Incidence, risk factors and mortality of nosocomial pneumonia in Intensive Care Units: A prospective study
© Alp et al; licensee BioMed Central Ltd. 2004
Received: 07 July 2004
Accepted: 15 September 2004
Published: 15 September 2004
To determine the frequency, risk factors and mortality of nosocomial pneumonia a prospective study was conducted in the intensive care units. In the study period, 2402 patients were included. The nosocomial pneumonia was defined according to the Centers for Disease Control Criteria. Overall, 163 (6.8%) of the patients developed nosocomial pneumonia and 75.5% (n = 123) of all patients with nosocomial pneumonia were ventilator-associated pneumonia. 163 patients who were admitted to the intensive care unit during the same period but had no bacteriologic or histologic evidence of pneumonia were used as a control group. The APACHE II score, coma, hypoalbuminemia, mechanical ventilation, tracheotomy, presence of nasogastric tube were found as independent risk factors. Crude and attributable mortality were 65% and 52.6%, respectively. The mortality rate was five times greater in the cases (OR: 5.2; CI 95%: 3.2–8.3). The mean length of stay in the intensive care unit and hospital in the cases were longer than controls (p < 0.0001). Patients requiring mechanical ventilation have a high frequency of nosocomial pneumonia.
Nosocomial pneumonia (NP) is the most frequent nosocomial infection in the intensive care units (ICU). The reported frequency varies with the definition, the type of hospital or ICU, the population of patients, and the type of rate calculated. In the recent studies, the incidence was reported as 6.8–27% [1–4]. In an one day point prevalence study in European ICUs, ICU-acquired pneumonia accounted for 46.9% of nosocomial infections . The National Nosocomial Infections Surveillance (NNIS) system reported that NP accounts for 31% of all nosocomial infections in intensive care units . The risk of pneumonia is increased in the intubated patients receiving mechanical ventilation (MV) and the ventilator associated pneumonia (VAP) frequencies varied between 7–70% in different studies [7–9]. NP developed at a rate of 0.9 cases per 1000 patient-days in non-ventilated patients versus rates of 20.6 cases per 1000 patient-ventilator-days and 14.8 cases per 1000 patient-days in patients who received any MV . NP is also associated with high morbidity and mortality in ICUs. The increasing incidence of infections caused by antibiotic-resistant pathogens contributes to the seriousness of these infections. The mortality rate reaches to 20–50%, and also NP caused by high-risk pathogens (Pseudomonas aeruginosa, Acinetobacter spp., Stenotrophomonas maltophilia) are associated with higher mortality [1, 11, 12]. Patients with NP, stay 1 to 2 weeks longer than those without NP and result in higher costs .
Studies on NP are mainly reported from the United States and European countries, whereas studies from around the world are missing. The aims of this study were to assess incidence, risk factors and mortality of NP in Eurasian intensive care units.
Between February 2001 and February 2002 a prospective study was conducted among intensive care units (ICU) patients of the Erciyes University Hospital. This university hospital is a teaching hospital and full time intensivists care the patients in ICUs. Patients from the surgical ICU (SICU) (24 beds), medical ICU (MICU) (9 beds) and burn unit (7 beds) were included. The SICU consist of 8 neurosurgical (NICU), 8 general surgery (GICU) and 8 cardiac surgery (CICU) beds. Patients older than 16 yr of age were included. The same infection control doctor collected data and intensivist reviewed the diagnosis of pneumonia. Data collection included physical examination findings, APACHE II scores on admission, consciousness, risk factors (intubation, MV, presence of nasogastric tube, enteral nutrition, tracheotomy), prior surgery, immunosuppression, prior antimicrobial and antacid or histamine type 2 (H2) blocker therapy, clinical outcome, length of stay in ICU and in the hospital.
163 patients who were admitted to the ICU during the same period but had no bacteriologic or histologic evidence of pneumonia were used as a control group.
In the ICUs infection control doctor collects active surveillance data routinely and empiric antibiotic therapy is directed at the most prevalent and virulent pathogens reported in these data. Appropriate antibiotic therapy included the administration of at least one antibiotic with in vitro activity against the bacterial pathogens isolated from the patient's respiratory secretions, as well as from blood and pleural fluid when applicable .
