• View in gallery

    Seasonal isolation of aquatic microorganisms from 2010 to 2017.

  • 1.

    Igbinosa IH, Igumbor EU, Aghdasi F, Tom M, Okoh AI, 2012. Emerging Aeromonas species infections and their significance in public health. ScientificWorldJournal 2012: 625023.

    • Search Google Scholar
    • Export Citation
  • 2.

    Vignier N, Barreau M, Olive C, Baubion E, Théodose R, Hochedez P, Cabié A, 2013. Human infection with Shewanella putrefaciens and S. algae: report of 16 cases in Martinique and review of the literature. Am J Trop Med Hyg 89: 151156.

    • Search Google Scholar
    • Export Citation
  • 3.

    Chen SC, Chan KS, Chao WN, Wang PH, Lin DB, Ueng KC, Kuo SH, Chen CC, Lee MC, 2010. Clinical outcomes and prognostic factors for patients with Vibrio vulnificus infections requiring intensive care: a 10-yr retrospective study. Crit Care Med 38: 19841990.

    • Search Google Scholar
    • Export Citation
  • 4.

    Hoel S, Vadstein O, Jakobsen AN, 2017. Species distribution and prevalence of putative virulence factors in mesophilic Aeromonas spp. isolated from fresh retail sushi. Front Microbiol 8: 931.

    • Search Google Scholar
    • Export Citation
  • 5.

    Araujo VS, Pagliares VA, Queiroz ML, Freitas-Almeida AC, 2002. Occurrence of Staphylococcus and enteropathogens in soft cheese commercialized in the city of Rio de Janeiro, Brazil. J Appl Microbiol 92: 11721177.

    • Search Google Scholar
    • Export Citation
  • 6.

    Praveen PK, Debnath C, Shekhar S, Dalai N, Ganguly S, 2016. Incidence of Aeromonas spp. infection in fish and chicken meat and its related public health hazards: a review. Vet World 9: 611.

    • Search Google Scholar
    • Export Citation
  • 7.

    Klontz KC, Lieb S, Schreiber M, Janowski HT, Baldy LM, Gunn RA, 1988. Syndromes of Vibrio vulnificus infections: clinical and epidemiologic features in Florida cases, 1981–1987. Ann Intern Med 109: 318323.

    • Search Google Scholar
    • Export Citation
  • 8.

    Levine WC, Griffin PM, 1993. Vibrio infections on the Gulf Coast: results of first year of regional surveillance Gulf Coast Vibrio working group. J Infect Dis 167: 479483.

    • Search Google Scholar
    • Export Citation
  • 9.

    Diaz JH, 2014. Skin and soft tissue infections following marine injuries and exposures in travelers. J Travel Med 21: 207213.

  • 10.

    McAuliffe GN, Hennessy J, Baird RW, 2015. Relative frequency, characteristics, and antimicrobial susceptibility patterns of Vibrio spp., Aeromonas spp., Chromobacterium violaceum, and Shewanella spp. in the northern territory of Australia, 2000–2013. Am J Trop Med Hyg 92: 605610.

    • Search Google Scholar
    • Export Citation
  • 11.

    Singer 2016. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA 315: 801810.

  • 12.

    EUCAST, 2017. European Committee on Antimicrobial Susceptibility Testing 2017. Available at: http://www.eucast.org/clinical_breakpoints/.

  • 13.

    Hou CC, Lai CC, Liu WL, Chao CM, Chiu YH, Hsueh PR, 2011. Clinical manifestation and prognostic factors of non-cholerae Vibrio infections. Eur J Clin Microbiol Infect Dis 30: 819824.

    • Search Google Scholar
    • Export Citation
  • 14.

    Lamy B, Kodjo, A, Laurent, F; colBVH Study Group, 2009. Prospective nationwide study of Aeromonas infections in France. J Clin Microbiol 47: 12341237.

    • Search Google Scholar
    • Export Citation
  • 15.

    Qamar FN 2016. Aeromonas-associated diarrhea in children under 5 years: the GEMS experience. Am J Trop Med Hyg 95: 774780.

  • 16.

    Aravena-Román M, Inglis TJ, Henderson B, Riley TV, Chang BJ, 2012. Antimicrobial susceptibilities of Aeromonas strains isolated from clinical and environmental sources to 26 antimicrobial agents. Antimicrob Agents Chemother 56: 11101112.

    • Search Google Scholar
    • Export Citation
  • 17.

    Chen PL, Ko WC, Wu CJ, 2012. Complexity of β-lactamases among clinical Aeromonas isolates and its clinical implications. J Microbiol Immunol Infect 45: 398403.

    • Search Google Scholar
    • Export Citation
  • 18.

    Héritier C, Poirel L, Nordmann P, 2004. Genetic and biochemical characterization of a chromosome-encoded carbapenem-hydrolyzing ambler class D beta-lactamase from Shewanella algae. Antimicrob Agents Chemother 48: 16701675.

    • Search Google Scholar
    • Export Citation
  • 19.

