Printer Friendly

Performance of the PIRO score for predicting mortality in patients with ventilator-associated pneumonia.

Ventilator-associated pneumonia (VAP) continues to be a leading cause of morbidity and mortality in the nosocomial setting. For those patients who acquire VAP, the likelihood of dying is twice as high as that observed in ventilated patients without VAP, with mortality rates ranging from 33 to 50% (1,2). To date, few scoring systems have been developed and validated to assess mortality risk in patients with VAP (3). The Acute Physiology and Chronic Health Evaluation (APACHE) II score is the most widely utilised score to assess risk of death in intensive care units and is often used in many clinical trials to stratify severity of illness, including studies in patients with VAP (4-6). However, the APACHE II score can be cumbersome to calculate and simpler, accurate scoring systems are still needed for the practising clinician. More recently, the new IBMP-10 score (Immunodeficiency, Blood pressure, Multilobar infiltrates, Platelet count, and 10-day hospitalisation) was found to be comparable to APACHE II in its ability to predict mortality for patients with VAP (7).

The PIRO (Predisposition, Insult, Response, Organ dysfunction) is a scoring concept introduced in 2003 to predict mortality among patients with sepsis (8). Most recently it has been adapted for VAP and community-acquired pneumonia (9,10). Lisboa and colleagues defined the VAP PIRO score based on four variables: presence of co-morbidities (chronic obstructive pulmonary disease, immunocompromised, heart failure, cirrhosis or chronic renal failure), bacteraemia, hypotension (systolic blood pressure <90 mmHg) and acute respiratory distress syndrome (9). One point is given for each variable present; the resultant score is the sum of these four variables. Moreover, the VAP PIRO score demonstrated better performance than the APACHE II in predicting mortality at end of intensive care unit (ICU) stay. As the PIRO score has only been validated by the original population used to derive the score criteria, the aim of this study was to validate the performance of the PIRO score for predicting ICU and 28-day mortality against APACHE II score (collected on the day of ICU admission) and VAP APACHE II score (collected on the day of VAP diagnosis) in an independent group of VAP patients.


We conducted a secondary analysis of data gathered from a previous study with 168 patients who developed VAP between July 2004 and August 2007 at our institution, Hartford Hospital, an 867-bed tertiary care medical centre located in Hartford, Connecticut, USA. The original study was a prospective, observational analysis of VAP patients who were initiated on an antibiotic clinical pathway and aimed to determine which antibiotic regimens would provide the greatest likelihood of a positive clinical response based on a pharmaco-dynamic modelling (11). This study was approved by the Hartford Hospital Institutional Review Board and informed consent was waived because all data were already collected and the study was observational in nature.

Study patients

Patients enrolled were [greater than or equal to] 17 years of age who developed VAP while admitted to one of the three ICUs (surgical, medical or neurotrauma) at our hospital. VAP was defined as an acute lower respiratory tract infection in a patient who had been mechanically ventilated for more than 48 hours or weaned from mechanical ventilation within the previous 48 hours, had a new or progressing infiltrate on chest radiograph and who met at least two of the following clinical criteria: body temperature >38[degrees]C or <36[degrees]C with no other recognised cause, white blood cell count >10,000 /mm3 or <5000 /[mm.sup.3], or a macroscopically purulent tracheal aspirate. Only the first VAP episode was included and only those patients who had a pathogen defined at baseline were evaluated as this represented a similar population to that used in the original study. VAP episodes were classified as early or late onset according to criteria established by the American Thoracic Society and the Infectious Diseases Society of America (1).


Patient demographics as well as clinical and microbiological data were recorded to calculate the APACHE II score within 24 hours of ICU admission and VAP PIRO score and VAP APACHE II score on day of VAP identification. As described by Lisboa and colleagues in the original presentation of VAP PIRO, mortality at the end of ICU stay was used as the primary endpoint (9). Mortality at 28 days after the VAP diagnosis date was also evaluated as this is a common endpoint used in VAP outcomes research.


