Rapid intestinal transit as a primary cause of severe chronic diarrhea in patients with amyloidosis

THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2003 by Am. Coll. of Gastroenterology Published by Elsevier Inc.

Vol. 98, No. 10, 2003 ISSN 0002-9270/03/$30.00 doi:10.1016/S0002-9270(03)00701-9

Rapid Intestinal Transit As a Primary Cause of Severe Chronic Diarrhea in Patients With Amyloidosis Michael J. Guirl, M.D., Christoph Ho¨genauer, M.D., Carol A. Santa Ana, B.S., Jack L. Porter, M.S., Katherine H. Little, M.D., Marvin J. Stone, M.D., and John S. Fordtran, M.D. Baylor University Medical Center, Dallas, Texas

OBJECTIVE: The cause of severe diarrhea in patients with systemic amyloidosis is obscure. We therefore performed pathophysiological studies in three such patients in an effort to determine the mechanism of amyloid diarrhea. METHODS: Epithelial cell absorption rate of electrolytes was measured during steady state GI perfusion of a salinemannitol solution. GI transit time of PEG and absorption of radiolabeled bile acid were measured simultaneously while subjects ingested three meals per day. To obtain a diarrhea control group for transit time and bile acid absorption, normal subjects were studied when they had diarrhea caused by ingestion of Milk of Magnesia (MOM). RESULTS: Diarrhea could not be explained by malabsorption of ingested nutrients, bacterial overgrowth, bile acid malabsorption, or epithelial cell malabsorption of electrolytes. However, 25% of polyethylene glycol (PEG) ingested with a standard meal was recovered in stool in 45 min, which is 10 times faster than in normal subjects with equally severe diarrhea caused by ingestion of MOM. All of the patients had autonomic neuropathy that remained unrecognized for 15–36 months after onset of chronic diarrhea; it seems likely that this was the cause of rapid transit. CONCLUSIONS: Severe chronic diarrhea in three patients with systemic amyloidosis was mediated by extremely rapid transit of chyme and digestive secretions through the intestine. (Am J Gastroenterol 2003;98:2219 –2225. © 2003 by Am. Coll. of Gastroenterology)

INTRODUCTION Severe chronic diarrhea can be the dominating clinical manifestation of systemic amyloidosis associated with autonomic neuropathy (1), and we recently studied three such patients. Previously suggested mechanisms for diarrhea in patients with amyloidosis have included extensive deposits of amyloid in intestinal mucosa, steatorrhea from pancreatic disease, bile acid malabsorption, and intestinal pseudoobstruction. None of these could explain the diarrhea in our patients. Incoordination of intestinal movements and incoordination of intestinal secretions, secondary to autonomic

neuropathy, have also been suggested as a cause of amyloid diarrhea (1), although no evidence has been produced to evaluate these two possibilities. Because the cause of diarrhea in our patients with amyloidosis was obscure, we performed pathophysiological studies in an effort to determine its mechanism. Diarrhea in our patients was associated with high concentrations of sodium and potassium in fecal fluid, indicating that electrolyte malabsorption was the cause of excess fecal water output. Electrolyte malabsorption is usually caused by abnormalities in secretion or absorption of electrolytes by intestinal epithelial cells, i.e., secretory diarrhea (2, 3). However, it is theoretically possible that electrolyte malabsorption could be caused by rapid intestinal transit (3, 4). To evaluate ion transport by epithelial cells, absorption/ secretion of electrolytes was measured during steady state GI perfusion (3). To evaluate rapid transit as a primary cause of electrolyte malabsorption, we measured the timed fecal recovery of orally ingested PEG, a nonabsorbable and water-soluble solute. Interpretation of transit times in any group of patients is problematic because diarrhea itself, regardless of etiology, would be expected to cause a secondary acceleration of intestinal transit as compared with transit in normal persons without diarrhea. This is because distention of the intestinal lumen with fluid promotes peristaltic contractions (3, 4). Therefore, to prove that rapid intestinal transit is a primary mechanism of diarrhea, it would be necessary to show that intestinal transit was more rapid than in normal subjects with equally severe diarrhea. To obtain normal values for transit time in persons with diarrhea, we studied normal volunteers in whom osmotic diarrhea was induced by ingestion of Milk of Magnesia.

