Intestinal Spirochetosis and Chronic Watery Diarrhea

Blackwell Publishing AsiaMelbourne, AustraliaJGHJournal of Gastroenterology and Hepatology0815 93192006 Blackwell Publishing Asia Pty Ltd20062113261333Original Article Intestinal spirochetosis and diarrheaM Esteve et al.

doi:10.1111/j.1440-1746.2006.04150.x

GASTROENTEROLOGY Intestinal spirochetosis and chronic watery diarrhea: Clinical and histological response to treatment and long-term follow up Maria Esteve,* Antonio Salas,† Fernando Fernández-Bañares,* Josep Lloreta,‡ Meritxell Mariné,§ Clara Isabel Gonzalez,† Montserrat Forné,* Jaume Casalots,† Rebeca Santaolalla,* Jorge Carlos Espinós,* Mohammed Arif Munshi,¶ David John Hampson¶ and Josep Maria Viver*

Departments of *Gastroenterology, †Pathology and §Internal Medicine, Mutua Terrassa Hospital, University of Barcelona, Terrassa, ‡Department of Pathology, Hospital del Mar, Barcelona, Catalonia, Spain and ¶School of Veterinary and Biomedical Sciences, Murdoch University, Murdoch, Western Australia, Australia

Key words Abstract chronic watery diarrhea, intestinal spirochetosis, metronidazole, microscopic Background: The clinical significance of intestinal spirochetosis is uncertain, therefore colitis, penicillin. the aim of the present paper was to assess the prevalence of histological intestinal spirochetosis in patients with and without chronic watery diarrhea and to evaluate its clinical Accepted for publication 2 July 2005. relevance. Methods: A prospective diagnostic work-up of intestinal spirochetosis was made on Correspondence biopsy samples taken from patients with chronic watery diarrhea submitted between 1994 Dr M Esteve, Department of Gastroenterology, and 2004 (1174 colonoscopies with multiple biopsies). Three other positive cases identified Hospital Universitari Mútua de Terrassa, from routine endoscopic biopsies also were reviewed. In addition, samples from 100 Plaça Dr Robert n° 5, 08221 Terrassa, asymptomatic control patients and a random sample of another 104 colonic specimens Barcelona, Catalonia, Spain. were reviewed for intestinal spirochetosis. The diagnosis was established by light and Email: [email protected] electron microscopy. Polymerase chain reaction (PCR) amplification of the 16S ribosomal RNA and reduced nicotinamide adenine dinucleotide (NADH) oxidase genes of the intestinal spirochetes Brachyspira aalborgi and Brachyspira pilosicoli was performed on tissue biopsies of the 11 positive patients. After diagnosis, treatment with penicillin benzatine (PB) or metronidazole was offered to all symptomatic patients and they were followed for a mean of 45.4 months (range: 37–113 months). Results: Eight patients with chronic watery diarrhea were positive for intestinal spirochetosis. Intestinal spirochetosis was not diagnosed in the controls. Histological resolution of the infection paralleled clinical recovery in six patients (following metronidazole treatment in three). Most patients showed mild, non-specific colonic inflammation. Invasion by the spirochetes was not demonstrated by electron microscopy. Brachyspira aalborgi and B. pilosicoli each were identified by PCR in two cases. Conclusions: Histological intestinal spirochetosis appears to be relatively uncommon in Catalonia (Spain) compared to previous reports from other countries, but was identified in patients (0.7%) with chronic watery diarrhea. Sustained clinical recovery after spontaneous or drug-induced spirochetal disappearance in these individuals suggests that intestinal spirochetosis may play a pathogenic role in chronic watery diarrhea. Treatment with metronidazole is advisable in patients with persistent symptoms.

4–9 Introduction mainly consisting of watery diarrhea. In addition, epidemiolog- ical studies have shown that chronic diarrhea or wet feces were The clinical relevance of intestinal spirochetosis (IS) is controver- signiﬁcantly associated with colonization by intestinal spirochetes sial.1–3 The occurrence of spirochetes attached to the luminal sur- in Australian Aboriginal children,10 and in people on the island of face of the colorectal mucosa has been documented in various case Bali.11 There also have been studies in which intestinal spirochete reports and studies of series of patients with intestinal complaints, strains isolated from humans have been used to infect and cause

