| |
Ventilator-associated pneumonia in an Italian pediatric intensive care unit: a prospective study
Maria Francesca Patria, Giovanna Chidini, Ludovica Ughi, Cinzia Montani, Edi Prandi, Carlotta Galeone, Edoardo Calderini, Susanna Esposito
Milan, Italy
Author Affiliations: Department of Pathophysiology and Transplantation, Università degli Studi di Milano, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy (Patria MF, Esposito S); Pediatric Intensive Care Unit, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy (Chidini G, Ughi L, Montani C, Prandi E, Calderini E); Department of Epidemiology, Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy (Galeone C); Department of Occupational Health, Clinica del Lavoro Luigi Devoto, Università degli Studi di Milano, Milan, Italy (Galeone C)
Corresponding Author: Susanna Esposito, MD, Department of Pathophysiology and Transplantation, Università degli Studi di Milano, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, via Commenda 9, 20122 Milan, Italy (Tel: +39-02-55032498; Fax: +39-02-50320206; Email: susanna.esposito@unimi.it)
doi: 10.1007/s12519-013-0444-y
Background: This study was undertaken to determine the prevalence, risk factors and outcomes associated with ventilator-associated pneumonia (VAP) in a European pediatric intensive care unit (PICU).
Methods: A total of 451 children who had been mechanically ventilated in the PICU for ≥48 hours during a 3-year period were enrolled in this prospective study.
Results: In comparison with children without VAP, 30 children (6.6%) who developed VAP had a longer PICU stay (P=0.0001) and hospital stay (P=0.0001), and a higher mortality rate (P=0.04). Logistic regression analysis showed that the need for re-intubation (P=0.0001), the presence of tracheostomy (P=0.04), and enteral feeding (P=0.02) were independent risk factors for VAP.
Conclusions: A relevant proportion of intubated children develop VAP, which is closely related to invasive procedures. As VAP is associated with increased medical costs and death, multicenter studies are urgently needed to improve the therapeutic approach to VAP and VAP prevention.
Key words: nosocomial infections; pneumonia; ventilator-associated pneumonia
World J Pediatr 2013;9(4):365-368
Introduction
Ventilator-associated pneumonia (VAP) is an important problem in pediatric intensive care units (PICUs) and has a prevalence ranging from 5% to 32%.[1,2] Data collected in the United States show that VAP is the second most common cause of nosocomial infections in PICU patients, accounting for 23% of all infections.[3] Published data suggested that it is associated with a longer PICU stay and a possibly higher mortality rate.[1,4,5] Despite the frequency and severity of this complication, there are few studies on pediatric VAP and these are often difficult to compare because of the lack of a standardized diagnostic approach, and the heterogeneity of the underlying diseases, types of treatment, ages and sample sizes. The aim of this prospective study was to determine the prevalence, risk factors and outcomes associated with VAP in a European PICU.
Methods
The study involved prospectively all of the children aged less than 14 years who were mechanically ventilated in our PICU for ≥48 hours between January 2007 and December 2010. Our PICU is a six-bed unit in a university hospital in Milan, Italy that admits about 350 patients a year. The study was approved by the Institutional Review Board of Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, and written informed consent was obtained from a parent or legal guardian of all of the participants.
The medical records of each patient (demographics, underlying illness, types of procedures, medication, duration of PICU, hospital stays, and mortality) were collected from an electronic database in order to evaluate patients' characteristics. VAP was defined on the basis of clinical, microbiological and radiologic criteria in relation to patient's age, and diagnosed in the presence of at least one of the manifestations such as new or progressive infiltrate, consolidation, cavitation or pleural effusion on chest radiography, plus at least one episode of fever (>38°C) with no other recognized cause, leukopenia [<4000 white blood cells (WBC)/mm3] or leukocytosis (≥12 000 WBC/mm3), and at least two signs of a new onset of purulent sputum: a change in sputum characteristics, increased respiratory secretions or increased suctioning requirements, new-onset or worsening cough, dyspnea or tachypnea, rales or bronchial breath sounds and worsening gas exchange [i.e. O2 desaturation (PaO2/FiO2 levels of ≤240), increased oxygen requirement, or increased ventilation demand], according to the criteria of the Center for Disease Control and Prevention.[6] In all the patients, respiratory cultures were performed and obtained by means of tracheal aspiration; a threshold of ≥105 colony-forming units (CFUs/mL) was considered significant.
Continuous data were analyzed using a two-sided Student's t test, after checking data were normally distributed (based on the Shapiro-Wilk statistics) and a two-sided Wilcoxon's rank-sum test otherwise. Categorical data were analyzed using the contingency table analysis with the Chi-square test or Fisher's exact test, as appropriate. The adjusted odds ratio (OR) of VAP and corresponding 95% confidence interval (CI) were derived using unconditional multiple logistic regression models in terms of sex, age (≤2 years, >2 years), underlying illness (yes/no) and clinical procedures. All tests were two-sided, and a P value ≤0.05 was considered statistically significant. Data analyses were conducted using SAS version 9.1 (cary, NC, USA).
