Zaigham Abbas  ( Department of Medicine, The Aga Khan University Medical Centre, Karachi. )
Abrar Aleem Qureshi  ( Department of Medicine, The Aga Khan University Medical Centre, Karachi. )
Hizbullah Sheikh  ( Department of Pathology, The Aga Khan University Medical Centre, Karachi. )
S. M. Wasim Jafri  ( Department of Medicine, The Aga Khan University Medical Centre, Karachi. )
Abdul Haleem Khan  ( Department of Medicine, The Aga Khan University Medical Centre, Karachi. )

Histological changes in 20 Giardia positive duodenal biopsies (Group A) were compared with 50, Giardia negative duodenal biopsies (Group B), taken during the same period. Stool examinations in Group B were negative for Giardia. Surface epithelium, villous and crypt architecture and cellular infiltrates were examined and compared between the groups. Atrophic changes in the villi were more common in Group A as compared to B(P<0.0001). lntraepithelial neutrophil infiltration (P<0.001), infiltration of the lamina propria with plasma cells (P<0.4), and presence of eosinophils in the lamina propria (PcO.001) were significant findings in group A. Some of the changes were related to the density of Giardia colonization e.g., the goblet celldepletion (P<0.05) and the density of plasma cell infiltration in lamina propria (PcO.01). Erosions and ulcerations were less commonly seen in group A. Thus we conclude that giardiasis manifests its peculiar features in the distal duodenal mucosa and a biopsy of this region is an important diagnostic tool for detection of this disease. (JPMA 44:206,1994).

Giardiasis is a disease of the small intestine caused by the flagellated protozoan, Giardialamblia. It has a worldwide distribution. There is an extreme variation in the clinical response to Giardia in man¹. Patients may remain asympto­matic, develop diarrhoea^2,3 and later on malabsorption⁴ or present with symptoms of dyspepsia such as bloating belching, nausea or epigastric discomfort. The histological^5-8 and immunological^9-13 responses are also variable. Available methods for diagnosis of giardiasis include stool examina­tion^14-17 the enterotest¹⁸, duodenal aspirate^19-21 endoscopic brush cytology²², biopsy examination^23-26 and inununodiag­nostic techniques^27-32. In the developing world where Giardi­asis still remains one of the important unresolved health problems, selection of a single and reliable diagnostic tech­nique remains elusive³³. In order to define the histopathologi­cal changes associated with giardiasis, we reviewed our cases diagnosed on distal duodenal biopsy alone.

Twenty consecutive cases of giardiasis (Group A), diagnosed between 1990 and 1992, on histopathological evaluation of distal duodenal biopsies, were included in this study. These biopsies were taken from the distal most accessible part of the duodenum during upper gastrointestinal endoscopy. Fifty cases were selected from all the distal duodenal biopsies done during the same period using simple random sampling (Group B). These patients had upper gastrointestinal complaints and duodenal biopsies were rou­tinely obtained. Both the biopsies and the stool examinations of these patients were negative for Giardia. The specimens were fixed in 10%bufferedformalinand embedded in paraffin wax. Atleast, seven sections were obtained on the microtome for each specimen, each 3-5 um thick, and stained with haematoxylin and eosin. In doubtful cases, sections were stained with Giemsa\\\'s. All the slides were reviewed under the light microscope prospectively in accord­ance with a designed proforma. Surface epithelium was reviewed for nature of cells, whether columnar or cuboidal, and for erosions orulcerations; focal ordiffuse. Erosions were defined as superficial denuded areas involving the upper third of mucosa while ulcerations were considered as excavating lesions involving more than two thirds of the mucosa. The pattern was defined semi-quantitatively as (a) normal, (b) shortened and widened, (c) partial atrophy and (d) subtotal atrophy. The crypts were examined for distortion, elongation, branching and irregularity and the crypt-villous ratio was recorded, normal being 1:3. Crypt hyperplasia was graded as mild, moderate or severe depending on the number of mitoses, hyperchromasia and cellular stratification. Type of cellular infiltration in the surface epithelium and lamina propria was noted and graded for presence o lymphocytes, plasma cells, eosinophils and neutrophils. Lymphoid follicles were defined as focal aggregates of lymphocytes with or without genninal centres. Two more histopathological features were defined for group A only - Goblet cell depletion in the surface epithelium and density of plasma cell infiltrate in the lainina propria. Goblet cells were graded semi-quantitatively as normal or reduced when viewed under high power and plasma cell infiltrates as normal, Grades I, II and Ill. The relative density of colonization by Giardia was determined as Grade I: <5, Grade II: 5-10, Grade Ill:>l0 organisms seen per high power field. Histopathological findings were analyzed using dBASE lIT PLUS and SPSS computer packages. P value was calcu­lated by applying Chi square with Yate’s correction.