All co-morbidity definitions were followed as described by the original validation study 9. Briefly, chronic obstructive pulmonary disease was defined as the presence of chronic bronchitis and emphysema. Immunocompromised was defined as immunodeficiency primary or secondary to radiation, use of cytotoxic drugs or steroids (daily dose greater than 20 mg of prednisolone or equivalent for >2 weeks), cancer or AIDS. Chronic heart failure was defined by patients classified as having New York Heart Association class III or IV heart failure on admission. Chronic liver disease was considered in patients with documented cirrhosis, portal hypertension, previous upper gastrointestinal bleeding secondary to portal hypertension, or encephalopathy. Chronic renal failure was defined in patients who were undergoing long-term haemodialysis. Bacteraemic episodes of VAP were defined as isolation of at least one positive blood culture not related to another source of infection, matching at least one positive respiratory sample obtained within 48 hours. At least one of the respiratory micro-organisms had to be isolated in blood cultures and all isolated micro-organisms were required to grow in simultaneous respiratory samples. Hypotension was defined as the presence of systolic blood pressure <90 mmHg within 24 hours of VAP diagnosis or need of vasopressor drugs to maintain the blood pressure above this value. Acute respiratory distress syndrome was defined according to the American-European Consensus Conference Committee criteria (12). We defined patients as a head injury if they were admitted to the neurotrauma ICU with stroke, brain anoxia or as a result of a motor vehicle accident or other trauma to the head.

Statistical analysis

Data were analysed using SigmaStat Version 2 (SYSTAT Software Inc, Chicago, IL, USA) software and area under the receiver operating characteristic (AUROC) curves were constructed by MedCalc Version 11 (MedCalc Software, Belgium). Continuous variables were analysed by Student's t-test or MannWhitney rank sum test as appropriate. Categorical variables were compared with chi-square and Fisher exact test as appropriate. Sensitivity, specificity, positive predictive value and negative predictive value for predicting ICU mortality and 28-day mortality for APACHE II, VAP APACHE II and VAP PIRO scores were determined. AUROC and 95% confidence intervals (CI) were compared between scores. To explore which baseline characteristics were predictive of ICU mortality in our independent population, logistic regression was performed utilising all variables with a P value <0.2 during univariate analyses. An a priori P value <0.05 was considered significant for all final analyses.


One hundred and sixty-eight VAP patients were evaluated and 20 were excluded because no baseline pathogen could be identified, leaving 148 patients for evaluation in the study. The baseline characteristics of patients stratified by survival to ICU discharge are displayed in Table 1. Overall ICU mortality was 25% and 28-day mortality was 25.7%. Survivors and non-survivors had median post-VAP ICU stays of 17 days (range 3 to 80) and 10 days (range 1 to 91), respectively.

Causative pathogens are provided in Table 2. Staphylococcus aureus (28.3%) was the most prevalent pathogen followed by Pseudomonas aeruginosa (27.7%) and Klebsiella pneumoniae (16.2%). The majority of pathogens were identified by tracheal aspirates (86.4%). Eight percent of patients had baseline pathogens indentified by bronchoalveolar lavage.

Most patients in this cohort had a mild to high PIRO score (0 to 1 and 2, respectively). Only 9 (6%) of the patients had a VAP PIRO score >2 and none of these patients met all four criteria identified by Lisboa and colleagues. The median (range) VAP PIRO score for non-survivors and survivors were 2 (0 to 3) and 1 (0 to 3), respectively, P=0.057. ICU mortality by PIRO score and corresponding APACHE II and VAP APACHE II ranges are provided in Figure 1.


ICU mortality gradually increased from 15.2% for patients with a PIRO score of 0 to 44.4% for a PIRO score of 3. Using increments of five points after an APACHE II and VAP APACHE II of 15, ICU mortality also gradually increased with increasing APACHE II and VAP APACHE II ranges.

A VAP PIRO score threshold >2 had the most consistent sensitivity and specificity, but these values were 57 and 59% respectively for ICU mortality, and 53 and 58% respectively for 28-day mortality (Table 3).

The AUROC for the VAP PIRO score (AUROC 0.605; 95% CI 0.521 to 0.684, P=0.03) was similar to APACHE II score (AUROC 0.631; 95% CI 0.548 to 0.709, P=0.01) for predicting ICU mortality (P value for difference between areas, P=0.68) (Figure 2).

The AUROC for the VAP APACHE II score (AUROC 0.724; 95% CI 0.645 to 0.795, P <0.0001) was also similar to both APACHE II AUROC (P=0.15) and VAP PIRO AUROC (P=0.054). The AUROC of the scores were also similar when assessing 28-day mortality as the endpoint. The VAP PIRO AUROC was 0.614 (95% CI 0.530 to 0.692, P=0.01), the APACHE II was 0.633 (95% CI 0.550 to 0.710, P=0.01) and the VAP APACHE II was 0.697 (95% CI 0.616 to 0.770, P=0.0002). Again, no differences between areas were observed (APACHE II vs VAP PIRO, P=0.77; APACHE II vs VAP APACHE II, P=0.35; and VAP PIRO vs VAP APACHE II, P=0.22).


