Aerosolized Ceftazidime for Prevention of Ventilator-Associated Pneumonia and Drug Effects on the Proinflammatory Response in Critically Ill Trauma Patients

G. Christopher Wood, Pharm.D., Bradley A. Boucher, Pharm.D., FCCP, Martin A. Croce, M.D., Scott D. Hanes, Pharm.D., Vanessa L. Herring, B.S., Timothy C. Fabian, M.D.


Pharmacotherapy. 2002;22(8) 

In This Article


A total of 59 patients were enrolled between October 1998 and June 2000. Seventeen patients were excluded from the analysis because they were extubated (10 patients), developed VAP (6 patients), or died (1 patient) before receiving 3 days of study treatment. One patient was excluded because treatment started more than 48 hours after admission, and one patient was excluded because of an invalid informed consent agreement. Forty patients completed the study and were eligible for evaluation. Patient characteristics are summarized in Table 1 . No statistically significant differences were observed between the groups in demographics or selected clinical factors that may influence VAP development (e.g., systemic antimicrobial therapy). However, clinically significant demographic differences may have existed, particularly with more patients with chest trauma in the placebo group than in the ceftazidime group. Alternatively, there were two more patients with severe traumatic brain injury in the ceftazidime group than in the placebo group.

The frequency of VAP is depicted in Figure 1. A total of 22 episodes of VAP occurred in 19 patients. All three patients who developed a second episode of VAP were in the placebo group. No statistically significant difference was noted in the number of patients developing VAP by ICU day 7 (p=0.22). However, for patients receiving aerosolized ceftazidime, significant reductions in VAP were noted at ICU day 14 (p=0.021) and throughout the ICU stay (p=0.022). A Kaplan-Meier estimation for VAP development over time in each treatment group is depicted in Figure 2. A post hoc analysis showed that VAP development in patients with chest trauma was similar between groups (4 of 9 vs 9 of 15, p>0.05). However, VAP development in other patients was lower in the ceftazidime group (2 of 12 vs 6 of 7, p=0.006). No differences in efficacy were found based on the duration of study therapy, and no adverse effects of the study treatment were reported.

Frequency of ventilator-associated pneumonia (VAP) by day 7, day 14, and duration of intensive care unit (ICU) stay for the ceftazidime (white bar) and placebo (black bar) groups.

Kaplan-Meier estimation for the frequency of ventilator-associated pneumonia (VAP) over time. Overall frequency in the ceftazidime group (dashed line) was 54% lower than that in the placebo group (solid line) (p=0.022).

The types of organisms causing VAP in the study patients were similar to those that historically cause VAP in the Presley Trauma Center ICU ( Table 2 ).[31] Thirty five organisms were isolated in quantities of at least 105 cfu/ml. Polymicrobial VAP developed in 1 of 6 patients in the ceftazidime group compared with 8 of 13 patients receiving placebo (p=0.14). Ceftazidime sensitivity patterns of gram-negative bacilli causing VAP in study patients were similar to those observed in the overall population of the Presley Trauma Center ICU.[36]

Duration of mechanical ventilation, ICU length of stay, total antibiotic therapy, and mortality were not significantly different between treatment groups ( Table 3 ). However, the mean duration of antimicrobial therapy to treat documented infections (therapeutic antibiotics) was 7 days shorter in the ceftazidime group than that in the placebo group (p=0.024).

Ceftazidime concentrations were detectable at follow-up BAL in 16 of 19 patients in the ceftazidime group (median 56 µg/ml, range 2.1-443.9 µg/ml) (Figure 3). Bronchoalveolar lavage procedures for determining ceftazidime concentrations were performed a mean of 7 hours (range 2.5-16 hrs) after the last dose of study drug. Three of 16 patients with measurable ceftazidime had concentrations less than the breakpoint for ceftazidime sensitivity (8 µg/ml); however, none of these patients developed VAP. One patient with an undetectable ceftazidime concentration developed VAP. Serum ceftazidime concentrations were undetectable in 17 of 18 patients and relatively low (1.2 µg/ml) in 1 patient.

Bronchoalveolar lavage (BAL) ceftazidime concentrations versus time after dosing. = patients who did not develop VAP; = patients who developed VAP.

Bronchoalveolar TNF- , IL-1 , IL-8, and IL-6 concentrations at each sampling time are reported in Table 4 and depicted in Figure 4. Four patients, two in each group, did not have complete BAL results because of death in one patient, hemodynamic instability in one, or undetectable urea concentrations in BAL fluid that precluded dilutional correction in two patients. The TNF- , IL-1 , and IL-8 concentrations were significantly attenuated in the ceftazidime group compared with the increases from baseline seen in the placebo group (p<0.001, p=0.02, p=0.003, respectively). Pulmonary IL-6 decreased in both treatment groups over time. Changes in TNF- and IL-1 over time were related directly to subsequent development of VAP (p<0.001 and p=0.023, respectively). The change in IL-8 showed a similar trend (p=0.08). The change in IL-6 was not related to VAP development.

Baseline and follow-up BAL cytokine concentrations in the ceftazidime group and the placebo group. Boxes represent median (indicated by a line) and 25-75% ranges. Whiskers are 10% and 90% ranges. Individual points are outliers beyond 10% and 90%.

In patients who developed VAP, there was no effect on cytokine concentrations based on the type of causative organism (gram-positive, gram-negative, mixed) or colony count from the BAL culture. Plasma cytokine concentrations and their change over time are summarized in Table 5 . Plasma samples could not be obtained in one patient in each group. Plasma cytokine concentrations were substantially lower than BAL concentrations, and no statistically significant differences were noted between the groups.