PEDIATRICS Vol. 107 No. 6 June 2001, p. e93

ELECTRONIC ARTICLE:
Macroamylasemia Attributable to Gluten-Related Amylase Autoantibodies: A Case Report

Graziano Barera, MD^*, Elena Bazzigaluppi, BSc, Matteo Viscardi, MD^*, Federica Renzetti, MD^*, Cesare Bianchi, MD^*, Giuseppe Chiumello, MD^*, and Emanuele Bosi, MD

From the Departments of ^* Pediatrics and  Medicine, Scientific Institute H San Raffaele, University of Milan, Milan, Italy.

	ABSTRACT

Top Abstract Methods Results Discussion References

Background.  Macroamylasemia (MA) is a benign condition caused by circulating macroamylase complexes of pancreatic or salivary amylase bound to plasma proteins, which cannot be cleared by the renal glomeruli. In most cases, the macromolecular amylase represents a complex of normal amylase and either immunoglobulin A or G and may be a specific antigen-antibody complex. Celiac disease (CD) is a permanent intolerance to ingested gluten that results in immunologically mediated inflammatory damage of the small intestinal mucosa. Several recent population-based serologic surveys have shown CD to be a common disorder, possibly affecting 1 in 200 to 250 individuals in most countries studied, including the United States, where overt CD is rare, indicating a high proportion of subclinical disease. The diagnosis of CD currently rests on the histological demonstration of the characteristic lesion in the small intestine and the subsequent clinical response to the introduction of a gluten-free diet. MA associated with CD has been described in adult patients, and in a few cases, MA decreased or resolved after a strict gluten-free diet. A few single cases of MA have been described in childhood, but no association with CD has been reported so far. We report a girl with CD, autoimmune thyroiditis, and MA, in whom CD-related antibodies to amylase and to exocrine pancreas tissue resolved with a gluten-free diet.

Case Report.  An 11-year-old girl was referred for chronic abdominal pain and growth retardation associated with persistent hyperamylasemia and suspected chronic pancreatitis. We confirmed elevated serum amylase, normal serum lipase, and very low 24-hour urine amylase and amylase clearance/creatinine clearance ratio, consistent with MA. Serologic tests for CD were positive, and the diagnosis was confirmed by small bowel biopsy showing subtotal villous atrophy. Thyroid function tests showed a pronounced hypothyroidism, associated with high titers of thyroid microsomal and thyroglobulin antibodies. Screening for other autoantibodiesincluding antinuclear, islet cell, glutamic acid decarboxylase, protein tyrosine phosphatase islet antigen 512, adrenal gland, and cytoplasmic neutrophil granulocyte antibodieswas negative. A diagnosis of CD, MA, and hypothyroidism attributable to autoimmune thyroiditis was made. A gluten-free diet and oral replacement with L-thyroxine was started with clinical improvement. Serum amylase and amylase clearance/creatinine clearance ratio normalized, consistent with resolution of MA.

Study Design and Methods.  The patient's serum samples were obtained at the time of CD diagnosis and at 3 and 12 months after instituting a gluten-free diet. Serum samples from 10 consecutive untreated celiac children were disease controls, and 39 participants with no gastrointestinal symptoms and no family history of CD served as healthy controls. The origin of MA as determined by complexes of amylase with circulating immunoglobulins was tested by the measurement of amylase on supernatants after precipitation of immune complexes with either protein A Sepharose or polyethylene glycol. The precipitation of >60% of amylase activity was consistent with the presence of MA. Immunoglobulin G (IgG) and immunoglobulin A (IgA) circulating autoantibodies to amylase were measured using recently developed enzyme-linked immunosorbent assay (ELISA), using porcine amylase as antigen. Results were expressed as arbitrary units (AUs). Statistical analysis was performed by Student's t test for unpaired data. IgA and IgG antibodies to exocrine pancreas tissue were detected by indirect immunofluorescence on human pancreas cryosections.