MATERIALS AND METHODS Clinical Presentation and Diagnosis The three patients studied were referred for evaluation of chronic diarrhea. Despite the extensive diagnostic studies listed in Table 1, the cause of diarrhea had not been discovered. The only abnormality in the tests listed in Table 1

2220

Guirl et al.

Table 1. Diagnostic Evaluation Stool ova and parasites, occult blood, leukocytes Stool cultures, tests for Clostridium difficile toxin Colonoscopy with biopsies of colonic and ileal mucosa Upper GI endoscopy, small bowel biopsy Upper GI and small bowel roentgenography Abdominal and pelvic sonography or computed tomography Serum thyroid function tests, gastrin Serum calcitonin, vasoactive intestinal polypeptide Urinary excretion of 5-hydroxy indoleacetic acid Stool and urine tests for laxatives

was rapid transit of barium during x-ray studies of the small bowel (Fig. 1) (5); this was noted in the radiological reports, but no pathophysiological or diagnostic significance was recognized. A multitude of therapeutic trials (including bile acid binding agents, tincture of opium and opiate derivatives, antibiotics, octreotide, and dietary changes) did not reduce the severity of diarrhea.

Figure 1. Abdominal x-ray taken 10 min after ingestion of barium sulfate by patient 1. Barium had already reached the transverse colon. Patients 2 and 3 had small bowel transit times of 20 and 30 min, respectively. Transit times in normal young adults aged 18 –25 yr averaged 2.96 h (range 0.65–9 h). The range and lower limits of transit times in older normal individuals have apparently not been defined.

AJG – Vol. 98, No. 10, 2003

At the time of referral, patient 1 had had diarrhea for 24 months, patient 2 for 36 months, and patient 3 for 15 months. The onset of their diarrhea was gradual and intermittent over a period of several months, after which diarrhea became continuous. Patient 2 had frequent episodes of fecal incontinence, possibly because of sphincter damage from previous episiotomies; patients 1 and 3 did not have fecal incontinence. The three patients had lost 29, 18, and 23 kg of body weight, respectively, since the onset of diarrhea. A significant fraction of weight loss was probably caused by reduced caloric intake, as each patient expressed a fear of eating because of increased diarrhea. There was no abdominal pain. History suggested autonomic neuropathy in patient 1 (decreased salivation and erectile dysfunction) and patient 3 (erectile dysfunction). Postural hypotension was present in patients 1 and 3. This was previously attributed to volume depletion from diarrhea; however, there was no increase in pulse rate upon standing, suggesting that postural hypotension was due to autonomic neuropathy rather than to volume depletion. All three patients had cardiovascular autonomic neuropathy by pulse rate variability testing, measured with electrocardiography (6). Because of the combination of diarrhea and autonomic neuropathy (1), and in the absence of diabetes, amyloidosis was suspected. Intestinal biopsy samples were stained with Congo red, which revealed sparse deposits of amyloid in duodenal, ileal, and colonic submucosa, adjacent to blood vessels. Further studies revealed that patient 1 had amyloid deposits of light chain origin. Patients 2 and 3 had amyloidosis caused by an abnormal transthyretin, with alanine substituted for threonine at position 60 (7, 8). Although this latter form of amyloidosis is usually inherited in an autosomal dominant fashion, neither of our patients had a family history of this illness and these patients likely represent sporadic cases. Other features of amyloidosis in each patient are presented in Table 2. While the patients ingested a diet containing approximately 100 g/day of fat, their stools were liquid in consistency, and stool weights were ⬎1 L/day. Diarrheal fluid contained high concentrations of sodium and potassium, with osmotic gap values of ⬍50 mEq/L (9), indicating either secretory diarrhea or diarrhea caused by rapid intestinal transit (3). Stool fat outputs were 12, 26, and 17 g/day, respectively, in patients 1, 2 and 3 (normal ⬍7 g/day). Fecal chymotrypsin activity (10) was normal in each patient, indicating normal exocrine pancreatic enzyme secretion. Patient Protection and Privacy Patients and normal subjects were studied by a protocol that was approved by the Institutional Review Board for Human Protection of Baylor University Medical Center. Informed consent was obtained from each participant. Electrolyte and Water Absorption During Steady State Perfusion Fasting subjects swallowed a single-lumen, mercuryweighted polyvinyl tube. A saline-mannitol solution was

AJG – October, 2003

Rapid Intestinal Transit and Amyloidosis

2221

Table 2. Characteristics of Patients With Amyloidosis*

Patient No.

