Toll-like receptor expression in chronic hepatitis C: correlation with pro-inflammatory cytokine levels and liver injury

BACKGROUND/AIMS

Toll-like receptors (TLR's) are critical receptors that promote innate immune responses to pathogen-associated molecular patterns. Activation of TLR's leads to production of pro-inflammatory cytokines such as tumour necrosis factor (TNF)-alpha. This study investigates whether peripheral blood monocyte expression of TLR's is disturbed in patients with chronic hepatitis C and whether levels of expression of these molecules are significantly correlated with hepatitis C virus (HCV) genotype, viral load, hepatic necroinflammatory activity, histological stage and circulating TNF-alpha concentrations.

METHODS

In 18 non-cirrhotic patients with biopsy-proven, virologically-confirmed chronic hepatitis C and 32 controls, we measured expression of TLR2 and TLR4 on peripheral blood monocytes. HCV genotype, viral load, serum alanine aminotransferase (ALT) levels, histological stage of disease and circulating TNF-alpha and endotoxin levels were also determined.

RESULTS

Peripheral blood monocyte expression of TLR2 and TLR4 were significantly increased in patients with chronic hepatitis C compared to controls, irrespective of HCV genotype or histological stage of disease. Circulating levels of TNF-alpha were also significantly increased in patients with chronic hepatitis C. In both the overall study cohort and patients with chronic hepatitis C, monocyte expression of TLR2, but not of TLR4, correlated significantly with serum TNF-alpha levels. In patients with chronic hepatitis C, monocyte expression of TLR2, but not of TLR4, also correlated significantly with serum ALT levels. Expression of TLR's was not significantly correlated with viral load.

CONCLUSIONS

Up-regulation of peripheral blood monocyte expression of TLR2 and TLR4 occurs in patients with chronic hepatitis C. Increased monocyte expression of TLR2, but not of TLR4, correlates significantly with both increased circulating TNF-alpha levels and hepatic necroinflammatory activity in this disorder.

Correlates of placental infection with cytomegalovirus, parvovirus B19 or human herpes virus 7

Vertical transmission of viruses is an important cause of morbidity in the fetus and neonate. Placental viral infection indicates risk of vertical transmission, but not always transmission to, or disease of the fetus. Specimens from mothers and babies from three groups-two prospective and one retrospective cohort-were tested for pathogens of teratogenic potential using multiplex PCR. Placental infection was present in 13% of the 105 samples collected. Assessment of the prospective cohorts showed cytomegalovirus (CMV) detected in 4% of placentae from unselected women, parvovirus B19 in 1% and Ureaplasma parvum in 1% of placentae. In a retrospective cohort of women at high risk of transmitting congenital infection due to seroconversion during pregnancy, miscarriage or stillbirth, CMV was detected in 64% and human herpes virus type 7 in 9% of placentae. Of 14 PCR-positive placentae, two were associated with the birth of a living symptomatic infant, two with stillbirth, one with miscarriage, and two with elective terminations of pregnancy. Directed laboratory assessment of women at high risk of transmitting congenital infection, on the basis of clinical or laboratory markers, is important for accurate diagnosis of adverse outcomes of pregnancy. However, routine screening for viruses in the placentae from women with a low-risk serological profile for transmitting congenital infection is unlikely to result in significant numbers of additional diagnoses and is confounded by inadequacy of current diagnostic methods. The major pathogen detected in all cases of placental infection associated with fetal death was human CMV.

Development of multiplex PCRs for detection of common viral pathogens and agents of congenital infections

Potential causes of congenital infection include Toxoplasma gondii and viruses such as cytomegalovirus (CMV), enterovirus, hepatitis C virus, herpes simplex virus types 1 and 2 (HSV-1 and -2), human herpesvirus types 6, 7, and 8, lymphocytic choriomeningitis virus, parvovirus, rubella virus, and varicella-zoster virus. Testing for each of these agents using nucleic acid tests is time consuming and the availability of clinical samples such as amniotic fluid or neonatal blood is often limited. The aim of this study was to develop multiplex PCRs (mPCRs) for detection of DNA and RNA agents in the investigation of congenital infection and an mPCR for the viruses most commonly requested in a diagnostic virology laboratory (CMV, Epstein-Barr virus, enterovirus, HSV-1, HSV-2, and varicella-zoster virus). The assays were assessed using known pathogen-positive tissues (cultures, placentae, plasma, and amniotic fluid) and limits of detection were determined for all the agents studied using serial dilutions of plasmid targets. Nested PCR was performed as the most sensitive assay currently available, and detection of the amplicons using hybridization to labeled probes and enzyme-linked immunosorbent assay detection was incorporated into three of the four assays. This allowed detection of 10 to 10(2) copies of each agent in the samples processed. In several patients, an unexpected infection was diagnosed, including a case of encephalitis where HSV was the initial clinical suspicion but CMV was detected. In the majority of these cases the alternative agent could be confirmed using reference culture, serology, or fluorescence methods and was of relevance to clinical care of the patient. The methods described here provide useful techniques for diagnosing congenital infections and a paradigm for assessment of new multiplex PCRs for use in the diagnostic laboratory.

Integrons and gene cassettes in the enterobacteriaceae

Integrons were detected in 59 of 120 (49%) urinary isolates of Enterobacteriaceae by PCR using degenerate primers targeted to conserved regions of class 1, 2, and 3 integrase genes. PCR sequencing analysis of the cassette arrays revealed a predominance of cassettes that confer resistance to the aminoglycosides and trimethoprim.

