Digestive Diseases and Sciences, Vol. 35, No. 9 (September 1990), pp. 1089-1097

Neurohumoral Control of Gallbladder Motility in Healthy Subjects and Diabetic Patients with or Without Autonomic Neuropathy STEFANO FIORUCCI, MD, RACHELE BOSSO, MD, LUCIANO SCIONTI, MD, SILVANA DISANTO, MD, BRUNO ANNIBALE, MD, GIANFRANCO DELLE FAVE, and ANTONIO MORELLI, MD

Patients affected by diabetes mellitus are reported to have an increased incidence of gallbladder abnormalities. The pathophysiologic mechanisms for this phenomenon are unclear. In the present study ultrasonography was used to determine gallbladder emptying in response to a meal Or separate cephalic or hormonal stimulation in 21 diabetic patients and 10 healthy subjects. Gallbladder emptying and refilling after a meal were similar in diabetic patients and healthy subjects. When diabetics were divided according to the presence or absence of cardiac autonomic neuropathy (AN), a significant reduction of gallbladder emptying in response to cephalic stimulation was found in diabetics with A N (P < 0.01 in comparison with diabetics without A N or healthy subjects). A dose-response curve of gallbladder emptying in response cerulein, a cholecystokinin analog, at concentrations of 0.25, 1, and 4 txg/kg/min was evaluated. No differences of gallbladder emptying were found in the three groups of subjects, indicating that gallbladder sensitivity to hormonal stimulation is not changed in diabetic patients with or without AN. Diabetic patients with A N have a significant reduction of gastric acid output and pancreatic polypeptide (PP) secretion in response to cephalic stimulation (P < 0.05 in comparison with diabetic patients without A N or healthy subjects). Cerulein-induced PP secretion was Similar in all three groups of subjects (P > 0.05). This study indicates that in diabetic patients with AN, gallbladder emptying as well as gastric acid and PP secretions induced by neural stimulation are markedly reduced in comparison to diabetics without AN. KEY WORDS: ultrasonography; vagal stimulation; pancreatic polypeptide; gastric acid secretion.

Abnormalities in gallbladder motility are thought to be common in diabetic patients and are considered Manuscript received December 12, 1989; revised manuscript received April 11, 1990; accepted April 12, 1990. From the Isituti di Clinica Medica I, Cattedra di Gastroenterologia, and Patologia Medica Speciale, Universit~ di Perugia and Cattedra di Gastroenterologia I, Universit~i " L a Sapienza," Roma, Italy. Address for reprint requests: Dr. Stefano Fiorucci, Clinica Medica I, Policlinico Monteluce, 06100 Perugia, Italy.

to contribute to the increased risk of gallbladder disease reported in this population (1, 2). Decreased gallbladder emptying may be a key factor in the pathogenesis of gallstones. Unfortunately, gallbladder motility has not been evaluated extensively in patients with diabetes mellitus, and available studies report conflicting results (3, 4). Moreover, in contrast with earlier views (1), population surveys and autopsy studies have failed to reveal any asso-

Digestive Diseases and Sciences, Vol. 35, No. 9 (September 1990)

0163-2i 16/9(I/0900-1089506.00/09 1990PlenumPublishingCorporation

1089

FIORUCCI ET AL TABLE 1.

CLINICALCHARACTERISTICSOF PATIENTS IN STUDY Diabetic population Without cardiac automatic neuropathy (N = 12)

With cardiac automatic neuropathy (N = 9)

Healthy subjects (N = 10)

Age (years)* Gender (female/male) Type of diabetes I II

35 (19-58) 7/5

42 (30-53) 5/4

33 (27-41) 7/3

9 3

7 2

Known duration of diabetes (years) Weight (% of ideal body) Diabetic complications Retinopathy Peripheral neuropathy Nephropathy HbAlct

10.6 (3-8) 106 --- 9

13.7 (1-26) 105 -+ 8

1/12 3/12 3/12 7.1 -+ 2.6

9/9 8/9 5/9 9.0 + 2.2

103 -+ 6

*Mean and range or mean -+ SD. tNormal values < 3.8-6.2%.