    Kim DM, Kang CI, Lee CS, Kim HB, Kim EC, Kim NJ, Oh MD, Choe KW, 2006. Treatment failure due to emergence of resistance to carbapenem during therapy for Shewanella algae bacteremia. J Clin Microbiol 44: 11721174.

    • Search Google Scholar
    • Export Citation
  • 20.

    Lamy B, Laurent F, Kodjo A, Roger F, Jumas-Bilak E, Marchandin H; colBVH Study Group, 2012. Which antibiotics and breakpoints should be used for Aeromonas susceptibility testing? Considerations from a comparison of agar dilution and disk diffusion methods using Enterobacteriaceae breakpoints. Eur J Clin Microbiol Infect Dis 31: 23692377.

    • Search Google Scholar
    • Export Citation
  • 21.

    Baron S, Granier SA, Larvor E, Jouy E, Cineux M, Wilhelm A, Gassilloud B, Le Bouquin S, Kempf I, Chauvin C, 2017. Aeromonas diversity and antimicrobial susceptibility in freshwater—an attempt to set generic epidemiological cut-off values. Front Microbiol 8: 503.

    • Search Google Scholar
    • Export Citation
  • 22.

    Brulliard C, Traversier N, Allyn J, Schaeffer C, Bouchet B, Allou N, 2017. Case report: disseminated Shewanella algae infection with meningoencephalitis in a traveler secondary to marine injury in Madagascar. Am J Trop Med Hyg 97: 10431044.

    • Search Google Scholar
    • Export Citation
  • 23.

    Wu CJ, Chen PL, Hsueh PR, Chang MC, Tsai PJ, Shih HI, Wang HC, Chou PH, Ko WC, 2015. Clinical implications of species identification in monomicrobial Aeromonas bacteremia. PLoS One 10: e0117821.

    • Search Google Scholar
    • Export Citation
  • 24.

    Wong KC, Brown AM, Luscombe GM, Wong SJ, Mendis K, 2015. Antibiotic use for Vibrio infections: important insights from surveillance data. BMC Infect Dis 15: 226.

    • Search Google Scholar
    • Export Citation
  • 25.

    Holt HM, Gahrn-Hansen B, Bruun B, 2005. Shewanella algae and Shewanella putrefaciens: clinical and microbiological characteristics. Clin Microbiol Infect 11: 347352.

    • Search Google Scholar
    • Export Citation
  • 26.

    Jacob-Kokura S, Chan CY, Kaplan L, 2014. Bacteremia and empyema caused by Shewanella algae in a trauma patient. Ann Pharmacother 48: 128136.

 

 

 

 

Waterborne Infections in Reunion Island, 2010–2017

View More View Less
  • 1 Réanimation Polyvalente, Centre Hospitalier Universitaire Félix Guyon, Allée des Topazes, Saint Denis, France;
  • 2 Bacteriologie, Centre Hospitalier Universitaire Felix Guyon, Allée des Topazes, Saint Denis, France;
  • 3 Bacteriologie, Centre Hospitalier Universitaire Sud Réunion, Saint Pierre, Saint Pierre, France

Gram-negative bacilli Vibrio spp., Aeromonas spp., and Shewanella spp. are a major cause of severe waterborne infection. The aim of this study was to assess the clinical and microbiological characteristics and prognosis of patients hospitalized in Reunion Island for a waterborne infection. This retrospective study was conducted in the two university hospitals of Reunion Island between January 2010 and March 2017. Patients diagnosed with a Vibrio, Aeromonas, or Shewanella infection were evaluated. Over the study period, 112 aquatic strains were isolated at Reunion Island: Aeromonas spp. were found in 91 patients (81.3%), Shewanella spp. in 13 patients (11.6%), and Vibrio spp. in eight patients (7.2%). The in-hospital mortality rate was 11.6%. The main sites of infection were skin and soft tissue (44.6%) and the abdomen (19.6%). Infections were polymicrobial in 70 cases (62.5%). The most commonly prescribed empiric antibiotic regimen was amoxicillin–clavulanate (34.8%). Eighty-four percent of the aquatic strains were resistant to amoxicillin–clavulanate and more than > 95% were susceptible to third or fourth generation cephalosporins and fluoroquinolones. After multivariate analysis, the only independent risk factor of in-hospital mortality was the presence of sepsis (P < 0.0001). In Reunion Island, the most commonly isolated aquatic microorganisms were Aeromonas spp. Sepsis caused by aquatic microorganisms was frequent (> 50%) and associated with higher in-hospital mortality. This study suggests that empiric antibiotic regimens in patients with sepsis or septic shock caused by suspected aquatic microorganisms (tropical climate, skin lesion exposed to seawater…) should include broad-spectrum antibiotics (third or fourth generation cephalosporins).

