Download View PDF

Large-area burns with pandrug-resistant Pseudomonas aeruginosa
infection and respiratory failure: clinical experience of five cases
Ning Fanggang1, Zhao Xiaozhuo1,Bian Jin2 and Zhang Guo’an1
1
Department of Burns, Beijing Jishuitan Hospital, 31 Dongjie, Xinjiekou, Xicheng
District, Beijing 100035, China
2
Division of Pharmacy, Beijing Jishuitan Hospital, 31 Dongjie, Xinjiekou, Xicheng
District, Beijing 100035, China
Correspondence to:
Zhang GA, Department of Burns, Beijing Jishuitan Hospital, 31, Dongjie, Xinjiekou,
Xicheng District, Beijing 100035, China
Key words: burns, pandrug-resistant Pseudomonas aeruginosa, respiratory failure
Abstract
Background: To summarize the treatment of large-area burns (60–80%) with
pandrug-resistant Pseudomonas aeruginosa (PDRPA) infection and respiratory failure
in our hospital over the last two years, and to explore a feasible treatment protocol for
such patients.
Methods: We retrospectively analyzed the treatment of five patients with large-area
burns accompanied by PDRPA infection and respiratory failure transferred to our
hospital from Burn Units in hospitals in other Chinese cities from January 2008 to
February 2010.
Results: Before PDRPA infection occurred, all five patients had open wounds with
large areas of granulation because of the failure of surgery and dissolving of scar
tissue; they had also undergone long-term administration of carbapenems. Four
patients recovered from burns and one died after therapy. This therapy was ventilatory
support, rigorous repair of wounds, and combined antibiotic therapy targeted at
drug-resistance mechanisms, including carbapenems, ciprofloxacin, macrolide
antibiotics and β-lactamase inhibitors.
Conclusions: First, compromised immunity caused by delayed healing of burn
wounds in patients with large-area burns and long-term administration of
carbapenems may be the important factors in the initiation and progression of PDRPA
infection. Second, if targeted at drug-resistance mechanisms, combined antibiotic
therapy using carbapenems, ciprofloxacin, macrolide antibiotics and β-lactamase
inhibitors could effectively control PDRPA infection. Third, although patients with
large-area burns suffered respiratory failure and had high risks from anesthesia and
surgery, only aggressive skin grafting with ventilatory support could control the
infection and save lives. Patients may not be able to tolerate a long surgical procedure,
so the duration of surgery should be minimized, and the frequency of surgery
increased.
Background
Drug-resistant infection has become a global concern in recent years as
broad-spectrum antibiotics have been brought into wide use. Pseudomonas
aeruginosa is a leading cause of nosocomial infection in burns patients.
Multidrug-resistant Pseudomonas aeruginosa (MDRPA) or pandrug-resistant
Pseudomonas aeruginosa (PDRPA) is being detected with increasingly frequency in
burns patients. Patients with large-area burns are particularly vulnerable to infection,
and standard treatment may be compromised if they are infected with MDRPA or
PDARPA. These patients often have signs of severe respiratory failure, and may
eventually die from resultant multiple-organ failure.
This contribution summarizes treatment and outcomes for large-area burns with
PDRPA infection and respiratory failure in the Department of Burns, Beijing Jishuitan
Hospital (Beijing, China) over the last 2 years. We used these data to explore a
feasible treatment protocol for such patients.
METHODS
Our department admitted five patients (4 males, 1 female; age range, 16–58
years) with large-area burns accompanied by PDRPA infection and respiratory failure
from January 2008 to February 2010. These patients were transferred to our hospital
from Burn Units in hospitals located in other cities in China. Of the five patients, one
died of severe arrhythmia due to cardiac dysfunction at 58 days post-admission. The
other four patients recovered and were discharged after hospitalization for 67–91
days.
In the present study, we retrospectively analyzed the medical conditions and
treatment of all the patients to summarize our clinical experiences.