ciations between diabetes mellitus and gallbladder stones (5, 6). Cardiac autonomic neuropathy (AN), a frequent complication of long-standing diabetes mellitus, is associated with a complex deterioration of parasympathic functions involving different systems (7). Several studies indicate, in diabetic patients with AN, the parasympathic (vagal) regulation of some gastrointestinal functions is impaired (7). The term " a u t o v a g o t o m y " has been suggested to characterize this condition (8). Despite the fact that the vagus plays a major role in the control of gallbladder motility, and the gallbladder contracts in response to cephalic stimulation (9), the effects of such autovagotomy on gallbladder emptying induced by neural stimulation in patients with diabetes mellitus are still unknown. The present study was designed to evaluate whether patients with diabetes mellitus, with and without AN, present any alteration of gallbladder motility in response to either a meal or separate cephalic or hormonal stimulations. Gastric acid secretion and pancreatic polypeptide (PP) release are both regulated by long vagal pathways (10-12). Gastric acid secretion and PP release induced by cephalic stimulation are reported to be reduced in patients with diabetes mellitus and AN (13). To determine if alterations of these functions parallel gallbladder dismotility, we evaluated gastric acid secretion and PP release during neural and hormonal stimulation of gallbladder emptying. MATERIALS AND METHODS Twenty-one diabetic patients with a mean age of 38.1 years (range 19-53 years) were recruited for the study

1090

(Table 1). Patients were selected who had a functioning gallbladder documented by visualization with ultras0nography of contractions after a meal. Criteria for exclusion from the study were obesity (> 125% ideal body weight) and gastrointestinal or hepatobiliary disease (documented by elevation of serum aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, or bilirubin and clinical or sonographic evidence of disease). Patients with gallstones or previous vagotomy were excluded. The presence and type (either type 1 or type 2) of diabetes mellitus was determined according to the criteria proposed by the National Diabetes Data Group (14). Of the 21 patients studied, 16 were affected by type 1 and five by type 2 diabetes mellitus. All type-1 diabetic patients were treated with insulin. Four of the type-2 diabetic patients were treated with oral hypoglycemic agents and one with diet alone. Ten healthy subjects (seven males and three females), mean age 33 years (range 27-41 years), free of gastrointestinal disease, were included in the study after informed consent had been given. The percentage of ideal body weight was 103 -+ 6%. All subjects volunteered and gave fully informed consent. Sonographic criteria for inclusion in the protocol were: (1) ellipsoid form of gallbladder during fasting, without angulation or septa, (2) fasting gallbladder volume >12 ml, and (3) gallbladder volume reduction after the mixed meal >50%. Control subjects and diabetic patients did not differ significantly in age, sex, and percentage of ideal body. Diabetic patients were classified into two groups according to the presence or absence of cardiac AN (7). Five cardiovascular tests were used to investigate autonomic function: (1) heart response to Valsava maneuver, (2) heart rate variation during deep breathing, (3) blood pressure response to sustained handgrip, (4) immediate heart rate response to standing, and (5) blood pressure response to standing. An autonomic score from 0-10 (0 normal, 10 worst) was derived for each patient. One or more of the following symptoms was present in each subject: impotence, postural hypotension, urinary symptoms. Impotence was not considered sufficient evidence of AN because a vascular rather than autonomic dysfunction could account for this Digestive Diseases and Sciences, Vol. 35, No. 9 (September 1990)

GALLBLADDER MOTILITY IN DIABETIC PATIENTS symptom. The presence of a peripheral neuropathy was determined by measuring motor and sensory conduction along the tibial nerve and sensory conduction of the sural nerves using standard techniques. Study Protocol. Subjects were studied on three nonconsecutive days. All drugs except insulin were discontinued 12 hr before the study. On the first day, after an overnight fast, gallbladder motility was evaluated after ingestion over 15 min of a solid-liquid meal of 800 kcal, consisting of a standard hospital food such as pasta, beef with sauce, and vegetables, and three glasses of water. Gallbladder contraction and refilling were monitored with sonography and images taken every 15 min for up to 150 min. In the second study, a nasogastric tube was passed into the stomach under fluoroscopic control. Gastric secretion was aspirated continuously using a peristaltic pump (Minipulse II, Gilson) and samples collected every 10 min. The titrable acidity was determined by titrating each sample to pH 7.0 with 0.1 M NaOH, using a titration apparatus (pH-Stat, ABU 80, and 75-pHMeter, all Radiometer, Copenhagen, Denmark). Modified sham-feeding (MSF) was carried out according to the method of Feldman et al (10) and as reported elsewhere (15). Briefly, after a 60-min basal period, each subject chewed and spat an appetizing meal consisting of 200 g of veal, 180 g of bread, and 300 ml of water during 30-min period. The meal was prepared in a separate building so that subjects could not see or smell the food until the time of sham feeding. Basal acid output (BAO) was measured as a sum of six consecutive 10-min outputs. Sham-feeding-stimulated acid output (SAO) was measured during six periods of 10 min each (30 min during and 30 rain after MSF). Gallbladder motility during MSF was monitored by ultrasound and images were taken every 10 min for 60 min. During all experiments a forearm vein was used for blood sampling. Blood samples for determination of pancreatic polypeptide (PP) concentrations in response to MSF stimulation were taken at - 10, 0, 10, 20, 30, 40, 50, and 60 min in relation to the meal. Samples were collected in EDTA plus aprotinin and plasma, separated by centrifugation, and stored at - 2 0 ~ C until assayed. In the third experiment, gallbladder emptying was evaluated during continuous infusion of cerulein (Takus, kindly donated by Farmitalia, Milan, Italy). Cerulein is a decapeptide extracted from frog skin, which shares the common C-terminal pentapeptide with cholecystokinin (CCK) and gastrin (16). Therefore, cerulein exhibits biological activities similar to CCK and stimulates gallbladder contraction as well as PP release in humans (17). A dose-response curve of gallbladder emptying was evaluated during intravenous infusion of 0.25, 1, and 4 Ixg/kg/ min of cerulein. Each dose of cerulein was infused for 20 min, and gallbladder emptying was monitored by sonography with images taken every 5 min. Venous blood samples for PP determination were taken at - 5 , 0, 5, 10, 15, and 20 min. Each 20-min infusion period was followed by a 60-min washout period. Each cerulein infusion was begun only when gallbladder volume had returned to basal. Plasma glucose concentrations were monitored during all studies by using Dextrostix reagents combined with a Glucometer reader (Ames Division, Miles LaboraDigestive Diseases and Sciences, Vol. 35, No. 9 (September 1990)