Results.  Serum immunoprecipitation with either protein A Sepharose or polyethylene glycol reduced amylase activity from 1698 to 89 U/L (94.8%) and to 75 U/L (95.6%), with only marginal reduction in control serum samples. The ELISA for autoantibodies to amylase detected high values, both IgA (3531 AU) and IgG (1855 AU), in the serum sample from the patient at CD diagnosis. IgA autoantibodies (mean ± standard deviation) were 3.4 ± 2.5 AU in healthy controls, and 2.1 ± 1.2 AU in celiac controls; IgG autoantibodies were 10 ± 4.8 AU in healthy controls and 8.5 ± 3.2 AU, respectively. Autoantibodies to exocrine pancreas tissue were documented in patient sera at the time of CD diagnosis, both IgA and IgG, but not in control groups. Preincubation of patient's serum with excess of -amylase specifically inhibited antibody binding to coated amylase in the ELISA, and partially inhibited immunoreactivity to exocrine pancreas. Autoantibodies to -amylase and to exocrine pancreas declined in CD patients after institution of a gluten-free diet.

Conclusions.  Few cases of MA have been described in children, and in all amylase determination was part of the clinical investigation for abdominal pain or trauma. We describe the first pediatric case report of MA associated with CD and autoimmune thyroiditis. The association of autoimmunity to exocrine pancreas tissue with CD is intriguing. CD is frequently found in association with pancreatic islet cell autoimmunity and with clinical type 1 diabetes. In our case, endocrine pancreas seemed to be spared by the autoimmune process, because we could not demonstrate islet cell and islet-specific autoantibodies. A relationship of pancreatic autoimmunity and CD is suggested by the parallel decline and disappearance of both CD and pancreas autoantibodies after gluten withdrawal. These finding are consistent with the increasing recognition of autoantibodies as a manifestation of CD and regression of these antibodies with treatment of a gluten-free diet. The mechanisms underlying the formation of MA in CD and the site of production of amylase-binding antibodies have been poorly defined and can only be hypothesized. It is possible that at the intestinal level, crossreactivity either with gluten-related or other antigens occurs, resulting in autoantibody formation to pancreas serum amylase. Nevertheless, our findings suggest a correlation between CD and MA, because it resolved with institution of a gluten-free diet. Because CD may be silent and undiagnosed, we suggest screening for the disease in patients with MA.  Key words:  childhood, celiac disease, thyroiditis, antiendomysium antibodies, antitransglutaminase antibodies, autoantigens, enzyme-immunoglobulin complex.

Macroamylasemia (MA) is a benign condition caused by circulating macroamylase complexes of pancreatic or salivary amylase bound to plasma proteins, which cannot be cleared by the renal glomeruli.¹ The finding of elevated serum amylase activity in the presence of low urinary amylase should alert the clinician to the presence of MA.^1,2 In most cases, the macromolecular amylase represents a complex of normal amylase and either immunoglobulin A (IgA)^1-6 or immunoglobulin G (IgG)^1,6 and may be a specific antigen-antibody complex.^1,2 Although MA has been described among patients with malabsorption, chronic alcoholism, cancer, autoimmune disorders, liver disease, and diabetes, these associations are thought to be incidental.

Celiac disease (CD) is a permanent intolerance to ingested gluten that results in immunologically mediated inflammatory damage of the small intestinal mucosa.⁷ The classic clinical presentation of malabsorption has become less common. Atypical features, such as nonspecific gastrointestinal complaints, short stature, iron deficiency anemia, and dental abnormalities, may predominate. However, the majority of patients have what is termed silent CD, which may remain undiagnosed because the condition has no symptoms. Several recent population-based serologic surveys have shown CD to be a common disorder, possibly affecting 1 in 200 to 250 individuals in most countries studied, including the United States, where overt CD is rare, indicating a high proportion of subclinical disease.⁷ The diagnosis of CD currently rests on the histologic demonstration of the characteristic lesion in the small intestine and the subsequent clinical response to the introduction of a gluten-free diet.⁸ Serologic studies are now used to further confirm the diagnosis and to screen populations at risk for the disease. These include the enzyme-linked immunosorbent assay (ELISA) for IgA antibodies to gliadin and the immunofluorescence test for IgA antibodies to endomysium, the presence of which is virtually pathognomonic for CD. Measurement of antitissue transglutaminase antibodies might be an alternative and more objective method for detecting antiendomysium antibodies, proving to be highly sensitive and specific.⁹

MA associated with proven^2,4,510-13 or likely^3,14,15 CD has been described in adult patients. In a few cases,^2,4,10,13,14 MA decreased or resolved after a strict gluten-free diet, whereas in others it did not.^5,11 A few single cases of MA have been described in childhood,^16-20 but no association with CD has been reported so far.