Age, Sex

Amyloid Type

1

53, M

AL

2

69, F

ATTR

3

55, M

ATTR

Initial Urinalysis Protein

SPE (g/dl)

Serum IFE

M-component 0.8, albumin 2.5 No Mcomponent, normal albumin No Mcomponent, normal albumin

Monoclonal IgG, ␭

Trace

Normal

Normal

24-h Urine Protein (mg/day)

Urine IFE

Bone Marrow Bx % Plasmacytosis‡ (Congo Red Stain)

TTE Suggestive of Amyloidosis

Albumin and ␭ BJP Normal

30% (negative)

No

Negative

1698, 92% albumin 134

2% (positive)†

Yes

Negative

378

Increased albumin

4% (negative)

Yes

AL ⫽ amyloidosis of light chain origin; ATTR ⫽ amyloidosis of transthyretin origin; BJP ⫽ Bence Jones protein; SPE ⫽ serum protein electrophoresis; IFE ⫽ immunofixation electrophoresis; TTE ⫽ transthoracic echocardiogram. * In all patients cardiovascular autonomic neuropathy was present by pulse rate variability testing (see Ref. 6), and skeletal survey was negative for lytic lesions. † Periosteum and blood vessels positive for amyloidosis. ‡ Normal % plasma cells is ⬍5%.

infused continuously into the stomach at a rate of 30 ml/min (11, 12). The solute composition of the test solution was: Na⫹ 110, Cl⫺ 74, K⫹ 4, HCO3⫺ 40, and mannitol 60 (all values are in mmol/L). The test solution also contained 2 g/L of PEG as a nonabsorbable marker. After steady state conditions were obtained, perfusate effluent was collected through a rectal tube during six consecutive 30-min periods. The effluent was analyzed by previously described methods, and absorption rates of water and electrolytes were calculated using standard nonabsorbable marker equations. Patient 1 was also studied by means of steady state segmental perfusion of the jejunum and ileum using a triple-lumen tube (13). Diarrhea in Normal Subjects Induced by Milk of Magnesia Normal subjects ingested different doses of Milk of Magnesia (MOM) (concentrate, Roxane Laboratories, Inc., Columbus, OH) by mouth with meals and a bedtime snack, for 3 consecutive days. The total daily doses of magnesium hydroxide were 60, 90, 123, or 239 mmol. The first day of MOM ingestion was used for equilibration and stools were not collected. GI PEG Transit Time and Radiolabeled Bile Acid Absorption Normal subjects and patients ingested a standard breakfast at 8 AM, consisting of eggs, hash-browned potatoes, orange juice, and water containing 2 g of PEG and 2.5 ␮Ci of [14C] taurocholic acid. In normal subjects with MOM-induced diarrhea, this was done on day 2 after 24 h of equilibration to a diarrheal state. During the 48-h period after ingestion of PEG and radiolabeled bile acid, each stool was collected in a separate container. The amount of PEG in each stool specimen was determined by the method of Hyden (14) and 14 C excreted in 24 h was determined after combustion (15).

GI PEG transit time is expressed as the time required for ⱖ25% (t1/4), ⱖ50% (t1/2), and ⱖ75% (t3/4) of ingested PEG to be excreted in stool. Bile acid excretion is expressed as the time required for 50% (t1/2) of ingested radiolabeled taurocholic acid to be excreted in stool (16). Statistical Analysis Results are expressed as the mean ⫾ SE. Differences were assessed by t test using Sigma Stat 2.03 (SPSS, Chicago, IL). In two instances we also used a one-tailed MannWhitney U test. A p value of ⬍ 0.05 was considered to be statistically significant.

RESULTS Steady State Intestinal Perfusion As shown in Table 3, patients with amyloid diarrhea absorbed water and electrolytes at a normal rate during steady state GI perfusion with a saline-mannitol solution. Jejunal and ileal absorption of electrolytes, water, and glucose was normal in patient 1 (data not shown). Effect of MOM-Induced Diarrhea on PEG Transit Times and Bile Acid Absorption in Normal Subjects PEG t ⁄ was approximately 48 h in two normal subjects who did not ingest MOM (Fig. 2). Ingestion of increasing amounts of MOM caused a progressive increase in stool weight, stool water content, and bowel movement frequency. In association with increasingly severe diarrhea, as measured by each of these changes in stool, there was a progressive reduction in PEG t1/2. Qualitatively similar results were obtained when transit time was based on recovery of one quarter of ingested PEG. Increasingly severe MOM-induced diarrhea also caused a progressive reduction in bile acid t1/2, indicating a progressive reduction in radiolabeled bile acid absorption. Diarrhea 12

2222

Guirl et al.

