Introduction

Ventilator associated pneumonia (VAP) is a serious and common complication of patients in intensive care units (ICUs). The A. baumannii has emerged as one of the most important nosocomial pathogens associated with VAP [1–5]. Resistance to all major classes of antibiotics in A. baumannii has essentially increased worldwide in the past decade [6]. Multidrug resistant (MDR) A. baumannii infections have an attributable mortality of 25 % associated with inappropriate antibiotic treatment [7]. Due to the limited treatment options, an old drug, colistin, has increasingly been used as salvage therapy in these infections [8]. However, clinical effectiveness of this old drug for treatment of pneumonia is doubtful because of its inadequate penetration into the pulmonary parenchyma. And the clinical cure rates are not very high with colistin monotherapy in VAP [3, 9]. As a result, new agents or combination regimens are investigated for the effective management of these life-threatening nosocomial infections. Sulbactam is the most active ß-lactamase inhibitor against MDR A. baumannii strains. There are experimental studies that show the efficacy of sulbactam alone or in combination therapy against MDR A. baumannii pneumonia in the literature [10, 11]. However, there are limited clinical studies about its efficacy in MDR A. baumannii pneumonia [12].

The objective of this study was to compare clinical and microbiological efficacy of colistin monotherapy and colistin/sulbactam combination therapy for the treatment of MDR A. baumannii VAP in ICUs.

Materials and methods

This retrospective study was carried out at the ICUs of Erciyes University Hospital between January and December 2011. This is a tertiary hospital with 213 ICU beds and approximately 4,787 admissions annually.

Design

In this study, we evaluated critically ill adult patients (>16 years old) who received IV colistin or colistin/sulbactam for the treatment of MDR A. baumannii VAP. VAP was defined according to the criteria of the American Thoracic Society Consensus Conference on VAP [13]. Patients with monomicrobial VAP and with no co-infection were included into the study. The records of the patients were accessed by scanning the surveillance data of the Infection Control Committee and patients’ charts. Patients’ demographic characteristics; underlying diseases, Acute Physiology and Chronic Health Evaluation (APACHE) II score on admission, severity of sepsis on the diagnosis of VAP, length of ICU stay, colistin and sulbactam doses, concomitant use of glycopeptide and aminoglycoside and adverse reactions were recorded. Sepsis was defined according to the 2001 SCCM/ESICM/ATS International sepsis definitions conference criteria [14].

Identification of MDR

The identification of the isolates as Acinetobacter was performed using Phoenix microscan system which is FDA (the United States Food and Drug Administration) approved for this use in clinical laboratories with recommended practices. The antimicrobial susceptibility was determined using the Kirby-Bauer disk diffusion test according to the Clinical and Laboratory Standards Institute (CLSI) (formerly NCCLS) [15]. Multidrug resistant A. baumannii isolates are defined as those resistant to more than three classes of antibiotics (aminoglycosides, antipseudomonal penicillins, carbapenems, cephalosporins, beta-lactam/beta-lactamase inhibitor, quinolones; in additional, colistin, tigecycline, were agents that were tested occasionally) [16].

Administration of antimicrobials

Colistimethate sodium (Colimycin, Kocak Farma) and sulbactam (B-Laktam, Mustafa Nevzat) were used in this study. For patients with normal renal function, the standard dosage was 2.5 mg/kg every 12 h. The dose was adjusted according to the creatinine clearance [17]. The dosage 2.5 mg/kg every 6 h was defined as a high dose. The dosage of sulbactam was 3 g every 8 h.

Evaluation of outcome

The clinical and microbiological responses were evaluated on the fifth day and at the end of the treatment. On the fifth day of the therapy, subsidence of symptoms and signs of VAP without an additional antimicrobial agent for VAP, was defined as a good response; progression of symptoms, signs and prescription of additional antimicrobial agents were defined as a poor response. At the end of the therapy, resolution of the symptoms and signs of VAP without antimicrobial maintenance therapy were defined as a clinical success; the persistence of symptoms and signs or recurrence of VAP after the discontinuation of colistin and sulbactam were defined as a clinical failure. Bacteriological clearance was defined as eradication of MDR A. baumannii and bacteriological failure was defined as the persistence of MDR A. baumannii on follow up culture [17]. During the therapy, patients’ creatinine and liver enzymes levels were monitored every other day to evaluate nephrotoxicity and hepatotoxicity. The creatinine clearance rate (CLcr) was calculated using the equation of Cockcroft [18]. The colistin related nephrotoxicity was defined according to the risk, injuy, failure, loss, end stage kidney disease (RIFLE) criteria [19].