We report a girl with CD, autoimmune thyroiditis, and MA, in whom CD-related antibodies to amylase and to exocrine pancreas tissue resolved with a gluten-free diet.

	CASE REPORT

An 11-year-old girl was referred for chronic abdominal pain and growth retardation associated with persistent hyperamylasemia and suspected chronic pancreatitis. Physical examination revealed height of 1.30 m (<third percentile), weight of 25 kg (<third percentile), prepubertal state, dry skin, and mild abdominal distension. We confirmed elevated serum amylase, normal serum lipase, and very low 24-hour urine amylase and amylase clearance/creatinine clearance (amCl/crCl) ratio (Table 1), consistent with MA. Abdominal ultrasonography and magnetic resonance cholangiopancreatography were normal. Serologic tests for CD (Table 1) were positive, and the diagnosis was confirmed by small bowel biopsy showing subtotal villous atrophy. Thyroid function tests showed a pronounced hypothyroidism with elevated thyroid-stimulating hormone (969 mU/L; normal values = 0.27-4.2), low free triiodothyronine (0.8 ng/L; normal values = 1.8-4.6), and low free thyroxine (1.3 pmol/L; normal values = 11.5-24.5), associated with high titers of thyroid microsomal and thyroglobulin antibodies. Screening for other autoantibodiesincluding antinuclear, islet cell, glutamic acid decarboxylase, protein tyrosine phosphatase islet antigen 512, adrenal gland, and cytoplasmic neutrophil granulocyte antibodieswas negative. A diagnosis of CD, MA, and hypothyroidism attributable to autoimmune thyroiditis was made. A gluten-free diet and oral replacement with L-thyroxine was started with clinical improvement (weight gain = 8 kg/year; height velocity = 12 cm/year). Serum amylase and amCl/crCl ratio normalized, consistent with resolution of MA.

	METHODS

Top Abstract Methods Results Discussion References

Patient and Control Sera

The patient's serum samples were obtained at the time of CD diagnosis and at 3 and 12 months after instituting a gluten-free diet. Serum samples from 10 consecutive untreated celiac children (mean age: 8.8 years; range: 6.4-11.8) fulfilling the European Society of Pediatric Gastroenterology, Hepatology, and Nutrition criteria⁸ were disease controls; thyroid function was normal in all. Thirty-nine participants (mean age: 10 years; range: 6.1-12.3) with no gastrointestinal symptoms and no family history of CD served as healthy controls. All studies of celiac and healthy control participants were performed in accordance with the ethic rules of the institute.

MA as a Complex of Amylase With Immunoglobulins

The origin of MA as determined by complexes of amylase with circulating immunoglobulins was tested by the measurement of amylase on supernatants after precipitation of immune complexes with either protein A Sepharose or polyethylene glycol. The precipitation of >60% of amylase activity was consistent with the presence of MA.²¹

Amylase Antibodies by ELISA

IgG and IgA circulating autoantibodies to amylase were measured using a recently developed ELISA,⁶ using porcine amylase as antigen. Results were expressed as arbitrary units (AUs). Statistical analysis was performed by Student's t test for unpaired data.

Exocrine Pancreas Antibodies by Immunofluorescence

IgA and IgG antibodies to exocrine pancreas tissue were detected by indirect immunofluorescence on human pancreas cryosections.²²

Specificity of Antibody Binding to Amylase

Specificity of antibodies for amylase was tested by inhibition of antibody binding with antigen excess in the ELISA and immunofluorescence reactivity assay on pancreas cryosections. All 3 patient's serum samples and serum samples of 4 control participants were retested in the 2 assays after competition with porcine amylase. Sera (25 µL) were incubated with amylase preparation (5 µg in 25 µL) for 1 hour at 4°C and then tested in the 2 assays.

	RESULTS

Top Abstract Methods Results Discussion References

Serum immunoprecipitation with either protein A Sepharose or polyethylene glycol reduced amylase activity from 1698 to 89 U/L (94.8%) and to 75 U/L (95.6%), with only marginal reduction in control serum samples.

Autoantibodies to -amylase declined in CD patients after institution of a gluten-free diet (Fig 1). Preincubation of patient's serum with excess of -amylase specifically inhibited antibody binding to coated amylase in the ELISA.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Autoantibodies to amylase at CD diagnosis and after gluten-free diet, compared with autoantibody concentration detected in healthy and celiac control groups (mean ± standard deviation). Differences between control groups were not significant.