Diagnosis of enteric pathogens in children with gastroenteritis

The aim of this study was to determine the isolation trends of common and emerging pathogens in children over a 12-month period. The study group included 412 children under 6 years with diarrhoea who were either hospitalised, or seen in the outpatients department of The Sydney Children's Hospital. Pathogens were detected in 137 (33%) samples, with rotavirus most common (40%), followed by adenovirus (26%), astrovirus (12%), Campylobacter jejuni (12%), Salmonella spp. (10%) and Giardia lamblia (< 1 %). Giardia-specific antigen (GSA) was detected in 11 of 382 (3%) using an enzyme immunoassay (EIA), and this included four samples in which cysts of G. lamblia were detected by microscopy. Using electron microscopy (EM), viruses were detected in 29 of 120 (24%) samples from hospitalised children and 53 of 171 (31%) outpatients (P = 0.23). Amongst this subset, Norwalk-like viruses (NLVs) were detected by RT-PCR in 10 samples including six of 14 with small round viruses, one of seven with small viral-like particles (SVLPs), and three of 126 EM-negative samples. Lactoferrin, detected by EIA, was 59% more likely to be positive in samples infected with salmonella/campylobacter than in samples in which bacterial pathogens were not isolated. As an indicator for infection with these bacterial agents, the assay showed a sensitivity and specificity of 95 and 40.3%, respectively. A routine microbiological analysis of stools from children of this age group should include a screen for foodborne bacterial agents and rotavirus. Tests for adenovirus, astrovirus and NLVs should be secondary. The cost-effectiveness of including the EIAs for lactoferrin and G. lamblia in the routine testing protocol needs to be evaluated.

Small intestinal mucosal immunity and morphometry in luminal overgrowth of indigenous gut flora

OBJECTIVE

The aim of this study was to investigate the separate effects of indigenous oropharyngeal- and colonic-type flora on small intestinal mucosal immunity and morphometry in small intestinal bacterial overgrowth (SIBO).

METHODS

A duodenal aspirate and random biopsies of underlying mucosa were obtained from 52 adult subjects (age range, 18-90 yr; median, 60 yr) without disorders that may otherwise disturb small intestinal histology or mucosal immunity. Villus height, crypt depth, villus/crypt ratios, counts of intraepithelial lymphocytes (IELs) and lamina propria total mononuclear cells, IgA, IgM, and IgG plasma cells, mast cells, and B and T lymphocytes were determined in relation to the presence or absence of SIBO and the nature of the overgrowth flora in all subjects. CD4+ve and CD8+ve T-cell counts were determined in 24 subjects.

RESULTS

SIBO was present in 26 of 52 (50%) subjects. Overgrowth flora included colonic-type bacteria in 20 subjects and oropharyngeal-type flora alone in 6 subjects. Lamina propria IgA plasma cell counts were significantly increased in subjects with SIBO, irrespective of whether the overgrowth flora comprised oropharyngeal-type flora alone or included colonic-type bacteria. Neither villus height, crypt depth, villus/crypt ratios, nor total or other mononuclear cell counts in lamina propria differed significantly between subjects with and without SIBO, irrespective of the nature of the overgrowth flora. IEL counts were significantly higher than in culture-negative subjects only when the overgrowth flora included colonic-type bacteria. Even then, IEL counts were within a range currently considered normal. A significant, inverse correlation between advancing age and IEL counts became apparent after adjusting for the effect of SIBO of colonic-type flora.

CONCLUSIONS

SIBO of oropharyngeal- and colonic-type flora are associated with differing disturbances of local duodenal mucosa. Nonetheless, these would not be readily apparent during routine histological assessment. Old age independently influences duodenal IEL counts.

Evaluation of the rice breath hydrogen test for small intestinal bacterial overgrowth

OBJECTIVES

The aims of this study were 1) to document the sensitivity, specificity, and predictive values of the rice breath hydrogen test for small intestinal bacterial overgrowth; 2) to determine the possible influence of concurrent gastric bacterial overgrowth and gastroduodenal pH on the efficacy of this test; and 3) to investigate whether reliability is limited by an inability of small intestinal luminal flora to ferment rice or its product of hydrolysis, maltose.

METHODS

Twenty adult subjects were investigated with microbiological culture of proximal small intestinal aspirate and a 3-g/kg rice breath hydrogen test. Gastroduodenal pH, the presence or absence of gastric bacterial overgrowth, and the in vitro capability of small intestinal luminal flora to ferment rice and maltose, its product of hydrolysis, were determined.

RESULTS

Sensitivity of the rice breath hydrogen test for small intestinal bacterial overgrowth was 33% and remained low even when subjects with small intestinal overgrowth with oropharyngeal-type (38%) and colonic-type flora (20%) and those with concurrent small intestinal and gastric bacterial overgrowth (40%) were considered separately. Sensitivity remained suboptimal despite favorable gastroduodenal luminal pH and documented ability of bacterial isolates to ferment rice and maltose in vitro. Specificity of the rice breath hydrogen test for small intestinal bacterial overgrowth was 91%. Positive predictive value, negative predictive value, and predictive accuracy were 75%, 63%, and 65%, respectively.

CONCLUSIONS

Clinical value of the rice breath hydrogen test for detecting small intestinal bacterial overgrowth is limited. The rice breath hydrogen test is not a suitable alternative to small intestinal intubation and culture of secretions for the detection of small intestinal bacterial overgrowth.

Characterisation of two new gene cassettes, aadA5 and dfrA17

Escherichia coli INS33 was isolated from the urinary tract of an infected patient. It was resistant to ampicillin, chloramphenicol, spectinomycin, streptomycin, sulfafurazole, tetracycline and trimethoprim. PCR screening revealed the presence of a class 1 integron that harboured two new gene cassettes, designated dfrA17 and aadA5. The new dfrA17 cassette was 91% identical to the known dfrA7 cassette. The aadA5 cassette was 95% identical over the first 830 bp to aadA4, but lacked the IS26 element found at the 3' end of this truncated cassette. Cloning and expression of the cassette region demonstrated that dfrA17 conferred high level resistance to trimethoprim but aadA5 conferred resistance to spectinomycin but not to streptomycin.

