Research Article

A fatal Vibrio cholerae O37 enteritis

  • 1UOC Microbiologia e Virologia, Azienda Ospedaliera ‘Ospedale San Carlo Borromeo’, Milano, Italy
  • 2UOC Istologia e Anatomia Patologica, Azienda Ospedaliera ‘Ospedale San Carlo Borromeo’, Milano, Italy
  • 3Dipartimento Malattie Infettive, Parassitarie e Immunomediate, Istituto Superore di Sanità, Roma, Italy
  • 4Laboratorio Nazionale di Riferimento Contaminazioni Batteriologiche dei Molluschi Bivalvi, Istituto Zooprofilattico Sperimentale dell'Umbria e delle Marche, Ancona, Italy
  • 5Direzione Sanitaria, Azienda Ospedaliera ‘Ospedale San Carlo Borromeo’, Milano, Italy
  • Correspondence
    : C. Farina
    (farina.claudio{at}sancarlo.mi.it)
  • Journal of Medical Microbiology 2010; 59(12):1538–1540 · https://doi.org/10.1099/jmm.0.023093-0

    View at publisher PubMed

    Abstract

    Isolates of non-O1 and non-O139 Vibrio cholerae are associated with sporadic diarrhoeal disorders, limited enteritis outbreaks or extraintestinal infections, especially in immunocompromised patients with haematological malignancies, ascites, cirrhosis and renal failure. In Italy, very few cases of intestinal and extraintestinal infections caused by non-O1 and non-O139 V. cholerae have been described in the past, as reported by Piersimoni et al. (1991), Farina et al. (1999, 2000) and Ottaviani et al. (2009).

    A 54-year-old Egyptian man living in northern Italy was admitted on 5 August 2008 to the Dialysis Unit of the AO ‘Ospedale San Carlo Borromeo’, Milano, Italy, to be suddenly submitted to haemodialysis (blood urea nitrogen 323 mg dl−1). The patient, with a well-known history of asthma, hypertension, hydronephrosis of the left kidney, urethral stenosis and uraemic pericarditis treated by pericardiotomy and Salmonella orchitis 1 year previously, had been regularly treated for 3 years with haemodialysis three times per week because of chronic renal failure. At admission, he declared his recent return from Alexandria, Egypt, where he stayed with his family for 3 weeks. Two days before coming back, he consumed undercooked fish and vegetables along the Nile river, and a few hours later he presented with vomiting and nausea.

    A few hours after haemodialysis, the patient presented with greenish diarrhoea and vomiting, abdominal tenderness and fever (temperature 38.5 °C). He had no skin lesions. Blood analysis showed a leukocyte count of 9690 ml−1 (neutrophils 73.3 %). Blood chemistry values were as follows: blood urea nitrogen, 164 mg dl−1; creatinine, 17.38 mg dl−1; troponin I, 0.55 ng ml−1; sodium, 137 mmol l−1; potassium, 4.94 mmol l−1. Prothrombin time international normalized ratio was 1.50, glutamic oxaloacetic transaminase was 55 U l−1 and glutamic pyruvic transaminase was 40 U l−1. The C-reactive protein level was 24.90 mg dl−1 and the erythrocyte sedimentation rate was 82 mm h−1.

    He was treated with parenteral rehydration, acidosis correction and antibiotic therapy (ampicillin, then imipenem plus sulfamethoxazole–trimethoprim). An abdominal examination revealed a diffuse tenderness in the right hypocondrium but no resistance to palpation. Abdominal ultrasound and abdominal tomography demonstrated a reduction of the renal cortex, a normal liver, but no ascites or splenomegaly.

    Nine blood samples were negative. Faeces were collected daily for 10 days: two samples yielded non-O1 and non-O139 V. cholerae, finally serotyped as V. cholerae O37. The strain was susceptible in vitro to several antimicrobial agents, including ampicillin, first-, second- and third-generation cephalosporins, imipenem, gentamicin, chloramphenicol, sulfamethoxazole–trimethoprim and quinolones, when tested by the agar disc diffusion method according to CLSI (2006) guidelines.

    The patient's condition became worse, and he died on 16 August because of hypovolaemic shock and ventricular fibrillation. The autopsy revealed end-stage kidney disease with haemodialysis-associated amyloidosis and left ureteral stenosis with stent. The gastrointestinal tract showed congestion and only a moderate infiltration of mononuclear inflammatory cells with focal superficial necrosis not directly related to V. cholerae. In addition, shock-related focal myocardial necrosis, diffuse vascular congestion, lung oedema, chronic uraemic pericarditis and endomyocarditis were observed. Culture of all tissues was negative for V. cholerae. Pus collected from the kidney and bladder yielded Salmonella enterica group C.

    The V. cholerae O37 strain was further characterized for different virulence properties by using phenotypic tests and PCR assays to detect the presence of toxin genes (Table 1).

    Isolates of non-O1 and non-O139 Vibrio cholerae are associated with sporadic diarrhoeal disorders, limited enteritis outbreaks or extraintestinal infections, especially in immunocompromised patients with haematological malignancies, ascites, cirrhosis and renal failure. In Italy, very few cases of intestinal and extraintestinal infections caused by non-O1 and non-O139 V. cholerae have been described in the past, as reported by Piersimoni et al. (1991), Farina et al. (1999, 2000) and Ottaviani et al. (2009).