NP was considered when new and persistent (more than 48 h) pulmonary infiltrates not otherwise explained appeared on chest radiographs. Moreover, at least two of the following criteria were also required: 1) fever >38°C; 2) peripheral leukocyte count >10 000/mm3; 3) purulent endotracheal secretions with a Gram stain showing one or more types of bacteria . VAP was considered when its onset occurred after 48 h of MV and was judged not to have been incubated before starting MV . Admission APACHE II score was used to determine the severity of the illness, and attributable mortality was registered, as were laboratory values, electrocardiogram, x-ray, and arterial blood gas values.
Extra length of stay was calculated comparing the extra stay after onset of pneumonia in the cases and after a reference date (the mean value of the extra stay after onset of pneumonia in the cases) in the control group.
Giemsa stains of sputum samples were performed for all patients. Sputum samples, containing more than 25 polimorphonuclear leukocyte (pnl) and less than 10 (×100) epithel were classified as purulent. If necessary, samples were obtained by nasotracheal aspiration. In that case, samples containing more than 10 pnl (×1000) were defined as purulent. Quantitative cultures of all purulent samples were performed using standard methods. Susceptibility testing was performed by disc diffusion method. In the absence of an alternative diagnosis a bronchoalveolar lavage was performed. In some case, pleural fluid was obtained by thoracentesis and examined for cell count, smear, Gram- and Giemsa-staining and microbiological culture.
All data were evaluated using SPSS. Parameters were compared using univariate and multivariate logistic regression and chi-square tests. Student t test was used to compare the extra length of stay. Data were given as mean ± SD and a p-value of <0.05 was accepted as significant.
Numbers of patients and length of stay in ICU
No. of beds
Total patient (n)
Total length of stay (d)
Mean length of stay (d)
Demographic Factors of Study Patients*
NP patients (n = 163)
Control (n = 163)
53.30 ± 16.05
51.50 ± 16.87
Admission APACHE II
10.86 ± 3.42
10.18 ± 4.56
Characteristics of patients
Age (mean ± SD)
52.9 ± 15.1
50.25 ± 17.57
58.28 ± 14.38
55.00 ± 14.68
53.3 ± 16.1
Admission APACHE II, (mean ± SD)
11.2 ± 3.3
10.43 ± 3.12
11.03 ± 3.67
11.20 ± 6.38
10.9 ± 3.4
NP APACHEII, (mean ± SD)
16.2 ± 5.1
14.73 ± 3.94
16.30 ± 4.79
14.80 ± 6.37
15.6 ± 4.7
Length of stay in ICU (d, mean ± SD)
21.3 ± 21.4
14.42 ± 8.87
19.47 ± 8.20
12.20 ± 3.96
18.0 ± 14.7
Length of hosp. stay (d, mean ± SD)
25.0 ± 22.5
21.60 ± 11.93
24.78 ± 12.13
18.40 ± 8.26
23.5 ± 16.4
Rate of VAP in ICUs
No. of patients required MV
No. of patients with VAP (%)
During the study period patients received 3128 ventilation days, with an average duration of 11.3 ± 10.0 days per ventilated patient. The device-related incidence rate for VAP was 39.3/1000 ventilation days. The incidence per 1000 ventilation days was 41.9 in SICU, 36.6 in MICU, 66.0 in NICU, 38.0 in GICU, and 9.1 in CICU patients. The mean onset day of NP after MV was 4.2 ± 3.9 days.
Results of univariate analysis of potential risk factors for NP
95% Confidence Interval
0.99 – 1.02
1.22 – 1.38
3.75 – 11.48
0.93 – 2.97
1.91 – 8.01
0.35 – 1.16
Central nervous system disorder
0.71 – 1.77
Presence of nasogastric
Previous antibiotic treatment
Antacids or H2 antagonist therapy
187 pneumonia episodes were observed during the study period, resulting in the isolation of 257 microorganisms. The most commonly isolated pathogens were Gram-negative bacteria (85.6%). Among these pathogens, A. baumannii (29.6%), P. aeruginosa (20.6%), Klebsiella pneumoniae (14.4%) were the most common.
