Self-report of extent of recovery and barriers to recovery after traumatic brain injury: a longitudinal study

OBJECTIVE

To examine the perspective of survivors of traumatic brain injury (TBI) regarding the extent and nature of their recovery over time.

DESIGN

Inception cohort, longitudinal study.

SETTING

Level I trauma center.

PARTICIPANTS

One hundred fifty-seven consecutively hospitalized individuals with TBI (mean age, 36.1 yr; 80% men) with a broad range of injury severity.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Participants reported the extent of their recovery and barriers to full recovery at 1, 6, and 12 months.

RESULTS

Participants reported a median return to normal at the 3 follow-up times of 65%, 80%, and 85%. After 1 month, self-reported extent of recovery correlated well with performance on the Glasgow Outcome Scale (p <.001 at 6 and 12 mo) and Wechsler Adult Intelligence Scale Performance IQ (p =.001 at 12 mo). The major reported barrier to recovery was physical difficulties, which constituted over half of the concerns at all time periods. Report of physical-related concerns decreased significantly (p =.002) over time whereas cognition-related concerns increased significantly (p =.02).

CONCLUSION

Brain injury survivors consider themselves to have greater recovery than previously reported by clinicians or family members, consider physical problems a significant factor in their recovery, and appear to become more aware of cognitive impairments over time.

Utility of a composite measure to detect problematic alcohol use in persons with traumatic brain injury

OBJECTIVES

To examine factors complicating the study of alcohol-related effects in traumatic brain injury (TBI) patients and to evaluate a composite measure to categorize such patients according to degree of alcohol-related problems.

DESIGN

Inception cohort.

SETTING

Level I trauma center.

PATIENTS

Consecutively hospitalized adult TBI patients (n = 156; 73% men; 87% Caucasian; mean age, 30yr; mean education, 12yr). Selection criteria required objective evidence of brain trauma; minimum survival of 1 month postinjury; age 15 years or older; and English speaking.

MAIN OUTCOME MEASURES

An index of problematic drinking based on a measure created by combining blood-alcohol level, quantity-frequency of consumption, and the Short Michigan Alcoholism Screening Test. Preinjury characteristics were obtained through structured interview.

RESULTS

Participants with highly problematic drinking showed poorer premorbid psychosocial functioning, including lower educational attainment, greater likelihood of problems with the law, lower perceived social support, and greater prevalence of other substance abuse.

CONCLUSION

The composite index is useful in identifying problematic drinkers among TBI patients. Results have implications for interpreting and planning research on the role of alcohol in TBI outcomes.

Antiepileptogenesis and seizure prevention trials with antiepileptic drugs: meta-analysis of controlled trials

PURPOSE

To synthesize evidence concerning the effect of antiepileptic drugs (AEDs) for seizure prevention and to contrast their effectiveness for provoked versus unprovoked seizures.

METHODS

Medline, Embase, and The Cochrane Clinical Trials Register were the primary sources of trials, but all trials found were included. Minimal requirements: seizure-prevention outcome given as fraction of cases; AED or control assigned by random or quasi-random mechanism. Single abstracter. Aggregate relative risk and heterogeneity evaluated using Mantel-Haenszel analyses; random effects model used if heterogeneity was significant.

RESULTS

Forty-seven trials evaluated seven drugs or combinations for preventing seizures associated with fever, alcohol, malaria, perinatal asphyxia, contrast media, tumors, craniotomy, and traumatic brain injury. Effective: Phenobarbital for recurrence of febrile seizures [relative risk (RR), 0.51; 95% confidence interval (CI), 0.32-0.82) and cerebral malaria (RR, 0.36; CI, 0.23-0.56). Diazepam for contrast media-associated seizures (RR, 0.10; CI, 0.01-0.79). Phenytoin for provoked seizures after craniotomy or traumatic brain injury (craniotomy: RR, 0.42; CI, 0.25-0.71; TBI: RR, 0.33; CI, 0.19-0.59). Carbamazepine for provoked seizures after traumatic brain injury (RR, 0.39; CI, 0.17-0.92). Lorazepam for alcohol-related seizures (RR, 0.12; CI, 0.04-0.40). More than 25% reduction ruled out valproate for unprovoked seizures after traumatic brain injury (RR, 1.28; CI, 0.76-2.16), and carbamazepine for unprovoked seizures after craniotomy (RR, 1.30; CI, 0.75-2.25).

CONCLUSIONS

Effective or promising results predominate for provoked (acute, symptomatic) seizures. For unprovoked (epileptic) seizures, no drug has been shown to be effective, and some have had a clinically important effect ruled out.

Incidence of intravenous site reactions in neurotrauma patients receiving valproate or phenytoin

OBJECTIVE

To determine the incidence of intravenous site reactions to phenytoin and valproate in a large population of patients with neurotrauma.

DESIGN

Retrospective chart review of two double-blind, randomized clinical trials evaluating the use of antiepileptic drugs to prevent posttraumatic seizures in patients with neurotrauma: phenytoin versus placebo (n = 390), and valproate versus phenytoin with placebo (n = 385). Information collected from the charts included the number, type, and location of intravenous lines and intravenous site events.

SETTING

Tertiary care trauma and university teaching hospital.

