Aerosolized plus Intravenous Colistin versus Intravenous Colistin Alone for the Treatment of Ventilator-Associated Pneumonia: A Matched Case-Control Study

.

There is limited information on the pharmacokinetics of colistin after intravenous (IV) administration, and its effectiveness for treatment of pneumonia has been questioned because of its inadequate penetration in the lung parenchyma [9].However, there are several reports of successful treatment of pneumonia with IV colistin [8][9][10].

Setting and Study Design
Patient population.This retrospective case-control matching study (ratio, 1:1) was performed at the intensive care unit (ICU) of the University Hospital of Heraklion (Crete, Greece), an 11-bed medical-surgical unit.All patients with culture-documented monomicrobial VAP due to Acinetobacter baumanii, Pseudomonas aeruginosa, or Klebsiella pneumoniae that was susceptible only to colistin during the period January 2005-December 2008 were potentially eligible for the study.
Eligible case patients had received у6 doses of AS therapy and у3 days of IV therapy (the AS-IV colistin group) [15].Control patients had to have had received IV colistin for у3 days without AS colistin therapy (the IV colistin group).Control patients were chosen according to the following matching criteria: age ‫5ע(‬ years) and Acute Physiology and Chronic Health Evaluation (APACHE) II score ‫4ע(‬ points) on the day of the introduction of colistin therapy.If 12 control patients were available, the date of ICU admission was used as an additional matching criterion.The additional matching criterion was necessary for the equal distribution of patients in the treatment groups during the 2 study periods.Thus, 11 patients were distributed in each arm during the first 2 years, and patients were distributed in each arm during the last 2 years of the study.In selecting control patients, the investigators were not aware of the outcome of the treatment.
Data from all case patients were reviewed independently by 2 infectious disease specialists (G.S. and D.P.K.) to check the clinical outcomes in case and control patients.In the event of a discrepancy, the 2 reviewers assessed the records again and reached a consensus decision.The response to treatment was assessed at the time of discharge from the ICU or at the end of antimicrobial therapy.The 2 investigators were not aware of the patient's therapy.
Definitions.Pneumonia was considered to be ventilatorassociated if the onset occurred after the patient was intubated for у48 h and the infection was judged not to have been incubating before the initiation of mechanical ventilation [8].Pneumonia was diagnosed on the basis of a radiographic finding of a new and progressive pulmonary infiltrate and at least 2 of the following clinical criteria: body temperature, 138ЊC or !35.5ЊC; leukocytosis (leukocyte count, 112,000 cells/mm 3 ) or leukopenia (leukocyte count, !4000 cells/mm 3 ); and clinical evidence suggestive of pneumonia, such as purulent bronchial secretions and a decrease in oxygenation [19].Microbiological diagnosis of VAP was established by positive cultures of bronchial secretions or bronchoalveolar lavage with isolation of an MDR gram-negative bacterium with a concentration of у10 4 CFU/mL [18].Bacteriologic sampling was performed for all patients on the day that VAP was suspected (day 0), before new antimicrobials were started.The severity of the clinical condition was assessed according to the APACHE II score [20].The response to treatment was assessed at the time of discharge from the ICU or at the end of antimicrobial therapy, especially if the patient remained hospitalized for a non VAP-related disease.
The primary end point of the study was the clinical outcome of VAP.As secondary end points, we evaluated microbiological outcome, VAP-related mortality, all-cause mortality, and the occurrence of adverse events during colistin treatment.
Clinical outcome was classified as clinical cure (ie, resolution of presenting symptoms and signs of infection by the end of colistin treatment), clinical improvement (ie, partial resolution of presenting symptoms and signs of infection), clinical failure (ie, persistence or worsening of presenting symptoms and/or signs of infection during colistin administration), and recurrence of infection (ie, occurrence of a new episode of infection at least 72 h after clinical resolution of a preceding episode).Clinical success was defined as clinical cure or clinical improvement.
Microbiological outcome was rated as eradication of the pathogen (ie, no growth of the pathogen in the final culture of specimens during the entire hospitalization), persistence of the pathogen (ie, persistent growth of the responsible pathogen regardless of the clinical outcome of the infection), recurrence (regrowth) of the pathogen (ie, reisolation of the same pathogen regardless of the clinical outcome of the infection), or colonization (ie, persistence or regrowth of the pathogen without symptoms and signs of infection).
VAP-related mortality was defined as death that occurred during the treatment period when the signs of pneumonia remained and as death due to septic shock.
In patients with normal renal function, nephrotoxicity was defined as a serum creatinine value 12 mg/dL; as a reduction in the calculated creatinine clearance of 50%, compared with the value at the start of treatment; or as a decline in renal function that prompted renal replacement therapy.In patients with preexisting renal dysfunction, nephrotoxicity was defined as an increase of 150% of the baseline creatinine level or as a reduction in the calculated creatinine clearance of 50% relative to the value at therapy initiation.All adverse effects related to AS colistin use, such as bronchoconstriction, cough, apnea, or chest tightness, and arterial hypoxemia were recorded.The following variables were also recorded demographic characteristics; cause of ICU admission; duration of ICU stay; comorbidities, including chronic lung disease, malignancy, diabetes mellitus, and renal failure; antineoplastic therapy; use of systemic corticosteroids and antibiotics 1 week prior to and/ or during the infectious episode; previous surgery; length and dosage of colistin treatment; simultaneous use of other antimicrobials; causative bacteria; source of diagnostic culture; antimicrobial susceptibility; the results of laboratory and imaging tests; treatment-associated adverse events; clinical and microbiological outcome; and VAP-associated and overall mortality.

