Trauma and Antimicrobial Resistance Are Independent Predictors of Inadequate Empirical Antimicrobial Treatment of Ventilator-Associated Pneumonia in Critically Ill Patients

Abstract

Background:

We aimed to assess independent risk factors for inadequate initial antimicrobial treatment (IAT) in critically ill patients with ventilator-associated pneumonia (VAP) treated in intensive care units (ICU) and to determine whether IAT is associated with adverse outcomes in patients with VAP.

Patients and Methods:

A prospective cohort study was performed and included 152 patients with VAP treated in an ICU for more than 48 hours. The main outcomes of interest were all-cause ICU mortality and VAP-related mortality. Other outcomes considered were: intra-hospital mortality, VAP-related sepsis, relapse, re-infection, length of stay in ICU (ICU LOS), and number of days on mechanical ventilation (MV).

Results:

One-third of patients (35.5%) received inadequate antimicrobial therapy. Trauma (odds ratio [OR], 3.55; 95% confidence interval [CI], 1.25–10.06) and extensively drug-resistant (XDR) causative agent (OR, 3.09; 95% CI, 1.23–7.74) were independently associated with inadequate IAT. Inadequate IAT was associated with a higher mortality rate (OR, 3.08; 95% CI, 1.30–7.26), VAP-related sepsis (OR, 2.39; 95% CI, 1.07–5.32), relapse (OR, 3.25; 95% CI, 1.34–7.89), re-infection (OR, 6.06; 95% CI, 2.48–14.77), and ICU LOS (β 4.65; 95% CI, 0.93–8.36). Acinetobacter spp., Pseudomonas aeruginosa and Klebsiella/Enterobacter spp. were the most common bacteria in patients with IAT and those with adequate antimicrobial therapy.

Conclusions:

This study demonstrated that inadequate IAT is associated with a higher risk of the majority of adverse outcomes in patients with VAP treated in ICUs. Trauma and XDR strains of bacteria are independent predictors of inadequate IAT of VAP in critically ill patients.

Ventilator-associated pneumonia (VAP) is the most common health-care–associated infection (HAI) in intensive care units (ICU). The incidence of VAP is 10%–30% among patients who require mechanical ventilation for longer than 48 hours [1]. Although mortality rates differ between hospitals, it is recognized that one-third to one-half of all hospital-acquired pneumonia (HAP)–related deaths are directly attributable to pneumonia [2].

Several risk factors associated with mortality of HAP and VAP have been recognized. In the literature, one of the most evident risk factor is the adequacy of initial antibiotic treatment (IAT) [3 –6]. Antimicrobial drugs are widely prescribed for patients on ventilators in the ICU because of the risk of VAP. International societies, such as the American Thoracic Society and Infectious Diseases Society of America (ATS/IDSA), European Respiratory Society, European Society of Intensive Care Medicine, European Society of Clinical Microbiology and Infectious Diseases, and Asociación Latinoamericana del Tórax (ERS/ESICM/ESCMID/ALAT) have developed guidelines for VAP treatment [1,7 –9]. Despite this, the mortality rates for patients with VAP who receive inadequate IAT is still high. In a multicenter study conducted in Central Europe, the 30-day mortality of patients with VAP receiving inadequate IAT was as high as 44.8% and considerably higher than in patients who had received adequate IAT [10].

Multi-drug–resistant (MDR) bacteria are the primary reason for IAT being inadequate in critically ill patients [11]. As previously shown, MDR gram-negative bacteria are the leading cause of early and late onset VAP and blood stream infections in severely injured ICU patients, which can be possibly associated with higher mortality in these patients [12 –14].

Therefore, the main aim of this study was to assess independent risk factors for inadequate IAT in critically ill patients as well as to determine if there is an association between inadequate antimicrobial treatment and adverse outcomes in patients with VAP in major referral emergency center in Serbia.

Patients and Methods

Study design and population

The prospective cohort study was conducted in a multidisciplinary, 24-bed ICU at the University Emergency Centre, Clinical Centre of Serbia, Belgrade. All consecutive patients admitted to the ICU between February 2017 and March 2019 who required mechanical ventilation (MV) for longer than than 48 hours and who developed VAP were eligible for the study. Exclusion criteria were: patients younger than 18 years, patients with recorded gastric aspiration, patients who received antibiotic therapy in the previous 90 days, recently hospitalized patients, patients who resided in a nursing home or extended care facility, patients at home care (including antibiotic agents and wound care), and underlying malignancy. The study was approved by the University of Belgrade, Medical School Institutional Review Board (no. 29/IV-14).

Evaluation on admission

Data collected at admission included patients' sociodemographic characteristics, the mechanism of injury (blunt or penetrating trauma), trauma distribution (head, thoracic, abdomen, etc.), the Acute Physiology and Chronic Health Evaluation (APACHE) II score, the Abbreviated Injury Scale (AIS) score, the Injury Severity Score (ISS), and the patients' underlying diseases. Factors such as shock, coma (Glasgow Coma Scale [GCS] score <9), multiple transfusions, use of vasopressors, need for intubation or cardiopulmonary resuscitation on admission, emergency or elective surgery, and the presence of pulmonary contusions on chest radiograph at admission were recorded.

