Clostridium tertium bacteremia: contamination or true pathogen? A report of two cases and a review of the literature

Article Outline

• Summary
• 1. Introduction
• 2. Case reports
  • 2.1. Case 1
  • 2.2. Case 2
• 3. Discussion
• References
• Copyright

Summary 

We observed two cases of Clostridium tertium bacteremia three months apart in the sterile unit of our department of hematology and oncology. One patient was being treated for first-relapse acute myeloblastic leukemia, while the second was receiving high-dose chemotherapy with hematopoietic stem cell support for non-Hodgkin lymphoma. At the time that C. tertium was identified, the first patient was completely asymptomatic, while the second was highly febrile. Both responded biologically and/or clinically to antibiotherapy. We discuss the epidemiology and pathology of C. tertium in the general and cancer patient population.

Keywords: Clostridium tertium, Bacteremia, Cancer

Back to Article Outline

1. Introduction 

Clostridium tertium is a spore-forming Gram-positive bacillus; it is anaerobic,¹ but aerotolerant.², ³, ⁴, ⁵, ⁶, ⁷, ⁸ This bacterium is found in soil, animal and human gastrointestinal tracts,⁶, ⁸, ⁹, ¹⁰ and the commensal flora of the mouth.², ⁶ C. tertium is considered to be a weak pathogen.², ³, ⁴, ⁶ Its pathogenicity is unclear, but it has been implicated in severe infections with specific characteristics. Here, we report two cases of consecutive bacteremia caused by C. tertium, and discuss C. tertium pathogenicity and predisposing factors, epidemiology, and treatment.

Back to Article Outline

2. Case reports 

2.1. Case 1 

A 51-year-old man was treated in 2006 for M2 acute myeloid leukemia (AML) without cytogenetic abnormalities. Cytological and molecular remission was achieved after a classical induction treatment including cytosine arabinoside and idarubicin, followed by consolidation with high-dose chemotherapy (cyclophosphamide), total body irradiation, and an autologous stem cell transplantation. Two years later, there was an AML relapse in the bone marrow. The patient was admitted to the sterile unit of the department of hematology and oncology in order to receive a new course of induction chemotherapy. Upon admission, blood tests demonstrated pancytopenia: 8.3g/dl hemoglobin (Hb; normal range 13–18g/dl), 0.98×10⁹ leukocytes/l (normal range 4–10.5×10⁹ leukocytes/l), and 10×10⁹ platelets/l (normal range 140–450×10⁹ platelets/l). The patient was completely asymptomatic. There were no signs of infection and he was not taking antibiotics or other medications. Before beginning the chemotherapy, a complete bacterial investigation was performed, including urine, stool, and expectoration analysis, nose and throat swabs, and three blood cultures. Forty-eight hours later, Gram-positive bacilli were isolated from all three blood cultures. The patient was on the third day of cytosine arabinoside and mitoxantrone chemotherapy. He did not have an elevated temperature and his C-reactive protein (CRP) level was normal. Broad-spectrum antibiotherapy was empirically initiated: ceftazidime (6g/24h continuous infusion) and intravenous (IV) teicoplanin (800mg followed by 400mg once a day). Laboratory examination confirmed the presence of C. tertium in all three blood cultures. Antibiotherapy was adapted according to the available antibiogram: teicoplanin was stopped and the patient began taking amoxicillin–clavulanic acid (4g/24h). In vitro, the bacterium was sensitive to penicillin, amoxicillin–clavulanic acid, erythromycin, and metronidazole, and was resistant to clindamycin). Stool analysis and an abdominal computed tomography (CT) scan did not reveal any abnormalities. One week after the initiation of antimicrobial therapy the patient's blood cultures were sterile. The patient remained completely asymptomatic during the entire hospitalization and no elevated temperature was ever reported. On day 28, the patient's bone marrow had recovered and his blood cell count results had improved: 1.16×10⁹ leukocytes/l, 190×10⁹ platelets/l, and 7.7g hemoglobin/dl. Treatment was halted after 26 and 23 days of ceftazidime and amoxicillin–clavulanic acid, respectively.