AJG – Vol. 98, No. 10, 2003

Table 3. Absorption of Water and Electrolytes During Steady-State GI Perfusion Absorption

Amyloidosis Patient 1 Patient 2 Patient 3 Mean ⫾ SE Normal subjects (n ⫽ 6) Mean ⫾ SE Range

Water ml/h

Na mEq/h

Cl mEq/h

HCO3 mEq/h

458 702 393 518 ⫾ 94

61 100 58 73 ⫾ 14

31 70 23 41 ⫾ 14

36 41 42 40 ⫾ 2

508 ⫾ 31 391–584

69 ⫾ 4 53–82

29 ⫾ 6 15–47

40 ⫾ 1 36–43

induced reduction in bile acid absorption is mediated by two factors. First, transit of ingested bile acid through the intestine is accelerated, resulting in decreased contact time between ingested bile acid and absorbing mucosal cells. Second, increased intraluminal volume dilutes the concentration of bile acid in fluid exposed to absorbing mucosal cells, and the lower bile acid concentration reduces the rate of bile acid absorption.

values. Table 4 also shows that transit times were shorter in patients with amyloid diarrhea than in four patients with severe diarrhea caused by other diseases. Bile acid malabsorption was more severe in amyloid diarrhea than in the control group. In patients with amyloid diarrhea, the severity of bile acid malabsorption correlated with the degree to which transit time was reduced.

GI PEG Transit Times and Bile Acid Absorption in Patients With Amyloid Diarrhea As shown in Table 4, average PEG transit times were shorter in amyloid diarrhea than in normal subjects with comparably severe diarrhea caused by ingestion of MOM. The average t1/4 value was most strikingly reduced, to one tenth of control values. The average t1/2 and t3/4 values in amyloid diarrhea were approximately one third of control

DISCUSSION

Figure 2. Effect of Milk of Magnesia (MOM)–induced diarrhea on PEG transit times and on 14C labeled bile acid absorption in normal subjects. Open circles represent normal subjects who did not ingest MOM. Closed circles represent normal subjects who ingested MOM with each of three meals and with a bedtime snack. Data represent the time required for fecal recovery of one half of the ingested dose of PEG and 14C labeled bile acid. PEG t1/2 correlated with stool weight (A), p ⬍ 0.001; percent stool water (B), p ⬍ 0.001; and bowel movement frequency (C), p ⬍ 0.001. Bile acid t1/2 correlated with stool weight (D), p ⫽ 0.004; qualitatively similar results were obtained when bile acid t1/2 was correlated with percent stool water and bowel movement frequency.

Previously suggested mechanisms for diarrhea in patients with amyloidosis have included amyloid deposition in intestinal mucosa, steatorrhea caused by pancreatic insufficiency, bile acid malabsorption, and intestinal pseudoobstruction resulting from autonomic neuropathy with bacterial overgrowth syndrome. It is unlikely that amyloid deposits in the mucosa could have interfered with absorption in our patients because the endoscopic and microscopic appearance of duodenal, ileal, and colonic mucosa was normal. No digestive or mucosal disease was present to explain the mild steatorrhea in our patients, and steatorrhea cannot be the cause of their diarrhea because even severe fat malabsorption causes only mild diarrhea (17). Although the patients had bile acid malabsorption, the fact that their diarrhea was not improved by bile acid binding agents argues against bile acid malabsorption as a primary cause of diarrhea. In addition, bile acid malabsorption causes only low-volume diarrhea, not the high-volume diarrhea exhibited by our patients (18). Intestinal pseudoobstruction, stasis and bacterial overgrowth were ruled out by small bowel x-ray and by lack of improvement with antibiotics directed against intestinal bacterial flora. Because all three patients had evidence of autonomic neuropathy, their diarrhea might have been caused by a reduction in epithelial cell absorption of electrolytes and water, which has been postulated to develop in response to sympathetic denervation of intestinal mucosa (19, 20). To assess epithelial function, we measured electrolyte absorption during high flow, steady state GI perfusion, in which all epithelial cells are continuously exposed to the perfusate (21). This technique readily detects the presence of reduced electrolyte absorption (12). However, the amyloid diarrhea patients absorbed water and electrolytes normally during