RESULTS
1. Treatment before hospital transfer
The five patients were hospitalized in the Burn Units for 17–38 days before
being transferred to our department. The total burn surface area in these patients was
60–80%; full-thickness burns accounted for 40–60%. All five patients had moderate
inhalation injuries, and thus underwent emergency tracheotomy. Urine output during
the shock phase was 50–100 ml/h. Four patients were treated by escharectomy,
autologous micro-skin grafting, and large-sheet allogeneic skin grafting. Two patients
underwent a single escharectomy of full-thickness burns on the four limbs, chest and
abdomen; the other two subjects underwent escharectomy of third-degree burns on
four limbs, chest and abdomen twice. In four patients, the micro-skin grafts had poor
survival, and large areas of granulated wounds were exposed. One out of the five
patients was not debrided. All five patients were given carbapenems for 14–24 days
before transfer to our department. All suffered respiratory failure and received
ventilatory support. Chest radiography and microbiological testing demonstrated
PDRPA infection before admission to our department. The Pseudomonas aeruginosa
strain was resistant to most antibiotics, including third-generation cephalosporin,
carbapenems and ciprofloxacin.
Table 1. Patient information before hospital transfer
Patient
Sex
Age
Total burn area/ area
Inhalation
Hospital
Number of auto-skin
Ventilatory
Use of
(years)
of third-degree
injury
stay
graft transplantation
support
carbapenems
burns (%)
(Y/N)
(days)
procedures
(days)
(days)
A
M
16
80/60
Y
17
2
5
17
B
M
32
70/50
Y
21
2
6
14
C
M
34
70/50
Y
21
1
8
14
D
M
58
60/50
Y
38
1
34
24
E
F
31
70/40
Y
25
0
5
19
2. Medical condition upon hospital transfer
2.1 Infection
2.1.1 Pulmonary infection
All patients were diagnosed with pulmonary infection by chest radiography and
sputum microbiological test. Sputum was aspirated from the deep airway through an
incision on the trachea using a suction catheter and sputum collector. The sputum was
collected for three days for culture purposes. PDRPA was cultured three times for all
five patients, and was resistant to all available antibiotics, including third-generation
cephalosporin, carbapenems and ciprofloxacin, but not polymyxin.
2.1.2 Wound infection
The burn wounds of the five patients were covered by a large amount of purulent
secretion. In the burn wound drainage culture, PDRPA was cultured for all five
patients, and two of them were combined with methicillin-resistant Staphylococcus
aureus (MRSA) infection.
2.1.3 Systemic infection
In the blood culture, PDRPA was detected in only one patient.
2.1.4 Fungal infection
No fungi were detected in the sputum, blood, or wound discharge in any patient.
2.2 Wound
Of the five patients, four had poor outcomes after autologous micro-skin grafting
that did not survive or expand to the desired extent. Granulation tissue accounted for
30–50% of the burn wound. The wound base was exposed to dirt, and a large amount
of purulent discharge was observed. In the one patient who did not undergo surgery,
the eschar was allowed to undergo autolytic debridement; purulent discharge was
noted underneath.
2.3 General condition
All five patients met the criteria for systemic inflammatory response syndrome
(SIRS) and acute respiratory distress syndrome (ARDS); all showed cardiac
dysfunction. One patient had cardiac arrest 2 h after hospital admission, but recovered
cardiac rhythm after resuscitation. All patients exhibited signs of hepatic dysfunction,
and renal dysfunction was observed in two patients. All patients showed anemia and
hypoproteinemia, as well as electrolyte disturbances of various degrees.
Table 2. Medical condition upon hospital admission
Patient
Granulation
Sputum
Blood
Discharge
SIRS
ARDS
Hepatic
Renal
area or scar
culture
culture
culture
(Y/N)
(Y/N)
dysfunction
dysfunction
dysfunction
(Y/N)
(Y/N)
(Y/N)
dissolution
Cardiac
area (%)
A
30
PDRPA
N
PDRPA +
Y
Y
Y
N
Y
Y
MRSA
B
40
PDRPA
N
PDRPA
Y
Y
Y
N
C
40
PDRPA
N
PDRPA
Y
Y
Y
Y
Y (cardiac
arrest once)
D
50
PDRPA
PDRPA
PDRPA +
Y
Y
Y
Y
Y
Y
Y
Y
N
Y
MRSA
E
40
PDRPA
N
PDRPA
PDRPA: pandrug-resistant Pseudomonas aeruginosa; MRSA: methicillin-resistant
Staphylococcus aureus; SIRS: systemic inflammatory response syndrome; ARDS:
acute respiratory distress syndrome
3. Treatment
3.1 Wound treatment
The burn wound is the origin of most problems in burn patients and, if the wound
cannot be adequately treated in a timely fashion, recovery will not occur. Although the
five patients were critically ill and had ventilatory support, we chose to treat the burn
wound rigorously.