INTRODUCTION

Gram-negative bacilli Aeromonas spp., Vibrio spp., and Shewanella spp. are the most common cause of waterborne infection. Found naturally in aquatic environments,13 these microorganisms develop mainly in the warm waters of tropical and subtropical regions.13 Of all aquatic bacteria, Aeromonas strains are the most commonly isolated; in some countries, they are detected even in food and running water.1,46 Vibrio and Shewanella strains are less common, and their presence in aquatic environments is favored by salt water.2,7,8 Aeromonas spp., Vibrio spp., and Shewanella spp. are responsible for all kinds of infections, especially skin and soft tissue infections.9 Waterborne infections often occur in immunocompromised patients—for instance, those suffering from diabetes or liver disease, or those undergoing immunosuppressive therapy.13 Few global epidemiological studies on aquatic bacteria have been published to date,10 and local epidemiological studies are usually limited to one type of infection2,3,7,8 and one type of bacteria.9 To fill this void, we present the most common types of waterborne infections in the Indian Ocean region, in particular in Reunion Island, a French overseas territory with a population of 850,000 inhabitants. The aim of this study was to assess the clinical and microbiological characteristics and prognosis of patients with a waterborne infection.

MATERIAL AND METHODS

This observational study was approved by the Institutional Review Board of the Committee of the French Intensive Care Society (CE SRLF17-27) and was declared to the Commission nationale de l’informatique et des liberté (CNIL MR-003, No. 2000694). The need for informed consent was waived because of the observational and retrospective nature of the study.

Selection of the study sample.

This retrospective study was conducted from January 2010 to March 2017 in the two university hospitals of Reunion Island (Félix Guyon and Saint Pierre).

We consecutively screened all hospitalized patients with a positive microbiological sample for Aeromonas spp., Shewanella spp., or Vibrio spp. The exclusion criteria were age < 18 years old and colonization by Aeromonas spp., Shewanella spp., or Vibrio spp. without infection.

Definitions.

Sepsis was defined as infection with an increase in the Sequential Organ Failure Assessment (SOFA) score ≥ 2 points from baseline.11

Septic shock was defined as sepsis requiring vasopressor support to maintain a mean arterial pressure ≥ 65 mm Hg and serum lactate levels ≥ 2 mmol/L in the absence of hypovolemia.11

Data collection.

The following demographic and clinical data were collected from consecutive patients infected with aquatic bacteria: demographic characteristics, site of infection, bacterial species, and death.

Rainfall data.

Rainfall data consisted in the average rainfall of Reunion Island over the 2010–2017 period (Météo France, La Réunion, Saint-Denis, Bureau of Meteorology).

Microbiological investigations.

All microorganism isolates were identifid at the species level using biochemical identification galleries (API® 20E; bioMérieux, Marcy l’Etoile, France) or matrix-assisted laser desorption-ionization time-of-flight (MALDI-TOF) mass spectrometry (Bruker Daltonics, Breme, Germany/bioMérieux, La Balme-les-Grottes, France).

Susceptibility testing and empiric antimicrobial therapy.

When an infection was suspected, empiric antimicrobial therapy based on local guidelines was initiated. The therapy was considered appropriate when all isolated bacteria were susceptible to at least one of the administered drugs after susceptibility testing.

Susceptibility testing was performed using the disk diffusion method based on the criteria of the European Committee on Antimicrobial Susceptibility Testing.12

Statistical analysis.

Results were expressed as total numbers (percentages) for categorical variables and as median (25th–75th percentiles) for continuous variables. Continuous variables were compared using the Mann–Whitney test or the Kruskal–Wallis test, as appropriate. Categorical variables were compared using the χ2 test or the Fisher’s exact test, as appropriate.

Risk factors found to be predictive of in-hospital mortality in the bivariate analysis with P < 0.05 were entered into a multivariate logistic regression analysis using backward selection with a criterion of P < 0.05. Collinearity between independent factors was investigated. When identified, the most clinically relevant factor was chosen for use in the multivariate model. A P value < 0.05 was considered significant. Analysis was performed using SPSS statistical software (8.2, Cary, NC).

RESULTS

Study population.

Over the study period, 156 patients were hospitalized with a positive microbiological sample for Aeromonas spp., Shewanella spp., or Vibrio spp. Among these patients, 44 were excluded (35 were < 18 years old and nine were colonized). Overall, 112 aquatic strains were isolated at Reunion Island in 112 infected patients: The majority of infections occurred in Reunion Island (N = 107, 95.5%). Four infections (3.6%) developed in Madagascar after swimming in salt water and one (0.9%) did so in the Comoros Archipelago (0.9%).

As shown in Figure 1, the largest number of Shewanella infections occurred during the rainy season, between December and June. Vibrio or Aeromonas infections were relatively well distributed throughout the year, with a peak of Aeromonas infections occurring at the end of the rainy season (Figure 1).

Figure 1.
Figure 1.

Seasonal isolation of aquatic microorganisms from 2010 to 2017.

Citation: The American Journal of Tropical Medicine and Hygiene 99, 3; 10.4269/ajtmh.17-0981

Patient characteristics at study inclusion are shown in Table 1. Median age was 59 [46–72] years. The most frequent underlying diseases were diabetes mellitus (32.1%), heavy alcohol use (23.2%), and cancer (22.3%) (Table 1).

