Walter Heise

Small Intestinal Structure and Function in Patients Infected with Human Immunodeficiency Virus (HIV): Evidence for HIV-Induced Enteropathy

STUDY OBJECTIVE To determine small intestinal mucosal structure and function in patients with human immunodeficiency virus (HIV) infection. DESIGN Prospective, consecutive sample study. SETTING Referral-based medical clinics at a municipal and a university medical center. PATIENTS Forty-five HIV-infected patients (44 men, 1 woman) with gastrointestinal complaints. INTERVENTIONS All patients had esophagogastroduodenoscopy. Distal duodenal biopsy samples were examined morphometrically and by quantitative enzyme histochemical techniques. Immunohistologic studies were done to determine whether HIV antigen p24 was present. Biopsy and stool samples were examined for enteric pathogens and patients were evaluated for malabsorption. MEASUREMENTS AND MAIN RESULTS Malabsorption was common in HIV-infected patients. In 15 of 38 patients mononuclear cells infected with HIV were found in the mucosa. In 15 of 25 patients there was no detectable lactase (beta-glucosidase) activity in the duodenal brush border; when measurable, lactase (beta-glucosidase) activity was decreased (P less than 0.02). Alkaline phosphatase activity was normal. Crypt depth was greater (P less than 0.05), villous surface area was slightly smaller, and mitotic figures per crypt were not different in HIV-infected patients compared with controls. Patients without additional intestinal infection had a reduced number of mitotic figures per crypt (P less than 0.05) and normal crypt depth. The reduction in mitotic figures was most pronounced in patients with mucosal HIV antigen p24. CONCLUSIONS The HIV-infected patients with gastrointestinal symptoms show low-grade small bowel atrophy and a maturational defect in enterocytes, which may be caused exclusively by HIV. An additional intestinal infection can mask this mucosal atrophy.

Intestinal Non-Hodgkin’s Lymphoma: A Multicenter Prospective Clinical Study From the German Study Group on Intestinal Non-Hodgkin’s Lymphoma

PURPOSE Intestinal non-Hodgkins lymphomas are not well characterized. We therefore studied prospectively their clinical features and response to standardized therapy. PATIENTS AND METHODS Fifty-six patients with primary intestinal lymphoma were included in a prospective, nonrandomized multicenter study. Lymphoma resection was recommended and staging was performed according to the Ann Arbor classification. Patients were scheduled to receive six cycles of cyclophosphamide, doxorubicin, vincristine, and prednisolone (CHOP) chemotherapy, and at stages EIII to EIV, they received additional involved-field radiotherapy. Corticosteroids were used in patients who could not receive chemotherapy. RESULTS Thirty-five patients had intestinal T-cell lymphoma (ITCL), 21 patients had intestinal B-cell lymphoma (IBCL; 18 diffuse large-cell lymphomas, two marginal-cell lymphomas, and one follicle-center lymphoma). Thirty-four patients at stages EI to EII (14 ITCL and 20 IBCL) and nine patients at stages EIII to EIV (all ITCL) received chemotherapy. No patient in stages EIII to EIV received radiotherapy, because death occurred in 12 of 14 patients. Two-year cumulative survival in patients with IBCL was 94% (95% CI, 82% to 100%) and higher than in patients with ITCL (28% [95% CI, 13% to 43%]; P <.0001), even when only stages EI to EII were considered (ITCL, 37.5% [95% CI, 16.5% to 58.5%]; P <.0001). IBCL patients compared with ITCL patients were at lower lymphoma stages (P <.01), had higher Karnofsky status (P <.005), had intestinal perforation less often (P <.05), required emergency operation less often (P <.05), received CHOP (P <.05) more often, and reached complete remission (P <.0005) more frequently. CONCLUSION IBCL patients at stages EI and EII respond well to chemotherapy, but the prognosis and treatment of ITCL patients is unsatisfactory.

Frequency of clonal intraepithelial T lymphocyte proliferations in enteropathy-type intestinal T cell lymphoma, coeliac disease, and refractory sprue.