The number of cases in Group A were 20, 14 males and 6 females, a ratio of 2.3:1 and a meanage of 38.6 years (Mean age ± standard deviation = 38.6±16.2), ages ranging between 22 to 66 years. The number of cases in group B were 50, 28 males and 22 females, a ratio of 1.3:1 and a mean age of 35.2 years (Mean age±standard deviation = 35.2±17.7), ages ranging between 17-72 years. The histological changes have been described below as a comparison between the two groups, along with their respective P values. (Table I, Figures 1-3).




Atrophic changes in the viIIi, crypt hyperplasia, in­tnepithelial neutrophil infiltration, plasma cell predominance in the lanilna propria along with lymphocytes and eosinophil presence were the significant features of group A (Figure 4).

Erosions were seen more commonly in group B. Group B had two cases of ciyptosporidiosis where some of the changes were similar to those of group A. Though lymphoid aggregates were present in some cases of Giardiasis, there was no significant difference between the gmups. Mild to moderate goblet cell depletion was observed while reviewing the slides of Group A (Figure 5).

The goblet cell depletion and the density of plasma cell infiltration in the lanîina propria had a statistically significant correlation with density of giardial colonization (Table II).

Such a relation could not be obtained for the other histophathological charac­teristics.

Giardia lamblia adheres to the upper small intestinal epithelial cells and induces variable degrees of enteropathy⁴. It may also invade the mucosa³⁴. Therefore, the biopsy of this region would not only demonstrate the organism but also the histological changes induced by this protozoan. If these changes were peculiar to giardiasis, then the presumptive diagnosis of giardiasis could be suggested even in the absence of Giardia in the biopsy specimen. Unfortunately, many such changes described in the literature are non-specific. The surface epithelium in our cases was columnar with regular nuclear alignment in 95%, which was significantly different from Group B where 66% specimens had a regular nuclear alignment The surface epithelium was intact in most cases of giardiasis. The presence of erosions and ulcerations was more common in the control Group B. ViIIi were shortened, widened or partially atrophic in 70% with an altered crypt-villous ratio in up to 45% of cases. Both these features were statistically significant when compared with Group B. The disease process is not fully understood35 but it appears that Giardia is not directly cytopathic to the intestinal cells as can be seen by the lack of erosions. It is probably more of an immunologically mediated process and some of its features resemble those of coeliac disease^26,36. The changes in the viIIi point towards an insidious onset and slow disease process, again signifying an immunological pathogenesis. The presence of intraepithelial lymphocyte infiltration has been described in the literature^5,6,37. In our study, the most frequent cellular infiltrate was of neutrophils, probably suggesting an acute inflammatory response to the organisms. This can also be explained by the fact that in our hospital, biopsies are taken earlier in the disease process, when the patient initially presents with gastrointestinal complaints. In a study of the time course of infection with Giardia muris, intraepithelial lymphocytes increased only after villous short­ening. By this time decrease in brush border disaccharidase had already taken place³⁸. Neutrophil infiltration could also be a response to the other organisms including bacteria which produce infection simultaneously³⁹. A striking increase of plasma cell infiltrates was observed in the laminaptopria. It should be noted that all of our cases, except one case of IPSID, were apparently immune­competent. This is then in contrast to IgA deficiency states where plasma cells may be reduced or absent, and if normal, produce IgM antibodies. Predominance of IgM containing cells in the intestinal mucosa has been described before with Giardia infection^23,40,41. The statistically significant presence of eosinophils in the lamina propria may suggest a hypersen­sitivity response and degranulation of these cells may be responsible for some of the histological changes described. We were not able to demonstrate organisms in the mucosa or lamina propria but invasion by giardia has been demonstrated before³⁴. Lymphoid follicles were present in few cases of both groups and their frequency did not differ significantly between the groups. There were some common features of giardiasis in the distal duodenal biopsies like intact surface epithelium with decreased goblet cells, a vaiying degree of villous atrophy, intraepitbelial neutrophils, dense plasma cell infiltration and presence of eosinophils in the lamina propria. A decrease in goblet cells and an increase in plasma cells in the lamina propria are directly related to the density of giardial coloniza­tion. Therefore, in the presence of constellation of these peculiar hi histopathological features, a careful search should be made for Giardia lamblia in the biopsy specimens.

