Long-term life expectancy in severe traumatic brain injury: a systematic review

Abstract

INTRODUCTION

Traumatic brain injuries (TBIs) pose significant challenges to public health, medicine, and society due to their substantial impact on victims, caregivers, and the community. While indicators like life expectancy or death rates provide insights into mortality and long-term outcomes, they fail to address how TBIs affect aging, neurological sequelae, cognitive impairment, and psychological or psychiatric disorders. Moreover, most studies are limited to North America, limiting the generalizability of findings across different social welfare systems. As a result, clinicians face difficulties in providing optimal care and prognosis, hindering the improvement of life quality for victims and caregivers and efficient public health service planning. This study aims to address these limitations by examining life expectancy, mortality rates, and long-term outcomes in severely injured individuals.

EVIDENCE ACQUISITION

PubMed/Medline, Web of Science, Cochrane Library, Google Scholar, and PEDro search engines were systematically searched for studies investigating life expectancy and long-term outcomes in severe traumatic brain injuries. The final search date for all sources/databases was July 31, 2023. We conducted a systematic review, and only original research articles published in English were eligible for inclusion. After the screening process, data were extracted about life expectancy, follow-up, and conclusions.

EVIDENCE SYNTHESIS

This study analyzed 24 studies out of 343 identified. Life expectancy in the TBI population is lower than that of the general population. Older age and severity of functional impairments are major risk factors for mortality. Mortality rates are particularly high in the first two months. Mortality trends suggest a bimodal distribution, with a peak in the first five years followed by no further deaths until nine years after injury. The most influential factors include age, sex, trauma severity, independence in walking and feeding, time since injury, ventilator dependence, and cognitive and communication impairments. Respiratory and circulatory complications are among the leading causes of TBI-related deaths, followed by epilepsy, suicide, and respiratory infections.

CONCLUSIONS

Further research is required, considering the different long-term outcomes after TBI and their impact on families and society, to accurately estimate the life expectancy necessary for clinicians, caregivers, national health institutions, and medico-legal settlements.

Introduction

Traumatic brain injury (TBI) ranks among the leading causes of death,^1-4 especially among young individuals. In the US alone, estimates indicate TBI-related fatalities range from 15 to 30 per 100,000 people, depending on the source.^5-8 In the last decades, there has been a notable increase in head injuries among the elderly (people over 65) in developed nations due to population aging.^{9, 10} Globally, an estimated 50 million people suffer from TBI annually, resulting in severe disability or death.³

The causes of head trauma vary, with factors such as context and age influencing its etiology.¹¹ For instance, in the USA, from 1997 to 2007, TBI was primarily caused by gunshots (34.8%), road accidents (31.4%), and accidental falls.^{5, 12} In Europe, road accidents were the main cause of TBI, followed by falls.^{5, 12}

Survivors of severe TBI often need protracted hospitalization and extensive rehabilitation, facing physical, cognitive, and psychological challenges that impact their relationships and ability to work.¹³ The economic and social burden of TBI extends to families and society^{14, 15} – in 2006, the cost of managing TBI patients was estimated to be 60 million dollars per year¹⁶ – with survivors experiencing lower life expectancies¹⁷ and increased risks of comorbidities¹⁸ and neurodegenerative diseases.¹⁹

Recent scientific research characterizes TBI as a chronic health condition rather than an acute event,^20-22 emphasizing the critical role of quality care throughout emergency treatment, ICU stays, and rehabilitation.¹⁵ Social and family support significantly influence outcomes, yet economic and social disparities can hinder access to resources and treatment, exacerbating the overall impact.¹³

While interest in understanding TBI outcomes grows, available data about life expectancy remain limited, particularly regarding cognitive and psychological sequelae along with TBI survivors’ lifestyles and quality of life.^{22, 23} Most of the data investigating the life expectancy after a TBI originate from the USA. Additionally, longitudinal studies and metanalyses are scarce, hindering comprehensive analyses of influential factors and complicating comparisons across countries and outcomes.²⁴ And yet accurate estimates of long-term survival post-TBI are essential for patients, caregivers, and clinicians to make evidence-based decisions about assessing prognosis, developing public health services, and dealing with forensic implications effectively.²⁵ Moreover, long-term cost-effectiveness assessments of TBI interventions are crucial to gauge their full health benefits and highlight, once more, the need for valid survival estimates specific to TBI victims.²⁶ This would help invest in the appropriate care for TBI patients, which could significantly improve their quality of life and ultimately result in savings compared to the costs of providing longer-term healthcare.²⁷

The present systematic review aims to provide a comprehensive overview of long-term life expectancy in patients following severe traumatic brain injury (sTBI). Specifically, our objectives are to: 1) summarize available data on survival rates, mortality trends, and causes of death in this patient population; 2) critically assess the quality of the included studies; 3) identify factors that may impact long-term life expectancy in individuals who have experienced sTBI. Our secondary objectives are to explore the clinical and research implications of these findings, considering the geographic origins of each study and related socioeconomic, welfare policy, and social contextual factors, verify the transferability of the available data, predominantly from North America to different countries, and contribute developing recommendations that address the long-term needs of TBI patients, improve their quality of life, and optimize resource allocation.