MAIN RESULTS

Intravenous site reactions occurred in 18% and 25% of patients receiving valproate or phenytoin, respectively, with the majority of events (70%) occurring in the first intravenous site. Patients received the neurosurgery study drug (NSSD) by either central or peripheral lines; all intravenous site reactions occurred in peripheral administration sites. When patients who received the drug by central line during the course of therapy were excluded, the estimated incidence of site reactions was 21% and 30% for valproate and phenytoin, respectively (p = 0.056). The time to the first event was shorter with phenytoin compared with valproate (2.0 +/- 1.3 vs. 3.0 +/- 1.9 d; p = 0.009). Fewer adverse events were noted with phenytoin in the phenytoin-without-valproate study than in the phenytoin-with-valproate study, with 4.3% and 8.2% of intravenous site events recorded in patients receiving placebo or phenytoin, respectively. There was no significant difference in the number of intravenous lines per patient used during NSSD drug infusion for phenytoin versus placebo or phenytoin versus valproate.

CONCLUSIONS

Both intravenous phenytoin and valproate resulted in intravenous site reactions, with the loading doses responsible for the majority of the events.

Neuropsychological effects of valproate in traumatic brain injury: a randomized trial

OBJECTIVES

To examine the neuropsychological side effects of valproate (VPA) given to prevent posttraumatic seizures.

METHODS

In a randomized, double-masked, parallel group clinical trial, we compared the seizure prevention and neuropsychological effects of 1 or 6 months of VPA to 1 week of phenytoin. We studied 279 adult subjects who were randomized within 24 hours of injury and examined with a battery of neuropsychological measures at 1, 6, and 12 months after injury. We examined drug effects cross-sectionally at 1, 6, and 12 months and longitudinally by examining differential change from 1 to 6 months and from 6 to 12 months as a function of protocol-dictated changes in treatment.

RESULTS

No significant adverse or beneficial neuropsychological effects of VPA were detected.

CONCLUSIONS

Valproate (VPA) appears to have a benign neuropsychological side effects profile, making it a cognitively safe antiepileptic drug to use for controlling established seizures or stabilizing mood. However, based on this study, VPA should not be used for prophylaxis of posttraumatic seizures because it does not prevent posttraumatic seizures, there was a trend toward more deaths in the VPA groups, and it did not have positive effects on cognition.

Side effects and mortality associated with use of phenytoin for early posttraumatic seizure prophylaxis

OBJECT

The goals of this study were to determine if the use of phenytoin to prevent early posttraumatic seizures following head injury was associated with significant adverse side effects and also to determine if the reduction in early posttraumatic seizures after phenytoin administration was associated with a change in mortality rates in head-injured patients.

METHODS

The authors performed a secondary analysis of the data obtained in a prospective double-blind placebo-controlled study of 404 patients who were randomly assigned to receive phenytoin or placebo for the prevention of early and late posttraumatic seizures. The incidence of adverse drug effects during the first 2 weeks of treatment, however, was low and not significantly different between the treated and placebo groups. Hypersensitivity reactions occurred in 0.6% of the patients in the phenytoin-treated group compared with 0% in the placebo group (p = 1.0) during week 1, and in 2.5% of phenytoin-treated compared with 0% of placebo-treated patients (p = 0.12) for the first 2 weeks of treatment. Mortality rates were also similar in both groups. Although the mortality rate was higher in patients who developed seizures, this increase was related to the greater severity of the injuries sustained by these patients at the time of the original trauma.

CONCLUSIONS

The results of this study indicate that the incidence of early posttraumatic seizure can be effectively reduced by prophylactic administration of phenytoin for 1 or 2 weeks without a significant increase in drug-related side effects. Reduction in posttraumatic seizure during the 1st week, however, was not associated with a reduction in the mortality rate.

Valproate therapy for prevention of posttraumatic seizures: a randomized trial

OBJECT

Seizures frequently accompany moderate to severe traumatic brain injury. Phenytoin and carbamazepine are effective in preventing early, but not late, posttraumatic seizures. In this study the authors compare the safety and effectiveness of valproate with those of short-term phenytoin for prevention of seizures following traumatic brain injury.

METHODS

The study was a randomized, double-blind, single-center, parallel-group clinical trial. Treatment began within 24 hours of injury. One hundred thirty-two patients at high risk for seizures were assigned to receive a 1-week course of phenytoin, 120 were assigned to receive a 1-month course of valproate, and 127 were assigned to receive a 6-month course of valproate. The cases were followed for up to 2 years. The rates of early seizures were low and similar when using either valproate or phenytoin (1.5% in the phenytoin treatment group and 4.5% in the valproate arms of the study; p = 0.14, relative risk [RR] = 2.9, 95% confidence interval [CI] 0.7-13.3). The rates of late seizures did not differ among treatment groups (15% in patients receiving the 1-week course of phenytoin, 16% in patients receiving the 1-month course of valproate, and 24% in those receiving the 6-month course of valproate; p = 0.19, RR = 1.4, 95% CI 0.8-2.4). The rates of mortality were not significantly different between treatment groups, but there was a trend toward a higher mortality rate in patients treated with valproate (7.2% in patients receiving phenytoin and 13.4% in those receiving valproate; p = 0.07, RR = 2.0, 95% CI 0.9-4.1). The incidence of serious adverse events, including coagulation problems and liver abnormalities, was similar in phenytoin- and valproate-treated patients.

CONCLUSIONS

Valproate therapy shows no benefit over short-term phenytoin therapy for prevention of early seizures and neither treatment prevents late seizures. There was a trend toward a higher mortality rate among valproate-treated patients. The lack of additional benefit and the potentially higher mortality rate suggest that valproate should not be routinely used for the prevention of posttraumatic seizures.

