Abstract
Traumatic brain injury (TBI) is a significant public health problem in the US. Specific preexisting medical illnesses delay recovery after TBI and increase mortality or risk of repeat TBI. This study examined the impact of preexisting illness and substance use on patient rehabilitation outcomes following TBI. The Functional Independence Measure total score and Disability Rating Scale score measured functional outcomes at discharge from inpatient rehabilitation, while the Trail Making Test A and B and Total Trials 1–5 of the California Verbal Learning Test–II measured neuropsychological outcomes in 128 TBI survivors with moderate or severe TBI. Results showed that the presence of a heart condition or diabetes/high blood sugar was associated with lower functional outcomes by discharge. A history of a heart condition, stroke, or respiratory condition prior to TBI was associated with reduced cognitive flexibility. Those with preexisting diabetes/high blood sugar demonstrated poorer visual attention, visuomotor processing speed, and ability to learn and recall verbal information. Those with pre-TBI cancer also had greater auditory-verbal memory deficits. The findings showed that specific preexisting medical conditions are independently associated with lower functional and cognitive outcomes for patients with TBI. By screening patients for preexisting medical conditions, multidisciplinary TBI rehabilitation teams can identify patients who require more aggressive treatments or greater length of stay.
The purpose of this study was to examine the relation of preexisting medical illnesses and substance use with both functional status and neurocognitive status at the time of discharge from inpatient rehabilitation in individuals with traumatic brain injury (TBI). Preexisting cancer, preexisting liver disease, and use of tobacco/cigarettes were included, as few studies have examined their association with post-TBI functional outcomes. No studies to date have evaluated the impact of preexisting medical illnesses on neuropsychological outcomes.
METHODS
This study was part of the Traumatic Brain Injury Model System (TBIMS), with data collected at a center in an urban metroplex in the Southern US (1). TBIMS is a longitudinal multicenter, prospective, observational study of patients aged ≥16 years who sustained a moderate to severe TBI due to blunt or penetrating injuries. Additional information regarding TBIMS inclusion criteria and the definition of TBI can be found in Dijkers, Harrison-Felix, and Marwitz's historical review of the TBIMS (1). Patients with concomitant spinal cord injury, whose inpatient rehabilitation course exceeded 90 days, or who died during their inpatient rehabilitation stay were excluded from this study. Collection of preexisting medical illnesses and substance use information was initiated in October 2012. Therefore, 138 patients enrolled in the TBIMS longitudinal study between October 1, 2012, and November 5, 2013, were screened for inclusion in this retrospective study. Of 138 screened patients, 128 (93%) had both complete functional outcome data and information regarding all preexisting health conditions in the TBIMS national database. Patients with preexisting medical illness (n = 39) were compared with those without any such illnesses (n = 89).
Of the 128 patients with preexisting health data, neuropsychological testing was obtained in 77 patients (60%) after repeated evaluation confirmed that the patients had emerged from posttraumatic amnesia (PTA) and were oriented 72 hours later during their rehabilitation course. The rest of the patients (n = 51) were still in PTA at or near the time of discharge from rehabilitation and hence were excluded from neuropsychological testing. Testing was completed by the neuropsychologist or psychometrist on the inpatient TBI service. Data were included for analysis even if patients were outside the TBIMS testing window of 4 weeks (±2 weeks) postinjury (2). Despite the longitudinal nature of the TBIMS, only discharge outcomes were examined in this study, given that neuropsychological data were only collected during inpatient rehabilitation.
Based on preexisting health status and duration of PTA, patients fell into one of four groups: 1) patients with preexisting illness who completed neuropsychological measures; 2) patients with preexisting illness who did not complete neuropsychological measures; 3) patients without a history of preexisting illness who completed neuropsychological measures; and 4) patients without a history of preexisting illness who did not complete neuropsychological measures.
Functional outcomes were measured using the total score of the Functional Independence Measure (FIM) (3, 4) and Disability Rating Scale (DRS) (3, 5) at discharge. Neuropsychological outcomes were measured during inpatient rehabilitation using the Trail Making Test (TMT) (6–8) completion time in seconds and the California Verbal Learning Test, 2nd edition (CVLT-II) (9) total number of items immediately recalled on Trials 1 to 5. These are the only neuropsychological measures currently available in the TBIMS National Database.
Briefly, the FIM comprises 18 items designed to measure functional independence in self-care, mobility, and communication and social cognition (3, 4). Performance on each item is rated between 1 and 7 (7 = complete independence). The maximum score possible on the FIM is 126 (4). Higher FIM scores reflect better functional status. FIM efficiency was defined as FIM change (FIMdischarge – FIMadmission) divided by total length of stay, in days.