Evidence acquisition

The review was per the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines.^{28, 29}

A literature search for journal articles published from January 1^st, 2000, up to July 31^st, 2023, was performed using the following databases: PubMed, Web of Science, Cochrane Library, Google Scholar, and PEDro. We used Rayyan software to create a database containing the search results. The search focused on these terms: ‘severe traumatic brain injury’, ‘long-term outcome’, and ‘life expectancy’. Detailed mesh term branching is reported in Supplementary Digital Material 1 (Supplementary Text File 1). The retrieved articles were independently scanned for inclusion criteria by three reviewers (A.D.T., S.B., D.S.); they were also scanned for additional papers that may have been missed in the original search. Results were limited to studies published in English.

Finally, to be included in this systematic review, the published manuscript must contain all necessary data.

Study selection

We employed the following inclusion criteria for our study: 1) Study designs encompassed randomized controlled trials, quasi-experimental design, cohort or historical cohort studies, case-control studies, or cross-sectional studies; systematic reviews were utilized to identify further relevant studies; 2) patients were ≥16 years old at the time of suffering the TBI; 3) brain injury severity was severe TBI (Glasgow Coma Scale score, GCS <9; Abbreviated Injury Scale, AIS ≥3; Injury Severity Score, ISS ≥16); 4) any mechanism causing traumatic brain injury was considered; 5) the study included a control or comparison population, which could be a comparable general population, or a group of individuals with other medical conditions not involving brain injury; 6) publications were published in English even if carried out in different countries; 7) mortality rates were provided stratified by Glasgow Coma Scale score and/or age, sex, race group; 8) a follow-up time ≥3 months was specified when assessing mortality; 9) study outcome included life expectancy, standardized mortality ratio, and mortality rate.

We applied the following exclusion criteria: 1) mortality rate data could not be accurately extracted from the article or mortality rates were not specifically provided for the control group; 2) studies that included individuals with TBI who died at the scene or before hospital admission were excluded; 3) nontraumatic mechanisms of injury, such as inflammation, infection, or autoimmunity; 4) studies involving veterans; 5) studies involving mild TBI cases; 6) letters, editorials, reviews. Veteran studies were excluded to maintain homogeneity in the study sample, ensuring broader applicability of the findings to the general population. Veterans often experience distinct risk factors, injuries, and access to specialized healthcare systems, which could introduce variables not representative of the wider population.

Risk of bias

The quality and level of evidence of each study were assessed independently by three authors (A.D.T., S.B., S.D.). The quality of evidence was assessed based on internal validity factors (such as study design, reporting quality, presence of selection and misclassification biases, and potential confounding) and external validity (generalizability). This evaluation was conducted using the “Methodological Index for NOn-Randomized Studies” (MINORS) quality assessment tool,³⁰ which assigns a score out of 24 total points to each study. The MINORS is a validated instrument designed to evaluate the methodological quality of non-randomized studies in both comparative and non-comparative contexts. It includes 12 criteria, with each criterion scored from 0 to 2, for a maximum score of 24 for comparative studies. The criteria assess aspects such as the clarity of the study aim, the inclusion of consecutive patients, the prospective collection of data, and the adequacy of follow-up.

Furthermore, each study’s evidence level was categorized according to the Oxford Centre of Evidence-Based Medicine (OCEBM) model.³¹ To adhere to study exclusion criteria, levels 1a, 2a, 3a (systematic reviews), 4 (case series), and 5 (opinion-based papers) were excluded from consideration. Any discrepancies in MINORS scoring or OCEBM categorization were resolved through consensus among the three reviewers responsible for assessing the study. All quality assessment scores are reported in Table I.^{1, 6, 7, 11, 20, 24, 32-49}

Table I. —Modified MINORS Scale scores.^{1, 6, 7, 11, 20, 24, 32-49}.