Detection of astrovirus gastroenteritis in children

A commercial enzyme immunoassay (EIA) for the detection of astrovirus antigen was used to detect the virus during a 12-month survey of enteric pathogens in children in outpatient (n = 238) and hospital (n = 176) settings. It was found to have a 100% sensitivity and 98.6% specificity. Nineteen astrovirus isolates were detected and confirmed by northern hybridization, cell culture, and RT-PCR. The virus was detected mainly amongst outpatients although a comparison of the detection rate with that in hospitalised children did not demonstrate a statistically significant difference (p = 0.1347). In contrast, there was a strong association between hospitalization and rotavirus infection (p = 0.0371), and a strong association between infection detected in outpatients and adenovirus infection (p = 0.0193). Strains of astrovirus were sequenced, genotyped and shown to be: type 1 (n = 11), type 3 (n = 1), and type 4 (n = 7). Maximum genetic variation in type 1 isolates was 8.6% and type 4 was 7.8%. Changes did not result in amino acid substitutions.

Serum immunoglobulin and soluble IL-2 receptor levels in small intestinal overgrowth with indigenous gut flora

Murine studies have demonstrated that the presence of indigenous gut flora is crucial for the induction of systemic immune hyporesponsiveness to antigens initially encountered within the gastrointestinal lumen. This study investigated whether increased titers of such flora, as occur in human small intestinal bacterial overgrowth, may be associated with increased suppression of systemic immune responsiveness and the possible relation between systemic and mucosal immunity in this setting. Serum total immunoglobulin (Ig), immunoglobulin subclass, and soluble interleukin-2 receptor levels and lamina propria IgA plasma cell counts were determined in 50 consecutive subjects with (N = 30) and without (N = 20) small intestinal bacterial overgrowth. Luminal IgA levels were measured in 35 of these subjects. Serum concentrations of IgG3, but not of other immunoglobulin isotypes or soluble interleukin-2 receptors, were significantly reduced in subjects with bacterial overgrowth (P < 0.0005). Small intestinal lamina propria IgA plasma cell counts (P < 0.0005) and luminal IgA concentrations (P = 0.001) were significantly increased in this group. Serum IgG3 levels were significantly inversely correlated with luminal IgA levels (P < 0.01) and fell below the lower limit of normal (0.41 g/liter) in 17/30 (56.7%) subjects with bacterial overgrowth compared to 1/20 (5.0%) subjects without (P < 0.0005). These findings document an association between small intestinal bacterial overgrowth with indigenous gut flora and reduced serum IgG3 reactivity in humans, possibly via an interaction with mucosa-related immunoregulatory mechanisms. The possibility of underlying small intestinal bacterial overgrowth should be considered in patients with serum IgG3 deficiency, especially those with compatible symptoms and/or known predisposition.

Interleukin-6 and small intestinal luminal immunoglobulins

Our aim was to determine the relationships between interleukin-6 and immunoglobulin levels within small intestinal luminal secretions. Twenty adult subjects with small intestinal bacterial overgrowth (N = 13), irritable bowel syndrome (N = 4), and nonulcer dyspepsia (N = 3) underwent endoscopic aspiration of secretions from the small intestinal mucosal surface for assessment of IL-6, IgA1, IgA2, IgM, IgG1, IgG2, IgG3, and IgG4 concentrations. Serum immunoglobulin concentrations and small intestinal histology were also determined. IgA2 and IgG3 were the predominant IgA and IgG subclasses in luminal secretions in 19/20 (95%) and 20/20 (100%) subjects, respectively. IgA1 and IgG1 predominated in serum in all subjects. No subject had villous atrophy. Luminal IL-6 concentrations correlated significantly with luminal IgA2, IgM, and IgG3 concentrations but not with IgA1 or any other IgG subclass levels. Conversely, luminal IL-6 or immunoglobulin concentrations did not correlate significantly with levels of any immunoglobulin isotype in serum. These observations suggest that important relationships exist between local IL-6 and IgA2, IgM, and IgG3 responses in human small intestinal luminal secretions. Local investigation is mandatory when assessing intestinal immune activity.

Liver damage in human small intestinal bacterial overgrowth

OBJECTIVE

Some rodent strains with experimental small intestinal bacterial overgrowth (SIBO) unrelated to jejunoileal bypass are susceptible to hepatic damage, possibly because of increased small intestinal permeability to proinflammatory bacterial polymers. However, data on the prevalence of hepatic damage in human subjects with SIBO in this setting are lacking. This study addressed this issue.

METHODS

Seventy adult subjects were investigated for possible SIBO and hepatic damage with bacteriological analysis of small intestinal aspirates and measurement of serum concentrations of alkaline phosphatase, gamma-glutamyl transpeptidase, aspartate aminotransferase, and alanine aminotransferase. Nutritional indices (serum albumin and anthropometry) and the urinary lactulose/mannitol ratio, an index of small intestinal permeability, were measured in all subjects with SIBO and liver damage.

RESULTS

SIBO was present in 40 of 70 subjects (57.1%). Overgrowth flora included salivary-type bacteria alone in 11 subjects and colonic-type bacteria in 29 subjects (facultative anaerobes [Enterobacteriaceae] alone in 21 subjects and both facultative and obligate anaerobes [Enterobacteriaceae and Bacteroides spp] in eight subjects). Biochemical evidence of liver damage was found in zero of 30 subjects without SIBO, zero of 11 subjects with SIBO with salivary-type bacteria alone, zero of 21 subjects with SIBO with facultative but not obligate anaerobic colonic-type bacteria, and in one of eight subjects (12.5%) with SIBO with obligate anaerobic colonic-type bacteria, in whom serum alkaline phosphatase and gamma-glutamyl transpeptidase levels were elevated. Nutritional indices were normal in this patient. Small intestinal permeability was increased and, along with liver enzyme abnormalities, normalized after eradication of SIBO. Small intestinal permeability was also increased in three of six patients (50.0%) with SIBO with obligate anaerobic colonic-type bacteria who had no evidence of liver damage.