BACKGROUND Clonal T cell receptor (TCR) gene rearrangements and loss of T cell antigens such as CD8 and TCR-β in intraepithelial lymphocytes (IELs) may indicate the development of an enteropathy-type intestinal T cell lymphoma (EITCL) in patients with refractory sprue. AIMS To define the diagnostic value of these markers in duodenal biopsies from patients with villous atrophy as a result of various underlying disorders. PATIENTS AND METHODS Duodenal biopsies from eight patients with coeliac disease and five patients with villous atrophy caused by defined disorders were compared with three patients with refractory sprue evolving into overt EITCL, two patients with ulcerative jejunitis, and with eight patients with overt EITCL, for expression of CD3, CD4, CD8, and TCR-β in IELs using immunohistochemistry and for clonal TCR-γ gene rearrangements using polymerase chain reaction. In addition, biopsies from six consecutive patients with refractory sprue of uncertain cause were examined. RESULTS Clonal TCR-γ gene rearrangements were found in all resected tumours of patients with EITCL, in 3/8 duodenal biopsies of patients with EITCL, in 2/2 patients with ulcerative jejunitis, in 2/3 patients with refractory sprue evolving into overt EITCL, and in 1/6 patients with refractory sprue. No rearrangements were found in biopsies from patients with refractory sprue caused by defined disorders or those with coeliac disease. Clonality in duodenal biopsies was associated with an abnormal phenotype of IELs in all cases and in all but one case in patients with evidence of underlying coeliac disease. Specificity for detection of an EITCL using immunohistology was 77% for CD8 and for TCR-β staining, and 100% for detection of a clonal TCR-γ gene rearrangement. Sensitivity was 62% for staining with CD8 and clonality investigation, while sensitivity reached 100% for TCR-β staining in all investigated patients with EITCL. CONCLUSIONS Clonal proliferations of phenotypically abnormal IELs in refractory sprue represent an early manifestation of EITCL, for which the term “sprue-like intestinal T cell lymphoma” is proposed. This constellation is also found in duodenal biopsies from patients with an overt EITCL and is not related to other sprue syndromes, resulting in a high specificity for detection of an EITCL or refractory sprue evolving into EITCL. Overt EITCL may develop directly from coeliac disease without a precursor lesion (refractory sprue with clonal IELs) being demonstrable in duodenal biopsies or via a “sprue-like intestinal T cell lymphoma”. This latter entity is a complication of coeliac disease.

Mucosal Atrophy Is Associated with Loss of Activated T Cells in the Duodenal Mucosa of Human Immunodeficiency Virus (HΙV)-Infected Patients

Gastrointestinal symptoms and malabsorption are frequent in HIV-infected patients even in the absence of opportunistic infections. In earlier studies we found indications that the gastrointestinal mucosa itself may be affected by HIV. Since there is evidence that the mucosal structure is influenced by changes in the gut-associated lymphoid tissue, we have investigated mucosal structure and immune cells in HIV-infected patients. Sixty patients (3 f, 57 m; age 21-61, median 37 years; 11 at CDC stage II or III, 49 at stage IV) with gastrointestinal complaints undergoing upper endoscopy were examined for enteric pathogens. Duodenal biopsies were labelled by immunohistology for HIV antigen p24 and for lymphocyte surface markers; mucosal architecture was studied by three-dimensional morphometry. Biopsies from HIV seronegative patients without abnormal findings served as controls. In 29 patients an enteric pathogen was identified. In 22 patients HIV-infected mononuclear cells were detected in the lamina propria. In the lamina propria CD25+ cells were decreased, CD3+ and CD8+ cells were increased in HIV-infected patients compared with controls, while the numbers of CD4+, Leu8+, and HML-1+ cells, and of macrophages were not different. Patients at stage IV had decreased numbers of CD4+ T cells compared with patients at stage II or III. Villus surface area was reduced in HIV-infected patients compared with controls. Crypt depth was increased in patients with intestinal infection compared with controls while numbers of mitotic figures were normal. Patients without intestinal infection and patients with mucosal HIV-infected cells had decreased numbers of mitotic figures and normal crypt depth compared with controls.(ABSTRACT TRUNCATED AT 250 WORDS)