Study name	A clearly stated aim	Inclusion of consecutive patients	Prospective collection of data	Endpoints appropriate to the aim of the study	Unbiased assessment of the study endpoint	Follow-up period appropriate to the aim of the study	Loss to f/up less than 5%	Contemporary group	Adequate statistical analysis	Score
Baguley33	2	2	2	2	0	2	2	2	2	16
Baguley32	2	2	2	2	0	2	2	2	2	16
Brooks6	2	2	2	2	0	1	2	2	2	15
Brooks24	2	2	2	2	0	2	2	2	2	16
Brooks35	2	2	2	2	0	2	2	2	2	16
Cameron34	2	2	2	2	0	2	2	2	2	16
Claridge36	2	2	2	2	0	2	2	2	2	16
Colantonio7	2	2	2	2	0	2	2	2	2	16
Dams-O’Connor37	2	2	2	2	0	2	2	2	2	16
Esterov39	2	2	2	2	0	2	2	2	2	16
Forslund20	2	2	2	2	0	2	2	2	2	16
Groswasser38	2	2	2	2	0	2	2	2	2	16
Harrison-Felix40	2	2	2	2	0	2	2	2	2	16
Harrison-Felix11	2	2	2	2	0	2	2	2	2	16
Harrison-Felix42	2	2	2	2	0	2	2	2	2	16
Majdan43	2	2	2	2	0	0	2	2	2	14
McCrea41	2	2	2	2	0	1	2	2	2	15
McMillan44	2	2	2	2	0	2	2	2	2	16
Nakase-Richardson45	2	2	2	2	0	2	2	2	2	16
Ratcliff1	2	2	2	2	0	0	2	2	2	14
Selassie46	2	2	2	2	0	2	2	2	2	16
Ventura48	2	2	2	2	0	2	2	2	2	16
Wiegers49	2	2	2	2	0	2	1	2	2	15
Wilkins47	2	2	2	2	0	2	2	2	2	16

Maximum points are 18 for each study.

Data extraction and reporting

A data extraction form was used to summarize these characteristics for each study: author, publication year, data collection year, country of origin, study design (retrospective versus prospective), population (TBI and control group), population details (including sample size, age, and sex distribution, GCS or analog scales, etc.), primary outcome, and follow-up. The most frequently reported outcomes were survival rates, life expectancy, mortality trends, cause of death, standardized mortality ratio, and mortality rate. The extracted data were then reviewed for completeness by three authors (A.D.T., S.B., D.S.). The same authors independently assessed the quality and level of evidence of each study. The quality of evidence underwent assessment based on both internal validity factors (such as study design, reporting quality, presence of selection and misclassification biases, and potential confounding) and external validity (generalizability). Discrepancies between reviewers were resolved through discussion and consensus.

Data analysis

The analysis aimed to synthesize the findings from the included studies to provide a comprehensive overview of long-term life expectancy in patients with severe traumatic brain injury (TBI). To analyze the extracted data, we utilized descriptive statistics. Relevant data was reported using median, means, and outcome ranges. Calculations, figures, and tables were made using Microsoft Excel and Microsoft Word.⁵⁰

Figure 1 illustrates the PRISMA flow diagram of our study selection process. The initial phase (Identification) involved searching multiple databases and other sources, such as the reference lists hand search, to identify potentially relevant studies. After removing duplicate records, we moved into the Screening phase and reviewed the titles and abstracts of the remaining to determine their relevance to our review. In the eligibility phase, we retrieved and assessed the full-text articles of the screened records to evaluate if they met the eligibility criteria. Articles that failed to meet the criteria were excluded, with specific reasons for exclusion documented. The final phase involved the inclusion of studies that met all the eligibility criteria in the systematic review. The total number of included studies was noted.

Figure 1.

—Flow diagram of the literature selection process.

Data synthesis

We summarized the extracted data, the quality, and the level of evidence for each study.

Evidence synthesis

Identification of studies

An overview of the study identification process is represented in Figure 1, while the detailed breakdown of the search procedure is summarized in Supplementary Digital Material 2 (Supplementary Table I).^{1, 6, 7, 11, 20, 24, 32-49} We identified a total of 9099 studies through our initial database search. Among them, 65 studies were deemed relevant for full-text review. Following a detailed assessment, 24 studies met the inclusion criteria and were included in the final analysis. Studies were mainly excluded for: 1) not reporting data on outcomes of interest; 2) reporting data on populations with mixed age and/or disease severity; 3) being reviews or editorial; and 4) reporting duplicate analysis. For instance, studies by McMillan et al.⁵¹ and Harrison-Felix et al.⁵² were removed since the data was retrieved from the same database as was used by McMillan et al.⁴⁴ and Harrison-Felix et al.,⁵³ respectively.