The DRS is an 8-item scale that incorporates a modification of the Glasgow Coma Scale items and other items assessing cognitive ability to manage activities of daily living, the need for assistance or supervision, and potential employability (3, 5). The maximum score possible on the DRS is 29, with zero indicating lack of disability. A lower DRS score reflects better functional status. DRS efficiency was defined as DRS change (DRSdischarge – DRSadmission) divided by total length of stay, in days.
The TMT consists of part A, which measures visual attention and visuomotor processing speed, and part B, which meas-ures the same functions as part A and mental flexibility. It is a brief and reliable measure that has been shown to be sensitive to cognitive impairments commonly exhibited in individuals with TBI (6–8). The score is based on completion time in seconds for each part (TMT:A and TMT:B) and is converted to a T score.
The CVLT-II is a comprehensive assessment of auditory-verbal learning and memory. The split-half reliability estimate ranges between 0.90 and 0.96. Construct validity for the CVLT-II is adequate (9). A composite score reflecting total number of items learned and immediately recalled across five exposure trials (Total List A Trials 1–5) was used in this study and was presented in the form of a T score. This is the only CVLT-II index available in the TBIMS database.
Age, sex, race/ethnicity, years of education, marital status, duration of PTA, length of stay, discharge disposition (home vs. care facility), and functional status at admission using the FIM and DRS were included in the analysis (3, 10–16).
Information on the presence or absence of medical conditions and substance use prior to TBI was obtained from patients' medical charts and by querying patients or family members. The conditions were predefined by the TBIMS: hypertension/high blood pressure, stroke, diabetes/high blood sugar/sugar in the urine, cancer, liver disease, emphysema/asthma/chronic obstructive pulmonary disease, use of tobacco/cigarettes, alcohol consumption, and use of illicit/nonprescription drugs. Congestive heart failure, myocardial infarction, and other heart conditions were merged into a single category, heart condition, due to the small number of patients in these individual heart conditions. Alcohol use was defined by the number of days per week or month alcoholic beverages (beer, wine, wine coolers, liquor) were consumed, the number of drinks consumed on average in a sitting, and the number of times the individuals consumed ≥5 drinks in a sitting the month prior to the TBI. Moderate use was 1 to 3 drinks and heavy use was >3 drinks, as defined by the National Institute on Alcohol Abuse and Alcoholism (17). Drug use was defined as use of any illicit or nonprescription drugs. This study was approved by Baylor University Medical Center's institutional review board.
Analyses were conducted using SAS, version 9.3 (SAS Institute Inc., Cary, NC). Linear regression was used to measure associations between preexisting medical illness and patient outcomes, while adjusting for patient demographics, severity of TBI, and functional status at admission, with P < 0.05 considered significant. All assumptions justifying use of linear regression were satisfied. Separate models were developed for each of the five outcomes. First, regression analyses were conducted examining the relationship of an individual preexisting illness with five outcomes: FIM at discharge, DRS at discharge, TMT:A, TMT:B, and CVLT-II. Next, the full regression model was examined to evaluate the impact of all preexisting illnesses combined on the five outcomes. The full regression model included all of the preexisting medical illnesses and substance use as covariates, as well as other potential confounders listed previously. The GLMSelect procedure was used to find the optimal model using the Schwarz Bayesian criteria.
RESULTS
Table 1 summarizes patient demographics for each of the four groups. The mean age of the overall study population was 47 years. Approximately 75% were men, 28% were racial/ethnic minorities, and 32% were married. The education level of these patients varied from less than high school (11%) to completion of graduate school (12%). Fifty-nine percent of patients engaged in moderate or heavy alcohol use prior to the TBI. Patients were admitted to rehabilitation centers at a mean of 2 weeks after sustaining injuries and had a mean inpatient rehabilitation stay of 19 days. Patients emerged from PTA a mean of 2 weeks from the time of injury (range = 1 to 22 days). Nearly 88% of the patients were discharged to their homes. Table 2 shows patients' functional status at admission and discharge and scores on three neuropsychological measures (TMT:A, TMT:B, and CVLT-II) during their inpatient stay.
Table 1.
