Background The optimal duration of antibiotic therapy for ventilator-associated pneumonia (VAP) remains controversial. Although current guidelines recommend 7–8 days of treatment, prolonged antibiotic use is still common in clinical practice. However, objective criteria to individualize and safely shorten treatment duration remain poorly defined. The aim of this study was to identify early clinical and laboratory response parameters associated with eligibility for short-term antibiotic therapy and to evaluate the predictive performance of composite criteria assessed on day 5 of treatment. Methods This retrospective study included intensive care unit (ICU) patients diagnosed with ventilator-associated pneumonia (VAP) according to Centers for Disease Control and Prevention (CDC) criteria between January 1, 2015 and June 30, 2018. A total of 147 VAP episodes in 111 patients with complete clinical and laboratory follow-up data were categorized according to treatment duration into the short-term treatment group (STTG; 7–10 days) and long-term treatment group (LTTG; > 10 days). Clinical and laboratory response parameters were assessed on treatment days 3, 5, and 7. Results Among 147 VAP episodes, 35 received 7–10 days and 112 received > 10 days of antibiotic therapy. Baseline demographic characteristics, APACHE II scores, CPIS values, PaO₂/FiO₂ ratios, and laboratory parameters at diagnosis were largely comparable between groups. On day 5 of treatment, the STTG had significantly lower rates of fever or hypothermia (8.6% vs. 33%; p = 0.004), lower CPIS scores (3.8 ± 1.2 vs. 5.0 ± 2.1; p = 0.003), lower CRP levels (10.7 ± 8.5 vs. 14.2 ± 8.5 mg/dL; p = 0.030), lower procalcitonin levels (2.0 ± 3.6 vs. 6.5 ± 12.5 μg/L; p = 0.009), and higher PaO₂/FiO₂ ratios (335.6 ± 63 vs. 274.9 ± 95.4 mmHg; p < 0.001). Appropriate empirical therapy was more frequent in the STTG (91.4% vs. 71.4%; p = 0.021), whereas VAP-associated bacteremia was less common (8.6% vs. 33.9%; p = 0.003). Seven criteria associated with eligibility for short-term therapy were identified: absence of fever on day 5, procalcitonin ≤ 0.5 μg/L, CRP < 15 mg/dL, PaO₂/FiO₂ > 300 mmHg, CPIS ≤ 4 on day 5, absence of bacteremia, and appropriate empirical therapy. The presence of ≥ 4 criteria yielded a sensitivity of 91.4% and specificity of 50.9%. When PaO₂/FiO₂ > 300 mmHg was included among these criteria, specificity increased to 75%. The presence of ≥ 5 criteria was associated with the highest predictive performance, with an AUC of 0.839, sensitivity of 82.9%, and specificity of 74.1%. Conclusions In patients with VAP receiving appropriate empirical therapy and without bacteremia, early clinical and laboratory response parameters assessed on day 5 —including fever status, procalcitonin, CRP, PaO₂/FiO₂ ratio, and CPIS— may help identify patients potentially suitable for shorter antibiotic therapy. The proposed criteria should be considered exploratory and require prospective validation before routine clinical implementation.

Individualized antibiotic therapy in patients with ventilator-associated pneumonia in the intensive care unit

期刊论文
出版模式：

开放获取
发表日期：

2026年6月23日

查看全文

Serhat Atalar,
Gürdal Yılmaz,
Ahmet Eroğlu,
İftihar Köksal

摘要

Background

The optimal duration of antibiotic therapy for ventilator-associated pneumonia (VAP) remains controversial. Although current guidelines recommend 7–8 days of treatment, prolonged antibiotic use is still common in clinical practice. However, objective criteria to individualize and safely shorten treatment duration remain poorly defined. The aim of this study was to identify early clinical and laboratory response parameters associated with eligibility for short-term antibiotic therapy and to evaluate the predictive performance of composite criteria assessed on day 5 of treatment.

Methods

This retrospective study included intensive care unit (ICU) patients diagnosed with ventilator-associated pneumonia (VAP) according to Centers for Disease Control and Prevention (CDC) criteria between January 1, 2015 and June 30, 2018. A total of 147 VAP episodes in 111 patients with complete clinical and laboratory follow-up data were categorized according to treatment duration into the short-term treatment group (STTG; 7–10 days) and long-term treatment group (LTTG; > 10 days). Clinical and laboratory response parameters were assessed on treatment days 3, 5, and 7.

Results

Among 147 VAP episodes, 35 received 7–10 days and 112 received > 10 days of antibiotic therapy. Baseline demographic characteristics, APACHE II scores, CPIS values, PaO₂/FiO₂ ratios, and laboratory parameters at diagnosis were largely comparable between groups.

On day 5 of treatment, the STTG had significantly lower rates of fever or hypothermia (8.6% vs. 33%; p = 0.004), lower CPIS scores (3.8 ± 1.2 vs. 5.0 ± 2.1; p = 0.003), lower CRP levels (10.7 ± 8.5 vs. 14.2 ± 8.5 mg/dL; p = 0.030), lower procalcitonin levels (2.0 ± 3.6 vs. 6.5 ± 12.5 μg/L; p = 0.009), and higher PaO₂/FiO₂ ratios (335.6 ± 63 vs. 274.9 ± 95.4 mmHg; p < 0.001). Appropriate empirical therapy was more frequent in the STTG (91.4% vs. 71.4%; p = 0.021), whereas VAP-associated bacteremia was less common (8.6% vs. 33.9%; p = 0.003).

Seven criteria associated with eligibility for short-term therapy were identified: absence of fever on day 5, procalcitonin ≤ 0.5 μg/L, CRP < 15 mg/dL, PaO₂/FiO₂ > 300 mmHg, CPIS ≤ 4 on day 5, absence of bacteremia, and appropriate empirical therapy. The presence of ≥ 4 criteria yielded a sensitivity of 91.4% and specificity of 50.9%. When PaO₂/FiO₂ > 300 mmHg was included among these criteria, specificity increased to 75%. The presence of ≥ 5 criteria was associated with the highest predictive performance, with an AUC of 0.839, sensitivity of 82.9%, and specificity of 74.1%.

Conclusions

In patients with VAP receiving appropriate empirical therapy and without bacteremia, early clinical and laboratory response parameters assessed on day 5 —including fever status, procalcitonin, CRP, PaO₂/FiO₂ ratio, and CPIS— may help identify patients potentially suitable for shorter antibiotic therapy. The proposed criteria should be considered exploratory and require prospective validation before routine clinical implementation.