Study characteristics and quality

The studies were published between January 1^st, 2000, and July 31^st, 2023, with the majority originating from the USA (N.=15, 63%), followed by Canada (N.=4, 17%), Europe (N.=2, 8%), the UK (N.=1, 4%), and Israel (N.=1, 4%). One study (4%) included data from multiple countries. Nine studies (37.5%) were retrospective, while six (25%) were prospective in design (Supplementary Table I), with a median sample size of 2545 individuals and a range from 97 to 18,988 across studies. The median follow-up duration was 9.5 years, varying from 0.5 to 25 years. Follow-up periods from 0 to 5 years were reported in eight studies (33%),^{6, 36, 41, 45-49} follow-ups from 5 to 10 years in four studies (17%),^{7, 20, 34, 37} and a follow-up longer than 10 years were reported in ten studies (42%).^{11, 24, 32, 33, 35, 38-40, 44, 53} Mechanisms of injury include falls, motor vehicle accidents, assault, blunt force trauma, and other/unknown causes, with subjects having severe injuries (GCS 3-8).

The MINORS scale was used to evaluate the methodological quality of each study. Due to the observational nature of the studies, three items on the MINORS did not apply to all included studies, preventing full scores. The mean MINORS score was 15.7 (range, 14-16 out of 18); the median was 16. Nineteen studies (79%) scored 16, three studies scored 15 (13%), and the remaining two studies scored 14 (8%). Most studies scoring lower than 16 exhibited common limitations, such as incomplete sample representation descriptions, limited information on characteristics of those lost to follow-up, use of unreliable measures, inadequate reporting on handling follow-up participants, and differing follow-up lengths in statistical analyses. In general, all studies were based on prospectively collected data with appropriate follow-up and adequate statistical analysis.

All reviewed studies demonstrated the highest level of evidence, categorized as level 2b according to the OCEBM Levels of Evidence Working Group,³¹ signifying individual cohort studies.

Outcome measures

The most investigated outcomes were life expectancy (N.=6, 25%), standardized mortality ratio (N.=11, 46%), and mortality rate (N.=16, 66%). Other outcomes mainly described were causes of death after TBI (N.=8, 33%) and factors related to survival after TBI (N.=5, 21%) (Supplementary Table I).

Life expectancy

Evidence suggests that survivors of TBI generally have a reduced life expectancy compared to the general population. This may be due to the injury itself or patient characteristics that predispose them to TBI and early mortality.²⁵ Life expectancy is influenced by factors such as age, sex, and severity of disability or functional status,^{6, 11, 24, 54} which represent reliable predictors of survival time after TBI.^{54, 55} Non-ambulatory individuals who can self-feed may have an estimate of their life expectancy varying from 46 years at age 10 to 19 years at age 50.⁵⁴ When an individual is unable to feed themselves or move independently, their life expectancy can decrease significantly. For instance, a 10-year-old’s life expectancy may be reduced by up to 27 years, while a 50-year-old’s life expectancy may decrease to just 11 years.⁵⁴ However, as soon as an individual can walk independently, their life expectancy improves markedly, often nearing that of the general population.⁵⁴ People in a vegetative state have a maximum average life expectancy of 12 years, regardless of their sex.⁵⁴ Shavelle and Strauss⁵⁴ also suggest that individuals in a minimally conscious state (MCS) may have slightly better life expectancies than those in a persistent vegetative state (PVS), even though data and clinical criteria for diagnosing MCS were not provided.⁵⁶ Another study⁴⁵ highlighted significant and ongoing recovery among patients admitted to inpatient rehabilitation in a minimally responsive state. This underscores the need for further research to identify individuals with the greatest potential for recovery and to assess the impact of different care systems at various stages on long-term outcomes.

Rate of death

Mortality rates reported in studies examining mortality patterns vary based on the time elapsed since the injury. The risk of death is notably elevated during the initial 1-2 months after the injury³⁴ and remains elevated up to 15 months post the lesion onset.⁴⁶ Another study corroborated these findings,⁴⁷ acknowledging the high mortality rate among severe TBI patients within 3 months, despite focusing on long-term recovery trajectories for survivors of the acute hospital phase. They observed a 46% mortality rate after 2 years, yet three-fourths of severe TBI survivors achieved favorable outcomes. McMillan and Teasdale⁵¹ highlighted a 23-fold increase in mortality risk during the first 1-2 months post-injury, with a subsequent decrease but still elevated risk in months 3-12 and 13-84. They also highlighted that the mortality risk was significantly associated with the age of the injury onset. They reported that the mortality risk was notably higher in individuals older than 54 years during the initial two months, while those younger than 54 exhibited a sevenfold increase in mortality risk between 13-84 months after the injury. Higher mortality risk within the first year was associated with injury severity,⁴¹ whereas pre-injury medical history impacted both early and late deaths,³⁸ albeit the risk remained higher even without injury. Suboptimal post-injury lifestyles contributed to later deaths, with primary causes aligning with those observed in the general population. McMillan et al.⁴⁴ demonstrated persistently high mortality rates among TBI patients admitted to hospitals, particularly pronounced in individuals under fifty-five for at least thirteen years post-admission. Esterov et al.³⁹ carried out a population-based case-matched study, revealing an increased risk of death only within the initial six months post-injury, primarily due to external causes. In another study, mortality rates were compared to an equivalent population sample, indicating a standardized mortality rate of 13.2 over an average post-injury period of 10.5 years. The distribution of mortality trends suggested two distinct peaks, with more deaths occurring within the first five years followed by a period of no further deaths until nine years post-injury, indicating varying contributory mechanisms to early versus late mortality.³²