M Esteve et al. Intestinal spirochetosis and diarrhea intestinal disease in chickens and pigs.12,13 In contrast, in studies Colonic surgery was performed due to colonic neoplasms (50 where colonic specimens or feces of a large series of patients were cases), diverticular disease (49 cases), irradiation colitis (1 case), assessed, IS was a frequent finding, ranging from 1.9 to Hirschsprung disease (one case) and inflammatory bowel disease 50%,3,10,11,14–17 being particularly common in developing commu- (three cases). nities.3,10,11,18,19 In some studies symptoms were not modified by IS eradication,14–17 leading to the suggestion that intestinal spiro- Biopsy handling and pathological diagnosis chetes are harmless commensals in humans, and thus specific treatment is not required. This apparent discrepancy in outcomes The biopsy specimens were immediately fixed in 4% phosphate- might be due either to a variable pathogenicity of different spiro- buffered formaldehyde, pH 7.0 and embedded in paraffin. Sections chete species or strain,20,21 or the immunological status of the were stained with hematoxylin and eosin (HE). The diagnosis of infected individuals.22 At least two species of spirochetes, Brach- IS was established by light microscopy and was confirmed by yspira pilosicoli23 and Brachyspira aalborgi,24 have been associ- periodic acid–Schiff stain (PAS) and Steiner silver32 stains. ated with IS in humans, and B. pilosicoli has been isolated from Because it has been suggested that up to half of IS cases may be the blood of critically ill patients.25 In addition, electron micro- overlooked on HE,17,33 PAS and Steiner silver stains were per- scopic analysis of specimens from some infected individuals has formed in all of the control biopsies in spite of a normal appear- shown spirochetal invasion of colonic epithelial cells, macroph- ance on HE. ages, goblet cells and Schwann cells.3,6,26–28 Biopsy specimens from four patients with IS were examined by The aims of the present study were (i) to assess the histological electron microscopy before and after treatment. Biopsies were prevalence of IS in adult patients attending a tertiary hospital in fixed in 2.5% cacodylate-buffered glutaraldehyde, pH 7.4 and Catalonia (Spain); (ii) to evaluate the clinical relevance of IS, embedded in epoxy resin. Two blocks from each case were pro- mostly diagnosed in the setting of a prospective evaluation of cessed and five ultra-thin sections from each block (10 grids per chronic watery diarrhea; and (iii) to determine the spirochete spe- case) were performed. Three additional colonic samples from IS cies present in IS cases. patients were recovered from formaldehyde-fixed archival tissue, and were deparaffinized and processed for electron microscopic Methods examination. Sections approximately 100 nm thick were cut with a diamond knife and stained with uranyl acetate and lead citrate, and examined under a Philips CM100 transmission electron Patients and controls microscope (Philips, Eindhoven, the Netherlands) On light microscopy, the following histological findings were Patients with chronic watery diarrhea recorded: (i) inflammatory infiltrate in the lamina propria; and (ii) A prospective diagnostic work-up was made of biopsies taken epithelial damage: flattening and detachment of the surface epite- from patients with chronic watery diarrhea but a normal appearing lium, crypt distortion with polymorphonuclear infiltration and mucosa. Procedures followed those established in the Gastroenter- increased number of intraepithelial lymphocytes. The inflamma- ology Department of Mutua Terrassa Hospital, Barcelona, which tory infiltrate was subjectively assessed. Care was taken to record have been previously described in detail.29–31 Briefly, total colonos- the situation when there was an even distribution of cells (lympho- copy was undertaken and 12 colonic biopsies of five areas of the cytes, plasma cells and eosinophils) extending into the deeper colon, from the cecum to the rectum, were taken during the course portion of the lamina propria, as opposed to the usual situation in of 1174 colonoscopies (one colonoscopy per patient) performed the normal mucosa, where cells are concentrated in the upper one- between 1 January 1994 and 31 December 2004. third of the lamina propria. On electron microscopy, IS was confirmed and the following features were recorded: (i) abnormalities in epithelial cell surface; Samples from routine biopsies (ii) intraepithelial spirochetal fragments; and (iii) presence of spi- During the same period, IS had been identified in biopsy samples rochetal remnants in the phagolysosomes of macrophages. from three patients without chronic diarrhea, and these cases also were reviewed. Identifying spirochete species DNA from paraffin-embedded tissue (PET) samples from the IS Controls patients was extracted using a modification of a previously One hundred patients attending the Endoscopic Unit due to rectal described method, with samples from positive and negative con- bleeding secondary to anorectal pathology or having a previous trol patients included in each analysis.24,34 Several sections history of colonic polyps were included as controls (54 men and approximately 10 µm thick were cut from the PET samples. Each 46 women; mean age 48.0 years; range 26–74 years). All these sample was dewaxed by adding 400 µL of xylene and placing on patients had either normal bowel habit or constipation. After a rocker at room temperature until the paraffin dissolved. A total informed consent was obtained, multiple biopsies were taken of 400 µL of 100% ethanol was then added, and the samples from five areas of the colon, following the same previously were centrifuged at 10 000 × g for 10 min, and the supernatant described protocol. In addition, a random sample of 104 colonic discarded. The samples were dried at 50°C in an oven for at least specimens of patients (67 men and 37 women; mean age 67.28 30 min. Twenty micrograms of Proteinase K (Boehringer Man- years; range 4–98 years) operated on within the study period were nheim, Mannheim, Germany) in 200 µL of 50 mmol/L Tris-HCl, selected and the slides were reviewed for IS identification. pH 8.3, was added and incubated for 1–2 h at 55°C. The samples

