July - September 2006: 
Volume 19, Issue 3

Click on the image to download the Issue in PDF format.


The incidence of infections caused by multi-drug resistant Acinetobacter baumannii in an ICU
Intensive Care Unit, General Hospital of Lamia, 2Microbiology Laboratory, General Hospital of Lamia Nosocomial infections caused by Acinetobacter baumannii are of increasing concern in critically ill patients and the emergence of multi-drug resistant strains (MDRAB) complicates their treatment. The purpose of this retrospective study was to determine the incidence of MDRAB infections, as well as PNEUMON Number 3, Vol. 19, July - September 2006 229 to investigate the effect of antibiotic treatment on mortality. Medical records of all ICU admissions through a 21-month period were reviewed (N 240); patients with MDRAB-positive cultures were eligible for evaluation (N 35). According to infection type and origin, eligible patients were divided into three groups: group A included patients with ventilator-associated pneumonia (VAP); group B patients with sepsis of non-respiratory origin; and group C patients with respiratory colonization. The overall incidence of MDRAB-positive cultures was 14.58%. Of them, 71.4% of the patients (N 25) had VAP; 14.2% had sepsis of non-respiratory origin (N 5); and 14.2% (N 5) had respiratory colonization. Mean age was 58.14 years; male-to-female ratio was 21/14; and mean APACHE II score at ICU admission 18.2. The most important risk factors were CHF, CAD, COPD and surgical procedure. Treatment included ampicillin/sulbactam and colimycin alone or in combination. Crude mortality was 25.71% (N 9) and disease-specific mortality 23.33% (N 7). Clinical improvement was noted in 70% (N 21) and chronic colonization developed (group C encountered) in 11.42% (N 4). The incidence of infections due to MDRAB among ICU patients is clinically significant. VAP is the most frequently encountered infection. Combination treatment with ambicillin/sulbactam plus colimycin seems to be more effective due to the synergistic effects of colimycin and sulbactam, as evidenced by a reduction in mortality, which was nevertheless not statistically significant. Our results are based on a small series of patients and larger well-designed studies are needed to determine the efficacy of colimycin in combination treatment. Pneumon 2006, 19(3):231-237.
Full text


Acinetobacter baumanni is a pathogen primarily associated with hospitalization. It causes nosocomial infections, as well as outbreaks in special clinical settings such as an ICU, with increasing incidence.1,2 In addition, the emergence of multi-drug resistant strains (multi-drug resistant Acinetobacter baumanni, MDRAB) is currently a worldwide problem making infection control and treatment extremely difficult.3

Indicatively, in an outbreak of infections caused by a certain Acinetobacter baumanni strain that affected all New York hospitals in 2002, resistance to carbapenems was 53% and resistance to all antimicrobials was 12%.4 Such outbreaks typically occur among mechanically ventilated patients.5 Increasing resistance to carbapenems commonly leads to the use of either colimycin6-9 or polymyxin-B,12 since reports of resistance to these agents are scarce. Furthermore, both clinical and in vitro studies suggest better outcomes with combination treatment using such agents as carbapenems, colimycin, rifampin, and ambicillin/sulbactam (AM/SB).10-12 The clinical effectiveness of such combination treatment regimens is under study.

Patients and Methods

This retrospective study aims at determining the incidence of MDRAB infections in patients admitted in the ICU of the General Hospital of Lamia during the 21-month period from September 2003 through May 2005, as well as evaluating the effectiveness of the applied treatment regimens. The setting is a multidisciplinary ICU with 6 beds. During the above-mentioned period, a total of 240 patients were admitted in our ICU, 35 of which had cultures positive for MDRAB. Multi-drug resistance is defined as lack of sensitivity to five major antimicrobial classes: (1) anti-pseudomonal penicillins; (2) cephalosporins; (3) monobactams; (4) aminoglycosides; and (5) carbapenems.

