Journal of Investigative Surgery, 27, 256–261, 2014 C 2014 Informa Healthcare USA, Inc. Copyright ISSN: 0894-1939 print / 1521-0553 online DOI: 10.3109/08941939.2014.892652
ORIGINAL ARTICLE
Independent Risk Factors for Ventilator-Associated Pneumonia After Cardiac Surgery Wei Sheng,1 Quan-sheng Xing,2 Wen-ming Hou,1 Long Sun,1 Zhao-zhuo Niu,1 Ming-shan Lin,1 Yi-fan Chi1 1
Department of Cardiovascular Surgery, Qingdao Municipal Hospital, Medical College of Qingdao University, Qingdao, Shandong, China, 2 Heart Center, Qingdao Children’s Hospital, Qingdao, Shandong, China
ABSTRACT Objective: To investigate the related factors and pathogens of ventilator-associated pneumonia (VAP) after heart surgery so as to provide evidences for clinical prevention and therapy. Methods: In total 1,688 cases were collected from January 2004 to January 2011. Overall 105 patients developed VAP. Retrospectively analyzed these patients after heart surgery to determine the clinical data, pathogens and treatment measures. Results: The frequency of ventilator-associated pneumonia was 6.2% (105/1 688), and mortality was 25.7% (27/105), 198 pathogen strains were isolated by bacterial culture, in which Gram negative bacteria accounted for 69.2% (137/198), Gram positive bacteria 27.8% (55/198), and fungi 3.0% (6/198). The independent risk factors for VAP after cardiac surgery were: age >70 (p < .01), emergent surgery (p < .01), perioperative blood transfusions (p < 0.01), reintubation (p < .01) and days of mechanical ventilation (MV) (p < .01). Median length of stay in the ICU for patients who developed VAP or not was, respectively, (24.7 ± 4.5) days versus (3.2 ± 1.5) days (p < .05), and mortality was, respectively, 25.7% versus 2.9% in both populations (p < .05). Conclusion: Age >70, emergent surgery, perioperative blood transfusions, reintubation and days of MV are the risk factors for VAP in patients following cardiac surgery. P. aeruginosa, P. klebsiella, S. aureus, and Acinetobacter baumannii were the main pathogens of VAP. According to the cause of VAP, active prevention and treatment measures should be developed and applied to shorten the time of MV and improve chances of survival. Keywords: ventilator-associated pneumonia (VAP); cardiac surgery; risk factor; pathogen
INTRODUCTION
surgery (MHS) are a particularly high-risk population for nosocomial infections during the postoperative period with a high incidence and related mortality [4]. However, information regarding the risk factors and outcome of VAP in this setting is scarcely available. Therefore, it is essential to investigate the risk factors of VAP after heart surgery so as to acquire prevention and reasonable effective treatment. We carried out a retrospective study conducted to identify potential risk factors amenable to intervention for VAP in patients undergoing cardiac surgery in our institution during a 7-year period from January 2004 to January 2011. Our study aims were to determine the incidence, etiology, risk factors, and outcome of VAP in a large sample of patients who have undergone MHS in our heart center.