Microbiological Testing
Susceptibility testing of gram-negative microorganisms was performed using an automated broth microdilution method (Organon Teknika Corp).The breakpoints used were those defined by the Clinical and Laboratory Standards Institute [21].Susceptibility to colistin was also tested using the Etest methodology (susceptibility, р2 mg/L; resistance, у4 mg/L) and the disk diffusion method with a 10-mg colistin sulfate disk.Gram-negative microorganisms were defined as MDR susceptible only to colistin if they were resistant to all of the 6 antipseudomonal classes of antimicrobial agents (antipseudomonal penicillins, cephalosporins, carbapenems, monobactams, fluoroquinolones, and aminoglycosides) for P. aeruginosa and K. pneumoniae, and in addition, resistance to ampicillin-sulbactam and tetracycline for A. baumanii.

Treatment Regimen
The daily dose of AS colistin was 2 million international units (IU) divided into 2 doses, whereas the daily dose of IV colistin was 9 million IU divided into 3 doses in patients with normal renal function.

Data Management and Statistics
Data were collected on forms and were computerized and analyzed using SPSS software, version 16.0 (SPSS).Variables for the matched case-control pairs were compared by Wilcoxon matched pairs test.The x 2 or Fisher exact test was used to assess differences in categorical variables, as appropriate.Differences in continuous variables were assessed by the Student t test or nonparametric Mann-Whitney U test.Kaplan-Meier curves were used to assess differences between the IV group and the IV plus AS group and overall mortality.The log-rank test was used to determine the level of statistical significance when comparing survival curves.Multivariate logistic regression analysis was used to assess the independent effect of therapy on each of the 2 outcomes (clinical cure and microorganism eradication).P values are 2-tailed, and P values !.05 were considered to be statistically significant.