Detecting significant change in neuropsychological test performance: a comparison of four models

A major use of neuropsychological assessment is to measure changes in functioning over time; that is, to determine whether a difference in test performance indicates a real change in the individual or just chance variation. Using 7 illustrative test measures and retest data from 384 neurologically stable adults, this paper compares different methods of predicting retest scores, and of determining whether observed changes in performance are unusual. The methods include the Reliable Change Index, with and without correction for practice effect, and models based upon simple and multiple regression. For all test variables, the most powerful predictor of follow-up performance was initial performance. Adding demographic variables and overall neuropsychological competence at baseline significantly but slightly improved prediction of all follow-up scores. The simple Reliable Change Index without correction for practice performed least well, with high error rates and large prediction intervals (confidence intervals). Overall prediction accuracy was similar for the other three methods; however, different models produce large differences in predicted scores for some individuals, especially those with extremes of initial test performance, overall competency, or demographics. All 5 measures from the Halstead-Reitan Battery had residual (observed--predicted score) variability that increased with poorer initial performance. Two variables showed significant nonnormality in the distribution of residuals. For accurate prediction with smallest prediction--confidence intervals, we recommend multiple regression models with attention to differential variability and nonnormality of residuals.

Aging effect on psychosocial outcome in traumatic brain injury

OBJECTIVE

To examine the effects of age on outcome in persons with traumatic brain injury.

DESIGN

Longitudinal cohort design.

SETTING

Level I trauma center.

PATIENTS

A total of 411 hospitalized subjects with mild to severe traumatic brain injury prospectively studied to 1 year; their age range was 18 to 89 years.

MAIN OUTCOME MEASURES

Glasgow Outcome Scale, living situation, and employment.

RESULTS

Increasing age is associated with increasing levels of psychosocial limitations, especially in persons 60 years of age and older. Part of the reason is the greater severity of injury sustained by older persons as reflected in longer coma (despite equivalent initial coma depth) and greater numbers of complications and surgeries for subdural hematomas. However, the consequences of traumatic brain injuries appear to worsen with increasing age at each level of brain injury severity examined, including the milder injuries.

CONCLUSIONS

Older adults clearly show less complete recovery 1 year after brain injury than younger adults, either because they have reduced reserves with which to tolerate brain injury or because their physiologic status creates a more destructive injury. Glasgow Coma Scale alone may underestimate the severity of brain injury in the aged as well as its associated consequences. Caution is advised in generalizing findings based principally on younger individuals to older adults with traumatic brain injuries.

Increases in metabolism of valproate and excretion of 6beta-hydroxycortisol in patients with traumatic brain injury

AIMS

The objectives of this study were to determine the effect of brain trauma on the multiple pathways of metabolism of valproate and to evaluate the use of the urinary 6beta-hydroxycortisol to cortisol ratio in predicting changes in hepatic metabolism induced by brain injury.

METHODS

Fourteen patients with severe head injuries received a 15 mg kg(-1) loading dose and a maintenance dose of valproate to maintain therapeutic plasma concentrations. A minimum of one steady state trough blood sample and one dosage interval urine were collected during days 3-6 and during days 7-14 post-injury. Total and unbound valproate plasma concentrations were determined by gas chromatography-flame ionization detection (GC-FID) with and without ultrafiltration. Urinary valproate metabolites were measured by gas chromatography/mass spectrometry (GC-MS) (n = 10). Urinary 6beta-hydroxycortisol and cortisol concentrations were determined by high performance liquid chromatography (h.p.l.c.) (n = 14). Total intrinsic clearance (CL[int]) for valproate and individual formation clearances (CL[f]) to its major metabolites were calculated. Data obtained during baseline (days 3-6) were averaged for each patient and were compared with averaged data obtained from days 7 to 14 using a paired t-test.

RESULTS

Statistically significant increases in the CL(int), CL(f) of VPA glucuronide, 2-ene-VPA, and 4-OH-VPA pathways and the 6beta-hydroxycortisol to cortisol ratio were found. The percent change in the 6beta-hydroxycortisol to cortisol ratio correlated significantly with the changes in the CL(int) of valproate.

CONCLUSIONS

Brain trauma results in induction of multiple pathways of valproate metabolism and increases in the 6beta-hydroxycortisol to cortisol ratio, suggesting a non-specific enzyme induction in response to head injury.

Risk of seizure recurrence after the first late posttraumatic seizure

OBJECTIVE

To determine the incidence and risk factors for seizure recurrence after the onset of late posttraumatic seizures (ie, seizures occurring more than 7 days after injury).

DESIGN

Longitudinal cohort design.

SETTING

Level 1 trauma center.

PATIENTS

Sixty-three moderately to severely head-injured adults who developed late posttraumatic seizures during the course of their participation in a randomized, placebo-controlled study of the effectiveness of prophylactic phenytoin (Dilantin) for prevention of posttraumatic seizures.

MAIN OUTCOME MEASURES

Time from the first unprovoked late seizure to time of seizure recurrence.

RESULTS

The cumulative incidence of recurrent late seizures was 86% by approximately 2 years. However, the frequency of recurrent seizures varied considerably across subjects: 52% experienced at least five late seizures, and 37% had 10 or more late seizures within 2 years of the first late seizure. The relative risk of recurrence was highest in patients with a history of acute subdural hematoma and prolonged coma (ie, longer than 7 days).