    A 54-year-old Egyptian man living in northern Italy was admitted on 5 August 2008 to the Dialysis Unit of the AO ‘Ospedale San Carlo Borromeo’, Milano, Italy, to be suddenly submitted to haemodialysis (blood urea nitrogen 323 mg dl−1). The patient, with a well-known history of asthma, hypertension, hydronephrosis of the left kidney, urethral stenosis and uraemic pericarditis treated by pericardiotomy and Salmonella orchitis 1 year previously, had been regularly treated for 3 years with haemodialysis three times per week because of chronic renal failure. At admission, he declared his recent return from Alexandria, Egypt, where he stayed with his family for 3 weeks. Two days before coming back, he consumed undercooked fish and vegetables along the Nile river, and a few hours later he presented with vomiting and nausea.

    A few hours after haemodialysis, the patient presented with greenish diarrhoea and vomiting, abdominal tenderness and fever (temperature 38.5 °C). He had no skin lesions. Blood analysis showed a leukocyte count of 9690 ml−1 (neutrophils 73.3 %). Blood chemistry values were as follows: blood urea nitrogen, 164 mg dl−1; creatinine, 17.38 mg dl−1; troponin I, 0.55 ng ml−1; sodium, 137 mmol l−1; potassium, 4.94 mmol l−1. Prothrombin time international normalized ratio was 1.50, glutamic oxaloacetic transaminase was 55 U l−1 and glutamic pyruvic transaminase was 40 U l−1. The C-reactive protein level was 24.90 mg dl−1 and the erythrocyte sedimentation rate was 82 mm h−1.

    He was treated with parenteral rehydration, acidosis correction and antibiotic therapy (ampicillin, then imipenem plus sulfamethoxazole–trimethoprim). An abdominal examination revealed a diffuse tenderness in the right hypocondrium but no resistance to palpation. Abdominal ultrasound and abdominal tomography demonstrated a reduction of the renal cortex, a normal liver, but no ascites or splenomegaly.

    Nine blood samples were negative. Faeces were collected daily for 10 days: two samples yielded non-O1 and non-O139 V. cholerae, finally serotyped as V. cholerae O37. The strain was susceptible in vitro to several antimicrobial agents, including ampicillin, first-, second- and third-generation cephalosporins, imipenem, gentamicin, chloramphenicol, sulfamethoxazole–trimethoprim and quinolones, when tested by the agar disc diffusion method according to CLSI (2006) guidelines.

    The patient's condition became worse, and he died on 16 August because of hypovolaemic shock and ventricular fibrillation. The autopsy revealed end-stage kidney disease with haemodialysis-associated amyloidosis and left ureteral stenosis with stent. The gastrointestinal tract showed congestion and only a moderate infiltration of mononuclear inflammatory cells with focal superficial necrosis not directly related to V. cholerae. In addition, shock-related focal myocardial necrosis, diffuse vascular congestion, lung oedema, chronic uraemic pericarditis and endomyocarditis were observed. Culture of all tissues was negative for V. cholerae. Pus collected from the kidney and bladder yielded Salmonella enterica group C.

    The V. cholerae O37 strain was further characterized for different virulence properties by using phenotypic tests and PCR assays to detect the presence of toxin genes (Table 1).

    Table 1.

    Virulence properties of the isolate and the control strains

    Cholera (the duba in Arabic) is caused by V. cholerae strains belonging only to serovars O1 and O139: it is well known in the Mediterranean area where it has been endemic since the 19th century when it was first described in Egypt (1832).

    V. cholerae non-O1 and non-O139 strains have been shown to cause gastroenteritis in humans often presenting with an underlying hepatic, renal or haematological chronic disease. The mortality of non-O1, non-O139 V. cholerae systemic infections is high at 24–62 % (Phetsouvanh et al., 2008).

    The patient in the case presented here had a long history of chronic renal failure and steroid treatment, and may have contracted the non-O1, non-O139 V. cholerae infection from undercooked fish and vegetables consumed along the Nile river.

    Previous studies (Lukinmaa et al., 2006) demonstrated that, in the absence of cholera toxin, toxin co-regulated pilus (tcpA and tcpI), zonula occludens toxin (zot) and non-O1 V. cholerae heat-stable enterotoxin (NAG-ST), non-O1, non-O139 V. cholerae still has the ability to cause diarrhoea by a mechanism entirely different from that of the toxigenic V. cholerae strains. In these cases, haemolysis, cytotoxicity and proteolytic activity were demonstrated to be the putative virulence factors (Iyer et al., 2000; Restrepo et al., 2006).

    In agreement, our isolate, characterized by the genotype hlyAET+ stn/sto ctx tcpA zot hlyA class, was cytotoxic, showed haemolytic and proteolytic activities and was able to colonize the intestine of a suckling mouse. This report confirms that non-toxigenic, non-O1, non-O139, O37 V. cholerae can be isolated from patients suffering from a cholera-like syndrome after consumption of contaminated food.

    Acknowledgments

    We are truly indebted to Eiji Arakawa, Department of Bacteriology, National Institute of Infectious Diseases, Tokyo, Japan, for serotyping.

    References