CONCLUSIONS

SIBO per se is not a major risk factor for liver damage in humans, even when the overgrowth flora includes obligate anaerobes. Liver damage is not a necessary consequence of increased small intestinal permeability in this setting.

Luminal antigliadin antibodies in small intestinal bacterial overgrowth

OBJECTIVE

Elevated antigliadin antibody levels in small intestinal luminal secretions of subjects with normal or only mildly abnormal small intestinal histology are considered indicative of "latent" or "potential" celiac disease. The purpose of this study was to determine whether small intestinal bacterial overgrowth (SIBO) might provide an alternative explanation for positive luminal antigliadin antibodies in such subjects.

METHODS

Twenty-six adult subjects without predisposition to disturbed mucosal immunity were investigated with culture of small intestinal luminal secretions. Luminal total IgA and IgA-antigliadin antibody concentrations were measured by radial immunodiffusion and indirect enzyme immunoassay, respectively. Local mucosal counts of IgA-plasma cells were determined by immunohistochemistry. Small intestinal histology and intraepithelial lymphocyte counts were assessed by light microscopy. Corresponding serum antigliadin antibody concentrations were determined.

RESULTS

SIBO was present in 17/26 (65.4%) subjects. No subject with SIBO had villous atrophy. Luminal total IgA concentrations (p < 0.0005), mucosal IgA-plasma cell counts (p < 0.01), and intraepithelial lymphocyte counts (p < 0.01) were significantly increased in subjects with SIBO. Luminal IgA-antigliadin antibodies were detected in 6/17 (35.3%) subjects with SIBO and 0/9 (0%) subjects without SIBO. Luminal IgA-antigliadin antibody concentrations correlated significantly with luminal total IgA levels (p < 0.01) but not with serum values (p < 0.1). Serum IgG-antigliadin antibody concentrations were elevated in 2/6 (33.3%) subjects with SIBO and positive luminal antigliadin antibodies.

CONCLUSIONS

SIBO may be an alternative explanation to "latent" or "potential" celiac disease for positive luminal antigliadin antibodies in subjects with either normal or only mildly abnormal small intestinal histology, even when serum antigliadin antibody concentrations are elevated. Positive luminal antigliadin antibodies in SIBO probably occur as epiphenomena in the context of a graded mucosal immune response to local bacterial antigens.

Luminal bacteria and small-intestinal permeability

BACKGROUND

The influence of luminal bacteria on small-intestinal permeability has not been fully assessed. This study addressed this issue.

METHODS

Thirty-four subjects (mean age 64 years; range 22-95 years) were investigated for possible small-intestinal bacterial overgrowth (SIBO) with culture of a small-intestinal aspirate. A lactulose/mannitol small-intestinal permeability test was performed, small-intestinal histology assessed and serum vitamin B12 concentrations measured in all subjects. Permeability was also assessed in a control group of 34 asymptomatic volunteers.

RESULTS

Urinary lactulose/mannitol ratios were significantly increased in subjects with SIBO with colonic-type flora (P < 0.0005), even in the absence of villous atrophy. Urinary lactulose/mannitol ratios were increased in this group due to significantly increased urinary lactulose concentrations (P < 0.0005) rather than reduced urinary mannitol levels, after correcting for inter-subject variations in renal function. Counts of intraepithelial lymphocytes of CD8 phenotype were significantly increased in this group (P = 0.003). Although a significant correlation was found between intraepithelial lymphocyte counts and small-intestinal permeability overall (P < 0.002), these counts were not significantly different in subjects with SIBO with colonic-type flora whose permeability values were < or = > 0.028, the upper limit of normal in asymptomatic controls. Serum vitamin B12 concentrations did not differ significantly between groups (P > 0.5). Ageing did not independently influence small-intestinal permeability (P > 0.5).

CONCLUSIONS

Small-intestinal permeability is increased in subjects with SIBO with colonic-type bacteria. This effect is independent of ageing and not mediated by vitamin B12 deficiency. Although counts of intraepithelial lymphocytes of CD8 phenotype are increased in this disorder, it is also unlikely that these cells play an important causative role in this process. Routine light microscopic assessment underestimates the prevalence of small-intestinal functional disturbance in this disorder.

Local and systemic complement activity in small intestinal bacterial overgrowth

It is unknown whether bacteriolysis due to luminal complement activation contributes to local defense mechanisms against small intestinal bacterial overgrowth, particularly with gram-negative bacteria. This study addressed this issue. Thirty adult subjects were investigated with culture of luminal secretions adherent to proximal small intestinal mucosa. Luminal and plasma concentrations of C3 and C3d and C3d/C3 ratios were determined. Activated terminal complement complex was sought in surface epithelium to which aspirated secretions had been adherent. Small intestinal bacterial overgrowth with gram-negative bacteria was present in 12/30 (40.0%) subjects. C3, C3d, and C3d/C3 profile indicated that increased local but not systemic C3 activation occurs in this group. Conversely, no activation of terminal complement complex was evident in this circumstance. Thus, complement-mediated bacteriolysis is unlike to contribute to local defense mechanisms against small intestinal bacterial overgrowth, even when overgrowth flora includes gram-negative bacteria. Factors preventing full local activation of the complement cascade in this circumstance require investigation.

Longitudinal study of rotavirus infection in child-care centres

OBJECTIVE

To describe the epidemiology of symptomatic and asymptomatic rotavirus infection among young children attending Sydney child-care centres during the 1994 rotavirus season.

METHODS

Children aged 0-36 months in 11 child-care centres participated in the study. A weekly stool specimen was collected from each subject and tested for rotavirus antigen by commercial enzyme immunoassay.