Diagnosis and definitions

The clinical, radiographic, and laboratory data of all patients were reviewed daily. The chest radiograhs were interpreted by a radiologist. The diagnosis of VAP was made by the intensive care physician based on chest radiograph, systemic infection, pulmonary criteria, and bacteriologic examination. The chest radiograph criteria included a new pulmonary infiltrate, worsening of radiographic findings in patients with a contusion, pleural effusion, or cavitation. The systemic criteria were temperature higher than 38.8°C, leukopenia (white blood cell count less than 4 × 10⁹ per milliliter), or leukocytosis (white blood cell count more than 12 × 10⁹ per milliter). The pulmonary criteria were new-onset purulent sputum (change in the sputum character, enhanced respiratory secretions, or increased suctioning requirement), worsening gas exchange (desaturation, increased oxygen requirement, or increased ventilator demand), new-onset or worsening cough, dyspnea or tachypnea, and rales or bronchial breath sounds.

Pneumonia was considered as early-onset if it occurred within two to four days. Patients with very early-onset pneumonia were defined as those who developed the infection within the first 48 hours and were not included in the study. Late-onset pneumonia began on day five or later after admission [7]. All patients were managed with a standardized protocol if VAP was suspected [7]. Relapse was defined as an episode of VAP caused by the same strain as initial infection whereas re-infection was defined as a new-onset of VAP occurred after eradication of the first strain and caused by a strain different from the initial infection.

In management of all patients with VAP, locally developed prevention bundles were used. The local bundle consisted of head elevation of bed (30–45 degrees), cuff pressure monitoring, daily sedation interruption, and assessment of wean readiness, oral care, hand hygiene, deep vein thrombosis prophylaxis, and peptic ulcer prophylaxis.

Disease severity was evaluated in all patients using the APACHE II score during the first 24 hours of ICU admission. The Sequential Organ Failure Assessment (SOFA) score and the Clinical Pulmonary Infection Score (CPIS) were measured on the day of VAP diagnosis. The CPIS considered clinical and radiographic data for determination of numerical value that predicts the presence or absence of VAP; a score higher than six had a good correlation with the presence of VAP [15]. Acute respiratory distress syndrome was diagnosed according to the Berlin definition criteria [16]. Shock was defined as systolic blood pressure lower than 90 mm Hg despite adequate fluid resuscitation and the need for vasopressor agents. Polytransfusion was defined as the need for more than 10 units of packed red blood cells within 24 hours. Pulmonary contusion was diagnosed by a radiologist by chest radiograph. The AIS was used to compare injuries. The AIS is the anatomic scale used most widely for rating severity of injury [17,18] and historically has been used in conjunction with the ISS to identify the effects of multiple injuries on trauma victims [19].

Before starting IAT, blood samples, tracheal aspirate, or bronchoalveolar lavage (BAL) for microbiology assessment were collected. The empiric antibiotic regimens for non-traumatized and traumatized patients were guided by local antibiograms and previous experience. Recommendations from guidelines were adjusted by intensive care physician, microbiologist, and epidemiologist. Patients received gram-positive antibiotic agents with methicillin-resistant Staphylococcus areus (MRSA) activity, such as vancomycin or linezolid, plus one gram-negative antibiotic with antipseudomonal activity, such as meropenem or cefepime. In patients with septic shock, one more gram-negative antibiotic with antipseudomonal activity was added, for example, amikacin or colistin. Traumatized patients with severe and extensive injury (especially with traumatic brain injury and chest injury) and high inflammatory parameters shortly after admission received vancomycin and meropenem earlier (in the first five days of admission) as our previous research has shown that such patients have a high risk of developing early VAP [13]. The length of antibiotic treatment usually lasts seven to 10 days except in situations that required longer antimicrobial treatment such as lack of improvement in laboratory parameters, clinical status, and chest radiogram.

Therapy was considered inadequate if no effective drug against the isolated pathogen(s) was included in the initial empirical antibiotic treatment within 24 hours of culture collection or the doses and pattern of administration were not in accordance with current medical standards. For patients with polymicrobial blood stream infections, all pathogens were required to be susceptible to the antimicrobial agents in the regimen [20]. Ventilator-associated-related mortality was determined by two clinicians (B.J. and A.H.) who have extensive clinical and research experience in VAP and critical care, and it was defind as death whose immediate or underlying cause was VAP or death in which VAP played a major role.

Outcomes

The main outcomes of interest were all-cause ICU mortality and VAP-related mortality. Other outcomes considered were: intra-hospital mortality, VAP-related sepsis, relapse, re-infection, length of stay in intensive care unit (ICU LOS), and number of days on MV.

Microbiologic assessment

Respiratory and blood samples were collected in all patients with signs and symptoms of VAP or sepsis. Respiratory samples for a bacteriologic examination were usually collected from tracheobronchial aspirates. Bronchoalveolar lavage (BAL) fluid was used only when BAL would not increase the hypoxia. Respiratory samples were quantitatively cultured according to standard microbiological laboratory procedures. Samples with less than 25 neutrophils or with more than 10 squamous epithelial cells in the field were defined as contaminated. Microbiologic confirmation of VAP was defined as the presence of at least one potentially pathogenic micro-organism in the respiratory sample at predefined thresholds (10⁴ colony forming units [CFU]/mL in BAL fluid or 10⁶ CFU/mL in a tracheobronchial aspirate). Samples with a number of micro-organisms below these thresholds were excluded because of low specificity. Change of antibiotic agents was based on the microbiologic findings of the BAL fluid or tracheobronchial aspirate. Strains that showed intermediate susceptibility and resistance to the specific antibiotic were considered resistant. Strains were denoted as multi-drug–resistant (MDR) if they showed non-susceptibility to at least one agent in three or more classes of antimicrobial drugs, extensively drug-resistant (XDR) as non-susceptibility to at least one agent in all but two or fewer antimicrobial categories, and pan-drug resistant (PDR) was defined as non-susceptibility to all agents in all antimicrobial categories [21].