Serum carotene deficiency in HIV-infected patients

ObjectiveTo assess total serum carotene concentration in HIV-infected patients as an indicator of fat malabsorption in correlation with diarrhoea, secondary enteric infections, and blood lymphocyte subsets. DesignProspective study. SettingTwo referral-based tertiary care centres in Berlin, Germany. PatientsA total of 33 controls and 116 HIV-infected patients who had complete microbiological evaluation of stools and biopsies obtained at upper endoscopy because of diarrhoea (n = 54), or other symptoms (n = 62), were studied. Main outcome measuresTotal serum carotene concentration was determined spectrophotometrically. ResultsTotal serum carotene concentration was abnormal (< 0.88 μmol/l) in 77% of HIV-infected patients and significantly decreased compared with controls [0.47μmol/l (range, 0.06–1.69 μmol/l) versus 1.37μmol/l (range, 0.88–2.92 μmol/l); P<0.0001]. Total serum carotene concentration did not differ between AIDS patients and patients at earlier disease stages, between patients with or without secondary enteric infections, or between patients with or without fever. In patients at earlier disease stages, but not in AIDS patients, total serum carotene concentration was lower for patients with than without diarrhoea. The percentage of CD4 lymphocytes (r = 0.364; P< 0.001), CD4 count (r = 0.28; P= 0.0013), and CD4/CD8 ratio (r = 0.38; P< 0.001) in the peripheral blood correlated with total serum carotene levels in HIV-infected patients. ConclusionHIV-infected patients frequently have abnormal total serum carotene concentrations indicating fat malabsorption which may contribute to diarrhoea. Furthermore, total serum carotene concentrations correlate with immunologic abnormalities in HIV infection.

Effects of zidovudine treatment on the small intestinal mucosa in patients infected with the human immunodeficiency virus

Zidovudine is associated with hematologic toxicity and may also impair the rapidly proliferating intestinal epithelium. However, patients with human immunodeficiency virus (HIV) infection receiving zidovudine gain body weight, indicating improved absorptive function. In the present study, 33 HIV-infected patients with gastrointestinal symptoms who were undergoing duodenoscopy and who had no detectable secondary intestinal pathogens were investigated; 12 of them received zidovudine. HIV antigen p24 was detected in duodenal biopsy specimens by immunohistology in 3 of 12 patients with zidovudine treatment and in 10 of 21 patients without zidovudine treatment. Morphometry of duodenal specimens showed reduced villus surface area (P less than 0.05) without crypt hyperplasia independent of zidovudine therapy and reduced numbers of crypt mitoses in patients with mucosal HIV infection (P less than 0.001) compared with controls. In the duodenal brush border, patients with mucosal HIV infection (P = 0.006) and patients without zidovudine treatment (P = 0.009) had absent lactase/beta-glucosidase activity more frequently than controls, and all HIV-infected patients (P less than 0.025) except zidovudine recipients had decreased alkaline phosphatase activity compared with controls. These findings show a hyporegenerative atrophy of the small intestine and enterocyte dysmaturation associated with mucosal HIV infection. Improved enterocyte maturation, indicated by increased brush border enzyme activity, may contribute to the clinical benefit of HIV-infected patients from zidovudine therapy.

HIV-1 p24 but not proviral load is increased in the intestinal mucosa compared with the peripheral blood in HIV-infected patients

Objective: To investigate differences in viral and proviral load between the peripheral blood and the intestinal mucosal immune system in HIV-infected patients. Design: HIV-1 p24 and HIV DNA content were compared in blood samples and intestinal biopsies from HIV-infected patients. Methods: Intestinal biopsies and peripheral blood were simultanously obtained from 27 HIV-infected patients undergoing diagnostic endoscopy. The p24 concentrations were measured in serum and homogenized intestinal biopsies by enzyme-linked immunosorbent assay after acid-dissociation of immune complexes. Proviral load was determined in blood and intestinal biopsies by a quantitative competitive polymerase chain reaction amplifying the HIV-1 nef gene from genomic DNA. Results: No significant differences were found in proviral load comparing HIV copies per 1.5 x 105 cell equivalents in blood [2650 (600–44 000)] and intestinal biopsies [4200 (1325–19 625)]. Paired analysis revealed a strong positive correlation between serum and mucosal proviral load. In contrast, HIV core protein p24 was detected in intestinal biopsies from 18 patients in much higher concentrations than in serum [858 (262–4111) pg/g versus 34 (9–242) pg/g; P < 0.005]. The p24 concentrations in serum and intestinal biopsies did not correlate and no significant correlation was observed in serum or intestinal biopsies between proviral load and p24 concentrations. No clear correlations were observed between clinical parameters and HIV DNA or HIV p24 levels in blood or biopsies. Conclusions: Our findings demonstrate a homogenous distribution of HIV proviral load in the peripheral blood and the intestinal mucosal immune system. The high viral antigen load in the intestine therefore indicates that mucosal HIV production is upregulated at the transcriptional and/or translational level. The intestinal mucosa is a major reservoir for HIV in HIV-infected patients.