Standard mortality ratio

Colantonio et al.⁷ conducted a retrospective cohort study to examine post-acute mortality rates and predictors among severe TBI patients one to nine years post-injury, covering an approximately two-year period (April 1993-March 1995). The study included 2721 patients with severe TBI and 557 control patients with lower extremity injuries. Post-acute death was defined as occurring one year or more post-discharge. Poisson regression modeling revealed that TBI predicted premature death, even after controlling for age and injury severity (Standard Mortality Ratio, SMR, of 2.90 for the TBI group vs. 2.26 for the control group, P=0.046). Age, comorbidities, injury severity, mechanism of injury, and discharge destination significantly predicted SMR. Consistent with previous studies by Ratcliff¹ and Shavelle et al.,⁵⁴ the study found that SMR decreased with increasing age, indicating a reduction in the ratio of raw death rate to expected death rate with older age. In a separate inception cohort study, Baguley et al.,³³ examined 2545 adults discharged from metropolitan tertiary post-acute inpatient rehabilitation services between 1990 and 2007 after primary TBI. They found a standardized mortality ratio of 3.19 (95% CI, 2.80-3.60) two years post-discharge, with elevated mortality risk persisting for at least 8 years.³³ Functional dependence at discharge, age at injury, pre-injury drug and alcohol misuse, pre-injury epilepsy, and discharge to aged care facilities were associated with increased mortality risk.³³ Adults with severe TBI faced significantly higher risks of death from external causes, respiratory and nervous system disorders (six to seven times higher), digestive system disorders, and mental and behavioral disorders (five times higher) compared to the general population.³⁹ Lastly, Harrison-Felix et al.⁴² employed a retrospective cohort design to analyze mortality rates, life expectancy, risk factors, and causes of death among 1678 TBI survivors admitted to acute rehabilitation within one year of injury between 1961 and 2002. TBI patients were 1.5 times more likely to die than individuals in the general population, with an average reduction in life expectancy of four years. Independent risk factors for death beyond one year post-injury included older age, male sex, lower education levels, longer hospital stays post-injury, earlier injury occurrence, and vegetative status upon rehabilitation discharge. The risk of death due to aspiration pneumonia was 49 times higher in TBI patients compared to the general population, with elevated risks for death from seizures, suicide, and digestive conditions as well (22, 3, and 2.5 times higher, respectively).^{40, 42}

Cause of death after TBI

The current literature indicates a notable increase in long-term mortality among severe TBI survivors.²⁵ Also, TBI-related mortality rates showed differences based on sex and age group.⁵ For instance, among individuals from 20–to 24 and those 75 years or older, firearm-related TBI deaths were most prevalent. Motor vehicle-related TBI fatalities were most common among those aged 15-24. For fall-related TBI deaths, the highest rates were observed in adults 75 years and older, with a notable rise correlating with age.^{5, 12} Remarkably, across all age categories and causes of injury, males consistently exhibited higher rates of TBI-related mortality compared to females.^{5, 43} Despite the general agreement about those findings, studies differ regarding the specific causes of excess mortality. In the very short term, brain damage itself is the primary cause of death.⁵⁷ Yet, in the long term, individuals with TBI face elevated death rates from various causes compared to the general population.⁵⁸ These causes, outlined in Table II, include epilepsy, suicide, respiratory infections like pneumonia, meningitis, and circulatory system diseases.

Table II. —Leading causes of death after a TBI.

