Healey, Letha M. MD*†; Gibert, Cynthia L. MD*†; Liappis, Angelike P. MD*
Originally described as a Campylobacter-like organism, Helicobacter cinaedi is a gram-negative enteric bacteria which has been described as a pathogen primarily among those immunocompromised because of age (neonates) and medications (ie, immunosuppressive and chemotherapeutic agents) and those with underlying diseases such as malignancy, chronic obstructive pulmonary disease, cirrhosis, diabetes, alcoholism, or hereditary immunodeficiencies such as cross-linked agammaglobulinemia.1-4
A rare cause of infection in the immunocompetent5 immunocompromised hosts commonly present with blood stream infection accompanied by fever and gastrointestinal symptoms, including abdominal pain and diarrhea. In addition, H. cinaedi bacteremia has often been described in association with recurrent skin involvement of the extremities.2,3,6
In 1999, we published a case of an HIV-infected patient with Helicobacter bacteremia.7 At that time, ribosomal testing identified the organism as a new Helicobacter species (VAH). Subsequent testing of our isolate and of other previously described Helicobacter species (including Helicobacter westmeadii and strain Mainz) has identified all of these as H. cinaedi.8
There were 4 additional H. cinaedi bacteremic infections at our institution over a 7-year period, with all cases occurring in HIV-infected men. We describe the clinical presentations and present a comprehensive review of 33 cases identified in a PubMed search (through July 2005) of H. cinaedi bacteremia or previously misidentified species in HIV-infected individuals.
In this review, we have explored the contribution of the host immunologic status and the organism's predilection for recurrent cellulitis, and have examined epidemiological features and potential routes of exposure which may contribute to H. cinaedi colonization and infection in HIV-infected adults.
A 46-year-old man with a CD4 count of 136 cells/μL and HIV RNA of 861 copies/mL presented with a 1-day history of fever (101°F), chills, persistent diarrhea, and 3 patches of erythema over both the upper and lower extremities, which were warm and tender to palpation. Examination also revealed a palpable effusion of the right knee, with limited range of motion on passive flexion.
Before this admission, the patient had been hospitalized on 3 separate occasions over the preceding 6 months with similar symptoms. At each admission, the presentation was marked by fever, chills, and cellulitis of his lower extremities. The patient had been treated at each hospitalization with a short course (3-5 days) of intravenous nafcillin and discharged to complete an additional 10 to 14 days of oral dicloxacillin after clinical improvement was noted. Multiple blood cultures obtained over the course of these 3 admissions had been negative.
Two sets of blood cultures were obtained, and empiric therapy with intravenous ciprofloxacin and clindamycin was initiated. On the fourth day, one culture bottle grew an organism which was later identified to be H. cinaedi through 16s ribosomal RNA (rRNA) analysis.
A transesophageal echocardiogram identified a linear mobile echodensity on the mitral valve, which was present on a transthoracic echocardiogram obtained during a previous admission and attributed to calcified chordae tendinae in the absence of positive blood cultures. Cultures of synovial fluid and stool as well as subsequent blood cultures, obtained while on therapy, failed to grow the organism. After a 6-week course of intravenous imipenem and gentamicin, no further episodes of cellulitis occurred.
A 26-year-old man with CD4 224 cells/μL was admitted with complaints of abdominal pain and myalgias. On presentation, the patient was febrile (103°F). The physical examination was significant for bilateral upper quadrant abdominal tenderness. There was no skin involvement. Two sets of blood cultures were obtained at admission. After 4 days of incubation, H. cinaedi was recovered from 2 bottles. The patient was treated with a 14-day course of ciprofloxacin without recurrence of bacteremia.
A 44-year-old man with CD4 count 386 cells/μL, HIV RNA of 64,514 copies/mL, and hepatitis C infection was admitted with acute change in mental status attributed to worsened hepatic encephalopathy after a 3-week history of fever, lower abdominal pain, and profuse diarrhea.