Table 1

Clinical characteristics of the 112 patients with waterborne infections

CharacteristicsTotalAeromonas spp.Shewanella spp.Vibrio spp.P value
(N = 112)(N = 91)(N = 13)(N = 8)
Age (years old)59 [46–72]60 [45–73]61[55–69]54 [42–64]0.568
Male gender74 (66.1)58 (63.7)8 (61.5)8 (100)0.108
Water exposure12 (10.7)8 (8.8)3 (23.1)3 (37.5)0.39
Underlying condition
 Body mass index > 27 kg/m214 (12.5)11 (12.1)2 (15.4)1 (12.5)0.945
 Heavy alcohol use26 (23.2)22 (24.2)2 (15.4)2 (25)0.775
 Biliary tract disease12 (10.7)12 (13.2)000.212
 Liver cirrhosis6 (5.4)6 (6.6)000.481
 Chronic respiratory failure7 (6.3)5 (5.5)1 (7.7)1 (12.5)0.716
 Cancer (< 3 months)25 (22.3)25 (27.5)000.024
 Diabetes mellitus36 (32.1)32 (35.2)3 (23.1)1 (12.5)0.23
 Immunodepression12 (10.7)12 (13.2)000.212
 Corticosteroid treatment6 (5.4)5 (5.5)01 (12.5)0.462
Organ failure
 Sepsis*60 (53.6)50 (54.9)8 (61.5)2 (25)0.22
 ICU admission29 (25.9)23 (25.3)4 (30.8)2 (25)0.903
 Catecholamines22 (19.6)17 (18.7)3 (23.1)2 (25)0.862
 Renal replacement therapy9 (8)6 (6.6)2 (15.4)1 (12.5)0.491
 Creatinine level (μmol/L)97 [70–168]97 [69–167]103 [63–264]93 [78–116]0.898
 Platelet count (G/L)194 [106–283]177 [102–274]222 [166–310]232 [198–312]0.244
 Leukocyte count (G/L)11.6 [6.7–16.3]11.9 [6.8–16.7]8.6 [6.4–12.3]7.4 [6.9–16.4]0.442
 Neutrophil count (G/L)9.1 [4.7–13.6]9.3 [5–14.1]8 [4.5–11.1]4.8 [3.7–13.9]0.571
 Glasgow Coma Scale score15 [15–15]15 [15–15]15 [11–15]15 [12–15]0.308
 Mechanical ventilation22 (19.6)17 (19.7)3 (23.1)2 (25)0.862
 Total bilirubin level (mg/dl)17 [9–41]20 [10–45]11 [8–48]9.5 [9–10]0.127
 Conjugated bilirubin level (μmol/L)9 [4–28]11 [5–33]6 [5–13]4 [2–5]0.038
 Prothrombin time (%)74 [55–87]73 [54–87]68 [59–79]84 [83–85]0.739
 C-reactive protein (mg/dl)127 [63–256]97 [69–234]318 [178–329]74 [41–119]0.039

ICU = intensive care unit; SOFA = sequential organ failure assessment. Results are expressed as median [25th–75th percentiles] or n (%), as appropriate.

Increase of SOFA score ≥ 2 points from baseline.

Distribution of microorganisms and sites of infection.

Most strains were identified phenotypically by extensive biochemical testing. The most commonly isolated strains were Aeromonas spp. in 91 cases (81.3%): Aeromonas hydrophila (54 strains), Aeromonas caviae (29 strains), and Aeromonas sobria (four strains). Shewanella spp. were isolated in 13 patients (11.6%): Shewanella putrefaciens (11 strains) and Shewanella algae (two strains). Vibrio spp. were isolated in eight patients (7.1%): Vibrio alginolyticus (five strains), Vibrio parahaemolyticus (one strain), Vibrio fluvialis (one stain), and Vibrio vulnificus (one strain).

The main sites of infection were skin and soft tissue in 50 cases (44.6%) and the abdomen in 22 cases (19.6%). Bacteremia was present in 39 cases (34.8%) (Table 2).

Table 2

Sites of infection

Sites of infectionTotalAeromonas spp.Shewanella spp.Vibrio spp.P value
(N = 112)(N = 91)(N = 13)(N = 8)
Pulmonary9 (8)4 (4.4)3 (23.1)2 (25)0.013
Catheter4 (3.6)4 (4.4)000.62
Skin and soft tissue50 (44.6)40 (44)6 (46.2)4 (50)0.941
Urinary tract8 (7)7 (7.7)01 (12.5)0.5
Abdominal or biliary tract22 (20)20 (22)2 (15.4)00.298
Bacteremia39 (32)32 (35.2)4 (30.8)3 (37.5)0.94
Bone7 (5)5 (5.5)1 (7.7)1 (12.5)0.716
Meningitis1 (0.9)0 (0)1 (7.7)0
Feces1 (0.9)1 (1.1)000.89

Results are expressed as number (%).