Medical and nursing records of all patients were reviewed to identify MDRAB cases and record the type of infection, as well as the applied treatment. Results of blood gas analyses, chest radiographs, physical therapy charts, and laboratory tests of renal (urea, creatinin) and liver (transaminase, alkaline phosphatase, and γ-GT, billirubin) function were also included in the evaluation of MDRAB-positive patients.

Over the study period, the agents used for the treatment of MDRAB infection were colimycin (Colistin, Norma, Athens) and AM/SB (Begalin, Pfizer, Greece) alone or in combination. Provided renal function was normal, colimycin dosing was 3·106 q8d IV or 106 IU q4d as inhalations, and AM/SB dosing was 9 g q8d IV.

Microbiology testing of all samples was performed in the Microbiology Laboratory of the General Hospital of Lamia. The preliminary identification was based on direct microscopy features, colony morphology in solid cultures, and microscopic observation of Gram-stained slides. The API 20 NE system was used for definite identification. Sensitivity testing was performed using the Kirby-Bauer method, in line with the NCCLS guidelines. MIC (μg/ml) values and possible presence of synergy were determined using E-test. The presence of identical MDRAB strains in bronchial secretion cultures of different patients was detected using DNA-fingerprinting in the Microbiology Laboratory of the Medical School of the University of Larissa.

Infection definitions

Patients were classified as cases of ventilatorassociated pneumonia (VAP) if they met the CPIS criteria (Clinical Pulmonary Infection Score; i.e. six independent clinicolaboratory variables rated at least 6 in total) provided that invasive mechanical ventilation had started at least 48 hours before and no other pathogen was cultured in their bronchial secretions, apart from MDRAB of course.

Sepsis14 due to central venous catheter infection (catheter-related sepsis, CRS)15 was diagnosed in septic patients meeting the criteria of the semiquantitative method, i.e. growth of >15 CFU at the tip of the catheter and detection of the same pathogen in blood cultures in the absence of other possible infection site.

Sepsis due to wound infection was diagnosed in patients with MDRAB-positive cultures of both blood and trauma fluid (pus or drainage fluid) in the absence of other possible infection site.

Finally, patients with MDRAB-positive bronchial secretion cultures not meeting the CPIS criteria were classified as having respiratory MDRAB-colonization.Those VAP patients who continued to have MDRABpositive bronchial secretion cultures after their clinical improvement, were classified as having chronic colonization. It should be noted that there were no positive urine cultures; this is the reason for not defining sepsis due to urinary tract infection.

Outcome definitions

A positive outcome (cure or improvement) of VAP was defined as clinical (fever remission, reduced bronchial secretions with almost normal appearance); radiological (less prominent alveolar infiltrations); microbiological (negative cultures); and laboratory (better ABG results, WBC normalization) restoration.

Apart from clinical improvement, the definition of positive outcome in cases of catheter-related sepsis or sepsis due to trauma infection also required negative cultures.

Treatment regimens

The review of treatment regimens indicated that patients with sepsis of non-respiratory origin (CRS or trauma infection) received only intravenous antimicrobials. On the other hand, 4 patients with VAP received only inhalations of nebulized colimycin.


This retrospective study reviews ICU patient records over a 21-month period. During this period, a total of 240 patients were admitted, 35 of which had cultures positive for MDRAB. Hence, MDRAB incidence is 14.5%.

Table 1 describes the general characteristics of MDRAB-positive patients. Mean age was 58.1 years, and male-to-female ratio 21:14. Mean APACHE II score on admission was 18.2%. Nine of the 35 MDRAB-positive patients died, which translates in a crude mortality of 25.7%. Note that total crude mortality in the ICU during the study period was 28.3%, as there were 68 deaths among the total of 240 patients.

According to the CPIS criteria, 85.7% (30) of the MDRAB-positive patients had infection, but 14.28% (5) just had respiratory colonization from the beginning. Of the subjects with infection, 70% (21) showed clinical improvement, whereas only 6.6% (2) evolved chronic colonization. Disease-specific mortality was 23.3% (7 out of 30 patients with infection died).