On hospital day 5, the patient was febrile to 102.3°F. Two sets of blood cultures were obtained. H. cinaedi was isolated from one bottle after 3 days of incubation. Stool cultures failed to demonstrate the organism. With antibiotic therapy, the patient improved clinically. On day 11, H. cinaedi was again isolated in 1 of 4 blood cultures. Echocardiogram failed to reveal a valvular lesion. A repeat set of blood cultures drawn on hospital day 16 was negative, and the patient was discharged on an oral fluoroquinolone to complete a 21-day course.
A 49-year-old man with a CD4 cell count of 105 cells/μL and HIV RNA of 280,043 copies/mL was admitted with newly diagnosed Burkitt lymphoma for chemotherapy. During treatment, the patient developed neutropenia followed by fever. The patient complained of right lower quadrant abdominal pain and loose stools. From 1 of 4 blood culture bottles, after 5 days of incubation, a curved gram-negative organism was isolated which was later identified as H. cinaedi. With intravenous ceftazidime therapy, the patient defervesced as his neutropenia resolved. The patient finished a 14-day total course of antibiotics with oral ciprofloxacin. All subsequent blood cultures remained negative, and the patient's abdominal pain and diarrhea resolved.
Five patients with H. cinaedi bacteremia were identified at our institution from a total of 35,260 blood cultures drawn on 6960 patients over a 7-year period. All H. cinaedi bacteremic infections were found to involve HIV-infected men.
Thirty-two additional cases of H. cinaedi bacteremia in HIV-infected patients have been reported in the literature since 1982. Table 1 summarizes the clinical and laboratory features of our cases (cases 1-4) as well as in the previously reported cases of H. cinaedi bacteremia found in HIV-infected individuals (cases 5-37).
Patients with H. cinaedi bacteremia (n = 37) ranged in age from 26 to 49 years (mean, 35 years). CD4 counts were reported for 19 (51.4%) of the 37 patients, with mean 112 ± 116 cells/μL (range, 1-386 cells/μL). Patient gender was reported in only 70% of the cases; of the reported cases where gender was identified, 25 (96%) of 26 were reported as male. The sexual orientation was reported for 13 (35%) of 37 patients. For 12 (92%) of 13, including all of our patients, men having sex with men (MSM) were identified as a risk factor for HIV acquisition.
Most patients presented with a febrile illness. Fever, defined as temperature ≥38.3°C, was documented in 32 (86.5%) of the 37 patients. Gastrointestinal symptoms were a common finding, encountered in 20 (54.1%) of 37 of patients. Symptoms reported at presentation or at the time of bacteremia included profuse diarrhea (43.2%), nausea/vomiting (18.9%), and mild to moderate abdominal pain (21.6%). Involvement of skin and soft tissues was seen in 13 (35.1%) of 37 patients, most frequently the lower extremities (Table 1).
Recurrence of infection was reported in 13 (35.1%) of 37. We defined recurrence as the return of clinical symptoms on antibiotic therapy or as the development of positive cultures (blood, synovial, or stool) after documented clearance of prior bacteremia.
For cases reported in the literature, to define recurrence, we relied on available clinical data or the author's determination. Information regarding outcomes was not available for 2 patients. For 5 (38.5%) of the 13 patients, multiple recurrences (>1 episode) of H. cinaedi infection were reported.
There was no difference between the CD4 counts in the recurrent (n = 13) and nonrecurrent (n = 24) patients (96 vs 118 cells/μL, P = 0.61 with 2-tailed unpaired t test), and there was a trend for patients with cellulitis to have recurrent disease (46.2% vs. 27.3%, P = 0.28 with 2-tailed χ2 test).
Our series is the first to comprehensively review the presentation and salient clinical and epidemiological features of H. cinaedi bacteremia among HIV-infected individuals.
The organism has a predilection for certain hosts. Among the thousands of bacteremic patients at our institution, only 5 men were found to have blood stream infection with H. cinaedi over a 7-year period. All 5 patients were HIV-infected MSM. In our review, bacteremic HIV-infected patients typically presented with a febrile illness and gastrointestinal symptoms (abdominal pain and diarrhea) not dissimilar to cases described in other immunocompromised hosts.