On multivariate logistic regression, only APACHE II score and presence of bacteraemic pneumonia were independently associated with ICU-mortality (Table 4).

The VAP PIRO score was not independently predictive of ICU mortality when assessed with or without APACHE II or VAP APACHE II scores in the model. In addition, by utilising VAP APACHE II instead of APACHE II, the same variables persisted in the model with bacteraemia with odds ratio 8.88, 95% CI 1.30 to 60.58, P=0.02 and VAP APACHE II with odds ratio 1.16, 95% CI 1.08 to 1.25, P <0.0001.


The present study sought to evaluate the ability of the VAP PIRO score to predict ICU and 28-day mortality, as this score has not been validated in an independent population of VAP patients. Our findings did not confirm the robust performance observed previously with the VAP PIRO score for predicting ICU mortality (9). In our population of VAP patients, the VAP PIRO score and APACHE II score were outperformed by VAP APACHE II score in their ability to predict ICU mortality and 28-day mortality. In addition, VAP PIRO score was not a significant predictor of ICU mortality on multiple logistic regression analysis, whereas both APACHE II and VAP APACHE II were.

The VAP PIRO score was developed in a population of 441 patients with a diagnosis of VAP based on clinical criteria in addition to quantitative cultures of tracheal aspirates or bronchoalveolar lavage (9). VAP PIRO demonstrated consistent mortality discrimination among subgroups of trauma versus non-trauma, medical versus surgical and early--versus late-onset VAP populations: furthermore, it outperformed the APACHE II score as demonstrated by AUROCs of 0.81 vs 0.53 (P <0.001), respectively. However, among our population of 148 culture positive VAP patients, AUROCs for predicting ICU mortality were similar for VAP PIRO and APACHE II (0.605 vs 0.631, P=0.68). Although mortality trended upwards with increasing VAP PIRO score (Figure 1), the sensitivity and specificity for a score threshold >2 were only 57% and 59% in our population, values much lower than the 79.8% sensitivity and 73.4% specificity for this threshold reported in the derivation population.

The reasons for the observed differences between the two studies may be severalfold. First, Lisboa's population was defined in part by quantitative cultures of tracheal aspirates (96% of patients) and bronchoalveolar lavage fluid (4% of patients). Although recommended by the American Thoracic Society/Infectious Diseases Society of America guidelines for the microbiological definition of VAP1, quantitative cultures are not considered by many to impart greater diagnostic sensitivity compared with qualitative tracheal aspirates (13). The randomised study in 740 patients with suspected VAP conducted by the Canadian Critical Care Trials Group found no difference in mortality between patients who received quantitative bronchoalveolar lavage versus endotracheal aspirates without quantitation (14). Quantitative cultures were not routinely conducted at our hospital at the time of this study; however, all included patients had qualitative cultures conducted by endotracheal aspiration or bronchoalveolar lavage. Of note, when VAP PIRO and APACHE II were analysed for our entire 168 patient population (including the 20 patients who were culture negative), APACHE II remained a significant predictor of ICU- and 28-day mortality (data not shown). Second, given the variation in critical illness and underlying co-morbidites among ICU patients, it would not be surprising that the VAP populations may have been different. Lastly, antibiotic therapy was not well described in the original study. Any of these variables may have greatly affected overall mortality, which was numerically higher in their population (37%) compared with ours (25%). Potential differences among VAP patients across hospitals and continents makes complete validation of the VAP PIRO score in an independent population even more important, since clinicians and researchers who will apply the score are unlikely to pay significant attention to these minor details.

Currently, there is a scarcity of scores for predicting mortality in VAP patients. VAP PIRO is an attractive mortality predictor mainly due to its simplicity as compared with other more cumbersome scores. Although APACHE II is often utilised, it has demonstrated problems with reproducibility in some cases (15). Conversely, in a recent study, the APACHE II score defined on the day of diagnosis of VAP (VAP APACHE II) was predictive of patient mortality with an AUROC of 0.81 (6). Another recent new score for predicting mortality in patients with VAP, IBMP-10, was introduced and demonstrated an AUROC of 0.824 for predicting 14-day mortality (7). Nonetheless, this score was completely validated in the original report and data from the same population utilised herein found IBMP-10 to be less predictive than APACHE II (16). Two recent studies have analysed the concept of PIRO in predicting in-hospital mortality in patients with sepsis (17,18). In these studies, PIRO presented AUROCs between 0.69 and 0.77. However, the components of the PIRO score in these studies were rather different from the ones used to predict mortality in VAP patients.

