Ventilator-associated Infection: The Role for Inhaled Antibiotics

Lucy B. Palmer


Curr Opin Pulm Med. 2015;21(3):239-249. 

In This Article

Clinical Trials Using Inhaled Therapy to Treat Ventilator-associated Tracheobronchitis

There are only two RCTs examining the effect of aerosolized antibiotics targeted on VAT, and both are by the same authors.[34,50] Patients were not excluded if they had co-existing VAP as the VAT-A definition was used. Prior work by this group had shown that with the optimization of delivery, peak antibiotic concentrations in respiratory secretions were 200-fold greater than the levels achieved in the blood of patients receiving systemic therapy.[51] The investigators hypothesized that treating the proximal airway infection (VAT-A) with very high concentrations of antibiotics would reduce or prevent signs and symptoms of respiratory tract infection and might also decrease bacterial resistance by eliminating bacterial populations.[34] Patients were randomized to aerosolized antibiotic or placebo targeted at the organisms on Gram stain in their tracheal aspirates. Systemic antibiotics were given by the responsible physicians, as indicated by clinical signs and symptoms of pneumonia. Both groups were on similar amounts of appropriate systemic antibiotics for the organism(s) in the tracheal aspirate. Patients receiving aerosolized antibiotics had significantly decreased signs of respiratory infection, were extubated more often, had decreased need for additional antibiotics, and had decreased bacterial resistance at the end of the treatment. In the seven patients with only VAT-A, none progressed to VAP. These findings suggest an important role for aerosolized therapy in the ICU for the treatment of patients with VAT and VAP.

Most recently, in another randomized, double-blind, placebo-controlled study, critically ill intubated patients with prolonged mechanical ventilation were randomized if they exhibited signs of respiratory infection (purulent secretions and Clinical Pulmonary Infection Score ≥ 6) and VAT-A.[50] Using a well characterized jet nebulizer aerosol delivery system with humidification turn off, inhaled antibiotic or saline placebo was given for 14 days or until extubation. The responsible clinician determined administration of systemic antibiotics for VAP and any other infection. Compared with placebo, inhaled antibiotics significantly reduced Clinical Pulmonary Infection Score (mean ± SEM, PRE 9.3 ± 2.7 to POST 5.3 ± 2.6 vs. PRE 8.0 ± 23 to POST 8.6 ± 2.10; P = 0.0008), and the volume of secretions (mean ± SEM, PRE 6.9 ± 4.7 ml/4 h decreased to POST 1.1 ± 1.3 ml/4 h vs. PRE 8.80.69 ml/4 h to POST 6.3 ± 4.3 ml/4 h; P < 0.001). The effects on bacterial growth are shown in Fig. 2.

Figure 2.

Bacterial growth from tracheal aspirates obtained at time of randomization, mid-treatment (Mid-Tx), and at end of treatment (EOT) for (a) aerosolized antibiotics (AA) and (b) placebo. Growth is quantified using a graded scale 0–4 from semiquantitative cultures: multidrug-resistant Gram-negative organisms (filled circles), nonresistant Gramnegative organisms (open circles), resistant Gram-positive organisms (filled squares), nonresistant Gram-positive organisms (open squares), and newly resistant organisms on treatment (X). Some patients had multiple isolates. At Mid-Tx, all the isolates with zero growth represent organisms detected at randomization that did not grow in isolates sampled at Mid-Tx. At EOT, the isolates with zero growth represent organisms detected at randomization and Mid-Tx that did not grow in samples obtained at EOT. There was a clear difference in pattern of bacterial growth between AA and placebo. Two AA isolates demonstrated persistent growth at EOT: one methicillin-resistant Staphylococcus aureus (filled square) that was not eradicated by AA, but had no Gram-positive cocci on Gram stain, and one persistent Acinetobacter (filled circle) with organisms present on Gram stain. More newly resistant organisms were seen in the placebo group. Reproduced from Palmer et al. [50].

Aerosolized Colistin for Ventilator-associated Infection

Highly resistant P. aeruginosa and A. baumanii have led to the reintroduction of colistin (polymyxin E) in an aerosolized form, as well as its prodrug, colistimethate sodium (CMS). Colistin's bactericidal activity works via destabilization of the lipopolysaccharide (LPS) of the outer membrane, and in addition, it neutralizes the LPS, thereby decreasing antiendotoxin activity.[52] Its i.v. use was discontinued for about 40 years because of its neurological and renal toxicity when used parenterally and the advent of less toxic antibiotics. There have been multiple small nonrandomized clinical trials, one RCT, one review, and one meta-analysis focused on aerosolized and i.v. colistin treatment for MDR Gram-negative species, in particular, Acinetobacter spp. and Pseudomonas spp..[20,22–25,27,28,43,45,46,48,49,53–56] Both these bacteria produce extended-spectrum β-lactamases, as well as metallo-β-lactamases. Acinetobacter is often sensitive only to polymyxin B or colistin (polymyxin E), and there are now reports of colistin resistance as well.

