Article Text

Download PDFPDF

Infections in crush syndrome: a retrospective observational study after the Wenchuan earthquake
  1. Chen Xiaolei,
  2. Zhong Hui,
  3. Fu Ping,
  4. Hu Zhangxue,
  5. Qin Wei,
  6. Tao Ye
  1. Department of Nephrology, West China Hospital, Sichuan University, Chengdu, China
  1. Correspondence to Dr Fu Ping, Department of Nephrology, West China Hospital, Sichuan University, Chengdu 610041, China; irenecxl{at}


Objectives To investigate the characteristics of infections in the casualties with crush syndrome after the Wenchuan earthquake.

Methods The clinical data of patients during their first 2 months of hospitalisation were analysed retrospectively. 58 crush syndrome patients were included. Demographic data, physical and laboratory findings, treatment and outcome were recorded.

Results The mean age of these patients was 32.8±20.7 years and the average duration under the rubble was 23.7±19.3 h. Fasciotomy and amputation were performed in 29 (50%) and 34 (58.6%) patients, respectively. Renal replacement therapies were administered to 47 patients (81%). Four patients died while all the others restored their renal function. Microbial pathogens were detected in the samples from 39 patients (67.2%), 26 of whom developed sepsis. The most vulnerable sites for infection were wounds (55.2%) and lungs (37.9%). Acinetobacter baumanii and Pseudomonas aeruginosa were the most common bacterial isolates from wound infections. 26 of 39 patients (66.7%) became infected after 48 h of admission. The infected group had a longer duration under the rubble (39.2 h vs 30.1 h, p=0.000) and more fasciotomies (61.5% vs 26.3%, p=0.012) compared with those without infections. Multivariate analysis indicated that duration under the rubble (p=0.012, OR 1.061), duration of renal impairment (p=0.015, OR 1.096) and fasciotomy (p=0.024, OR 5.470) were risk factors for infection.

Conclusions Infectious complications are common in crush syndrome. In order to improve patient outcomes, vigorous care and strict surveillance are required.

  • Abdomen—non-trauma
  • clinical care
  • continuous renal replacement therapy
  • crush syndrome
  • fasciotomy
  • infection
  • major incident
  • renal

Statistics from

Crush syndrome, which is characterised by severe muscle injuries, myoglobinuria and acute renal failure (ARF), is commonly seen in earthquake patients. Despite great developments in critical care, mortality from crush syndrome remains high.1 2 Infections, which carry the risk of increasing morbidity and mortality when complicated by sepsis, are the most frequent medical complications in these patients.3 Open injuries or wounds as a result of fasciotomy, impaired nutritional status, compromised immune system and various indwelling catheters may all contribute to the outbreak of infections, especially nosocomial events. Therefore, a profound understanding of the characteristics of infectious complications is important for medical staff to provide comprehensive treatment.

On 12 May 2008, a catastrophic earthquake struck Wenchuan County of Southwest China's Sichuan Province. The earthquake registered 8.0 on the Richter scale and caused heavy damage to an area of over 100 000 km2. According to the official reports, the earthquake killed more than 80 000 and injured over 30 000 people, of whom many were transferred to the West China hospital located approximately 92 km away from the epicentre. The hospital accepted nearly 2600 patients from the disaster area and approximately 1700 of them were hospitalised.

The aim of this report was to provide an overview of the clinical and bacteriological characteristics of infectious complications in crush syndrome patients admitted to our hospital after the Wenchuan earthquake.

Patients and methods

Medical records of patients with earthquake-related injuries or diseases during their first 2 months of hospitalisation were analysed retrospectively. Data concerning patient demographics, physical and laboratory findings, diagnoses, treatment modalities (including dialysis and surgical procedures) and outcomes were recorded. As a leading medical institution after the disaster, this hospital received patients who were relatively severely injured. One hundred forty-six patients with crush injuries were admitted to the hospital in the first 2 weeks after the earthquake. Among them, patients diagnosed with crush syndrome were included in this study and those with previous chronic renal disease were excluded.