Additionally, we performed multiple logistic regression to see independent predictors of ICU mortality in our population. Interestingly, only bacteraemia and APACHE II or VAP APACHE II scores were found as independent variables associated with ICU mortality in our cohort, including or not including VAP PIRO in the analysis. In a recent study, APACHE II score >16 was also an independent predictor of mortality. Moreover, in that study, APACHE II score had a high AUROC (0.81) (6). Differently from the original study where comorbidities were found as predictors of ICU mortality, the sole VAP PIRO component found as significant in our study was bacteraemia. This variable was also previously found as a predictor of mortality in two recent studies with VAP patients (19,20). The importance of this observation is severalfold.

Our study has limitations. First, as a single-center retrospective analysis our data might not be generalised to all settings. Second, in our cohort no patient had a VAP PIRO score of 4 and only nine patients had a score of 3. Indeed, our relatively low number of patients might have affected the results of the study.

In conclusion, our study demonstrated that APACHE II was the sole score that independently predicted mortality for an independent population of culture positive VAP patients. VAP PIRO score was not a good predictor of ICU- and 28-day mortality. Its low sensitivity and specificity of VAP PIRO score preclude its use clinically as a predictor of ICU mortality.


This study was undertaken with funds from the Center for Anti-Infective Research and Development at Hartford Hospital, Hartford, CT, USA. G. H. Furtado is supported by Coordenacao de Aperfeicoamento de Pessoal de Nivel Superior, Brasilia, Brazil.


(1.) 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.

(2.) Safdar N, Dezfulian C, Collard HR, Saint S. Clinical and economic consequences of ventilator-associated pneumonia: a systematic review. Crit Care Med 2005; 33:2184-2193.

(3.) Napolitano LM. Use of severity scoring and stratification factors in clinical trials of hospital-acquired and ventilator-associated pneumonia. Clin Infect Dis 2010; 51 (Suppl 1):S67-80.

(4.) Niskanen M, Kari A, Nikki P, Iisalo E, Kaukinen L, Rauhala V et al. Acute physiology and chronic health evaluation (APACHE II) and Glasgow coma scores as predictors of outcome from intensive care after cardiac arrest. Crit Care Med 1991; 19:1465-1473.

(5.) Grmec S, Gasparovic V. Comparison of APACHE II, MEES and Glasgow Coma Scale in patients with nontraumatic coma for prediction of mortality. Acute Physiology and Chronic Health Evaluation. Mainz Emergency Evaluation System. Crit Care 2001; 5:19-23.

(6.) Gursel G, Demirtas S. Value of APACHE II, SOFA and CPIS scores in predicting prognosis in patients with ventilator-associated pneumonia. Respiration 2006; 73:503-508.

(7.) Mirsaeidi M, Peyrani P, Ramirez JA. Predicting mortality in patients with ventilator-associated pneumonia: The APACHE II score versus the new IBMP-10 score. Clin Infect Dis 2009; 49:72-77.

(8.) Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003; 31:1250-1256.

(9.) Lisboa T, Diaz E, Sa-Borges M, Socias A, Sole-Violan J, Rodriguez A et al. The ventilator-associated pneumonia PIRO score: a tool for predicting ICU mortality and health-care resources use in ventilator-associated pneumonia. Chest 2008; 134:1208-1216.

(10.) Rello J, Rodriguez A, Lisboa T, Gallego M, Lujan M, Wunderink R. PIRO score for community-acquired pneumonia: a new prediction rule for assessment of severity in intensive care unit patients with community-acquired pneumonia. Crit Care Med 2009; 37:456-462.

(11.) Nicasio AM, Eagye KJ, Nicolau DP, Shore E, Palter M, Pepe J et al. Pharmacodynamic-based clinical pathway for empiric antibiotic choice in patients with ventilator-associated pneumonia. J Crit Care 2010; 25:69-77.

(12.) Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149:818824.