All five patients (including the subject with cardiac arrest upon hospital admission)
underwent skin grafting within four days after admission. One patient who did not
undergo debridement before admission did so with autologous micro-skin and
large-sheet allogeneic skin grafting. The other four patients with large-area
granulating wounds were treated with wound expansion; stamp-shaped autologous
skin grafting expanded 2:1. Due to a shortage of autologous skin, the remaining
wounds were covered with allogeneic skin. After the initial procedure, the five
patients underwent stamp-shaped skin grafting every 7–10 days with skin grafts taken
predominantly from the head. The first surgical procedure was carried out within 1.5 h,
and subsequent procedures within 1 h. Each patient underwent 4–6 surgical
procedures in total. After surgery, the wound was covered with silver-containing
dressings.
Of the five patients, one suffered severe arrhythmia due to cardiac dysfunction at
58 days post-admission: he died after resuscitation attempts arrest failed. At that time,
the residual open wound area was ~5%. The residual open wound areas of the other
four patients were <1% after hospitalization for 40–50 days. These patients recovered
and were discharged after hospitalization for 67–91 days.
3.2 Administration of antibiotics
After a clear diagnosis of PDRPA, patients were given infusions of imipenem
combined with azithromycin and ciprofloxacin. Two of the patients infected with
MRSA were given infusions of vancomycin. Forty-eight hours after administration,
all five patients demonstrated a significant improvement in five main areas. First,
body temperature decreased by >1°C. Second, respiratory function improved
significantly; when the partial pressure of oxygen in blood was maintained at >60
mmHg, the essential oxygen concentration could be reduced ≥10%. Third, the test
of white blood cell counts, the classification of neutrophils and chest radiography
showed significant improvement. Fourth, symptoms of severe infection (e.g., bloating)
improved.
The types of antibiotics used were limited because we wanted to prevent adverse
events and the risk of drug-resistance. The antibiotic combinations were converted to
meropenem + azithromycin, ciprofloxacin + azithromycin, or ciprofloxacin +
cefoperazone sodium. The administration time of each combination did not exceed 15
days. After replacing the combination of antibiotics, the condition of the five patients
remained stable.
Though fungi were not detected in the patients, the risk of fungal infection was
high. Patients were therefore given fluconazole (Diflucan) via the oral route to
prevent fungal infection until discontinuation of broad-spectrum antibiotics. All
patients were given (via the oral route) live combined Bifidobacterium, Lactobacillus,
and Enterococcus preparation.
Antibiotic administration was continued for 39–58 days for all patients. The
patient died later administrated 58days. For the four survival patients, after body
temperature and the WBC returned to normal levels for 5–7 days, and ventilatory
support was discontinued, antibiotics were discontinued, regardless of sputum culture
results. Three patients did not show recurrent infection after discontinuation of
antibiotics, body temperature were normal, and repeated chest radiography test and
blood test showed no obvious signs of infection. One patient showed an increase in
body temperature and WBC as well as respiratory deterioration after discontinuation.
We, therefore, continued administration of antibiotics for 10 days, after which
recovery was observed.
When antibiotic administration was discontinued in the four survival patients,
However, a small amount of PDRPA growth was observed in the sputum culture, and
the result turned negative 20–30 days after discontinuation of antibiotic
administration.
3.3 Ventilatory support
Ventilatory support was initiated after the five patients were admitted to our
department. In the process of assisted ventilation, to reduce damage to the airway due
to long-term high concentrations of oxygen and high airway pressure, high blood
oxygen pressure and oxygen saturation should not be pursued. The modes of the
ventilator were initially set to SIMV+PS+PEEP. The parameters were set as:
frequency, 15–20/min; tidal volume, 8 mL/kg; support pressure, 15–18 CmH; PEEP,
10–12 CmH; oxygen concentration, 60–80%. The modes of removal of the ventilator
were set to PSV+PEEP. The parameters were set as: support pressure: 6–8 CmH;
PEEP, 3–5 CmH; oxygen concentration, 35–40%.
