|Year : 2023 | Volume
| Issue : 1 | Page : 94-100
Hospital-acquired pneumonia pattern in the intensive care units of a governmental hospital: A prospective longitudinal study
Mina Yakoub1, Fayek Elkhwsky1, Ayman El Tayar2, Iman El Sayed1
1 Department of Biomedical Informatics and Medical Statistics, Medical Research Institute, Alexandria University, Alexandria, Egypt
2 Department of Intensive Care Medicine, Damanhour Medical National Institute, General Organization of Teaching Hospitals and Institutes, Cairo, Egypt
|Date of Submission||26-Aug-2021|
|Date of Decision||04-Apr-2022|
|Date of Acceptance||07-Jun-2022|
|Date of Web Publication||24-Jan-2023|
165 El-Horreya Avenue, El-Hadara, Alexandria
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: Epidemiological data on Hospital-Acquired Pneumonia (HAP) are scarce inside Intensive Care Units (ICUs). Aim: This study aims to quantify the incidence of HAP, determine the predictors of HAP, calculate HAP-related mortality risk ratio as well as pinpoint the different risk factors contributing to mortality. Subjects and Methods: A prospective longitudinal study was conducted at a governmental hospital's general ICUs over 12 months. We included adult patients admitted for at least 72 h before signs appear. We utilized a logistic regression model for fatality outcome and cox proportional hazard model for HAP outcome. Results: Of 356 patients, 133 patients developed Ventilated-Acquired Pneumonia (VAP), 76 patients with Non-Ventilated HAP (NV-HAP), as well as 147 patients did not acquire HAP. The incidence of HAP was 28 cases of HAP per 1000 person-days, as well as the mortality rate was 74 per 100 days, while the Attributable Risk Percentage (ARP) was 85%. This high fatality rate was clarified by independent predictors as reintubation (odds ratio [OR] = 8.99, P < 0.001), ICU duration ≥5 days (OR = 7.29, P = 0.02), HAP outcome (OR = 6.49, P = 0.001), diabetes mellitus (DM) (OR = 2.98, P = 0.004), APACHE II ≥17 (OR = 2.76, P = 0.004), as well as neurological diseases (OR = 2.20, P = 0.03). The most common independent HAP predictors were Pseudomonas aeruginosa (Hazard Ratio [HR] = 2.27, P < 0.001), Klebsiella pneumoniae (HR = 1.81, P = 0.003), tracheostomy (HR = 1.72, P = 0.04), and APACHE II ≥17 (HR = 1.54, P = 0.04). Conclusion: High incidence rate of HAP was linked with P. aeruginosa, K. pneumoniae, tracheostomy, and APACHE II ≥17. Furthermore, a high mortality rate was strongly correlated with reintubation, duration in ICU ≥5 days, HAP outcome, DM, APACHE II ≥17, and neurological diseases.
