Altered Small Bowel Motility in Patients

with Liver Cirrhosis Depends on Severity of

Liver Disease

ANA MARIÂ A MADRID, MD, FRANCISCO CUMSILLE, and CARLOS DEFILIPPI, MD

Abnormal small bowel motility has been described in patients with liver cirrhosis but themechanisms involved are unknown. The aim was to investigate a possible relationshipbetween the severity of liver failure and the intensity of small intestinal abnormalit ies.Motility was studied during fasting, by means of perfused catheters and external transducers,on 33 cirrhotics with different etiologies; 8 were at Child-Pugh stage A, 12 stage B, and 13stage C. Both abnormalit ies of MMC and increased clustered activity were recorded. Absenceof cycling activity was most frequently observed in Child-Pugh stage C patients compared toChild-Pugh stage A cirrhotics. A signi® cant increase in clustered contractions from 4.7 60.4/hr in stage A patients to 11.3 6 1.1 in stage C was recorded. The frequency and amplitudeof contractions was also increased from Child-Pugh stage A to stage C. Our ® ndings might berelated to a delayed transit time observed in these patients and a higher prevalence ofbacterial overgrowth in cirrhotics with more advanced liver disease.

KEY WORDS: liver cirrhosis; small bowel motility.

Altered small intestinal motility was described in pa-

tients with liver cirrhosis by Chesta et al (1). Their

study revealed a prolonged duration of the migrating

motor complex (MMC), related in large part to a

longer phase II, in cirrhotics when compared to nor-

mal controls. However, the most striking ® nding was

a predominance of clustered contractions during

phase II. This pattern featured repetitive bursts of

3± 6 rhythmic contractions separated by quiescence.

The prevalence of these motor patterns appeared

unrelated to alcohol intake; ® ve of the 16 cirrhotic

patients were nonalcoho lics. No differences in motor

functions were seen between patients with and with-

out bacterial overgrowth, and treatment with tetracy-

cline did not modify this altered small intestinal mo-

tili ty (2). The occurrence of clusters was also

unrelated to glucagon plasma levels (3) or histological

abnormalit ies in neurons of the enteric nervous sys-

tem (4).

Considerable variability in motility was evident,

however, within this patient population with respect

to both the MMC alterations and the intensity of

clustered contractions. Although this study failed to

demonstrate a correlation between dysmotilit y and

Child-Pugh stage, the size of the individual patient

Child-Pugh groups may have been too small for ade-

quate statistical analysis. The aim of our study was:

(1) to reassess, in a larger group of patients with

cirrhosis, the prevalence of altered small intestinal

motility and (2) to explore the in¯ uence of liver

disease severity, on the different patterns of altered

motility.

Manuscript received May 23, 1996; revised manuscript receivedNovember 15, 1996; accepted December 9, 1996.

From the Gastroenterology Center, Department of Medicine,University Hospital, University of Chile, Santiago, Chile.

Address for reprint requests: Dr. A.M. Madrid, Gastroenterol-ogy Center, University Hospital J.J. Aguirre, University of Chile,Santos Dumont 999, Gastroenterology Section, Santiago, Chile.

Digestive Diseases and Sciences, Vol. 42, No. 4 (April 1997), pp. 738 ± 742

738 Digestive Diseases and Sciences, Vol. 42, No. 4 (April 1997)

0163-2116/97/0400-0738$12.50/0 Ñ 1997 Plenum Publishing Corporation

MATERIALS AND METHODS

Patients. Thirty-three patients entered the study (24 menand 9 women; mean age 55.8 years, range: 28 to 77).Cirrhosis was of alcoholic origin in 25 subjects; in theremaining eight patients the etiology was related to hepa-titis C virus infections (HVC) in three, to both excessivealcohol intake and HCV antibodies in two patients, and theetiology remained unknown in three. In this study nohealthy subjects were studied as a control group. The diag-nosis in each case was con® rmed by appropriate investiga-tions including laboratory, liver biopsy, and endoscopy.According to the Child-Pugh criteria (5), 8 patients wereclassi® ed as stage A; 12 as stage B, and 13 as stage C.Patients with diabetes mellitus, marked serum electrolytedisturbances, previous abdominal surgery and renal failure,or those who had been treated with lactulose, antibiotics, orprokinetic drugs during the past 30 days were excluded. Thecharacteristics of the patient groups are shown in Table 1.

Motility Studies. Duodenal motor activity was studied bymeans of a multilumen perfused catheter assembly whichincorporated four polyvinyl tubes (0.9 mm ID) glued to-gether with tetrahydrofuran and with side holes spaced at 3cm from each other. They were continuously infused withbubble-free distilled water at a rate of 0.4 ml/min with apneumohydraulic capillary infusion system (ArndorferMedical Specialities, Greendale, Wisconsin). The mano-metric catheters were attached to external TP-400t pressuretransducers and connected to a Nihon Kohden polygraph(Nihon Kohden Co., Tokyo, Japan).