RESULTS

One hundred and seventy-eight children (76 girls and 102 boys) with a mean age of 18.2 +/- 6.5 (SD) months were enrolled for a total of 2249 child-weeks. Of 1653 weekly faecal specimens, 59 (3.6%) were positive for rotavirus antigen. Positivity for rotavirus antigen peaked at 8.0% and 7.4% of specimens in weeks 6 and 7, respectively. The 59 positive specimens were obtained from 44 children in eight of the 11 study centres. One child appeared to suffer a second episode. Eighty-two per cent of episodes were associated with symptoms of gastroenteritis. Overall, 32% of the children in the eight affected centres were infected; 52% of those < 12 months were infected compared to 26% of older children. Secondary spread to household contacts was also documented.

CONCLUSIONS

Rotavirus infection poses a significant health problem in under-3-year-old children attending child-care centres in Sydney. We believe that these results are applicable to all Australian children of this age attending group child care.

Small intestinal bacterial overgrowth in the symptomatic elderly

OBJECTIVE

1) To determine the prevalence of small intestinal overgrowth with colonic-type bacteria in symptomatic elderly subjects, particularly those without important "clues" such as clinically apparent predisposition or vitamin B12 deficiency, and 2) to investigate defense mechanisms such as gastric acidity, small intestinal motility, and luminal IgA in this setting.

METHODS

Fifty-two symptomatic subjects without vitamin B12 deficiency or clinically apparent predisposition to bacterial overgrowth or disturbed mucosal immunity, including 22 subjects > or = 75 yr old, underwent culture of small intestinal luminal secretions. Indicator paper was used to measure fasting gastric pH. The presence of bacteria of confirmed nonsalivary origin in small intestinal secretions served as an index of small intestinal dysmotility. Small intestinal luminal IgA concentrations were measured by radial immunodiffusion.

RESULTS

Small intestinal overgrowth with colonic-type flora was not present in any subject investigated for dyspepsia, irrespective of age. In subjects with chronic diarrhea, anorexia, or nausea, overgrowth with colonic-type flora (Enterobacteriaceae) was present in 0/12 (0%), 1/10 (10.0%), and 9/14 (64.3%) subjects aged < 50 yr, 50-74 yr, and > or = 75 yr, respectively. Enterobacteriaceae were not concurrently recovered from saliva of any subject > or = 75 yr old with small intestinal overgrowth with these bacteria. Fasting hypochlorhydria was present in only 1/9 (11.1%) such subjects. Luminal IgA concentrations were significantly greater in subjects > or = 75 yr old with bacterial overgrowth than in culture-negative subjects (p < or = 0.003).

CONCLUSIONS

Small intestinal overgrowth with colonic-type bacterial should be considered in subjects > or = 75 yr old with chronic diarrhea, anorexia, or nausea, even in the absence of clues such as clinically apparent predisposition or vitamin B12 deficiency. Small intestinal dysmotility, rather than fasting hypochlorhydria or mucosal immunosenescence, probably is responsible for the prevalence of bacterial overgrowth in this group.

The expression of complement protein 4 and IgG3 in luminal secretions

BACKGROUND

Factors regulating proximal small-intestinal luminal concentrations of IgG3, the predominant IgG subclass at this site, are unclear. This study determined whether luminal IgG3 concentrations are related to those of complement protein 4 (C4), an acute-phase reactant predominantly derived from local mucosa.

METHODS

Proximal small-intestinal luminal and peripheral blood IgG subclass and C4 concentrations were measured by radial immunodiffusion in 30 adult subjects without predisposition to disturbed mucosal immunity. Mucosal C4 immunoreactivity and the presence or absence of small-intestinal bacterial overgrowth were determined in all subjects. Caecal luminal concentrations of IgG3 and C4 were measured in a separate cohort of eight asymptomatic subjects.

RESULTS

Proximal small-intestinal luminal C4 and IgG subclass concentrations were not significantly influenced by the presence of absence of small-intestinal bacterial overgrowth (P > 0.2). Nor did plasma C4 levels significantly influence C4 concentrations in small-intestinal luminal secretions (P > 0.2). Mucosal immunoreactivity for C4 was present in every subject. A significant correlation was found between C4 and IgG3 concentrations in proximal small-intestinal luminal secretions (P < 0.0005) and also in caecal secretions (P < 0.05) but not in peripheral blood (P > 0.1).

CONCLUSIONS

Common factors, not including the presence or absence of small-intestinal bacterial overgrowth, regulate luminal concentrations of C4 and IgG3. Local investigation is mandatory when assessing mucosal immune mechanisms.

Bacteriological method for detecting small intestinal hypomotility

OBJECTIVE

Small intestinal hypomotility is an important cause of small intestinal bacterial overgrowth, yet assessment of small intestinal motility in this setting is problematic. This study was performed to investigate the validity of a bacteriological method for detecting small intestinal hypomotility.

METHODS

Twenty-five subjects without previous gastric surgery were studied with (i) concurrent bacteriological analyses of fasting saliva and gastric and proximal small intestinal aspirates, (ii) measurement of gastric pH, and (iii) scintigraphic assessment of small intestinal transit rates of a liquid test meal. The reproducibility of bacteriological analyses of saliva and small intestinal secretions was determined in 12 subjects.

RESULTS

Serial bacteriological analyses of saliva and proximal small intestinal secretions yielded reproducible results over time periods of up to 1 month. Eleven subjects were deemed to harbor Enterobacteriaceae of nonsalivary origin in proximal small intestinal secretions. Orocaecal transit, but not gastric emptying, of a liquid test meal was significantly delayed in this group (p = 0.002 and p = 0.84, respectively), suggesting the presence of small intestinal hypomotility. Impaired gastric acidity unlikely confounded assessment of the origin of small intestinal Enterobacteriaceae in any instance.

CONCLUSIONS

The presence of Enterobacteriaceae of nonsalivary origin in proximal small intestinal secretions may be taken to reflect the presence of small intestinal hypomotility. The presence of impaired gastric acidity does not confound this approach. Because small intestinal intubation and culture of aspirate are required anyway to accurately diagnose small intestinal bacterial overgrowth, the simple addition of concurrent bacteriological analysis of saliva may allow small intestinal hypomotility to be detected at the same time as the presence or absence of small intestinal bacterial overgrowth itself is established, thus streamlining the investigation of subjects for this disorder and its possible causes.