| Abstract in French|| |
Contexte: Les données épidémiologiques sur la pneumonie acquise dans les hôpitaux (HAP) sont rares dans les unités de soins intensifs (ICUs). Objectif: ce L'étude vise à quantifier l'incidence du HAP, à déterminer les facteurs prédictifs du HAP, à calculer le ratio de risque de mortalité lié au HAP ainsi qu'à identifier les différents facteurs de risque contribuant à la mortalité. Sujets et méthodes: Une étude longitudinale prospective a été menée à les unités de soins intensifs générales d'un hôpital gouvernemental sur 12 mois. Nous avons inclus les patients adultes admis depuis au moins 72 h avant l'apparition des signes. Nous ont utilisé un modèle de régression logistique pour les résultats en matière de décès et un modèle de risque proportionnel de Cox pour les résultats HAP. Résultats: Sur 356 patients, 133 patients ont développé une Pneumonie Acquise sous Ventilation (VAP), 76 patients avec une NV-HAP, ainsi que 147 patients n'a pas acquis HAP. L'incidence du HAP était de 28 cas de HAP pour 1000 jours-personnes, ainsi que le taux de mortalité, de 74 pour 100 jours, alors que le pourcentage de risque attribuable (ARP) était de 85 %. Ce taux de mortalité élevé a été clarifié par des prédicteurs indépendants comme réintubation (odds ratio [OR] = 8,99, P < 0,001), durée de ICU ≥ 5 jours (OR = 7,29, P = 0,02), résultat HAP (OR = 6,49, P = 0,001), le diabète sucré (DM) (OR = 2,98, P = 0,004), APACHE II ≥17 (OR = 2,76, P = 0,004), ainsi que les maladies neurologiques (OR = 2,20, P = 0,03). Les prédicteurs indépendants de HAP les plus courants étaient Pseudomonas aeruginosa (Hazard Ratio [HR] = 2,27, P < 0,001), Klebsiella pneumoniae (HR = 1,81, P = 0,003), trachéotomie (HR = 1,72, P = 0,04) et APACHE II ≥ 17 (HR = 1,54, P = 0,04). Conclusion: Le taux d'incidence élevé de HAP était lié à P. aeruginosa, K. pneumoniae, trachéotomie et APACHE II ≥17. De plus, un taux de mortalité élevé était fortement corrélé à la réintubation, à la durée en ICU ≥ 5 jours, au résultat HAP, au DM, à l'APACHE II ≥17, et maladies neurologiques.
Mots-clés: Pneumonie nosocomiale, incidence, unités de soins intensifs, facteurs de risque, trachéotomie
Keywords: Hospital-acquired pneumonia, incidence, intensive care units, risk factors, tracheostomy
|How to cite this article:|
Yakoub M, Elkhwsky F, El Tayar A, El Sayed I. Hospital-acquired pneumonia pattern in the intensive care units of a governmental hospital: A prospective longitudinal study. Ann Afr Med 2023;22:94-100
|How to cite this URL:|
Yakoub M, Elkhwsky F, El Tayar A, El Sayed I. Hospital-acquired pneumonia pattern in the intensive care units of a governmental hospital: A prospective longitudinal study. Ann Afr Med [serial online] 2023 [cited 2023 Feb 1];22:94-100. Available from: https://www.annalsafrmed.org/text.asp?2023/22/1/94/368399
| Introduction|| |
Hospital-Acquired Pneumonia (HAP) refers to pneumonia that occurs 48 h or more following admission, excluding pneumonia that happens on admission., Instead, Ventilated-Acquired Pneumonia (VAP), is a subcategory of HAP that appears at least 2 days after endotracheal intubation. The incidence of HAP in intensive care unit (ICU) ranges from 6.24% to 42.7% with differences relating to different ICUs and mortality ranges from 31.4% to 82%.,
Detailed information about HAP pattern among patients is crucial to achieve effective prevention and enhance ICU quality of care. Hence, our specific aims were to quantify the HAP incidence, as well as ascertain HAP and mortality predictors.
| Subjects and Methods|| |
It was a 12-month prospective longitudinal study at a governmental hospital's general ICUs. Approval to carry out this research was obtained from the hospital. The minimum required sample size, using PASS 2002 (Power Analysis and Sample Size Software) (NCSS, LLC. Kaysville, Utah, USA), was 147 patients with HAP, according to the previous literature's 35% frequency of nosocomial pneumonia, and a margin of error of 5%. Of 1023 individuals admitted to the ICU, we collected 209 patients, who acquired HAP, and 147 patients without HAP in the ICUs. Over 1 year (July 2018–June 2019), we recruited all the patients admitted to our hospital in the Beheira governorate. All information from patients' records were kept strictly restricted.
Inclusion criteria to our study were adult individuals (males as well as females) developing pneumonia after at least 72 h to guarantee infection was gained following ICU admission. We disqualified patients with proven pneumonia at the time of admission or during the first 2 days following ICU admission. Patients were monitored daily till they gained HAP, died, or were discharged from the ICUs. For all suspected HAP patients, sputum and blood cultures were collected with aseptic techniques and sent to the microbiology laboratory.