Procedures. Following an overnight fast, the catheterassembly was passed by mouth to the stomach and ad-vanced under ¯ uoroscopic control until the tip of the as-sembly had reached the angle of Treitz, the recording sitesbeing located in the third and fourth part of the duodenum.Studies were performed in the fasting state only.

Analysis of Motor Activity Data. The frequency andamplitude of individual contractions were determined bydirect visual inspection of the chart recording paper. Anintestinal motility index (IMI) was calculated as the productof the mean amplitude and frequency of contractionsthroughout the study. The different phases of the MMCwere identi ® ed using the following criteria: phase I wascharacterized by the complete absence of contractions;phase II consisted of irregular phasic contractions, whichculminated in a burst of rhythmic phasic contractions at afrequency of 12 cpm (phase III). To de® ne MMC activity,we required the presence of at least one phase III precededby phase II and followed by phase I. Clustered contractionswere de® ned as a sequence of 3± 10 rhythmic contractionspreceded and followed by a quiescent period of 1± 5 min.

According to previous studies in the fasting period (6),more than six clusters per hour occurring at all four siteswas considered abnormal.

Statistics. In order to compare the amplitude, frequency,IMI, and cluster frequency among the groups, analysis ofvariance methodology was used. When the null hypothesiswas rejected, the Tukey’ s studentized range test for multiplecomparison at 0.05 level was conducted. SAS (7) softwarewas used in the data analysis.

RESULTS

The average duration of individual studies was

240 6 11 min (range 135± 432). Two types of changes

were observed: (1) abnormalit ies of the cyclic activity

of the MMC, and (2) the appearance of an abnormal

pattern of contractions. Cyclic activity was observed

in 17 patients; in the remaining 16, a pattern of

continuous phase II type contractive activity was re-

corded over 226 6 8.5 min. Cyclic activity was ob-

served in 5 (62.5%) of Child-Pugh class A, in 6 (50%)

Child-Pugh class B, and in 6 (46.1%) of Child-Pugh

class C (P , 0.05 Child-Pugh class A vs C).

An abnormal pattern consisting of an increased

number of clustered contractions was seen in both the

presence and absence of the MMC. Abnormal clus-

tering was not observed in any Child-Pugh class A

patients, but was present in 6 (50%) of Child-Pugh

class B and in 10 (77%) of Child-Pugh class C patients

(P , 0.05 A vs B, A vs C). Frequency of clusters

signi® cantly increased from 4.7 6 0.4 clusters/hr in

Child-Pugh class A patients to 11.3 6 1.1 clusters/hr

in class C cirrhotics (P , 0.05) (Table 2).

In terms of the overall organizatio n of motor activ-

ity, four separate patterns were therefore observed

(Figure 1): (1) normal MMC activity with sequential

phases I, II, and III; (2) cyclic activity with predomi-

nant clustered contractions during phase II of the

MMC; (3) absent cyclic activity: persistent phase II-

type activity; and (4) absent cyclic activity with pre-

dominant clustered contractions throughout the re-

cording. A characterist ic type 2 pattern is shown in

Figure 2.

The prevalence of each one of these patterns in

TABLE 1. CHARACTERISTICS OF CIRRHOTIC PATIENTS ACCORDING TO CHILD-PUGH STAGE

Child-Pugh stage N Age (yr)

Sex Etiology*

M F Alcohol HVC Alcohol 1 HVC ?

A 8 49.5 6 3.8 7 1 7 1 0 0

B 12 59.1 6 2.5 9 3 8 2 2 1C 13 58.8 6 3.5 8 5 10 0 0 2

* HVC 5 Hepatic C virus infections; ? 5 cryptogenic cirrhosis.

SMALL BOWEL MOTILITY AND CIRRHOSIS

739Digestive Diseases and Sciences, Vol. 42, No. 4 (April 1997)

each Child-Pugh stage group is shown in Table 3.

Normal motor activity was the most common pattern

in Child-Pugh stage A cirrhotics; in contrast, absent

cyclic activity and a predominance of clustered con-

tractions was most frequently observed in Child-Pugh

stage C patients.