Crush syndrome was defined as the presence of crush injuries, that is muscle injury caused by prolonged limb compression and myoglobinuria with acute impairment of renal function. The latter diagnosis was described as at least one of the following criteria: urine output less than 400 ml/day and/or blood urea nitrogen greater than 40 mg/dl, serum creatinine greater than 2.0 mg/dl, uric acid greater than 8.0 mg/dl, potassium greater than 6.0 mEq/l, phosphorus greater than 8.0 mg/dl, bicarbonate less than 15 mEq/l and/or serum total calcium less than 8.0 mg/dl.4 5

For each patient, a complete blood count and blood chemistry were tested at the first day of admission and thereafter. Changes in the values of the indicators such as haemoglobin, white blood cells, creatinine and creatine kinase were observed.

Test for infections

In order to analyse infectious complications, we collected microbiological data of the patients on various specimens from blood, wound, sputum, urine, drainage material and vascular catheters.

Samples from wounds as a result of open injury or surgical operations such as fasciotomy and amputation were collected on the first visit from all the patients, whereas subsequent cultures were obtained only from those who presented with signs of infections, such as purulent discharge or darkening colour of the muscles.

Blood was cultured in patients with fever or systemic signs of sepsis, such as rigors, alteration of consciousness, tachycardia and drop in arterial blood pressure. Blood was obtained from three different vein punctures at different times and was inoculated for aerobic and anaerobic cultures.

A first morning specimen was used in the sputum collection from patients suspected of having lung infection. This procedure was repeated at least three times.

If patients with indwelling vascular catheters got fever accompanied by swelling and tenderness at the catheter site for 3 days or got rigors soon after the initiation of haemodialysis, the catheters were removed as soon as possible and their tips were cultured semiquantitatively. Colonisation of the catheter was defined as more than 15 colony-forming units in cultures obtained by rolling proximal or distal catheter segments on plates.6

In patients with infections, sepsis was defined by two or more of the following conditions: temperature greater than 38°C or less than 36°C, pulse greater than 90 beats/min, a drop in systolic arterial blood pressure despite adequate fluid replacement (<90 mm Hg), respiratory rate greater than 20 breaths/min and white blood cell count greater than 12 000 cells/mm3 or less than 4000 cells/mm3.7

Renal replacement therapy

As acute renal failure is common in crush syndrome, renal replacement therapy is important in the treatment. In our hospital, haemodialysis is the first choice for these patients. The indications to initiate haemodialysis were as follows: hyperkalaemia (potassium >6.5 mmol/l) accompanied by oliguria or anuria, myoglobinuria or creatin kinase greater than 1000 IU/l, sever metabolic acidosis (pH <7.2) which could not be alleviated by alkaline infusions. The duration of dialysis was determined by the return of renal function. We used three patterns of haemodialysis: intermittent haemodialysis, haemodiafiltration and continuous renal replacement therapy (CRRT). CRRT was preferred for the severe victims with unstable vital signs, whereas the other two were chosen in mild conditions.

Statistical analysis

Results are given as mean±SD for numerical variables and as proportions for categorical variables. Comparisons of means and percentages were carried out with the unpaired two-tailed Student's t test and the χ2 test, respectively. Multiple logistic regression analysis was performed to determine what kinds of factors were independently correlated with infection. Two-tailed p values less than 0.05 were considered statistically significant.


Clinical characteristics

According to the definition of crush syndrome, 58 patients (35 male, 23 female) were included in our study. The mean age of these patients was 32.8±20.7 years (9–83 years). Age groups were as follows: 0–15 years, 10.3%; 16–30 years, 46.6%; 31–45 years, 17.2%; 46–60 years, 12.1%; 61–85 years, 13.8%. The average duration under the rubble was 23.7±19.3 h (2–73 h). A total of 53 patients (91.4%) were trapped under the rubble over 4 h. The most frequently injured parts were extremities (93.1%). Fasciotomies were performed on 29 patients (50%) and 34 (58.6%) had amputations. There was a significant difference in the incidence of infection between the fasciotomised and non-fasciotomised groups (82.8% vs 51.7%, p=0.012).