Patient characteristics by group among 128 patients with traumatic brain injury
| Characteristic | Mean ± SD or % (n) | |||
|---|---|---|---|---|
| Group 1 (n = 18) | Group 2 (n = 21) | Group 3 (n = 59) | Group 4 (n = 30) | |
| Age (years) | 59.4 ± 18.8 | 65.9 ± 20.0 | 39.4 ± 16.9 | 43.1 ± 19.1 |
| Days from injury to rehab admission | 10.9 ± 6.3 | 13.6 ± 7.5 | 14.4 ± 6.5 | 13.2 ± 7.0 |
| Length of rehab stay (days) | 19.3 ± 11.4 | 26.8 ± 18.4 | 19.1 ± 11.9 | 15.4 ± 15.3 |
| Posttraumatic amnesia (days) | 8.2 ± 11.9 | 17.8 ± 22.2 | 13.7 ± 11.4 | 17.3 ± 17.9 |
| Discharge to home | 83.3% (15) | 71.4% (15) | 89.8% (53) | 96.7% (29) |
| Gender: Male | 66.7% (12) | 52.4% (11) | 78.0% (46) | 90% (27) |
| Racial/ethnic minority | 38.9% (7) | 23.8% (5) | 22.0% (13) | 36.7% (11) |
| Married | 33.3% (6) | 52.4% (11) | 25.4% (15) | 33.3% (10) |
| Alcohol | ||||
| No alcohol use | 50.0% (9) | 52.4% (11) | 22.0% (13) | 46.6% (14) |
| Moderate alcohol use | 33.3% (6) | 23.8% (5) | 52.5% (31) | 23.3% (7) |
| Heavy alcohol use | 16.7% (3) | 14.3% (3) | 23.7% (14) | 26.7% (8) |
| Not reported | 0 | 9.5% (2) | 1.7% (1) | 3.3% (1) |
| Education | ||||
| Less than high school | 22.2% (4) | 14.3% (3) | 5.1% (3) | 13.3% (4) |
| High school graduate | 22.2% (4) | 42.9% (9) | 33.9% (20) | 23.3% (7) |
| Some college | 27.8% (5) | 14.3% (3) | 33.9% (20) | 26.7% (8) |
| College graduate | 11.1% (2) | 14.3% (3) | 15.3% (9) | 30.0% (9) |
| Graduate school | 16.7% (3) | 14.3% (3) | 11.9% (7) | 6.7% (2) |
| Tobacco use | 16.7% (3) | 23.8% (5) | 30.5% (18) | 30.0% (9) |
| Drug use | 11.1% (2) | 0 | 25.4% (15) | 20.0% (6) |
Group 1 indicates patients with preexisting illness who completed neuropsychological measures; Group 2, patients with preexisting illness who did not complete neuropsychological measures; Group 3, patients with no preexisting illness who completed neuropsychological measures; Group 4, patients with no preexisting illness who did not complete neuropsychological measures.
Table 2.
Functional and neuropsychological measurements (outcome variables) in patients with and without preexisting illnesses before traumatic brain injury
| Outcome | Group 1 (n = 18) | Group 2 (n = 21) | Group 3 (n = 59) | Group 4 (n = 30) | ||||
|---|---|---|---|---|---|---|---|---|
| Mean ± SD | IQR | Mean ± SD | IQR | Mean ± SD | IQR | Mean ± SD | IQR | |
| FIM at admission | 54.8 ± 18.2 | 47 to 68 | 50.8 ± 19.9 | 38 to 59 | 57.6 ± 21.0 | 46 to 76 | 65.5 ± 23.2 | 53 to 79 |
| FIM at discharge | 89.2 ± 13.6 | 81 to 95 | 80.9 ± 19.8 | 72 to 94 | 98.6 ± 14.2 | 88 to 110 | 96.6 ± 16.4 | 86 to 112 |
| DRS at admission | 9.2 ± 3.1 | 7 to 9 | 10.9 ± 4.5 | 8 to 14 | 9.9 ± 3.5 | 7 to 11 | 8.9 ± 3.5 | 7 to 10 |
| DRS at discharge | 5.8 ± 1.7 | 5 to 6 | 6.1 ± 2.9 | 5 to 7 | 5.1 ± 1.4 | 4 to 6 | 5.5 ± 2.1 | 4 to 6 |
| FIM efficiency* | 2.1 ± 0.9 | 1.4 to 2.6 | 1.6 ± 1.2 | 0.8 to 1.9 | 2.6 ± 1.5 | 1.7 to 3.3 | 2.9 ± 1.7 | 1.5 to 3.6 |
| DRS efficiency* | 0.2 ± 0.1 | 0.1 to 0.3 | 0.2 ± .2 | 0.1 to 0.3 | 0.3 ± 0.3 | 0.1 to 0.4 | 0.3 ± 0.2 | 0.1 to 0.5 |
| TMT Part A T score | 30.1 ± 11.0 | 21 to 37 | – | – | 32.3 ± 16.6 | 20 to 43 | – | – |
| TMT Part B T score | 27.5 ± 12.7 | 18.5 to 39 | – | – | 32.4 ± 13.1 | 24 to 41 | – | – |
| CVLT trials 1–5 T score | 31.9 ± 14.5 | 19.5 to 41 | – | – | 35.3 ± 18.7 | 20 to 48 | – | – |
Efficiency is calculated as improvement.
CVLT indicates California Verbal Learning Test; DRS, Disability Rating Scale; FIM, Functional Independence Measure; Group 1, patients with preexisting illness who completed neuropsychological measures; Group 2, patients with preexisting illness who did not complete neuropsychological measures; Group 3, patients with no preexisting illness who completed neuropsychological measures; Group 4, patients with no preexisting illness who did not complete neuropsychological measures; IQR, interquartile range; SD, standard deviation; TMT, Trail Making Test.