One patient was sustained on the ventilator for 58 continuous days before his
death, whereas the other four patients were sustained for 28–38 days. Weaning from
the ventilator was attempted unsuccessfully in four patients on 1–3 occasions due to
retention of carbon dioxide.
Of the five patients, one suffered severe arrhythmia due to cardiac dysfunction at
58 days post-admission; he died after resuscitation for cardiac arrest failed. The other
four patients recovered and were discharged after hospitalization for 67–91 days.
Table 3. Treatment information
Patient
Number of auto-skin
Duration of
graft transplantation
Antibiotic combination
Duration of
Times of failure
Hospital stay
Outcome
antibiotic use
ventilatory
in stopping the
(days)
procedures
(days)
support (days)
ventilator
A
4
40
combination 1, combination 2, combination 3
28
1
67
Recovery
B
5
47
combination 1, combination 2, combination 3
37
3
81
Recovery
C
5
39
combination 1, combination 2, combination 3
33
2
78
Recovery
D
6
58
combination 1, combination 2, combination 3,
58
0
58
Death
38
1
91
Recovery
combination 4
E
4
54
combination 1, combination 2, combination 3,
combination 4
Antibiotic combination 1: Imipenem + cilastatin sodium, azithromycin, ciprofloxacin
Antibiotic combination 2: meropenem + azithromycin
Antibiotic combination 3: ciprofloxacin + azithromycin
Antibiotic combination 4: cefoperazone/sulbactam + ciprofloxacin
DISCUSSION
Infection due to MDRPA and PDRPA has become a challenge in clinical practice.
Risk factors linked to infection with MDRPA include illness, severity of illness,
bedridden state, invasive devices, and exposure to antibiotics (particularly β-lactams
and fluoroquinolones) (1,2,3,4).
In the present study, five patients with PDRPA infection had open wounds with
large areas of granulation or scar tissue. Toxin absorption due to wound discharge and
infection causes severe physiological damage and compromises immunity. Long-term
inhalation injury with ventilatory support and administration of carbapenems also
complicates PDRPA infection and ARDS.
Respiratory failure is also accompanied by multiple-organ failure in such patients,
making development of a treatment protocol difficult. If the wound is not treated in a
timely fashion, uncontrollable infection can result. Risks from anesthesia and surgery
are high due to ARDS and multiple-organ failure. We think that the underlying cause
of these problems is the burn wound. For such patients, aggressive skin grafting
should be conducted to treat open burns. Patients may not be able to tolerate a long
surgical procedure, so the duration of surgery should be minimized. In the present
study, the duration of surgery was <1.5 h initially, and <1 h in subsequent procedures.
The frequency of surgery was once every 7–10 days, and we demonstrated that the
duration of surgery was safe and effective. After initial surgery, respiration
deteriorated in all patients, but recovered in 3–5 days after the parameters of the
ventilator were adjusted. Respiration deteriorated to some extent in the subsequent
procedures but, overall, improved as the size of the open wound was gradually
decreased. The residual open wound area was only 5% and respiration clearly
improved in one patient before death, suggesting that the ventilator could have been
removed. However, the patient had a history of coronary disease and cardiac
dysfunction. Severe arrhythmias and cardiac failure occurred suddenly and the patient
died after the failure of resuscitation.
PDRPA is resistant to multiple antibiotics excluding polymyxin, so antibiotic use
is a major concern. Some clinicians try to reapply polymyxin to combat MDRPA
infection (5). However, polymyxin is not usually incorporated into the routine
treatment protocol because of its severe toxicity and limited availability. Selection of
multiple antibiotics targeting drug-resistance mechanisms seems to be achievable. The
drug resistance of PDRPA has been widely researched. The mechanisms for multidrug
and pandrug resistance appear to be biofilm and the efflux pump. With respect to the
former, macrolides may be used to damage the biofilm; however, this is not
bactericidal, it instead facilitates the effect of other antibiotics (6). The efflux pump
appears to cause an increase in the antibiotic dose, and a short period of treatment
may help to kill bacteria. Cilastatin sodium and sulbactam suppress broad-spectrum
β-lactamases.