Statistical analysis

Data are presented as the mean ± standard deviation or median (interquartile range [IQR]) for continuous variables and as the frequency (%) for categorical variables. The t-test or Man-Whitney U test was used for the two-group mean comparisons. Categorical data were compared using Pearson χ² test or Fisher exact test. Each patient characteristic (age, gender, and pre-existing medical conditions), injury characteristics and severity (type and number of injuries, injured body regions, AIS, and ISS), and clinical characteristics (presence of hemorrhagic shock, coma, or pulmonary contusion on admission and sustaining re-animation, multiple transfusions, surgery, or vasopressor use) were analyzed in a univariable logistic regression analysis. Parameters that were associated with IAT in the univariable analysis (p < 0.20) and based on clinical and biologic plausibility, were entered in a multivariable model, and stepwise logistic regression was performed. Odds ratios with 95% confidence intervals were computed, and Pearson goodness-of-fit test was performed to assess overall model fit. All parameters that entered the multivariable models were assessed for collinearity and tested for interaction terms. Best-fitting model was chosen based on the area under the receiver operating characteristics curve (C statistic) and the Hosmer–Lemeshow goodness-of-fit test. For the association of IAT with adverse outcomes we did not use propensity analysis, instead we utilized multivariable regression with models adjusted for parameters that would have made a propensity score because it has been shown that is not inferior to propensity score matching [22]. The statistical analysis was performed using Stata, version 13 (StataCorp, College Station, TX).

Results

Of 1,584 patients admitted to the ICU during the study period, 363 patients required MV for longer than 48 hours (69.1% were traumatized patients), of whom 152 (41.9%) developed VAP and were included in the analysis. Ventilator-associated pneumonia was confirmed by tracheal aspiration in 86 (56.58%) patients and by bronchoalveolar lavage in 66 (43.42%) patients. There were 78 (51.3%) patients with early-onset VAP and 74 (48.7%) with late-onset VAP. Fifty-four (35.5%) patients received inadequate IAT.

The mean age of patients was 51.6 ± 19.5 years and 75.2% were male. Half of all patients (49.9%) had one or more comorbidity, of which the most common were cardiovascular diseases (31.7%) and diabetes mellitus (7.7%) (Table 1). Frequency of comorbidities was different between treatment groups in patients with VAP.

Table 1.

Sociodemographic and Clinical Characteristics of Patients

	Appropriate IAT	Inappropriate IAT	RR (95% CI)
Characteristic	n = 98	n = 54
Age^a	53.8 ± 19.8	49.6 ± 21.1	—
Gender (male)	68 (69.4)	42 (77.8)	1.12 (0.92–1.36)
Presence of comorbidities	54 (55.1)	20 (37.0)	0.67 (0.45–0.99)
Comorbidity
COPD	4 (4.1)	1 (1.9)	0.45 (0.05–3.96)
Restrictive lung diseases	2 (2.0)	1 (1.9)	0.91 (0.08–9.78)
CKD	6 (6.1)	3 (5.6)	0.91 (0.24–3.48)
Cirrhosis	4 (4.1)	1 (1.9)	0.45 (0.05–3.96)
Immunocompromised	2 (2.0)	2 (3.7)	1.8 (0.26–12.52)
Diabetes mellitus	10 (10.2)	3 (5.6)	0.54 (0.16–1.89)
CVD	41 (41.8)	11 (20.4)	0.49 (0.27–0.87)
Reason for ICU admission
Post-operative admission	20 (20.4)	5 (9.3)	Ref.
Underlying disease	6 (6.1)	0	—
Trauma	72 (73.5)	49 (90.7)	2.02 (0.9–4.6)
Intubation
In ICU	31 (31.6)	15 (27.8)	Ref.
In ED or in OR	49 (50.0)	32 (59.3)	1.27 (0.82–1.98)
Another institution	18 (18.4)	7 (13.0)	0.76 (0.39–1.48)
Previous hospitalization	27 (27.6)	17 (31.5)	1.14 (0.69–1.90)
Stage on admission to ICU
Hemorrhagic shock	20 (20.4)	8 (14.8)	0.72 (0.34–1.54)
Vasopressors	18 (18.4)	6 (11.1)	0.60 (0.25–1.43)
Massive blood transfusion	29 (29.6)	16 (29.6)	1.00 (0.60–1.67)
Coma	35 (35.7)	25 (46.3)	1.30 (0.88–1.92)
Urgent surgery	36 (36.7)	16 (29.6)	0.81 (0.49–1.31)
Pulmonary contusion	21 (21.4)	18 (33.3)	1.56 (0.91–2.66)
VAP
Early	49 (50.0)	29 (53.7)	1.07 (0.78–1.47)
Late	49 (50.0)	25 (46.3)	Ref.
Polymicrobial infection	35 (35.7)	23 (42.6)	1.20 (0.78–1.85)
ISS	27.3 ± 13.5	30.1 ± 13.6
Trauma severity (ISS categories)
Mild and moderate (<16)	21 (29.2)	7 (14.3)	Ref.
Severe (16–24)	7 (9.7)	12 (24.5)	0.7 (0.28–1.62)
Critical (25–75)	44 (61.1)	30 (61.2)	1.6 (0.81–3.26)
GCS^a	9.0 (9)	8.0 (11)	—
APACHE II score^a	16.7 ± 5.6	16.3 ± 6.1	—
SOFA score^a	9.3 ± 3.3	9.3 ± 3.5	—
CPIS score^a	6.9 ± 1.4	6.8 ± 1.5	—