tories, Elkhart, Indiana). To avoid hypoglycemic episodes, a 5% glucose solution was infused intravenously if necessary. PP concentrations in plasma were determined by radioimmunoassay as previously described (18) using the antiserum 146,10 (kindly donated by Dr. R.A. Chance, Eli Lilly and Company, Indianapolis, Indiana) at a final dilution of 1:3,000,000. The detection limit of the assay was 2.75 pg/tube. Sonographic Studies. Sonographic studies were performed using an AUC 940 sonographic machine (Ansaldo, Genoa, Italy) equipped with a 3.5-MHz convex probe. Gallbladder emptying and refilling were determined. Transverse and longitudinal scans at the level of the right hypochondrium were obtained with the patients in the supine and left posterior oblique positions. The gallbladder volume was determined by the ellipsoid model (19): volume = ~r/6 x [L x W x D], where L is the maximum longitudinal diameter, W is the width, and D is the depth of gallbladder. Gallbladder emptying at each time interval was determined by calculating the ejection fraction (EF) using the following formula:

EF = [1 - (RV/FV)] x 100 where RV is residual volume at any prefixed time, and FV is the fasting gallbladder volume. Statistical Analysis. Results are expressed as the means + SE. The effects of various treatments on gallbladder volume were analyzed using the analysis of variance (ANOVA). Plasma PP output in response to stimulants was expressed as integrated PP response (IPPR) and calculated according to Taylor et al (11). Area under curve (AUC) was determined using the trapezoid rule (20). Student's t test was used when appropriate.

RESULTS Fasting gallbladder volume was 18.9 +-- 1.3 ml in diabetic patients and 18.1 --- 1.6 ml normal subjects (P > 0.05) (Figure 1). W h e n diabetic patients were divided according to the p r e s e n c e of A N , a slightly higher volume was found in diabetics with A N (21.1 - 2.1 ml) in c o m p a r i s o n to diabetic patients without A N (17.5 --- 1.3 ml, P > 0.05). After a mixed meal, the gallbladder did not completely e m p t y in any subject, the EFs at 45 min were 60.1 --- 5.6, 64.3 --3.7, and 69.3 +- 7.0% in diabetic patients with or without A N and healthy volunteers, respectively (P < 0.01 in c o m p a r i s o n with basal values). Gallbladder refilling was almost complete 150 min after the meal in all three groups (P > 0.05 in c o m p a r i s o n with basal volume) (Figure 1). N o difference in gallbladder emptying was seen w h e n diabetic patients were divided according to p r e s e n c e of peripheral n e u r o p a t h y (data not shown). To evaluate if gallbladder motility in r e s p o n s e to neural (cephalic) stimulation was reduced in patients with diabetes mellitus, we monitored gall-