AJG – October, 2003

Rapid Intestinal Transit and Amyloidosis

2223

Table 4. GI PEG Transit Time in Patients With Amyloid Diarrhea, Compared to Results in Normal Persons With Milk of Magnesia– Induced Diarrhea PEG Transit Time (h)

Amyloidosis Patient 1 Patient 2 Patient 3 Mean ⫾ SE Milk of Magnesia–induced diarrhea 1 2 3 4 Mean ⫾ SE p value by t test Other causes Surreptitious bisacodyl ingestion Polyglandular syndrome 1 Ileal resection Chloridorrhea

Stool Frequency (BM/day)

% Stool Water*

14

t1/4

t1/2

t3/4

Stool Weight (g/day)

0.33 1.33 0.58 0.75 ⫾ 0.3

1.5 4.5 4.1 3.4 ⫾ 0.9

2.0 7.5 7.2 5.6 ⫾ 1.8

1242 1223 1421 1295 ⫾ 63

6 9 5 7⫾1

90.7 89.1 93.0 90.9 ⫾ 1.1

3.6 9.0 7.0 6.5 ⫾ 1.6

7.7 5.3 9.9 6.1 7.3 ⫾ 1.0 0.003

7.8 11.8 14.3 6.1 10.0 ⫾ 1.9 0.037

10.1 11.8 23.1 7.8 13.2 ⫾ 3.4 0.135‡

1354 1355 1299 1540 1387 ⫾ 53 0.313

4 8 8 8 7⫾1 0.838

93.3 92.8 95.6 93.0 93.7 ⫾ 0.7 0.075

25 23 100 122 68 ⫾ 26 0.057‡

4.5

5.2

7.1

1008

9

94.7

24

11.6 21.9 23.2

11.6 23.8 23.2

26.3

1205 1286 1789

4 8 3

75.7 79.2 93.5

47 16 599

C Bile Acid t1/2 (h)

* Percent stool water was determined by weighing, before and after lyophilization. ‡ Even though there was no overlap between values in the two groups, differences were not statistically significant by t test. However, p values by a one-tailed Mann-Whitney U test were ⬍0.05. The justification for a one-tailed test is that we had a specific hypothesis of short transit time and decreased bile acid absorption.

steady state perfusion, indicating normal epithelial absorption of electrolytes and water. As a prerequisite for evaluating transit time as a possible cause of diarrhea, it was necessary to determine transit times in normal individuals with diarrhea (see Introduction). To do this, osmotic diarrhea was induced in normal volunteers by administration of Milk of Magnesia (MOM) in various dosages. The results revealed that even mild diarrhea in normal subjects was associated with a substantial reduction in PEG transit times, and when diarrhea was severe, there was a marked reduction in transit times. The fall in transit times with increasingly severe diarrhea was well correlated with increasing stool weight, decreasing stool consistency (as measured by percent stool water), and bowel movement frequency. Each of these cardinal features of diarrhea could have contributed to the reduction in PEG transit times. Therefore, when evaluating transit time as a cause of diarrhea in patients, the results in patients must be compared with those in a control group of otherwise normal subjects with comparable severity of diarrhea as measured by stool weight, stool consistency, and frequency of bowel movements. We found that patients with diarrhea due to amyloidosis had much shorter PEG transit times than those of normal individuals with equally severe diarrhea. To illustrate the magnitude of this abnormality, the time required for fecal recovery of 25% of ingested PEG in patients with amyloid diarrhea was only one tenth that in normal persons, and 75% of PEG was recovered in amyloid diarrhea patients before 25% of PEG was recovered in normal persons with induced diarrhea. Patients with amyloid diarrhea also had shorter

