons were done using Pearson's chi-square tests for categorical variables and the Stee–Dwass multiple comparisons procedure for continuous and ordinal variables. A significance level of 0.008 (0.05/6) was considered for these.

3.2 Comparison of serum 5-HT Level among the study groups

The 5-HT levels were comparable among the 4 study groups (Table 2 and Fig. 1). The fasting median serum 5-HT level was 94.0 ng/ml (IQR, 70.0,128.1], 89.2 ng/ml (IQR, 54.2,155.9), 90.3 ng/ml (IQR, 49.7,164.1), and 77.9 ng/ml (IQR, 54.7,129.0) for normal pouch, functional pouch, inflammatory pouch, and surgical complication groups, respectively (p = 0.91). The details relating to the diagnostic groups are presented in Figs. 2 and 3. SSRI use was recorded in 4 patients in the IPS, 22 in the inflammatory pouch group, 7 in the surgical complication group and none in the normal pouch group. The factor of SSRI use was taken into consideration in the proportional odds regression analysis for the association between the serum 5-HT and PDAI symptoms and endoscopic subscores (Table 3).

Figure 1

Serum 5-HT level by diagnostic groups. 5-HT levels are comparable among different diagnostic groups, P = 0.91.

Figure 2

Serum 5-HT level at different symptom scores in different diagnostic groups.

Figure 3

Serum 5-HT level at different endoscopic scores in different diagnostic groups.

View this table:
Table 3

Association between 5-HT levels and inflammation and symptom scores in different diagnostic groups: proportional odds regression analysis.

ScoreNormalInflammatoryFunctionalSurgical complications
OR (95% CI)p-valueOR (95% CI)P-valueOR (95% CI)P-valueOR (95% CI)P-value
Symptom score1.00 (0.02, 63.7)0.991.5 (0.99, 2.4)0.0571.1 (0.49, 2.6)0.782.5 (0.76, 8.3)0.13
Pouch body endoscopy subscore1.00 (0.02, 63.7)0.991.9 (1.2, 2.9)0.0071.05 (0.13, 8.7)0.961.07 (0.35, 3.3)0.91
Afferent-limb endoscopy subscore2.8 (0.05, 156.0)0.611.1 (0.69, 1.8)0.640.47 (0.04, 6.2)0.57
Cuff endoscopy subscore1.4 (0.85, 2.3)0.181.8 (0.56, 5.9)0.32
Total endoscopy score1.7 (0.08, 36.4)0.731.8 (1.2, 2.8)0.0081.05 (0.13, 8.7)0.961.3 (0.49, 3.6)0.56
  • OR: odds ratio; CI: confidence interval; –: data not enough to perform statistical analysis.

  • Reported OR (95% CI) and p-values are for 100 ng/mL increase in 5-HT levels and are obtained from proportional odds regression models.

  • For each disease group, each score was modeled as the outcome. Adjustment for SSRI was not done for the normal pouch group since no patient in that group was on SSRI, Adjustment for SSRI was done for the other three groups.

3.3 Correlation between serum 5-HT Level and symptom and endoscopy subscores

Proportional odds regression analysis for the association of PDAI subscores and disease category/5-HT in the 4 groups is summarized in Table 3. The serum 5-HT level significantly correlated with the PDAI endoscopy subscore in the pouch (odds ratio [OR] = 1.9; 95% confidence interval [CI]: 1.2, 2.9, p < 0.05) and total endoscopy score (OR = 1.8; 95% CI: 1.2, 2.8, p < 0.05) in the inflammatory complication group only. There was no correlation between serum 5-HT level and symptoms subscore in all 4 groups.

4 Discussion

Restorative proctocolectomy with IPAA has become the surgical treatment of choice for UC patients who require colectomy. Close to 40% IPAA patients had IPS13 and 60% have pouchitis,22 the underlying mechanisms of which are not completely understood. We previously observed that patients with functional pouch disorder had EC cell hyperplasia in biopsies taken from the pouch and evidence of visceral hypersensitivity, based on barostat evaluation.14,15 This study intended to expand on the knowledge gained by our previous observations, specifically by evaluating the relation between fasting serum 5-HT and the PDAI symptom and endoscopy subscores among different diagnostic groups of pouch patients. No significant difference in 5-HT was found among the normal pouch, IPS, inflammatory pouch and surgical complication groups at the baseline level. However, there was a significant association between serum 5-HT level and the mPDAI pouch endoscopy subscore and total endoscopy score. These findings suggest that 5-HT as a key hormone may be involved in the inflammatory process of the pouch, despite it has no direct strong association with the clinical symptoms.

The exact mechanism of the regulation of 5-HT pathway in IBS patients is still unclear. As established in previous studies, 5-HT modulates the immune response, influences intestinal inflammation by acting on serotonergic receptors in lymphocytes, monocytes, macrophages and dendritic cells,23 and inhibits apoptosis of immune cells.24 Multiple inflammatory markers have been found to be involved in IBS symptoms including IL-1β, IL-6, IL-8, TNF alpha.2532 The evidence suggests that an elevated blood 5-HT level is associated with IBS symptoms.58,33 However, whether this alteration in the 5-HT level is due to an increased biosynthesis or an decreased degradation and reuptake is still controversial. Studies by Dunlop et al.4 that included a small group of IBS patients and Atkinson et al.7 demonstrated a similar proportion of increase in the postprandial level of plasma 5-HT in IBS-diarrhea predominant patients when compared to controls, which suggests the possibility of a decreased enterocytes SERT activity or a defect in mono-amino-oxidase as the underlying mechanism rather than an increased biosynthesis.