1091

FIORUCCI ET AL Meal

~-~ 24-~

MSF

28

I'--I I

16! 1~111~\T

/'~zi

Neuropathy

/

+//

/ / A / /

Normal subjects

8

I 0

I 15

I 30

I 45

I 60

I 75

0

4

I I I I I 90 105 120 135 150

bladder emptying during MSF. Since MSF is known to stimulate gastric acid and PP secretions by long vagal pathways, we measured gastric acid secretion and PP plasma concentrations. Fasting gallbladder volume was 19.7 --- 1.7 ml in diabetic patients (22.0 --+ 3.1 ml and 18.7 -+ 3.3 in diabetic patients with or without AN, respectively, P > 0.05), and 18.5 -+ 2.0 ml in the healthy subjects. In healthy volunteers, gallbladder residual volume fell to I 1.5 --- 1.3 after 30 min (P < 0.01 in comparison with basal values). The EF at 30 min was 38.5 - 5.5%, approximately half that seen after a meal (Figure 2). Considering all diabetic patients together, the R V and EF were 17.8 - 1.5 ml and 13.5 -+ 7%, respectively. Both values were significantly higher than in healthy volunteers (P < 0.01). When diabetic patients were considered according to the presence of AN, a significant reduction in gallbladder emptying induced by MSF was found only in patients with AN (Figure 2). In these patients MSF induced a paradoxical increase in gallbladder volume, which was 25.4 --- 2.6 ml after 60 min with an increase of approximately 16% (15.8 -+ 8.5%). By contrast, MSF markedly reduced gallbladder volume in diabetics without AN. The R V and EF at 30 rain were 12.5 - 1.5 ml and 35.8 -+ 5.5%, respectively (P < 0.01 in comparison with diabetics with AN, by ANOVA). Gallbladder emptying induced by MSF did not correlate with the type of diabetes or with glycemic control, as indicated by HBA1 c values. Among the 21 diabetic patients included in this

1092

-

r

/

T

9

i

9 All diobeHcs

I--I

9

He a l t h y s u b j e c t s

0

Time (min)

Fig 1. Gallbladder emptying and refilling after a meal in diabetic patients with (9 patients) or without (12 patients) autonomic neuropathy and in 10 healthy subjects. Gallbladder volume was determined by ultrasonography and images taken every 15 min before, during, and after meal. In all three groups of subjects gallbladder volume was significantly reduced, in comparison with basal, 15 rain after meal (P < 0.05). Values given are means -+ SE. No differences in gallbladder emptying were found among the three groups of subjects (by ANOVA).

~

1

/

1

,

l

9 --

(.9 0 -15

l

9

12

o

4

~+

:>

No Neuropathy

"0 "0

a (,.9

s

::3

///1

'~?XT ~"'~II

12

-

"

o/I

Jl

t

I

10

20

No neuropathy

9

Neuropothy

11--0

~q-

I

I

I

30 40 ]~me (min)

50

60

Fig 2. Gallbladder emptying and refilling after MSF in all diabetic patients, in diabetics with (9 patients) or without (12 patients) autonomic neuropathy, and in healthy subjects. Gallbladder volume was determined by ultrasonography and images taken every 10 min during and after the MSF. Values given are means -+ SE. *P < 0.01 in comparison with all diabetic patients and diabetic patients with AN (by ANOVA).

study, 11 were affected by peripheral neuropathy. MSF failed to stimulate gallbladder emptying in patients with peripheral neuropathy (Figure 3). When patients with peripheral neuropathy were divided according to presence of AN, we found that MSF failed to stimulate gallbladder emptying only in seven patients affected by both AN and peripheral neuropathy, while significant reduction of gallbladder volume (P < 0.05) was seen in four diabetic patients affected by peripheral neuropathy without AN (Table 2). BAG (Figure 4) in diabetic patients with or with28

v

E o~

E

MSF

24 20

o > n~ ~

8

o 0

0--0

Peripheral

13--

[3 No peripheral

I

I

10

20

I

Neuropothy neuropothy I

I

30 40 50 Time ( m i n )

I

I

60 min

Fig 3. Gallbladder emptying and refilling after modified shamfeeding in diabetics with (i1 patients) or without (10 patients) peripheral neuropathy. Gallbladder volume was determined by ultrasonography and images taken every 10 min during and after the MSF. Values given are means -+ sE. Gallbladder volume was significantly reduced only in diabetic patients without peripheral neuropathy (P < 0.01 in comparison with diabetics with peripheral neuropathy, by ANOVA). Digestive Diseases and Sciences, Vol. 35, No. 9 (September 1990)

GALLBLADDER MOTILITY IN DIABETIC PATIENTS TABLE 2.