Empiric antibiotic therapy was based on previous surveillance cultures and the Gram stain results. Therapy was adjusted according to the reports of susceptibility testing.
Comparisons of outcomes between NP and control group
NP group n (%)
Control group n (%)
Mortality rates in high risk pathogens
Appropriateness of empiric therapy and mortality
Appropriate n (%)
Inappropriate n (%)
The mean length of stay in the ICU and hospital for the patients with NP were 18.04 ± 14.74 days and 23.49 ± 16.44 days, respectively. The mean length of stay in the ICU and hospital for the control group 3.10 ± 3.03 and 9.64 ± 5.08 days, respectively. This difference was statistically significant (p < 0.0001). The extra stay in the control group was 4.36 ± 3.87 and 17.04 ± 14.17 in the patients (p < 0.001).
The incidence of NP was reported different in different studies, which may be justified by the presence of different populations with variable ages, underlying diseases, and other associated risk factors. Incidence ranges from 6.8 to 27% [1–4] and also in this study it was 6.8%. Development of NP varies according to the different type of ICUs. Craven et al.  reported that the rate of pneumonia was higher in MICU but the difference was not significant. In the present study, the rate of NP and VAP was significantly higher in MICU than SICU, possible due to the differences in the proportion of patients that needed MV and the duration of MV.
MV increases the risk of NP by 3- to 10-fold [1, 18–23], resulting in an VAP incidence of 7 to 70% [7–9]. Generally, the duration of mechanical ventilation increases the risk of pneumonia. Cook et al.  reported that the rate of VAP increased 3% per day in the first week of ventilation, 2% per day in the second week, and 1% per day in the third week. In this study, 75.5% of the cases with NP occurred in ventilated patients. From 724 patients who required MV 123 (17%) developed VAP. Accordingly, patients on MV had a 3-fold higher risk to develop NP than the non-ventilated patients. Consequently, the use of non-invasive MV should be preferred whenever possible, since it has lower rates of nosocomial infections [25–27].
Coma was described as another important risk factor for NP. In these patients, local defense mechanisms of the respiratory airway are altered, allowing microorganisms to better attach to and colonize the mucosal surface. Furthermore, depression of the level of consciousness significantly increases the chance of aspiration, and as a result development of NP [3, 28]. In our study, comatose patients had a 2-fold increased risk of NP.
The causative agents of NP differ by the study population and diagnostic techniques but generally Gram-negative bacteria are the most common ones [3, 4, 28–33]. Colonization of the oropharynx, trachea or stomach with Gram-negative pathogens has been identified as a risk factor for NP [15, 31]. Also in our study, the most common pathogens were Gram-negative bacteria. Furthermore, prior antibiotic therapy and COPD, leading to colonization with Gram-negative aerobic pathogens, were reported to be risk factors for the development of NP [11, 28, 30, 34]. In our patient population, univariate analysis suggested that previous antibiotic treatment and COPD increased the risk of pneumonia, but interestingly they were not independent risk factors in multivariate analysis. Furthermore, the presence of a naso-gastric tube was found to be a risk factor in our study population. Naso-gastric tubes impair the function of the gastroesophageal sphincter and increase the risk of maxillary sinusitis, oropharyngeal colonization and reflux, all of which may lead to migration of bacteria . Accurate evaluation of nutritional status and early initiation of enteral feeding is important in ICUs patients and can aid to preserve the gastrointestinal epithelium and prevent bacterial colonization. However, it may also increase the risk of gastric distention, colonization, aspiration, and pneumonia. Though, to reduce the risk of NP, it is important to avoid unnecessary enteral nutrition . In univariate analysis, we found enteral feeding as a risk factor, but in multivariate analysis it was not an independent risk factor. For a long time it was assumed that increased gastric pH levels e.g. after the use of antacids, would allow Gram-negative microorganisms to multiply in the stomach, and consequently lead to an increased rate of NP. Our study results confirm what was reported by George et al. , namely that the use antacids or H2 antagonists did not increase the risk of NP.
In the literature, tracheotomy is described as a significant risk factor for NP. Bronchial colonization during the procedure and (prolonged) continuation of sedation after the procedure will furthermore increase the occurrence of NP , a fact that was also seen in our patients. Patients with tracheotomy had a 7-fold increased risk of NP.