We are grateful to Mr. Sulaiman S. Gilani and Ms.Farida S. Mithani for their help in the preparation of this manuscript.

1. Wolfe, M.S. Giardiasia. N. Engi. J. Met, 1978,298:319-21.
2. Moore, CT., Cross, W.M., McGuire, D. etal. EpidemicGiardiasis ala Ski resost. N. Engl. J. Med., 1969;281 :402-7.
3. Eastham, E.J., Watson, A. J. and Douglas A.P. Diagnosis of Giardia lambha infection as a cause of diarrhoea. Lancet, 1976;11:950-51.
4. Brasitus, T.A. Parasites and malabsorption. Ant J. Med. 1979;67: 1058-65.
5.  Obemhuber, G and Stolte, M. Giardiasia: analysis of histological changes in biopsy specimens of 80 patients. J. Clin. pathol., 1990;43:641-43.
6.  Wright, S.C. and Tombkins, A.M. Quantitation of the lymphocytic infiltrate in jejuna epithelium ingiardiasia. Gin. Exp. Immunol., 1977,29:408-12.
7. Ament, ME. and Ruhin, C.E. Relation of giardiasis to abnormaal intestinal structure and function in gastrointestinal immunodeficiency syndromes. Gasteroenterology, 1972;62:216-26.
8. Yardley, J.H., Takano, J. and Hendrix, T.R. Epithelial and other nsucosst lesions of the jejunurn in gisrdissis. Jejuna biopsystudies. Bull. Johns Hopkins Hosp. 1964;15:389-406.
9. Editosial. Baffles against Giardis ingutmucoaa. Lancet, 19:527-28.
10. Kraft, S.C. The inteatinal insmunc response in giardissis. Gastroenterology, 1979;76:877-9.
11. Webster, A.D.B., Kenwright, S., Ballard, J/. et at Nodular lymphoid hyperplasia of the bowel in primary hypogammagl obulinemia; study of in vivo and in vitro lymphocyte functiion. Gut, 1977;18:364-72.
12. Nagura, H., Kobles PP. and Brown, WR. lmmunohistochemscal chsrscterizssion of tlw lymphocytes innodular lysnphoid hyperplasis ofthe bowel. Lab. Invest., 1979;40:66-73.
13. Kanwsr, S.S. Ganguly, N.K., Walia, B.N.S., et al. Direct and antibody dependent ccli mediated cytotoxicity against Giardia lamblia by splenic and intestinal lymphoid cells in mice. Gut 1986;27:73-77.
14. Danciger, M. and Lopez, M. Number of giardia in the faecesofinfectedchildren. Anti. Trop. Med., Hyg., 1975;24:237-42.
15. Kamatch, K.R. and Murugasu, R. A comparative studyoffourmethods for detecting Giardla lamblis in children with diarrhoeal disease and malsbsorption. Gastroenterolog% 1974;66:16-21.
16. Kscrnsrski, E.B. and Jones, D.M. Letsleeping Giardislie. Lancet, 1989;11:872.
17. Gilt, A.P. Let sleeping Giardialie. Lancet 1989;11:1106-7
18. Rosenthal, P and Liebman, W.M. Comparative study of stool examination, duodenal aspiration andpediatsic Entero-test forgisrdiasis in children. J.Pediatr., l980;96:278-9.
19. Zafar, MN., Baqai, K., Loch, T.Z., et al. Giardia lamblia in patients undergoing upper Gl endoscopy. J.Pak.Med.Assoc., 1991;41:74-5.
20. Madangopalan, N., Probhaksn, RU., Somssssnderanz A. et al. A correlative study of duodenal aspirate and faeces examination on giardissis before and after treatment with metronidazole. Curr. Met Ret. Opin., 1975;3:99-103.
21. Gordts, B., Hemelof, W, Retort, P., et al. Routine culture of Gisrdialamblis trophozites from husnan dssodenal aspirates. Lancet, 1984;11:137-8.