Intestinal spirochetosis and diarrhea M Esteve et al. were then boiled for 8 min, and 2 µL of each resultant extract Patients with chronic watery diarrhea was used as template DNA for polymerase chain reaction (PCR) Intestinal spirochetosis was identified in biopsies from eight of the analysis. 1174 patients with chronic watery diarrhea (0.7%). This preva- Specific PCR procedures amplifying portions of the 16S riboso- lence was not statistically different from the prevalence in the mal RNA (16S rRNA) and reduced nicotinamide adenine dinucle- control group (P = 0.6133). Seven of the IS patients were male. otide (NADH) oxidase (nox) genes of B. aalborgi and B. pilosicoli The mean age of the eight patients was 47 years (range 29– were applied to DNA extracted from these tissue biopsies.24,34 The 72 years). Six of these IS cases (cases 1–5, 7) were diagnosed PCR products were subjected to electrophoresis in 1.5% (wt/vol) between January and October 2000. Clinical, endoscopic and his- agarose gels in 1× Tris-acetate buffer, stained with ethidium bro- topathological findings and follow-up data for the patients (cases mide and viewed over ultraviolet (UV) light. 1–8) are summarized in Table 1. One of these patients had been diagnosed with acquired immunodeficiency syndrome (AIDS) Clinical assessment of the IS patients and 4 years before IS diagnosis, and was admitted to the hospital due follow up to a reactivation of visceral Leishmaniasis (case 6). Except for this patient, the hematological and biochemical parameters of all Following diagnosis, a prospective clinical evaluation, treatment patients were normal. The AIDS patient, and one other patient and follow up were offered to all IS patients. Sources of infection (case 3) were homosexual. The mean number of bowel movements were prospectively investigated: sexual habits, employment, travel by the patients per day was 4.8 (range 3–8). The mean duration of to endemic areas and possible ingestion of non-potable water. diarrhea at the time of diagnosis was 12 months (range 1– Evaluation of patients at diagnosis included anamnesis, asking 48 months) and, with the exception of the AIDS patient, it was about drug consumption (non-steroidal anti-inflammatory drugs, well-tolerated, and not of major clinical concern. antibiotics, etc.), symptoms (fever, weight loss, abdominal pain, diarrhea, number of bowel movements, duration of diarrhea) and physical examination. In addition, bacterial culture for pathogenic Cases identified in routine biopsy examination aerobic bacteria and fecal examination for ova and parasites Intestinal spirochetosis was identified in biopsies taken from three (assessed by stool concentration in samples taken on 3 consecutive other patients with non-specific ulceration of the ileocecal valve, days), routine blood analysis and examination for antibody to ischemic ulcers, or colonic polyps (cases 9,10 and 11, respec- + human immunodeficiency virus (HIV; 1 2) were performed. tively: one female and two male; Table 2). Patient 11 worked in a When informed consent was obtained, symptomatic patients were pig and chicken slaughterhouse, where he drank non-potable treated with penicillin benzathine 2.4 MU i.m. in a single dose as water. a first option or oral metronidazole 500 mg q.i.d. for 10 days as a second option, when persistence of IS was histologically demonstrated. The asymptomatic patients were simply followed up. A Other potential pathogens colonoscopy with multiple biopsies was performed, to evaluate In all 11 IS cases, fecal bacterial cultures for pathogens and exam- healing of the disease, at least 1 month after the end of treatment ination for ova and parasites were negative. or the disappearance of symptoms. After cure, patients were vis- ited every 3 months during the follow up for assessment of treat- Histopathological findings ment response. Multiple biopsies were available from 10 patients. In six patients a homogeneous distribution of IS was observed throughout the Data analysis colon. By light microscopy, the presence of a layer of spirochetes The prevalence rates for IS in patients with chronic watery diar- attached to the colonic epithelium on HE, PAS and Steiner silver rhea and in the controls were compared using Fisher’s exact test. stains was observed in all cases (Fig. 1). All except three patients Other results were expressed as mean and range. showed increased inflammatory infiltrate of the lamina propria, which was mild in most cases. Histological findings are detailed in Tables 1 and 2. Ethical considerations Electron microscopy confirmed the presence of a dense band of The study was performed according to the 1995 Declaration of spirochetes adherent to the colonic surface (Fig. 2). Mild degener- Helsinki ethical guidelines, and was approved by the research and ative changes of the epithelium, consisting of stunting and frag- ethical committees of Mutua Terrassa Hospital, Barcelona. mentation of microvilli, were detected in one case. Electron microscopy was also performed on tissue retrieved from the paraf- Results fin blocks of the two patients with ulcers in the colonic mucosa. No invasive spirochetes were detected in the enterocytes, goblet cells, macrophages or Schwann cells in any case. Controls Histopathological follow up was available for eight of the 11 IS Intestinal spirochetosis was not identified in any of the control patients (four by electron microscopy). After IS disappearance endoscopic or surgical biopsy specimens. The use of PAS and (both spontaneous or therapy-induced), most histopathological Steiner silver stain did not identify any additional cases. The abnormalities persisted on light microscopy (Fig. 3). Intestinal control individuals recruited in the endoscopic unit had a normal spirochetosis was not identified by electron microscopy in any of colonic mucosa on histological examination. the specimens obtained during the follow up (Fig. 4).

