Wound Infections in Repatriated Survivors of the Tsunami Disaster

On December 26, 2004, a strong earthquake off the west coast of northern Sumatra triggered a devastating tsunami that hit 8 countries in south Asia, southeast Asia, and east Africa. The tsunami disaster caused more than 280,000 victims, and many of the survivors subsequently died from infection. Measles, diarrhea, viral hepatitis, acute respiratory infections, malaria, and malnutrition are the main causes of death in complex emergencies.^1,2 Usually, epidemics of other infectious diseases are expected,^1,2 but in this case, only a few cases (clusters) of various communicable diseases were reported, and no major outbreaks were reported.³ Patients who survived the disaster were often severely injured, and the incidence of wound infections was high.^4,5 Several reports appeared in the medical literature, on the Internet, and in the lay press of patients being at risk of deadly fungal infections and infections caused by waterborne and highly resistant pathogens.^6–8 These reports were based on 1 or 2 repatriated patients and raw data from the Thai Ministry of Public Health regarding the results of cultures taken from 33 patients in Phuket during a surveillance program.⁹ A report on a group of 17 victims repatriated to Germany and a recent article about a large retrospective survey in Thailand provided more detailed information on culture results and resistance patterns in larger groups of tsunami victims.^8,10
In the Netherlands, a Major Incident Hospital was established in the early 1990s during the first Gulf War. The facility is also used as a quarantine unit for repatriated groups of injured Dutch patients as well as a first response unit for (pan)epidemic outbreaks of infectious diseases. On January 1, 2005, 5 days after the tsunami disaster, 23 Dutch tsunami victims were admitted to this hospital. In this report, the authors describe the wound infections, causative organisms, and resistance patterns found in these patients with 6- to 8-month clinical follow-up.

Patients and Methods

On January 1, 2005, 5 days after the tsunami disaster, 23 Dutch tsunami victims were collectively repatriated from 7 hospitals in Bangkok and Phuket, Thailand. They were admitted to the Major Incident Hospital in Utrecht for evaluation and further treatment.
On admission, all wounds and other injuries were systematically reviewed and documented along with previously received treatment. Superficial wounds were defined as wounds that did not involve or penetrate the muscular fascia and therefore involved strictly skin and subcutaneous tissue. Deep wounds were defined as wounds that penetrated these barriers. Diagnosis of clinical wound infection was judged by a physician based on the presence of 1 or more signs, such as pus, foul smell, necrosis, gangrene/subcutaneous emphysema, cellulitis, erythema, swelling or infiltration of adjacent tissue, pain, and delayed healing. Three sets of culture swabs were taken from all wounds and 3 sets of surveillance cultures were taken from the pharynx, perineum, and nose; 1 rectal swab was taken. Urine, blood, or sputum samples were taken only when indicated. Specimens were cultured and tested for antimicrobial susceptibility according to 2003 National Committee Clinical Laboratory Standards (NCCLS)^11,12 using an automated method for bacterial identification (PHOENIX™ Automated Microbiology System, BD Diagnostics, Franklin Lakes, NJ). In some patients, additional E-tests were performed to confirm antibiotic resistance, especially to aminoglycosides. Cultured microorganisms and resistance patterns were evaluated in relation to the clinical picture, current as well as previous antibiotic treatment, and coexisting conditions.
The repatriated patients were treated in quarantine to prevent transmission of multi-resistant microorganisms (eg, methicillin-resistant Staphylococcus aureus [MRSA]), if present, to other patients or healthcare workers. All treatment, including surgical procedures, was performed in separate facilities within the Major Incident Hospital.