The role of advanced age and high APACHE II scores as risk factors of NP are still under discussion. While Kollef et al.  report them as significant risk factors, earlier investigations do not support this [1, 30]. In our present study, the APACHE II score was a significant risk factor for the development of NP; suggested that the severity of the general condition of the patient was important. Besides, uremia was found as a risk factor in univariate analysis.
Patients with NP have a significantly higher morbidity and mortality [12, 35, 38]. Heyland et al.  reported the crude mortality rate of VAP 23.7% and an attributable mortality rate 32.3%. However, numerous studies have demonstrated that severe underlying illness predisposes patients in the ICU to the development of pneumonia, and their mortality rates are, as a result, high. Survival in patients with NP primarily by the degree of severity of illness at the time of diagnosis [23, 39, 40]. On the other hand, this does not exclude the possibility that certain subgroups of patients, such as patients with VAP caused by antibiotic resistant bacteria may have had extra attributable mortality rate . In our study, the crude mortality rates for cases and controls were 65.0% and 26.4%, respectively and the mortality rates were highest in high risk pathogens. The mortality rate was five times greater in cases and attributable mortality of NP was 52.6%. Recent clinical investigations suggest that patients receiving inappropriate initial therapy have a greater mortality rate compared to patients receiving antibiotics to which the isolated bacteria were sensitive. However, in this study there was no statistically difference between the mortality rates of the patients who received appropriate and inappropriate initial therapy.
As a result of the increased morbidity, patients with VAP remain hospitalized for 4–17 days longer than controls [35–38]. This observation was confirmed in the present study. The incidence of NP and VAP in MICU were significantly higher than in SICU patients and consequently the length of stay in the MICU was significantly higher than in the SICU.
In conclusion, NP is a major cause of morbidity and mortality in ICU patients. Especially patients on mechanical ventilation are at high risk. Studies determining the impact of "old" and "new" risk factors of NP should repeatedly be performed in order to effectively guide the implementation of preventive measures methods.
- Fagon JY, Chastre J, Hance AJ, Montravers P, Novara A, Gibert C: Nosocomial pneumonia in ventilated patients: A cohort study evaluating attributable mortality and hospital stay. Am J Med. 1993, 94: 281-288. 10.1016/0002-9343(93)90060-3View ArticlePubMedGoogle Scholar
- Vanhems P, Lepape A, Savey A, Jambou P, Fabry J: Nosocomial pulmonary infection by antimicrobial-resistant bacteria of patients hospitalized in intensive care units: risk factors and survival. J Hosp Infect. 2000, 45: 98-106. 10.1053/jhin.2000.0752View ArticlePubMedGoogle Scholar
- Rello J, Ausina V, Castella J, Net A, Prats G: Nosocomial respiratory tract infections in multiple trauma patients. Influence of level of consciousness with implications for therapy. Chest. 1992, 102: 525-529.View ArticlePubMedGoogle Scholar
- Richards MJ, Edwards JR, Culver DH, Gaynes RP: Nosocomial infections in medical intensive care units in the United States. National Nosocomial Infections Surveillance System. Crit Care Med. 1999, 27: 887-893. 10.1097/00003246-199905000-00020View ArticlePubMedGoogle Scholar
- Vincent JL, Bihari DJ, Suter PM: The prevalence of nosocomial infection in intensive care units in Europe. Results of the European Prevalence of Infection in Intensive Care (EPIC) Study. EPIC International Advisory Committee. JAMA. 1995, 274: 639-644. 10.1001/jama.274.8.639View ArticlePubMedGoogle Scholar