ACID SECRETION AND PANCREATIC POLYPEPT1DE RELEASE 1N RESPONSE TO M S F

BAO (mMIhr)

SAO (mM/hr)

PP

IPPR

(pglml)

(nM/liter)

Diabetics without A N 3.4 + 1.3 14.2 --- 2.8 41.8 -+ 6 7.1 - 0.5 Diabetics with A N 2.1 + 1.2 6.3 --- 1.7' 71.0 --- 20.5* 0.8 -+ 0.3* Healthy subjects 3.5 --- 1.2 17.3 - 3.0 35.0 -+ 7.5 8.0 -+ 0.6 *P < 0.01 in comparison with healthy subjects and diabetic patients without AN.

out AN and in healthy volunteers was comparable: 2.1 -+ 0.6 mmol/hr, 3.4 +- 0.4 mmol/hr, and 2.3 - 0.5 mmol/hr, respectively. None of the differences were statistically significant (Table 3). MSFstimulated acid secretion was significantly reduced in diabetic patients with AN in comparison with patients without AN or healthy subjects (P < 0.01). Mean basal concentrations of plasma PP were higher in diabetic patients than healthy subjects (P < 0.05). In healthy volunteers and diabetic patients without AN, MSF induced a rapid and significant increase of plasma PP concentrations, which was maintained until 60 min (Figure 5, Table 3). By contrast, as previously reported (13), MSF failed to increase PP concentrations in patients with AN. As a consequence, the IPPR in these patients was significantly lower than in healthy subjects (P < 0.05) or diabetic patients without AN (P < 0.05). No correlation was found between the AUC of plasma PP concentrations and blood glucose during MSF. Finally, to determine if gallbladder emptying induced by cerulein was reduced in patients with iI

OE

i

4..

.~

3-

~

2-

O

-Q

9~
n-

hi t21 t23

5m _J .< 0

4020-

0

I 0.25

I

I

1

4

CERULEIN CONCENTRATION ( u g / k g / m i n )

Fig 7. Correlation between maximal gallbladder emptying and concentration of cerulein infused in diabetic patients with and without AN and healthy subjects. Data from Figure 5.

1094

4 ug/ki/min

400 l

"- ~2oof 0

~" 100

~

t

12

"o "o mt~ _o

infusion

1 ug/kg/min

5~176 T

9 Neuropothy

16

-6 >

ug/kg/mln

rl,

0~:-- ,-5

5

15

-

I

I

5

15

-5

I

5

I

I

15 rain

Time

Fig 8. Effect of intravenous infusion of cerulein at the doses of 0.25, 1, and 4 ixg/kg/min on blood concentrations of pancreatic polypeptide in diabetic patients with or without AN and in healthy subjects. Cerulein was infused for 20 min, from time 0 to 20. Values given are means _+ SE. In all three experiments PP values were significantly higher than basal (P < 0.01) after 5 min of cerulein infusion.

were found in the pattern of secretion or in the AUC of plasma PP concentrations in the three groups of subjects (Figure 8). Once again the CDs0 of PP release in response to cerulein administration was comparable in all three groups. DISCUSSION Gallbladder emptying in response to a meal is regulated by several different mechanisms such as the physical characteristics of the meal (eg, volume and osmolality) or meal composition (22). Eating food is believed to stimulate gallbladder emptying by three major physiological mechanisms: (1) cephalic-vagal stimulation; (2) gastric and intestinal distension, which stimulates cholinergic reflexes; and (3) chemical reaction of food and digestive products with gastrointestinal mucosa, causing release of hormones such as CCK, gastrin, motilin, and secretin, which modulate gallbladder contraction by activating specific receptors (22). The frequency of gallbladder disorders in patients with diabetes mellitus is still a matter of controversy (1-6). Despite the fact that the incidence of gallstone in diabetic patients is not increased, enlargement of fasting gallbladder volume, as well as reduction of gallbladder emptying induced by meal (23) or by hormonal stimulation (4), has been reported. In the present study we evaluated gallbladder emptying induced either by a meal or by separate cephalic and hormonal stimulation. Our results indicate, in diabetic patients and healthy subjects, Digestive Diseases and Sciences, Vol. 35, No. 9 (September 1990)