Luminal immunity in small-intestinal bacterial overgrowth and old age

BACKGROUND

The independent influences of small-intestinal bacterial overgrowth and old age on mucosal immunoglobulin production and secretion have not been assessed. This is an important issue, since luminal IgA deficiency may exacerbate small-intestinal bacterial overgrowth, the prevalence of which is high in selected elderly populations.

METHODS

Proximal small-intestinal aspirates were obtained from 33 subjects for bacteriologic analysis and measurement of total IgA, IgM, total IgG. IgG subclass, and IgD concentrations. IgA subclasses were measured in 24 unselected subjects. Serum immunoglobulin and salivary IgA concentrations were measured in all subjects.

RESULTS

IgA2 and IgG3 were predominant IgA and IgG subclasses in proximal small-intestinal luminal secretions. Luminal concentrations of IgA2 and IgM, but not IgG3 or any other IgG subclass, were significantly increased in small-intestinal bacterial overgrowth, which was present in 19 of 33 (57.6%) subjects. Old age did not influence these levels. Luminal immunoglobulin concentrations did not correlate significantly with either serum or salivary values. IgD was not measureable in proximal small-intestinal secretions.

CONCLUSIONS

Increased luminal concentrations of the secretory immunoglobulins, IgA2 and IgM, occur in small-intestinal bacterial overgrowth. Local investigation is mandatory when assessing the mucosal immunopathology of this disorder. Luminal IgG3 is unlikely to be predominantly derived from serum. Old age does not independently influence luminal immunity.

Mucosal cytokine production in small-intestinal bacterial overgrowth

BACKGROUND

Mucosal production of interferon-gamma, interleukin-6, and tumour necrosis factor-alpha is increased in inflammatory bowel disease and parallels disease activity. Interferon-gamma production is also increased in coeliac disease. Conversely, local cytokine profiles have not been investigated in small-intestinal bacterial overgrowth. This study addressed this issue.

METHODS

Eighteen adult subjects were studies with culture of proximal small-intestinal luminal secretion and measurement of luminal interferon-gamma, interleukin-6, and tumour necrosis factor-alpha concentrations by enzyme-linked immunosorbent assay. Small-intestinal histology was assessed by light microscopy.

RESULTS

Interferon-gamma, interleukin-6, and tumour necrosis factor-alpha were measurable in proximal small-intestinal luminal secretions of all subjects, even in the absence of light microscopic evidence of enteropathy. Small-intestinal bacterial overgrowth was present in 12 of 18 (66.7%) subjects. Luminal concentrations of neither interferon-gamma nor tumour necrosis factor-alpha differed significantly in subjects with and without small-intestinal bacterial overgrowth (P + 0.06 and P = 1.0, respectively). Conversely, luminal interleukin-6 concentrations were significantly increased in subjects with this disorder (P = 0.02). Multivariate linear regression analysis suggested that colonic-type rather than salivary-type flora mediated this increased interleukin-6 response (P = 0.02 and P = 0.64, respectively). No correlation was found between luminal interleukin-6 and tumour necrosis factor-alpha concentrations, even after the confounding influence of colonic-type bacteria was excluded (P = 0.60).

CONCLUSIONS

These findings suggest that increased mucosal production of interleukin-6 occurs in small-intestinal bacterial overgrowth, particularly when the overgrowth flora includes colonic-type bacteria. Conversely, luminal levels of neither interferon-gamma nor tumour necrosis factor-alpha are increased in the circumstance, distinguishing the local cytokine profile in this disorder from those that occur in coeliac disease and inflammatory bowel disease.

The lactulose breath hydrogen test and small intestinal bacterial overgrowth

OBJECTIVES

To i) document the sensitivity and specificity of a combined scintigraphic/lactulose breath hydrogen test for small intestinal bacterial overgrowth and ii) investigate the validity of currently accepted definitions of an abnormal lactulose breath hydrogen test based on "double peaks" in breath hydrogen concentrations.

METHODS

Twenty-eight subjects were investigated with culture of proximal small intestinal aspirate and a 10-g lactulose breath hydrogen test combined with scintigraphy. Gastroduodenal pH, the presence or absence of gastric bacterial overgrowth, and the in vitro capability of overgrowth flora to ferment lactulose were determined.

RESULTS

Sensitivity (16.7%) and specificity (70.0%) of the lactulose breath hydrogen test alone for small intestinal bacterial overgrowth were poor. Combination with scintigraphy resulted in 100% specificity, because double peaks in serial breath hydrogen concentrations may occur as a result of lactulose fermentation by cecal bacteria. Sensitivity increased to 38.9% with scintigraphy, because a single rise in breath hydrogen concentrations, commencing before the test meal reaches the cecum, may occur in this disorder. Sensitivity remained suboptimal irrespective of the definition of small intestinal bacterial overgrowth used, the nature of the overgrowth flora, favorable luminal pH, the presence of concurrent gastric bacterial overgrowth, or the in vitro ability of the overgrowth flora to ferment lactulose.

CONCLUSIONS

Definitions of an abnormal lactulose breath hydrogen test based on the occurrence of double peaks in breath hydrogen concentrations are inappropriate. Not even the addition of scintigraphy renders this test a clinically useful alternative to culture of aspirate for diagnosing small intestinal bacterial overgrowth.

Factors influencing the 1-g 14C-D-xylose breath test for bacterial overgrowth

OBJECTIVES

To document the sensitivity of the 1-g 14C-D-xylose breath test for bacterial overgrowth and to investigate luminal and nonluminal factors that may influence breath 14CO2 levels and impact on the clinical utility of this test.

