Quantitative culture sampling of tracheal aspirates for diagnosis of nosocomial pneumonia in the ITU
*Aleksander Zeliaś1, Alicja Budak2,3, Dorota Włodarczyk2, Paweł Wodziński1
1Department of Intensive Therapy and Anaesthesiology, District Specialist Hospital in Cracow
2Department of Pharmaceutical Microbiology, Collegium Medium, Jagiellonian University in Cracow
3Department of Laboratory Diagnostics, District Specialist Hospital in Cracow
- Table 1. Bacteria isolated from the endotracheal aspirate on the diagnosis of nosocomial pneumonia
- Table 2. Bacteria isolated from the endotracheal aspirate on day 6 of treatment of hospital-acquired pneumonia
- Table 3. The course of treatment and hospital mortality
- Fig. 1. ITU admissions – referring units NO – Neuro-orthopaedics. HED – Hospital Emergency Department. ID – Internal Diseases. Neur – Neurology. Sur – Surgery. Nephro – Nephrology. Ortho – Orthopaedics. O – Other hospitals
- Fig. 2. The number of insufficient organs and hospital mortality
- Fig.3. Age of patients versus hospital mortality
- Fig. 4. Antibiotic therapy of nosocomial pneumonia
Background. Nosocomial pneumonia is defined as a pneumonia developing within 48 hours of admission in a patient with previously normal lung tissue. Ventilator-associated pneumonia (VAP) due to migration of microorganisms from the nose, throat and digestive tract is typical in ITU patients. Microbiological diagnosis is difficult, and there is no single method approved for VAP diagnosis. We have assessed the usefulness of Quantitative Endotracheal Aspirate (QETA) for this purpose, basing on its reported high sensitivity (mean 76%±9%) and specificity (mean 75±28%).
Methods. Endotracheal aspirates were sampled from ITU patients fulfilling the following criteria: (1) new or progressive infiltrations in the lungs plus hyperthermia (>38.3 °C) or hypothermia (<36 oC); (2) leukocytosis (>12 G L-1) or leukopenia (<4 G L-1); or (3) purulent sputum.
Results. Nosocomial pneumonia was diagnosed in 40 out of 312 patients treated in the ITU during the study period. VAP was diagnosed in 33 patients. Multi-drug resistant bacteria: Pseudomonas aeruginosa (16), Acinetobacter baumannii (19), ESBL pathogens (8) and MRSA (1) were cultured from QETA in 32 patients. The sensitivity of QETA was 86%.
Conclusion. The results allowed for early and accurate diagnosis of nosocomial pneumonia and institution of the best possible treatment.
Patients of intensive therapy units are at the highest risk of nosocomial infections, which is associated with their bad general status, multiorgan failures, impaired immunity and various invasive diagnostic and therapeutic methods used. Moreover, the multi-drug resistant (MDR) bacterial flora present in ITUs, which results from the common use of wide-spectrum antibiotics and occurrence of various mechanisms of bacterial resistance, is equally important. The skin and mucous membranes (of the airway, urinary and digestive systems) are quickly colonized with typical opportunistic bacterial strains, which invade the organism through probes, catheters, endotracheal or tracheostomy tubes and are often spread among patients via hands or clothes of the staff .
One of the commonest types of ITU infections is hospital-acquired pneumonia (HAP)  developing 48 h after admission of the patient with no earlier diagnosis of pulmonary parenchyma infection. Another group of pneumonias classified together with HAP due to similar bacterial aetiology is health care associated-pneumonia (HCAP). Ventilator-associated pneumonia (VAP) is the commonest variant of HAP in ITUs, which develops within 48-72 h after the onset of mechanical lung ventilation or intubation (accounting for even 90% of HAP cases treated in ITUs) [3, 4]. The main cause of VAP is the migration of bacteria from the oropharygeal cavity around the endotracheal tube cuff; the infection of pulmonary parenchyma is often preceded by colonization of the patient`s airway. The mortality rates in HAP are high and mainly depend on the promptness of effective antibiotic therapy. The lack of confirmation of HAP and its aetiological factor (no proper bacteriological tests performed) is thus one of the major causes of therapeutic failures .