GALLBLADDER MOTILITY IN DIABETIC PATIENTS gallbladder emptying induced by a regular meal was comparable, suggesting that the mechanisms that regulate postprandial gallbladder emptying, mainly CCK, are fully operational in diabetic patients with or without AN. A reduction of gallbladder motility has been described before in patients with diabetes mellitus, but the results of these studies are conflicting. A greater fasting gallbladder volume was described earlier in female diabetic patients by Gitelson et al using cholecystography (23). In this study, approximately a 20% reduction of gallbladder emptying induced by a fatty meal was found in patients with peripheral neuropathy and retinopathy. Using the same technique, Grodzki et al found a normal fasting gallbladder volume in 17 diabetic patients (24). In recent years several accurate imaging techniques have been developed to evaluate gallbladder motility. Both ultrasonography and scintigraphy are used primarily (22). Using ultrasonography, Braveman recently reported a reduction of 21% in gallbladder emptying induced by a meal in a group of 12 diabetic patients (25). Unfortunately, he did not group the patients according to the type of diabetes, nor did he examine the role of potential factors such as weight, degree of diabetic control, and presence of AN, which might affect gallbladder emptying. By contrast, Keshavarzian et al documented normal gallbladder emptying in response to a meal in a population of 47 male diabetic patients (3). In this study, gallbladder emptying induced by a meal was comparable in patients with or without AN or peripheral neuropathy. Conflicting results have been obtained by Stone et al using cholescintigraphy (4). In Stone's study, gallbladder emptying in response to CCK was evaluated in 30 diabetic patients, nine of whom had AN. A reduction of 21% of gallbladder emptying in response to infusion of 20 ixg/kg/hr CCK-octapeptide was documented in diabetics patients with AN in comparison with patients without AN. Unfortunately, gallbladder emptying in response to a meal was not evaluated. As seen by Keshavarzian et al, we also found that gallbladder emptying induced by an ordinary meal was comparable in diabetics and healthy subjects. We were, however, unable to confirm any difference in gallbladder emptying induced by hormonal stimulation. The discrepancy with the study of Stone et al can be explained in several ways. First, the techniques used in the two studies were different (scintigraphy versus ultrasonography). Using ultrasonography, gallbladder emptying is deterDigestive Diseases and Sciences, Vol. 35, No. 9 (September 1990)

mined only indirectly, by measuring volume changes. It is generally assumed that reduction in gallbladder volume indicates ejection of the bile in the duodenum. The quantitative relationship, however, between change of gallbladder volumes and bile fluxes is still unknown. With cholescintigraphy, direct measurement of gallbladder volume cannot be obtained while bile fluxes are being measured. Since Stone et al documented only a 20% reduction of bile fluxes (4), it might be argued that such reduction was too small to affect gallbladder volumes measured with ultrasonography. Second, in Stone's study, 16 of the diabetic patients were obese. Obesity is a known predisposing factor for gallstones formation in humans and has been reported to reduce gallbladder emptying. In a recent study by Vezina et al, obese patients and large nonobese males were found to have 40-50% gallbladder reduction in response to a meal, in comparison to 70-80% reduction observed in normal subjects (26). The results of Vezina et al appear to be in conflict with Stone's findings, since the latter reported no difference in gallbladder emptying between normal subjects and nondiabetic obese patients, as well as no difference between obese and nonobese diabetic patients. The reasons for these differences are unclear. Stone's results seem to indicate that gallbladders of patients with diabetes mellitus and AN are less sensitive to hormonal stimulation than gallbladders of diabetics without AN, but a dose-response curve was not evaluated. In our study, in which a dose-response curve to cerulein infusion was determined either in normal subjects or diabetic patients with and without AN, we found comparable CDs0 values. Thus, the sensitivity of gallbladder to hormonal stimulants in our diabetic patients with AN appears to be the same as in diabetics without AN or healthy subjects. We do not have a clear explanation for the differences between our study and that of Stone et al. In the present study we found that MSF induced approximately a 30% reduction of gallbladder volume. This same magnitude of emptying has been reported previously by Fisher et al (9) using cholescintigraphy. The mechanisms that regulate gallbladder emptying in response to MSF are still unclear but seem to involve cholinergic pathways, as gallbladder emptying is prevented by vagotomy and atropine (9). Our major finding is that gallbladder emptying in response to cephalic stimulation is abolished in diabetic patients with AN. This parallels reductions

1095

FIORUCCI ET AL of gastric acid secretion and PP release induced by vagal stimulation. Thus, gastrointestinal functional abnormalities in diabetics with AN are analogous to abnormalities of the gastrointestinal tract observed after vagotomy. It is of interest that after truncal vagotomy, gallbladder emptying in response to a meal appears to be normal. In contrast to diabetics, dilated fasting gallbladder was a uniform finding after truncal vagotomy (27), In the present study we were unable to find any differences in fasting gallbladder volume between diabetic patients and healthy subjects. However, similar to previous reports, the largest gallbladders were seen in diabetic patients with AN. It is difficult to evaluate the clinical consequences of a reduction of gallbladder motility in response to cephalic stimulation, since after a meal hormonal factors play a major role in regulating gallbladder emptying and can compensate for defective neural mechanisms. Recently Baxter et al reported that vagal pathways mediated the early-phase gallbladder contraction that precedes gastric emptying after a solid meal (28). This pattern is abolished by vagotomy. The physiologic role of this early gallbladder emptying is unclear, but alteration of the timing of delivery of bile to the duodenum may play a role in pathogenesis of diarrhea observed after vagotomy. It is still unknown if an alteration of the correlation of gastric and gallbladder emptying is present in diabetic patients with AN. Gallbladder emptying occurs cyclically in the interdigestive period, which correlates with the interdigestive migrating motor complex (IMMC) (22). The mechanisms responsible for IMMC regulation of gallbladder motility are unclear. Neural and hormonal factors (motilin) seem to be involved (22). Since the absence of IMMC is one of the notable abnormalities described in patients with diabetes mellitus, it cannot be excluded that alteration of gallbladder motility may take place during the interdigestive period in diabetics with visceral autonomic neuropathy (29). Gastric acid secretion in response to MSF was markedly reduced in diabetics with AN. Buysschaert et al found that gastric acid secretion induced by MSF was reduced in diabetic patients with AN while acid secretion induced by pentagastrin infusion was comparable to the secretion observed in healthy subjects or patients without AN (13). Similarly, a reduction of gastric acid secretion induced by MSF was reported in patients with long-standing

