Keywords
traumatic brain injury - nosocomial infections - infections
Introduction
Traumatic brain injury (TBI) is one of the leading causes of morbidity and mortality
all over the world. Approximately 50 million cases of TBI are recorded annually all
over the world and it contributes to approximately 30% of all injury-related deaths.[1]
[2]
[3]
[4]
Most patients with moderate-to-severe TBI require hospitalization, and mortality of
such patients may be as high as 40%.[5] For inpatient TBI-related deaths, multiple risk factors for mortality have been
identified, including age, presence of existing comorbidities, hypotension, injury
severity, pupillary abnormalities on presentation, and abnormal imaging findings.[6]
Infections are more common in severe TBI and are an independent risk factor for in-hospital
death and prolonged length of stay. Increased risk of infections in these patients
is due to the increasingly recognized phenomenon of peripheral immunosuppression because
of alterations in the innate and adaptive immune systems.[7]
[8] Infections commonly occur in the acute phase after TBI; however, they may even affect
patients during rehabilitation and chronic recovery phase and are associated with
poorer outcomes overall.[9]
Despite significant impact on the overall outcome, infections in TBI patients are
largely understudied and underreported. The purpose of this study was to study the
profile of infectious complications in patients with TBI and impact of these infections
on the hospital outcome of these patients.
Materials and Methods
It was a hospital-based observational retrospective study jointly conducted by the
Infectious Diseases Division of the Department of Internal Medicine and Neurosurgery
Department of Sher-i-Kashmir Institute of Medical Sciences (SKIMS) Srinagar, Kashmir,
a leading tertiary care institute in north India.
The medical records of the all the patients with a diagnosis of TBI admitted via emergency
department of the hospital from January 2017 to January 2020 were reviewed. These
medical records were screened for presence of any predefined infection (defined by
latest CDC criteria[10]) acquired at any time during their hospital stay. Patients who had evidence of any
nosocomial infection were included in the study.
For each included patient, the following data was collected and entered on a standard
form: age, gender, days of hospital stay, mode of trauma, type of injury, Glasgow
coma scale (GCS) score at the time of presentation, comorbidities, focus of infection,
results of cultures, antibiotic sensitivity of isolated organism, results of common
laboratory tests, and hospital outcome in terms of all-cause mortality.
The recorded data was compiled and entered in a spreadsheet (Microsoft Excel) and
then exported to data editor of SPSS Version 20.0 (SPSS Inc., Chicago, Illinois, USA).
Continuous variables were expressed as mean ± SD and categorical variables were summarized
as frequencies and percentages.
Results
A total of 60 patients were included in 3-year study period. The mean age of patients
was 36.6 (± SD 12.3) years. As much as 75% of the patients were males and 25% were
females. Mean hospital stay of the patients was 10.3 (± SD 5.9) days. Forty-three
patients had road traffic accident (RTA) as the cause of TBI, while 17 patients sustained
injuries because of a fall.
Subdural hemorrhage was the predominant type of injury in these patients, followed
by subarachnoid hemorrhage (SAH), as shown in [Table 1].
Table 1
Type of injury in study population
|
Type of injury
|
Frequency
|
Percentage
|
|
Subdural hematoma
|
24
|
40.0
|
|
Subarachnoid hemorrhage
|
14
|
23.3
|
|
Brain contusions
|
11
|
18.3
|
|
Extradural hematoma
|
8
|
13.3
|
|
Intraparenchymal hemorrhage
|
3
|
5.0
|
|
Intraventricular hemorrhage
|
2
|
3.3
|
|
Diffuse axonal injury
|
1
|
1.6
|
The average GCS score of study patients was 9.6. Nine patients had a GCS of less than
7, 31 patients had GCS from 8 to 10, and 20 patients had GCS score of 11 or above.
The respiratory tract was the predominant focus of infection in these patients, followed
by urinary tract, as is shown in [Table 2].
Table 2
Focus of infection in study patients
|
Focus
|
Frequency
|
Percentage
|
|
Respiratory tract
|
49
|
81.6
|
|
Urinary tract
|
13
|
21.6
|
|
Skin and soft tissue
|
09
|
15.0
|
|
Primary blood stream
|
08
|
13.3
|
|
others
|
03
|
5.0
|
A total of 28 (46.6%) patients had a positive blood culture and 13 (21.6%) patients
had positive urine cultures. [Tables 3] and [4] show the microbiology of positive blood and urine cultures.
Table 3
Microbiology of positive blood cultures
|
Organism
|
Frequency
|
Percentage (of total patients)
|
|
Burkholderia cepecia
|
11
|
18.3
|
|
Acinetobacter baumannii
|
7
|
11.6
|
|
Klebsiella pneumoniae
|
4
|
6.6
|
|
Staphylococcus aureus
|
3
|
5.0
|
|
Enterococcus sps
|
2
|
3.3
|
|
Escherichia coli
|
1
|
1.6
|
Table 4
Microbiology of positive urine cultures
|
Organism
|
Frequency
|
Percentage (of total patients)
|
|
Escherichia coli
|
6
|
10.0
|
|
Klebsiella pneumoniae
|
3
|
5.0
|
|
Enterococcus sps
|
3
|
5.0
|
|
Acinetobacter baumannii
|
1
|
1.6
|
A higher rate of antimicrobial resistance among different Gram-positive organisms
was found in our study. [Table 5] shows the sensitivity of different bacterial isolates to commonly used antimicrobials.
Table 5
Sensitivity of different microorganisms to commonly used antibiotics
|
Organism
|
No. of isolates
|
Percentage of organisms sensitive to particular antibiotic
|
|
Pipercillin/ tazobactam (%)
|
Levofloxacin (%)
|
Ceftriaxone (%)
|
Meropenem
|
Vancomycin (%)
|
Linizolid (%)
|
Amikacin (%)
|
Polymyxin B (%)
|
|
Burkholderia cepecia
|
11
|
45.4
|
27.2
|
18.1
|
90.9
|
–
|
–
|
63.6
|
100.0
|
|
Acinetobacter baumannii
|
8
|
25
|
12.5
|
12.5
|
50.0
|
–
|
–
|
37.
|
100.0
|
|
Klebsiella pneumoniae
|
7
|
28.5
|
14.2
|
0.0
|
28.5
|
–
|
–
|
57.1
|
100.0
|
|
Staphylococcus aureus
|
4
|
0.0
|
0.0
|
0.0
|
25.0
|
100.0
|
100.0
|
–
|
–
|
|
Enterococcus sps
|
5
|
40.0
|
0.0
|
0.0
|
20.0
|
80.0
|
100.0
|
–
|
–
|
|
Escherichia coli
|
7
|
28.4
|
14.2
|
14.2
|
71.4
|
–
|
–
|
85.7
|
100.0
|
Fourteen patients out of the total study cohort died during the hospital stay, thus
constituting an all-cause hospital mortality of 23.3% in these patients.
Discussion
This study has provided important glimpses about the different types of infections
seen early in the hospital course of patients admitted with TBI. Most of these patients
are young and tend to have a prolonged hospital stay. Our study demonstrated that
majority of patients who develop hospital-acquired infection early in their course
of illness had a low GCS score, thereby reaffirming the fact that more severe TBI
is associated with more chances of developing nosocomial infections. We do believe
that for prevention of infection, natural immunity as well as barrier immunity has
utmost importance. The lower the GCS score at admission of a traumatic patient, more
vulnerability to infections.
Pulmonary infections were most found to be most common, comprising more than 80 percent
of total cases. Previous works by Helling et al[11] and Kourbeti et al[12]
[13] also found that pulmonary infections were most common in patients with head trauma.
Multiple factors are responsible for development of respiratory tract infections in
these patients, which include colonization of lower respiratory tract with microorganisms,
continuous microaspirations and subsequent overwhelming of lung defense mechanisms,
and frequent use of endotracheal intubation and mechanical ventilation in these patients.[14]
[15] In addition, it has been studied earlier that dampening of cough reflex, as is expected
in low GCS patients, constitutes the single most important factor in breaching of
host defenses for development of infections in traumatic brain patients. A less emphasized
fact in such patients is reflux of gastrointestinal (GI) sections into respiratory
tract makes a nidus for percolation of infections. An added factor being dampening
of secretions due to impaired swallowing in low GCS patients changes the ecobiology
of oropharynx and thereby predisposes such patients to infections.[15]
Majority of the patients with moderate to severe TBI require catheterization and are
at risk of developing urinary tract infections. Our study demonstrated that urinary
tract infections were the second most common nosocomial infections, which is in concordance
with previously published literature in trauma patients.[16] TBI patients often have associated peripheral injuries, and these open wounds provide
additional access routes of infection in these patients. Our study showed that skin
and soft-tissue infections were common in these patients and comprised 15% of total
infections. Many of these patients develop retention of urine due to neurologic dysfunction,
and retained urine becomes a source of infection in some percentage of patients.
This study has also provided important insights into local trends and microbiological
agents of infection in these patients. In our study, Burkholderia cepecia and Acinetobacter
baumannii were predominant isolates from blood; this is in contrast with a previously
published study on such patients wherein coagulase-negative Staphylococcus and S.
aureus were more frequent.[17]
The blood stream infections are most of the times neglected in such cohort of patients,
but it should be meticulously sought. The present study provides us an insight that
Gram-negative organisms predominate blood infections in such patients at our center
as against Gram-positive infections seen in such patients in Western countries.
Furthermore, the antibiotic sensitivity profile of these organisms demonstrates that
they are less susceptible to commonly used antibiotics, which is an alarming trend.
This changing trend of organisms and their resistance pattern may be attributed to
the ubiquitous nature of these in hospital settings and inadequate infection control
practices in this part of the world. Rational use of antibiotics and formulation of
antibiotic policy, based on changing trend of identified isolates, is urgently needed
for this group of patients, so as to improve their overall outcome.
Infections are associated with increased mortality, financial liability, and poor
outcome in patients admitted with traumatic injuries. Prevention of many hospital-acquired
conditions by implementing interventions aiming to decrease the incidence of hospital-acquired
infections may have a better outcome in traumatic brain patients. Further, larger
studies need to be done in order to find the mitigation factors, their application,
and resultant outcome of application of such practices for better survival of traumatic
patients.
Conclusions
Patients with TBI are more susceptible to develop hospital- acquired infection; more
severe TBI, as demonstrated by low GCS score, poses more risk of infection. Respiratory
tract and urinary tract are the most common sites of infections in these patients
and Gram-negative antibiotic resistant organisms are more frequently involved in this
part of the world. Prompt recognition and effective antimicrobial therapy are key
to improve hospital outcome in these patients.
Limitations
Small sample size and retrospective nature were major limitations in our study. Lack
of a control group of patients with TBI without infection, missing data regarding
exact timing of acquisition of infection during the hospital stay, and lack of follow-up
of these patients were some other limitations.