CONCLUSIONS

When late seizures develop after severe head injury, the probability of recurrence is high, which suggests that patients be treated aggressively with anticonvulsant medication after a first unprovoked late seizure.

Revised Winter-Tozer equation for normalized phenytoin concentrations in trauma and elderly patients with hypoalbuminemia

OBJECTIVE

To develop a revised equation reflecting the current practice of measuring unbound phenytoin at room temperature, and to evaluate the revised Winter-Tozer method of predicting normalized total phenytoin concentrations in two groups of patients with hypoalbuminemia-elderly nursing home patients and critically ill head trauma patients.

DESIGN

Albumin, unbound phenytoin, and total phenytoin concentrations were obtained from two sources: prospectively from a group of elderly nursing home patients and by a retrospective chart review of trauma patients enrolled in a previous double-blind, placebo-controlled study.

SETTING

Community nursing homes; a university-affiliated urban teaching hospital.

PARTICIPANTS

Elderly nursing home patients (n = 46) taking chronic phenytoin therapy and patients enrolled in a double-blind, placebo-controlled study (n = 58) evaluating the use of phenytoin to prevent posttraumatic seizures.

MAIN OUTCOME MEASURES

Prediction error analysis was performed by using the methods proposed by Sheiner and Beal. Bias and precision were evaluated by calculating the mean prediction error (MPE) and root mean squared error (RMSE), respectively.

RESULTS

The Winter-Tozer equation consistently overpredicted the normalized phenytoin concentration in the elderly nursing home population (MPE = 3.2, RMSE = 5.9) and the trauma patients (MPE = 3.3, RMSE = 4.8). The equation was revised to reflect the increased protein binding of phenytoin with decreased temperature and resulted in significantly decreased bias in both groups of patients.

CONCLUSIONS

The revised equation is useful in predicting normalized phenytoin concentrations in both elderly nursing home patients and critically ill trauma patients.

The impact of posttraumatic seizures on 1-year neuropsychological and psychosocial outcome of head injury

This study examined the relationship of posttraumatic seizures and head injury severity to neuropsychological performance and psychosocial functioning in 210 adults who were prospectively followed and assessed 1 year after moderate to severe traumatic head injury. Eighteen percent (n = 38) of the patients experienced 1 or more late seizures (i.e., seizures occurring 8 or more days posttrauma) by the time of the 1-year followup. As expected, the head injured patients who experienced late posttraumatic seizures were those with the most severe head injuries, and they were significantly more impaired on the neuropsychological and psychosocial measures compared to those who remained seizure free. However, after the effects of head injury severity were controlled, there were no significant differences in neuropsychological and psychosocial outcome at 1 year as a function of having seizures. These findings suggest that worse outcomes in patients who develop posttraumatic seizures up to 1 year posttrauma largely reflect the effects of the brain injuries that cause seizures, rather than the effect of seizures.

Causes, prevention, and treatment of post-traumatic epilepsy

Post-traumatic seizures often occur after severe head injury. Acutely, these seizures complicate management of the head-injured patient by increasing intracranial pressure and causing postictal decreases in level of consciousness. In the long term, epilepsy can have a negative effect on the patient's functioning and integration into society. The more severe the head injury, the more likely that post-traumatic seizures will occur. The risk of late seizures exceeds 30% for patients with penetrating head injury, intracerebral hematoma, subdural hematoma, depressed skull fracture, or seizure within the first week after injury. Late post-traumatic seizures are treated the same as any epileptic seizures of the same type. Phenytoin and carbamazepine are effective in preventing seizures in the first week after head injury, but are not effective in preventing late seizures. Both additional antiepileptic drugs and neuroprotective agents that may lessen the damage that leads to seizures are being investigated to determine if they are effective in preventing the occurrence of post-traumatic epilepsy.

Alcohol use before and after traumatic head injury

STUDY OBJECTIVE

To determine (1) the significance of blood alcohol level in the emergency department in history of alcohol abuse and (2) the significance of habitual alcohol use in head-injured patients before and after injury.

DESIGN

Inception cohort study with 1-year follow-up.

SETTING

Level I trauma center.

PARTICIPANTS

One hundred ninety-seven hospitalized adult head-injury survivors with a broad spectrum of head-injury severity.

RESULTS

Alcohol use and behavioral problems associated with alcohol use were assayed before injury and in the month and year after injury. The patients' blood alcohol levels in the ED were also examined. Preinjury alcohol abuse was frequent; 42% of the subjects were legally intoxicated while in the ED. The amount of drinking and magnitude of reported preinjury alcohol problems decreased soon after the injury but was followed by an increase by 1 year, although the amount of drinking did not return to the preinjury level (P < .0001). Patients with more severe head injuries decreased their drinking more than did those with less severe head injuries. The patients' blood alcohol levels in the ED were a good indicator of the magnitude of their preinjury alcohol problems (r = .51 to .59; each, P < .001).

CONCLUSION

Preinjury habitual alcohol abuse is frequent in head-injured patients. Blood alcohol levels in the ED are indicative of history of problem drinking and might serve as a basis for treatment referral. The first weeks after injury in hospitalized patients may provide an opportunity to begin interventions because head-injured patients drink less at that time.