1096

diabetes, while gastric acid output induced by intragastric infusion of meal was normal (30). PP levels in patients with diabetes mellitus were higher than in healthy subjects. This finding may be explained by the higher levels of blood glucose in diabetic patients, particularly in those with AN. As previously reported, patients with poor glucose control show higher PP basal levels, since there is a direct correlation between blood glucose and PP concentrations (31). PP secretion in response to MSF also was reduced in diabetics with AN, confirming previous observations (13). PP secretion is largely dependent on vagal innervation (12). Vagotomy and atropine reduce PP secretion in response to MSF as well as the early peak of PP observed after meal (11, 12, 32). In a previous study in dogs, Guzman et al found that vagotomy reduces PP secretion in response to stimulation with CCK octapeptide, indicating that vagal innervation modulates the sensitivity of pancreatic PP cells to hormonal stimulants (21). Responsiveness of pancreatic PP cells to CCK or CCK analogs has never been evaluated before in patients with diabetes mellitus. CCK and cerulein are both powerful stimulants of PP secretions in humans, as well as in dogs (21, 33). We found that PP secretion in response to cerulein was comparable in diabetic patients with and without AN and healthy subjects. The CDSs0 observed in the three groups of subjects were comparable, indicating that in diabetic patients pancreatic PP cells were fully responsive to hormonal stimulation. This is confirmed by the fact that after meal, only the early (vagally mediated) peak of PP is reduced in diabetic patients with AN, while the sustained (hormone-mediated) secretion is maintained (32). ACKNOWLEDGMENTS

The authors thank Byron Cryer, MD, and Rossana De Palma, MD, for reading the manuscript and helpful suggestions. REFERENCES 1. Lieber MM: The incidence of gallstone and their correlation with other diseases. Ann Surg 135:394-404, 1952 2. Stone BG, Van Thiel DH: Diabetes mellitus and liver. Sem Liver Dis 5:8-28, 1985 3. Keshavarzian A, Dunne M, Iber FL: Gallbladder volume and emptying in insulin-requiring male diabetics. Dig Dis Sci 32:824-828, 1987 4. Stone BG, Gavaler JS, Belle SH, Shreiner DP, Peleman RR, Sarva RP, Yingvorapant N, Van Thiel DH: Impairment of Digestive Diseases and Sciences, Vol. 35, No. 9 (September 1990)