RESULTS
During the 4-year study period, 151 patients with MDR-gram negative VAP treated with colistin were identified.Ninety-five patients were available as control patients, and there were 56  potentially eligible case patients who had received simultaneous AS and IV colistin treatment.
Among them, we evaluated 43 case patients with MDR VAP due to gram-negative bacteria who received simultaneous AS and IV colistin and 43 corresponding control subjects who received IV colistin alone matched for age and APACHE II score on the day of introduction of colistin for VAP; their baseline characteristics are shown in Table 1.The median duration of hospitalization in the ICU was similar for both groups: 18 days (range, 3-75 days) for the IV colistin group and 20.5 days (range, 3-93 days) for the AS-IV colistin group (P p .676).
A separate analysis of VAP cases due to A. baumannii, which was the most common pathogen in both arms, was performed to exclude a potential effect of other pathogens on outcome.No significant differences between the 2 arms in terms of clinical and microbiological outcome or mortality were found.
The median duration of therapy was comparable between the 2 treatment groups: 10 days (range, 4-36 days) for the IV colistin group and 13 days (range, 5-56 days) for the AS-IV colistin group (P p .840).
The clinical and bacteriological outcomes for the 2 treat-ment groups are summarized in Table 2.No significant differences were found in bacteriological outcome between the 2 treatment groups.Eradication of the causative microorganism was achieved in 17 (50%) of 34 patients in the IV colistin group and in 19 (45%) of 42 patients in the AS-IV colistin group.The patients in AS-IV colistin group had a marginally better rate of clinical cure than did patients in IV colistin group (23 [54%] of 43 patients vs 14 [32.5%] of 43 patients; P p .05).However, we found no statistically significant difference in terms of clinical success (26 [60%] of 43 patients in the IV colistin group vs 33 (74%) of 43 patients in AS-IV colistin group; P p .10).
To investigate the role of simultaneous AS and IV colistin treatment as an independent predictor of clinical cure, a logistic regression model was used, with adjustments for renal failure, presence of diabetes mellitus, malignancy, prior immunosuppressive treatment, and prior antibiotic or corticosteroid use.No statistically significant better clinical cure rate was observed in association with AS-IV colistin treatment (odds ratio, 2.375; 95% confidence interval, 0.901-6.258;P p .080).
Overall, the mortality rate in the ICU was 42% (18 of 43 patients) in the IV colistin, compared with and 24% (10 of 43 patients) in the AS-IV colistin group (P p .066).The VAPrelated mortality rates were 26% (11 of 43 patients) and 16% (7 of 43 patients), respectively (P p .289).Kaplan-Meier curves revealed no statistically significant differences in either all-cause mortality (P p .888, by log-rank test) or VAP-related mortality (P p .268, by log-rank test) (Figures 1 and 2).Eight patients (19%) in each treatment group presented with renal dysfunction.One patient in the IV colistin group had preexisting mild chronic renal disease.In all 16 patients, the dose of colistin was reduced; none of the patients required renal replacement therapy or discontinuation of colistin treatment.No adverse events, such as bronchoconstriction, apnea, or chest tightness, were associated with AS colistin therapy.In addition, neurotoxic adverse effects were not observed in any patient in either treatment group.