Classification and regression trees (CART) for prediction of function at 1 year following head trauma

A cohort of 514 hospitalized head-injury survivors was identified based on their injury and 448 (87%) of them were followed for 1 year. Comprehensive neurobehavioral testing was performed 1 month and 1 year after injury. The authors developed predictions of six neuropsychological and two psychosocial outcomes 1 year after head injury. Prediction trees are presented for verbal IQ, Halstead's Impairment Index, and work status at 1 year. Early predictors of neurobehavioral outcome in survivors are similar to previously reported predictors of mortality. Extent (both depth and length) of coma and age are the medical and demographic variables most predictive of late outcome. Adding 1-month scores substantially improves prediction of neuropsychological variables. The classification and regression tree is a useful technique for predicting long-term outcome in patients with head injury. The trees are simple enough to be used in a clinical setting and, especially with 1-month scores, predictions are accurate enough for clinical utility.

One year psychosocial outcome in head injury

Psychosocial outcome at one year post-injury was examined prospectively in 466 hospitalized head-injured subjects, 124 trauma controls, and 88 friend controls. The results indicate that head injury is associated with persistent psychosocial limitations. However, the presence and extent of limitations are related to the demographics of the population injured, to other system injuries sustained in the same accident, and particularly to the severity of the head injury. More severe head injuries are associated with limitations implying greater dependence on others including poorer Glasgow Outcome Scale (GOS) ratings, dependent living, unemployment, low income, and reliance on family and social subsidy systems. Head injury severity is more closely related to more objective indices of psychosocial outcome (e.g., employment) than to self-perceived psychosocial limitations, such as measured by the Sickness Impact Profile (SIP).

Neuropsychological outcome in relation to head injury severity. Contributions of coma length and focal abnormalities

Neuropsychological test performances of 102 consecutive head-injured patients were evaluated at 1 mo and 1 yr after injury. The results of the study indicated that both coma length and the presence of focal abnormalities on computed tomography (CT) scans contribute independently to neuropsychological outcome. The effects of coma length are stronger than the effects of focal abnormalities evident on CT scans and continue to exert a stronger influence on neuropsychological outcome over the year postinjury. These results suggest that the extent of diffuse pathology may be a more important determinant of long-term behavioral outcome than the presence of focal lesions.

Decreased plasma protein binding of valproate in patients with acute head trauma

1. One hundred and ten plasma samples were obtained from 50 patients treated with valproate for prophylaxis of post-traumatic head injuries. The samples were selected to include a wide range of albumin concentrations and were assayed for free and total valproate concentrations. Valproate binding parameters were determined from the Scatchard equation for one binding site using reweighted least squares analysis. 2. Plasma albumin concentrations were measured in 130 patients with head trauma. They started to decrease immediately after trauma, reaching a minimum at 5-7 days of approximately 24% of baseline value and did not return to normal until 1 month. 3. The free fraction of valproate varied six to seven-fold as albumin concentration ranged from 1.5 to 4.8 g 100 ml-1 (218-696 mumol l-1). The mean association constant for binding (Ka) was 0.008 mumol l(-1) and the mean number of binding sites (N) was 2.0. There values were similar to those reported for valproate in otherwise healthy patients with epilepsy. 4. Because of saturable protein binding of valproate, hypoalbuminaemia may necessitate the monitoring of free valproate concentrations to avoid toxicity when valproate is used in patients with acute head injury.

Employment following traumatic head injuries

OBJECTIVE

Determine rates of, and factors predictive of, return to work in patients with civilian traumatic head injuries.

DESIGN

Inception cohort study with 1- to 2-year follow-up.

SETTING

Hospitalized patients in a level I trauma center.

PATIENTS

Three hundred sixty-six hospitalized head-injured subjects who were workers before injury and 95 comparison subjects participated in prospective, longitudinal investigations of employment following head injury. Head-injured and comparison subjects were similar on basic demographics and preinjury employment status. The comparison subjects consisted of patients who sustained traumatic injury to the body but not to the head.

MAIN OUTCOME MEASURE

Time taken to return to work following head injury.

RESULTS

Survival methodology was used for analysis. Whether patients returned to work and when related to both the characteristics of the injured patients (eg, education, preinjury work history), the severity of head injury and associated neuropsychologic problems, and severity of other system injuries. More precise predictions were possible using the multivariate model.

CONCLUSIONS

The present study provides a means of assessing employment potential predictively. This can be useful for clinical and research purposes. The results should be used cautiously and should stimulate discussions of appropriate use of services and resources to meet individual patients' needs.

Psychosocial recovery after head injury

Psychosocial recovery after head injury was prospectively examined at 1 and 12 months postinjury in a group of 102 hospitalized adult head-injured patients representing a broad range of head injury severity. For comparison purposes, 102 friend controls were used. Outcome was assessed with a battery of psychosocial measures including the Sickness Impact Profile, the Head Injury Symptom Checklist, and the Modified Function Status Index. The results indicate that head-injury patients experience significant psychosocial problems (eg, ability to return to work, resume leisure activity, concentrate and remember information, feelings of irritability) at both 1 and 12 months postinjury but these difficulties improve over time. Whereas improvements occur in both psychosocial and physical areas of functioning, improvements are greater in the physical area. The nature and extent of difficulties seen vary as a function of head injury severity, and time from injury to observation. Finally, the results suggest that not all problems reported by head-injured patients are solely related to the injury (eg, irritability, anxiety, fatigue, or headaches).