We gathered information from patients' files such as detailed demographics, comorbidities, diagnosis, APACHE II score, Charlson Comorbidity Index, time to HAP occurrence, the onset of fatality, ICU duration, as well as microbiological laboratory reports. These data were collected anonymously. Non-Ventilated HAP (NV-HAP) and Ventilated Acquired Pneumonia (VAP) were clinically defined according to the centers for Disease Control and Prevention/National Healthcare Safety Network.
The normality of quantitative data distribution was examined by the Kolmogorov–Smirnov test. These quantitative data were represented in the form of mean ± standard deviation, median, and interquartile range. P < 0.05 was contemplated as significant results, and we stated the 95% Confidence Interval (CI) of that fraction. Comparisons for nonparametric variables were made using the Man–Whitney (U) or Kruskal–Wallis (H) test, whereas for parametric quantitative variables, we applied one way-analysis of variance (ANOVA). As well, pairwise comparisons were performed using Tukey's test or Dunnett's test to assess significant results from ANOVA or Kruskal–Wallis (H) test, respectively. To investigate the relationship between qualitative variables, the Chi-square test was used as well as Fisher's exact test was reported when small counts were analyzed (P value of Chi-square is not valid).
Variables that are statistically significant and clinically relevant were included in our multivariate analysis to assess the independent contributions of each significant predictor throughout the outcome of interest. To identify the independent factors predicting the mortality, we applied a multivariate logistic regression model. We addressed Odds Ratio (OR) with 95% CI. We also utilized a multivariate cox regression model to inform HAP's independent predictors, as well as we reported the Hazard Ratio (HR) with 95% CI. To assess the impact of different categorical predictors on time till mortality, we used Log-Rank test with Kaplan–Meier curve for bivariate analysis, and statistically significant predictors from the bivariate analysis were inserted into a multivariate cox regression. SPSS software for Windows, version 23.0, was used (Armonk, NY: IBM Corp.USA).
| Results|| |
Over the studied year (2018–2019), 1023 patients were admitted to our hospital's ICUs. Six hundred and sixty-seven individuals were ineligible for our prospective study. In total, 356 individuals participated in our study. Of 356 individuals, we found that 37.4% of individuals developed Ventilated-Acquired Pneumonia (VAP) (n = 133), 21.3% Non-Ventilated HAP (NV-HAP) (n = 76), while 41.3% individuals without HAP (n = 147). Consequently, HAP's incidence rate is 28 cases of HAP per 1000 person-days (95% CI 24–32 per 1000 person-days).
The study population's characteristics (356 individuals) are presented in [Table 1]. The HAP individuals had an average age of 56 ± 19 years, with 113 (54.1%) males, and 96 (45.9%) females. As well, HAP individuals spent an average of 31.60 ± 26.18 days in the ICUs. Early-onset HAP was found among 30% of individuals (n = 63), while late-onset HAP was found in 70% of individuals (n = 146). The most common reason for admission was neurological disorders (37%). Furthermore, HTN is the major comorbidity in 53% of HAP patients.
|Table 1: Discrepancy in patient characteristics among the studied groups|
Click here to view
In the current analysis, we revealed that 54 (25.8%) individuals who developed HAP were alive, and 155 (74.2%) individuals with HAP were dead, whereas there were 16 (10.9%) dead as well as 131 (89.1%) alive in the control group (n = 147). In terms of mortality outcome, there was a highly statistically significant difference between HAP individuals and the control (P < 0.001). Therefore, we are 95% confident that when comparing the HAP patients compared to non-HAP patients, the mortality risk ratio ranges between 6.34 and 7.28 (Mortality Risk Ratio = 6.81, 95% CI 6.34–7.28).
HAP was responsible for nearly between 74% and 96% of deaths among HAP individuals in the ICUs (ARP = 85%, 95% C. I. 74%–96%). Hence, the overall risk of mortality enacted by HAP in the general population is approximately 37% (PAR = 37%, 95% C. I. 31.04%–40.96%).