2.2. Case 2 

A 23-year-old woman with poor-prognosis diffuse non-Hodgkin B-cell lymphoma was admitted to the sterile unit of the department of hematology and oncology for high-dose chemotherapy treatment. She received the BEAM regimen (300mg/m² BCNU on day −6, 200mg/m² VP16 and 200mg/m² aracytine from day −5 to day −2, and 140mg/m² melphalan on day −1), followed by hematopoietic stem cell (HSC) support on day 0. On the second day of chemotherapy, she presented with a fever and shivering; empirical antibiotherapy with amoxicillin–clavulanic acid (4g/24h) was initiated. The patient had anemia (8.6g/dl Hb) and mild leukopenia (2.87×10⁹ leukocytes/l, 99% neutrophils and 1% lymphocytes), but her platelet count was normal (140×10⁹ platelets/l). Inflammation parameters were normal and all of the bacterial cultures remained sterile. Three days later, her temperature was again elevated and CRP levels had dramatically increased to 20mg/dl (normal range <1mg/dl). At that time, C. tertium was isolated from one of three blood cultures. The patient had no intestinal symptoms and no abdominal CT imaging was performed. The antibiotherapy was continued unchanged; her temperature disappeared and inflammatory parameters decreased. HSC were normally infused. Later, an antibiogram showed that the bacterium was resistant to amoxicillin–clavulanic acid. This antibiotic was stopped after 15 days (day 9 post-HSC infusion), when fever recurred during neutropenia, and was replaced by IV amikacin (25mg/kg followed by 15mg/kg daily) and ceftazidime (6g/24h as continuous infusion). Multiple bacteriologic samples including blood cultures, nose and throat swabs, and urine and stool analysis remained negative. Antibiotics were halted on day 14 post-HSC infusion after hematological recovery and complete disappearance of all signs of infection. The patient was discharged on day 15. Later evaluations confirmed she was in complete clinical and radiological remission, with no signs of persisting infection.

Back to Article Outline

3. Discussion 

C. tertium was first described in 1917 by Henry, who isolated it from the wounds of World War I soldiers.⁴ It grows best under anaerobic conditions,⁴, ⁶ and forms spores only under anaerobic conditions.⁴ In contrast with other clostridia, it does not produce toxins.¹, ³ When cultured, it is frequently mistaken for a Bacillus species.⁶, ¹⁰ The differential diagnosis is based on the ability of Bacillus species to form spores in an aerobic environment and the presence of a catalase in Bacillus species.³, ⁴, ⁵, ⁶, ¹¹ In addition to its presence in soil, animal and human guts,⁶, ⁸, ⁹, ¹⁰ and the commensal flora of the mouth,², ⁶ C. tertium has also been identified in traumatic and surgical wounds.², ³, ⁵, ⁹, ¹¹ In wounds, it is usually considered to be a contaminant, however a case of necrotizing fasciitis and gangrene has been described.⁸

Infections with C. tertium are rare and the etiology varies.² Multiple clinical presentations have been described in the literature. It has been identified in premature infants and young children with enterocolitis,² and in children and adults with septicemia, pneumonia with empyema,⁵, ⁶, ⁷, ⁸ spontaneous peritonitis,⁵, ⁶, ⁷, ⁸ colitis,⁶, ⁸, ¹¹ cerebral abcesses,⁶, ⁷, ⁸, ¹⁰, ¹¹ abdominal¹⁰ and cutaneous abscesses,⁷, ¹¹, ¹² meningitis,¹¹ gas gangrene,⁸ and septic arthritis.³, ⁸, ⁹, ¹² However, C. tertium is considered to be a weak pathogen.², ³, ⁴, ⁶ C. tertium isolates are usually found with other bacteria.³, ⁶, ⁷, ⁸, ¹¹ Less frequently, it is the only isolate.⁷, ¹³ Those patients who died and who had C. tertium in their blood cultures had severe underlying diseases², ³, ⁵, ⁶, ¹³ that were potentially fatal in the short term. As a result, the bacterium's virulence has not yet been clearly determined.⁶ Nevertheless, one case of lethal bacteremia with shock and multiple organ failure was reported in a woman with colitis, after surgery for ileus.³ C. tertium was the only pathogen isolated from multiple bacterial samplings. This case challenges the idea that C. tertium's pathogenicity is limited to severely ill or neutropenic patients,⁶ in whom it can cause often-fatal intestinal necrosis.⁷, ¹³ Because C. tertium does not produce toxins, the pathogenic mechanism of its virulence is unknown.⁸ Because of the potentially dramatic consequences of infection, it would be unwise to leave patients with C. tertium untreated, which further complicates our understanding of its pathogenesis.