(13.) Niederman MS. The argument against using quantitative cultures in clinical trials and for the management of ventilator-associated pneumonia. Clin Infect Dis 2010; 51 (Suppl 1):S93-99.

(14.) Canadian Critical Care Trials Group .A randomized trial of diagnostic techniques for ventilator-associated pneumonia. N Engl J Med 2006; 355:2619-2630.

(15.) Booth FV, Short M, Shorr AF, Arkins N, Bates B, Qualy RL et al. Application of a population-based severity scoring system to individual patients results in frequent misclassification. Crit Care 2005; 9:R522-529.

(16.) Wiskirchen DE, Kuti JL, Nicolau DP. Validation of the IBMP 10 score for predicting mortality in patients with ventilator associated pneumonia (abstract #K-2189). 50th Interscience Conference on Antimicrobial Agents and Chemotherapy, Boston, MA, USA, 12 September 2010.

(17.) Rubulotta F, Marshall JC, Ramsay G, Nelson D, Levy M, Williams M. Predisposition, insult/infection, response, and organ dysfunction: A new model for staging severe sepsis. Crit Care Med 2009; 37:1329-1335.

(18.) Moreno RP, Metnitz B, Adler L, Hoechtl A, Bauer P, Metnitz PGH. Sepsis mortality prediction based on predisposition, infection and response. Intensive Care Med 2008; 34:496-504.

(19.) Agbaht K, Diaz E, Munoz E, Lisboa T, Gomez F, Depuydt PO et al. Bacteremia in patients with ventilator-associated pneumonia is associated with increased mortality: A study comparing bacteremic vs. nonbacteremic ventilator-associated pneumonia. Crit Care Med 2007; 35:2064-2070.

(20.) Siempos II, Vardakas KZ, Kyriakopoulos CE, Ntaidou TK, Falagas ME. Predictors of mortality in adult patients with ventilator-associated pneumonia: a meta-analysis. Shock 2010; 33:590-601.

G. H. FURTADO *, D. E. WISKIRCHEN ([dagger]), J. L. KUTI ([double dagger]), D. P. NICOLAU ([section])

Center for Anti-Infective Research and Development, Division of Infectious Diseases, Hartford Hospital, Hartford, Connecticut, USA

* MD, Infectious Diseases Physician and Research Fellow, Center for AntiInfective Research and Development, Division of Infectious Diseases, Hartford Hospital and Division of Infectious Diseases, Hospital Sao Paulo, Federal University of Sao Paulo, Sao Paulo, Brazil.

([dagger]) PharmD, Infectious Diseases Research Fellow.

([double dagger]) PharmD, Vice-Director.

([section]) PharmD, Director.

Address for correspondence: Dr G. H. Furtado, Division of Infectious Diseases, Federal University of Sao Paulo, 690 Napoleao de Barros Street, 2nd floor, Sao Paulo, SP 04024-002, Brazil. Email:

Accepted for publication on January 3, 2012.
Table 1
Patient baseline characteristics and univariate analysis for ICU
mortality among the 148 VAP patients *

Variables Survivors, n=111 Non-survivors, n=37

Age, median (range) 60 (17-94) 64 (20-87)
Male gender 74 (66.7) 27 (72.9)
Length of stay after VAP, 17 (3-80) 10 (1-91)
 days, median (range)
APACHE II score, mean, SD 18 ([+ or -] 8.04) 22 ([+ or -] 9.56)
PIRO score
 Mild (0-1) 66 (59.5) 16 (43.2)
 High (2) 40 (36) 17 (46)
 Very high (3-4) 5 (4.5) 4 (10.8)
 Cancer 12 (10.8) 4 (10.8)
 Chronic heart failure 10 (9) 7 (18.9)
 Chronic liver disease 5 (4.5) 5 (13.5)
 Chronic renal failure 1 (0.9) 2 (5.4)
 COPD 14 (12.6) 6 (16.2)
 Immunocompromised 19 (17.1) 8 (21.6)
Multilobar pneumonia 69 (62.1) 26 (70.2)
Head injury 20 (18) 11 (29.7)
ARDS 45 (40.5) 14 (37.8)
Bacteraemia 2 (1.8) 4 (10.8)
Hypotension 54 (48.6) 21 (56.7)
Late-onset VAP 79 (71.1) 20 (54)
Polymicrobial VAP 27 (24.3) 8 (21.6)
Type of ICU
 MICU 22 (19.8) 9 (24.3)
 SICU 45 (40.5) 14 (37.8)
 NTICU 44 (39.6) 14 (37.8)
Pseudomonas aeruginosa 29 (26.1) 10 (27)
 at baseline
MRSA at baseline 10 (9) 5 (13.5)