Cause of death	Description
Neurological complications	Infections, cerebral hemorrhage, brain edema
Respiratory failure	Hypoxemic, hypercapnic/ventilatory, secondary to cardiovascular instability
Cardiovascular complications	Hypotension, cardiogenic shock
Pulmonary embolism	Blood clots that move to the lungs
Choking	Difficulty in swallowing or managing airways
Multiorgan injuries	Damage to multiple vital systems
Nosocomial infections	Infections contracted during hospitalization
Thromboembolic events	Deep vein thrombosis, pulmonary embolism
Cardiac arrest	Sudden cessation of cardiac activity

Harrison-Felix et al.⁵³ carried out a retrospective cohort study on causes of death in TBI individuals, utilizing data from a federally funded TBI database and vital status data from multiple sources. They found that individuals with TBI were significantly more likely to die from seizures, septicemia, pneumonia, respiratory conditions, digestive conditions, and external causes of injury/poisoning compared to the general population.⁵³ However, the study’s follow-up averaged only 3.1 years from the first post-injury anniversary and did not analyze the severity of injury and the extent of the patient’s functional impairment, apart from stating that the average age was 37 years, with 76% being male, 50% Caucasian, and 42% having a severe TBI based on an emergency department Glasgow Coma Scale (GCS) score between 3 and 8.⁵³

Walker et al.⁵⁷ proposed that a traumatic brain injury renders the entire body more susceptible to aging-related stressors. Conversely, Lewin et al.⁵⁹ challenged this hypothesis, noting that only a few causes of death appeared elevated in their TBI population. In contrast, Shavelle et al.⁶⁰ found increased mortality from most major causes other than cancer, which may support the view of Walker et al. Ultimately, TBI has been linked to an increased likelihood of facing neurodegenerative diseases.¹³

Factors related to survival after TBI

Various factors influence the long-term survival of individuals following severe TBI, with clinical measures like the Glasgow Coma Scale and post-traumatic amnesia duration being primary predictors in the short term.⁶¹ However, Walker et al.,⁶¹ focused on factors indicative of long-term survival. Apart from age,^{36, 40} key determinants include trauma severity and residual disability, particularly motor dysfunction, which encompasses the ability to walk, use hands, and self-feed.^{6, 24, 35, 37, 54} Other factors correlated with survival time post-TBI include age, time since injury, ventilator support, cognitive impairments, respiratory problems like pneumonia, circulatory system diseases, and post-traumatic epilepsy.^{54, 62-64} Additionally, ethnicity, socioeconomic background, and destination after hospital discharge (home versus institution) also play roles in survival outcomes.^{65, 66}

The issue of care quality relative to life expectancy carries significant clinical and forensic implications. Defining good quality care versus suboptimal care poses challenges, considering medical and psychosocial aspects. Quality of care also hinges on caregiver expertise, health monitoring access, and intensity of care provided.⁶⁷ Remarkably, all these variables are influenced by financial resources and healthcare systems in which individuals are taken care of and they will vary from country to country.⁶⁸ For instance, Strauss et al.⁶⁹ showed that mortality rates differ between state facilities (lower) and private group or family homes, suggesting quality disparities in the care provided. Different studies report contrasting results.¹³ McMillan et al.,⁷⁰ based on a study carried out in a cohort of homeless TBI patients in Scotland, reported a different piece of evidence and demonstrated that appropriate and prolonged care, coupled with family support, can mitigate TBI consequences. However, these interventions require structured care systems and financial resources. Individuals lacking financial and social support face heightened TBI consequences, as evidenced by homeless TBI patients experiencing elevated mortality rates post-discharge. In fact, in the above-mentioned study, the authors showed homeless have a 5.4 times increased risk of TBI compared with the normal population.⁷⁰ After discharge and back to their condition of homelessness, they had double the mortality rate compared with homeless cases not hospitalized.⁷⁰ Although care quality impacts survival time, its overall impact on severely disabled persons post-TBI remains unclear: factors like the degree of immobility may outweigh the quality of care in determining survival outcomes.⁵⁴

Poor education and maladaptive behaviors, including substance abuse and suicide attempts, also influence survival rates, albeit negatively.⁷¹ Maladaptive behaviors, in general, have all been shown to be associated with increased mortality. Harris and Barraclough⁷² showed an increase in deaths due to unnatural causes, including accidents, suicide, and homicide.

To conclude, there are variables relevant to life expectancy in the general population that are also relevant for persons with TBI, such as pre-injury health status, history of smoking, alcohol and drug abuse, and obesity.⁵⁴ However, there are clinical factors unique to TBI, such as chewing and swallowing, contractures, bowel and bladder dysfunction, frequency of infections, and psychological diseases (i.e., depression, post-traumatic stress disorder, anxiety, and psychosis) that further affect mortality risk.^{54, 73, 74}

Discussion

This systematic literature review has identified several key findings regarding long-term life expectancy in patients with severe traumatic brain injury (sTBI). The main outcomes include significant variability in life expectancy, mortality rate, standardized mortality ratio (SMR), causes of death, and factors related to survival after sTBI. These findings highlight the complexity and heterogeneity of long-term outcomes for sTBI patients.