There is no current golden standard for the diagnosis of HAP . The classic radiological and clinical criteria (infiltrates on chest radiographs, fever, leucocytosis, purulent secretion in the airway) are not sufficiently sensitive and specific . The modern diagnostics of HAP should be supported with microbiological methods. Since the airway colonization precedes the development of pneumonia, the presence of bacteria in the trachea (qualitative examination) alone is not enough to diagnose HAP. Therefore quantitative methods were introduced which enable accurate determinations of the number of bacterial cells in one mL of the secretion collected (CFU – Colony Forming Units), hence the infection may be distinguished from colonization and needless use of antibiotics avoided . For each available quantitative microbiological method, a different threshold value for the diagnosis of pneumonia has been empirically determined . There is no consensus whether the material for examinations should be collected invasively, i.e. using the bronchoscope, or non-invasively with the use of blind aspiration [10, 11, 12]. One of the arguments in favour of invasive methods is that pneumonia is the process developing in the pulmonary alveoli, i.e. distally to terminal bronchioli, thus the invasively collected material should be more representative for identification of the actual pathogenic flora of pneumonia in a given patient . Since there is no convincing evidence that invasive methods increase the survival of patients with HAP, the American Thoracic Society (ATS) allows non-invasive methods as an alternative .
The aim of the present study was to assess the usefulness of quantitative endotracheal aspirate (QETA) cultures for the diagnosis of hospital-acquired pneumonia and the efficacy of its treatment.
The study encompassed patients hospitalized in ITUs between 1.06.2006 and 30.09.2007 with the suspicion of HAP, intubated, mechanically ventilated who met the following criteria:
- the presence of new or progression of earlier infiltrates on chest radiographs together with two out of three of the following symptoms :1/ fever >38.3º C or hypothermia <36º C, 2/ leucocytosis >12 G L-1 or leucopoenia <4 G L-1 or 3/ purulent secretion from the bronchial tree,
- at least one QETA sampling performed; the material was collected with the sterile catheter to a special container using dry aspiration; if the secretion was extremely concentrated, about 5 mL of normal saline was administered to the endotracheal tube.
The first sampling was carried out before the modification or initiation of antibiotic therapy, and at 3-day intervals until the therapy completion or extubation. Based on the findings and clinical status of patients (demographic data and past medical history), the empiric and targeted antibiotic therapies were determined.
Since the microbiological tests for pneumonia are performed routinely, no special consent was obtained except for the consent for treatment. Conscious patients were informed about the tests.
The statistical analysis was based on the Student’s t test for independent groups, X2, and X2 with a Fisher’s modification for small-size samples as well as the Cramer`s V correlation coefficient. The level of significance was set at p=0.05.
Forty (12.8%) out of 312 patients hospitalized in ITUs were enrolled in the study, including 28 males and 12 females with the diagnosis of pneumonia according to the accepted criteria. Their age range was 60.75±20.11 years. The majority of patients were referred to ITUs from the neuro-orpthopaedic and emergency departments (Fig. 1). Over 30% of patients were diagnosed with the failure of 4 organs.
A large group consisted of patients with spinal injuries – 10 (25%) and COPD – 7 (17.5%). Fifteen (37.5%) patients met the criteria of severe sepsis or septic shock.
Amongst all pneumonia cases, early VAP (=5 days after initiation of lung ventilation) was diagnosed in 10 patients whereas late VAP (> 5 days) in 23. In the remaining group of 7 patients, two were diagnosed with HCAP and 5 with HAP complicated with respiratory insufficiency and necessary intubation. The mean duration of ITU stay before the diagnosis of HAP was 14.18 ±16.5 days.
In 30 patients, aspirates were sampled according to the ATS recommendations, i.e. on diagnosis of HAP, before the initiation or modification of antibiotic therapy. Seven patients were sampled between day 1 and 3 of therapy; in three cases, such a sampling was impossible due to lack of information about the antibiotics used earlier.
In the group with the tracheal aspirates sampled before the institution or modification of antibiotic therapy, the sensitivity of QETA sampling for the diagnosis of HAP was 86% at the determined diagnostic threshold of =106 CFU mL-1. The species of bacteria isolated from endotracheal aspirates on diagnosis are presented in Table 1. During the treatment of HAP, the same amounts of bacteria were found in the majority of patients, mostly multi-resistant ones.