Studies of adjunctive therapy from 2007 through 2011 have been reviewed previously and will not be covered in detail here.[31,57] Recent colistin studies are included in Table 2 .

In a retrospective cohort study, Arnold et al.[45] assessed the treatment of VAP caused by P. aeruginosa and A. baumannii VAP. Patients with only i.v. therapy (n = 74) were compared to those who received adjunctive inhaled colistin (n = 19) and adjunctive inhaled tobramycin (n = 10). This was a retrospective study and the cohort group had a much lower Acute Physiology and Chronic Health Evaluation II (APACHE II) than the group that received inhaled therapy. Drugs were aerosolized via a nebulizer that generated particles that were 1–5 μm for the delivery of the aerosolized antibiotic over a 15–20-min time period.

Those treated with inhaled antibiotics had more MDROs (52.6 vs. 14.9%; P = 0.001) and higher APACHE II scores (21.4 ± 5.7 vs. 17.5 ± 5.3; P = 0.004). Despite these marked differences in patient acuity and bacterial susceptibility, the Kaplan–Meier curves for the probability of 30-day survival from VAP onset demonstrated that patients receiving an aerosolized antibiotic had statistically greater survival (P = 0.030 by the log-rank test).

The effects of high-dose nebulized colistin (5 million international units every 8 h) were investigated by Lu et al.[46] in a prospective observational comparative study testing the efficacy of nebulized colistin for treating VAP caused by MDR Gram-negative vs. susceptible organisms treated with i.v. antibiotics.[46] One hundred and sixty-five patients with VAP caused by P. aeruginosa and A. baumannii were enrolled. There were 122 patients having VAP caused by P. aeruginosa and A. baumanii susceptible to β-lactams, aminoglycosides, or quinolones and treated with i.v. antibiotics for 14 days. The second group included 43 patients having VAP caused by MDR P. aeruginosa and A. baumannii and treated with nebulized colistin (5 million international units every 8 h) either in monotherapy (n = 28) or combined to a 3-day i.v. aminoglycoside treatment for 7–19 days. The primary endpoint was clinical cure rate. Aerosol was delivered using vibrating plate nebulizer with no humidification. The clinical cure rate was 66% in the sensitive strain group and 67% in the MDR strain group (difference -1%, lower limit of 95% CI for difference -12.6%). There was no difference in the inhaled monotherapy group vs. the inhaled therapy + 3 days of i.v. aminoglycoside therapy. Mortality was not different between groups. This investigation demonstrated a therapeutic effect that was noninferior to i.v. β-lactams associated with aminoglycosides or quinolones for treating VAP caused by susceptible P. aeruginosa and A. baumannii. This study raises the question: If the patient had received both i.v. and inhaled therapy throughout the treatment, would the outcome have been more robust?

The question was addressed by Tumbarello et al.[49] in a retrospective 1 : 1 matched case-control study, to evaluate the efficacy and safety of aerosolized + i.v. colistin vs. i.v. colistin alone in 208 patients in the ICU, with VAP caused by colistin only susceptible (COS) A. baumannii, P. aeruginosa, or Klebsiella pneumoniae. The medication was delivered with jet or ultrasonic nebulizers. The aerosolized antibiotic-intravenous (AA-i.v). colistin cohort had a higher clinical cure rate (69.2 vs. 54.8%; P = 0.03) and required fewer days of mechanical ventilation after VAP onset (8 vs. 12 days; P = 0.001). One hundred and sixty-six patients had post-treatment cultures. Eradication of the causative organism was more common in the AA-i.v. colistin group (63.4 vs. 50%; P = 0.08), although the difference was not significant. No differences in ICU mortality, length of ICU stay after VAP onset, or rates of acute kidney injury (AKI) during colistin therapy were seen between the i.v. or the AA-i.v. arms.

In a similar study, organisms were MDR, but not exclusively COS. Doshi et al.[48] conducted a retrospective multicenter cohort study comparing i.v. colistin alone vs. colistin given in aerosolized and i.v. forms. Baseline characteristics were similar between the two groups. Twenty patients (39.2%) receiving i.v. and 24 (54.5%) receiving i.v. + aerosolized colistin achieved clinical cure (P = 0.135). There was no difference in microbiologic cure rates between the i.v. and the i.v. + aerosolized colistin groups (40.7 vs. 44.4%; P = 0.805). The i.v. group demonstrated a trend towards higher mortality (70.4 vs. 40%; P = 0.055) attributable to pneumonia.

In the subgroup analysis of patients with high-quality respiratory cultures (bronchoalveolar lavage), there was a significantly higher clinical cure rate for those in the i.v. + aerosolized group compared to the i.v. group (57.1 vs. 31.3%; P = 0.033).