Hypertension was the most common condition, found in 28% of patients (Table 3). About 16% had some form of heart condition, and about 10% had diabetes mellitus/high blood sugar. Few patients were diagnosed with stroke, respiratory conditions (emphysema/asthma/chronic obstructive pulmonary disease), cancer, or liver disease. Eighteen percent of the patients consumed >3 alcoholic beverages daily, while the rest consumed ≤3 alcoholic beverages daily (nonsignificant relationship with outcomes and therefore alcohol consumption was not included in Table 4). About 13% used illicit substances and 34% used tobacco or nicotine before their injury, but these predictors were not associated with outcomes.
Table 3.
Preexisting medical conditions among 128 patients with traumatic brain injury
| Condition | Frequency | Percentage |
|---|---|---|
| Heart condition | 20 | 15.6% |
| Hypertension/high blood pressure | 35 | 28.0% |
| Stroke | 6 | 4.8% |
| Diabetes/high blood sugar | 12 | 9.5% |
| Cancer | 5 | 4% |
| Liver disease | 2 | 1.6% |
| Emphysema, asthma, COPD | 8 | 6.4% |
| Alcohol consumed per sitting | ||
| >3 drinks | 21 | 18.0% |
| 1–3 drinks | 49 | 41.9% |
| None | 47 | 40.1% |
| Illicit drug use | 17 | 13.4% |
| Tobacco/cigarette use | 35 | 34% |
COPD indicates chronic obstructive pulmonary disease.
After adjusting for demographic variables, severity of TBI, and functional status at admission, regression analyses revealed that at discharge, patients with two or more of any type of preexisting medical illness performed a mean of 13 points lower on the FIM than patients without two or more illnesses. None of the preexisting illnesses alone demonstrated a significant relationship with any of the five outcomes (Table 4).
Table 4.
Regression analyses of individual preexisting conditions associated with functional and neuropsychological outcomes among 128 patients with traumatic brain injury
| Preexisting condition | Mean (SD) | |||||
|---|---|---|---|---|---|---|
| FIM total at discharge | DRS at discharge | TMT Part A T score | TMT Part B T score | CVLT trials 1–5 T score | ||
| Diabetes/high blood sugar | Present | 83.6 (22.5) | 5.8 (1.8) | 31.7 (9.0) | 32.8 (16.6) | 30.8 (9.2) |
| Absent | 95.0 (16.0) | 5.5 (2.0) | 31.6 (15.9) | 31.1 (13.1) | 34.6 (18.3) | |
| P value | 0.095 | 0.65 | 0.15 | 0.29 | 0.96 | |
| Heart condition | Present | 86.6 (15.5) | 6.0 (2.8) | 33.9 (10.5) | 32.4 (8.3) | 35.5 (16.0) |
| Absent | 95.2 (16.8) | 5.4 (1.7) | 31.4 (15.9) | 31.1 (13.6) | 34.5 (18.1) | |
| P value | 0.13 | 0.55 | 0.48 | 0.79 | 0.70 | |
| Emphysema, asthma, COPD | Present | 85.4 (11.1) | 5.8 (1.5) | 29.5 (12.0) | 21.0 (12.7) | 42.0 (35.4) |
| Absent | 94.4 (17.3) | 5.5 (2.0) | 31.6 (15.7) | 31.4 (13.2) | 33.8 (17.4) | |
| P value | 0.24 | 0.59 | 0.25 | 0.36 | 0.66 | |
| Cancer | Present | 78.4 (12.1) | 6.8 (2.4) | 26.0 (15.8) | 23.5 (17.6) | 22.0 (7.9) |
| Absent | 94.5 (16.9) | 5.4 (1.9) | 31.9 (15.4) | 31.7 (12.9) | 34.8 (17.9) | |
| P value | 0.55 | 0.50 | 0.07 | 0.49 | 0.37 | |
| Stroke | Present | 86.0 (19.6) | 6.5 (2.2) | 27.6 (9.0) | 19.0 (12.8) | 33.0 (17.5) |
| Absent | 94.2 (16.9) | 5.5 (1.9) | 31.9 (15.9) | 32.1 (12.9) | 34.1 (17.9) | |
| P value | 0.40 | 0.35 | 0.87 | 0.63 | 0.10 | |
| Liver disease | Present | 94.5 (0.7) | 5.0 (0) | N/A | N/A | N/A |
| Absent | 93.8 (17.2) | 5.5 (1.9) | 31.6 (15.5) | 31.1 (13.2) | 34.0 (17.7) | |
| P value | 0.65 | 0.46 | N/A | N/A | N/A | |
| Hypertension/high blood pressure | Yes | 92.5 (21.8) | 5.6 (2.2) | 31.6 (15.0) | 34.6 (16.9) | 27.5 (13.5) |
| No | 94.3 (14.8) | 5.5 (1.8) | 34.9 (14.8) | 30.4 (15.0) | 32.3 (13.1) | |
| P value | 0.34 | 0.47 | 0.07 | 0.80 | 0.09 | |
| Any condition | Present | 84.7 (17.5) | 6.0 (2.4) | 30.1 (11.0) | 27.5 (12.7) | 31.9 (14.5) |
| Absent | 97.9 (14.9) | 5.2 (1.7) | 32.3 (16.6) | 32.4 (13.1) | 35.3 (18.7) | |
| P value | 0.02 | 0.66 | 0.09 | 0.31 | 0.62 | |
COPD indicates chronic obstructive pulmonary disease; CVLT, California Verbal Learning Test; DRS, Disability Rating Scale; FIM, Functional Independence Measure; TMT, Trail Making Test.