Results

Fourteen men and 9 women with a mean age of 41 years (range 16–64) were admitted. Five patients had relevant comorbidity or preexisting conditions, namely, diabetes, pregnancy, cystic fibrosis, use of prednisone, and atrophy of the injured limb due to a congenital pes varus (clubfoot).
Wounds, other diagnoses, and treatment prescribed in Thailand are summarized in Table 1. Nearly all patients had superficial lacerations and cuts/scratches, predominantly on the extremities. Nineteen patients had also sustained more extensive wounds of the skin and subcutis (18 patients) and/or deep injuries penetrating the fascia (13 patients). Clinically, 13 patients had clearly infected wounds on admission. Nine patients had already undergone surgical debridement in Thailand. In 4 additional patients, wounds had been opened after suturing (2 patients) or superficially debrided on the ward (3 patients). Eight patients needed further surgical wound debridement in the hospital, 2 of whom had not been operated on in Thailand.
Nineteen of the 23 repatriated patients had been treated in Thailand with antibiotics for 2 to 5 days for aspiration pneumonia (n = 2), open femur fracture (n = 1), and wound infections (n = 16). In 8 of these patients, antibiotic treatment could be discontinued on admission in the Netherlands. The results of the wound cultures are presented in Tables 2 and 3. Ten patients had no positive wound cultures. Cultures for the other 13 patients revealed bacteria and resistance patterns that differed from those usually found in wound infections in the Netherlands.¹³ Most cultures were polymicrobial. Gram-negative and waterborne bacteria dominated the spectrum. Few cultures were positive for Staphylococcus aureus, especially those from patients treated with either oxacillin or clindamycin in Thailand. Only 1 patient still had S aureus-positive cultures, and 1 other patient appeared to have methicillin-resistant S aureus (MRSA).
The MRSA surveillance cultures initially detected 1 patient whose nasal culture was MRSA positive. One patient was negative for MRSA in the 3-fold inventory cultures but became MRSA positive 3 days later after her antibiotic treatment was terminated. Although both patients were repatriated from the same Thai hospital, they were infected with different strains of S aureus.
Rectal swabs revealed several multi-drug-resistant gram-negative bacteria that differed in their resistance to those detected in the Dutch surveillance inventory (SWAB),¹³ such as multi-drug-resistant E coli, Achromobacter species, Acinetobacter species, and Proteus species (Table 3). The resistance of E coli to aminoglycosides in the inventory cultures was higher than expected, with mean E-test values for gentamicin of 256 mg/L (range 16–1024 mg/L) and tobramycin 64 mg/L (range 0.25–256 mg/L).
After 5 days, 17 patients were discharged to home, and the remaining 6 patients were transferred to other facilities (1 nursing home and 5 hospitals) where they were kept in quarantine during the processing of culture results in order to clear the Major Incident Hospital. During this time, no complications were observed in the patients. All responded well to treatment, and the wound infections had not deteriorated. Fifteen patients attended a follow-up assessment 6–8 months later. By this time, all wounds had healed, although some slowly. No adverse events were reported.