METHODS

Thirty-five adult subjects were investigated for bacterial overgrowth by culture of gastric and small intestinal aspirates and by a 1-g 14C-D-xylose breath test. Body weight, gastroduodenal pH and the in vitro capability of overgrowth flora to ferment D-xylose were assessed. Serial breath 14CO2 levels were also recorded before and after the resolution of malabsorption in a subject with celiac disease to determine the importance of postabsorptive metabolism of this substrate.

RESULTS

Gastric and small intestinal bacterial overgrowth were present in 19/35 (54.3%) and 21/35 (60.0%) subjects, respectively. The positivity rate of culture of aspirate exceeded that of the 1-g 14C-D-xylose breath test. Endogenous CO2 production independently influenced breath 14CO2 levels. After excluding this influence, sensitivity of the 1-g 14C-D-xylose breath test for gastric bacterial overgrowth or small intestinal bacterial overgrowth was poor, even when overgrowth with specific "marker organisms" was considered. Poor sensitivity could not be explained by unfavorable luminal pH. Overgrowth flora were proven capable of in vitro D-xylose fermentation in 81.8% of subjects. Systemic and/or colonic metabolism of 1-g 14C-D-xylose appear to be important factors influencing results of the 1-g 14C-D-xylose breath test, especially in partial gastrectomy subjects.

CONCLUSIONS

The 1-g 14C-D-xylose breath test is not a suitable alternative to culture of aspirate for the investigation of subjects for bacterial overgrowth.

Beta-D-Glucuronidase activity among prototrophic and auxotrophic variants of Escherichia coli and other Enterobacteriaceae commonly implicated in urinary tract infections

Glucuronidase (GUD) activity of 102 prototrophic, 91 cysteine-requiring, and 19 thymidine-requiring strains of Escherichia coli was examined using growth from MacConkey, CLED, and enriched brain heart infusion (BHI) agars. After 24 h incubation, GUD activity was detected in 92%-96% of prototrophic strains and a similar proportion of thymidine-requiring strains with most reactions detectable in shorter incubation periods. GUD activity among strains requiring cysteine was significantly less than that found amongst prototrophic strains. The effects of different sources of inocula were evident in the shorter incubation periods. Other strains of the Enterobacteriaceae and oxidative strains frequently implicated in urinary tract infection were also tested. Here, positive reactions were detected among Citrobacter and Enterobacter spp. and a strain of Klebsiella oxytoca, but only after 24 h incubation. GUD activity was not detected among the oxidative strains tested under the same conditions. Although an incubation time of 24 h is necessary to detect activity in a small number of "slow hydrolyzing" E. coli, the increased sensitivity thus attained compromises the specificity of the test for this organism by simultaneously enhancing detection of the enzyme in other enterobacteria.

Bacteriologic analysis of mucosal biopsy specimens for detecting small-intestinal bacterial overgrowth

BACKGROUND

Although culture of luminal secretions is regarded as the most accurate diagnostic test for small-intestinal bacterial overgrowth, obtaining an aspirate is often difficult owing to the sparseness of luminal secretions present at the time of aspiration. Obtaining a mucosal biopsy specimen for bacteriologic analysis would overcome this problem.

METHODS

Culture of small-intestinal and gastric aspirates and unwashed small-intestinal mucosal specimens was performed in 51 adult subjects investigated for small-intestinal overgrowth.

RESULTS

Highly significant (r = 0.85-0.90; p < 0.0005) correlations were found between viable bacterial counts in small-intestinal luminal secretions and biopsy specimens. Small-intestinal bacterial overgrowth was present in 60.8% of subjects. When specimens weighing 4.0-84.0 mg were suspended in diluent, total aerobic and/or anaerobic bacterial counts > or = 10(2) CFU/ml were found to have 90.3% sensitivity and 100% specificity for small-intestinal bacterial overgrowth.

CONCLUSION

Culture of an unwashed small-intestinal mucosal biopsy specimen is a useful alternative to culture of a small-intestinal aspirate for detecting subjects with small-intestinal bacterial overgrowth, especially when luminal secretions are scanty at the time of aspiration.

An appraisal of a 'string test' for the detection of small bowel bacterial overgrowth

The efficacy of a string test for the detection of small bowel bacterial overgrowth (SBBO) was determined by comparison with a sterile endoscopic method for sampling small bowel secretions in 15 subjects investigated for SBBO. Clinical value was found to be limited by poor sensitivity, specificity and positive predictive value. The string test is not an adequate substitute for oro-duodenal intubation for the detection of SBBO.

Fasting breath hydrogen concentrations in gastric and small-intestinal bacterial overgrowth

BACKGROUND

Although elevated fasting breath hydrogen concentrations have been reported in small-intestinal bacterial overgrowth, this diagnosis has been presumptive or based on definitions that vary from study to study. The influence of gastric bacterial overgrowth and gastroduodenal pH has not been documented. Conflicting evidence exists as to the reproducibility of breath hydrogen measurements.

METHODS

Forty-two subjects underwent culture of gastric and duodenal aspirates. The pH was measured by indicator paper. Paired fasting breath hydrogen concentrations were measured by gas chromatography within 7 days of endoscopy.

RESULTS

Paired fasting breath hydrogen concentrations differed in terms of normality or abnormality in 21% of subjects. Paired concentrations correlated significantly in overgrowth but not in culture-negative subjects. Sensitivity for bacterial overgrowth was 4-29%, and specificity 71-100%. No correlation with gastroduodenal pH was found.

CONCLUSIONS

The clinical relevance of a single fasting breath hydrogen concentration is limited. The efficacy of paired measurements for gastric or small-intestinal bacterial overgrowth is poor.

Virulence and other phenotypic characteristics of urinary isolates of cysteine-requiring Escherichia coli

Urinary isolates of cysteine-requiring Escherichia coli were found to be generally lacking in virulence factors commonly associated with uropathogenic strains. The proportion of auxotrophic strains showing type-1 fimbriation, haemolysin production, motility and sensitivity to normal human serum was significantly less than that of a comparable number of urinary isolates of prototrophic E. coli, although the proportion in both groups possessing K1 antigen was similar. Furthermore, the biotyping and serogrouping of these and other strains from systemic infections demonstrated a high degree of phenotypic diversity. This is further evidence that infection with these auxotrophs results from a combination of decreased host resistance and a physiological condition conducive to the random selection of these auxotrophs in vivo.