Ventilator-associated pneumonia (VAP) is a severe hospital-acquired pneumonia (HAP) and is a hazardous complication in intensive care unit (ICU) among patients receiving mechanical ventilation (MV) [1]. VAP is associated with a significant increase in mortality, morbidity, lengths of MV and ICU stay, prolonged hospitalization, and increased healthcare costs. Between 10% and 20% of patients who receive more than 48 hr of MV develop VAP [2]. The mortality rates for VAP range from 20% to 75% according to the different hospital and different study population [3]. Despite the progress in surgery and anesthesia, the risk of developing nosocomial infections remains a real threat. Further more, patients undergoing major heart
Received 11 December 2013; accepted 5 February 2014. Address correspondence to Dr. Yi-fan Chi, Department of Cardiovascular Surgery, Qingdao Municipal Hospital, Medical College of Qingdao University, Qingdao, Shandong, China. E-mail: [email protected]
256
Independent Risk Factors for VAP
MATERIAL AND METHODS All patients (1,688 patients) who were admitted to the heart surgery ICU after MHS at the Department of Cardiovascular Surgery in Qingdao Municipal Hospital between January 2004 and January 2011 were included in the study. This study was conducted in accordance with the principles of the Helsinki Declaration and was approved by the ethics committee of Qingdao Municipal Hospital. Informed consent for participation in this investigation was obtained from all subjects. All patients received mechanical ventilation with endotracheal intubation (or tracheostomy) and survived for at least 48 hr after the operations. Clinical data collection and statistics were accomplished by cardiac surgeons in our heart center and infectious diseases physicians including preoperative, intraoperative, and postoperative clinical data. Its incidence, mortality, etiology, and risk factors were analyzed. Patients were followed up until death or discharge. Overall, 105 patients developed VAP of 1,688 patients in this study. VAP incidence was 6.2%, mortality rate was 25.7% (27/105). There were 78 male patients and 27 female patients of these 105 VAP patients, age from 3 years to 87 years, mean age 54 ± 13.5 years, cardiopulmonary bypass (CPB) time was 60 min to 540 min, mean time (125 ± 34.5) min, MV time was 48 hr to 58 days, mean time (130 ± 28.6) hr, ICU stay time was 7 days to 60 days, mean time (10 ± 2) d. Preoperative cardiac function were evaluated by New York Heart Association functional classification (NYHA), grade I–II 25 cases, grade III 65 cases and grade IV 15 cases. The surgical risk was evaluated by the European system for cardiac operative risk evaluation (EuroSCORE) [5]. Including eight cases of congenital heart disease correction surgery, valve replacement surgery in 45 cases, 18 cases of aortic surgery, coronary artery bypass grafting in 25 cases, coronary artery bypass grafting and valve replacement surgery nine cases. Ventilator-associated pneumonia (VAP) diagnostic criteria were defined in accordance with the new guidelines of the American Thoracic Society and the Infectious Diseases Society of America [6]. VAP was diagnosed in patients who received MV for 48 hr when the presence of a new and/or progressive radiographic pulmonary infiltrate was detected plus two or more of the following clinical features: fever (temperature greater than 38.0◦ C), leukocytosis or leukopenia, purulent tracheobronchial secretions. Sampling of the lower respiratory tract in cases suspected of VAP was performed by endotracheal aspiration (ETA) of respiratory secretions. Sampling quantitative bacterial culture is the preferred method to determine the VAP pathogens. The lower respiratory tract deep sputum samples were obtained by disposable aseptic suction tube connected to the specimen collection bottles from the artificial airway C
2014 Informa Healthcare USA, Inc.
257
(endotracheal intubation or tracheotomy tube) for suspicious patients with VAP. The entire operation complied with sterile principle, and then, quantitative bacterial culture was accomplished. Negative results of ordinary bacterial culture contribute to exclude VAP, but prone to false positives because the specimens were often contaminated by bacterial colonization of the oropharynx and upper airway. Endotracheal aspirate for quantitative bacterial culture helps to distinguish infections (high colony counts) and colonization (low colony counts). We used 105 cfu/ml as the cutoff point of infection and contamination. Twenty samples were obtained by bronchoalveolar lavage (BAL) method in some cases. We performed direct sampling in the open condition to the lesion site according to the chest X-ray. In these cases, the cutoffs for positive samples were ≥104 cfu/ml. So this can improve the accuracy of diagnosis. Obtained specimens were cultured by the central laboratory of our hospital according to clinical laboratory standard recommendations. All data were analyzed by SPSS version 16.0 software package. Normal distribution of measurement data were measured by mean ± standard deviation (x¯ ± s), nonnormal distribution of measurement data with a median (25 percentile, 75 percentile), short for M (P25, P75). Comparison of the two data sets: Normal distribution of measurement data were compared using the t-test for equal variance data, and t test for unequal variance data. Nonnormal distribution of measurement data were compared using the Mann–Whitney rank sum test, and Count data comparison with chi-square test. So we can screen the possible risk factors. All possible risk factors as independent variables were analyzed by multivariate logistic regression to calculate odds ratios and 95% confidence intervals, and to analyze the independent risk factors for VAP occurrence. All statistical tests were two-tailed. The level of significance was set at p < .05 for all the tests.