Bacteremia and septicemia caused by C. tertium were first described in 1963 by King et al.¹ Since then, other cases have been reported. In 1971, 17 patients with C. tertium-positive blood cultures were identified,⁴ followed by two more in 1975.¹⁴ More recent reports of Clostridium bacteremia have been published; globally, Clostridium perfringens is the most common isolate. C. tertium is rare, but is considered to be an emerging pathogen.⁴, ¹⁴ The largest series of C. tertium bacteremia was published in 2001 by Miller et al.,² and comprised 32 patients.

Some common characteristics of C. tertium infection can be extracted from the literature. Patients may present with elevated temperature⁶, ⁷ or gastrointestinal complaints, such as abdominal pain, diarrhea or constipation, rectal bleeding, and nausea.⁶, ⁷, ⁸, ¹¹, ¹³ In the series of Miller et al., all of the patients developed a fever.² However, in some cases, the infection was otherwise perfectly asymptomatic, and abdominal lesions were found only after further investigations such as imaging, or even at autopsy.², ⁷, ⁹

Patient medical histories have frequently revealed blood diseases such as leukemia or marrow aplasia;⁴, ⁷, ¹² inflammatory bowel diseases such as Crohn's disease or ulcerative colitis;⁶, ¹² hepatic failure;⁶, ⁸ abdominal surgeries such as gastrostomy;², ³, ⁴, ⁶, ¹² or immune disorders, such as systemic lupus, that were treated with high doses of corticosteroids.³ Bacteremia preferentially affects patients with immune deficiencies, either systemic or localized. For example, in the gastrointestinal tract, mucosal breaches allow bacterial translocation.², ³, ⁴, ⁸, ¹¹ Moreover, chemotherapy can lead to inflammation and abrasion of the intestinal mucosa.³, ⁹, ¹¹ This may partially explain the increased susceptibility to infection of patients with hematological diseases who are undergoing chemotherapy.

A large number of the patients described in the literature were neutropenic when bacteremia was diagnosed,², ³, ⁶, ⁷, ⁸, ⁹, ¹¹, ¹², ¹³ but there are exceptions.², ⁶ Typically, the few patients who developed bacteremia without neutropenia presented with gastrointestinal symptoms.⁶, ¹⁰ Interestingly, most patients with C. tertium bacteremia were neutropenic without a defined source of infection.² Regarding predisposing medications, third-generation cephalosporins², ³, ⁴, ⁸, ¹² and aminoglycosides appear to be risk factors for the development of C. tertium infections; the bacterium is resistant to these antibiotics, and is thus selected for.⁶, ⁹, ¹¹ Administration of broad-spectrum antibiotics also facilitates the growth of resistant bacteria.⁷

No prophylaxis is recommended to prevent C. tertium growth due to the low incidence of infection and the uncertainty regarding its pathogenicity. C. tertium is resistant to beta-lactams and clindamycin,², ³, ⁴, ⁵, ⁹, ¹¹ and sensitive to quinolones,², ³, ⁷ imipenem,³, ⁶, ⁹, ¹¹, ¹² vancomycin,², ³, ⁶, ⁷, ¹¹, ¹² and trimethoprim–sulfamethoxazole.², ³, ⁷, ⁹ It can be either sensitive⁵, ⁶, ¹¹, ¹² or resistant², ⁵, ⁶, ⁷, ⁹ to metronidazole. Some authors have observed that it is sensitive to penicillin.¹⁰, ¹² Therapy with two or three antibiotics is generally initiated, but there are no guidelines.¹, ⁴, ⁶, ⁷ The duration of antibiotherapy is rarely mentioned in the literature, but appears to range from 15 to 27 days.⁴, ¹⁰ In our first case report, the patient received intravenous antibiotherapy for the entire duration of marrow aplasia, from the time that bacteremia was first noted (24 days in total). Interestingly, in contrast with the literature, C. tertium in our patient was sensitive to beta-lactams. For our second patient, the antibiogram matched the data in the literature; even so, amoxicillin–clavulanic acid was administered successfully, although the bacterium was resistant in vitro. In this case, antibiotherapy was continued for nearly 21 days.

Some authors have doubts about the clinical importance of the identification of C. tertium.¹, ⁴, ⁶, ¹¹ However, in the series of Miller et al., four patients died within a week after C. tertium was identified.² In the two cases that we report here, we considered the possibility of culture contamination because of the atypical clinical presentation (the first patient was completely asymptomatic) and because the bacterium was isolated from only one of three blood cultures from the second patient. In the latter case, the infection clinically and bacteriologically improved despite in vitro evidence of inadequate antibiotherapy. C. tertium is aerotolerant and present in the commensal flora of the mouth, increasing the likelihood of contamination. However, both patients were at risk (because of hematological malignancies, leukopenia, and chemotherapy) for C. tertium bacteremia, and were treated in the same sterile unit within a short period of time (less than three months); thus, contamination seems unlikely.