In particular, 71.4% (25) of all MDRAB-positive patients with infection had VAP, 14.2% (5) had sepsis of non-respiratory origin, and 14.2% (5) had respiratory colonization. Table 2 presents detailed information on patient characteristics and predisposing factors. As we can see, the mean age of VAP patients was 58.8 years; male-to-female ratio 15:10; and APACHE II score on admission 18. Predisposing factors included, with decreasing order of frequency, CHF, prior surgical procedure, CAD, head injury, COPD, and DM. It is noteworthy that the majority of the patients presented VAP in the second week of ICU stay, as 56% (14) of the patients were found to have the infection between day 7 and day 14. However, the time distribution of infection onset is extremely wide, ranging from day 3 to day 44 of ICU stay. This may be due to variations in the level of compliance with the general measures of infection prevention and control. Finally, as regards VAP diagnosis, the mean CPIS score was 7.6.

Among patients with sepsis of non-respiratory origin, 60% had MDRAB cultured in samples (pus) obtained from the site of operation (2 patients had peptic surgery and 1 had orthopedic surgery), as well as in blood samples, whereas 40% had catheter-related sepsis (CRS). Mean age was 55.2 years; male-to-female ratio 3:2; and APACHE II score at admission 22.6. The most common predisposing factors were DM and CRF.

Lastly, regarding patients with respiratory MDRAB colonization, mean age was 57.4 years; male-to-female ratio 4:1; and APACHE II score on admission 13.4. It should be noted that APACHE II score at admission varied widely, with two patients having low values, i.e. 2 and 6, and three patients having much higher values, i.e. 16, 18 and 25. ICU admission was necessary due to acute respiratory insufficiency in 80% of the cases. 40% of the cases had undergone surgery prior to ICU admission.

Regarding the outcome of VAP, mortality was 24%; however, 72% of the patients improved and 4% of the patients evolved chronic colonization.

Based on the applied treatment regimen (Table 3), patients were divided into three groups. Group I included 10 patients who received colimycin as monotherapy. Of them, 4 received nebulized colimycin, 3 received intravenous colimycin, and 3 received both nebulized and intravenous colimycin.

In group II, 13 patients were treated with the combination of AM/SB given IV and colimycin given either IV (6 patients) or as inhalations (7 patients). Lastly, group III consists of two patients treated with intravenous AM/SB alone. Mean patient age in the three treatment groups was 57.2, 58.6 and 68.5 years, respectively. The respective values of APACHE II score on admission were 17, 19 and 21.

Our data show that 60% of the patients (groups II and III) who received AM/SB had poorer prognosis, as assessed by an APACHE II score of 19.8, which is predictive of mortality rates up to 40% for non-surgical conditions. However, observed mortality was 30% (3 deaths) in group I and 23% (3 deaths) in group II. No death was recorded in group III, although this cannot be reliably evaluated due to the small number of patients (2 patients).

Response to treatment was evaluated separately in each diagnostic group. In group A, 60% of the patients improved and 10% evolved chronic colonization. Improvement was achieved in 76.9% of the patients in group B. In group C, both patients improved. Hence, treatment with AM/SB may be associated with favorable prognosis.

In patients with sepsis of non-respiratory origin, mortality rate was 20%; this corresponds to the death of the oldest patient (82 years old), who had the highest APACHE II score on admission (25), three major predisposing factors (autoimmune disease, liver and renal failure) and foreign body contamination (bone prosthesis), which was the cause of sepsis in the first place. Improvement was demonstrated in 60% of the patients, whereas 20% (1 patient with biliary tract-skin fistula) showed chronic colonization. It should be noted that complete response was achieved in the two CRS cases; one of the patients received monotherapy with AM/SB and the other combination therapy with colimycin plus AM/SB intravenously.