We observed that all HAP individuals were given acid-suppressive medications, whereas 138 (94%) of the control group received acid-suppressive medications without developing HAP (39.63a, P < 0.001). Furthermore, when compared to the control individuals, 61% of HAP patients (n = 127) did not use nasogastric feeding (NGF), while 39% of HAP individuals (n = 82) used NGF (116.24a, P < 0.001).
Predictors of mortality
On the multivariate logistic regression [Table 2], the independent risk factors associated with mortality were HAP individuals (OR = 6.49, CI 2.21–19.02, P = 0.001), neurological disorders (OR = 2.20, CI 1.07–4.52, P = 0.03), ICU duration ≥5 days (OR = 7.29, P = 0.02), DM as comorbidity (OR = 2.98, CI 1.43–6.23, P = 0.004), APACHE II ≥17 (OR = 2.76, CI 1.38–5.55, P = 0.004), and reintubation (OR = 8.99, CI 2.80–28.92, P < 0.001).
Predictors of hospital-acquired pneumonia
Using multivariate cox regression model [Table 3]a and [Table 3]b, various factors were analyzed. We found that APACHE II (≥17) (HR = 1.54, CI 1.03–2.31, P = 0.04), tracheostomy (HR = 1.72, CI 1.03–2.85, P = 0.04), Klebsiella pneumoniae (HR = 1.81, CI 1.22–2.69, P = 0.003), as well as Pseudomonas aeruginosa (HR = 2.27, CI 1.45–3.57, P < 0.001) were the significant predictors of HAP [Figure 1].
|Figure 1: A chart of Hazard Ratio values of the effect of the most independent predictors correlated with hospital-acquired pneumonia|
Click here to view
| Discussion|| |
Epidemiology of hospital-acquired pneumonia
Our HAP incidence is 28 cases of HAP per 1000 person-days. This rate is comparable to a study conducted in China, which computed the HAP rate was 39.2 per 1000 patient-days in trauma ICUs. In the contrary, a domestic surveillance study that measured the healthcare-associated infections' rate in Egypt, they realized that HAP's rate was 5.6 per 1000 patient days. Due to this variation, comparison is difficult. Discrepancy in the incidence may be due to different subsets of patients, management policy of various ICUs, and the number of patients in each study. Similarly, a heavy workload and deficient nursing staff in our hospital may boost the healthcare-associated infections.
According to our study, the overall HAP fatality rate is 74%. Nonetheless, the HAP fatality rate in the analyses performed by Costa, et al.(18.3%), Ranjan et al.(48.33%), and Schwebel C, et al.,, (82%) was fluctuating. Tracheostomy, intubation, MDR, and prolonged length of stay are likely the reasons for these fluctuations. Furthermore, published data, on considering the HAP attributable mortality risk as well as HAP population-attributable risk percentage in an individual Egyptian adult ICU setting, were scarce.
We realized that neurological disorders are among the predictive factors of ICU fatality. A research was performed by Yuan et al. to investigate the characteristics of fatality among severe stroke patients, analyze their causes of death. According to the findings of this study, HAP was the commonest cause of death in stroke patients. Another study was carried out to determine the risk factors for fatality in patients with Gallian–Barre syndrome. Mechanical ventilation was identified as one of the highly risk factors of fatality (P < 0.0001).
Another independent factor is the duration in ICU ≥5 days. Moitra et al. conducted a similar study on the relationship between ICU length of stay and fatality in elderly patients, which found that longer ICU duration is associated with higher fatality for HAP patients. Although Kumar et al. found that HAP patients with ICU duration ≥5 days had no statistically significant relationship with fatality.
Diabetes mellitus (DM) is another independent factor. A prospective cohort study of VAP conducted to determine risk factors for mortality. It addressed that DM and Ventilated-Acquired Pneumonia (VAP) were independent risk factors of in-hospital mortality. Nevertheless, the ICU duration, tracheostomy, reintubation, and mechanical ventilation were not addressed in this study.