The life expectancy of an individual is the average number of additional years of life in a large group of similar persons and, in many cases, it can be estimated with some precision.⁷⁵ Estimates of life expectancy are crucial in many fields but distinct from predicting the actual survival time, which remains challenging even in the uninjured general population.⁵⁴ Various predictors, including sex, injury severity, age, and motor and cognitive function can help estimate life expectancy. Other factors, such as pre-existing health conditions, healthy habits, socioeconomic status, education level, and family/social support should be considered as well. However, caution is necessary when using population data, as it may not fully account for individual circumstances.⁵⁴ The review indicates that survivors of TBI generally experience a reduced life expectancy compared to the general population. This reduction can be attributed to the injury itself or pre-existing patient characteristics that predispose them to TBI and early mortality. Factors such as age, sex, and severity of disability or functional status emerge as significant predictors of survival time post-TBI. Mobility and self-care skills are crucial predictors of survival, with ambulation and self-feeding linked to better outcomes.^{6, 24, 37} The life expectancy of individuals who can walk and self-feed experiences only a slight decrease compared to the general population.⁵⁴ Conversely, those who are non-ambulatory and have limited self-care abilities face a significantly greater reduction in life expectancy.⁵⁴ Trauma literature, though informative, may be biased and lacks comprehensive control for confounding variables like initial injury severity and comorbidities. For instance, most data originating from registries and care series are susceptible to selection bias. As Thompson et al.⁷⁶ pointed out, only a few researchers, in their studies, have employed various statistical techniques like multiple or logistic regression, matching, or other methods to address crucial confounding factors, such as initial injury severity, comorbidities, and baseline health status when assessing the impact of chronological age on TBI outcomes. Moreover, the same authors underlined that the considerable variability in mortality rates among older adults likely stems from methodological and conceptual differences across studies, including varying definitions of “older adults” and discrepancies in sample sizes. They also stressed how many studies examining age as a predictor of outcome have insufficient representation of older adults and an overrepresentation of younger adults, making it challenging to accurately determine mortality rates and outcomes in older TBI patients.^{75, 76}

Mortality rates following TBI are higher than in the general population, especially in the initial months following injury, with a continued elevated risk up to 15 months post-injury.^{33, 48, 53} Studies have demonstrated that the risk of death is significantly higher within the first 1-2 months post-injury, particularly among older individuals.^{33, 48, 53}

Complications post-TBI, such as persistent motor and sensory deficits, worsen morbidity and mortality. Stocchetti and Zanier¹³ suggested these kinds of impairments can result from direct trauma to the nervous system with prolonged immobilization during hospitalization as a possible cause that can exacerbate the damage, potentially leading to complications like peri-articular calcification. Additionally, bladder and sphincter control may be also affected, worsening the quality of life of TBI survivors.¹³ Despite the high mortality rates, a substantial portion of severe TBI survivors achieve favorable outcomes. Key determinants of mortality risk include age at injury, severity of the injury, and pre-injury medical history. Additionally, suboptimal post-injury lifestyles contribute to late deaths, with primary causes aligning with those observed in the general population. Notably, mortality rates remain elevated for several years post-injury, particularly among individuals under 55 years of age.

Several studies have reported elevated SMRs among TBI patients compared to the general population, indicating a higher-than-expected death rate post-injury. Age, comorbidities, injury severity, and discharge destination significantly predict SMR. For example, one study reported an SMR of 3.19 within two years post-discharge, with an elevated mortality risk persisting for at least eight years. Factors like functional dependence at discharge, pre-injury substance misuse, and discharge to aged care facilities were associated with increased mortality risk. Adults with severe TBI face significantly higher risks of death from external causes, respiratory and nervous system disorders, digestive system disorders, and mental and behavioral disorders compared to the general population.

Long-term mortality among severe TBI survivors is notably higher than in the general population, with differences observed based on sex and age group. Common causes of death include seizures, pneumonia, respiratory and digestive conditions, and external causes of injury/poisoning. Males consistently exhibit higher rates of TBI-related mortality compared to females across all age categories and causes of injury. Notably, the risk of death due to aspiration pneumonia is significantly higher in TBI patients. The increased mortality from various causes, including neurodegenerative diseases, underscores the heightened vulnerability of TBI survivors to multiple health issues.