If targeted at the mechanisms of drug resistance, combined antibiotic therapy
(including carbapenems, ciprofloxacin, macrolide antibiotics and β-lactamase
inhibitors) could effectively control infection by PDRPA.
Infection cannot be controlled in a short period of time if a large open wound is
not covered and the patient is maintained on the ventilator. In these circumstances,
antibiotics should be used for a longer period of time, and rotated periodically to
reduce toxicity and prevent an increase in drug resistance. In the meantime,
precautions should be taken to prevent ecologic imbalance and fungal infection.
Elimination of PDRPA infection through antibiotic use is not realistic if the
granulating wound is large and the patient has reduced immunity. However,
appropriate use of antibiotics inhibits infection, and “buys” time for wound recovery.
Once the overall wound is treated, the physiological condition and immunity will
improve and PDRPA infection can be controlled. When antibiotic administration was
discontinued in the four patients in the present study, a small amount of PDRPA
growth was observed in the sputum culture. The patients recovered in these situations,
indicating that their immune system could kill bacteria without long-term use of
powerful broad-spectrum antibiotics.
If patients are sustained on a ventilator for long periods, respiratory dependence
and airway injury tends to occur. The time spent on the ventilator should therefore be
minimized. The oxygen saturation level should be kept to >95%, and the oxygen
concentration and other parameters maintained at a low level.
In the present study, despite several attempts, four patients could not be weaned
from the ventilator due to the retention of carbon dioxide which was probably caused
by low respiratory capacity and a heavy discharge of respiratory system. This problem
could be solved by physical exercise, intermittent removal of ventilatory support, and
strengthening respiratory care.
Conclusion
We can draw three main conclusions from the present study. First, compromised
immunity caused by delayed healing of burn wounds in patients with large-area burns
and long-term administration of carbapenems may be the important factors in the
initiation and progression of PDRPA infection. Second, if targeted at drug-resistance
mechanisms, combined antibiotic therapy using carbapenems, ciprofloxacin,
macrolide antibiotics and β-lactamase inhibitors could effectively control PDRPA
infection. Third, although patients with large-area burns suffered respiratory failure
and had high risks from anesthesia and surgery, only aggressive skin grafting with
ventilatory support could control the infection and save lives. Patients may not be able
to tolerate a long surgical procedure, so the duration of surgery should be minimized,
and the frequency of surgery increased.
References
1.Defez C, Fabbro-Peray p, Bouziges N, et al. Risk factors for multidrug-resistant
Pseudomonas aeruginosa nosocomial infection. J Hosp Infect . 2004;57:209–216.
2.Mohr JF, Jones A, Ostrosky-Zeichner L, et al. Associations between antibiotic use
and changes in susceptibility patterns of Pseudomonas aeruginosa in a private,
university-affiliated teaching hospital: an 8-years-experience:1995-2002. Int J
Antimicrob Agents 2004; 24:346–351.
3.Paramythiotou E, Lucet JC, Timsit JF, et al. Acquisition of multidrug-resistant
Pseudomonas aeruginosa in patients in intensive care units: role of antibiotics with
antipseudomonal activity. Clin Infect Dis. 2004;38:67–677.
4.Ortega B, Groeneveld AB, Schultsz C. Endemic multidrug-resistant Pseudomonas
aeruginosa in critically ill patients. Infect Control Hosp Epidemiol. 2004;25:825–831.
5.Kallel H, Hergaft L, Bahloul M, et al. Safety and efficacy of colistin compared with
imipenem in the treatment of ventilator-associated pneumonia: a matched case-control
study. Intensive Care Med.2007;33:1162–1167.
6.Nagata T, Mukae H, kadota J, et al. Effect of erythromycin on chronic respiratory
infection caused by Pseudomonas aeruginosa with biofilm formation in an
experimental murine model. Antimicrob Agents Chemother 2004;48:2251–2259.