Reintubation was also an independent predictor for in-hospital fatality in multivariate analysis. Gao et al. conducted a systematic review and meta-analysis study to recap the risks of reintubation to VAP and mortality. They revealed that reintubation could endanger survival as well as enhance the risk of VAP. Another study was done by Frutos-Vivar et al. to evaluate the outcome of patients who require reintubation after elective extubation. It noted that in a large cohort of scheduled extubated patients, one-third of patients experienced extubation failure, with half of them needed reintubation. Furthermore, reintubation was linked to enhanced fatality because of the new development of complications following reintubation.
Hospital-acquired pneumonia outcome
One more independent HAP predictor was APACHE II ≥17. In a related research, Kumar et al. noticed that APACHE II score ≥17 as well as septic shock were independent predictors of fatality. On the other hand, we observed that septic shock was not statistically significant in either the univariate or multivariate analysis. This is due to the small number of patients with septic shock who developed HAP.
Tracheostomy is another independent predictive factor of HAP occurrence. Othman et al. agreed with our findings. In fact, they stated that the HAP's independent predictors were tracheostomy and reintubation. However, reintubation (HR = 1.52, P = 0.02) was statistically significant on univariate analysis in our study. Chaari et al. showed another analysis was to assess VAP's predictive factors following tracheostomy in trauma patients. In addition, this analysis revealed that delayed tracheostomy is an independent factor of VAP occurrence. Controversial to these studies, Nseir et al. noticed that tracheostomy is independently linked with decreased risk for ventilated-acquired pneumonia. Although tracheostomy might shield against VAP because of the reduction of long-term epithelial injury unlike endotracheal intubation, it might enhance the risk of VAP due to direct injury as well as bacterial insertion to the airways.
Our analysis had a larger sample size of 356 patients. As well, the selection of patients was done with regards to strict selection criteria. However, this study was a mono-center-based design. We did not test for viruses. Another limitation is that neither Bronchoalveolar Lavage nor lung biopsy was done to confirm our findings. One more restriction is that we were unable to fully comprehend all risk factors relevant to HAP.
| Conclusion|| |
High incidence rate of HAP was linked with P. aeruginosa, K. pneumoniae, tracheostomy, and APACHE II ≥17. Furthermore, high mortality rate was strongly correlated with reintubation, duration in ICU ≥5 days, HAP outcome, DM, APACHE II ≥17, and neurological diseases. More research should be conducted to reassess the impact of HAP in nongovernmental ICU settings.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Ekren PK, Ranzani OT, Ceccato A, Li Bassi G, Munoz Conejero E, Ferrer M, et al
. Evaluation of the 2016 Infectious Diseases Society of America/American Thoracic Society guideline criteria for risk of multi-drug resistant pathogens in hospital- acquired and ventilator-associated pneumonia patients in the Intensive Care Unit pervin. Am J Respir Crit Care Med 2018;197:826-30.
Ferrer M, Torres A. Epidemiology of ICU-acquired pneumonia. Curr Opin Crit Care 2018;24:325-31.
Leone M, Bouadma L, Bouhemad B, Brissaud O, Dauger S, Gibot S, et al.
Hospital-acquired pneumonia in ICU. Anaesth Crit Care Pain Med 2018;37:83-98.
Herkel T, Uvizl R, Doubravska L, Adamus M, Gabrhelik T, Htoutou Sedlakova M, et al.
Epidemiology of hospital-acquired pneumonia: Results of a Central European multicenter, prospective, observational study compared with data from the European region. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 2016;160:448-55.
Schwebel C, Reignier J, Kallel H, Ruckly S, de Montmollin E, Mourvillier B, et al
. A comparison of the mortality risk associated with ventilator-acquired bacterial pneumonia and nonventilator ICU-acquired bacterial pneumonia. Crit Care Med 2018;1:346.
Bhadade R, Harde M, deSouza R, More A, Bharmal R. Emerging trends of nosocomial pneumonia in Intensive Care Unit of a tertiary care public teaching hospital in Western India. Ann Afr Med 2017;16:107-13.