The full regression model (Table 5), which was also adjusted for demographic variables, severity of TBI, and functional status at admission, revealed that preexisting heart condition was associated with lower functional outcomes, as measured by a 7.4-point lower discharge FIM score on average (β = −7.37, P = 0.02). Similarly, the presence of preexisting diabetes mellitus/high blood sugar was associated with a 12-point lower discharge FIM score (β = −11.99, P = 0.003).
Table 5.
Multivariate regression analyses of preexisting conditions associated with higher or lower functional and neuropsychological outcomes
| Predictors | Parameter (P value) | ||||
|---|---|---|---|---|---|
| FIM DC | DRS DC | TMT:A | TMT:B | CVLT-II | |
| Heart condition | −7.37 (0.02)* | −5.57 (0.15) | 5.20 (0.64) | 87.98 (0.01)* | −5.48 (0.46) |
| Hypertension | 3.48 (0.30) | 0.07 (0.86) | −6.93 (0.42) | −21.68 (0.38) | −5.35 (0.30) |
| Stroke | −5.25 (0.40) | 0.35 (0.65) | 6.79 (0.61) | 57.56 (0.02)* | 5.67 (0.58) |
| Diabetes/high blood sugar | −11.99 (0.003)* | 0.10 (0.86) | 25.25 (0.02)* | −92.19 (0.40) | −17.33 (0.005)* |
| Cancer | −4.97 (0.47) | 0.53 (0.53) | −1.57 (0.92) | 9.55 (0.83) | −20.43 (0.01)* |
| Liver | −1.62 (0.87) | 0.15 (0.90) | * | * | * |
| Emphysema/asthma/COPD | −5.21 (0.41) | −0.49 (0.52) | −7.62 (0.72) | 159.36 (0.005)* | 15.27 (0.26) |
| 1–3 alcoholic drinks | 0.70 (0.86) | 0.02 (0.96) | −5.68 (0.62) | −0.02 (1.00) | −8.17 (0.26) |
| >3 alcoholic drinks | 5.88 (0.11) | −0.66 (0.15) | 2.78 (0.79) | 48.64 (0.11) | −3.27 (0.61) |
| Illicit drug use | 4.41 (0.28) | −0.40 (0.42) | −0.60 (0.95) | 4.02 (0.88) | −6.05 (0.29) |
| Tobacco/cigarette use | −1.40 (0.64) | −0.15 (0.69) | 0.72 (0.93) | 19.95 (0.42) | −2.01 (0.67) |
Significant: factor associated with lower outcomes. Note: There were no observations with both preexisting liver disease and neuropsychological outcomes.
COPD indicates chronic obstructive pulmonary disease; DC, discharge; NS, outcomes not significant; CVLT-II, California Verbal Learning Test, 2nd edition; DRS, Disability Rating Scale; FIM, Functional Independence Measure; TMT, Trail Making Test.
Individuals with a history of a heart condition (β = 87.98, P = 0.01), stroke (β = 57.56, P = 0.02), or respiratory condition (β = 159.36, P = 0.005) demonstrated reduced cognitive flexibility. As an example, patients with a preexisting respiratory condition took 159 seconds longer on average to complete the TMT:B task (longer completion time on TMT:A and TMT:B equates to poorer performance). Patients with TBI who had preexisting diabetes/high blood sugar demonstrated poorer visual attention, visuomotor processing speed (TMT:A: β = 25.25, P = 0.02), and ability to learn and recall auditory-verbal information (CVLT-II: β = −17.33, P = 0.005). A history of cancer prior to TBI was associated with lower auditory-verbal learning and memory (CVLT-II: β = −20.43, P = 0.01). The final models explained varying magnitudes of variability in the outcomes (using R2 for linear regression): FIM at discharge, 60%; DRS at discharge, 48%; TMT:A, 49%; TMT:B, 69%; and CVLT-II, 43%.