There are limited and conflicting published data regarding the role of 5-HT in IBD patients. In 1976, Ahonen reported a lower mean number of EC cells per crypt in patients with UC than controls.34 Magro et al. demonstrated a significantly lower 5-HT level in the inflamed mucosa of 22 CD and 21 UC patients.35 Other factors in the 5-HT pathway have been studied by Coats et al. in 2004.36 They illustrated that mucosal 5-HT, tryptophan hydroxylase-1 mRNA, SERT mRNA were all significantly reduced in UC, as was the EC cell population. On the contrary, two other studies reported that EC cells were significantly increased in patients with UC and CD when compared with controls.37,38 This discrepancy could be due to the complexity of tissue sampling error, the use of immunosuppressants as well as differences in inclusion and exclusion criteria used, and sample sizes.

Specifically in the circumstances where IBS condition coexists with IBD,8,9 a significantly higher level of tryptophan hydroxylase-1, the key synthetic enzyme for 5-HT has been identified in the colon of IBD patients in remission with IBS symptoms when compared to patients without IBS symptoms.8 However, a more recent study found a significantly elevated calprectin level in IBD patients in remission with IBS like symptoms. A concern for occult inflammation in this population rather than IBS has been discussed.39 Unfortunately the level of blood and tissue 5-HT was not measured in this study. The studies in the animal enterocolitis models have provided some insight into the role of 5-HT in the post infectious GI conditions and IBD. An earlier study in 1999 by Oshima et al. in a rat colitis model with dextran sodium sulfate showed EC hyperplasia and increased 5-HT content.10 In a mouse colitis model from Citrobacter rodentium infection, evidence showed increased 5-HT release from decreased number of EC cells, and a decreased SERT level in the inflamed mucosa of the colon.40 The same group also demonstrated similar findings in the Guinea-pig TNBS ileitis model although with an increased number of EC cells.41 Wheatcroft et al. reported similar findings in a mouse model post Trichinella spiralis infection with increased small bowel EC cells and reduced SERT expression.42 Instead, Linden et al.11,43 documented in mouse and guinea-pig TNBS colitis model that an increase in mucosal 5-HT content could be due to decreased SERT level only rather than increased 5-HT releasing EC cells. In contrast, mice lacking a functional T-cell receptor α chain spontaneously develop colitis and these mice were found to have reduced numbers of 5-HT expressing cells in the colon.44 None of the studies in animal colitis models or IBD patients measured the blood 5-HT level, which could have missed an important aspect of its regulation as illustrated in IBS patients.

The patient population of our study with ileal pouches and a background of UC shares aspects of both IBS and IBD. Serum 5-HT level was shown to be positively associated with pouch endoscopic inflammation subscore and total endoscopy score in the inflammatory group only. This suggests the 5-HT signaling pathway was being involved in the inflammatory process in pouchitis and CD of the pouch, although these data will not adequately support the use of serum serotonin as the non-invasive marker for pouchitis, it certainly sheds some light on the potential role of 5-HT receptor inhibitors as a new potential therapeutic option for these groups of patients, especially those with antibiotic refractory pouchitis. On the other hand, contrary to our original hypothesis, our findings do not support a significant contribution of serum 5-HT to functional pouch disorders. The lack of tissue 5-HT levels and its correlation with serum 5-HT in this study could explain this conflicting result from our previous publications.15 Further study on the tissue level will help to further clarify the role of 5-HT on the inflammatory mechanism of the pouch.

There are limitations to our study. As potential confounding factors, personal and family history of IBS was not incorporated in our analysis due to the concern of unreliable data extraction from inadequate documentation in the retrospective records. The patient population at our tertiary care center could have referral and selection bias. Since multiple adverse pouch complications were measured, the study might have been underpowered with type II errors. The findings of this hypothesis-generating project will enable us to do future studies by modifying the sample size and adding additional markers. Important downstream components of 5-HT pathway, such as the 5-HT receptors and SERT could also play a critical role, which were not included in this study. Therefore, a larger, prospective study to evaluate the expression level of 5-HT receptors and SERT in the tissue level and fecal 5-HT would expand our understanding of the role of 5-HT pathway in the pouch diseases as well as concomitant IBD and IBS.

In conclusion, serum 5-HT levels were comparable among different diagnostic groups of the pouch. Serum 5-HT level was not found be associated with symptoms subscores in all pouch disorders, however, appeared to be associated and correlated with a higher pouch endoscopic inflammation subscore in patients with pouchitis, CD of the pouch, and cuffitis, suggesting a potential role for serum 5-HT as either a contributing factor for the inflammation in the pouch or instead an effect of disease activity in these patients. Further research at a mucosal level is needed to determine whether this is an epi-phenomenon or is potentially causative.


Yinghong Wang carried out the acquisition and interpretation of data, and drafted the manuscript. Hua Gong performed the measurement of serum serotonin level. Rocio Lopez and Lei Lian participated in the statistical analysis. Ravi Kiran and Edy Soffer revised the draft and made intellectual input on the study design. Bo Shen designed the study, revised the draft and finalized the manuscript to be submitted. All authors read and approved the final manuscript.

The authors are grateful to the American College of Gastroenterology research award to Yinghong Wang for the financial support of the project.


  • Conflict of interest: The authors declared no financial conflict of interest. Hua Gong is employed by the Prometheus Laboratories.

Crohn's disease
inflammatory bowel disease
irritable bowel syndrome
ileal pouch-anal anastomosis
irritable pouch syndrome
non-steroidal anti-inflammatory drugs
the Pouchitis Disease Activity Index
primary sclerosing cholangitis
serotonin reuptake transporters
ulcerative colitis.


View Abstract