RESULTS Clinical Data Analysis During the study period, a total of 1,735 consecutive patients underwent MHS. Overall, 47 patients were excluded because of intraoperative or early postoperative (48 hr) death. Overall 1,688 patients remained in the study. General data regarding the population intervened including the demographic and descriptive data of the patients are listed in Table 1.
Incidence and Etiology of VAP The cumulative incidence of VAP in the study cohort was 6.2% (105/1,688 patients), and VAP frequency was
258
Wei Sheng et al.
TABLE 1 Preoperative and surgical characteristics of patients who underwent major heart surgery Characteristics Preoperative Number of patients Mean age in years (SD) Sex, male/female Underlying conditions (%) Myocardial infarction (70 years, intraoperative IABP auxiliary, NYHA grade IV, the EuroSCORE, emergent surgery, inotropic support, cardiopulmonary bypass time (CPBT), blood transfusions, reintubation, mechanical ventilation days, and reoperation were associated with VAP. But the Logistic multivariate analysis showed that age >70 year old, emergent surgery, blood transfusions, reintubation, the days of mechanical ventilation were independent risk factors of VAP. As shown in Table 4.
70.6 ± 28.7
SD = Standard deviation, EF = Ejection fraction, EuroSCORE = European system for cardiac operative risk evaluation
126. Of the 126 total episodes, 10 patients had more than one VAP episode. The incidence density of VAP in our population was 23.1 episodes per 1,000 days of mechanical ventilation. The number of patients whose mechanical ventilation more than 48 hr was 198, and VAP incidence was 53.0% in this subgroup of patients. In these patients, VAP occurred about 4.5 day (3 days8 days) after the MV. The average MV time in these VAP patients was 10 days (2 days–58 days). One hundred and ninety-eight strains of pathogenic bacteria from the sputum were isolated and cultivated from these 105 VAP patients, among them the gram negative (G− ) bacillus 137 plants, accounting for 69.2% of the total plants, mainly is the P. aeruginosa, P. klebsiella, Acine-
Outcome The ICU stay time was significantly longer in patients with VAP (24.7 ± 4.5 days) than in patients without VAP (3.2 ± 1.5 days) (p < .05). Overall in-hospital mortality in our study population was 4.3% (73/1,688), with a mortality rate in patients with VAP of 25.7% (27/105) versus 2.9% (46/1,583) in patients without VAP (p < .05).
DISCUSSION Ventilator-associated pneumonia is a special type of hospital-acquired pneumonia. In recent years, VAP also gradually increased with the rapid development of the technology of mechanical ventilation. The study shows that if mechanical ventilation increases 1 day, the incidence of VAP will increase by 1%–3% [7]. VAP has Journal of Investigative Surgery
Independent Risk Factors for VAP
259
TABLE 3 Univariate analysis of preoperative, operative, and postoperative risk factors of VAP in patients underwent MHS
Preoperative risk factors Age >70 years Intra-aortic balloon Severe pulmonary hypertension Diabetes mellitus COPD Creatinine levels >1.5 mg/dl NYHA IV EuroScore(SD) Operative risk factors Emergent surgery Intra-aortic balloon Perioperative myocardial infarction Inotropic support CPBT(SD) Postoperative risk factors Blood transfusions (ml) (IQR) Reintubation Days of MV median (IQR) Reoperation