- Richards MJ, Edwards JR, Culver DH, Gaynes RP: Nosocomial infections in combined medical-surgical intensive care units in the United States. Infect Control Hosp Epidemiol. 2000, 21: 510-515.View ArticlePubMedGoogle Scholar
- Fagon JY, Chastre J, Domart Y: Nosocomial pneumonia in patients receiving continous mechanical ventilation: prospective analysis of 52 episodes with use of a protected specimen brush and quatitative culture techniques. Am Rev Respir Dis. 1989, 139: 877-884.View ArticlePubMedGoogle Scholar
- Craven DE, Steger KA, Barber TW: Preventing nosocomial pneumonia: state of the art and perspectives for the 1990s. Am J Med. 1991, 91 (3B): 44S-53S. 10.1016/0002-9343(91)90343-VView ArticlePubMedGoogle Scholar
- Andrews CP, Coalson JJ, Smith JD, Johanson WG: Diagnosis of nosocomial bacterial pneumonia in acute, diffuse lung injury. Chest. 1981, 80: 254-258.View ArticlePubMedGoogle Scholar
- George DL: Nosocomial pneumonia. In: Hospital Epidemiology and Infection Control. Edited by: Mayhall CG. 1996, 175-195. Baltimore: Williams & WilkinsGoogle Scholar
- Kollef MH: Ventilator-associated pneumonia. JAMA. 1993, 270: 1965-1970. 10.1001/jama.270.16.1965View ArticlePubMedGoogle Scholar
- Kollef MH, Silver P, Murphy DM, Trovillion E: The effect of late-onset ventilator-associated pneumonia in determining patient mortality. Chest. 1995, 108: 1655-1662.View ArticlePubMedGoogle Scholar
- Strausbaugh LJ: Nosocomial Respiratory Infections. In: Principles and Practice of Infectious Diseases. Edited by: Mandell GL, Bennett JE, Dolin R. 2000, 3020-3028. Churchill Livingstone, PhiladelphiaGoogle Scholar
- Blot S, Vandewoude K, Colardyn F: Nosocomial bacteremia involving Acinetobacter baumannii in critically ill patients: a matched cohort study. Intensive Care Med. 2003, 29: 471-475.PubMedGoogle Scholar
- Centers for Disease Control and Prevention: Guidelines for prevention of nosocomial pneumonia. MMWR Recomm Rep. 1997, 46 (RR-1): 1-79.Google Scholar
- Bonten MJM: Controversies on diagnosis and prevention of ventilator-associated pneumonia. Diagn Microbiol Infect Dis. 1999, 34: 199-204. 10.1016/S0732-8893(99)00040-1View ArticlePubMedGoogle Scholar
- Craven DE, Kunches LM, Lichtenberg DA: Nosocomial infection and fatality in medical and surgical intensive care units patients. Arch Intern Med. 1988, 148: 1161-1168. 10.1001/archinte.148.5.1161View ArticlePubMedGoogle Scholar
- Bell RC, Coalson JJ, Smith JD, Johanson WG: Multiple organ system failure and infection in adult respiratory distress syndrome. Ann Intern Med. 1983, 99: 293-298.View ArticlePubMedGoogle Scholar
- Chevret S, Hemmer M, Carlet J, Langer M: Incidence and risk factors of pneumonia acquired in intensive care units. Results from a multicenter prospective study on 996 patients. European Cooperative Group on Nosocomial Pneumonia. Intensive Care Med. 1993, 19: 256-264.View ArticlePubMedGoogle Scholar
- Langer M, Mosconi P, Cigada M, Mandelli M: Long-term respiratory support and risk of pneumonia in critically ill patients. Intensive Care Unit Group of Infection Control. Am Rev Respir Dis. 1989, 140: 302-305.View ArticlePubMedGoogle Scholar
- Celis R, Torres A, Gatell JM, Almela M, Rodriquez-Roisin R, Agusti-Vidal A: Nosocomial pneumonia. A multivariate analysis of risk and prognosis. Chest. 1988, 93: 318-324.View ArticlePubMedGoogle Scholar
- Cross AS, Roup B: Role of respiratory assistance devices in endemicnosocomial pneumonia. Am J Med. 1981, 70: 681-685. 10.1016/0002-9343(81)90596-9View ArticlePubMedGoogle Scholar
- Chastre J, Fagon JY: Ventilator-associated pneumonia. Am J Respir Crit Care Med. 2002, 165: 867-903.View ArticlePubMedGoogle Scholar