] [Full text]
Moolchandani K, Sastry AS, Deepashree R, Sistla S, Harish BN, Mandal J. Antimicrobial resistance surveillance among Intensive Care Units of a tertiary care hospital in South India. J Clin Diagn Res 2017;11:DC01-7.
Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008;36:309-32.
Tao L, Hu B, Rosenthal VD, Gao X, He L. Device-associated infection rates in 398 Intensive Care Units in Shanghai, China: International Nosocomial Infection Control Consortium (INICC) findings. Int J Infect Dis 2011;15:e774-80.
Talaat M, El-Shokry M, El-Kholy J, Ismail G, Kotb S, Hafez S, et al.
National surveillance of health care-associated infections in Egypt: Developing a sustainable program in a resource-limited country. Am J Infect Control 2016;44:1296-301.
Costa RD, Baptista JP, Freitas R, Martins PJ. Hospital-acquired pneumonia in a multipurpose Intensive Care Unit : One-year prospective study. Acta Med Port 2019;32:746-53.
Ranjan N, Chaudhary U, Chaudhry D, Ranjan KP. Ventilator-associated pneumonia in a tertiary care Intensive Care Unit: Analysis of incidence, risk factors and mortality. Indian J Crit Care Med 2014;18:200-4.
] [Full text]
Yuan MZ, Li F, Fang Q, Wang W, Peng JJ, Qin DY, et al.
Research on the cause of death for severe stroke patients. J Clin Nurs 2018;27:450-60.
van den Berg B, Bunschoten C, van Doorn PA, Jacobs BC. Mortality in Guillain-Barré syndrome. Neurology 2013;80:1650-4.
Moitra VK, Guerra C, Linde-Zwirble WT, Wunsch H. Relationship between ICU length of stay and long-term mortality for elderly ICU survivors. Crit Care Med 2016;44:655-62.
Kumar S, Jan RA, Fomda BA, Rasool R, Koul P, Shah S, et al.
Healthcare-associated pneumonia and hospital-acquired pneumonia: Bacterial aetiology, antibiotic resistance and treatment outcomes: A study from North India. Lung 2018;196:469-79.
Tamayo E, Álvarez FJ, Martínez-Rafael B, Bustamante J, Bermejo-Martin JF, Fierro I, et al.
Ventilator-associated pneumonia is an important risk factor for mortality after major cardiac surgery. J Crit Care 2012;27:18-25.
Gao F, Yang LH, He HR, Ma XC, Lu J, Zhai YJ, et al.
The effect of reintubation on ventilator-associated pneumonia and mortality among mechanically ventilated patients with intubation: A systematic review and meta-analysis. Heart Lung 2016;45:363-71.
Frutos-Vivar F, Esteban A, Apezteguia C, González M, Arabi Y, Restrepo MI, et al.
Outcome of reintubated patients after scheduled extubation. J Crit Care 2011;26:502-9.
Othman HA, Gamil NM, Elgazzar AE, Fouad TA. Ventilator associated pneumonia, incidence and risk factors in emergency Intensive Care Unit Zagazig university hospitals. Egypt J Chest Dis Tuberc 2017;66:703-8.
Chaari A, Kssibi H, Zribi W, Medhioub F, Chelly H, Algia NB, et al.
Ventilator-associated pneumonia in trauma patients with open tracheotomy: Predictive factors and prognosis impact. J Emerg Trauma Shock 2013;6:246-51.
] [Full text]
Nseir S, Di Pompeo C, Jozefowicz E, Cavestri B, Brisson H, Nyunga M, et al
. Relationship between tracheotomy and ventilator-associated pneumonia: A case control study. Eur Respir J 2007;30:314-20.
Veelo D, Binnekade J, Schultz M. Tracheostomy – Causative or preventive for ventilator-associated pneumonia? Curr Respir Med Rev 2010;6:52-7.
[Table 1], [Table 2], [Table 3]