Electrocorticography in patients with medically intractable temporal lobe seizures. I. Quantification of epileptiform discharges prior to resective surgery

We studied retrospectively the intraoperative preresection electrocorticograms (ECoGs) of 72 patients undergoing surgery for medically intractable, mostly complex partial, temporal lobe seizures (TLS). Quantification of interictal epileptiform discharges (EDs) detected visually at each electrode location in 2 min recording epochs included computations of ED rates (EDs/min) and cumulative voltages (CuVs) (microV/min). Of 6388 EDs, 81% involved the infratemporal surface, 18% the lateral temporal surface and 1% the orbital frontal area. Forty-eight patients (67%) demonstrated multiple (up to 5 or more), temporally independent foci. Dominant foci in medial and lateral infratemporal locations were about equally common and were significantly more frequent than in lateral temporal locations. Rankings of ED CuVs and rates at individual cortical locations defined 4 areas of "relative interictal cortical epileptogenicity." These were arranged in an orderly pattern with the anterior parahippocampal gyrus and the inferomedial surface of the temporal tip displaying the highest and the lateral temporal and posterior infratemporal cortices showing the lowest propensity to the interictal epileptiform discharge. Individual areas were not characterized by distinct clinical seizure manifestations. Preresection ECoGs provide information on the epileptogenic dysfunction that involves most of the temporal lobe of patients with medically intractable TLS.

Electrocorticography in patients with medically intractable temporal lobe seizures. II. Quantification of epileptiform discharges following successive stages of resective surgery

We quantified retrospectively the interictal epileptiform discharges (EDs) detected visually in the electrocorticograms (ECoGs) of 42 patients undergoing successive stages of anterior temporal lobectomy for medically intractable temporal lobe seizures (TLS). Following first resection sparing the hippocampus (H) and the parahippocampal gyrus (PHG), EDs were recorded on both structures in all patients and by far exceeded in amount those on residual lateral infratemporal and lateral temporal cortices. Frequently, EDs occurred apparently simultaneously but with opposite polarities on the H and the PHG, but more complex relationships were also evident in most individuals. These features likely reflected abnormal post-synaptic activity generated at different locations and cortical depths within the H, PHG, or both. Quantification of epileptiform activity and the effects of selective anterior hippocampectomy or parahippocampectomy suggested that both the H and PHG had remarkable epileptogenic potential. Levels of epileptiform activity were not significantly different in the H and PHG and in the H of subjects with and without H sclerosis. After final resection, including the amygdaloid nucleus (AN), anterior H and PHG, interictal EDs were present, although markedly diminished, in 35 patients. Postresection foci were significantly less numerous and extensive, and attained smaller maximal voltages, than did foci before and after first resection.

Long-term treatment with gabapentin for partial epilepsy

Gabapentin was studied as an open-label 'add-on' antiepileptic drug in 35 patients with partial seizures. Follow-up at 6 months, 12 months, 18 months, and 24 months is reported. There was a trend toward improvement in simple (SPS) and complex partial seizures with it reaching significance for SPS at 12 and 24 months and for the weighted combination of seizures at 3 months. Five of nine patients were subsequently successfully converted to gabapentin monotherapy. Of those five, one is now seizure free and three are significantly improved since baseline. One remains with unchanged seizure frequency compared to baseline, but is experiencing less toxicity than at that time. This long-term observation suggests that the short-term effect demonstrated in blinded studies continues and that indeed some patients with refractory epilepsy can be maintained on gabapentin alone. Based on these findings, double-blind monotherapy trials of this drug are presently being conducted.

Effect of statistical methodology on normal limits in nerve conduction studies

Mean +/- 2 standard deviations (SD), which relies on a Gaussian distribution, has traditionally been used to derive normal limits for nerve conduction studies. Our purpose was to examine skew in nerve conduction study (NCS) parameters, and to compare normal limits derived by several alternative methods. We examined 22 NCS parameters from 75 asymptomatic, nondiabetic men (controls). The coefficient of skewness (g1) was significantly positive (P less than 0.10, two-tailed test) in 5 of 8 amplitude and 6 of 8 latency measurements. Transformation reduced g1 in 19 of 22 parameters, and was optimal when g1 was closest to zero. For each measurement, ideal normal limits were defined as mean +/- 2 SD of the optimally transformed data of the control subjects. The percentage of 66 diabetic subjects classified as abnormal by the raw data, but normal by the ideal normal limits, was the positive misclassification rate; while the percentage considered normal by the raw data, but abnormal by the ideal normal limits, was the negative misclassification rate. Mean +/- 2 SD of the raw data produced up to 11% positive misclassifications and 12% negative misclassifications. When the range of observed values was used, up to 6% positive misclassifications and 13% negative misclassifications were found, while the 2.5 or 97.5 percentile values produced up to 10% positive misclassifications and 13% negative misclassifications. We conclude that analyses using the raw data to derive normal limits result in an unacceptable rate of misclassification. Normal limits should be derived from the mean +/- 2 SD of the optimally transformed data.

Management of head injury. Posttraumatic seizures

Posttraumatic seizures are relatively common among patients with severe head injuries, with major risk factors being penetrating head wound, hematoma, depressed skull fracture, and, for late seizures, early seizures. Management of late posttraumatic seizures, if they do develop, follows the treatment of patients with epilepsy. Their treatment should be determined by the type of seizure (i.e., partial or generalized) and the individual responsiveness of the patient to drug therapy. Prophylactic administration of antiepileptic drugs to prevent posttraumatic epilepsy has been frequently tried. The data supports a short-term but not a long-term effect of the most commonly used drug, phenytoin. A decision of whether to use prophylaxis, with what, and for how long needs to consider the likely benefit (i.e., the chance of seizures if untreated and the likelihood that the proposed treatment will substantially reduce that chance) and risk (i.e., medical or behavioral adverse effects) of this treatment strategy.