At our institution, blood subculture revealed narrow, curved, or gull wing-shaped, gram-negative organisms; the identification of H. cinaedi was later confirmed through 16S rRNA analysis. In other reviews of Helicobacter infection, difficulty in visualizing the organism on Gram stain prompted the use of additional staining techniques (acridine orange, Giemsa staining, or dark-field examination) to assist in organism identification from positive cultures.3,9-11
Culturing the organism from blood requires rich nonselective media (blood or chocolate agar) in a microaerophilic atmosphere, at 37°C.3 The yield from stool and other specimens (such as rectal swabs) may be increased by plating directly onto selective media and by prolonging culture incubation 10 to 14 days.3,6,9
For those patients with available data, we were unable to find a correlation between the incidence of H. cinaedi bacteremia and either the level of immunosuppression (absolute CD4 cell count), the use of antiretroviral therapy, or prophylactic antibiotic therapy against HIV-related opportunistic infection. However, among the previously published cases we reviewed, it should be noted that laboratory and clinical information was not always uniformly reported.
To date, studies have failed to clearly define the risk factors for acquiring Helicobacter infection. Exposure to animals known to harbor the organisms has been speculated to be the source of disease.3,12 H. cinaedi has been isolated from gerbils, rhesus monkeys, and dogs, but direct transmission from animal to human has not been documented.13,14
Person-to-person transmission via the fecal-oral route is also a potential source for acquiring infection. As with other infectious diseases with this mode of transmission, exposure routes include contaminated water and unprotected anal-receptive intercourse.3,15 Acquisition of H. cinaedi through sexual contact is supported by our findings. In the cases where sufficient clinical information was available, we found that infections occurred predominately in men who have sex with men. The organism has been isolated from the stool of MSM with proctocolitis, suggesting that fecal carriage rates of this organism may be higher among homosexual men.15
The relative frequency of infectious and noninfectious gastroenteritis in the HIV-infected population may facilitate invasion by colonizing organisms, such as Helicobacter species.6 The antibody response to pathogens may be altered by the immune dysregulation of HIV infection, increasing the risk for bacteremia related to either chronic gastrointestinal colonization or to more acute exposures (sexual or environmental) to Helicobacter.
Exposure through sexual contact, combined with defects in humoral immunity, may predispose HIV-infected MSM to gastrointestinal colonization and subsequent invasive disease.2,3,6 Bloodstream infections with Campylobacter species occur more commonly in patients with deficiencies in humoral immunity.16 The microbial killing and clearance of Campylobacter species in the bloodstream are accomplished by complement-mediated proteins and bactericidal antibodies. An intact immune response is also important in defense against invasion of Campylobacter through the gastrointestinal tract.17
Approximately 10% of the HIV patients presenting with H. cinaedi bacteremia had a concurrent diagnosis of an underlying malignancy (lymphoma and Kaposi sarcoma). In the cases we reviewed, other causes of immunosuppression included persistent parvovirus B19 requiring immunoglobulin treatment, chronic alcohol abuse, and pyomyositis requiring prolonged steroid therapy. In our cohort of 5 patients, one had advanced cirrhosis caused by hepatitis C, one had end-stage renal disease, and another had chemotherapy-related neutropenia.
As reported among other immunocompromised individuals, we found that HIV-infected adults with H. cinaedi bacteremia commonly presented with multifocal cellulitis and arthritis. In our review, a diagnosis of cellulitis was made in approximately 40% of the bacteremic patients, with soft tissue involvement of the extremities being the most common location identified.
Recurrence of infection after a course of appropriate therapy has been previously reported with H. cinaedi infection.3,10,18-20 Recurrent disease was a feature of over one third of the cases of bacteremic infection among our patients, most presenting with multiple episodes.
In subsequent analysis of those with recurrent disease, we were not able to identify differences in absolute CD4 counts or of the use of antiretroviral therapy. Nor were we able to ascertain differences in antibiotic therapies used from available data. The length of therapy and the choice of antibiotic were not included in most cases reported. In the cases where the information was available, β-lactam antibiotics and fluoroquinolones (predominately ciprofloxacin) were the most frequently used therapies. Approximately one third of cases of H. cinaedi bacteremia in this series involved recurrence of either bacteremia or cellulitis despite antibiotic treatment.
