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


In this study, a short course of prophylactic aerosolized ceftazidime in critically ill trauma patients decreased the frequency of VAP by 73% at ICU day 14 and by 54% for the entire ICU stay. The decrease in VAP seen in this study was similar to the effects seen in other successful trials of prophylactic aerosolized antibiotics.[19,20] The overall frequency of VAP in the placebo group was high (65%), but consistent with the injury severity observed in the study population.[12] No statistically significant difference in VAP was observed during the first week of ICU stay. A larger study population would be needed to show a significant decrease in the frequency of VAP at that time point. Visual examination of the Kaplan-Meier estimation for VAP development over time seems to indicate an approximately 10-day window after the end of study treatment when VAP is markedly prevented by aerosolized ceftazidime (Figure 2). If such a protective period exists, possible mechanisms include persistence of therapeutic ceftazidime concentrations in the lung owing to slow clearance of antibiotics relative to systemic clearance and/or attenuation of the proinflammatory cascade.[21,38,39]

A key finding is that aerosolized ceftazidime did not adversely affect bacterial flora and sensitivity patterns ( Table 2 ). Adverse effects on bacterial flora is a primary reason that VAP chemoprophylaxis has not been widely accepted.[33,40,41,42] Limits on patient selection and duration of therapy were the major steps taken to avoid adverse changes in ICU flora in this study, and they appear to have been successful. Compared with previous data from the study site, the percentage of VAP caused by P. aeruginosa or Acinetobacter calcoaceticus was approximately inverted. However, the overall frequency for these two organisms remained the same. These findings parallel recent changes documented in our hospital.[36]

An additional benefit of prophylactic aerosolized ceftazidime is decreased treatment with therapeutic systemic antibiotics by an average of 7 days ( Table 3 ). This decrease in antibiotic treatment was likely due to a lower frequency of VAP in the ceftazidime group. Other secondary outcomes including duration of mechanical ventilation, length of ICU stay, and mortality were not affected by aerosolized ceftazidime ( Table 3 ); however, this study was not designed or powered to detect such differences.

A potential limitation of the study is that the placebo group may have been at a higher risk for VAP at baseline. Although no statistically significant differences in type or severity of injury were observed between study groups at baseline, the higher number of patients with pulmonary contusions and rib fractures in the placebo group may be of clinical significance. A subgroup analysis showed that aerosolized ceftazidime did not decrease the frequency of VAP in patients with chest trauma but was effective in patients with other injuries. These data suggest that preventing VAP may be more difficult in patients with chest trauma than in those with other types of trauma, possibly due to impaired distribution of ceftazidime to injured lung fields. However, these results must be interpreted cautiously owing to low event numbers and use of post hoc analysis.

A strength of this study is its robust design, including randomization, blinding, placebo control, limited duration of therapy, selection of high-risk patients, and use of quantitative cultures in the definition of VAP. Previous trials of this type lacked desirable elements (e.g., blinding, randomization), did not include quantitative cultures in the definition of pneumonia, had indiscriminate patient selection, or had suboptimal drug administration. In this study, careful attention was paid to drug administration, including a nebulizer and ventilator combination that would be expected to provide good drug delivery.[43] This is important because using an appropriate nebulizer, ventilator, and administration technique can increase the amount of a nebulized dose that reaches the distal airways of the lung by up to 500%.[32,33] The administration technique used in this study produced high ceftazidime concentrations in the distal pulmonary tree and undetectable serum concentrations, and was well tolerated.

Ceftazidime was chosen for this study because it was known to be active against most gram-negative organisms that cause VAP in our ICU, and limited pharmacokinetic data were available to guide dosing.[36,38] Aminoglycosides were not chosen because they lack gram-positive activity and have poor efficacy as monotherapy for serious infections.[44,45] In this study, pulmonary ceftazidime concentrations observed a mean of 7 hours after dosing were several times higher than previously reported peak pulmonary concentrations from intravenous administration.[38] Noteworthy is that significant interpatient variability existed, and patients who developed VAP tended to have lower concentrations early in the dosing interval than those of patients who did not develop VAP (Figure 3). Although the dosing regimen studied seems acceptable based on the clinical results, further optimization may be needed.

In this study, aerosolized ceftazidime appeared to attenuate the increase in pulmonary TNF- , IL-1 , and IL-8 during the first week of ICU stay. These findings are presumed to be beneficial since other critically ill populations have been shown to have poor outcomes when pulmonary cytokines remain elevated over time.[26] Increases in TNF- and IL-1 were related to an increased frequency of VAP, whereas decreases in these markers were related to a decreased frequency of VAP. These results are similar to previous data that showed increases in pulmonary TNF- , IL-6, and IL-8 are predictive of subsequent development of pneumonia.[24,25] Plasma cytokine changes were similar between the groups and did not parallel the BAL results. These findings were similar to those of previous trials of patients with thoracic trauma or pneumonia.[27,46,47]

Unfortunately, interpretation of our data is somewhat limited by baseline disparities in BAL TNF- , IL-1 , and IL-8 concentrations ( Table 4 ). This finding was surprising given the clinical similarities between groups including the timing of the baseline BAL. Indeed, the opposite may have been expected since the placebo group may have had more thoracic trauma at baseline.[46] It is possible that varying antiinflammatory processes in this group may have attenuated the proinflammatory response; however, no such mediators (e.g., IL-4, IL-10) were measured. It could be argued that pulmonary cytokines had more of an opportunity to decrease over time in patients treated with ceftazidime because of higher baseline values. Nonetheless, the results were relatively consistent for TNF- , IL-1 , and IL-8 over time and seem to indicate that patients receiving aerosolized ceftazidime had an attenuation of these cytokines.

A second limitation is the wide variability in the pulmonary cytokine concentrations. This is common in studies of this type and may be due to factors such as interpatient genetic differences or type of injury.[25,26,27] As other authors have done, the variability was accounted for in this study by using nonparametric statistical methods for data that were not normally distributed.[25] Finally, it is unknown why attenuation of the most pleotropic mediators (TNF- , IL-1 ) had a relatively smaller effect on cytokines later in the inflammatory cascade (IL-6, IL-8).

Despite these limitations, the results of this study add to the growing body of research that suggests antibiotics may have advantageous immunomodulatory effects independent of antimicrobial activity. Cephalosporins have been shown to directly attenuate TNF- and IL-6 release in vitro, and imipenem decreased bacterial-induced lung injury in an animal model.[23,48] Interestingly, aerosolized imipenem was superior to intravenous administration for preventing bacterial-induced lung injury.[22] The authors speculated that this effect may have been due to higher concentrations achieved with aerosol administration. A recent clinical trial showed decreased pulmonary neutrophil concentrations in patients with mechanical ventilation who received aerosolized aminoglycosides.[21] An alternative explanation for the cytokine changes seen in this study could be that the antimicrobial effect of ceftazidime did not allow bacterial growth and subsequent activation of the proinflammatory cascade. Further studies are required to fully elucidate the contribution of immunomodulation to the efficacy of aerosolized ceftazidime.