CC BY
9КРЫМСКИЙ ЖУРНАЛ ЭКСПЕРИМЕНТАЛЬНОЙ И КЛИНИЧЕСКОЙ МЕДИЦИНЫ
UDC 616.24-002:616-022:616-036
EPIDEMIOLOGY OF HOSPITAL-ACQUIRED PNEUMONIA AND VENTILATOR-ASSOCIATED PNEUMONIA IN NEUROSURGICAL PATIENTS
А. A. Birkun
Department of Medicine of Emergency Conditions and Anesthesiology Medical Academy named after S. I. Georgievsky of V. I. Vernadsky Crimean Federal University Correspondence: 5/7, Lenin Avenue, Simferopol, 295006, Republic of Crimea, Russian Federation E-mail: birkunalexei@gmail.com
SUMMARY
Background. Hospital-acquired pneumonia (HAP) and ventilator-associated pneumonia (VAP) are recognized as important causes of morbidity and mortality in neurosurgical patients. Successful management of HAP and VAP should rely upon detailed knowledge of specific epidemiological pattern in a given medical facility. The study was aimed to evaluate epidemiological features of HAP in one of the leading neurosurgical centers in the southern Ukraine. Materials and methods. A retrospective (2006-2011) analysis of 4,528 medical records including 115 cases complicated with HAP was performed. Results. HAP was diagnosed in 2.5% (n = 115) of neurosurgical patients on record. Fatal outcome and recovery were registered for 41.7% and 37.4% of HAP patients, respectively. Most commonly isolated pathogen, Pseudomonas aeruginosa (26.4%), exhibited increased resistance to antipseudomonal drugs, namely ciprofloxacin (98%), ceftazidime (97%), gentamicin (96%), cefepime (95%) and amikacin (92%). Conclusions. Overall, increased mortality and high incidence of HAP, low bacteriologic coverage of HAP cases and excessive antimicrobial resistance of Pseudomonas were revealed. The results should be utilized to remove shortcomings in local practice of HAP treatment and microbiologic testing in the neurosurgical center.
ЭПИДЕМИОЛОГИЯ ГОСПИТАЛЬНОЙ И ВЕНТИЛЯТОР-АССОЦИИРОВАННОЙ ПНЕВМОНИИ У ПАЦИЕНТОВ НЕЙРОХИРУРГИЧЕСКОГО ПРОФИЛЯ
А. А. Биркун
РЕЗЮМЕ
Вступление. Госпитальная пневмония (ГП) и вентилятор-ассоциированная пневмония (ВАП) считаются важными причинами повышенной заболеваемости и смертности пациентов нейрохирургического профиля. Успешное лечение ГП и ВАП должно базироваться на углубленных знаниях особой эпидемиологической обстановки в конкретном лечебном учреждении. Цель исследования состояла в оценке эпидемиологических свойств ГП в одном из ведущих нейрохирургических центров юга Украины. Материалы и методы. Был выполнен ретроспективный (2006-2011) анализ 4528 историй болезни, включая 115 случаев, осложненных ГП. Результаты. ГП была диагностирована у 2,5% (n = 115) пациентов нейрохирургического профиля. Летальный исход и выздоровление наблюдались у 41,7% и 37,4% пациентов с ГП, соответственно. Наиболее распространенный возбудитель, Pseudomonas aeruginosa (26,4%), характеризовался повышенной резистентностью к антисинегнойным препаратам, а именно к ципрофлоксацину (98%), цефтазидиму (97%), гентамицину (96%), цефепиму (95%) и амикацину (92%). Выводы. В целом, были выявлены повышенный уровень смертности и высокая распространенность ГП, низкий бактериологический охват случаев ГП, а также чрезмерная резистентность Pseudomonas к антибиотикам. Полученные результаты следует использовать для устранения недостатков в локальной практике лечения и микробиологической диагностики ГП в нейрохирургическом центре.
Key words: pneumonia, multi-drug resistance, neurosurgery, antibacterial, Pseudomonas
Hospital-acquired pneumonia (HAP) consistently holds one of the leading positions among the causes of in-hospital morbidity and mortality all over the world. Showing incidence of 5-20 cases per 1,000 admissions, HAP is a second most frequent nosocomial infection with overall mortality rates of 27 to 51% [2,3,6]. Endotracheal intubation and mechanical ventilation significantly increase a risk of HAP, and ventilator-associated pneumonia (VAP) represents the most frequent ICU-acquired infection associated with poor clinical outcomes [1,2].
Pulmonary complications, including HAP and VAP, are known to be important causes of morbidity and mortality in neurosurgical patients [4,7]. Whereas it is generally recognized that effective prevention and treatment of HAP should be based on detailed knowledge of specific epidemiological pattern in a given medical facility [5], a prevalence and characteristic features of HAP and VAP in specific neurosurgical clinics are not extensively studied.
The aim of the study was to evaluate local epidemiological profile of HAP in one of the leading
neurosurgical centers in the southern Ukraine, including HAP incidence, outcomes, range of causative agents and their susceptibility to antibiotics.
MATERIALS AND METHODS 2.1. SOURCE DATA AND DEFINITIONS
Medical records (n=4,528) of patients who were withdrawn (discharged/deceased) from one of the major neurosurgical centers in the southern Ukraine from January 2006 to December 2011 were reviewed to identify cases of HAP based on the presence of respective pulmonologist's diagnostic confirmatory statement.
HAP case was defined by clinical and radiologic signs of pneumonia which occurred 48 hours and more after admission to the hospital, including presence of a new or progressive radiographic infiltrate and at least two of three clinical features: fever (core temperature >38.3°C), blood leukocytosis (>10,000 leucocytes/mm3) or leucopenia (<4,000/mm3), and purulent tracheal secretions. VAP corresponded to a pneumonia that developed 48 hours or more after endotracheal intubation. Pneumonia cases that developed <5 days after hospitalization were considered as early-onset. Multidrug resistance (MDR) was defined as non-susceptibility to at least one agent in three or more antimicrobial categories.
A total of 126 cases of HAP (including 99 cases of VAP) were identified in 115 patients.
Following data were extracted from the medical records and included in subsequent statistical analysis: age, sex, neurosurgical
diagnosis, comorbid conditions, duration of hospital stay, outcome, dates of HAP diagnosis and resolution, date of endotracheal intubation (tracheostomy), bacteriologic data, antimicrobial susceptibility, antimicrobials used, duration of antimicrobial therapy, risk factors for colonization and infection with MDR pathogens according to the ATS & IDSA Guidelines.4
2.2. STATISTICS
Most data are presented as a mean ± standard deviation. Data were compared using the un-paired Student's t-test and analysis of variance (ANOVA; univariate). A probability level of p<0.05 was considered to be statistically significant. Statistical analysis was performed using STATISTICA software version 10.0 (Stat Soft Inc., Tulsa, OK, USA).
RESULTS
HAP was diagnosed in 115 (2.5%) of 4,528 patients withdrawn from the center throughout the period of six years. Twenty five (22%) were female and 90 (78%) were male patients. The mean age of the patients with HAP was 47.0 ± 19.0 years (hereinafter: mean ± standard deviation) (range 16-91 years). Male patients with HAP were significantly younger than female ones (p<0.05) (Table 1). According to the results of univariate analysis (one-way ANOVA), duration of hospital stay and duration of pneumonia did not depend on age (F=1.12, p=0.344 and F=1.73, p=0.165, respectively). Predominant causes of neurosurgical center admission for patients with HAP were closed and open craniocerebral injuries (Figure 1).
Figure 1. Primary neurosurgical causes of admission.
Table 1.
Epidemiologic features of HAP with by-group distribution.
Characteristics Age (years) Duration of hospital stay, crude (days) Duration of hospital stay, survivors (days) Period from admission till initial HAP diagnosis (days) Period from intubation (tracheostomy) tiU HAP diagnosis (days) Duration of HAP, crude (days) Duration of HAP, cases of resolution (days) Duration of antimicrobial therapy prior to initial HAP diagnosis (days)
Gender Male 44.56±17.43 (90) 35.00±26.36 (90) 43.71±23.04 (45) 11.53+11.85 (90) 9.49±8.85 (68) 14.22±16.15 (98) 14.11±17.10 (44) n/s
Female 55.64±21.98 (25) 28.56±12.43 (25) 35.50±11.45 (6) 15.24±11.12 (25) 12.14±10.12 (21) 7.64±6.30 (28) 6.90±4.51 (10) n/s
Concomitant injury* Present n/s 32.10±28.17 (31) n/s 8.87±8.01 (31) 8.29±8.01 (24) 16.65±18.63 (34) n/s n/s
Absent n/s 34.15±22.62 (84) n/s 13.62±12.65 (84) 10.78±9.54 (65) 11.33±12.91 (92) n/s n/s
CCI CCI n/s 37.66±28.62 (47) n/s 9.83±7.57 (47) 8.08±6.60 (36) 17.20±19.07 (51) 18.45±21.85 (22) n/s
non-CCI n/s 30.79±20.22 (68) n/s 14.07±13.71 (68) 11.49±10.41 (53) 9.75±10.05 (75) 8.88±7.95 (32) n/s
OCI OCI n/s 28.06±19.08 (36) n/s 12.86±12.01 (36) 12.55±10.84 (29) 10.34±10.20 (38) n/s n/s
non-OCI n/s 36.13±25.82 (79) n/s 12.10±11.69 (79) 8.93±8.09 (60) 13.81±16.32 (88) n/s n/s
Early-onset/ late-onset pneumoniaf Early-onset 44.08±20.38 (25) 23.32±17.54 (25) 34.45±13.82 (11) n/s n/s 10.52±9.93 (25) 8.73±7.00 (11) n/s
Late-onset 47.08±18.38 (101) 39.10±25.69 (101) 47.47±24.58 (45) n/s n/s 13.32±15.75 (101) 13.81±17.23 (43) n/s
Outcome Death 50.11±19.58 (64) n/s n/s 13.53±13.36 (64) 10.49±9.95 (55) n/s 11.86±20.19 (22) 11.80±11.97 (64)
Survival 43.02±17.58 (51) n/s n/s 10.84±9.25 (51) 9.50±7.85 (34) n/s 13.41±12.16 (32) 10.59±9.47 (51)
Risk of MDR Present 47.28±18.88 (95) n/s 44.40±23.08 (42) n/s n/s 13.30±15.46 (106) n/s n/s
Absent 45.45±19.82 (20) n/s 35.00±15.35 (9) n/s n/s 9.90±10.40 (20) n/s n/s
MDR occurrence Present 46.87±18.09 (60) 41.35±22.62 (60) 49.25±22.24 (28) 15.00±13.79 (60) 12.42±10.32 (57) 12.73±12.37 (71) 11.88±12.18 (41) 14.02±12.14 (60)
Absent 44.31±22.56 (16) 42.44±33.01 (16) 49.38±24.22 (8) 13.00±11.02 (16) 9.10±5.43 (10) 24.19±26.35 (16) 22.50±35.67 (6) 12.56±12.05 (16)
to o
o\
H
a\ \
to
«
E
§
0 «
s »
N
1
>
W
«
o =
M
s
§
M
a
H
>
--
V
a
0
»
s «
1 s a
s £
M
o «
o
»
§
M Ö s
G S
a
s-
o>
Notes. All data are presented as a mean ± standard deviation. Data were compared using the un-paired Student's t-test. Results characterized by statistically significant difference are bolded. A probability level of p<0.05 was considered to be statistically significant. ^Concomitant injury represents cases of craniocerebral injury combined with spinal, thoracic, abdominal and/or skeletal trauma. tPneumonia cases that developed <5 days after hospitalization were considered as early-onset. CCI - closed craniocerebral injury; HAP - hospital-acquired pneumonia; MDR - multidrug resistance; n/s - not specified; OCI - open craniocerebral injury
The mean duration of hospital stay for patients with HAP who survived was 42.8 ± 22.1 patient-days. Univariate analysis showed a relation of longer hospital stay with greater duration of HAP (F=12.69, p<0.000), period from admission till initial HAP diagnosis (F=8.81, p<0.000), period from endotracheal intubation (tracheostomy) till pneumonia diagnosis (F=3.42, p=0.021) and duration of antimicrobial therapy prior to HAP diagnosis (F=10.05, p<0.000).
Early-onset HAP was diagnosed in 25 (22%) patients. Patients with late-onset HAP resided in the hospital for a longer time than patients with early-onset pneumonia (p<0.05), however there was no significant difference in the duration of HAP itself between early- and late-onset groups (p>0.05).
Whereas a crude mortality reached 55.7% (64 of 115 patients), a proportion of patients who died when pneumonia was ongoing amounted to 41.7% (48 of 115 patients). The patients who died were much older than those who survived (p<0.05).
Cases of pneumonia relapse were confirmed for 9 of 115 (8%) patients with HAP. Clinical resolution of pneumonia occurred in 43% (54 of 126) of HAP cases (37.4% of HAP patients). The
Epidemiologic feat
mean duration of the period from admission till initial HAP diagnosis was 12.3 ± 11.8 days. This period was shorter in patients with concomitant injury (craniocerebral injury combined with spinal, thoracic, abdominal and/or skeletal trauma) and closed craniocerebral injury (p<0.05), but not in patients with open craniocerebral injury (p>0.05).
The mean duration of pneumonia for cases of HAP with clinical resolution reached 12.8 ± 15.8 days. Pneumonia persisted significantly longer in males as compared with female patients (p<0.05). There was no relation of HAP duration with length of the period from admission till initial HAP diagnosis (F=0.65, p=0.582), as well as the period from endotracheal intubation till pneumonia diagnosis (F=0.25, p=0.863) or duration of antimicrobial therapy prior to HAP diagnosis (F=1.13, p=0.341).
In 89 of 115 patients (77%) HAP was presented by VAP. Overall incidence of VAP was 2.0%. For patients with VAP the mean duration of the period from endotracheal intubation till pneumonia diagnosis amounted 10.1 ± 9.2 days. Main epidemiologic features of VAP were generally comparable with those of HAP (Table 2).
Table 2
es of VAP and HAP.
Features for patients with pneumonia VAP patients (n = 89) HAP patients (n = 115)
Male patients 68 (76%) 90 (78%)
Age (years) 47.09 ± 18.94 46.97 ± 18.97
Age of male patients (years) 43.72 ± 17.54 44.56 ± 17.43
Age of female patients (years) 53.86 ± 21.76 55.64 ± 21.98
Duration of hospital stay for survivors (patient-days) 46.71 ± 21.53 42.75 ± 22.08
Early-onset pneumonia 21 (24%) 25 (22%)
Mortality of patients with pneumonia 47.2% 41.7%
Features for pneumonia cases VAP cases (n = 99) HAP cases (n = 126)
Clinical resolution of pneumonia 45 (45%) 54 (43%)
Mean duration of period from admission till initial pneumonia diagnosis (days) 12.53 ± 12.23 12.34 ± 11.75
Mean duration of pneumonia with clinical resolution (days) 13.51 ± 17.07 12.78 ± 15.77
Mean duration of period from endotracheal intubation (tracheostomy) till pneumonia diagnosis (days) 10.11 ± 9.17 n/a
Notes. Most data are presented as mean ± standard deviation; n/a - not applicable
2016, т. 6, №2
КРЫМСКИЙ ЖУРНАЛ ЭКСПЕРИМЕНТАЛЬНОЙ И КЛИНИЧЕСКОЙ МЕДИЦИНЫ
Bacteriologic testing has been performed in 76 of 115 patients with HAP (66%). Overall, 603 microbiologic assays were done, including 119 (20%) blood cultures and 484 (80%) sputum or tracheal aspirate cultures. The longer was duration of hospital stay, the higher was quantity of bacteriologic tests
Distribution of bacteric
(F=32.63, p<0.000). There was no microbial growth in 7% and 75% of tracheal and blood cultures, respectively. The most common pathogens isolated from tracheal aspirate and sputum samples were Pseudomonas aeruginosa (26%), Candida spp. (13%) and Klebsiella pneumoniae (10%) (Table 3).
Table 3
gic findings by species.
Isolated microorganisms Tracheal aspirate or sputum culture (n = 484) Blood culture (n = 119) MDR-positive cases (n = 60)
Pseudomonas aeruginosa 128 (26.4%) _ 25 (41.7%)
Candida spp. 65 (13.4%) _ _
Klebsiella pneumoniae 49 (10.1%) 1 (0.8%) 23 (38.3%)
Proteus mirabilis 44 (9.1%) _ 15 (25.0%)
Staphylococcus haemolyticus 43 (8.9%) 21 (17.6%) 12 (20.0%)
Staphylococcus aureus 33 (6.8%) 2 (1.7%) 14 (23.3%)
No growth 32 (6.6%) 89 (74.8%) _
Escherichia coli 21 (4.3%) 1 (0.8%) 12 (20.0%)
Streptococcus epidermidis 20 (4.1%) _ 8 (13.3%)
Citrobacter freundii 17 (3.5%) _ 7 (11.7%)
Proteus vulgaris 13 (2.7%) _ 7 (11.7%)
Enterobacter cloacae 6 (1.2%) _ _
Citrobacter diversus 5 (1.0%) _ 4 (6.7%)
Enterococcus spp. 3 (0.6%) 1 (0.8%) _
Streptococcus pneumoniae 2 (0.4%) 4 (3.4%) _
Providencia rettgeri 1 (0.2%) _ _
Enterobacteraerogenes 1 (0.2%) _ _
Morganella morganii 1 (0.2%) _ _
Note. The percentages are rounded to the nearest tenths.
Overall incidence of MDR-positive cases in patients with HAP was 52% (60 of 115 patients), whereas risk factors for MDR pathogens causing HAP were identified in 95 of 115 patients (83%). Seventy one (75%) and 24 (25%) patients had 2 and >3 risk factors, respectively.
MDR pathogens were identified in 58% (55 of 95) of patients from MDR risk category, including 50% (12 of 24) of patients with higher risk of MDR (>3 risk factors). Presence of MDR demonstrated no significant influence on the duration of hospital stay and duration of HAP itself (p>0.05).
Most commonly encountered MDR bacteria were P. aeruginosa and K. pneumoniae (41.7% and 38.3% of MDR-positive cases, respectively). P. aeruginosa demonstrated a high level of resistance to antipseudomonal
agents, including: ciprofloxacin - 98% (84 of 86 evaluations), ceftazidime - 97% (107 of 110), gentamicin - 96% (76 of 79), cefepime -95% (88 of 93) and amikacin - 92% (87 of 95).
Antimicrobial drugs were utilized prior to HAP diagnosis in all cases but three (98%), with a mean duration of 11.6 days. The duration of pre-HAP antibiotic use was comparable both for MDR-positive and MDR-negative cases (p>0.05). The most common antibiotics used before initial HAP diagnosis were ceftriaxone (mean duration of therapy = 5.6 days, n = 76), ciprofloxacin (5.9 days, n = 66), metronidazole (5.9 days, n = 56), amikacin (7.5 days, n = 42), furazidin (7.3 days, n = 28), ceftazidime (7.0 days, n = 25), cefoperazone (7.3 days, n = 23), levofloxacin (7.8 days, n = 23).
в
DISCUSSION AND CONCLUSIONS
The study revealed increased mortality and high incidence of both HAP and VAP in the given neurosurgical facility. Prolonged hospital length of stay was associated with increased duration of HAP and increased duration of antimicrobial therapy prior to HAP diagnosis. However, HAP duration itself was not influenced by terms of antibiotic use before pneumonia diagnosis. Presence of concomitant injury and closed craniocerebral trauma determined shorter period from admission till initial HAP diagnosis.
Furthermore, low bacteriologic coverage of HAP cases, high incidence of MDR-positive cases and excessive resistance of P. aeruginosa to the commonly used antipseudomonal agents were revealed.
This results should be utilized to remove a shortcomings in local practice of HAP treatment and microbiologic testing in the neurosurgical center. Also they may be helpful for improvement of relevant protocols of HAP and VAP management in other surgical facilities.
REFERENCES
1. American Thoracic Society and Infectious Diseases Society of America. Guidelines for the management of adults
with hospital-acquired, ventilator-associated, and health care-associated pneumonia. Am J Respir Crit Care Med 2005; 171:388-416.
2. Barbier F. Hospital-acquired pneumonia and ventilator-associated pneumonia: recent advances in epidemiology and management / Barbier F, Andremont A, Wolff M, Bouadma L. // Curr Opin Pulm Med 2013; 19(3):216-28.
3. Chawla R. Epidemiology, etiology, and diagnosis of hospital-acquired pneumonia and ventilator-associated pneumonia in Asian countries / Chawla R. // Am J Infect Control 2008; 36(4 Suppl):S93-100.
4. Guleria R. Pulmonary complications in neurosurgical patients / Guleria R and Madan K. // Indian J Neurosurg 2012; 1(2):175-80.
5. Niederman MS. Nosocomial and Ventilator-Associated Pneumonia. VAP: approach to therapy / Niederman MS. // Eur Respir Mon 2011; 53:36-47.
6. Rello J. Determinants of prescription and choice of empirical therapy for hospital-acquired and ventilator-associated pneumonia / Rello J, Ulldemolins M, Lisboa T et al. // Eur Respir J 2011; 37(6):1332-9.
7. Sogame LC. Incidence and risk factors for postoperative pulmonary complications in elective intracranial surgery / Sogame LC, Vidotto MC, Jardim JR, Faresin SM. // J Neurosurg 2008; 109(2):222-7