DISCUSSION
The main finding of the present study was that the addition of AS to IV colistin did not provide any additional therapeutic benefit to patients with MDR VAP due to gram-negative bacteria.In addition, adverse events associated with systemic and AS use of colistin, such as nephrotoxicity, neurotoxicity, and direct toxicity on airways, were not observed.
VAP is the most frequent nosocomial infection in the ICU; it affects up to 27% of patients undergoing mechanical ventilation and is associated with considerable morbidity and mortality [8,22].The increasing rate of VAP due to gram-negative MDR strains resistant to almost all available antimicrobials has led to the reintroduction of polymyxin antimicrobials (polymyxin B and colistin), which were discontinued from use because of their nephrotoxicity and neurotoxicity [8,9].
However, the efficacy of colistin for treatment of pneumonia has been questioned because of its inadequate penetration into lung parenchyma [9].AS colistin appears to be a suitable option for the treatment of patients with VAP, because it achieves higher pulmonary concentrations with ignorable systemic absorption and toxicity [9,23].Indeed, sputum and lung tissue antibiotic levels achieved after inhalation are higher than those obtained after IV administration [24,25].In addition, topical antibiotics may reduce sputum volume and sputum bacterial growth in patients undergoing ventilation who are at risk of developing VAP [24].However, there has been extensive experience with administration of AS colistin only for patients with cystic fibrosis [9,24].
By contrast, AS colistin as adjunctive to IV antimicrobials for the treatment of VAP caused by MDR gram-negative pathogens has been evaluated only in a few studies [11,[15][16][17][18]26].These reports have shown encouraging results, with a high favorable clinical response and microbiological eradication (180%) and mortality rates ranging from 12.5% to 46.7%.However, all of these studies included a small number of patients, and with the exception of the study by Korbila et al [18], they did not include a control arm [8,11,[15][16][17]26].
Furthermore, AS colistin as monotherapy has been used only in a case series of 5 patients with nosocomial pneumonia.A favorable clinical outcome was observed in 4 of these patients, suggesting that AS-only colistin therapy may be effective in certain cases [27].Finally, despite these encouraging data, current guidelines do not recommend AS colistin treatment.
The results of the present case-control study are in contrast with results of previous case studies, because the addition of AS colistin did not offer any benefit.Furthermore, our results partially contrast with the results of a recent comparative cohort study by Korbila et al [18].That study revealed a statistical significant difference in favor of combination treatment regarding disease resolution.However, Korbila and colleagues did not find improvements in the mortality rate between the 2 study groups, as was also shown by the data from our study.
The microbiological results showed that AS therapy had no impact on bacterial growth.It should be noted that 45% of the patients in our study had confirmed pathogen eradication, a percentage considerably lower than that previously reported with inhaled colistin [11,15,17,26] but comparable with data from reports including patients who received only IV colistin [10].
In terms of clinical outcome, there was a marginal benefit of combined AS-IV colistin therapy for clinical cure (P p .05).However, although several clinical factors that could affect patient's outcome appears to be present in higher percentages among patients who received IV colistin alone, a multivariate analysis including potential confounding factors showed no difference between the 2 treatment groups.Moreover, clinical success of VAP infection did not reveal any significant difference and was in accordance with previous reports for both IV [9] and AS colistin use [11,15,17,26].
Both VAP-related mortality and the all-cause mortality rates did not differ between the 2 groups, and the mortality rates were in agreement with those reported in the literature [10,11,15,17,18,26].
With regard to adverse events, the incidence of nephrotoxicity-the major limiting factor in the use of colistin in the past-was found to be lower in recent studies than in studies from the 1960s and early 1970s [10,11,16,17,26].In the present study, renal dysfunction was observed in 19% of the patients, which is comparable to the rate in previous reports [10,11,15,17].None of the patients discontinued colistin treatment or required renal replacement therapy because of renal dysfunction.It is noteworthy that the renal dysfunction rate was similar in the 2 treatment groups, suggesting that the addition of AS colistin does not cause systemic adverse events.Neurotoxicity and neuromuscular blockade, which were frequently reported in early studies from the 1960s, was been observed among the patients in our study.Furthermore, no major toxicity of colistin administered via the respiratory tract (bronchospasm, chest tightness, or apnea) was noted, a finding in agreement with other studies [11,15,16,26], with the exception of the study by Kwa et al [17], which reported that 1 AS colistin recipient experienced bronchospasm.
There are some drawbacks regarding the AS use of antimicrobials.Although, in normal lungs, AS antimicrobials have shown good penetration, little is known about the penetration of these agents into infected tissues from the airway lumen [28].Another concern is the impending emergence of resistant strains due to the use of inhaled antibiotics [29].In the present study, no colistin-resistant pathogens were isolated after the AS treatment.
This report has several limitations.First, this is a singlecenter, retrospective study with a relatively small number of patients and is thus vulnerable to b error.Second, relatively few P. aeruginosa and K. pneumoniae organisms were isolated.Therefore, our results provide limited information and cannot be generalizable with regard to these 2 pathogens.Finally, we did not monitor the volume of respiratory secretions, which is a marker of airway inflammation [25].However, to our knowledge, this is the first matched case-control study to have tried to address the role of AS in addition to IV colistin in patients with VAP.Also, to our knowledge, this study and the study by Korbila et al [18] are the only studies to have directly compared AS-IV colistin with IV colistin for the treatment of VAP.
In conclusion, the present study has revealed that the addition of AS colistin did not add any clinical, microbiological, or survival benefit for patients with VAP caused by gram-negative MDR pathogens susceptible only to colistin.Because of the differences between the present findings and those of other investigators, randomized, controlled trials are needed to examine the efficacy and safety of AS colistin therapy in addition to IV treatment in patients with VAP.