22. Marshall, J.B., Kelley, D.H. and Vogele, K.A. Giardiasis: diagnosis by cndoscopic brush cytology of the duodenum. Am. J. Gastroenterol., 1984;79:517-29.
23. Ridley, M.J. and Ridley, D.S. Semm antibodies and jejunal histologyin giardissls associated With malsbaorption. J. Chit Pathol., 1976,29:30-34.
24. Atoned, S. DisgnoaisofGisrdiasis. J. pat Met Assoc. 1991 ;41 :73-4.
25. Igne, P.M.G., Edson, C.M. and Farthing, M.J.G. Attachment of Giardis lamblis to rat intestinal epithelial cells. Gut 1988;29:795-801.
26. Levinson, J.D. and Nastro, L.J. Giardiasia with total villoua atrophy. Gastroenterology, 1978;74:271-75.
27. Coke, A.K.J., Rolston,D.D.K., Mathan, V.I., et al. Diagnosis of giardiasis by specific 1gM antibody enzyme-linked immunosorbent assay. Lancet, 1986;11:184-86.
28. Ungar, B.L.P., Yolken, R.H., Nash, T.E., et al. Enzyme-linked immunosorbent assay for the detection of Giardia lamblia in fseca! specimens. J.tnfeet. Dis., 1984;149:90-97.
29. Green, EL., Miles, M.A. and Warhurst, D.C. Immunodiagnostic detection of Giardia antigen in faeces by arapidvisual enzyme- linkedimmunosorbentassay. Lancet, 1985;ll :691-93.
30. Vsaveavara, OS., Smith, PD., Hea!y, O.K., et al. An immunofluorescenece test to detectserum antibodies to Giardialamblia. Ann. Intern. Med., 1980;93:802-5.
31. Smith, PD., Gillin, FIX, Brown, W.R., etal. IgO antibody to Gisrdia lamblia detected by enzyme-linked immunosorbent assay. Gastroenttrology, 1981 80:1476-80.
32. N6ash,tG., Harsington, D.A. and Levine, M.M. Usefulness and enzyme-linked immunosor­bent assay for detection of Giardia lamblia antigen in faeces. J.Clin. Microhiol., 1987;25: 1169-71.
33. Bsqsi, K. Diagnosis of giardissis (editorial) J. Pak. Met Asaoc., 1993;43:250.
34. Saha, T.K. and Ghosh, T.K. Invasion of small intestinal mucosa by Gisrdia lamblia in man. Gastroenterology, 1977;72:402-5.
35. Katelaris, PH. and Farthing, M.J.G. Diarrhoea and ma!absorption in giardiaais: A multifictorial process? Gut, 1992;33:295-7.
36. Cortner, J.A. and Tenn, N. Giardiasis; a cause of Celiae syndrome. Am. J. Dis. Child., 1959;98:311-16.
37. Ferguson, A., McClure, J.P./ and Toconley., RAW. Intrsepithelisl lymphocyte counts in small intestinal biopsies from children with diarrhoea. Acts. Paediatr. Scant, 1976;65:541-6.
38. Gillon, J., Al-Thamery, D. and Ferguson, A. Features of small intestinal pathology (epithelia! cell Kinetics intsaepitheliat lymphocytea, disaocharidases) in a primary Giardia murla infection. Gut, 1982;23:498-506,
39. Tandon, SN., Tandon, R.K., Satpathy, B.K., et al, Mechanism of malabaorption in giardiasis; a study of bacterial flora and bile salt deconjugation in upper jejunwn. Gut, 1977;18: 176-81.
40. Baveja, U.K. and Warhurat, D.C. Humoral response in giardiasis. J. Common. Dis. 1983;15:256-60.
41. Thompson, A., Rowland, K., Hecker, K., et al. Immunoglobulin bearing cells in giardiasia. J. Clin. Pathol., 1977;30:292-94.