VAP (n = 105)
No VAP (n = 1583)
χ 2 /t
p
71 (67.6) 5 (4.8) 12 (11.4) 31 (29.5) 4 (3.8) 9 (8.6) 15 (14.3) 10.5 ± 4.1
677 (42.7) 41 (2.6) 162 (10.2) 412 (26.0) 52 (3.3) 102 (6.4) 136 (8.6) 6.7 ± 3.1
24.646 1.752 0.152 0.622 0.084 0.726 3.920 9.332
ORIGINAL ARTICLE
Independent Risk Factors for Ventilator-Associated Pneumonia After Cardiac Surgery Wei Sheng,1 Quan-sheng Xing,2 Wen-ming Hou,1 Long Sun,1 Zhao-zhuo Niu,1 Ming-shan Lin,1 Yi-fan Chi1 1
Department of Cardiovascular Surgery, Qingdao Municipal Hospital, Medical College of Qingdao University, Qingdao, Shandong, China, 2 Heart Center, Qingdao Children’s Hospital, Qingdao, Shandong, China
ABSTRACT Objective: To investigate the related factors and pathogens of ventilator-associated pneumonia (VAP) after heart surgery so as to provide evidences for clinical prevention and therapy. Methods: In total 1,688 cases were collected from January 2004 to January 2011. Overall 105 patients developed VAP. Retrospectively analyzed these patients after heart surgery to determine the clinical data, pathogens and treatment measures. Results: The frequency of ventilator-associated pneumonia was 6.2% (105/1 688), and mortality was 25.7% (27/105), 198 pathogen strains were isolated by bacterial culture, in which Gram negative bacteria accounted for 69.2% (137/198), Gram positive bacteria 27.8% (55/198), and fungi 3.0% (6/198). The independent risk factors for VAP after cardiac surgery were: age >70 (p < .01), emergent surgery (p < .01), perioperative blood transfusions (p < 0.01), reintubation (p < .01) and days of mechanical ventilation (MV) (p < .01). Median length of stay in the ICU for patients who developed VAP or not was, respectively, (24.7 ± 4.5) days versus (3.2 ± 1.5) days (p < .05), and mortality was, respectively, 25.7% versus 2.9% in both populations (p < .05). Conclusion: Age >70, emergent surgery, perioperative blood transfusions, reintubation and days of MV are the risk factors for VAP in patients following cardiac surgery. P. aeruginosa, P. klebsiella, S. aureus, and Acinetobacter baumannii were the main pathogens of VAP. According to the cause of VAP, active prevention and treatment measures should be developed and applied to shorten the time of MV and improve chances of survival. Keywords: ventilator-associated pneumonia (VAP); cardiac surgery; risk factor; pathogen
INTRODUCTION
surgery (MHS) are a particularly high-risk population for nosocomial infections during the postoperative period with a high incidence and related mortality [4]. However, information regarding the risk factors and outcome of VAP in this setting is scarcely available. Therefore, it is essential to investigate the risk factors of VAP after heart surgery so as to acquire prevention and reasonable effective treatment. We carried out a retrospective study conducted to identify potential risk factors amenable to intervention for VAP in patients undergoing cardiac surgery in our institution during a 7-year period from January 2004 to January 2011. Our study aims were to determine the incidence, etiology, risk factors, and outcome of VAP in a large sample of patients who have undergone MHS in our heart center.