Further studies of clinical isolates of cysteine-requiring Escherichia coli and Klebsiella and possible mechanisms for their selection in vivo

Cysteine-dependent (cys-) Escherichia coli and Klebsiella spp., defective in sulphate assimilation, were isolated from urine and stool samples of infected patients. These isolates reverted to prototrophy under conditions of cysteine deprivation but the revertant strains and a prototrophic wild-type E. coli strain became auxotrophic for cysteine in a cysteine-enriched medium. This suggested that excess cysteine acts as a repressor of the cys HIJ operon known to control aspects of cysteine biosynthesis. A group of mostly elderly patients infected with cys- strains suffered a disproportionate amount of renal impairment as compared with a control group. In renal impairment, sulphur compounds, including cysteine, are retained. This raises the possibility that these raised levels of cysteine and related compounds may enhance the selection of cys- strains in vivo.

Study of growth requirements other than cysteine of naturally occurring Escherichia coli and Klebsiella spp. auxotrophic for cysteine

Cysteine remains the preferred supplement for cultivation of Cys- auxotrophs in vitro. Methionine, which reduced cysteine requirements, and branched-chain amino acids, which decreased cysteine toxicity, were identified as the components of casein hydrolysate responsible for growth enhancement by this additive. Glutathione and DL-homocysteine can be substituted for cysteine. Accumulation of these compounds in patients with renal impairment may favor selection of Cys- strains in vivo.

In vitro susceptibilities of clinical isolates of cysteine-requiring Escherichia coli to 12 antimicrobial agents

The MICs of 12 antimicrobial agents for 42 cysteine-requiring strains of Escherichia coli showed a high concordance when determined on three different media, one of which was supplemented with cysteine. Differences in the MICs of several agents were detected between 18 prototrophic revertants and their parent auxotrophs. A total of 64.7% of the isolates were fully susceptible to all agents, and no particular resistance pattern was evident.

Assessment of conventional and commercial methods for identification of clinical isolates of cysteine-requiring strains of Escherichia coli and Klebsiella species

Cysteine-requiring strains of the family Enterobacteriaceae that are auxotrophic for this amino acid because of defects in the sulfur assimilatory pathway account for about 1.5% of urinary tract isolates of Escherichia coli and Klebsiella species. Forty Escherichia and eight Klebsiella cysteine-requiring strains were used to test the ease with which various test systems identified clinical isolates of cysteine auxotrophs. In a preliminary experiment, the growth yield of 10 cysteine-requiring E. coli in 10 solutions of commercially available peptones was in each instance less than that of prototrophic control and showed that these sources of nutrients were suboptimal for these strains. A significant proportion of the cysteine-requiring strains were not adequately identified by growth-dependent tests which used various peptones as a nutrient source. Problems were encountered with all test systems examined, which were as follows: conventional methods; the API 20E, Microbact, and Vitek systems; and two rapid methods for the identification of E. coli, the Rapidec coli and the beta-D-glucuronidase tests. The performance of the test systems was only partly improved when inocula were derived from appropriately supplemented media. However, the problems of the growth-dependent tests were resolved when a cysteine-supplemented suspension was used to inoculate each test system.

Characteristics of cysteine-requiring strains of Klebsiella isolated from urinary tract infections

Clinical and bacteriological findings in seven cases of urinary tract infection with cysteine-requiring strains of Klebsiella are described. The organisms were isolated from patients with long-standing urinary tract abnormalities and grew as small (c. 1 mm) colonies on MacConkey agar. The organisms failed to grow in a minimal medium supplemented with sodium sulphate but grew when the medium was supplemented with cysteine sulphinic acid, sodium sulphide or L-cysteine. The smallest amount of cysteine required for optimal growth in a chemically defined medium was 20 mg/L. Cysteine-requiring strains of Escherichia coli had previously been shown to require a similar amount of cysteine and to be unable to reduce sulphate to sulphite; this suggests a common influence in the selection of cysteine auxotrophs in vivo. However, the amino acid inhibited the growth of E. coli at concentrations which only slightly altered growth of the Klebsiella strains. Problems with the isolation, identification and sensitivity testing of cysteine-requiring Klebsiella were also observed and methods by which these may be minimised are suggested.

Cysteine requirements of naturally occurring cysteine auxotrophs of Escherichia coli

The requirements for cysteine of naturally occurring cysteine auxotrophs of Escherichia coli were determined in a defined liquid medium. Maximal growth was obtained in the presence of cysteine concentrations between 20 and 250 mg/l. At concentrations below 20 mg/l growth of the auxotrophs, but not the prototrophic control, was suboptimal in this system. In the presence of cysteine concentrations in excess of 250 mg/l, growth of both auxotrophic and prototrophic E. coli was inhibited with lower growth yields, a decreased specific growth rate and an extended lag phase being observed. These effects were minimised in the presence of 2 mM L-leucine, L-isoleucine and L-valine.

Septicaemia caused by cysteine-requiring isolates of Escherichia coli

The clinical and bacteriological findings in five cases of septicaemia with cysteine-requiring isolates of Escherichia coli are reported. Infections with these nutritionally-dependent organisms have been found previously in the urinary tract only, associated usually with chronic rather than acute conditions. The urinary tract was considered to be the source of the septicaemia in our patients and that site should be investigated when such strains are isolated from blood cultures. When first isolated the organisms characteristically form small translucent colonies on media deficient in appropriate growth factors. Their nutritional requirement for cysteine can be determined by a simple auxanographic technique, thereby enabling the appropriate supplementation of media necessary for reliable identification and antibiotic-sensitivity testing.