IgA and IgG class autoantibody results to exocrine pancreas tissue, detected in patient's sera, are reported in Table 1; they were undetectable in either control group. Preincubation with excess of -amylase only partially inhibited immunoreactivity to exocrine pancreas.

	DISCUSSION

Top Abstract Methods Results Discussion References

Few cases of MA have been described in children,^16-20 and in all cases, amylase determination was part of the clinical investigation for abdominal pain or trauma. Amylase-IgA¹⁶ or amylase-IgG^18,19 immunocomplexes were documented by immunoelectrophoresis^16,18 or fast protein liquid chromatography.¹⁹ No pancreatic damage was demonstrated, and MA persisted despite the disappearance of abdominal pain.^16-19 We describe the first pediatric case report of MA associated with CD and autoimmune thyroiditis. We also quantitatively measured IgA and IgG autoantibodies to amylase by ELISA and to human exocrine pancreas by immunofluorescence.

The association of autoimmunity to exocrine pancreas tissue with CD is intriguing. CD is frequently found in association with pancreatic islet cell autoimmunity²³ and with clinical type 1 diabetes.²⁴ In our case, endocrine pancreas seemed to be spared by the autoimmune process, because we could not demonstrate islet cell and islet-specific autoantibodies. A relationship of pancreatic autoimmunity and CD is suggested by the parallel decline and disappearance of both CD and pancreas autoantibodies after gluten withdrawal. These findings are consistent with the increasing recognition of autoantibodies as a manifestation of CD^23,25,26 and with regression of these antibodies with treatment of a gluten-free diet.²³ Autoimmune diseases are also quite common in CD, and a pathogenetic role of gluten has been suggested as a major risk factor.²⁴ Although according to recent literature^23,25,26 up to 24% to 52.6% of patients may demonstrate an organ-specific autoantibody, we showed that amylase antibodies are not a common feature of CD, because they were undetectable in 10 untreated celiac controls.

The mechanisms underlying the formation of MA in CD and the site of production of amylase-binding antibodies have been poorly defined and can only be hypothesized. It is possible that at the intestinal level crossreactivity either with gluten-related or other antigens occurs, resulting in autoantibody formation to pancreas serum amylase. Direct sensitization attributable to pancreatic tissue destruction seems less likely because no other signs of tissue damage were found in this case. Nevertheless, our findings suggest a correlation between CD and MA, because it resolved with institution of a gluten-free diet. Interestingly, in the few cases of MA described in childhood, it was reported to persist,^16-20 because no pathogenetic mechanisms and effective treatment were recognized. Therefore, we hypothesize that some patients with MA are associated with silent CD. Because CD may be asymptomatic,⁷ we suggest screening for CD in children with MA.

	ACKNOWLEDGMENT

We thank Dr Franco Cazzulani for referring this patient for study and advice.

	FOOTNOTES

Reprint request to (G.B.) Pediatric Department, University of Milan, Scientific Institute H San Raffaele, Via Olgettina, 60, 20132 Milan, Italy. E-mail: barera.graziano{at}hsr.it

Received for publication Jul 3, 2000; accepted Jan 17, 2001.