transit times than our patients with severe diarrhea caused by diseases other than amyloidosis. Although GI PEG transit time does not define transit time in specific regions of the intestine, the extremely low values for PEG t1/4 (0.75 h on average) mean that transit time must be very short in all segments of the GI tract. Given that all regions of the intestine normally absorb electrolytes and water, a short transit time through any segment of the intestine could contribute to fluid malabsorption. That transit time is abnormally reduced in our three patients with amyloid diarrhea is undeniable. However, it would be reasonable to question whether reduced transit time could be a primary cause of severe diarrhea. When normal individuals eat three meals per day, 9 L of electrolyte rich fluids enter the upper small intestine per day, derived from oral intake and from digestive secretions. Of this fluid, 7 L is absorbed in the small intestine, and all but 100 ml of the remaining 2 L is absorbed by the colon (17, 22). Fluids traverse various parts of the intestine in boluses, which make transient contact with sequential segments of intestinal epithelial cells. The volume of fluid absorbed depends on the absorption rate by exposed epithelial cells, multiplied by the contact time (which is reflected by transit time) of fluid boluses with those epithelial cells (11, 22). Normally, GI transit time is about 48 h. In our opinion, it is reasonable to suggest that extreme shortening of transit time, as seen in our three patients with amyloid diarrhea, could result in malabsorption of enough fluid to cause severe diarrhea. Malabsorption of only 14% of the 9 L of fluid entering the upper small bowel per day would result in 1260

2224

Guirl et al.

AJG – Vol. 98, No. 10, 2003

Table 5. Factors Contributing to Diagnostic Delay in Our Three Patients With Amyloid Diarrhea Failure of physicians to recognize autonomic neuropathy, which probably would have suggested the correct diagnosis ● Characteristic symptoms of autonomic neuropathy, other than diarrhea, were subtle and not volunteered by patients or sought by physicians ● Attributed postural hypotension to volume depletion from diarrhea, when it was actually due to autonomic neuropathy ● Lack of physical examination to detect autonomic neuropathy ● Failure to recognize that very rapid intestinal transit of barium may indicate autonomic neuropathy. Interpreting normal urine protein excretion as evidence against amyloidosis Amyloid deposits in intestinal biopsy samples are sparse and may be overlooked even after staining by Congo red. Confusing the clinical picture of amyloid diarrhea, which begins gradually, with the clinical picture of self-limited chronic idiopathic diarrhea (possibly a sporadic variant of Brainerd diarrhea), which begins acutely and generally is less severe (see Ref. 34, 35). Although there was some overlap between the two groups, weight loss ⬎17 kg, stool weight ⬎1000 g/day, and stool fat ⬎20 g/day should suggest that spontaneous resolution of idiopathic diarrhea is unlikely and that a serious disease may have been overlooked. Low index of suspicion because literature does not emphasize diarrhea as a GI disorder that should raise suspicion of amyloidosis (see Ref. 33).

g/day of diarrheal fluid, approximately the same as the stool weights in our three patients. We therefore conclude that diarrhea in our patients with amyloidosis was caused by abnormally rapid transit of chyme and digestive secretions through the intestine. Almost all of the osmolality of the diarrheal fluid was made up by sodium and potassium salts, and this is one of the main clinical features of secretory diarrhea. However, by definition, secretory diarrhea is caused by a primary abnormality in intestinal epithelial cell absorption or secretion of electrolytes (3, 17). Inasmuch as epithelial absorption rates of electrolytes were normal, these patients do not have secretory diarrhea; rather, they have severe electrolyte and water malabsorption due to a primary abnormality in intestinal motility, mediated by rapid intestinal transit. The patients also had malabsorption of radiolabeled bile acid, above and beyond malabsorption that can be attributed to diarrhea itself. Because bile acids are absorbed almost exclusively in the ileum, bile acid malabsorption implies ileal dysfunction of some type. There is no reason to suspect a primary abnormality of bile acid transport by ileal epithelial cells, because ileal biopsy results were normal and because ileal absorption of electrolytes and glucose was normal in the one patient who could be studied by means of steady state ileal perfusion. We therefore believe that bile acid malabsorption in patients with amyloid diarrhea was also caused by rapid transit through the ileum. This could be mediated by previously described propagated contractions that propel fluid through the ileum and across the ileocecal sphincter at enormous speeds (23). Our patients also had greater malabsorption of fat and protein than in normal subjects with equally severe diarrhea caused by MOM (data not shown). As there were no primary absorptive or digestive defects, we believe that fat and protein malabsorption was also secondary to rapid intestinal transit. Given that fat and protein are normally absorbed mainly in the jejunum, these results imply rapid transit in the proximal small intestine as well as in the ileum. Intestinal autonomic neuropathy is, in our opinion, the most likely cause of rapid intestinal transit in our patients. First, previous autopsy studies have revealed amyloid de-

