Impaired gastrocolonic response and peristaltic reflex in slow-transit constipation: role of 5-HT3 pathways

Colonic motility is modulated by the 5-hydroxytryptamine (5-HT)3-dependent gastrocolonic response and 5-HT3-independent peristaltic reflex. We compared descending colon tone responses to antral distension, duodenal lipid perfusion, and colonic distension after double-blind placebo or granisetron in 13 healthy volunteers and nine slow-transit constipated patients. Antral distension (100–300 ml) and duodenal lipids (3 kcal/min) evoked increases in colon tone in volunteers, which were blunted in constipated patients ( P < 0.05). Granisetron (10 μg/kg) reduced responses to antral distension and lipids in volunteers and to lipids in constipated patients ( P < 0.05). The ascending contraction of the peristaltic reflex was blunted in constipated patients ( P < 0.05), whereas descending responses were similar. Granisetron did not modify the peristaltic reflex. Colonic responses to bethanechol were similar in patients and volunteers. In conclusion, antral distension- and duodenal lipid-activated gastrocolonic responses and ascending contractions of the peristaltic reflex are impaired with slow-transit constipation with loss of both 5-HT3-dependent and -independent function. Thus abnormalities of neural reflex modulation of colonic motor function may play pathophysiological roles in slow-transit constipation.

Differential 5-HT3 mediation of human gastrocolonic response and colonic peristaltic reflex

Colonic motor function is modulated by extended and local neural reflexes involving unknown mediators. To test the role of serotonin (5-HT3) pathways, increases in colonic tone during antral distension and duodenal lipid perfusion (gastrocolonic responses) and changes in orad and caudad colonic tone in response to local colonic distension (peristaltic reflex) were measured after double-blind granisetron (10 μg/kg) or placebo infusion in healthy human volunteers. Antral distension evoked increases in colonic tone, which were blunted by granisetron ( P < 0.05) without effects on antral compliance. Intraduodenal lipid perfusion also evoked increased colonic tone, which was reduced by granisetron ( P < 0.05). In contrast, orad colonic contractions and caudad relaxations and contractions during colonic distension were unaffected by granisetron. In conclusion, 5-HT3 receptor antagonism blunts both the mechano- and chemoreceptor components of the human gastrocolonic response without altering antral compliance. In contrast, 5-HT3 pathways play no role in the ascending or descending components of the colonic peristaltic reflex. These findings demonstrate different roles for 5-HT3 receptors in the control of colonic motor function by the proximal gastrointestinal tract and by local neural reflexes.