Neurobehavioral effects of phenytoin prophylaxis of posttraumatic seizures

In order to determine potential negative neurobehavioral effects of phenytoin given to prevent the development of posttraumatic seizures, 244 subjects were randomized to phenytoin or placebo. They received neurobehavioral assessments at 1 and 12 months postinjury while receiving their assigned drug and at 24 months while receiving no drugs. In the severely injured, phenytoin significantly impaired performance at 1 month. No significant differences were found as a function of phenytoin in the moderately injured patients at 1 month or in either severity group at 1 year. Patients who stopped receiving phenytoin according to protocol between 1 and 2 years improved more than corresponding placebo cases on several measures. We conclude that phenytoin has negative cognitive effects. This, combined with lack of evidence for its effectiveness in preventing posttraumatic seizures beyond the first week, raises questions regarding its use for long-term prophylaxis. Our findings do not negate phenytoin's proven efficacy in controlling established seizures nor do they indicate that its cognitive effects are worse than other anticonvulsant drugs.

A randomized, double-blind study of phenytoin for the prevention of post-traumatic seizures

BACKGROUND

Antiepileptic drugs are commonly used to prevent seizures that may follow head trauma. However, previous controlled studies of this practice have been inconclusive.

METHODS

To study further the effectiveness of phenytoin (Dilantin) in preventing post-traumatic seizures, we randomly assigned 404 eligible patients with serious head trauma to treatment with phenytoin (n = 208) or placebo (n = 196) for one year in a double-blind fashion. An intravenous loading dose was given within 24 hours of injury. Serum levels of phenytoin were maintained in the high therapeutic range (3 to 6 mumol of free phenytoin per liter). Follow-up was continued for two years. The primary data analysis was performed according to the intention to treat.

RESULTS

Between drug loading and day 7, 3.6 percent of the patients assigned to phenytoin had seizures, as compared with 14.2 percent of patients assigned to placebo (P less than 0.001; risk ratio, 0.27; 95 percent confidence interval, 0.12 to 0.62). Between day 8 and the end of year 1, 21.5 percent of the phenytoin group and 15.7 percent of the placebo group had seizures; at the end of year 2, the rates were 27.5 percent and 21.1 percent, respectively (P greater than 0.2 for each comparison; risk ratio, 1.20; 95 percent confidence interval, 0.71 to 2.02). This lack of a late effect could not be attributed to differential mortality, low phenytoin levels, or treatment of some early seizures in patients assigned to the placebo group.

CONCLUSIONS

Phenytoin exerts a beneficial effect by reducing seizures only during the first week after severe head injury.

Motor speed is a contaminating factor in evaluating the "cognitive" effects of phenytoin

Many studies suggestive of adverse effects of phenytoin (PHT) on mental abilities have used testing procedures which have timed or motor speed elements. Therefore, to what degree the motor speed element alone may have resulted in attributing adverse higher level intellectual or cognitive effects to PHT instead of the identified construct to be measured (e.g., memory, abstraction, decision making) is not clear. To help distinguish "motor" effects from these more complex "cognitive" effects, neuropsychological data on 70 adult PHT monotherapy patients were reanalyzed. Initially, a series of statistically significant differences favored the low serum level group over the high serum level group in neuropsychologic performance. However, when a simple measure of motor speed (Finger Tapping Test) was covaried out, all statistically significant differences between the groups disappeared. Thus, losses in cognitive abilities could not be associated with PHT even though markedly elevated blood levels had been achieved.

General versus disease-specific measures. Further work on the Sickness Impact Profile for head injury

Three modifications devised to make the Sickness Impact Profile more sensitive to head injury are evaluated in 202 head-injured and 132 general trauma patients 1 month and 12 months after injury. The modifications consist of adding items, deleting nonapplicable items, and reweighting areas of function. Each of the modifications, and especially all three combined, slightly but significantly improve discrimination of head-injured and comparison subjects and increase correlations with neurologic and neuropsychologic severity indexes. These slight improvements occur more often at 12 months than at 1 month and among those without rather than with pre-existing conditions. No improvements are found in the ability to classify patients into subgroups. The modifications fail to make improvements sufficiently large or consistent to provide a practical advantage over the SIP. The standard SIP provides a reasonable measure of psychosocial functioning following head injury. It relates to head injury and other system injury severity and reflects recovery with time. The SIP score relates to emotional functioning even after injury severity has been taken into account. Until other factors, such as emotional status and responses style, are better controlled, little benefit is likely to be obtained from creating disease-specific psychosocial measures.

Comparison of alternative outcome measures for antiepileptic drug trials

Studies to determine the efficacy of antiepileptic drugs often use seizure frequency as an outcome measure. Time to kth seizure (k up to 12) was investigated as an alternative endpoint. Monte Carlo simulations, based on seizure behavior in previous clinical trials, were used to evaluate crossover studies with these endpoints. exhibited the highest power. However, tests on time to the 12th seizure, for a sample size of 50, approached the power of tests on seizure frequency with a sample size of 20. Including patients with less severe epilepsy (two vs four seizures per month) did not change the power of tests on time to the kth seizure and lowered it only moderately for tests on seizure frequency. The simulation methodology presented can be adapted to evaluate other design variations.