Results
Of 1446 children admitted in the PICU, 451 mechanically ventilated children (31.2%) were enrolled during the study period. Their mean age (±standard deviation) was 3.8 (±4.5) years (age range, 0.5-14 years; median age, 4.5 years). Cerebral delay (82/451, 18.2%), chronic lung disease (61/451, 13.5%) and gastrointestinal malformation (55/451, 12.2%) were the most common pre-existing diseases, but 144 children (31.9%) had no chronic underlying disease and were admitted for acute infectious disease (58 children for severe community-acquired pneumonia; 33 for complicated bronchiolitis, 30 for sepsis, 23 for meningitis). In children with chronic underlying disease, 96 (31.3%) were surgical patients who received prophylactic antibiotics. No traumatic patient was enrolled. Central venous catheterization (334/451, 74%) and enteral nutrition through a nasogastric tube (271/451, 60.1%) were the most frequently performed procedures. One hundred and fifteen patients (25.5%) had received non-invasive ventilation trial before intubation. Mean pediatric index of mortality (PIM2)±standard deviation was -2.82±1.37.
Thirty patients (6.6%) developed VAP as defined above. Univariate analysis showed the results of the intubated patients with and without VAP (Table 1). The significant risk factors associated with VAP included previous non-invasive ventilation (P=0.0003), re-intubation before VAP diagnosis (P=0.0001), enteral feeding (P=0.0001), transfusion (P=0.01) and the presence of a central line (P=0.01). No significant differences were related to the underlying illness. Patients with VAP had a longer PICU (P=0.0001) and hospital stay (P=0.0001) as well as a higher mortality rate (P=0.04), regardless of PIM2.
Table 2 shows the results of a logistic regression analysis adjusted for gender, age, underlying illness and clinical procedures. The multivariate ORs of VAP were 9.46 (95% CI=3.34-26.78) for re-intubation (P=0.0001), 4.44 (95% CI=1.01-19.96) for the presence of tracheostomy (P=0.04), and 13.2 (95% CI=1.53-114.17) for enteral feeding.
A total of 33 tracheal aspiration cultures from the 30 VAP patients were micro-organism positive. The most frequently isolated microorganisms were Staphylococcus aureus (6/33, 18%) and Pseudomonas aeruginosa (5/33, 15%), followed by Stenotrophomonas maltophilia (3/33, 9%), fungi (3/33, 9%), Escherichia coli (2/33, 6%), Enterobacter species (2/33, 6%), and Klebsiella pneumoniae (2/33, 6%). None of the VAP patients had negative tracheal cultures and no viral pathogen was found.
Discussion
We prospectively collected data to determine the prevalence, risk factors and outcomes associated with pediatric VAP. Although the frequency of VAP has decreased over the last 20 years as a result of the implementation of preventive strategies, it remains a considerable problem associated with PICU admission and has an incidence ranging from 0.6 to 2.3/1000 ventilator-days, depending on the type of PICU.[4,5] VAP occurred in 6.6% of our cohort of 451 mechanically ventilated children, a prevalence that is much lower than that reported by Srinivasan et al (32% of 58 pediatric patients admitted to their PICU)[2] and Casado et al (10.7% of 336 intubated children).[7] These differences can be attributed to differences in VAP definitions as well as study methods and populations, although a low prevalence is always found in PICUs in which efforts have been made to provide education on preventing nosocomial infections.[5] In our PICU, we have a written protocol for infection (including VAP) prevention, and Health Direction of the Hospital regularly monitors nosocomial infections which occur in the various units and provides educational courses on their prevention.
Previous non-invasive ventilation, re-intubation, enteral feeding, transfusion and the presence of a central line were significant risk factors for VAP, but there was no association with the underlying illness. However, logistic regression analysis showed that enteral feeding and re-intubation before the diagnosis of VAP were determined and that tracheostomy was the only significant independent risk factor. During tracheostomy, we used cuffed endotracheal tubes, and indications for tracheotomy included bypassing airway obstruction, providing access for prolonged mechanical ventilation, and facilitating tracheobronchial toilet. Our findings were partially consistent with those of other studies although the data were difficult to compare because of heterogeneous procedures. Healthcare-associated pneumonia was suggested and VAP was considered as an equivalent. Srinivasan et al[2] identified continuous enteral feeding, narcotic use, female gender and post-surgical admission as independent risk factors for VAP in children. In another prospective surveillance study of 361 subjects, continuous enteral feeding, re-intubation, bronchoscopy and prior use of antibiotics were identified as independent predictors.[8] Many of the studies showed the risk of aspiration and by-passing of airway defences of the patient, thus allowing bacteria to enter the lungs.[6] This indicates that reducing the risk of aspiration and shortening the duration of mechanical ventilation are crucial in reducing the risk of VAP. However, it remains to be clarified why non-invasive ventilation is associated with an increased risk of VAP.