M Esteve et al. Intestinal spirochetosis and diarrhea

Table 1 IS patients diagnosed in the prospective diagnostic work-up of chronic diarrhea (cases 1–8)

Clinical features and Lamina propria Epithelial PMN Patient response to treatment Colonic distribution of IS inﬁltrate lesion activity

Case 1 (male, 29 years) Initial assessment Intermittent chronic diarrhea Right colon sparing No No No After PB treatment Persistence of symptoms – – – Case 2 (male, 55 years) Initial assessment Watery chronic diarrhea Rectal sparing Increase Yes† No After PB treatment Partial improvement Patchy distribution Increase Yes† No After MZ treatment Symptom disappearance Cure Increase Yes† No Case 3 (male, 43 years) Initial assessment Watery chronic diarrhea Entire colon Increase Yes Yes (focal) After PB treatment Partial improvement Entire colon Mild increase No No After MZ treatment Symptom disappearance Cure Mild increase No Yes (focal) Case 4 (male, 57 years) First infection Initial assessment Watery chronic diarrhea Entire colon Mild increase No No After PB treatment Symptom disappearance Isolated IS on EM No No No Second infection Initial assessment Watery chronic diarrhea Entire colon No No No After MZ treatment Symptom disappearance Cure No No Yes (focal) Case 5 (male, 60 years) Initial assessment Watery chronic diarrhea Entire colon Mild increase No No Follow up (not treated) Self-limited symptoms Spontaneous cure Mild increase No No Case 6 (male, 30 years) Initial assessment Watery chronic diarrhea Entire colon + Leishmanias Increase No No After pentamidine Improvement Cure Mild increase No No Case 7 (male, 30 years) Initial assessment Watery chronic diarrhea Patchy distribution Mild increase No No Right colon predominance Follow up (not treated) Persistence of symptom – – – – Case 8 (female, 72 years) Initial assessment Self-limited diarrhea Entire colon No No No (2 months duration) Follow up 12 months (not treated) Asymptomatic Entire colon No No No

EM, electron microscopy; IS, intestinal spirochetosis; MZ, metronidazole 2 g/day (500 mg q.i.d.) for 10 days; PB, penicillin benzathine 2.4 MU i.m. in a single dose; PMN, polymorphonuclear cells. †Lymphocytic exocytosis diagnostic of lymphocytic colitis.

Table 2 IS patients diagnosed in routine biopsy examination (cases 9–11)

Colonoscopic Clinical features and response Colonic distribution Lamina propria Epithelial PMN Patient ﬁndings to treatment of IS inﬁltrate lesion activity

Case 9 (male, 65 years) Initial assessment Non-speciﬁc ulceration of Malaise and weightloss Melena IS in the edge of Mild increase No No the ileocecal valve due to duodenal ulcer the cecal ulcer Follow up Normal Recovery. Cecal and duodenal Spontaneous cure No No No (not treated) ulcer healing (multiple biopsies) Case 10 (female, 87 years) Initial assessment Extensive areas of Abdominal pain Ischemic ulcers Increase Yes Yes No follow up (death) inﬂammation. Ulcers and entire colon thickened folds Case 11 (male, 41 years) Initial assessment Normal (adenomatous polyp) Asymptomatic IS in the right colon No No No Follow up 18 months Normal Asymptomatic Sigmoid and rectal No No No (not treated) sparing

IS, intestinal spirochetosis; PMN, polymorphonuclear cells.