Nephrotoxicity was evaluated for patients who received the therapy at least for five days. Hepatotoxicity was evaluated with elevation of liver enzymes. Also, skin reactions were followed-up during the study period.

Ethics

The study was approved by the ethics committee of Erciyes University (date 20.9.2011, number 2011/28).

Statistical analysis

The collected information was processed using version 20.0 of the Statistical Package for Social Sciences (SPSS) for Windows. The Shapiro–Wilk test was performed to check the normality assumption of the data. The Mann–Whitney U-test was used for the comparison of continuous variables. All the analyses were performed with the level of significance set at p < 0.05.

Results

During the study period, a total of 89 critically ill patients with MDR A. baumannii VAP meet the study criteria. Antibiotic resistance rates of MDR A. baumannii are shown in Table 1. Carbapenem resistance rates were >90 % and all the isolates were susceptible to colistin. Moreover, 26 (24.7 %) isolates were resistant to tigecycline. Fifty-two patients (58.4 %) received colistin alone, 37 (41.6 %) patients were treated with colistin/sulbactam. Seven patients continued to receive colistin or colistin/sulbactam therapy despite that the isolates were susceptible to carbapenems. This is because those patients received carbapenem before the diagnosis of VAP and had a good response to colistin or colistin/sulbactam therapy.

Table 1 In vitro antibiotic resistance rates of 87 A. baumannii strains isolated from endotracheal aspirates or bronchial lavage samples
Full size table

The demographic and clinical data of patients are shown in Table 2. The median APACHE II score was higher and diabetes mellitus was more common in the colistin/sulbactam group than the colistin group (p < 0.05). However, other demographic characteristics were not statistically significant between the two groups. Eighty-six (96.6 %) patients received antibiotics before colistin or colistin/sulbactam therapy. Seventy-two (80.9 %) patients had severe sepsis or septic shock due to VAP. In the monotherapy group, 26 (50.0 %) patients had a standard dosage of colistin, whereas 11 (21.2 %) had adjusted dosage according to their CLcr and 15 (28.8 %) patients had a high dosage. In the combination group, 30 (81.1 %) patients had a standard dosage of colistin and 7 (18.9 %) patients had an adjusted dosage according to their CLcr. None of the patients had a high dosage of colistin in the combination group. The median duration of therapy was 14 days (range 2–22 days) in both groups. In colistin/sulbactam combination group, ten (27 %) patients received glycopeptide and two (5.4 %) patients received aminoglycoside with colistin/sulbactam therapy. In the colistin group, 23 (44.2 %) patients received glycopeptide and 15 (28.8 %) patients received aminoglycoside with colistin therapy. The concomitant use of glycopeptides was not statistically significant between the groups; however, aminoglycoside use was statistically more prevalent in the colistin group (p = 0.006).

Table 2 The characteristics of patients who were treated with colistin and colistin/sulbactam
Full size table

The clinical and bacteriological outcomes for the two treatment groups are summarized in Table 3. On the fifth day of treatments, 21 patients (40.4 %) had a good response in the colistin monotherapy group and 16 patients (43.2 %) had a good response in the combination group (p = 0.84). Forty-seven patients in the colistin group and 35 patients in the colistin/sulbactam group completed the therapy. Clinical cure rates were 29.8 and 40.0 % in colistin and colistin/sulbactam groups, respectively. The bacteriological clearance rates were 72.3 and 85.7 % in the colistin and colistin/sulbactam groups, respectively. Although clinical and bacteriological response rates were higher in the colistin/sulbactam group, it was not statistically significant (p > 0.05). Nephrotoxicity was evaluated in 82 patients and 22 (26.8 %) had nephrotoxicity. The nephrotoxicity rate was higher (six of 15 patients-40 %) in patients who received a high dose than a standard dose (15 of 56 patients-26.8 %) and adjusted dose (one of 18 patients-5.6 %). However the difference was not statistically significant. The nephrotoxicity evaluation according to the RIFLE criteria is shown in Table 4. Nephrotoxicity was observed in eight (36.4 %) patients who received concomitant glycopeptides and four (18.2 %) patients who received aminoglycoside (p = 1.00).