Infections were polymicrobial in 70 cases (62.5%). They were more frequently polymicrobial in patients with an Aeromonas infection (63.7%) or a Shewanella infection (76.9%) than in patients with a Vibrio infection (25%, P = 0.049).

A total of 126 co-pathogens were cultured (Table 3). The most commonly isolated co-pathogens were Enterobacteriaceae (50.9%), enterococci (16.1%), and Staphylococcus aureus (14.3%). No significant difference was found between the three groups of patients with respect to the cultured co-pathogens, with the exception of Enterobacteriaceae (P = 0.01).

Table 3

Presence of co-pathogens

MicroorganismsTotalAeromonas spp.Shewanella spp.Vibrio spp.P value
(N = 112)(N = 91)(N = 13)(N = 8)
Staphylococcus aureus16 (14.3)12 (13.1)3 (23.1)1 (12.5)0.628
Streptococci6 (5.4)5 (5.5)1 (7.7)00.742
Non-fermenting Gram-negative bacilli12 (10.7)9 (9.9)3 (23.1)00.212
Enterobacteriaceae57 (50.9)49 (53.8)8 (61.5)00.01
Anaerobes8 (7.1)6 (6.6)2 (15.4)00.37
Enterococci18 (16.1)13 (14.3)4 (30.8)1 (12.5)0.305
Fungi4 (3.6)4 (4.4)000.620
Other5 (4.5)3 (3.3)2 (15.4)00.117

Results are expressed as n (%).

Empiric antibiotic therapy and susceptibility of microorganisms.

In the study cohort, more than 95% of aquatic strains were susceptible to a third or fourth generation cephalosporin, 94.3% to piperacillin/tazobactam, 96.2% to ciprofloxacin, and 96.3% to meropenem—with no significant difference between Aeromonas spp., Shewanella spp., and Vibrio spp. (Table 4). Susceptibility to amoxicillin–clavulanate was 15.9% and was higher in the Vibrio spp. group than in the other two groups (P < 0.0001) (Table 4).

Table 4

Antibiotic susceptibility of clinical Aeromonas spp., Shewanella spp., and Vibrio spp. isolates

TotalAeromonas spp.Shewanella spp.Vibrio spp.P value
(N = 112)(N = 91)(N = 13)(N = 8)
Amikacin111/111 (100)90/90 (100)13/13 (100)8/8 (100)0.89
Ampicillin3/46 (6.5)0/38 (0)0/1 (0)3/7 (42.9)< 0.0001
Amoxicillin–clavulanate7/44 (15.9)2/37 (5.4)0/1 (0)5/6 (83.3)< 0.0001
Aztreonam63/88 (71.6)52/72 (72.2)7/12 (58.3)4/4 (100)0.267
Cefotaxim46/47 (97.9)38/39 (97.4)1/1 (100)7/7 (100)0.901
Ceftazidime106/111 (95.5)85/90 (97.4)13/13 (100)8/8 (100)0.543
Ceftriaxone25/26 (96.2)20/21 (95.2)0/05/5 (100)0.248
Cotrimoxazole90/100 (90)77/86 (89.5)8/9 (88.9)5/5 (100)0.745
Ciprofloxacin102/106 (96.2)83/86 (96.5)12/13 (92.3)7/7 (100)0.656
Cefepime89/90 (98.9)72/73 (98.6)10/10 (100)7/7 (100)0.889
Colistin28/33 (84.8)21/25 (84)2/2 (100)5/6 (83.3)0.826
Fosfomycin41/53 (77.4)37/40 (92.5)0/8 (0)4/5 (80)< 0.0001
Gentamicin108/110 (98.2)88/89 (98.9)12/13 (92.3)8/8 (100)0.234
Imipenem98/107 (91.6)79/87 (90.8)11/12 (91.7)8/8 (100)0.669
Meropenem52/54 (96.3)42/43 (97.7)8/9 (88.9)2/2 (100)0.429
Ofloxacin41/43 (95.3)36/37 (97.3)1/1 (100)4/5 (80)0.221
Piperacillin60/84 (71.4)43/63 (68.3)11/13 (84.6)6/8 (75)0.48
Piperacillin–tazobactam99/105 (94.3)81/87 (93.1)13/13 (100)5/5 (100)0.518
Ticarcillin15/111 (13.5)2/91 (2.2)11/13 (84.6)2/7 (28.6)< 0.0001
Ticarcillin–clavulanate56/100 (56)41/82 (50)11/13 (84.6)4/5 (80)0.035
Tobramycin97/100 (97)82/84 (97.6)12/12 (100)3/4 (75)0.028

Results are expressed as no. positive/no. tested (%).

The rate of appropriate empiric antimicrobial therapy was 54.8%.

The most commonly prescribed empiric antibiotic regimen was amoxicillin–clavulanate (34.8%). Third generation cephalosporins, piperacillin/tazobactam, and carbapenems were prescribed in 24 patients (21.4%), 21 patients (18.8%), and two patients (1.8%), respectively. A combination therapy with either aminoglycosides, fluoroquinolones, or vancomycin was used in 24 patients (21.4%), seven patients (6.3%), and two patients (1.8%), respectively.