Intestinal spirochetosis and diarrhea M Esteve et al.

Figure 1 (a–c) Continuous layer of spirochetes covering the luminal aspect of the surface epithelium. (a) HE; (b) PAS; (c) Steiner silver stain.

benzathine. In one patient (case 1) the diarrhea persisted after treatment, but he refused an additional colonoscopy and treatment. At the end of follow up, persistence of intermittent watery diarrhea was recorded. Two patients (cases 2 and 3) showed a partial clinical improvement, but follow-up biopsies showed persistence of IS infection on light microscopy in patchy areas of the colon. In these two cases, treatment with metronidazole resolved the diarrhea and follow-up biopsies showed the disappearance of spirochetes from the colonic surface. The remaining patient (case 4) was first diagnosed with IS in 1994, and treated with penicillin benzathine. A good clinical and histological response was observed after treatment. Six years later he was again re-evaluated due to chronic diarrhea over the previous 8 months. Multiple biopsies of the colon again showed the presence of IS in all the samples. Biopsies taken in the previous IS infection were then recovered from formaldehyde-fixed archival tissue and examined by electron microscopy, confirming the presence of small numbers of spiro- Figure 2 Abundant, partly degenerated spirochetes are attached to chetes on the colonic surface, in biopsies taken after penicillin the surface of enterocytes (electron microscopy ×3400). benzathine treatment. The patient was treated with metronidazole, and clinical and histopathological cure occurred. Two more patients showed self-limited symptoms (cases 5 and Polymerase chain reaction analysis 9), and control biopsies showed spontaneous resolution of IS. In Spirochete DNA was amplifiable in only four of the 11 cases. the AIDS patient (case 6), gastrointestinal involvement by Leish- For cases 2 and 11, the PCR were positive for B. pilosicoli, mania (Fig. 5) was considered to be the main cause of chronic while the PCR were positive for B. aalborgi in cases 3 and 8. diarrhea, and consequently he was treated with pentamidine 4 mg/ Brachyspira pilosicoli-specific amplification was also obtained in kg per day. Colonic biopsies taken 2 months later did not identify a sample from case 2 after unsuccessful treatment with penicillin either IS or Leishmania. In two patients (cases 7 and 10), clinical benzathine. and pathological follow up was not available. One patient (case 7) refused to participate in the study and did not receive treatment. At the end of follow up he was contacted by telephone and reported Response to treatment and follow up persistence of mild chronic diarrhea. The other patient (case 10) All symptomatic patients were followed for a mean of 45.4 months died during hospital admission due to complications related to (range: 37–113 months). Four patients were treated with penicillin intestinal ischemia.

M Esteve et al. Intestinal spirochetosis and diarrhea

Figure 3 Colonic biopsies of the same patient (a) before and (b) after treatment. The bacterial layer has disappeared but surface epithelial damage and inﬂammation persist (H&E).

Figure 4 (a) No spiral-shaped microorganisms were detected after treatment (electron microscopy [EM] ×4600). (b) Prominent lysosomes are present in the macrophage from the lamina propria (EM ×25 000).

Patients 8 and 11 remained asymptomatic in spite of IS persis- in the control and chronic diarrhea groups were not statistically tence for more than 1 year after diagnosis. No specific treatment different. A much larger control group (>650 individuals) would was administered. have been required to obtain a statistically significant difference between the groups. Most previous studies assessing the histolog- Discussion ical prevalence of IS have been performed using biopsies of the rectosigmoid area14,17 or in segmental resection surgical speci- To our knowledge this is the first study evaluating the prevalence mens.16,35 In the present study multiple biopsies of five areas of the of histological IS in both patients with and without chronic watery colon were taken from all the patients with chronic diarrhea and diarrhea. In addition, it includes the longest follow up of symp- from the healthy controls. Thus, taking into account that IS may tomatic patients. No IS was diagnosed in 204 control subjects, have a patchy distribution with rectal sparing, differences in prev- while IS was diagnosed in eight (0.7%) of 1174 patients evaluated alence with previous reports could be even higher. for chronic watery diarrhea. The prevalence in the present study is In contrast to previous studies, the routine use of PAS and much lower than previously reported in Western countries (1.1– Steiner silver stains did not increase the number of diagnosed 5%),3,14,16,17,35,36 and, perhaps as a result of this, the prevalence rates cases in the present study.17,33 In fact, the characteristic fuzzy blue

Intestinal spirochetosis and diarrhea M Esteve et al.