Bold values are statistically significant.

Values are number (%) if not specified otherwise.

IAT = initial antimicrobial treatment; RR = relative risk; CI = confidence interval; COPD = chronic obstructive pulmonary disease; CKD = chronic kidney disease; CVD = cardiovascular disease; ICU = intensive care unit; ED = emergency department; OR = operating room; VAP = ventilator-associated pneumonia; ISS = Injury Severity Score; GCS = Glasgow Coma Score; APACHE II = Acute Physiology and Chronic Health Evaluation II; SOFA = sequential organ failure assessment; CPIS = clinical pulmonary infection score.

Continuous variables are presented as mean ± standard deviation or median (interquartile range).

Table 2 shows the outcome of patients who did and those who did not receive adequate IAT. Substantially higher risk for intra-hospital mortality (RR, 1.67; 95% CI, 1.23–2.28), VAP-related sepsis (RR 2.0; 95% CI, 1.20–3.31), relapse (RR, 2.66; 95% CI, 1.51–4.69), and re-infection (RR, 3.15; 95% CI, 1.83–5.41) was observed for patients who received inadequate IAT. These patients also had longer ICU stays (17 vs. 13 days; p = 0.027) and days on MV (14 vs. 11 days; p <0.001) in comparison with patients who received adequate IAT.

Table 2.

Patient Outcomes by the Appropriateness of Initial Antimicrobial Treatment

	Appropriate	Inappropriate	RR (95% CI)
Outcome	n = 98	n = 54	RR (95% CI)
Hospital mortality	39 (39.8)	36 (66.7)	1.67 (1.23–2.28)
VAP-related mortality	22 (22.4)	15 (27.8)	1.24 (0.70–2.18)
VAP-related sepsis	20 (20.4)	22 (40.7)	2.0 (1.20–3.31)
Relapse	15 (15.3)	22 (40.7)	2.66 (1.51–4.69)
Re-infection	15 (15.3)	26 (48.1)	3.15 (1.83–5.41)
ICU LOS^a	13.0 (11)	17 (19)	—
Ventilator-free days^a	1 (6)	0 (4.3)	—

Bold values are statistically significant.

RR = relative risk; CI = confidence interval; VAP = ventilator-associated pneumonia.

ICU LOS = intensive care unit length of stay.

Values are median (interquartile range).

The isolated pathogens according to the adequacy of antimicrobial treatment are presented in Table 3. Acinetobacter spp., Pseudomonas aeruginosa, and Klebsiella/Enterobacter spp. were the most common bacteria in both groups. Interestingly, the percentage of MDR strains of Pseudomonas aeruginosa, Klebsiella pneumoniae, Serratia spp., and Providencia spp. were higher in the group of patients with adequate IAT whereas the percentage of MDR strains of Acinetobacter spp., Klebsiella/Enterobacter spp., Escherichia coli, and Proteus spp. were higher in patients who received inadequate IAT. However, rates of XDR isolates were higher for patients with inadequate IAT. All cases of relapse were caused by MDR and XDR Acinetobacter spp. and Pseudomonas aeruginosa, whereas re-infections were in addition caused by MDR and XDR MRSA (data not shown). Pseudomonas spp. was the most common cause of re-infection (63.4%), followed by Acinetobacter spp. (22%). Methicillin-resistant Staphylococcus aureus (9.8%) and Providentia spp. (4.9%) accounted for the rest of the re-infections. Comparing two groups of the initial therapy, we observed that Acinetobacter spp. was two times more common in the adequate IAT group (33.3% vs. 15.4%) and that rate of MRSA was almost two times higher in the inadequate IAT group (11.5% vs. 6.7%), whereas Pseudomonas spp. and Providentia spp. were similarly distributed in two groups. This is likely to be compatible with chance given the small numbers of reinfections and the result of the official statistical test (p = 0.419), and therefore should be interpreted with caution.

Table 3.