Statistically signi® cant differences in amplitude,

frequency of contractions, and IMI were observed

among patients with different Child stages: a statisti-

cally signi® cant increase of amplitude (P , 0.05) was

observed between cirrhotics in Child-Pugh stage A

compared to stages B and C. A progressive increase

TABLE 2. PARAMETERS OF SMALL INTESTINAL CONTRACTILE ACTIVITY IN PATIENTS

WITH LIVER CIRRHOSIS ACCORDING TO CHILD-PUGH STAGE*

Child-Pugh stage

Amplitude

(mm Hg)

Frequency

(C/min) IMI

Clusters

(Cl/hr)

A 24.1 6 4.3² ³ 0.9 6 0.2² 25.7 6 9.5² 4.7 6 0.4²

B 37.2 6 2.3 1.20 6 0.13 45.1 6 5.2 8.3 6 1.3C 37.9 6 1.4 1.6 6 0.2 60.9 6 6.2 11.3 6 1.1

* Abbreviations: IMI 5 intestinal motor index; C 5 contractions; Cl 5 clustered

contractions.² P , 0.05, A vs C.

³ P , 0.05, A vs B.

Fig 1. Schematic representation of the different patterns of abnormal small bowel motility observed inpatient with liver cirrhosis. Type 1 pattern: normal MMC, type 2: cyclic activity with predominant

clustered contractions during phase II of the MMC, type 3: absent cyclic activity with persistent irregularcontractions, type 4: absent cyclic activity with predominant clustered contractions.

Fig 2. Tracing showing phase III of the MMC followed by a typical phase I. The preceding phase II is characterizedby the presence of clusters of contractions separated by periods of quiescence (type 2 pattern).

MADRID ET AL

740 Digestive Diseases and Sciences, Vol. 42, No. 4 (April 1997)

of frequency of contractions and of IMI was observed

from Child-Pugh stage A to C. Statistical ly signi® cant

differences were seen between stage A and C of the

Child-Pugh classi® cation (P , 0.05).

DISCUSSION

Our study has con® rmed the previous observations

of Chesta et al of altered small intestinal motor ac-

tivity in patients with liver cirrhosis (1). We thus

observed both alterations in cyclic activity during fast-

ing and the appearance of prominent clustered con-

tractions. We have additional ly observed an overall

increase in both the frequency and amplitude of con-

tractions during fasting.

In contrast to the original communication (1), we

noted in this study a direct relationship between these

changes in motility and the severity of liver disease.

Thus, the preservation of normal cyclic activity was

signi® cantly greater in Child-Pugh stage A patients, in

comparison to the other stages. The most striking

difference was related to the appearance of clustered

contractions: they were observed in a normal fre-

quency in Child-Pugh stage A patients; in contrast

77% of Child-Pugh stage C patients demonstrated an

increased number of these contractions. These differ-

ences with the previous results may be simply ex-

plained by the larger number of patients, both in total

and within each Child-Pugh group included in the

study. When the presence of both cyclic activity ab-

normalities and clusters of contractions were ana-

lyzed together, a clearer relationsh ip to altered he-

patic function was observed. The amplitude and

frequency of contractions, as well the frequency of

clusters, also increased from Child-Pugh stage A to C.

The mechanisms involved in the motor changes

observed in patients with liver cirrhosis are unknown.

Divergent results have been observed in studies of

plasma glucagon levels in patients with liver cirrhosis

and several portal hypertension animal models (8).

Previous studies have shown that motor abnormalit ies

are unrelated to plasma glucagon levels (3). Another

study did not show changes in the number of neurons

of the enteric nervous system (4). Our observation

also con® rms that changes were unrelated to the

etiology of liver disease.

Since our study was not designed to analyze the

mechanisms involved, we can only speculate about

the relationship between chronic liver failure and

altered motility. Our observation s indicate that the

motor changes are unlikely to be myogenic, since the

amplitude of contractions increased with advancing

liver disease. The possibility that hypoalbum inemia

and edema of the small intestinal wall might be in-

volved is dif® cult to accept because a decreased am-

plitude of contractions would be expected in this case,

rather than the appearance of such discrete patterns

as clusters of contractions. The patterns of motility

observed in these patients are more reminiscent of

those associated with altered functions of the enteric

or autonomic nervous systems. Clear evidence of

autonomic dysfunction has been found in patients

with end-stage chronic liver disease of various etiol-

ogies. These abnormalit ies improve following liver

transplantation, suggesting that they are directly re-

lated to abnormal liver function (9). Are the changes

in motility dependent on portal hypertension or on

metabolic derangements that occur in these patients

as a consequence of portosystemic venous shunting

and/or liver failure? Reilly et al (10) showed that

portal hypertension induced by staged portal vein

ligation was followed by a slowing of small intestinal

transit in rats, indirectly suggesting an altered motor

activity. Van Thiel et al (11) have suggested that

similar neuronal alterations, as observed in the cen-

tral nervous system, might be present in neurons of

the enteric nervous system in patients with advanced

liver disease and portosystemic encephalopathy. To

con® rm this hypothesis , the relationships between

ammonia, benzodiazepine-like compounds, amino ac-

ids, and motility need to be examined.