Prognosis.

Sixty patients had sepsis (53.6%), 22 had septic shock (19.6%), and 29 were hospitalized in intensive care unit (25.9%), with no significant difference between the three groups of microorganisms.

In-hospital mortality was 11.6%, with no significant difference between the three groups of patients (P = 0.895). After univariate analysis, the factors associated with in-hospital mortality were the following: male gender (P = 0.025), SOFA score ≥ 2 (P = 0.0004), intensive care unit admission (P < 0.0001), catecholamines (P < 0.0001), renal replacement therapy (P = 0.001), higher lactatemia (P = 0.004), lower platelet count (P = 0.003), lower Glasgow Coma Scale score (P = 0.004), mechanical ventilation (P < 0.0001), higher bilirubin level (P = 0.002), and bacteremia (P = 0.03).

After multivariate analysis, SOFA score ≥ 2 (OR: 7.5; 95% confidence interval [CI]: 1.8–37.1; P < 0.0001) was the only independent risk factor of in-hospital mortality. The Hosmer–Lemeshow goodness-of-fit test P value was 0.76. The Nagelkerke and Cox/Snell R squares were 0.332 and 0.17, respectively.

DISCUSSION

To date, only one study, conducted in Australia, has examined the epidemiology of waterborne infections.10 In our study of waterborne infections in Reunion Island, the main findings were as follows: 1) the aquatic bacteria most likely to cause infection are Aeromonas spp., 2) waterborne infections are often severe (> 50%), and 3) amoxicillin–clavulanate is the most commonly prescribed probabilistic antibiotic treatment (35%), but it is often ineffective on aquatic bacteria (84%).

The most frequently isolated aquatic bacteria in our study were Aeromonas spp., a finding that was also reported by Mc Auliffe et al.10 Comorbidities such as diabetes, alcoholism, neoplasia, and immunosuppression were common in our patients, as has been observed elsewhere.2,3,9,10

In our series, 38% of patients with a Vibrio infection had a cutaneous entry point after contact with seawater. This is a common finding in studies that specifically evaluate non-cholera Vibrio infections.13 Skin and soft tissue infections were the most common in our series, with an incidence comparable to that found in the study by Lamy et al.,14 which focused specifically on Aeromonas infections (44%). However, the incidence of skin and soft tissue infection was significantly lower than that reported by Mc Auliffe et al. (80%).11

Whereas the very low incidence of waterborne gastroenteritis in our study (< 1%) is similar to that found by Mc Auliffe et al.10 (5.4%), it is much lower than that reported by Lamy et al.14 (19%). This low incidence may be because of the fact that unlike the situation for children, the disease is not usually severe in adults, and hence rarely leads to hospitalization.15

In sensitivity analysis, only a small proportion of the isolated strains were sensitive to amoxicillin–clavulanate (16%), which makes sense given that the majority of these strains were Aeromonas spp.14,16 However, this poses a therapeutic problem because the most of the waterborne infections are skin and soft tissue infections, these being often treated probabilistically with amoxicillin–clavulanate. In addition, waterborne infections are most often polymicrobial, and there are no specific signs associated with aquatic bacteria.2,10 In cases of severe waterborne skin and soft tissue infection, antibiotic therapy should include broad-spectrum antibiotics such as third or fourth generation cephalosporins, which are effective on these bacteria in the vast majority of cases. In fact, antibiotic therapy may require a combination of molecules because infections are often polymicrobial, but also because Aeromonas strains can produce AmpC β-lactamase, which has been shown to occur frequently.17

Shewanella strains are most often susceptible to broad-spectrum antibiotics, such as third or fourth generation cephalosporins and carbapenems. It should be noted, however, that these strains can produce oxacillinase, which makes them resistant to carbapenems18 and may, therefore, lead to therapeutic failure.19 The study by Wong et al. suggests that fluoroquinolones can lead to a decrease in mortality from Vibrio infections.20 In the case of patients with marine injury with sepsis or septic shock, the use of fluoroquinolone should be discussed. Cefepime may be an interesting molecule for the treatment of Aeromonas infections. Indeed, cefepime is little affected by cephalosporinase, and minimal inhibitory concentrations of the molecule required for the treatment of Aeromonas and Shewanella infections are very low.2022

Mc Auliffe et al. did not evaluate the prognosis of patients in their study. By contrast, we found that waterborne infections are often severe: sepsis was present in more than 50% of the examined cases, often leading to hospitalization in intensive care (25.9%), with an in-hospital mortality rate of 11.6%. The different causative organisms did not seem to be associated with significantly different virulence. Nevertheless, Wu et al.23 have found that mortality is higher with Aeromonas dhakensis strains than with other Aeromonas strains. Similarly, mortality has been found to be higher with Vibrio vulinificius strains than with other Vibrio strains,24 and S. algae strains have been shown to be more virulent than S. putrefaciens strains.25

This study has many limitations. Its retrospective nature is clearly a weakness. Moreover, the number of infections is relatively low, which limits the power of statistical analysis. The typing of the strains may be criticized because in most of the cases it was performed phenotypically, and hence without genetic analysis. Indeed, it has been shown that many strains identified as S. putrefaciens by MALDI-TOF are identified as S. algae after analysis of 16S RNA.26

This study was performed on a tropical island in the Indian Ocean and cannot provide any definitive conclusions for other much larger island continents, such as Australia, and other tropical regions.