Our data have shown that VAP is associated with prolonged PICU or hospital stay regardless of illness severity,[4,5] and with increased mortality. Univariate analysis by Bigham et al[5] showed a similar increase in mortality in children with VAP although other studies have concluded that VAP is not independently associated with death.[9] However, VAP is thought to be associated with high medical costs,[5,6] and its prevention should be considered a priority in PICU patients.
The most frequently isolated organisms in our study were Staphyococcus aureus and Pseudomonas aeruginosa, although gram-negative bacteria were more frequently isolated than gram-positive bacteria. These findings are in consistent with those of other studies, and are similar in adult and pediatric series.[2,4]
One limitation of our study is that the culture specimens were obtained by means of endotracheal aspiration, so the results might reflect pathogens colonizing the upper airways. Gauvin et al[10] found that blind bronchoalveolar lavage (BAL) with a bacterial index of >5 was the most reliable means of diagnozing VAP in a 27-month prospective study of 30 intubated children. However, the risks of invasive methods may rule out the routine use of BAL in critically-ill children. Our cultures were analyzed quantitatively using a high threshold count of bacterial growth in order to improve specificity in distinguishing infection and colonization of the lower airways. Using BAL as the reference standard, Sachdev et al[11] have shown that tracheal aspirates with a threshold value of 105 CFUs/mL are sensitive and specific. However, more studies are required to compare different methods of diagnozing VAP and different pediatric therapeutic regimens.
In brief, our findings show that a small but not negligible proportion of intubated children develop VAP, and that this is closely related to invasive procedures. VAP increases PICU and hospital stays and is also a risk factor for death. Broad-spectrum antibiotics should be considered in the initial empiric approach to this severe disease, but more homogeneous studies are needed to define the diagnostic and therapeutic approach to children with suspected VAP and evaluate methods for its prevention.
Funding: This study was supported by a Ricerca Corrente grant. Ethical approval: The study was approved by the Institutional Review Board of Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico. Written informed consent was obtained from the parents or legal guardians of the study participants.
Competing interest: The authors declare that they have no conflict of interest.
Contributors: Patria MF was the consultant in pneumology, analyzed the data and wrote the first draft of the manuscript. Chidini G, Ughi L, Prandi E and Montani C visited the patients, collected the data and performed the data entry. Galeone C performed statistical analyses. Calderini E supervised the study and criticallly reviewed the manuscript. Esposito S was the consultant in pediatric infectious diseases, supervised data collectionn and co-wrote the manuscript.
References
1 Elward AM, Warren DK, Fraser VJ. Ventilator-associated pneumonia in pediatric intensive care unit patients: risk factors and outcome. Pediatrics 2002;109:758-764.
2 Srinivasan R, Asselin J, Gildengorin G, Wiener-Kronish J, Flori HR. A prospective study of ventilator-associated pneumonia in children. Pediatrics 2009;123:1108-1115.
3 Richards MJ, Edwards JR, Culver DH, Gaynes RP. Nosocomial infections in pediatric intensive care units in the United States. National Nosocomial Infections Surveillance System. Pediatrics 1999;103:e39.
4 Principi N, Esposito S. Ventilator-associated pneumonia (VAP) in pediatric intensive care units. Pediatr Infect Dis J 2007;26:841-843.
5 Bigham MT, Amato R, Bondurrant P, Fridriksson J, Krawczeski CD, Raake J, et al. Ventilator-associated pneumonia in the pediatric intensive care unit: characterizing the problem and implementing a sustainable solution. J Pediatr 2009;154:582-587.
6 Department of Health and Human Services. Criteria for defining nosocomial pneumonia, August 23, 2006. http://www.cdc.gov/ncidod/hip/NNIS/members/pneumonia/Final/PneumoCriteriaV1.pdf (accessed December 15, 2011).
7 Casado RJA, de Mello MJ, de Aragão RC, de Albuquerque Mde F, Correia JB. Incidence and risk factors for health care-associated pneumonia in a pediatric intensive care unit. Crit Care Med 2011;39:1968-1973.
8 Almuneef M, Memish ZA, Balkhy HH, Alalem H, Abutaleb A. Ventilator-associated pneumonia in a pediatric intensive care unit in Saudi Arabia: a 30-month prospective surveillance. Infect Control Hosp Epidemiol 2004;25:753-758.
9 Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Respir Crit Care Med 2002;165:867-903.
10 Gauvin F, Dassa C, Chaïbou M, Proulx F, Farrell CA, Lacroix J. Ventilator-associated pneumonia in intubated children: comparison of different diagnostic methods. Pediatr Crit Care Med 2003;4:437-443.
11 Sachdev A, Chugh K, Sethi M, Gupta D, Wattal C, Menon G. Diagnosis of ventilator-associated pneumonia in children in resource-limited setting: a comparative study of bronchoscopic and nonbronchoscopic methods. Pediatr Crit Care Med 2010;11:258-266.
Received May 26, 2012 Accepted after revision September 28, 2012
|