DISCUSSION
After accounting for demographic variables, severity of TBI, and functional status at admission, a history of a preexisting respiratory condition, cancer, and conditions known to be cerebrovascular risk factors (heart condition, stroke, diabetes/high blood sugar) remained significant predictors of functional outcomes and neurocognitive functioning at discharge from inpatient rehabilitation. In other words, worse outcomes were predicted beyond the direct effects of TBI.
The findings of this study relating to functional outcomes are similar to those in the study of Lew, Lee, and Zeiner, despite the fact that some different preexisting conditions and the DRS were additionally examined in the present study (18). Lew and his colleagues found that individuals with preexisting seizure disorder, congestive heart failure, diabetes mellitus, parkinsonism, and chronic low back pain obtained lower total FIM scores at admission and discharge from inpatient rehabilitation (18). These findings are relevant because half of TBI patients with preexisting medical conditions experience acute medical problems during their inpatient rehabilitation stay, which may further compromise gains (18).
Ensuring comprehensive medical treatment of these conditions early in TBI patients' rehabilitation course may impact their functional outcomes. Medical interventions may vary from those previously prescribed to patients due to TBI-induced changes in pressure-volume dynamics that influence the relationship of blood pressure with intracranial pressure changes, and changes in oxygen, mineral, and water metabolism that can adversely affect renal, heart, and immune system functioning (19, 20).
Identification of these patients is also important because these risk factors are predictive of physical dependence and the potential of suffering a superimposed stroke/vascular event following the TBI (21). Cardiovascular risk factors (hypertension, coronary artery disease, diabetes mellitus, smoking, alcohol intake) have been shown to be associated with development of white matter lesions or small subcortical infarcts, which tend to be correlated with reduced premorbid cognition (21–26). Changes in executive function are most pronounced (21, 25). TBI is not currently identified as a risk factor for stroke. However, one trauma database found that 37% of trauma patients are admitted secondary to TBI and, in a 2½-year follow-up period, patients who suffered a TBI had a 31% increased risk of being hospitalized secondary to ischemic stroke over non-TBI trauma patients (27).
Since these specific preexisting conditions have the potential to alter the course of rehabilitation, an important implication is that patients with TBI should be screened for the presence of preexisting medical illness at the time of admission to inpatient rehabilitation. Knowledge of a preexisting respiratory condition will allow rehabilitation care teams to better personalize that patient's care by prompting ancillary referrals. Augmentation of care via aggressive respiratory therapy may enable these patients to make functional gains that would be delayed or lost due to oxygen insufficiency. Receiving information regarding a preexisting history of diabetes/high blood sugar sooner can facilitate immediate placement of a patient on an American Diabetes Association–approved diet. Family members can be trained to routinely check the patient's blood glucose levels and offer snacks depending on their readings, or a glycosylated hemoglobin test can be initiated for cognitively impaired individuals. Additional treatment/therapies for heart or respiratory problems may extend length of stay beyond what is indicated by the patient's payer. This information may play a role in TBI advocacy efforts and may indicate a need for follow-up care following discharge from inpatient rehabilitation.
Evidence that a subset of individuals with TBI experience progressive functional decline in the years following TBI may explain why researchers in TBI and the Institute of Medicine are beginning to consider TBI a chronic health condition (28, 29). These findings highlight a need to study any reversible risk factors that may be associated with further cognitive deterioration in patients with TBI and additional medical illnesses that may develop in the months and years following discharge from inpatient rehabilitation. Early identification and treatment of these medical conditions is an important part of TBI rehabilitation care.
The results should be interpreted in light of the study's limitations. This study involved a small sample size, and neuropsychological test data were available for only about half of the patients. Selection of neuropsychological outcome measures was limited by availability of specific measures and indices in the TBIMS national database. Data entry of one specific CVLT-II index is required of TBIMS centers. A future study examining the relationship between preexisting conditions and other indices of the CVLT-II that measure executive functioning and additional executive measures assessing word generation, problem-solving, and inhibition would be beneficial. Only discharge outcomes were examined, given that neuropsychological variables are collected during inpatient rehabilitation but not at follow-up intervals.
No data were available concerning services received by patients during inpatient rehabilitation and the amount of time spent in each rehabilitation discipline. Therefore, the impact of receiving speech, physical, and occupational therapies on outcomes in this study is not known. This was a single TBIMS center study, and hence the findings may not be generalizable to other rehabilitation centers or to the entire population of individuals with TBI. Patient functional outcome meas-urements varied by the tool used despite moderate correlations between the DRS and FIM (r = −0.58). As noted in a previous study by these authors, this highlights the need for a composite score using multiple existing tools to measure functional outcomes (30). Inclusion of predictors was not exhaustive, and variables such as family support and income were not available but may have influenced the strength of the relationship between preexisting medical illness and outcomes in this study.