CONCLUSION

In Reunion Island, Aeromonas spp. were the most commonly isolated aquatic microorganisms. Infections were often severe and associated with sepsis in > 50% of cases. Amoxicillin–clavulanate was the most frequently prescribed antibiotic, but was often ineffective on aquatic microorganisms. This study suggests that empiric antibiotic regimens in patients with sepsis or septic shock caused by suspected aquatic microorganisms (tropical climate and skin lesion exposed to seawater) should include broad-spectrum antibiotics such as third or fourth generation cephalosporins.

REFERENCES

  • 1.

    Igbinosa IH, Igumbor EU, Aghdasi F, Tom M, Okoh AI, 2012. Emerging Aeromonas species infections and their significance in public health. ScientificWorldJournal 2012: 625023.

    • Search Google Scholar
    • Export Citation
  • 2.

    Vignier N, Barreau M, Olive C, Baubion E, Théodose R, Hochedez P, Cabié A, 2013. Human infection with Shewanella putrefaciens and S. algae: report of 16 cases in Martinique and review of the literature. Am J Trop Med Hyg 89: 151156.

    • Search Google Scholar
    • Export Citation
  • 3.

    Chen SC, Chan KS, Chao WN, Wang PH, Lin DB, Ueng KC, Kuo SH, Chen CC, Lee MC, 2010. Clinical outcomes and prognostic factors for patients with Vibrio vulnificus infections requiring intensive care: a 10-yr retrospective study. Crit Care Med 38: 19841990.

    • Search Google Scholar
    • Export Citation
  • 4.

    Hoel S, Vadstein O, Jakobsen AN, 2017. Species distribution and prevalence of putative virulence factors in mesophilic Aeromonas spp. isolated from fresh retail sushi. Front Microbiol 8: 931.

    • Search Google Scholar
    • Export Citation
  • 5.

    Araujo VS, Pagliares VA, Queiroz ML, Freitas-Almeida AC, 2002. Occurrence of Staphylococcus and enteropathogens in soft cheese commercialized in the city of Rio de Janeiro, Brazil. J Appl Microbiol 92: 11721177.

    • Search Google Scholar
    • Export Citation
  • 6.

    Praveen PK, Debnath C, Shekhar S, Dalai N, Ganguly S, 2016. Incidence of Aeromonas spp. infection in fish and chicken meat and its related public health hazards: a review. Vet World 9: 611.

    • Search Google Scholar
    • Export Citation
  • 7.

    Klontz KC, Lieb S, Schreiber M, Janowski HT, Baldy LM, Gunn RA, 1988. Syndromes of Vibrio vulnificus infections: clinical and epidemiologic features in Florida cases, 1981–1987. Ann Intern Med 109: 318323.

    • Search Google Scholar
    • Export Citation
  • 8.

    Levine WC, Griffin PM, 1993. Vibrio infections on the Gulf Coast: results of first year of regional surveillance Gulf Coast Vibrio working group. J Infect Dis 167: 479483.

    • Search Google Scholar
    • Export Citation
  • 9.

    Diaz JH, 2014. Skin and soft tissue infections following marine injuries and exposures in travelers. J Travel Med 21: 207213.

  • 10.

    McAuliffe GN, Hennessy J, Baird RW, 2015. Relative frequency, characteristics, and antimicrobial susceptibility patterns of Vibrio spp., Aeromonas spp., Chromobacterium violaceum, and Shewanella spp. in the northern territory of Australia, 2000–2013. Am J Trop Med Hyg 92: 605610.

    • Search Google Scholar
    • Export Citation
  • 11.

    Singer 2016. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA 315: 801810.

  • 12.

    EUCAST, 2017. European Committee on Antimicrobial Susceptibility Testing 2017. Available at: http://www.eucast.org/clinical_breakpoints/.

  • 13.

    Hou CC, Lai CC, Liu WL, Chao CM, Chiu YH, Hsueh PR, 2011. Clinical manifestation and prognostic factors of non-cholerae Vibrio infections. Eur J Clin Microbiol Infect Dis 30: 819824.

    • Search Google Scholar
    • Export Citation
  • 14.

    Lamy B, Kodjo, A, Laurent, F; colBVH Study Group, 2009. Prospective nationwide study of Aeromonas infections in France. J Clin Microbiol 47: 12341237.

    • Search Google Scholar
    • Export Citation
  • 15.

    Qamar FN 2016. Aeromonas-associated diarrhea in children under 5 years: the GEMS experience. Am J Trop Med Hyg 95: 774780.