bacitracin.14,17 Penicillin benzathine has been used previously with success,38 and this was the first treatment used in the present study due to its simplicity, which ensured therapeutic compliance. How- ever, persistence of infection was documented in all cases. Although a partial therapeutic effect of penicillin was observed, the metronidazole regimen resolved the infection, and thus metronidazole seems to be the drug of choice for treatment of IS. No consistent risk factors for infection were identified but, in agreement with previous observations, colonization was more commonly found in male patients.35 Two of the nine male IS patients were homosexual, and homosexuality is a known risk factor for IS.17 Because there was an aggregation of cases in January–October 2000, an environmental source of infection at that time was suspected, but could not be confirmed. Potential external sources of infection include contaminated water and animals. Two recent reports have demonstrated that colonization with intestinal spirochetes is more frequent in individuals consuming contaminated well water.11,19 In addition, B. pilosicoli is a common Figure 5 Case 6: intracellular structures corresponding to presumptive cause of IS in pigs, dogs and birds, and in some cases B. pilosicoli 20 amastigotes diagnostic of Leishmaniasis were detected in macroph- infection may be zoonotic. It was interesting that one patient ages of the lamina propria. Fuzzy blue band on the luminal surface of the (case 11) who was positive for B. pilosicoli worked in a slaughter- epithelium due to the attachment of spirochetes (HE). house where he was exposed to both pigs and chickens, and where he also drank non-potable water. It seems likely that he was infected from this environment. band on the luminal surface of the epithelium due to IS may be Most IS patients with normal appearing colonic mucosa and clearly distinguished from the turquoise blue of intestinal mucus. chronic watery diarrhea had mild microscopic abnormalities on Moreover, in one case, persistence of IS in post-treatment biopsies light and electron microscopy without signs of invasiveness, and was only retrospectively diagnosed by electron microscopy using these abnormalities remained generally unchanged after IS clear- archival tissue, while persistent infection was not detected in his- ance. This type of abnormality has been referred to as ‘micro- tological sections stained with HE, PAS or Steiner silver stains. scopic colitis not otherwise specified (NOS)’ to differentiate it Specific sequences of B. aalborgi and B. pilosicoli, the main from the well-defined ‘microscopic colitis’ (lymphocitic, collage- spirochetal infections involved in human IS,23,24 were amplified in nous colitis and focal active colitis).41 In some patients with these biopsies of only four patients. In part, this low identification rate histopathological features, intestinal infections or drugs have been may have been due to the prolonged storage of the samples prior to recognized as possible etiologic agents. On light microscopy, one PCR amplification in 2004, with resultant degradation of the patient had a characteristic histopathological pattern of lympho- DNA. Another possibility is that the remaining cases may have cytic colitis and two other patients had focal active colitis. Persis- contained other species of spirochetes that have yet to be charac- tence of colonic damage in patients with microscopic colitis, in terized. The existence of novel Brachyspira species in human IS clinical remission, previously has been described.29–31,42 This is has been reported.37 probably because the reduced absorptive capacity of the colon in No significant differences were found in the occurrence of IS microscopic colitis becomes clinically obvious only in those infection between patients with and without chronic diarrhea, patients with other concomitant pathological conditions30 such as while only three cases of IS were obtained from >20 000 routine celiac disease, bile acid malabsorption or, as in the present study, biopsies, where IS had been recorded as an incidental finding. The IS infection. best evidence that IS was a cause of the chronic watery diarrhea In conclusion, histological IS is rarely diagnosed in the Catalo- among the symptomatic patients was the fact that disappearance of nia area of Spain, but should be considered as a differential diag- the infection, whether spontaneous or following treatment, paral- nosis in patients with otherwise normal-appearing colonic mucosa leled a clinical sustained recovery. This is consistent with what has and chronic watery diarrhea. been previously described in a number of studies.6–9,38–40 It is important not to minimize the importance of IS infection, because invasive forms have been reported.3,6,22,25–28 It is not known if IS Ackowledgment infection had any influence on the outcome of the patient with The authors thank the nursing team at the endoscopic unit of the intestinal ischemia who subsequently died, but electron micros- Hospital Universitari Mútua de Terrassa for their technical assis- copy did not show invasion of the colonic mucosa. Thus, IS should tance and kind support. be treated in patients with persistent symptoms, and histological follow up is advisable to confirm the cure. In contrast, a wait-and- see policy can be followed in asymptomatic patients with normal colon appearance. References Several therapeutic regimens have been proposed for IS eradica- 1 Christie JD. Intestinal spirochetes. Organisms in search of a disease? tion, including penicillin,38 metronidazole,6 and neomycin plus Am. J. Clin. Pathol. 2003; 120: 820–1.