Microbiology of Causative Agents of VAP According to the Adequacy of Initial Antimicrobial Treatment

	Appropriate			Inappropriate
Pathogen	Total	MDR	XDR	Total	MDR	XDR
Acinetobacter spp.	58 (42.1)	36 (62.1)	3 (5.1)	37 (45.7)	25 (67.6)	8 (21.6)
Pseudomonas aeruginosa	24 (16.7)	11 (45.8)	6 (25)	8 (9.9)	3 (37.5)	5 (62.5)
Klebsiella/Enterobacter spp.	20 (14.5)	10 (50)	2 (10)	9 (11.1)	8 (88.9)	—
MRSA	15 (10.9)	15 (100)	—	9 (11.1)	9 (100)	—
MSSA	6 (4.4)	—	—	3 (3.7)	—	—
Klebsiella pneumoniae	3 (2.1)	2 (66.7)	—	5 (6.2)	3 (60)	—
Serratia spp.	2 (1.5)	2 (100)	—	3 (3.7)	1 (33.3)	—
Escherichia coli	1 (0.7)	—	—	2 (2.5)	1 (50)	1 (50)
Haemophilus influenzae	2 (1.4)	—	—	1 (1.2)	—	—
Providencia spp.	3 (2.2)	2 (66.7)	—	—	—	—
Sternotrophomonas	1 (0.7)	—	—	2 (2.5)	—	—
Streptococcus pneumoniae	2 (1.4)	—	—	—	—	—
Citrobacter	1 (0.7)	—	—	1 (1.2)	—	—
Moraxella catharalis	1 (0.7)	—	—	—	—	—
Proteus spp.	—	—	—	1 (1.2)	1 (100)	—
Total	139 (100)	77 (55.4)	11 (8.0)	81 (100)	50 (61.7)	15 (18.5)

There were no pan-drug–resistant (PDR) isolates.

VAP = ventilator-associated pneumonia; MDR = multi-drug resistant; XDR = extremely drug resistant; MRSA = methicillin-resistant Staphylococcus aureus; MSSA = methicillin-sensitive Staphylococcus aureus.

Of those parameters that were associated with inadequate IAT in univariable analysis (comorbidities, trauma, and level of antimicrobial resistance [AMR]), only trauma (odds ratio [OR], 3.55; 95% CI, 1.25–10.06) and XDR causative agent (OR, 3.09; 95% CI, 1.23–7.74) remained independently associated with inadequate IAT in the multivariable logistic regression (Table 4).

Table 4.

Univariate and Multivariate Logistic Regression of Predictors of Inappropriate Initial Antimicrobial Treatment

	Univariable		Multivariable
Characteristic	OR	CI	OR	CI
Comorbidities	0.479	0.24–0.95
Trauma	3.539	1.27–9.85	3.55	1.25–10.06
Level of AMR
Sensitive	ref.
MDR	1.97	0.81–4.81
XDR	5.08	1.64–15.56	3.09	1.23–7.74

OR = odds ratio; CI = confidence interval; AMR = antimicrobial resistance; MDR = multidrug–resistant; XDR = extremely drug-resistant.

Results of logistic regression analysis utilized to estimate the association between IAT and adverse outcomes are shown in Table 5. In all three models (crude, adjusted for trauma and level of AMR, and adjusted for trauma, level of AMR and other statistically and clinically important parameters), inadequate IAT was associated with the occurrence of all adverse outcomes, i.e., intra-hospital mortality, VAP-related sepsis, relapse, reinfection, and ICU LOS.

Table 5.

Association of Initial Antimicrobial Treatment and Adverse Outcomes

Outcome	Mortality	VAP-related sepsis	Relapse	Reinfection	ICU LOS
Outcome	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)^a
Inadequate IAT^b	3.03 (1.51-6.07)	2.68 (1.29–5.58)	3.80 (1.76–8.24)	5.14 (2.39–11.06)	4.71 (1.06–8.35)
Inadequate IAT^c	2.53 (1.23–5.23)	2.26 (1.06–54.86)	2.93 (1.31–6.55)	4.96 (2.23–11.04)	4.48 (0.6–8.30)
Inadequate IAT^d	3.08 (1.30–7.26)	2.39 (1.07–5.32)	3.25 (1.34–7.89)	6.06 (2.48–14.77)	4.65 (0.93–8.36)

Values in bold are significant.

VAP = ventilator-associated pneumonia; ICU LOS = intensive care unit length of stay; OR = odds ratio; CI = confidence interval; IAT = initial antimicrobial treatment.

Linear regression analysis of the association of empirical therapy and continuous outcomes.

Crude (unadjusted)

Adjusted for trauma and resistance of microorganism.

Adjusted for age, gender, Acute Physiology and Chronic Health Evaluation II, initial therapy, trauma, comorbidities, onset of VAP, polymicrobial infection and resistance of micro-organism.

Discussion

Ventilator-associated pneumonia is the most prevalent hospital-acquired infection in ICUs [23,24]. Knowing that antimicrobial agents are the most commonly prescribed drugs in the ICU [25], it is not surprising that inadequacy of the empiric antimicrobial prescription and delayed adequate treatment contribute not only to the higher VAP-associated mortality but also to the overall ICU and hospital mortality [4,20,26,27]. Despite this, studies addressing empirical antimicrobial treatment adequacy and factors contributing to the inadequate IAT prescribing in critically ill and traumatized patients who develop VAP are scarce in the literature.

In our study, more than one-third (35.5%) of critically ill patients with VAP received inadequate IAT. In the literature, the rate of inadequate IAT ranges between 10% and 73% [5,6,20,27 –31]. Such a tremendous difference can be partially explained by a different definition of the adequacy of IAT, i.e., whether it is defined as a favorable clinical response or by using in vitro susceptibility of recovered organisms.