The specific preexisting conditions included as predictors were those that were available in the database. Had preexisting conditions such as seizure disorder/epilepsy and pain been available, they could have been included in the analyses as well. Information regarding preexisting psychological illnesses was also unavailable, but it would be interesting to examine their impact on functional and neurocognitive outcomes as well. This study was further limited by a lack of information regarding subtypes and anatomic location of the diseases. It was unknown whether a patient was diagnosed with a brain malignancy versus colon cancer or whether a patient had type 1 versus type 2 diabetes. It was unknown if preexisting conditions were controlled or uncontrolled at the time of injury or were severe, chronic, or reversible.
References
- 1.Dijkers MP. Harrison-Felix C. Marwitz JH. The Traumatic Brain Injury Model Systems: history and contributions to clinical service and research. J Head Trauma Rehabil. 2010;25(2):81–91. doi: 10.1097/HTR.0b013e3181cd3528. [DOI] [PubMed] [Google Scholar]
- 2.Traumatic Brain Injury Model Systems. National Data and Statistical Center. Available at http://www.tbindsc.org; accessed September 15, 2014.
- 3.Penna S. Novack TA. Carlson N. Grote M. Corrigan JD. Hart T. Residence following traumatic brain injury: a longitudinal study. J Head Trauma Rehabil. 2010;25(1):52–60. doi: 10.1097/HTR.0b013e3181c29952. [DOI] [PubMed] [Google Scholar]
- 4.Wright J. Introduction to the FIMTM. Available at http://www.tbims.org/combi/FIM; accessed January 27, 2015.
- 5.Rappaport M. Hall KM. Hopkins K. Belleza T. Cope DN. Disability rating scale for severe head trauma: coma to community. Arch Phys Med Rehabil. 1982;63(3):118–123. [PubMed] [Google Scholar]
- 6.Hester RL. Kinsella GJ. Ong B. McGregor J. Demographic influences on baseline and derived scores from the Trail Making Test in healthy older Australian adults. Clin Neuropsychol. 2005;19(1):45–54. doi: 10.1080/13854040490524137. [DOI] [PubMed] [Google Scholar]
- 7.Wilde EA. Whiteneck GG. Bogner J. Bushnik T. Cifu DX. Dikmen S. French L. Giacino JT. Hart T. Malec JF. Millis SR. Novack TA. Sherer M. Tulsky DS. Vanderploeg RD. von Steinbuechel N. Recommendations for the use of common outcome measures in traumatic brain injury research. Arch Phys Med Rehabil. 2010;91(11):1650–1660.e17. doi: 10.1016/j.apmr.2010.06.033. [DOI] [PubMed] [Google Scholar]
- 8.Reitan RM. Wolfson D. The Halstead-Reitan Neuropsychological Test Battery. Tucson, AZ: Neuropsychology Press; 1985. [Google Scholar]
- 9.Delis D. Kaplan E. Kramer J. Ober B. California Verbal Learning Test-II. San Antonio, TX: Psychological Corporation; 2000. [Google Scholar]
- 10.Brown AW. Malec JF. McClelland RL. Diehl NN. Englander J. Cifu DX. Clinical elements that predict outcome after traumatic brain injury: a prospective multicenter recursive partitioning (decision-tree) analysis. J Neurotrauma. 2005;22(10):1040–1051. doi: 10.1089/neu.2005.22.1040. [DOI] [PubMed] [Google Scholar]
- 11.Cowen TD. Meythaler JM. DeVivo MJ. Ivie CS., III Lebow J. Novack TA. Influence of early variables in traumatic brain injury on functional independence measure scores and rehabilitation length of stay and charges. Arch Phys Med Rehabil. 1995;76(9):797–803. doi: 10.1016/s0003-9993(95)80542-7. [DOI] [PubMed] [Google Scholar]
- 12.Frankel JE. Marwitz JH. Cifu DX. Kreutzer JS. Englander J. Rosenthal M. A follow-up study of older adults with traumatic brain injury: taking into account decreasing length of stay. Arch Phys Med Rehabil. 2006;87(1):57–62. doi: 10.1016/j.apmr.2005.07.309. [DOI] [PubMed] [Google Scholar]
- 13.Lehmkuhl LD. Hall KM. Mann N. Gordon WA. Factors that influence costs and length of stay of persons with traumatic brain injury in acute care and inpatient rehabilitation. J Head Trauma Rehabil. 1993;8(2):88–100. [Google Scholar]
- 14.Ratcliff JJ. Greenspan AI. Goldstein FC. Stringer AY. Bushnik T. Hammond FM. Novack TA. Whyte J. Wright DW. Gender and traumatic brain injury: do the sexes fare differently? Brain Inj. 2007;21(10):1023–1030. doi: 10.1080/02699050701633072. [DOI] [PubMed] [Google Scholar]
- 15.Sandhaug M. Andelic N. Vatne A. Seiler S. Mygland A. Functional level during sub-acute rehabilitation after traumatic brain injury: course and predictors of outcome. Brain Inj. 2010;24(5):740–747. doi: 10.3109/02699051003652849. [DOI] [PubMed] [Google Scholar]
- 16.Schopp LH. Shigaki CL. Johnstone B. Kirkpatrick HA. Gender differences in cognitive and emotional adjustment to traumatic brain injury. J Clin Psychol Med Settings. 2001;8(3):181–188. [Google Scholar]
- 17.National Institute on Alcohol Abuse and Alcoholism. Drinking levels defined. Available at http://www.niaaa.nih.gov/alcohol-health/overview-alcohol-consumption/moderate-binge-drinking; accessed September 15, 2014.