Ventilator-associated pneumonia (VAP) is a severe hospital-acquired pneumonia (HAP) and is a hazardous complication in intensive care unit (ICU) among patients receiving mechanical ventilation (MV) [1]. VAP is associated with a significant increase in mortality, morbidity, lengths of MV and ICU stay, prolonged hospitalization, and increased healthcare costs. Between 10% and 20% of patients who receive more than 48 hr of MV develop VAP [2]. The mortality rates for VAP range from 20% to 75% according to the different hospital and different study population [3]. Despite the progress in surgery and anesthesia, the risk of developing nosocomial infections remains a real threat. Further more, patients undergoing major heart
Received 11 December 2013; accepted 5 February 2014. Address correspondence to Dr. Yi-fan Chi, Department of Cardiovascular Surgery, Qingdao Municipal Hospital, Medical College of Qingdao University, Qingdao, Shandong, China. E-mail: [email protected]
256
Independent Risk Factors for VAP
MATERIAL AND METHODS All patients (1,688 patients) who were admitted to the heart surgery ICU after MHS at the Department of Cardiovascular Surgery in Qingdao Municipal Hospital between January 2004 and January 2011 were included in the study. This study was conducted in accordance with the principles of the Helsinki Declaration and was approved by the ethics committee of Qingdao Municipal Hospital. Informed consent for participation in this investigation was obtained from all subjects. All patients received mechanical ventilation with endotracheal intubation (or tracheostomy) and survived for at least 48 hr after the operations. Clinical data collection and statistics were accomplished by cardiac surgeons in our heart center and infectious diseases physicians including preoperative, intraoperative, and postoperative clinical data. Its incidence, mortality, etiology, and risk factors were analyzed. Patients were followed up until death or discharge. Overall, 105 patients developed VAP of 1,688 patients in this study. VAP incidence was 6.2%, mortality rate was 25.7% (27/105). There were 78 male patients and 27 female patients of these 105 VAP patients, age from 3 years to 87 years, mean age 54 ± 13.5 years, cardiopulmonary bypass (CPB) time was 60 min to 540 min, mean time (125 ± 34.5) min, MV time was 48 hr to 58 days, mean time (130 ± 28.6) hr, ICU stay time was 7 days to 60 days, mean time (10 ± 2) d. Preoperative cardiac function were evaluated by New York Heart Association functional classification (NYHA), grade I–II 25 cases, grade III 65 cases and grade IV 15 cases. The surgical risk was evaluated by the European system for cardiac operative risk evaluation (EuroSCORE) [5]. Including eight cases of congenital heart disease correction surgery, valve replacement surgery in 45 cases, 18 cases of aortic surgery, coronary artery bypass grafting in 25 cases, coronary artery bypass grafting and valve replacement surgery nine cases. Ventilator-associated pneumonia (VAP) diagnostic criteria were defined in accordance with the new guidelines of the American Thoracic Society and the Infectious Diseases Society of America [6]. VAP was diagnosed in patients who received MV for 48 hr when the presence of a new and/or progressive radiographic pulmonary infiltrate was detected plus two or more of the following clinical features: fever (temperature greater than 38.0◦ C), leukocytosis or leukopenia, purulent tracheobronchial secretions. Sampling of the lower respiratory tract in cases suspected of VAP was performed by endotracheal aspiration (ETA) of respiratory secretions. Sampling quantitative bacterial culture is the preferred method to determine the VAP pathogens. The lower respiratory tract deep sputum samples were obtained by disposable aseptic suction tube connected to the specimen collection bottles from the artificial airway C
2014 Informa Healthcare USA, Inc.
257
(endotracheal intubation or tracheotomy tube) for suspicious patients with VAP. The entire operation complied with sterile principle, and then, quantitative bacterial culture was accomplished. Negative results of ordinary bacterial culture contribute to exclude VAP, but prone to false positives because the specimens were often contaminated by bacterial colonization of the oropharynx and upper airway. Endotracheal aspirate for quantitative bacterial culture helps to distinguish infections (high colony counts) and colonization (low colony counts). We used 105 cfu/ml as the cutoff point of infection and contamination. Twenty samples were obtained by bronchoalveolar lavage (BAL) method in some cases. We performed direct sampling in the open condition to the lesion site according to the chest X-ray. In these cases, the cutoffs for positive samples were ≥104 cfu/ml. So this can improve the accuracy of diagnosis. Obtained specimens were cultured by the central laboratory of our hospital according to clinical laboratory standard recommendations. All data were analyzed by SPSS version 16.0 software package. Normal distribution of measurement data were measured by mean ± standard deviation (x¯ ± s), nonnormal distribution of measurement data with a median (25 percentile, 75 percentile), short for M (P25, P75). Comparison of the two data sets: Normal distribution of measurement data were compared using the t-test for equal variance data, and t test for unequal variance data. Nonnormal distribution of measurement data were compared using the Mann–Whitney rank sum test, and Count data comparison with chi-square test. So we can screen the possible risk factors. All possible risk factors as independent variables were analyzed by multivariate logistic regression to calculate odds ratios and 95% confidence intervals, and to analyze the independent risk factors for VAP occurrence. All statistical tests were two-tailed. The level of significance was set at p < .05 for all the tests.
