In the present study, we evaluated the small intestinal motility and transit in patients with liver cirrhosis and clinically significant portal hypertension. We used a novel magnet-based MTS-1 that provides detailed information on intestinal motility. Our main finding was that patients with cirrhosis and portal hypertension had faster-than-normal postprandial transit through the proximal small intestine. Thus, our hypothesis of reduced small bowel motility and prolonged segmental gastrointestinal transit times was rejected.
The basic and highly characteristic contraction frequencies of the stomach and small intestine were unaltered by portal hypertension, but the distance was covered faster both during the long periods with relatively slow transit and during the short bursts of very fast movement. Madsen et al. studied eight patients with cirrhosis and portal hypertension using scintigraphy
. They found no difference in the GE of solids or small intestinal transit times compared to healthy controls. The scintigraphy technique is considered the gold standard for measuring GI transit times. However, standard 30 minute-intervals between each image were used, and smaller differences in either GE or small intestinal transit times could thus have been missed. In the same study, cirrhosis patients showed faster-than-normal colonic transit times
. It has previously been demonstrated that patients with cirrhosis have a prolonged phase II of the fasting small intestinal migrating motor complex (MMC)
. We obtained data on small intestinal motility in the postprandial state, but our data also suggest an abnormally active small intestine in patients with liver cirrhosis.
Some previous studies have suggested that patients with liver cirrhosis have prolonged GE
[14, 15], while others found no difference
. The main focus of the present study was small intestinal motility, aiming for an exact determination of pyloric passage that MTS-1 provides
. As a detailed evaluation of GE should be performed in both the solid and the liquid phases, MTS-1 is not ideal for the description of GE. Large particles will typically pass through the pylorus during phase III of the migrating motor complex
. Unless magnet intake is standardised with respect to MMC or with a meal, a large variation in solid-state GE should be expected, as observed in the present study.
Our findings are in contrast with some previous studies that demonstrated prolonged gastrointestinal transit times in patients with liver cirrhosis. However, those studies investigated the total orocaecal transit times with radiopaque markers
 or lactulose breath tests
. These methods do not differentiate between GE and small intestinal transit time, and differences in small intestinal transit may not be detected, especially if GE is prolonged. Furthermore, patients with liver cirrhosis often suffer from small intestinal bacterial overgrowth, making the lactulose breath test unreliable
. Other studies have shown both prolonged and reduced colonic transit times in cirrhotic patients
[6, 20]. However, GITT mainly reflects colonic transit time, and in our study, there was no difference in GITT between patients with cirrhosis and controls.
It has previously been shown that a history of SBP correlates to the degree of GI alterations in patients with liver cirrhosis
. Additionally, there are indications that abnormalities in small intestinal motility are related to the degree of chronic liver failure
. In accordance with the present study, Madsen et al. found no correlation between transit times and HVPG
. However, this may be a type II error due to the small number of patients included in both studies.
Our findings were surprising and contrary to our hypothesis. It is unclear through which mechanisms portal hypertension may affect gastrointestinal motility. It may be that portal vein and intestinal wall blood stasis may cause an overactive bowel. We studied small intestinal motility in the postprandial state. In the fasting state, a marked change in the contraction patterns of phase II and the MMC has been previously observed
. Whether this behaviour indicates an overactive bowel is unknown. In healthy individuals, intestinal peristalsis, gastric acid and mucosal immunity act to protect the small intestine from bacterial overgrowth and translocation. Breath tests for small intestine bacterial overgrowth are not reliable for research purposes
, and cultures from small intestinal fluid require invasive investigation. For these reasons, we did not investigate whether patients in the present study had small intestine bacterial overgrowth.
Previous studies have suggested that the aetiology of liver cirrhosis may influence GI transit times
[3, 23]. In the present study, we included patients consecutively and independently of cirrhosis aetiology. The mixed aetiology and differing representation of Child-Pugh classes in a small sample size are major limitations of the study. The main aetiology of cirrhosis was alcohol consumption, and it is well known that alcohol may affect gastrointestinal motility. However, all but two patients had abstained from drinking for at least 3 months prior to the investigation. Pancreatic insufficiency may also reduce small intestinal transit. However, none of the patients had a history of acute or chronic pancreatitis or malabsorption, and no signs of calcification in the pancreatic area were noticed on abdominal radiographs. Furthermore, autonomic dysfunction in patients with liver cirrhosis may lead to gastroparesis and an abnormal frequency of gastric contractions
. Compared to healthy controls, we found no differences in the contraction frequency in either the stomach or the small intestine.
There was a difference in gender between our patients and the control group. However, when stratifying the data in each group by gender, we could not detect any differences in GE, small intestinal velocity or colonic transit time. Ten out of 15 patients had a history of bleeding oesophagus varices. Oesophago-gastro-doudenoscopy had been performed in all patients, and 14 had oesophageal varices. We found no association between the presence of varices and motility patterns.