Infectious manifestation

Microbial pathogens were detected in the samples from 39 patients (67.2%) and sepsis developed in 26 of them. The most vulnerable sites for infection were wounds (55.2%) and lungs (37.9%), followed by blood (22.4%) and urinary tract (10.3%). A total of 18 patients (31.0%) got multisite infections. Acinetobacter baumanii, Pseudomonas aeruginosa, Enterobacter cloacae and Escherichia coli were isolated in 19, 17, 13 and nine patients, respectively. Twenty-eight patients (48.3%) were infected with more than one pathogen. For wound infections, the most common bacterial isolates were A baumanii and P aeruginosa, and those from pulmonary infections were A baumanii and E cloacae (table 1). A baumanii resistant to carbapenem were isolated from the specimens of 12 patients.

Table 1

Isolations from different bacterial specimens

Most of the patients with infections (26 of 39, 66.7%) got positive specimen cultures at a time over 48 h after admission. Therefore, they were all diagnosed as hospital-acquired infection.

Compared with the non-infected group, the infected group had a younger mean age and longer duration under the rubble. Also, the incidence of fasciotomy in this group was much higher than that of those without infections (table 2).

Table 2

Clinical data of patients with crush syndrome


Antibiotics were used in all of the patients diagnosed with infection (39 patients). Considering the multiple infections, the joint use of antibiotics was common. Each individual who got a surgical wound was given anti-anaerobe antibiotics empirically on admission. During the course of treatment, the adjusting of antibiotics was guided by microbial culture results and antibiotic susceptibility data. The commonly used antibiotics included cephalosporins in 24 patients (24/39, 61.5%), carbapenems in 11 patients (11/39, 28.2%), quinolones in 10 patients(10/39, 25.6%) and anti-staphyloccal antibiotics in eight patients (8/39, 20.5%).

Forty-seven patients (81%) received haemodialysis for acute kidney injury resulting from crush syndrome. CRRT was administered to 37 seriously ill patients. In our hospital, continuous veno-venous haemofiltration (CVVH) was the most often used modality of CRRT. The rest were treated by intermittent haemodialysis combined with haemodiafiltration.

At the end of this study, four patients died of liver rupture, splenic rupture, pulmonary contusion and multiple organ dysfunction, respectively. All the survivors restored their renal function after a mean impairment period of 22.0±13.5 days (4–55 days). Compared with the patients without infections, the duration of renal impairment was longer in the infected group (table 2).

In the multivariate analysis, time under the rubble, duration of renal impairment and fasciotomies were risk factors for infection. However, age was not found to be associated with the prevalence of infection in these patients (table 3).

Table 3

Logistic regression analysis for risk factors of infection


Infection is one of the most commonly encountered clinical entities in crush syndrome patients after a devastating earthquake, which contributes to a high morbidity and mortality when complicated by sepsis.3 8 All of the following factors make it easy for these patients to develop infectious complications: (1) Open injuries or fasciotomies can inevitably cause infection due to bacterial contamination. Also, a long duration under the rubble leads to more severe crush injuries and increases the risk of exposure to pathogens. Consequently, wounds become the most susceptible sites for infections. (2) The immune system is sure to be compromised after crush syndrome as a result of organ dysfunction and malnutrition.9 The underlying mechanisms were attributed to auto-oxidative receptor injury or changes in natural killer cell activity.10 (3) Diverse indwelling catheters potentially break down the anatomical barriers to infection.11 (4) In the chaos after a catastrophe, it is always difficult to implement effective surveillance on nosocomial infection and antibiotic prescription. As reported after the Marmara earthquake, which hit northwestern Turkey in 1999, the rate of infection complicated with crush syndrome ranged from 25.8% to 66.6% in different medical centres,4 5 12 whereas it was 67.2% in our study. The discrepancy may be explained by the fact that patients admitted to our hospital were relatively severely injured or had multiple organ dysfunction requiring longer hospitalisation.