Relapsing infection, characterized by recurrent bacteremia and/or cellulitis of the extremities, may indicate underlying endovascular disease. Superficial thrombophlebitis and cellulitis have been described in cases of bacteremia associated with 2 species closely related to Helicobacter, Campylobacter fetus, and Campylobacter jejuni.21,22 Bacteremia with these organisms, as well as H. cinaedi, has been associated with endovascular lesions.5,23,24 Endovascular involvement with H. cinaedi has been reported in 4 HIV-infected patients of the cases reviewed. Of these, 2 patients had recurrent infections.
A limitation of this series is the retrospective nature of the data collected. Several of the historic cases dated to the early 1980s, when the diagnosis of HIV infection was first being recognized. Not all of the clinical information was uniformly available from these early case reports and case series, limiting the statistical power of some comparisons. However, we found that the clinical course, treatment, and presentation were well documented for most patients.
We conclude from our case review that H. cinaedi is a rare pathogen in HIV-infected patients. Providers caring for HIV-infected patients must have a high clinical suspicion to properly identify H. cinaedi disease. The diagnosis of H. cinaedi bacteremia should be considered in the evaluation of the febrile HIV-infected patient with skin involvement, in addition to the more commonly encountered bacterial pathogens, Staphylococcus and Streptococcus. A history of recurrent disease should also prompt consideration of this organism in the differential diagnosis of potential causative pathogens.
Among the bacteremic HIV-infected patients in our series, we could not identify a definitive source of exposure to H. cinaedi. There are data to suggest a possible association between gastrointestinal colonization and infection. We suspect that the relative frequency of H. cinaedi bacteremia among HIV-infected men may result from increased exposure to the organism combined with the altered humoral response to infection in HIV disease. The role of sexual transmission in the acquisition of H. cinaedi and the contribution of immune suppression to the development of disease will require prospective studies, with larger patient cohorts.
The authors thank Dr Steven Fischer at the Department of Laboratory Medicine, National Institutes of Health, for their assistance with the 16s rRNA analysis of our isolates. They also thank Dr Virginia Kan, Dr Robert Williamson, and Ms Sabiha Zubairi at the Washington, DC, Veterans Affairs Medical Center.
1. Orlicek SL, Welch DF, Kuhls TL. Septicemia and meningitis caused by Helicobacter cinaedi in a neonate. J Clin Microbiol. 1993;31(3):569-571.
2. Tee W, Street AC, Spelman D, et al. Helicobacter cinaedi bacteremia: varied clinical manifestations in three homosexual males. Scand J Infect Dis. 1996;28:199-203.
3. Kiehlbauch JA, Tauxe RV, Baker CN, et al. Helicobacter cinaedi-associated bacteremia and cellulitis in immunocompromised patients. Ann Intern Med. 1994;121(2):90-93.
4. Simons E, Spacek LA, Lederman HM, et al. Helicobacter cinaedi presenting as macules in an afebrile patient with X-linked agammaglobulinemia. Infection. 2004;32(6):367-368.
5. Lasry S, Simon J, Marais A, et al. Helicobacter cinaedi septic arthritis and bacteremia in an immunocompetent patient. Clin Infect Dis. 2000;31:201-202.
6. Burman WJ, Cohn DL, Reves RR, et al. Multifocal cellulitis and monoarticular arthritis as manifestations of Helicobacter cinaedi bacteremia. Clin Infect Dis. 1995;20:564-570.
7. Weir SC, Gibert CL, Gordin FM, et al. An uncommon Helicobacter isolate from blood: evidence of a group of Helicobacter spp. pathogenic in AIDS patients. J Clin Microbiol. 1999;37(8):2729-2733.
8. VanDamme P, Harrington CS, Jalava K, et al. Misidentifying Helicobacters: the Helicobacter cinaedi example. J Clin Microbiol. 2000;38(6):2261-2266.
9. Cimolai N, Gill MJ, Jones A, et al. "Campylobacter cinaedi" bacteremia: case report and laboratory findings. J Clin Microbiol. 1987;25(5):942-943.
10. Ng VL, Hadley WK, Fennell CL, et al. Successive bacteremias with "Campylobacter cinaedi" and "Campylobacter fennelliae" in a bisexual male. J Clin Microbiol. 1987;25(10):2008-2009.