	ABBREVIATIONS

MA, macroamylasemia; IgA, immunoglobulin A; IgG, immunoglobulin G; CD, celiac disease; ELISA, enzyme-linked immunosorbent assay; AmCl/crCl, amylase clearance/creatinine clearance; AU, arbitrary unit.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Kanno T, Sudo K Properties of amylase-linked immunoglobulins. Clin Chim Acta 1977; 76:67-77 [CrossRef][Medline]
2. Hodgson HJ, Whitaker KB, Cooper BJ, Malabsorption and macroamylasemia: response to gluten withdrawal. Am J Med 1980; 69:451-457 [CrossRef][Medline]
3. Levitt MD, Goetzl EJ, Cooperbrand SR Two forms of macroamylasaemia. Lancet 1968; 1:957-958 [Medline]
4. Larvol L, Chagnon JP, Cerf M Celiac disease and macroamylasaemia [letter]. Am J Gastroenterol 1991; 86:788 [Medline]
5. Zaman Z, Van Orshoven A, Marrien G, Every J, Blanckaert N Simultaneous macroamylasemia and macrolipasemia. Clin Chem 1994; 40:939-942 [Abstract/Free Full Text]
6. Hortin LG, Summerfield AL, Wilhite TR, Detection of autoantibodies to amylase by ELISA: comparison of detection of macroamylase and free autoantibody. Clin Chem 1994; 40:2254-2259 [Abstract/Free Full Text]
7. Bransky D, Troncone R Celiac disease: a reappraisal. J Pediatr 1998; 133:181-187 [CrossRef][Medline]
8. The Working Group of the European Society for Paediatric Gastroenterology and Nutrition Revised criteria for diagnosis of coeliac disease. Arch Dis Child 1990; 65:909-911 [Free Full Text]
9. Bazzigaluppi E, Lampasona V, Barera G, Comparison of tissue transglutaminase-specific antibody assays with established antibody measurements for celiac disease. J Autoimmun 1999; 12:51-56 [CrossRef][Medline]
10. Van Gossum A, Cremer M Macroamylasaemia disappearance after gluten withdrawal. Digest Dis Sci 1989; 34:964-965
11. Rajvanshi P, Chowdhury JR, Gupta S Celiac sprue and macroamylasaemia: potential clinical and pathophysiological implications. J Clin Gastroenterol 1995; 20:304-306 [Medline]
12. Garcia-Gonzales M, Defarges-Pons V, Monescillo A, Hernandez F, Cano-Ruiz A Macrolipasemia and celiac disease. Am J Gastroenterol 1995; 12:2233-2234
13. Bonetti G, Serricchio G, Giudici A, Hyperamylasemia due to macroamylasemia in adult gluten enteropathy. Scand J Clin Lab Invest 1997; 57:271-274 [Medline]
14. Wilding P, Cooke WT, Nicholson GI Globulin-bound amylase: a cause of persistently elevated levels in serum. Ann Intern Med 1964; 60:1053-1059 [Abstract/Free Full Text]
15. Levitt MD, Cooperbrand SR Hyperamylasaemia from the binding of serum amylase by an 115 IgA globulin. N Engl J Med 1968; 278:474-479
16. Larcher WF, Tanner MS, Mowat AP, Bamforth F, Keenan JR Macroamylasaemia and hepatitis in a twelve-year-old boy. Lancet 1980; i:680-681
17. Colon AR, Felder RA, Ryan TM Macroamylasemia. J Pediatr 1980; 96:64-66 [CrossRef][Medline]
18. Catassi C, Guerrieri A, Natalini G, Busco F, Giorgi L Macroamylasaemia and selective IgA deficiency. Arch Dis Child 1986; 61:704-706 [Abstract/Free Full Text]
19. D'Avanzo M, Cobbaert C, Tolone C, Macroamylasemia in a 5-year-old girl. J Pediatr Gastroenterol Nutr 1992; 14:104-106 [Medline]
20. Gleghorn EE, Thomas DW Macroamylasemia in a child with abdominal trauma. J Pediatr Gastroenterol Nutr 1993; 16:102-103 [Medline]
21. Levitt MD, Ellis C A rapid and simple assay to determine if macroamylase is the cause of hyperamylasemia. Gastroenterology 1982; 83:387-392
22. Lendrum R, Walker G Serum antibodies in human pancreatic disease. Gut 1975; 16:365-371 [Abstract/Free Full Text]
23. Ventura A, Neri E, Ughi C, Leopaldi A, Città A, Not T Gluten-dependent diabetes-related and thyroid-related autoantibodies in patients with celiac disease. J Pediatr 2000; 137:263-265 [CrossRef][Medline]
24. Ventura A, Magazzù G, Greco L Duration of exposure to gluten and risk for autoimmune disorders in patients with celiac disease. SIGEP Study Group for Autoimmune Disorders in Celiac Disease. Gastroenterology 1999; 117:297-303 [CrossRef][Medline]
25. Volta U, De Franceschi L, Molinaro N, Tetta C, Bianchi FB Organ-specific autoantibodies in coeliac disease: do they represent an epiphenomenon or the expression of associated autoimmune disorders? Ital J Gastroenterol Hepatol 1997; 29:18-21 [Medline]
26. Toscano V, Conti FG, Anastasi E, Importance of gluten in the induction of endocrine autoantibodies and organ dysfunction in adolescent celiac patients. Am J Gastroenterol 2000; 95:1742-1748 [CrossRef][Medline]