Discussion

In this report, the authors described the wounds, wound infections, and microbiological pattern of wound cultures of repatriated Dutch tsunami victims. The main findings were that the wounds of the tsunami victims were infected with polymicrobial, multi-drug-resistant gram-negative microorganisms different from those typically found in infected wounds in the Netherlands (Tables 2 and 3). The pattern of pathogens probably reflects how the injuries were caused, the difficulty in providing optimal wound care under the extraordinary circumstances, and the selection of microorganisms as a result of antibiotic usage.
The etiology of the injuries is different from that of the average non-tsunami trauma patient. Most tsunami victims were injured by debris and polluted seawater, which contributed to wound contamination and further tissue damage. Bacteria, such as Pseudomonas, Aeromonas, and Shewanella, were probably washed into wounds. Of note, no infections caused by Vibrio or Burkholderia species were encountered.^6,14–19 The large number of enteric pathogens detected suggests either surface contamination by seawater or true polymicrobial infection.
Aeromonas hydrophila is present in salt, brackish, and fresh water and has previously been reported to cause rapidly progressing wound infections.²⁰ In a Thai surveillance program, Aeromonas hydrophila was recovered in 2 of 33 patients.⁹ In the retrospective large series conducted in southern Thailand, Aeromonas species were the most commonly detected bacteria and were isolated from 22% of the patients.¹⁰ In the authors’ group, 3 of 23 patients tested positive for Aeromonas species. One patient with Aeromonas hydrophila had a clinically evident wound infection on admission 5 days after the tsunami despite prior surgical debridement and antibiotic therapy. Repeated surgical debridement and antibiotic treatment finally resulted in complete wound healing by secondary intention.
Shewanella putrefaciens can be found in seawater, fresh water, fish, and soil. It is not usually found in human wounds, although incidental case reports have been published.¹⁶Shewanella putrefaciens was not cultured in the survey in Thailand,¹⁰ but the pathogen was cultured from a patient with an infected heel wound in the present study. This patient had initially been treated with amoxicillin-clavulanate, to which this strain was sensitive, but the wound had been sutured in Thailand after initial cleaning. The wound was opened and surgical debridement was performed in the airplane, and subsequent open-wound treatment without antibiotic support at the authors’ center cleared the infection. This emphasizes the need for adequate surgical treatment and the relatively adjuvant role of antibiotics in treatment of infected wounds.
Aside from the environmental factors present at the time of injury, surgical wound management plays a key role in the development of wound infections. The results of the authors’ cultures, taken on admission in the Netherlands, are definitely influenced by prior surgical and antibiotic treatment. The initial flood of victims was enormous and had to be managed under extraordinary circumstances with too few staff and insufficient surgical equipment and antibiotics,^4,5,9,21,22 which posed challenges to early medical treatment. Survivors mainly suffered from aspiration or trauma (eg, wounds, fractures, head injuries), and many needed surgical treatment. Occasionally, wounds were sutured despite contamination. Many patients rapidly developed foul-smelling wounds with frank pus, and surgical treatment may have been suboptimal under these harsh conditions.^4,5,9,21,22 Empirical antibiotic treatment was given without knowledge of the infecting species or antibiotic resistance, and availability may have played a role in the selection of the antibiotics that were used. A recent article discussed the pathogens expected in infected wounds according to the type of exposure and proposed a framework for choosing empiric antibiotics.²³ However, the wounds of the patients in the present study were of mixed etiology and exposure. Skin contamination as well as exposure to seawater, fresh water, and soil played a role.
Although S aureus is the most common microorganism that causes wound infections,²⁴ relatively few patients in the present study were infected with this pathogen. This could be because most patients had already been treated with oxacillin or clindamycin prior to admission to the Major Incident Hospital, and therefore, selection of remaining microorganisms might have occurred. Patients treated with a third-generation cephalosporin had fewer wound infections, but because there were few patients, this observation should be interpreted with caution. Empirical treatment with a third- or fourth-generation cephalosporin, such as carbapenem or piperacillin with tazobactam, appears to be the preferred treatment, as most microorganisms collected from wound swabs were sensitive to these antibiotics (Table 3). However, these results might be influenced by previous treatment, as some pathogens would have been selected by the previous antibiotic treatment in Thailand. On the basis of the culture results, treatment with only beta-lactam antibiotics, second-generation cephalosporins, and fluoroquinolones appears to be inadequate. Third-generation cephalosporins are effective against most of the pathogens isolated except Pseudomonas species, whereas fourth-generation cephalosporins are not effective against some gram-positive pathogens, such as S aureus, which still is an important causative agent in normal wound infections. Carbapenem and piperacillin with tazobactam appeared to be effective against most causative pathogens and would be sufficient for empirical treatment.
Most wounds were infected by several pathogens. Given the large number of enteric pathogens isolated, it is difficult to say whether this reflects surface contamination by tsunami water or true polymicrobial infection. Polymicrobial infections are frequently seen in traumatic injuries of different causes.^25,26 Because of the synergistic nature and necrotizing tendency, polymicrobial wound infections tend to be severe.²⁴ Indeed, all patients infected with multiple pathogens had infected wounds on their arrival in the Netherlands. The authors’ observation of polymicrobial infections is consistent with the Thai results, where 72% of the wound infections were polymicrobial.^9,10 However, in these studies, no surveillance cultures were done, so the relation to clinical wound infection is not clear.
The authors of the present study also found that gram-negative bacteria had different resistance patterns from those usually observed in the Netherlands.¹³ Additionally, the cultures from rectal swabs revealed several multi-drug-resistant gram-negative bacteria that differed in their resistance from those of the Dutch surveillance inventory, such as multi-resistant E coli, Achromobacter species, Acinetobacter species, and Proteus species. This might reflect colonization by local Thai bacteria, influenced by local antibiotic policies. However, many of these isolated organisms are innately resistant to antibiotics when isolated from their natural environment. This may occur partly because some of the molecules that are in antibiotics also occur in nature. Therefore, resistance patterns cannot be attributed to antibiotic use only.
In general, approximately 18.2% of all patients repatriated to the Netherlands were shown to be carriers of drug-resistant microorganisms.²⁷ Travel from Asia is a risk factor for carrier-state of resistant gram-negative pathogens. Additionally, the antibiotics given to these patients may, in part, have caused this resistance. However, a previous Dutch analysis showed that the country from which a patient is repatriated is more significant than the antimicrobial treatment received by the individual patient in a foreign hospital for the acquisition of multi-drug-resistant gram-negative microorganisms.²⁷ The observation that a patient who initially tested MRSA negative and subsequently became MRSA positive after stopping vancomycin treatment may have consequences for MRSA quarantine policy in repatriated patients. In the Netherlands, the incidence of MRSA is low (< 1%) whereas approximately 4.7% of patients repatriated to a Dutch hospital are carriers of MRSA.²⁷

Conclusion

Wound infections in tsunami victims show a different spectrum of pathogens from what is normally encountered. A combination of aerobic-anaerobic and waterborne bacteria and a high prevalence of gram-negative bacteria and enteric commensals should be anticipated. In the present study, most wound infections were polymicrobial, and bacteria were often multi-drug resistant. This should be anticipated when starting empirical antibiotic treatment and stresses the importance of strict quarantine and appropriate microbiological assessment on admission of such patients.