- Cook DJ, Walter SD, Cook RJ: Incidence of and risk factors for ventilator associated pneumonia in critically ill patients. Ann Intern Med. 1998, 129: 433-440.View ArticlePubMedGoogle Scholar
- Antonelli M, Conti G, Rocco M: A comparison of noninvasive positive-pressure ventilation and conventional mechanical ventilation in patients with acute respiratory failure. N Engl J Med. 1998, 339: 429-435. 10.1056/NEJM199808133390703View ArticlePubMedGoogle Scholar
- Nourdine K, Combes P, Carton MJ, Beuret P, Cannamela A, Ducreux JC: Does noninvasive ventilation reduce the ICU nosocomial infection risk? A prospective clinical survey. Intensive Care Med. 1999, 25: 567-573. 10.1007/s001340050904View ArticlePubMedGoogle Scholar
- Girou E, Schortgen F, Delclaux C: Association of noninvasive ventilation with nosocomial infections and survival in critically ill patients. JAMA. 2000, 284: 2361-2367. 10.1001/jama.284.18.2361View ArticlePubMedGoogle Scholar
- Ewig S, Torres A, El-Ebiary M: Bacterial colonization patterns in mechanically ventilated patients with traumatic and medical head injury. Am J Respir Crit Care Med. 1999, 159: 188-198.View ArticlePubMedGoogle Scholar
- Georges H, Leroy O, Guery B, Alfandari S, Beaucaire G: Predisposingfactors for nosocomial pneumonia in patients receiving mechanical ventilation and requiring tracheotomy. Chest. 2000, 118: 767-774. 10.1378/chest.118.3.767View ArticlePubMedGoogle Scholar
- Memish ZA, Cunningham G, Oni GA, Djazmati W: The incidence and riskfactors of ventilator-associated pneumonia in a Riyadh Hospital. Infect Control Hosp Epidemiol. 2000, 21: 271-273.View ArticlePubMedGoogle Scholar
- Garrouste-Orgeas M, Chevret S, Arlet G: Oropharyngeal or gasric colonization and nosocomial pneumonia in adult intensive care unit patients. Am J Respir Crit Care Med. 1997, 156: 1647-1655.View ArticlePubMedGoogle Scholar
- Rello J, Torres A: Microbial causes of ventilator-associated pneumonia. Semin Respir Infect. 1996, 11: 24-31.PubMedGoogle Scholar
- Rello J, Quintana E, Ausina V: Incidence, etiology, and outcome of nosocomial pneumonia in mechanically ventilated patients. Chest. 1991, 100: 439-444.View ArticlePubMedGoogle Scholar
- Soler N, Torres A, Ewig S: Bronchial microbial patients in severe exacerbations of chronic obstructive pulmonary disease (COPD) requiring mechanical ventilation. Am J Respir Crit Care Med. 1998, 157: 1498-1505.View ArticlePubMedGoogle Scholar
- Leal-Noval SR, Marquez-Vacaro JA, Garcia-Curiel A: Nosocomial pneumonia in patients undergoing heart surgery. Crit Care Med. 2000, 28: 935-940. 10.1097/00003246-200004000-00004View ArticlePubMedGoogle Scholar
- George DL, Falk PS, Wunderink RG: Epidemiology of ventilator-acquired pneumonia based on protected bronchoscopic sampling. Am J Respir Crit Care Med. 1998, 158: 1839-1847.View ArticlePubMedGoogle Scholar
- Kollef MH, Prentice D, Shapiro SD: Mechanical ventilation with or without daily changes of in-line suction catheters. Am J Respir Crit Care Med. 1997, 156: 466-472.View ArticlePubMedGoogle Scholar
- Heyland DK, Cook DJ, Griffith L, Keenan SP, Brun-Buisson C: The attributable morbidity and mortality of ventilator-associated pneumonia in the critically ill patient. Am J Respir Crit Care Med. 1999, 159: 1249-1256.View ArticlePubMedGoogle Scholar
- Rello J, Rue M, Jubert P: Survival in patients with nosocomial pneumonia: Impact of the severity of illness and the etiologic agent. Crit Care Med. 1997, 25: 1862-1867. 10.1097/00003246-199711000-00026View ArticlePubMedGoogle Scholar
- Rello J, Diaz E: Pneumonia in the intensive care unit. Crit care Med. 2003, 31: 2544-2551. 10.1097/01.CCM.0000089928.84326.D2View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.