Finally, among patients with respiratory MDRAB colonization, crude mortality rate was 40% (2 deaths); the patients who died had high APACHE II scores at admission (18 and 16) and older age (72 and 75 years, respectively). One colonized patient received nebulized colimycin due to a high APACHE II score on admission (25), relatively young age (49 years), and admission due to respiratory causes. Among those who survived, negativity of bronchial secretion cultures was achieved only in 33.3%, i.e. the younger patient with the lower APACHE II score on admission. The rest were considered to have chronic colonization.


Acinetobacter baumannii is an important nosocomial pathogen, primarily affecting mechanically ventilated patients with severe underlying diseases. It typically develops multiple resistance to antimicrobials. Furthermore, it has been responsible for outbreaks in intensive care units and units for burns worldwide. It is for these reasons that effective management is difficult, yet so necessary. Reports of global mortality range from 61.9% to 66.7%,7,16 making authors like Landman D and Quale JM4 wonder if the pre-antibiotic era has returned. Efforts to prevent and treat these infections primarily aim at determining predisposing factors for the emergence of MDRAB VAP or bacteremia, so as to avoid them whenever possible; and assessing clinical effectiveness of combination treatment.

In this context, we retrospectively reviewed medical records of 240 patients admitted in the ICU of the General Hospital of Lamia from September 2003 through May 2005. The incidence of MDRAB was 14.5%, with cultures positive for this pathogen in 35 patients. Infection and, in general, contamination with MDRAB primarily affects the respiratory system, as evidenced by the occurrence of VAP in 25 patients and the colonization of 5 additional patients. To a great extent, this relates to mechanical ventilation17 and also suggests inadequate compliance with infection control and prevention measures. As regards predisposing factors, there was no differentiation from data reported in the international literature,18,19 with surgical procedures and cardiopulmonary disorders being most commonly encountered.

Study patients had good outcome; both crude and disease specific mortality were too low, 25.1% and 23.3%, respectively, especially when compared to international reports. These results indicate that combination treatment with colimycin and AM/SB is probably effective.

In addition, among VAP patients who received combination treatment mortality rates were lower and evolution of chronic colonization was less common. However, the number of study subjects is too small to establish statistical significance and larger more powerful studies are required. Considering in vitro evidence of greater susceptibility to this β-lactamase inhibitor, as described in the work of Higgins PG and Wisplinghoff H,20 the synergistic effect of sulbactam probably accounts for a more favorable prognosis.

The observed differentiation in mortality rates as compared to rates reported in the international literature –with the exception of a recent work of Michalopoulos AS and Kasiakou SK9 reporting improvement in 7 out of 8 patients with VAP who received inhaled colimycin– may be attributable to additional independent determinants of prognosis, apart from treatment. It has been suggested that clinical severity at the time of pneumonia diagnosis affects mortality. Hence, a relatively moderate APACHE II score on admission (18.2) may have had a favorable effect on mortality, although it is similar to that reported in the study of Garnacho-Montero J and Ortiz-Leyba C,16 in which mortality was about 65%.

Moreover, it is possible that certain predisposing factors, such as prior treatment with imipenem,16,17 which were not examined in our study, are also important. It is clear that prospective randomized studies in larger patient series are required to verify the results discussed here.

Conclusively, the incidence of MDRAB among ICU patients is clinically important. The majority of these infections affect the respiratory system. Combination treatment with AM/SB and colimycin is effective, as disease-specific mortality is particularly low. Evidence of better response warrants further evaluation of the effectiveness of combination treatment including AM/SB.