RESULTS Clinical Data Analysis During the study period, a total of 1,735 consecutive patients underwent MHS. Overall, 47 patients were excluded because of intraoperative or early postoperative (48 hr) death. Overall 1,688 patients remained in the study. General data regarding the population intervened including the demographic and descriptive data of the patients are listed in Table 1.
Incidence and Etiology of VAP The cumulative incidence of VAP in the study cohort was 6.2% (105/1,688 patients), and VAP frequency was
258
Wei Sheng et al.
TABLE 1 Preoperative and surgical characteristics of patients who underwent major heart surgery Characteristics Preoperative Number of patients Mean age in years (SD) Sex, male/female Underlying conditions (%) Myocardial infarction (70 years, intraoperative IABP auxiliary, NYHA grade IV, the EuroSCORE, emergent surgery, inotropic support, cardiopulmonary bypass time (CPBT), blood transfusions, reintubation, mechanical ventilation days, and reoperation were associated with VAP. But the Logistic multivariate analysis showed that age >70 year old, emergent surgery, blood transfusions, reintubation, the days of mechanical ventilation were independent risk factors of VAP. As shown in Table 4.
70.6 ± 28.7
SD = Standard deviation, EF = Ejection fraction, EuroSCORE = European system for cardiac operative risk evaluation
126. Of the 126 total episodes, 10 patients had more than one VAP episode. The incidence density of VAP in our population was 23.1 episodes per 1,000 days of mechanical ventilation. The number of patients whose mechanical ventilation more than 48 hr was 198, and VAP incidence was 53.0% in this subgroup of patients. In these patients, VAP occurred about 4.5 day (3 days8 days) after the MV. The average MV time in these VAP patients was 10 days (2 days–58 days). One hundred and ninety-eight strains of pathogenic bacteria from the sputum were isolated and cultivated from these 105 VAP patients, among them the gram negative (G− ) bacillus 137 plants, accounting for 69.2% of the total plants, mainly is the P. aeruginosa, P. klebsiella, Acine-
Outcome The ICU stay time was significantly longer in patients with VAP (24.7 ± 4.5 days) than in patients without VAP (3.2 ± 1.5 days) (p < .05). Overall in-hospital mortality in our study population was 4.3% (73/1,688), with a mortality rate in patients with VAP of 25.7% (27/105) versus 2.9% (46/1,583) in patients without VAP (p < .05).
DISCUSSION Ventilator-associated pneumonia is a special type of hospital-acquired pneumonia. In recent years, VAP also gradually increased with the rapid development of the technology of mechanical ventilation. The study shows that if mechanical ventilation increases 1 day, the incidence of VAP will increase by 1%–3% [7]. VAP has Journal of Investigative Surgery
Independent Risk Factors for VAP
259
TABLE 3 Univariate analysis of preoperative, operative, and postoperative risk factors of VAP in patients underwent MHS
Preoperative risk factors Age >70 years Intra-aortic balloon Severe pulmonary hypertension Diabetes mellitus COPD Creatinine levels >1.5 mg/dl NYHA IV EuroScore(SD) Operative risk factors Emergent surgery Intra-aortic balloon Perioperative myocardial infarction Inotropic support CPBT(SD) Postoperative risk factors Blood transfusions (ml) (IQR) Reintubation Days of MV median (IQR) Reoperation
VAP (n = 105)
No VAP (n = 1583)
χ 2 /t
p
71 (67.6) 5 (4.8) 12 (11.4) 31 (29.5) 4 (3.8) 9 (8.6) 15 (14.3) 10.5 ± 4.1
677 (42.7) 41 (2.6) 162 (10.2) 412 (26.0) 52 (3.3) 102 (6.4) 136 (8.6) 6.7 ± 3.1
24.646 1.752 0.152 0.622 0.084 0.726 3.920 9.332