11. Husmann M, Gries C, Jehnichen P, et al. Helicobacter sp. strain Mainz isolated from an AIDS patient with septic arthritis: case report and nonradioactive analysis of 16S rRNA sequence. J Clin Microbiol. 1994;32(12):3037-3039.
12. Gebhart CJ, Fennell CL, Murtaugh MP, et al. Campylobacter cinaedi is normal intestinal flora in hamsters. J Clin Microbiol. 1989;27:1692-1694.
13. Misawa N, Kawashima K, Kondo F, et al. Isolation and characterization of Campylobacter, Helicobacter, and Anaerobiospirillum strains from a puppy with bloody diarrhea. Vet Microbiol. 2002;87(4):353-364.
14. Fernandez KR, Hansen LM, Vandamme P, et al. Captive rhesus monkeys (Macaca mulatta) are commonly infected with Helicobacter cinaedi. J Clin Microbiol. 2002;40(6):1908-1912.
15. Quinn TC, Goodell SE, Fennell C, et al. Infections with Campylobacter jejuni and Campylobacter-like organisms in homosexual men. Ann Intern Med. 1984;101:187-192.
16. Blaser MJ, Smith PF, Kohler PF. Susceptibility of Campylobacter isolates to the bactericidal activity of human serum. J Infect Dis. 1985;151:227-235.
17. Sullivan AK, Nelson MR, Walsh J, et al. Recurrent Helicobacter cinaedi cellulitis and bacteraemia in a patient with HIV infection. Int J STD AIDS. 1997;8:59-60.
18. Fleisch F, Burnens A, Weber R, et al. Helicobacter species strain Mainz isolated from cultures of blood of two patients with AIDS. CID. 1998;28:526-527.
19. Mammen MP, Aronson NE, Edenfield WJ, et al. Recurrent Helicobacter cinaedi bacteremia in a patient infected with human immunodeficiency virus: case report. CID. 1995;21:1055.
20. Schmidt U, Chmel H, Kaminski ZK, et al. The clinical spectrum of Campylobacter fetus infections: report of five cases and review of the literature. Q J Med. 1980;49:431-442.
21. Skirrow MB, Jones DM, Sutcliffe E, et al. Campylobacter bacteraemia in England and Wales, 1981-91. Epidemiol Infect. 1993;110:567-573.
22. Carbone KM, Heinrich MC, Quinn TC. Thrombophlebitis and cellulitis due to Campylobacter fetus spp. fetus. Medicine. 1985;64:244-250.
23. Guerrant RL, Lahita RG, Winn RB Jr, et al. Campylobacteriosis in man: pathogenic mechanisms and review of 91 bloodstream infections. Am J Med. 1978;65:584-592.
24. vanderVen AJAM, Kullberg BJ, Vandamme P, et al. Helicobacter cinaedi bacteremia associated with localized pain but not with cellulitis. Clin Infect Dis. 1996;22:710-712.
25. Hung CC, Hsueh PR, Chen MY, et al. Bacteremia caused by Helicobacter cinaedi in an AIDS patient. J Formos Med Assoc. 1997;96(7):558-559.
26. Trivett-Moore NL, Rawlinson WD, Yuen M, et al. Helicobacter westmeadii sp. nov., a new species isolated from blood cultures of two AIDS patients. J Clin Microbiol. 1997;35(5):1144-1150.
27. Decker CF, Martin GJ, Barham WB, et al. Bacteremia due to Campylobacter cinaedi in a patient infected with the human immunodeficiency virus. Clin Infect Dis. 1992;15:178-179.
28. Aboulafia D, Mathisen G, Mitsuyasu R. Case report: aggressive Kaposi's sarcoma and Campylobacter bacteremia in a female with transfusion associated AIDS. Am J Med Sci. 1991;301(4):256-258.
29. Sacks LV, Labriola AM, Gill VJ, et al. Use of ciprofloxacin for successful eradication of bacteremia due to Campylobacter cinaedi in a human immunodeficiency virus-infected person. Rev Infect Dis. 1991;13:1066-1068.
© 2006 Lippincott Williams & Wilkins, Inc.