1. Urban C, Segal-Maurer S. Considerations in control and treatment of nosocomial infections due to multi drug resistant Acinetobacter baumannii. Clin Infect Dis 2003; 36(10):1268-1274.
2. Villegas MV, Hartstein Al. Acinetobacter outbreaks, 1977-2000. Infect Control Hosp Epidemiol 2003; 24(4):284-295.
3. Rello J, Sa-Borges M. Variations in etiology of ventilator- associated pneumonia across four treatment sites: implications for antimicrobial prescribing practices. Am J Respir Crit Care Med 1999; 160:608-613.
4. Landman D, Quale JM. Citywide clonal outbreak of multi resistant Acinetobacter baumannii and Pseudomonas aeruginosa in Brooklyn, NY: The preantibiotic era has returned. Arch Intern Med 2002; 162(13):1515-20.
5. Husni RN, Goldstein LS. Risk factors for an outbreak of multi drug resistant Acinetobacter nosocomial pneumonia among intubated patients. Chest 1999; 115:1378-1382.
6. Levin AS, Barone AA. Intravenous colistin as therapy for nosocomial infections caused by multi drug resistant Pseudomonas aeruginosa and Acinetobacter baumannii. Clin Infect Dis 1999; 28(5):1008-1011.
7. Garnacho-Montero J, Ortiz-Leyba C. Treatment of multi resistant Acinetobacter baumannii ventilatorassociated pneumonia with intravenous colistin: a comparison with imipenem susceptible episodes. Clin Infect Dis 2003; 36:1111-1118.
8. Michalopoulos AS, Tsiodras S. Colistin treatment of patients with ICU-acquired infections due to multi resistant Gram-negative bacteria: the renaissance of an old antibiotic. Clin Microbiol Infect 2005; 11(2):115-21.
9. Michalopoulos AS, Kasiakou SK. Aerolised colistin for the treatment of nosocomial pneumonia due to multi drug resistant Gram-negative bacteria in patients without cystic fibrosis. Crit Care 2005; 9(1): R53-R59.
10. Wood GC, Hanes SD. Comparison of ampicillinsulbactam and imipenem-cilastin for treatment of Acinetobacter ventilator-associated pneumonia. Clin Infect Dis 2002; 34:1425-1430.
11. Giamarellos-Bourboulis EJ, Xirouchaki E. Interactions of colistin and rifampin on MDRAB. Diagn Microbiol Infect Dis 2001; 40(3):117-120.
12. Yoon J, Urban C. In vitro double and triple synergistic activities of Polymyxin B, imipenem and rifampin against MDRAB. Antimicrob Agents Chemother 2004; 48(3):753-757.
13. Baughman RP. Diagnosis of ventilator associated pneumonia. Microbes Infect. 2005 Feb;7(2):262-7.
14. American College of Chest Physicians/Consensus Committee (1992). Definitions of sepsis and organ failures and guidelines for the use of innovative therapies in sepsis. Chest 101:1658-1662.
15. Maki DG, Weise CE. A semi-quantitative culture method for identifying intravenous catheter-related infections. N Eng J Med 1977; 296:1305-1309.
16. Garnacho-Montero J, Ortiz-Leyba C. Acinetobacter baumannii ventilator-associated pneumonia: Epidemiological and clinical findings. Intensive Care Med 2005; 31:649-655.
17. Gomez J, Simarro E. Six year prospective study of risk and prognostic factors in patients with nosocomial sepsis caused by Acinetobacter baumannii. Eur J Clin Microbiol Infect Dis 1999; 18:358-361.
18. Garcia-Garmendia JL, Ortiz-Leyba C. Risk factors for Acinetobacter baumannii nosocomial bacteremia in critically ill patients: a cohort study. Clin Infect Dis 2001; 33:939-946.
19. Mulin B, Talon D. Risk factors for nosocomial colonization with multi resistant Acinetobacter baumannii. Eur J Clin Microbiol Infect Dis 1995; 14:569-576.
20. Higgins PG, Wisplinghoff H. In vitro activities of the β-lactamase inhibitors Clavulanic acid, Sulbactam and Tazobactam alone or in combination with β-Lactams against epidemiologically characterized MDRAB strains. Antimicrob Agents Chemother 2004; 48(5):1586-1592.