- 18.Lew HL. Lee E. Date ES. Zeiner H. Influence of medical comorbidities and complications on FIM change and length of stay during inpatient rehabilitation. Am J Phy Med Rehabil. 2002;81(11):830–837. doi: 10.1097/00002060-200211000-00005. [DOI] [PubMed] [Google Scholar]
- 19.Bouma GJ. Muizelaar JP. Bandoh K. Marmarou A. Blood pressure and intracranial pressure-volume dynamics in severe head injury: relationship with cerebral blood flow. J Neurosurg. 1992;77(1):15–19. doi: 10.3171/jns.1992.77.1.0015. [DOI] [PubMed] [Google Scholar]
- 20.Taylor RW. Dellinger RP. Preexisting medical problems in the trauma patient: do they matter? Int Anesthesiol Clin. 1987;25(1):143–161. doi: 10.1097/00004311-198702510-00010. [DOI] [PubMed] [Google Scholar]
- 21.Mok VC. Wong A. Lam WW. Fan YH. Tang WK. Kwok T. Hui AC. Wong KS. Cognitive impairment and functional outcome after stroke associated with small vessel disease. J Neurol Neurosurg Psychiatry. 2004;75(4):560–566. doi: 10.1136/jnnp.2003.015107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Breteler MM. van Swieten JC. Bots ML. Grobbee DE. Claus JJ. van den Hout JH. van Harskamp F. Tanghe HL. de Jong PT. van Gijn J. Hofman A. Cerebral white matter lesions, vascular risk factors, and cognitive function in a population-based study: the Rotterdam Study. Neurology. 1994;44(7):1246–1252. doi: 10.1212/wnl.44.7.1246. [DOI] [PubMed] [Google Scholar]
- 23.Debette S. Markus HS. The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis. BMJ. 2010;341:c3666. doi: 10.1136/bmj.c3666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Grysiewicz R. Evaluation of vascular cognitive impairment: to white matter and beyond. American Heart Association: Cardiovascular Daily. 2002;16(1):41–49. [Google Scholar]
- 25.Kramer JH. Reed BR. Mungas D. Weiner MW. Chui HC. Executive dysfunction in subcortical ischaemic vascular disease. J Neurol Neurosurg Psychiatry. 2002;72(2):217–220. doi: 10.1136/jnnp.72.2.217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Schmidt R. Fazekas F. Offenbacher H. Dusek T. Zach E. Reinhart B. Grieshofer P. Freidl W. Eber B. Schumacher M. Koch M. Lechner H. Neuropsychologic correlates of MRI white matter hyperintensities: a study of 150 normal volunteers. Neurology. 1993;43(12):2490–2494. doi: 10.1212/wnl.43.12.2490. [DOI] [PubMed] [Google Scholar]
- 27.Burke JF. Stulc JL. Skolarus LE. Sears ED. Zahuranec DB. Morgenstern LB. Traumatic brain injury may be an independent risk factor for stroke. Neurology. 2013;81(1):33–39. doi: 10.1212/WNL.0b013e318297eecf. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Corrigan JD. Hammond FM. Traumatic brain injury as a chronic health condition. Arch Phys Med Rehabil. 2013;94(6):1199–1201. doi: 10.1016/j.apmr.2013.01.023. [DOI] [PubMed] [Google Scholar]
- 29.Institute of Medicine. Long-Term Consequences of Traumatic Brain Injury. Washington, DC: National Academies Press; 2009. [Google Scholar]
- 30.Dahdah MN. Barisa MT. Schmidt K. Barnes SA. Dubiel R. Dunklin C. Harper C. Callender L. Wilson A. Diaz-Arrastia R. Shafi S. Comparative effectiveness of traumatic brain injury rehabilitation: differential outcomes across TBI model systems centers. J Head Trauma Rehabil. 2014;29(5):451–459. doi: 10.1097/HTR.0b013e3182a61983. [DOI] [PubMed] [Google Scholar]