Stool Viruses, Coinfections, and Diarrhea in Hiv-infected Patients

To examine the prevalence of stool viruses and their role in the pathogenesis of diarrhea in HIV infection, we evaluated biopsies and repeated stool samples of 256 HIV-infected patients undergoing diagnostic endoscopy because of diarrhea (n = 136) or other symptoms (n = 120) for bacterial, protozoal, and viral enteropathogens. In 70% of the patients with diarrhea, at least one potential enteropathogen was detected. Stool virus was detected by electron microscopy in 17% (44 of 256), adenovirus in 6.6% (17 of 256), and coronavirus in 11.3% (29 of 256) of the patients. Adenovirus and coronavirus were detected more frequently in patients with diarrhea than in patients without diarrhea [adenovirus 10% (13 of 136) vs. 3.3% (4 of 120), p = 0.0129; coronavirus 15% (21 of 136) vs. 6.6% (8 of 120), p = 0.0142]. Sixty-one percent of patients harboring stool virus were coinfected by another enteropathogen. Pathogens other than stool virus were detected more frequently in patients harboring adenovirus (82%) than in patients without stool virus (48%, p < 0.025). Adenovirus and coronavirus are frequently detected in stools of HIV- infected patients and may contribute to diarrhea. Adenovirus infection may facilitate the occurrence of other intestinal pathogens. Due to frequent coinfections, detection of stool viruses reduces the rate of diarrhea of unknown origin only by approximately 5%.

Mucosal abnormalities in microsporidiosis.

Objective:To determine the prevalence of microsporidiosis in HIV-infected patients with and without diarrhoea and to characterize alterations in mucosal architecture and brush border enzyme activities in patients with microsporidiosis. Patients:A total of 259 HIV-infected patients undergoing oesophagogastroduodenoscopy because of diarrhoea (n = 123) or other symptoms (n = 136) were studied. Methods:Patients were evaluated for the presence of microsporidia by electron microscopy of duodenal biopsies. Brush border enzyme activities were measured by histochemistry and mucosal architecture was determined by three-dimensional morphometry in biopsies from patients with microsporidiosis and compared with biopsies from a subgroup of HIV-infected patients with or without other enteropathogens. Results:Enterocytozoon bieneusi was detected in 17 patients and Encephalitozoon intestinalis was detected in two patients. Microsporidiosis was significantly more frequent in patients with chronic diarrhoea (19.1%; P< 0.0001) or in patients with acute diarrhoea (7.2%; P = 0.04) than in patients without diarrhoea (1.5%). Microsporidiosis was associated with lactase deficiency (P= 0.03) and a reduced activity of alkaline phosphatase (P= 0.028) and α-glucosidase (P= 0.025) at the basal part of the villus compared with brush border enzymes in patients without enteropathogens. Patients with microsporidia had reduced villus height (P= 0.043) and a villus surface reduced by 40% (P = 0.004) compared with patients with enteropathogens other than microsporidia. Conclusions:Our study confirms the association between microsporidia and diarrhoea. The pathophysiologic mechanism by which microsporidia cause diarrhoea appears in part to be malabsorption, caused by a reduction of absorptive mucosal surface and impairment of enterocyte function.

Gastrointestinal symptoms in patients infected with human immunodeficiency virus: relevance of infective agents isolated from gastrointestinal tract.

The correlation of gastrointestinal symptoms and infections in 186 consecutive patients with human immunodeficiency virus (HIV) infection undergoing diagnostic endoscopy (oesophagogastroduodenoscopy, n = 124; colonoscopy, n = 37; both, n = 25) was investigated. Biopsy and stool samples were examined for infective agents. Only weight loss (p = 0.003) and dysphagia (p = 0.027) were more common in patients at stage CDC IV compared with earlier stages. In three of 27 patients at stage II/III and in 93 of 159 patients at stage IV an infective agent was identified in stool or gastrointestinal biopsy specimen (p < 0.001). Cytomegalovirus (n = 35), Candida sp (n = 28), M avium complex (n = 10), and Cryptosporidium (eight) were the most frequent agents detected. At stage IV, diarrhoea was more frequent in infected compared with non-infected patients (p = 0.006); however, an infective agent was also found in 39 of 82 patients at stage IV without diarrhoea. The frequency of gastrointestinal symptoms was not consistently increased in patients harbouring specific infective agents compared with non-infected patients. Our findings indicate that the pathogenic relevance of a gastrointestinal infection in HIV infected patients has to be verified and indirectly support the existence of an HIV associated enteropathy.