Altered motility might in turn be related to the

well-established observation of a delayed transit

through the small intestine in patients with liver cir-

rhosis (12, 13), which in turn might facilitate the

appearance of bacterial overgrowth. Alterations in

jejunal micro¯ ora have been described in patients

with liver cirrhosis (2, 14, 15). The observation by

Casafont and De las Heras (16) that the prevalence of

bacterial overgrowth was signi® cantly higher in cir-

rhotics with Child-Pugh class C than in patients with

class A are in accordance with the observation of the

TABLE 3. RELATIONSHIPS BETWEEN OVERALL ORGANIZATION OF

SMALL INTESTINAL MOTILITY AND CHILD-PUGH STAGES

Child-Pugh stage

Patterns*

1 2 3 4

A 5² 0 3 0B 0 6 4 2

C 1 5 2 5

* 1 5 normal MMC, 2 5 MMC with predominant clustered con-

tractions, during phase II, 3 5 continuous irregular contractions,MMC absent, 4 5 predominant clustered contractions, MMC

absent.² Number of patients.

SMALL BOWEL MOTILITY AND CIRRHOSIS

741Digestive Diseases and Sciences, Vol. 42, No. 4 (April 1997)

present study of more intense abnormalit ies of small

intestinal motility in patients with more advanced

disease.

We conclude that several changes of small intesti-

nal motility are frequently observed in patients with

liver cirrhosis; the frequency and intensity of these

abnormalit ies increase with advanced stages of liver

disease.

REFERENCES

1. Chesta J, De® lippi C, De® lippi C: Abnormalities in proximal

small bowel motility in patients with cirrhosis. Hepatology

17:828 ± 832, 1993

2. Chesta J, Silva M, Thompson L, Del Canto E, De® lippi C:

Small intestine bacterial overgrowth in patient with hepatic

cirrhosis. Rev Med Chile 119:626 ± 632, 1991

3. Chesta J, Generini G, Antezana C, De® lippi C: Are distur-

bances of small bowel motility in patients with liver cirrhosis

related to hyperglucagonemia? Rev Med Chile 120:988 ± 1002,

1992

4. Chesta J, Smok G: Myenteric plexuses of the small intestine in

patients with liver cirrhosis: A histological study. Rev Med

Chile 121:139 ± 143, 1993

5. Christensen E, Schlichting P, Fauerholdt L, Glu C, Andersen

PK, Juhl E, Poulsen H, Tygstrup N: Prognostic value of Child-

Turcotte criteria in medically treated cirrhosis. Hepatology

3:430 ± 435, 1984

6. Summers RW, Anuras S, Green J: Jejunal manometry patterns

in health, partial intestinal obstruction, and pseudoobstruction.

Gastroenterology 85:1290 ± 1300, 1983

7. SAS Institute Inc: SAS/STAT User’ s Guide, Version, 4th ed.,

vol 1. Cary, North Carolina, SAS Inc, 1989, 943 pp

8. Johnson TJ, Quigley EMM, Adrian TE, Jin G, Pikkers LF:

Glucago n, stress, and portal hypertension. Dig Dis Sci

40:1816 ± 1823, 1995

9. Hendrickse MT, Thululath PJ, Triger DR: Natural history of

automic neuropathy in chronic liver disease. Lancet 399:1462±

1464, 1992

10. Reilly JA, Quigley EMM, Forst C, Rikkers LF: Small intestinal

transit in the portal hypertensive rat. Gastroenterology

100:670 ± 674, 1991

11. Van Thiel DH, Fagiuoli S, Wright HI, Chien MC, Gavaler JS:

Gastrointestinal transit in cirrhotic patients: Effect of hepatic

encephalopathy and its treatment. Hepatology 19:67± 71, 1994

12. Chesta J, Lillo R, De® lippi C, Jouanne F, Massone MA,

Maulen M, Zavala A: Mouth to caecum transit time and solid

meal gastric emptying in patients with liver cirrhosis. Rev Med

Chile 119:1284, 1991

13. Galati JS, Holdeman KP, Bottjen PL, Quigley EMM: Delayed

small intestinal transit in chronic liver disease. Gastroenterol-

ogy 106:A896, 1994

14. Lol D, Gorbach SL, Levitan R: Intestinal micro¯ ora in patients

with alcoholic cirrhosis: Urea-splitting bacteria and neomycin

resistance. Gastroenterology 62:275± 279, 1972

15. Bade CH, Kolepke R, Schofer K, Babe JC: Hydrogen breath

test in patients with alcoholic liver disease: Evidence for bac-

terial overgrowth. J Hepatol 11(suppl 2):S9, 1990

16. Casafont F, De las Heras G: Small bowel bacterial overgrowth

in patients with alcoholic cirrhosis. Dig Dis Sci 40:1252± 1256,

1995

MADRID ET AL

742 Digestive Diseases and Sciences, Vol. 42, No. 4 (April 1997)