1332 Journal of Gastroenterology and Hepatology 21 (2006) 1326–1333 © 2006 Journal of Gastroenterology and Hepatology Foundation and Blackwell Publishing Asia Pty Ltd M Esteve et al. Intestinal spirochetosis and diarrhea

2van Mook WNKA, Koek GH, van der Ven AJAM, Ceelen TL, Bos RP. 22 Kostman JR, Patel M, Catalano E, Camacho J, Hoffpauir J, DiNubile Human intestinal spirochetosis: any clinical significance? Eur J. Gas- MJ. Invasive colitis and hepatitis due to previously uncharacterized troenterol. Hepatol. 2004; 16: 83–7. spirochetes in patients with advanced human inmunodeficiency virus 3 Körner M, Gebbers JO. Clinical significance of human intesti- infection. Clin. Infect. Dis. 1995; 21: 1159–65. nal spirochetosis: a morphologic approach. Infection 2003; 31: 23 Trivett-Moore NL, Gilbert GL, Law CLH, Trott DJ, Hampson DJ. 341–9. Isolation of Serpulina pilosicoli from rectal biopsy specimens showing 4 Douglas JG, Crucioli V. Spirochaetosis: a remediable cause of diarrhea evidence of intestinal spirochetosis. J. Clin. Microbiol. 1998; 36: 261– and rectal bleeding? Br. Med. J. 1981; 283: 1362. 5. 5Teglbjaerg PS. Intestinal spirochaetosis. Curr. Top. Pathol. 1990; 81: 24 Mikosza ASJ, La T, Brooke CJ et al. PCR amplification from fixed 247–56. tissue indicates frequent involvement of Brachyspira aalborgi in 6 Pheghini PL, Guccion JG, Sharma A. Improvement of chronic diarrhea human intestinal spirochetosis. J. Clin. Microbiol. 1999; 37: 2093–8. after treatment for intestinal spirochetosis. Dig. Dis. Sci. 2000; 45: 25 Trott DJ, Jensen NS, SaintGirons I et al. Identification and character- 1006–10. ization of Serpulina pilosicoli isolates recovered from critically ill 7 Marthinsen L, Willen R, Carlen B, Lindberg E, Varendh G. Intestinal patients. J. Clin. Microbiol. 1997; 35: 482–5. spirochetosis in eight pediatric patients from Southern Sweden. APMIS 26 Padmanabhan V, Dahlstrom J, Maxwell L, Kaye G, Clarke A, Barratt 2002; 110: 571–9. PJ. Invasive intestinal spirochetosis: a report of three cases. Pathology 8 Heine RG, Ward PB, Mikosza ASJ, Bennett-Wood V, Robins-Browne 1996; 28: 283–6. RM, Hampson DJ. Brachyspira aalborgi infection in four Australian 27 White J, Roche D, Chan Y-F, Mitchell EA. Intestinal spirochetosis in children. J. Gastroenterol. Hepatol. 2001; 16: 872–5. children. Report of two cases. Pediatr. Pathol. 1994; 14: 191–9. 9 Amat Villegas I, Borobio Aguilar E, Beloqui Perez R, de Llano Varela 28 Guccion JG, Benator DA, Zeller J, Termanini B, Saini N. Intestinal P, Oquiñena Legaz S, Matinez-Peñuela Virseda JM. Colonic spiro- spirochetosis and acquired immunodeficiency syndrome: ultrastruc- chetes: an infrequent cause of adult diarrhea. Gastroenterol. Hepatol. tural studies of two cases. Ultrastruct. Pathol. 1995; 19: 15–22. 2004; 27: 21–3. 29 Fernández-Bañares F, Salas A, Forné M et al. Incidence of collagenous 10 Lee JI, Hampson DJ. Intestinal spirochaetes colonizing aborigines and lymphocytic colitis: a five-year population based study. Am. J. from communities in the remote north of Western Australia. Epidemiol. Gastroenterol. 1999; 94: 418–23. Infect. 1992; 109: 133–41. 30 Fernández-Bañares F, Esteve M, Salas A et al. Bile acid malabsorption 11 Margawani KR, Robertson ID, Brooke CJ, Hampson DJ. Prevalence, in microscopic colitis and in previously unexplained functional diar- risk factors and molecular epidemiology of Brachyspira pilosicoli in rhea. Dig. Dis. Sci. 2001; 46: 2231–8. humans on the island of Bali, Indonesia. J. Med. Microbiol. 