Another factor contributing to the different frequency of inadequate IAT could be different underlying conditions of patients, i.e., referral of patients to the ICU from different medical wards (surgical or clinical) and type of patient (surgical, medical, or trauma). It was demonstrated that 60% of traumatized patients had multi-drug-resistant organisms (MDRO) compared with medical and surgical patients, who had 33.6% and 32.5% MDRO, respectively [32].

It has also been shown that MDROs cause a greater number of VAP and HAPs in surgical than in trauma ICU [33]. However, the most likely reason for the relatively high frequency of inadequate IAT in our and other studies irrespective of the definition used or patients' medical background, is antimicrobial selective pressure and the sensitivity profile of the suspected pathogens [5]. In fact, we found a higher overall rate of MDR and XDR strains as well as a higher rate of XDR strains of Acinetobacter spp. and Pseudomonas aeruginosa in a group of patients who received inadequate IAT compared with those who received adequate IAT. This is supported by results of logistic regression in which, in addition to trauma, XDR causative agent was an independent risk factor for inadequate IAT. This is also in line with results of previous studies that showed that patients with VAP caused by MDR strains had up to a seven-fold higher chance for inadequate IAT [6,34].

As mentioned previously, the multivariable logistic regression model demonstrated that trauma and XDR causative agent are independent risk factors for inadequate IAT. Although the fact that adverse effects of high bacterial resistance on inadequate IAT have been described previously [6,11], our finding that trauma is an independent risk factor for inadequate IAT is surprising because other authors previously found no difference in inadequate IAT between trauma and non-trauma patients [35,36]. The explanation for this discrepancy may be the difference in IAT protocols between traumatized and non-traumatized patients. In our center, traumatized patients receive different IAT therapy than non-traumatized patients. Often on admission to our ICU traumatized patients have elevated inflammatory parameters or confirmed other infection, and initially receive more potent antibiotic agents such as meropenem and vancomycin. In addition to disrupted barriers and numerous invasive devices, the altered immune response can put those who survive initial trauma at particular risk of developing HAI over the course of ICU treatment. New insight into the imbalance of the immune system, which occurs shortly after trauma described by Xiao et al. [37] and Gentile et al. [38] can support our findings. According to this theory, traumatized patients suffer persistent inflammation, immunosuppression, and catabolism syndrome (PICS) [38]. Consequently, they can be more vulnerable to MDR and XDR nosocomial infection than other critically ill patients in the ICU despite equal exposure to invasive devices and to the particular local distribution of microbial flora. It has been shown that VAP is an independent predictor of death, even in patients with less severe trauma [39]. Considering all above adequate IAT of VAP represent one of the key elements for better survival among traumatized patients in ICUs.

We have shown that patients with inadequate IAT had higher crude and adjusted odds ratio for mortality. However, causative agents and their resistance level greatly influence the mortality rate (i.e., patients with VAP caused by Acinetobacter spp. and Pseudomonas aeruginosa had higher mortality rates than patients with VAP caused by other bacteria) [40]. However, there is a study that demonstrated that there is no difference in mortality between pneumonia caused by MDRO and pan-sensitive organisms among trauma patients [41]. It is worth noting that some other IAT and infection characteristics, such as multiple versus single infection episode, time to IAT admission, and changing from inadequate to adequate antimicrobial therapy, which we could not assess in our study, could have influenced this association. It has been shown that traumatized patients who received multiple episodes of inadequate IAT had four-fold higher risk for death [42] and that the mortality rate was higher in patients receiving delayed appropriate therapy than patients receiving adequate treatment [4]. Furthermore, changing from inadequate to adequate antimicrobial therapy once microbiologic culture results has been received, can improve outcomes but not to the same extent as initially adequate IAT [43].

Some limitations of this study should be emphasized. An important limitation is that majority of VAP diagnoses were based on tracheal aspirate samples and many patients with systemic inflammatory response syndrome could have been misdiagnosed as having VAP based solely on this sampling technique. However, one recent meta-analysis showed that tracheal aspirate samples have similar diagnostic accuracy as BAL and protected specimen brush (high sensitivity and low specificity) highlighting the general uncertainty and the need for better tools to diagnose VAP [44]. Generalizability of results is another important limitation of the study and the present findings regarding the incidence of isolation of pathogens are not applicable to other centers. The geographic variations and environmental characteristics of patients may predispose certain populations to infection in the setting of trauma, which makes the prevention of such infections particularly challenging [45]. In this study, the sample size is relatively small, and therefore, the study has limited power to detect the difference between groups. This also prevented us from stratifying patients into the groups depending on the presence of trauma or time of VAP onset (early and late VAP) and to investigate the association of IAT and outcomes with respect to trauma or VAP onset. Residual confounding due to unmeasured confounders such as previous antimicrobial therapy, previous hospitalization, de-escalation therapy, and therapy delay could have potentially influenced the strength of association. One of the limitations is also survival bias caused by lack of information on time to the event, because we could not exclude patients who died soon after initial therapy. However, our study was conducted in a level 1 trauma center, which is the major referral academic emergency center in our country with a catchment area of whole central Serbia and some parts of the neighbour countries. To the best of our knowledge this is one of the rare studies in southeastern Europe to delineate risk factors for IAT and the influence of IAT on adverse outcomes in critically ill patients with VAP, and as such allows a comprehensive insight into the complex association of initial treatment adequacy, antimicrobial resistance and adverse outcomes of VAP treatment.