posits in the nerves, ganglia, and enteric plexuses that provide autonomic innervation of the gut (1, 24, 25). Amyloid deposits in intraneural blood vessels may also cause ischemic autonomic neuropathy of the intestine (26). Second, all three of our patients had evidence of cardiovascular autonomic neuropathy, increasing the likelihood of autonomic neuropathy of the intestine. Third, there is precedence for suggesting that abnormalities in the autonomic nervous system might cause rapid transit. According to traditional concepts, the sympathetic nerves provide inhibitory impulses to the gut and thereby hold intestinal motility in check (27–29). Destruction of the sympathetic nervous system might therefore leave the intestine unprotected by the normal “inhibitory grip” of sympathetic/splanchnic “brakes” and might thereby provoke hyperperistalsis and rapid transit (27, 29). This is the proposed explanation for the severe diarrhea that may occur after surgical sympathectomy in animals or patients. It is also possible that autonomic neuropathy causes rapid transit by reducing or eliminating fed motor activity after ingestion of food (30, 31). Although the evidence presented here indicates that diarrhea in patients with amyloidosis is caused primarily by reduced transit time, their diarrhea was not ameliorated by antidiarrheal drugs (such as loperamide or deodorized tincture of opium), the main effect of which is mediated by prolongation of intestinal transit time. Destruction of the nerves that normally contain receptors for these drugs (32) may explain this paradox. The diagnosis of amyloidosis is often difficult, and the correct diagnosis is delayed on average by 1–2 yr after the onset of first symptoms (33). The main factors that contributed to diagnostic delay in our three patients are outlined in Table 5.

ACKNOWLEDGMENTS Alan Solomon, M.D., graciously identified the molecular abnormality in transthyretin in patients 2 and 3. Richard Meyer, M.D., interpreted the intestinal biopsies, and Michael Smerud, M.D., gave helpful advice on intestinal transit as measured by X-ray after ingestion of barium. Michael

AJG – October, 2003

Emmett, M.D., and Larry Schiller, M.D., provided much appreciated advice and suggestions about the manuscript. Carol Lee prepared the manuscript. This work was supported by Unites States Public Health grant 5-R01DK37172 from the National Institute of Diabetes and Digestive and Kidney Diseases and by the Southwest Digestive Disease Foundation.

Rapid Intestinal Transit and Amyloidosis

15. 16.

17. Reprint requests and correspondence: John S. Fordtran, M.D., Baylor University Medical Center, 3500 Gaston Avenue, Dallas, TX 75246. Received Jan. 29, 2003; accepted May 12, 2003.

REFERENCES 1. French JM, Hall G, Parish DJ, Smith WT. Peripheral and autonomic nerve involvement in primary amyloidosis associated with uncontrollable diarrhoea and steatorrhoea. Am J Med 1965;39:277–84. 2. Binder HJ. The pendulum of medical progress and intestinal fluid secretion: 1890 –1997. Gastroenterology 1997;112: 1748 –9. 3. Fordtran JS, Santa Ana CA, Morawski SG, et al. Pathophysiology of chronic diarrhoea: Insights derived from intestinal perfusion studies in 31 patients. In: Krejs GJ, ed. Clinics in gastroenterology: Diarrhoea, 15. Philadelphia: WB Saunders, 1986:477–90. 4. Read NW. Diarrhe´ e Motrice. In: Krejs GJ, ed. Clinics in gastroenterology: Diarrhoea, 15. Philadelphia: WB Saunders, 1986:657–86. 5. Lonnerblad L. Transit time through the small intestine. Acta Radiol 1951;88(suppl):1–85. 6. McLeod JG, Tuck RR. Disorders of the autonomic nervous system: Part 2: Investigation and treatment. Ann Neurol 1987; 21:519 –29. 7. Stone MJ. Amyloidosis. A final common pathway for protein deposition in tissues. Blood 1990;75:531–45. 8. Benson MD, Wallace MR, Tejada E, et al. Hereditary amyloidosis: Description of a new American kindred with late onset cardiomyopathy (Appalachian amyloid). Arthritis Rheum 1987;30:195–200. 9. Eherer AJ, Fordtran JS. Fecal osmotic gap and pH in experimental diarrhea of various causes. Gastroenterology 1992; 103:545–51. 10. Haverback BJ, Dyce BJ, Gutentag PJ, Montgomery DW. Measurement of trypsin and chymotrypsin in stool: A diagnostic test for pancreatic exocrine insufficiency. Gastroenterology 1963;44:588 –97. 11. Schiller LR, Davis GR, Santa Ana CA, Morawski SG, Fordtran JS. Studies of the mechanism of the antidiarrheal effect of codeine. J Clin Invest 1982;70:999 –1008. 12. Arrambide KA, Santa Ana CA, Schiller LR, et al. Loss of absorptive capacity for sodium chloride as a cause of diarrhea following partial ileal and right colon resection. Dig Dis Sci 1989;34:193–201. 13. Fordtran JS, Rector FC, Carter NW. The mechanisms of sodium absorption in the human small intestine. J Clin Invest 1968;47:884 –900. 14. Hyden S. A turbidimetric method for the determination of