The types of bacteria isolated from endotracheal cultures on day 6 are illustrated in Table 2. In total, the MDR pathogens in the number of =106 CFU mL-1 were found in the endotracheal material in 32 patients, including Pseudomonas aeruginosa in 16, Acinetobacter baumannii in 19, ESBL strains in 8 patients, and MRSA only in 1 patient. Positive cultures from the pleural exudate were obtained in 4 (10%) patients with HAP; in 3 of them, the pathogens isolated were the same as the ones from the endotracheal aspirates. Positive blood cultures were found in 11 (27.5%) patients with HAP, yet only in three cases (7.5% of all patients with HAP), the bacteria cultured from blood were the same as those from tracheal cultures.
Amongst 7 deaths during ITU therapy, two were due to progressive pneumonia, 2 – intra-abdominal infections, 2 – irreversible CNS damage and 1 – lung cancer. Only 23 out of 40 patients survived until discharge. In the group of 28 patients assessed in ITUs as cured or partially improved cases, 11 died.
The mortality rates were higher in the male group – 46% compared to the female group – 33%. Differences in hospital mortality rates were also observed with respect to HCAP -100% (2/2), HAP with necessary intubation – 20% (1/5), VAP=5 days – 30% (3/10), VAP>5 days – 47.8% (11/23). However, the differences were not significant.
Moreover, no significant differences were found in hospital mortality between the groups of patients with MDR pathogens =106 CFU mL-1 (46.8%) and <106 CFU mL-1 (25%).
A significant correlation was observed between hospital mortality rates and the number of insufficient organs (p=0.0025) (Fig. 2) and age of patients (p=0.001) (Fig. 3). Furthermore, the duration of mechanical lung ventilation was a relevant mortality-related factor (Table 3).
The mean duration of antibiotic therapy in the group of non-survivors did not differ from that in survivors. Empiric antibiotic therapy of HAP was adequately chosen (bacteria isolated from the endotracheal aspirates susceptible to the antibiotic used) in 27 cases. During antibiotic therapy, in 26 patients, the number of bacteria exceeding 106 CFU mL-1 was observed in the endotracheal aspirates = 3 days after the institution of therapy. In this group, 6 patients died in ITUs. Amongst 11 patients with repeated cultures demonstrating <106 CFU mL-1 of bacteria, only one died in ITU. Three patients were not sampled for further quantitative cultures. The kinds of antibiotics used for empiric and targeted therapy are listed in Fig. 4.
In the present study, the non-invasive quantitative endotracheal aspirate method for microbiological diagnostics was used. According to the literature data, the sensitivity of this method for diagnosis of nosocomial pneumonia is 38%-82% and its specificity – 72-85% at the cut-off threshold =106 CFU mL-1 . In our study, quantitative endotracheal aspirate cultures increased the reliability of clinical diagnosis of pneumonia and were characterized by high sensitivity – 86%.
It can be assumed that if the quantitative endotracheal aspirate cultures alone and less restrictive qualification criteria for this examination (which is closer everyday practice) were used, the number of diagnosed pneumonia cases would be substantially higher as the chest X-ray lesions observed in ITU patients are often of a non-infective nature (atelectasis, embolism, overhydration, alveolar bleeding). In such cases, negative quantitative cultures allow to exclude pneumonia with high probability and to search for other sources of infection .
Moreover, quantitative cultures shortened the duration of antibiotic therapy – in our study the duration of therapy was significantly shorter in patients with low CFU value at successive endotracheal cultures (on average 10 days) compared to patients with high CFU value ( on average more than 19 days). The decreased number of bacteria at successive cultures demonstrates that the treatment provided was effective.
Amongst our patients with the number of endotracheal bacteria =106 CFU mL-1 in repeated cultures, the risk of ITU death was higher than in those with lower numbers of bacteria, although the difference was not significant.
In the majority of cases, the MDR pathogens were aetiological factors of pneumonias. Although on diagnosis the pulmonary bacterial flora was extremely varied, during the therapy, other bacteria were outnumbered by MDR pathogens. The quantitative aspirate culture findings after 6 days of therapy indicated the elimination of other strains caused by the antibiotic therapy administered.