G A L L B L A D D E R M O T I L I T Y IN D I A B E T I C P A T I E N T S gallbladder emptying in diabetes mellitus. Gastroenterology 95:170-176, 1988 5. Friedman GD, Kannel WB, Dawber KR: The epidemiology of gallbladder disease: Observations in the Framingham study. J Chron Dis 19:273-292, 1966 6. Jaspan JB: The neuropathies of diabetes. In Endocrinology. LJ De Groote (ed). Philadelphia, WB Saunders, 1989, pp 1475-1511 7. Ewing DJ, Clarke BF: Autonomic neuropathy: its diagnosis and prognosis. Clin Endocrinol Metab 15:855-888, 1986 8. Feldman M, Schiller LR: Disorders of gastrointestinal motility associated with diabetes mellitus. Ann Intern Med 98:378-384, 1983 9. Fisher RS, Rock E, Malmud LS: Gallbladder emptying respons e to sham-feeding in humans. Gastroenterology 90:1854-1857, 1986 10. Feldman M, Richardson CT, Fordtran J: Experience with sham-feeding as a test for vagotomy. Gastroenterology 79:792-795, 1980 11. Taylor JL, Feldman M, Richardson CT, Walsh JH: Gastric and cephalic stimulation of human pancreatic polypeptide release. Gastroenterology 75:432-437, 1978 12. Schwartz TW: Pancreatic polypeptide. A hormone under vagal control. Gastroenterology 85:1411-1425, 1983 13. Buysschaert M, Donckier J, Dive A, Keteslegers JM, Lambert AE: Gastric acid and pancreatic polypeptide responses to sham feeding are impaired in diabetic subjects with autonomic neuropathy. Diabetes 34:1181-1185, 1985 14. National Diabetes Data Group: Classification and diagnosis of diabetes me!litus and other categories of glucose intolerance. Diabetes 28:1039-1052, 1982 15. Fiorucci S, Bosso R, FarineUi M, Cascetta R, Morelli A: Cephalic stimulation of gallbladder motility. Effect of naloxone, an opioid receptor antagonist. Ital J Gastroenterol 20:62-65, 1988 16. Anastasi A, Ersparmer V, Endean R: Isolation and aminoacid sequence of cerulein, an active decapeptide of the skin of Hyla caerulea. Arch Biochem 125:57-68, 1968 17. Adrian TE, Bestermen HS, Cooke TJC, Bloom SR, Barnes AJ, Russel RCG. Mechanism of pancreatic polypeptide release in man. Lancet 1:161-163, 1977 18. Kohn A, Annibale B, Suriano G, Severi C, Spinella S, Delle Fave G: Gastric acid and pancreatic polypeptide responses to modified sham feeding: indication of an increased basal vagal tone in a subgroup of duodenal ulcer patients. Gut 26:776-782, 1985 19. Dodds WJ, Groth WJ, Darweesh RMA, Lawson TL, Kisher SMA, Kern MK: Sonographyc measurement of gallbladder volume. Am J Radiol 145:1009-1011, 1985

Digestive Diseases and Sciences, Vol. 35, No. 9 (September 1990)

20. Bailer JA: Testing for equality of area under the curves using destructive measurement techniques. J Pharmacokinet Biopharm 6:303-310, 1988 21. Guzman S, Lonovics J, Devitt P, Hejtmancik KE, Rayford PL, Thompson JC: Hormone-stimulated release of pancreatic polypeptide before and after vagotomy in dogs. Am J Physiol 240:G114--G121, 1981 22. Ryan JP: Motility of the gallbladder and biliary tree. In Physiology of Gastrointestinal Tract. LR Johnson (ed). New York, Raven Press, 1987, pp 695-721 23. Gitelson S, Oppenheim D, Schwartz A: Size of the gallbladder in patients with diabetes mellitus. Diabetes 18:493-498, 1969 24. Grodzki M, Mazurkiewicz-Rozynska E, Czyzyk A: Diabetic cholecystopathy. Diabetologia 4:345-348, 1968 25. Braveman DZ: The lack of effect of metoclopramide on gallbladder volume and contraction in diabetic cholecystoparesis. Am J Gastroenterol 81:960-961, 1986 26. Vezina WC, Paradis RL, Grace MD, Zimmer RA, Lamont DD, Rycroft KM, King ME, Hutton LC, Chey WY: Increased volume and decreased emptying of the gallbladder in large (morbidly obese, tall normal, and muscolar normal) people. Gastroenterology 98:1000-1007, 1990 27. Shaffer EA: The effect of vagotomy on gallbladder function and bile composition in man. Ann Surg 195:413-428, 1982 28. Baxter JN, Grime JS, Critchley M, Jenkins SA, Shields R: Relationship between gastric emptying of a solid meal and emptying of the gallbladder before and after vagotomy. Gut 28:855-863, 1987 29. Ackem-Karam SR, Funakoshi A, Vinik AI, Owyang C: Plasma motilin concentration and interdigestive migrating motor complex in diabetic gastroparesis: Effect of metoclopramide. Gastroenterology 88:492-499, 1985 30. Feldman M, Corbett DB, Ramsey EJ, Walsh JH, Richardson CT: Abnormal gastric function in longstanding, insulindependent diabetic patients. Gastroenterology 77:12-17, 1979 31. Marco J, Hedo JA, Villanueva ML: Control of pancreatic polypeptide secretion by glucose in man. J Clin Endocrinol Metab 46:140-145, 1978 32. Gambardella S, Felici MG, Annibale B, Delle Fave G, Jacoangeli F, Spallone V, Menzinger G: Pancreatic polypeptide response to a protein-rich meal in diabetic patients with and without neuropathy. J Endocrinol Invest 9:1-4, 1986 33. Meyer FD, Gyr K, Hacki WH, Beglinger C, Jeker L, Varga L, Kayasseh L, Gillessen D, Stadler GA: The release of pancreatic polypeptide by CCK-octapeptide and some analogues in the dog. Gastroenterology 80:742-747, 1981

1097