18. 19. 20. 21. 22. 23. 24.

25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35.

2225

higher polyethylene glycols in biological materials. Ann Agr Coll Sweden 1955;22:139 –45. Arnfred T, Fogh K, Pedersen L. A simple method for combustion of faecal samples prior to liquid scintillation counting. Scand J Gastroenterol 1974;9:325–6. Schiller LR, Hogan RB, Morawski SG, et al. Studies of the prevalence and significance of radiolabeled bile acid malabsorption in a group of patients with idiopathic chronic diarrhea. Gastroenterology 1987;92:151–60. Fine KD, Krejs GJ, Fordtran JS. Diarrhea. In: Sleisenger MH, Fordtran JS, eds. Gastrointestinal disease: Pathophysiology, diagnosis, and management, 4th ed. Fromm H, Malavolti M. Bile acid-induced diarrhoea. In: Krejs GJ, ed. Clinics in gastroenterology: Diarrhoea. Philadelphia: WB Saunders, 1986:567–82. Sjovall H. Sympathetic control of jejunal fluid and electrolyte transport. Acta Physiol Scand 1984;535(suppl):1– 63. Chang EB, Fedorak RN. Diabetic diarrhea. In: Field M, ed. Diarrheal diseases. New York: Elsevier, 1991:413–27. Dillard RL, Eastman H, Fordtran JS. Volume-flow relationship during the transport of fluid through the human small intestine. Gastroenterology 1965;49:58 –66. Debongnie JC, Phillips SF. Capacity of the human colon to absorb fluid. Gastroenterology 1978;74:698 –703. Quigley EMM, Borody TJ, Phillips SF, et al. Motility of the terminal ileum and ileocecal sphincter in healthy humans. Gastroenterology 1984;87:857–66. Ikeda S, Yanagisawa N, Hongo M, Ito N. Vagus nerve and celiac ganglion lesions in generalized amyloidosis: A correlative study of familial amyloid polyneuropathy and AL-amyloidosis. J Neurol Sci 1987;79:129 –39. Brody IA, Wertlake PT, Laster L. Causes of intestinal symptoms in primary amyloidosis. Arch Intern Med 1964;113: 512–8. Strich SJ, Wade G. Primary amyloidosis presenting with peripheral neuritis and intractable heart-failure. Lancet 1953;2: 70 –1. Cannon WB. The motor activities of the stomach and small intestine after splanchnic and vagus section. Am J Physiol 1907;17:429 –42. Alvarez WC. The extrinsic nerves of the digestive tract and their functions. In: An introduction to gastro-enterology, 4th ed. New York: Paul B Hoeber, 1948:246 –93. Bingham JR, Ingelfinger FJ, Smithwick RH. The effects of sympathectomy on the motility of the human gastrointestinal and biliary tracts. Gastroenterology 1950;15:6 –13. Thompson DG, Ritchie HD, Wingate DL. Patterns of small intestinal motility in duodenal ulcer patients before and after vagotomy. Gut 1982;23:517–23. Wingate D, Hongo M, Kellow J, et al. Disorders of gastrointestinal motility: Towards a new classification. J Gastroenterol Hepatol 2002;17:S1–14. Schiller LR. Anti-diarrhoeal pharmacology and therapeutics. Aliment Pharmacol Ther 1995;9:87–106 (review article). Gertz MA. Diagnosing primary amyloidosis. Mayo Clin Proc 2002;77:1278 –9 (editorial). Afzalpurkar RG, Schiller LR, Little KH, et al. The self-limited nature of chronic idiopathic diarrhea. N Engl J Med 1992;327: 1849 –52. Mintz ED, Parsonnet J, Osterholm MT. Chronic idiopathic diarrhea. N Engl J Med 1993;328:1713–4.