Conclusions

In this study, we showed that trauma and XDR strains of bacteria are independent predictors of inadequate IAT of VAP in critically ill patients. Traumatized patients and patients with isolated XDR strains of bacteria have three times higher odds to receive inadequate IAT of VAP, respectively. This research demonstrated that inadequate IAT is associated with a higher mortality rate, VAP-related sepsis, relapse, re-infection, and ICU LOS in patients with VAP.

It is essential to implement and adhere to the newest ATS/IDSA or ERS/ESICM/ALT guidelines for VAP management in clinical practice and adjusted them with respect to local epidemiologic and microbiologic data. To support this, it is crucial to conduct surveillance studies on epidemiologic characteristics of VAP in trauma patients and to evaluate adequacy of current antimictobial regimens in order to inform development, periodic re-evaluation or improvent of existing local guidelines for VAP management in this specific subpopulation of critically ill patients.

Financial Information

This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia (grant No. 200110).

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Torres

, Ferrer

, Badia

. Treatment guidelines and outcomes of hospital-acquired and ventilator-associated pneumonia. Clin Infect Dis, 2010; 51:48–53.

Heininger

, Krueger

, Doring

, Unertl

. Ventilator-associated pneumonia. Curr Opin Anaesthesiol, 2002; 15:153–159.

Kollef

. Inadequate antimicrobial treatment: An important determinant of outcome for hospitalized patients. Clin Infect Dis, 2000; 31:131–138.

Luna

, Aruj

, Niederman

, et al. Appropriateness and delay to initiate therapy in ventilator-associated pneumonia. Eur Respir J, 2006; 27:158–164.

Piskin

, Aydemir

, Oztoprak

, et al. Inadequate treatment of ventilator-associated and hospital-acquired pneumonia: Risk factors and impact on outcomes. BMC Infect Dis, 2012; 12:268.

Teixeira

PJZ

, Seligman

, Hertz

, et al. Inadequate treatment of ventilator-associated pneumonia: risk factors and impact on outcomes. J Hosp Infect, 2007; 65:361–367.

American Thoracic Society; Infectious Diseases Society of America. Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med, 2005; 171:388–416.

Torres

, Niederman

, Chastre

, et al. International ERS/ESICM/ESCMID/ALAT guidelines for the management of hospital-acquired pneumonia and ventilator-associated pneumonia: Guidelines for the management of hospital-acquired pneumonia (HAP)/ventilator-associated pneumonia (VAP) of the European Respiratory Society (ERS), European Society of Intensive Care Medicine (ESICM), European Society of Clinical Microbiology and Infectious Diseases (ESCMID) and Asociación Latinoamericana del Tórax (ALAT). Eur Respir J, 2017; 50.

Kalil

, Metersky

, Klompas

, et al. Management of adults with hospital-acquired and ventilator-associated pneumonia: 2016 Clinical Practice Guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clin Infect Dis, 2016; 63:61–111.

10.

Herkel

, Uvizl

, Doubravska

, et al. Epidemiology of hospital-acquired pneumonia: Results of a Central European multicenter, prospective, observational study compared with data from the European region. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub, 2016; 160:448–455.

11.

Niederman

. Use of broad-spectrum antimicrobials for the treatment of pneumonia in seriously ill patients: Maximizing clinical outcomes and minimizing selection of resistant organisms. Clin Infect Dis, 2006; 42:72–81.

12.

Jovanovic

, Milan

, Djuric

, et al. Twenty-eight-day mortality of blunt traumatic brain injury and co-injuries requiring mechanical ventilation. Med Princ Pract, 2016; 25:435–441.

13.

Jovanovic

, Milan

, Markovic-Denic

, et al. Risk factors for ventilator-associated pneumonia in patients with severe traumatic brain injury in a Serbian trauma centre. Int J Infect Dis, 2015; 38:46–51.

14.

Djuric

, Markovic-Denic

, Jovanovic

, Bumbasirevic

. High incidence of multiresistant bacterial isolates from bloodstream infections in trauma emergency department and intensive care unit in Serbia. Acta Microbiol Immunol Hung, 2019; 66:307–325.

15.

Kalanuria

, Ziai

, Mirski

. Ventilator-associated pneumonia in the ICU. Crit Care, 2014; 18:208.

16.

Ferguson

, Fan

, Camporota

, et al. The Berlin definition of ARDS: An expanded rationale, justification, and supplementary material. Intensive Care Med, 2012; 38:1573–1582.

17.

Civil

, Schwab

. The Abbreviated Injury Scale, 1985 revision: A condensed chart for clinical use. J Trauma, 1988; 28:87–90.

18.

MacKenzie

, Shapiro

, Eastham

. The Abbreviated Injury Scale and Injury Severity Score. Levels of inter- and intrarater reliability. Med Care, 1985; 23:823–835.

19.

Baker

, O'Neill

, Haddon

, Long

. The injury severity score: A method for describing patients with multiple injuries and evaluating emergency care. J Trauma, 1974; 14:187–196.

20.

Kollef

, Sherman

, Ward

, Fraser

. Inadequate antimicrobial treatment of infections: A risk factor for hospital mortality among critically ill patients. Chest, 1999; 115:462–474.

21.

Magiorakos

A-P

, Srinivasan

, Carey

, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: An international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect, 2012; 18:268–281.

22.

Biondi-Zoccai

, Romagnoli

, Agostoni

, et al. Are propensity scores really superior to standard multivariable analysis?. Contemp Clin Trials, 2011; 32:731–740.

23.

Borgatta

, Rello

. How to approach and treat VAP in ICU patients. BMC Infect Dis, 2014; 14:211.

24.

Kuti

, Patel

, Coleman

. Impact of inappropriate antibiotic therapy on mortality in patients with ventilator-associated pneumonia and blood stream infection: A meta-analysis. J Crit Care, 2008; 23:91–100.

25.

Krivoy

, El-Ahal

, Bar-Lavie

, Haddad

. Antibiotic prescription and cost patterns in a general intensive care unit. Pharm Pract (Granada), 2007; 5:67–73.

26.

Rello

, Rué

, Jubert

, Muses

, et al. Survival in patients with nosocomial pneumonia: Impact of the severity of illness and the etiologic agent. Crit Care Med, 1997; 25:1862–1867.

27.

Alvarez-Lerma

. Modification of empiric antibiotic treatment in patients with pneumonia acquired in the intensive care unit. ICU-Acquired Pneumonia Study Group. Intensive Care Med, 1996; 22:387–394.

28.

Khan

, Khan

. Breaking the spell: Combating multidrug resistant “superbugs.”. Front Microbiol, 2016; 7:174.

29.

Leroy

, Meybeck

, d'Escrivan

, et al. Impact of adequacy of initial antimicrobial therapy on the prognosis of patients with ventilator-associated pneumonia. Intensive Care Med, 2003; 29:2170–2173.

30.

Dupont

, Mentec

, Sollet

, Bleichner

. Impact of appropriateness of initial antibiotic therapy on the outcome of ventilator-associated pneumonia. Intensive Care Med, 2001; 27:355–362.

31.

Evans

, Sharpe

, Swanson

, et al. Keeping it simple: Impact of a Restrictive Antibiotic Policy for Ventilator-Associated pneumonia in trauma patients on incidence and sensitivities of causative pathogens. Surg Infect, 2018; 19:672–678.

32.

Martin-Loeches

, Torres

, Rinaudo

, et al. Resistance patterns and outcomes in intensive care unit (ICU)-acquired pneumonia. Validation of European Centre for Disease Prevention and Control (ECDC) and the Centers for Disease Control and Prevention (CDC) classification of multidrug resistant organ. J Infect, 2015; 70:213–222.

33.

Becher

, Hoth

, Neff

, et al. Multidrug-resistant pathogens and pneumonia: Comparing the trauma and surgical intensive care units. Surg Infect, 2011; 12:267–272.

34.

Deconinck

, Meybeck

, Patoz

, et al. Impact of combination therapy and early de-escalation on outcome of ventilator-associated pneumonia caused by Pseudomonas aeruginosa. Infect Dis (Lond), 2017; 49:396–404.

35.

Agbaht

, Lisboa

, Pobo

, et al. Management of ventilator-associated pneumonia in a multidisciplinary intensive care unit: Does trauma make a difference?. Intensive Care Med, 2007; 33:1387–1395.

36.

Kashuk

, Moore

, Price

, et al. Patterns of early and late ventilator-associated pneumonia due to methicillin-resistant Staphylococcus aureus in a trauma population. J Trauma, 2010; 69:519–522.

37.

Xiao

, Mindrinos

, Seok

, et al. A genomic storm in critically injured humans. J Exp Med, 2011; 208:2581–2590.

38.

Gentile

, Cuenca

, Efron

, et al. Persistent inflammation and immunosuppression: A common syndrome and new horizon for surgical intensive care. J Trauma Acute Care Surg, 2012; 72:1491–1501.

39.

Magnotti

, Croce

, Fabian

. Is ventilator-associated pneumonia in trauma patients an epiphenomenon or a cause of death?. Surg Infect, 2004; 5:237–242.

40.

Kollef

, Silver

, Murphy

, Trovillion

. The effect of late-onset ventilator-associated pneumonia in determining patient mortality. Chest, 1995; 108:1655–1662.

41.

Rai

, Stephen

, Lu

, Heffernan

. Impact of multi-drug-resistant pneumonia on outcomes of critically ill trauma patients. Surg Infect, 2020; 21:422–427.

42.

Mueller

, Hanes

, Croce

, et al. Effect from multiple episodes of inadequate empiric antibiotic therapy for ventilator-associated pneumonia on morbidity and mortality among critically ill trauma patients. J Trauma, 2005; 58:94–101.

43.

Davey

, Marwick

. Appropriate vs. inappropriate antimicrobial therapy. Clin Microbiol Infect, 2008; 14:15–21.

44.

Fernando

, Tran

, Cheng

, et al. Diagnosis of ventilator-associated pneumonia in critically ill adult patients: A systematic review and meta-analysis. Intensive Care Med, 2020; 46:1170–1179.

45.

Zarzaur

, Bell

, Croce

, Fabian

. Geographic variation in susceptibility to ventilator-associated pneumonia after traumatic injury. J Trauma Acute Care Surg 201;75:234–240.