Variables P value

Age, median (range) 0.08
Male gender 0.61
Length of stay after VAP, 0.003
 days, median (range)
APACHE II score, mean, SD 0.005
PIRO score 0.14
 Mild (0-1)
 High (2)
 Very high (3-4)
 Cancer 1.00
 Chronic heart failure 0.18
 Chronic liver disease 0.13
 Chronic renal failure 0.15
 COPD 0.78
 Immunocompromised 0.71
Multilobar pneumonia 0.48
Head injury 0.19
ARDS 0.92
Bacteraemia 0.03
Hypotension 0.44
Late-onset VAP 0.08
Polymicrobial VAP 0.91
Type of ICU 0.84
Pseudomonas aeruginosa 0.91
 at baseline
MRSA at baseline 0.63

* Data are presented as number (% of total), unless otherwise noted.
ICU=intensive care unit, VAP=ventilator-associated pneumonia,
APACHE=Acute Physiology and Chronic Health Evaluation,
PIRO=Predisposition, Insult, Response, Organ dysfunction,
COPD=chronic obstructive pulmonary disease, ARDS=acute respiratory
distress syndrome, MICU=medical intensive care unit, SICU=surgical
intensive care unit, NTICU=neurotrauma intensive care unit,
MRSA=methicillin-resistant Staphylococcus aureus.

Table 2
Micro-organisms isolated among 148patients with VAP *

Pathogens Overall prevalence, n = 148 ([dagger])

Staphylococcus aureus 42 (28.4)
Pseudomonas aeruginosa 41 (27.7)
Klebsiella species 24 (16.2)
Enterobacter species 20 (13.5)
Escherichia coli 11 (7.4)
Acinetobacter baumannii 6 (4.1)
Stenotrophomonas maltophilia 4 (2.7)
Other ([dagger][dagger]) 34 (23.0)

* Data are presented as number (% of total). ([dagger]) 148 of the
168 included patients had a baseline organism identified.
([dagger][dagger]) Other includes: Proteus mirabilis, Citrobacter
species, Haemophilus influenzae, Streptococcus pneumoniae,
Providencia rettgeri, Burkholderia cepacia, Moraxella catarrhalis,
Serratia species, Morganella species and Kluyvera species.

Table 3
Identification and prediction characteristics of VAP PIRO score
with differing score thresholds for ICU mortality in the
148 VAP patients

PIRO Sensitivity, Specificity, PPV, NPV,
score % % % %

[greater than or equal to] 1 86 25 27.8 84.8
[greater than or equal to] 2 57 59 31.8 80.5
[greater than or equal to] 3 11 95 44.4 76.3
4 0 100 NA 75.0

VAP=ventilator-associated pneumonia, PIRO=Predisposition,
Insult, Response, Organ dysfunction, ICU=intensive care unit,
PPV=positive predictive value, NPV=negative predictive value,
NA=not applicable.

Table 4
Variables associated with ICU mortality among the 148 VAP patients
after multivariate logistic regression*

Variable Odds ratio 95% CI P value

Bacteraemia 7.16 1.19-42.98 0.03
APACHE II score 1.06 1.01-1.11 0.006

* The following variables were tested in the multivariate logistic
regression: PIRO score as ordinal variable, age, APACHE II
score, chronic heart failure, chronic liver disease, head injury,
bacteraemia and late-onset VAP. ICU=intensive care unit,
VAP=ventilator-associated pneumonia, CI=confidence interval,
APACHE=Acute Physiology and Chronic Health Evaluation,
PIRO=Predisposition, Insult, Response,Organ dysfunction.
COPYRIGHT 2012 Australian Society of Anaesthetists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:predisposition, insult, response, organ dysfunction
Author:Furtado, G.H.; Wiskirchen, D.E.; Kuti, J.L.; Nicolau, D.P.
Publication:Anaesthesia and Intensive Care
Article Type:Report
Geographic Code:3BRAZ
Date:Mar 1, 2012
Previous Article:An observational study exploring amplitude-integrated electroencephalogram and Spectral Edge Frequency during paediatric anaesthesia.
Next Article:Lignocaine plasma levels following topical gel application in laparoscopic and hysteroscopic procedures.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters