Infection is common among hospitalized patients and associated with substantially increased health care costs and worsened outcomes. Nosocomial infections affect approximately 2 million patients in the United States each year, increasing overall patient care costs by an estimated $4.5 to $5.7 billion [1, 2]. Patients who develop an infection while in hospital are more likely to have longer hospital stays and adverse outcomes than those who do not [3–7]. Moreover, the development of a severe infection or sepsis has been associated with an increased risk of physical disability, permanent organ damage, cognitive impairment, and death [3, 8, 9].
Approximately 5% to 10% of patients admitted to an acute care hospital, and 9% to 37% of those admitted to an intensive care unit (ICU), will acquire one or more infections [1, 10]. Patients with neurological injury, such as stroke and traumatic brain injury (TBI), appear to be particularly susceptible to infection [6, 7, 11–14]. Although aspiration due to a decreased level of consciousness may explain the development of pneumonia among some patients, research also suggests that catecholamines released as a result of brain injury-induced sympathetic activation may modulate cells of the immune system and induce systemic immunosuppression [15–20]. While this immune suppression may protect the brain from further inflammatory damage, it may also increase susceptibility to infection among those with acquired brain injury [16, 17, 19].
Among patients with stroke, a systematic review and meta-analysis reported that the pooled incidence of infection was 30%, with pneumonia and urinary tract infections being most common . This study also reported that the incidence of infection for patients admitted to the ICU after stroke was significantly higher (45% versus 28%) . In comparison to stroke, patients with TBI have an even higher reported incidence of infection. It is estimated that approximately 50% of patients with severe TBI develop at least one infectious complication during hospitalization [12–14, 21]. Among those who develop infection, the most frequent location is the lung, with reported incidences of pneumonia ranging between 41% and 74% [12–14, 21]. Moreover, sepsis has been found to affect between 10% and 41% of patients with severe TBI during hospitalization [6, 13].
As patients with severe TBI (Glasgow Coma Score ≤8) have a significantly higher incidence of infection and sepsis compared to patients with mild or moderate TBI (Glasgow Coma Score >8), the risk of infection may correlate with severity of brain injury [6, 11, 12]. Among ICU patients, reported risk factors for infection include mechanical ventilation, presence of indwelling invasive devices, administration of immunosuppressive drugs, long-term or repeated use of antibiotics, and decreased host defenses due to poor chronic health status and/or acute disease processes . It was found that patients with TBI who spent more than 7 days in an ICU had an increased risk for infection as compared to those who were not admitted or spent ≤7 days in an ICU .
To date, there has been no formal attempt to systematically review the published literature describing the epidemiology of infection after TBI. Therefore, it remains unknown whether the high reported estimates of infection in this population are homogenous across the literature. Moreover, many of the relevant published studies to date have varied in the type and severity of TBI among included patients as well as in their design (randomized versus non-randomized), risk of bias, and setting (hospital ward versus ICU). Thus, their reported results vary considerably. Therefore, the purpose of this systematic review and meta-analysis is to estimate the cumulative incidence, incidence rate, point prevalence, and occurrence rate of overall infection, as well as pneumonia (community-acquired, hospital-acquired, ventilator-associated), urinary tract infection, central nervous system infection, bloodstream infection, sepsis, severe sepsis, and septic shock among patients hospitalized after TBI. We also aim to synthesize and compare the incidence, prevalence, and occurrence rate of infection observed across randomized versus non-randomized studies. We will identify study and population characteristics associated with infection, define risk groups based on TBI severity (mild, moderate, severe), and identify sources of heterogeneity across these studies. A better understanding of the risk of infection among patients with TBI could assist healthcare providers in identifying patient subgroups that may benefit from preventative or early treatment efforts and may provide evidence to support priority setting for the allocation of scarce healthcare resources and research funds.