In conclusion, we report two cases of C. tertium bacteremia observed in the sterile unit of our department of hematology and oncology within a short period of time. Both patients were susceptible to C. tertium infection because of their hematological disease and neutropenia. In the first patient, who was completely asymptomatic, the bacteremia was discovered during routine bacteriologic sampling before starting chemotherapy. The symptoms of the second patient improved despite the patient receiving an antibiotic to which the bacterium was resistant in vitro. C. tertium's pathogenicity appears to be low, and it is not entirely clear if C. tertium is a true pathogen or merely a bystander contaminant. In addition, the appropriate duration of antibiotherapy is still a matter of debate.

Conflict of interest: No conflict of interest to declare.

Back to Article Outline

References 

1. King BM, Ranck BA, Daugherty FD, Rau CA. Clostridium tertium septicemia. N Engl J Med. 1963;269:467–469
   • View In Article
   • MEDLINE
   • CrossRef
2. Miller DL, Brazer S, Murdoch D, Reller B, Corey R. Significance of Clostridium tertium bacteremia in neutropenic and non neutropenic patients: review of 32 cases. Clin Infect Dis. 2001;32:975–978
   • View In Article
   • MEDLINE
   • CrossRef
3. Tappe D, Dirks J, Müller R, Brederlau J, Abele-Horn M, Suberbaum S. Fatal Clostridium tertium septicaemia in a non neutropenic patient. J Infect. 2005;50:76–80
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (180 KB)
   • CrossRef
4. Thaler M, Gill V, Pizzo PA. Emergence of Clostridium tertium as pathogen in neutropenic patient. Am J Med. 1986;81:596–600
   • View In Article
   • Abstract
   • Full-Text PDF (494 KB)
   • CrossRef
5. Leegaard TM, Sandven P, Gaustad P. Clostridium tertium: 3 case reports. Scand J Infect Dis. 2005;37:230–240
   • View In Article
   • MEDLINE
   • CrossRef
6. Gosbell IB, Johnson CG, Newton PJ, Jelfs J. Clostridium tertium bacteremia: 2 cases and review. Pathology. 1996;28:70–73
   • View In Article
   • MEDLINE
   • CrossRef
7. Speirs G, Warren RE, Rampling A. Clostridium tertium septicemia in patient with neutropenia. J Infect Dis. 1988;158:1336–1340
   • View In Article
   • MEDLINE
   • CrossRef
8. Ray P, Das A, Singh K, Bhansali A, Yadav TD. Clostridium tertium in necrotizing fasciitis and gangrene. Emerg Infect Dis. 2003;9:1347–1348
   • View In Article
   • MEDLINE
9. Valtonen M, Sivonen A, Elonen E. A cluster of seven cases of Clostridium tertium septicemia in neutropenic patient. Eur J Clin Microbiol Infect Dis. 1990;9:40–42
   • View In Article
   • MEDLINE
   • CrossRef
10. Severin A, Paulet R, Berth A, Balgone L, Coudray JM, Thyrault M. Pancréatite aiguë compliquée d’une septicémie à Clostridium tertium. Presse Med. 2005;34:446–447
    • View In Article
    • MEDLINE
    • CrossRef
11. Steyaert S, Peleman R, Vaneechoutte M, De Baere T, Claeys G, Verschraegen G. Septicemia in neutropenic patients infected with Clostridium tertium resistant to cefepime and other expanded-spectrum cephalosporins. J Clin Microbiol. 1999;37:3778–3779
    • View In Article
    • MEDLINE
12. Gredlein CM, Silverman ML, Downey MS. Polymicrobial septic arthritis due to Clostridium species: case report and review. Clin Infect Dis. 2000;30:590–594
    • View In Article
    • MEDLINE
    • CrossRef
13. Rampling A, Speirs G, Warren RE. Clostridium tertium and the gut. Lancet. 1988;1:534
    • View In Article
    • MEDLINE
14. Gorbach SL, Thadepalli H. Isolation of Clostridium in human infections: evaluation of 114 cases. J Infect Dis. 1975;131:S81–S85
    • View In Article
    • CrossRef