  • 16.

    Aravena-Román M, Inglis TJ, Henderson B, Riley TV, Chang BJ, 2012. Antimicrobial susceptibilities of Aeromonas strains isolated from clinical and environmental sources to 26 antimicrobial agents. Antimicrob Agents Chemother 56: 11101112.

    • Search Google Scholar
    • Export Citation
  • 17.

    Chen PL, Ko WC, Wu CJ, 2012. Complexity of β-lactamases among clinical Aeromonas isolates and its clinical implications. J Microbiol Immunol Infect 45: 398403.

    • Search Google Scholar
    • Export Citation
  • 18.

    Héritier C, Poirel L, Nordmann P, 2004. Genetic and biochemical characterization of a chromosome-encoded carbapenem-hydrolyzing ambler class D beta-lactamase from Shewanella algae. Antimicrob Agents Chemother 48: 16701675.

    • Search Google Scholar
    • Export Citation
  • 19.

    Kim DM, Kang CI, Lee CS, Kim HB, Kim EC, Kim NJ, Oh MD, Choe KW, 2006. Treatment failure due to emergence of resistance to carbapenem during therapy for Shewanella algae bacteremia. J Clin Microbiol 44: 11721174.

    • Search Google Scholar
    • Export Citation
  • 20.

    Lamy B, Laurent F, Kodjo A, Roger F, Jumas-Bilak E, Marchandin H; colBVH Study Group, 2012. Which antibiotics and breakpoints should be used for Aeromonas susceptibility testing? Considerations from a comparison of agar dilution and disk diffusion methods using Enterobacteriaceae breakpoints. Eur J Clin Microbiol Infect Dis 31: 23692377.

    • Search Google Scholar
    • Export Citation
  • 21.

    Baron S, Granier SA, Larvor E, Jouy E, Cineux M, Wilhelm A, Gassilloud B, Le Bouquin S, Kempf I, Chauvin C, 2017. Aeromonas diversity and antimicrobial susceptibility in freshwater—an attempt to set generic epidemiological cut-off values. Front Microbiol 8: 503.

    • Search Google Scholar
    • Export Citation
  • 22.

    Brulliard C, Traversier N, Allyn J, Schaeffer C, Bouchet B, Allou N, 2017. Case report: disseminated Shewanella algae infection with meningoencephalitis in a traveler secondary to marine injury in Madagascar. Am J Trop Med Hyg 97: 10431044.

    • Search Google Scholar
    • Export Citation
  • 23.

    Wu CJ, Chen PL, Hsueh PR, Chang MC, Tsai PJ, Shih HI, Wang HC, Chou PH, Ko WC, 2015. Clinical implications of species identification in monomicrobial Aeromonas bacteremia. PLoS One 10: e0117821.

    • Search Google Scholar
    • Export Citation
  • 24.

    Wong KC, Brown AM, Luscombe GM, Wong SJ, Mendis K, 2015. Antibiotic use for Vibrio infections: important insights from surveillance data. BMC Infect Dis 15: 226.

    • Search Google Scholar
    • Export Citation
  • 25.

    Holt HM, Gahrn-Hansen B, Bruun B, 2005. Shewanella algae and Shewanella putrefaciens: clinical and microbiological characteristics. Clin Microbiol Infect 11: 347352.

    • Search Google Scholar
    • Export Citation
  • 26.

    Jacob-Kokura S, Chan CY, Kaplan L, 2014. Bacteremia and empyema caused by Shewanella algae in a trauma patient. Ann Pharmacother 48: 128136.

Author Notes

Address correspondence to Nicolas Allou, Hôpital Félix Guyon, Réanimation Polyvalente, Bellepierre, Saint-Denis 97405, France. E-mail: nicolas.allou@chu-reunion.fr

Financial support: This work was internally funded.

Authors’ addresses: Nicolas Allou, Aurélien Soubeyrand, Romain Persichini, Caroline Brulliard, Dorothée Valance, Olivier Martinet, and Jérôme Allyn, Réanimation Polyvalente, Centre Hospitalier Universitaire Félix Guyon, Allée des Topazes, Saint Denis, France, E-mails: nicolas.allou@hotmail.fr, aurelien.soubeyrand@chu-reunion.fr, romain.persichini@chu-reunion.fr, caroline.brulliard@chu-reunion.fr, dorothee.valance@chu-reunion.fr, olivier.martinet@chu-reunion.fr, and jerome.allyn@chu-reunion.fr. Nicolas Traversier and Olivier Belmonte, Bacteriologie, Centre Hospitalier Universitaire Felix Guyon, Allée des Topazes, Saint Denis, France, E-mails: nicolas.traversier@chu-reunion.fr and olivier.belmonte@chu-reunion.fr. Sandrine Picot, Bacteriologie, Centre Hospitalier Universitaire Sud Réunion, Saint Pierre, Saint Pierre, France, E-mail: sandrine.picot@chu-reunion.fr.

Save