Table 3 Clinical and microbiological evaluation of patients on colistin and colistin/sulbactam therapy
Full size table

The hepatotoxicity rate was seen in two (5.4 %) patients in the colistin/sulbactam combination group and eight (15.4 %) patients in the colistin monotherapy group. The difference was not statistically significant between the two groups. No skin reaction was observed during the study period due to sulbactam.

Table 4 Patients’ relationship to RIFLE criteria
Full size table

The overall mortality rate was 60.7 % in MDR A. baumannii VAP patients. The mortality rate was 73.0 % in the colistin/sulbactam group, whereas it was 51.9 % in the colistin group. However, when the mortality rates were adjusted for APACHE II score, the difference was not statistically significant (p = 0.53). The survival graphic of the two groups is shown in Fig. 1.

Fig. 1
figure 1

Survival plot of groups

Full size image

Discussion

Colistin/sulbactam combination therapy is promising in severe MDR A. baumannii infections. In this study, although it was not statistically significant, clinical cure rates (40.0 vs. 29.8 %) and bacteriological clearance rates (85.7 vs. 72.3 %) were better than colistin monotherapy. As well as in experimental studies, sulbactam improved the survival and the sterilization of lungs and blood compared with the controls. And the investigators showed that sulbactam is as effective as other susceptible antimicrobials in severe A. baumannii infections. [20]. Also, Montero et al. [21] showed that colistin monotherapy did not appear as a good option for treatment of patients with pneumonia due to carbapenem-resistant A. baumannii strains in an experimental study. And in these experimental studies, other alternative therapeutic options, including combinations were offered for better clinical results. Sulbactam is one of the antimicrobials that was recommended in these experimental studies [7, 10, 20–23]. Also, there are some clinical studies with carbapenem/sulbactam and ampisillin/sulbactam that showed more potent antimicrobial activity [10, 12]. On the other hand, increasing use of colistin for treating infections due to MDR A. baumannii worldwide will inevitably increase the recovery rate of colistin-resistant isolates in the future. Kempf et al. [24] reported the in vitro synergistic activity of a combination of colistin with sulbactam against colistin-resistant A. baumannii strains and suggested the use of sulbactam added to colistin to avoid the emergence of colistin-resistant A. baumannii strains.

The other strategy to improve the outcome of serious infections with colistin therapy is using higher dosage. However, nephrotoxicity still remains a major concern with colistin and reported as 9–50 % in the literature [10, 14, 25]. Nephrotoxicity rates may increase with excessive doses to 80 % [26]. In this study, the overall nephrotoxicity rate was 26.8 % and 15 patients received high doses with higher nephrotoxicity rates (40.0 %) than standard doses (26.8 %). This may be associated with excessive dosing or the severity of the condition of the patients. However, colistin shows a concentration-dependent bactericidal activity and renal toxicity depends on the drug concentration and treatment duration. Higher doses and longer dosing-intervals, along with loading doses, have been proposed to obtain a more effective killing. Dalfino et al. [27] showed that in severe infections due to MDR gram-negative bacteria, the high dose extended-interval colistin regimen (loading colistin dose of 9 MU, followed by a maintenance dose of 4.5 MU every 12 h) has a high efficacy (82.1 %), without significant nephrotoxicity (17.8 %).

The hepatotoxicity rate was 5.4 % in the sulbactam combination group, whereas it was 15.4 % in the colistin monotherapy group. These rates may be associated with other drugs used in critically ill patients or for severity of illness (80.9 % of patients had severe sepsis or septic shock on the diagnosis of VAP), although mortality rates (73.0 vs. 51.9 %) were high in both groups due to the severity of patients and infection.

To our knowledge, this is the first clinical study that compares the effectiveness of colistin vs. colistin/sulbactam therapy, although our study has some limitations. Despite the acceptable number for the study population, it is a retrospective and not a randomized study. Other factors that can affect the clinical outcome and adverse reaction of colistin were not well-controlled.

In conclusion, colistin/sulbactam combination therapy is encouraging in severe A. baumannii VAP. However, large, prospective, randomized and well-controlled studies are needed to show the effectiveness of colistin/sulbactam combination therapy.