In accordance with the other literature,3 4 the most vulnerable sites for infection were wounds (55.2%) in the present study. Both the break down of the mucocutaneous barrier and necrotic tissue make bacteria colonise easily. Besides, a large volume of protein-rich exudates from the wound also aggravates malnutrition. Consequently, fasciotomy per se increases the risk of infection. Open wounds resulting from fasciotomies have recently captured the attention of clinicians. Some investigators have reported high infection rates from their fasciotomy series.13 14 In our study, the infection rate in the fasciotomised group was significantly greater than in the non-fasciotomised group (82.8% vs 51.7%). Fasciotomy also served as an independent risk factor for infection in the multivariable analysis. Debates on the necessity for fasciotomy still continue now. Some believe that many patients undergoing fasciotomy still need amputation in the long run, and it is not worth taking the high risk of infection and improper wound care thus.15 Nevertheless, others emphasise the early relief of intracompartmental pressure to avoid necrosis or palsies.4 Duman et al 16 also thought highly of the preservation of limbs as a result of fasciotomies. Fasciotomies should thus be performed on clear indications so that undesirable outcomes might be avoided. In general, the indications for fasciotomy are clinical evidence of compartment syndrome or elevated intracompartment pressure (>35 mm Hg). The clinical evidence of compartment syndrome is concluded as ‘5P’ (ie, pallor, paresthesia, pulseless, paralysis and pain). For those with irreversible ischaemic injuries, amputation should be performed directly. As it was difficult to measure the intracompartment pressure in the survivors, fasciotomies were routinely determined by the clinical evaluation of surgeons in our hospital. However, there are still no classic clinical findings that enable a successful prediction of the need for treatment of compartment syndrome. Randomised prospective investigations are still needed to determine the timing and necessity of fasciotomies.

In our study, A baumanii and P aeruginosa were the major bacterial isolates from wounds, which were essentially considered to be hospital pathogens.17 It was reported that as the hospitalisation period lengthened as a result of acute renal failure, nosocomial infections supervened.4 Consistently, we also discovered that the infected group was characterised by a longer duration of renal impairment. To improve the outcome of critically injured patients, extensive support therapy should be given with the aim of benefiting renal recovery. With respect to nosocomial infection control, admission to the particular rooms occupied by infected patients should be restricted and any indwelling catheters (eg, vascular and urinary catheters) should be removed as soon as possible. In addition, wide spectrum antibiotics are not recommended in empiric therapy. Upon any sign of infection, appropriate microbial samples should be obtained and antibiotics can be changed accordingly.

Renal replacement therapy is essential for the management of crush syndrome patients characterised by acute renal failure. In the light of convenience, haemodialysis was preferred in our hospital. For critically injured patients with unstable haemodynamics, CRRT was recommended as the first choice. CVVH was the most commonly used modality. By contrast, the moderately ill were given intermittent haemodialysis combined with haemodiafiltration. Myoglobin, which is released during rhabdomyolysis, is recognised as an important pathogenic factor of renal dysfunction by producing renal tubular obstruction, lipid peroxidation within the tubular cells and renal vasoconstriction.18 Given a molecular mass of 17 KDa, it requires removal by convection. CVVH is thus more effective than conventional haemodialysis in the clearance of myoglobin. In addition, continuous therapies incorporate several advantages including improved haemodynamic stability, optimal fluid balance, gradual urea removal, elimination of septic mediators and the possibility of unlimited parenteral nutrition.19 In the present study, most casualties with sepsis or even septic shock showed an obvious improvement soon after the application of CVVH. Despite encouraging results, the superiority of continuous modalities with regard to outcome is still controversial. It remains to be determined when to start or stop CRRT and whether the transition of modalities is beneficial.

Age was not found to be associated with the prevalence of infection in patients in the multivariate analysis, although the infected group had a younger mean age. The fact that young patients accounted for most of the survivors may provide selection bias in this study.

In conclusion, infectious complications are common in crush syndrome and contribute to high morbidity and mortality. It is prudent to implement fasciotomy by clear indications, in that it is a major predictor of infection. As the hospitalisation period lengthens, nosocomial infections become a troublesome medical entity necessitating vigorous care and strict surveillance.



  • CX and ZH contributed equally to this paper.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.