2004; 53: 31 Fernández-Bañares F, Salas A, Esteve M et al. Collagenous and lym- 325–32. phocytic colitis: evaluation of clinical and histological features, 12 Trott DJ, McLaren AJ, Hampson DJ. Pathogenicity of human and response to treatment, and long-term follow-up. Am. J. Gastroenterol. porcine intestinal spirochetes in one-day-old specific-pathogen-free 2003; 98: 340–7. chicks: an animal model of intestinal spirochetosis. Infect. Immun. 32 Swisher BL. Modified Steiner procedure for microwave staining for 1995; 63: 3705–10. spirochetes and nonfilamentous bacteria. J. Histotechnol. 1987; 10: 13 Trott DJ, Huxtable CR, Hampson DJ. Experimental infection of newly 241. weaned pigs with human and porcine strains of Serpulina pilosicoli. 33 Wu ML, Cortina G. Special studies help diagnose intestinal spiroche- Infect. Immun. 1996; 64: 4648–54. tosis in HIV-positive patients. Am. J. Clin. Pathol. 2001; 115: 613–15. 14 Nielsen RH, Orholm M, Pedersen JO, Hovind-Hougen K, Teglbjaerg 34 Mikosza ASJ, La T, de Boer WB, Hampson DJ. The comparative PS, Thaysen EH. Colorectal spirochetosis: clinical significance of the prevalence of Brachyspira (Serpulina) pilosicoli and Brachyspira aal- infestation. Gastroenterology 1983; 85: 62–7. borgi as the etiologic agents of histologically identified intestinal spi- 15 Delladetsima K, Markaki S, Papadimitriou K. Intestinal spirochaeto- rochetosis in Australia. J. Clin. Microbiol. 2001; 39: 347–50. sis. Light and electron microscopic study. Pathol. Res. Pract. 1987; 35 Lindboe CF. The prevalence of human intestinal spirochetosis in 182: 780–2. Norway. Anim. Health Res. Rev. 2001; 2: 117–19. 16 Lindboe CF, Tostrup NE, Nersund R, Rekkavik G. Human intestinal 36 Mikosza ASJ, Hampson DJ, Koopmans MPG, van Duynhoven YTHP. spirochaetosis in mid-Norway. APMIS 1993; 101: 858–64. Presence of Brachyspira aalborgi and B. pilosicoli in feces of patients 17 Surawicz CM, Roberts PL, Rompalo A, Quinn TC, Holmes KK, with diarrhea. J. Clin. Microbiol. 2003; 41: 4492. Stamm WE. Intestinal spirochetosis in homosexual men. Am. J. Med. 37 Mikosza ASJ, Munshi MA, Hampson DJ. Analysis of genetic variation 1987; 82: 587–92. in Brachyspira aalborgi and related spirochaetes determined by partial 18 Trott DJ, Combs BG, Mikosza AS et al. The prevalence of Serpulina sequencing of the 16S rRNA and NADH oxidase genes. J. Med. pilosicoli in humans and domestic animals in the Eastern Highlands of Microbiol. 2004; 53: 333–9. Papua New Guinea. Epidemiol. Infect. 1997; 119: 369–79. 38 Kaplan LR, Takeuchi A. Purulent rectal discharge associated with a 19 Munshi MA, Traub RJ, Robertson ID, Mikosza ASJ, Hampson DJ. nontreponemal spirochete. JAMA 1979; 24: 52–3. Colonization and risk factors for Brachyspira aalborgi and Brachys- 39 Rodgers FG, Rodgers C, Shelton AP, Hawkey CJ. Proposed patho- pira pilosicoli in humans and dogs on tea estates in Assam, India. genic mechanism for the diarrhea associated with human intestinal Epidemiol. Infect. 2004; 132: 137–44. spirochetes. Am. J. Clin. Pathol. 1986; 86: 679–82. 20 Lee JI, Hampson DJ. Genetic characterisation of intestinal spirocha- 40 Lo TC, Heading RC, Gilmour HM. Intestinal spirochaetosis. Postgrad. etes and their association with disease. J. Med. Microbiol. 1994; 40: Med. J. 1994; 70: 134–7. 365–71. 41 Warren BF, Edwards CM, Travis SPL. ‘Microscopic colitis’: classifi- 21 Dassanayake RP, Caceres NE, Sarath G, Duhamel GE. Biochemical cation and terminology. Histopathology 2002; 40: 374–6. properties of membrane-associated proteases of Brachyspira pilosicoli 42 Salas A, Fernández Bañares F, Casalots J et al. Subepithelial myofi- isolated from humans with intestinal disorders. J. Med. Microbiol. broblasts and tenascin expression in microscopic colitis. Histopathol- 2004; 53: 319–23. ogy 2003; 42: 48–54.