Several factors influence the long-term survival of individuals following severe TBI. Clinical measures such as the Glasgow Coma Scale and duration of post-traumatic amnesia are primary predictors of short-term survival. However, long-term survival is influenced by age, injury severity, functional status, and residual disabilities. Other significant factors include cognitive impairments, psychiatric disorders, and maladaptive behaviors that are common post-TBI, and also affect rehabilitation and employment prospects.⁷⁷ TBI leads to deficits in attention, memory, speed of information processing, and executive functioning.¹³ Comprehensive neuropsychological assessments of moderate to severe TBI cases have revealed impairments in sustained attention, associative learning, and reaction time.^{13, 78} Additionally, psychiatric disorders such as depression, anxiety, and psychosis, along with maladaptive behaviors and personality changes, can arise post-TBI.⁷⁴ A meta-analysis has shown an increased incidence of psychiatric disorders after TBI, with depression and bipolar disorders being particularly prevalent.⁷⁹ Lastly, long-term psychiatric disorders are associated with a higher risk of substance abuse.⁷⁹ Altogether, physical and cognitive disabilities significantly impede daily functioning and hinder the individual’s ability to resume a normal lifestyle.¹³ Additionally, the combination of these deficits results in a heightened rate of unemployment among TBI survivors.¹³ Returning to work is challenging for TBI survivors, particularly in severe cases, leading to financial strain and social isolation.^80-82 Patients recovering from severe TBI are sometimes offered a sheltered working environment, while return to preinjury work positions is unlikely, and locating new employment is very difficult.^{13, 68} Even individuals of working age who seem to have achieved positive outcomes encounter challenges in returning to their jobs. For instance, in a study conducted in Norway 10 years after the injury, only 58% of participants were employed.⁸¹

Physical, cognitive, and emotional deficits represent a significant barrier to reintegrating into the community. Decreased social interactions, depression, financial constraints, unemployment, and physical limitations can profoundly disrupt existing social relationships and make social and leisure activities impossible.⁸³ Quality-of-life assessments consistently reveal poorer outcomes in TBI patients,⁸⁴ especially older individuals.⁸⁵

Overall, multiple factors influence long-term outcomes post-TBI with motor, socio-cognitive functioning, and mental health being among the most affected areas.⁸⁶

Conclusion and future directions

The reviewed studies present several limitations. Many are retrospective and observational, raising concerns about potential bias. Moreover, conflicting results across studies likely stem from variations in inclusion criteria, methodologies, and study populations. Additionally, the literature lacks a comprehensive exploration of factors influencing long-term life expectancy, such as the quality of care, which varies by country, pre-existing conditions, and socioeconomic factors. Notably, much of the data on long-term survival after severe head trauma originates from North America, presenting a distinct healthcare system compared to Europe or low-income countries. In one study characteristics, management, and outcomes of TBI patients were compared across three different countries (Australia, the United Kingdom, and Europe) and despite some differences in management and discharge policies, mortality did not differ between regions.⁴⁹ Even though, it is unlikely that healthcare efficacy or healthcare systems do not affect the survival outcomes after a TBI, the data are insufficient to draw any conclusions.

Moreover, the literature has yet to thoroughly investigate important variables like premature brain aging post-trauma. Research indicates TBI can cause chronic diseases in approximately a quarter of TBI cases, with a functional decline observed in 7-13 years following the injury,⁸⁷ suggesting that TBI may be the substrate for the induction of neurodegenerative chronic processes.¹³ TBI has been linked to cognitive decline, encompassing conditions like Alzheimer’s disease and chronic traumatic encephalopathy.^87-89 Moreover, individuals with TBI face increased risks for Parkinson’s disease.⁹⁰ However, further research is needed to clarify these associations. Behavioral, cognitive, and psychological sequelae are also underexplored. Severe TBI patients may experience anxiety, depression, PTSD, and psychosis, impacting their and their caregivers’ quality of life. Epidemiological evidence supporting these claims is limited, even though quality of life should be a critical consideration, as patients often contend with physical, cognitive, and psychological challenges, resulting in substantial economic costs.⁴

To address these limitations, future prospective, controlled studies are required. These studies should include diverse patient populations, ensure sufficient follow-up, employ standardized criteria and validated methodologies, and assess all factors impacting long-term life expectancy. Additionally, meta-analyses are warranted to mitigate methodological limitations, including data scarcity, result heterogeneity, and bias risks, providing valuable insights for clinicians and researchers working with severe TBI patients all around the world.

Supplementary Digital Material 1

Supplementary Text File 1

Search strategy

Supplementary Digital Material 2

Supplementary Table I

Characteristics of the studies included.^{1, 6, 7, 11, 20, 24, 32-49}