In the majority of cases, empiric therapy involved penicillin with inhibitors of ß-lactamases, cephalosporins and macrolides whereas targeted therapy was mainly based on carbapenems combined with aminoglycosides, which was associated with the predominance of MDR bacterial flora as an aetiological factor of HAP.
As expected, a significant correlation was demonstrated between the hospital mortality and the number of insufficient organs on diagnosis of HAP as well as between the hospital mortality and the age of patients. A high proportion of deaths after discharge to other wards is worth stressing (more than half of all death cases). Although, there were no sufficient data to assess the degree that HAP was responsible for that, it might be supposed that bad general health status of patients, which made their survival outside ITUs impossible, was of much importance.
Quantitative culture sampling of endotracheal aspirates optimizes the diagnosis and management of hospital-acquired pneumonia.
1. Dzierzanowska D.: Zakazenie szpitalne. a-medica press, Bielsko-Biala, 2007.
2. Vincent JL, Biharki D, Suter PM, Bruining HA, White J, Nicolas-Chanoin MH, Wolff M, Spencer RC, Hemmer M: The prevalence of nosocomial infection in intensive care units in Europe – the results of the EPIC study. JAMA 1995; 274: 639-644.
3. Vincent JL: Ventilator-associated pneumonia. J Hosp Infect 2004; 57: 272-280.
4. Torres A, Anzar R, Gatell JM, Jimenez P, Gonzales J, Ferrer A, Celis R, Rodriquez-Roisin R: Incidence, risk and prognosis factors of nosocomial pneumonia in mechanically ventilated patients. Am Rev Resp Dis 1990; 142: 523-528.
5. Bonten MJ, Chastre J, Craig WA, Fagon J-Y, Hall J, Jacoby A, Kollef MH, Luna CM, Mandell LA, Torres A, Wunderink RG: Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Resp Crit Care Med 2005; 171: 388-416.
6. Waterer GW: The diagnostic dilemma in suspected ventilator-associated pneumonia: one size will never fit all. Chest 2003; 123: 335-337.
7. Grossman RF, Fein A: Evidence-based assessment of diagnostic tests for ventilator-associated pneumonia: executive summary. Chest 2000; 117: 791-801.
8. Torres A, Ewig S: Diagnosing ventilator-associated pneumonia. N Engl J Med 2004; 350: 433-435.
9. Ioanas M, Ferrer R, Angrill J, Ferrer M, Torres A: Microbial investigation in ventilator-associated pneumonia. Eur Resp J 2001; 17: 791-801.
10. Fagon J-Y, Chastre J, Wolff M, Gervais C, Parer-Aubas S, Stéphan F, Similowski T, Mercat A, Diehl JL, Sollet JP, Tenaillon A: Invasive and noninvasive strategies for management of suspected ventilator-associated pneumonia. Atrial. Ann Int Med 2000; 132: 621-630.
11. vSanchez-Nieto JM, Torres F, Garcia-Cordoba F, El-Ebiary M, Carrillo A, Ruiz J, Nuñez ML, Niederman M: Impact of invasive and noninvasive quantitative culture sampling on outcome of ventilator pneumonia. Am J Resp Crit Care Med 1998; 157: 371-376.
12. Riuz M, Torres A, Ewig S: Noninvasive versus invasive microbial investigation in ventilator-associated pneumonia. Am J Resp Crit Care Med 2000; 162: 119-125.
13. Chastre j, Fagon J-Y: Ventilator-associated pneumonia. State of the art. Am J Resp Crit Care Med 2002; 165: 867-903.
14. Koenig SM, Truwit JD: Ventilator-associated pneumonia: diagnosis, treatment and prevention. Clin Microbiol Rev 2006; 19: 637-657.
15. Rello J, Ollenford DA, Oster G, Montserrat V, Bellm L, Redman R, Kollef MH: Epidemiology and outcomes of ventilator-associated pneumonia in large US database. Chest 2002; 122: 2121.
Wojewódzki Szpital Specjalistyczny
im. L.Rydygiera w Krakowie
os. Zlotej Jesieni 1, 31-826 Kraków
tel.: 0-12 646 85 89, fax: 0-12 646 89 15