Neuropsychologic outcome at one-month postinjury

Neuropsychologic outcome at one-month postinjury was investigated in a group of 102 adult patients with head injury representing a broad spectrum of severity. A group of friends of the patients with head injury was selected for comparison purposes. A comprehensive battery of measures assessing various abilities determined the adequacy of neuropsychological functions. Results support the following conclusions: head injury is associated with early deficits observable on measures assessing a broad spectrum of functions, ranging from simple to complex and motoric to abstraction skills; use of appropriate control groups is essential for determining head injury related deficits; the degree of neuropsychologic deficits depends on the severity of head injury; and the severity indices of time to following commands and depth of coma relate more closely and systematically to adequacy of one-month neuropsychologic outcome than does retrospectively assessed posttraumatic amnesia.

Stress as a risk factor for seizures among adults with epilepsy

This study examines the effects of major life events, daily hassles and uplifts, and daily stress levels as they increase or decrease the risks of having seizures and estimates risk ratios for specific stressors and perceived stress levels. Utilizing a prospective design, 12 adults with severe epilepsy monitored the occurrence of seizures, stressors, and stress levels over a 3-month period. In within-individual analyses, high stress levels and stressful events were associated with more frequent seizures for most participants. The association between higher stress levels and increased seizures was confirmed in group analyses. This study provides empirical evidence of the association between stress and seizures and describes the use of a statistical model that is useful for investigating risk factors as they influence physical and mental illness.

Cognitive-behavioral treatment of depressed affect among epileptics: preliminary findings

Controlled study showed skills training for the management of depression in an epilepsy population to be effective. Cognitive-behavioral methods were utilized in a structured learning format with 13 clinically depressed epileptic "students." Significantly greater reductions in dysphoria/depression as measured by the Depression Adjective Checklist and the Generalized Contentment Scale occurred among Ss in the treatment group than among control group Ss. Significant decreases in anger and anxiety/stress and increases in social activities were noted on the Community Adjustment Questionnaire (CAQ). Similar trends were evident on the Beck Depression Inventory and the CAQ depression scale.

Filtration for free drug level monitoring: carbamazepine and valproic acid

Free carbamazepine and valproic acid monitoring using the EMIT FreeLevel filtration system was evaluated and compared with reference equilibrium dialysis and gas chromatographic (GC) techniques. For carbamazepine, free levels after filtration or dialysis were essentially identical (mean 1.74 vs. 1.77 mg/L, r = 0.940, n = 28, EMIT assay). Free levels were 16% higher by EMIT than by GC, possibly due to cross-reaction with carbamazepine-10,11-epoxide. Free fractions were not significantly different using any combination of filtration or dialysis with EMIT or GC (means 0.24-0.26). There was a significant correlation between epoxide and parent-drug free fractions (r = 0.642). Free fraction varied from 0.20 to 0.41 among 61 patient samples and was independent of total drug concentration. For valproic acid, there was a strong correlation between filtration and dialysis results for free level (r = 0.974) and free fraction (r = 0.892), but filtration values were 6-7% higher. Free fraction was concentration dependent (r = 0.597), and lower free fractions by dialysis were attributed to dilution of total drug concentration. Free fraction varied from 0.01 to 0.14 among 50 patient samples. For carbamazepine and valproic acid the EMIT FreeLevel filtration system compared favorably with equilibrium dialysis, and had the advantage of being rapid.

Program evaluation in epilepsy rehabilitation

The Vocational Services Program of the University of Washington Regional Epilepsy Center is described, and data relating to the first 106 clients who entered the program are examined. A major emphasis of the study was characteristic differences between those clients later competitively employed and program dropouts. Other study purposes related to examining client satisfaction ratings of different aspects of services and establishing whether a relationship existed between seizure occurrences and job loss. Stepwise discriminant function analysis indicated that "months employed in the last 24" appeared to be the stable discriminator between the employed and dropout groups. The group later employed averaged 12 of the prior 24 months in employment, whereas dropouts approximated 7. Associated psychiatric/addictions treatment was initially a key outcome discriminator, but it did not hold up on cross-validation. Subjects were more satisfied with individualized client services (e.g., counseling sessions) than with group activities (e.g., Job Club). As opposed to seizures, emotional/attitudinal difficulties with this rehabilitation population appeared to be the primary reason for job loss. A work adjustment or job station program can be critical for these individuals and others with neurological impairments. The importance of counselor follow-up after the initial job placement is also underscored. This program, with about half of its clients entering unsubsidized jobs, basically replicated the results of other community-oriented epilepsy rehabilitation programs.

The behavioral sequelae of head injury

Twenty patients with mild and, in a few cases, moderately severe head injuries were examined 3 days and 1 month postinjury. The results indicate that these patients, as compared to appropriate controls, show significant neuropsychological difficulties at 3 days, but not at 1 month postinjury. Postconcussional symptoms are endorsed, on the other hand, at both 3 days and 1 month. Absence of significant neuropsychological findings at 1 month is contrary to some of the previous reports. A number of reasons for this discrepancy were discussed. Some of these included: (a) inappropriate controls used in previous research; (2) failure to screen for pre-existing conditions in prior studies, therefore confounding the effects of the injury with pre-injury factors; (3) possible practice effects in our research; and (4) differences in the neuropsychological measures used across different studies.

The Seattle evaluation of computerized drug profiles: effects on prescribing practices and resource use

Since 1979, all outpatient pharmacy transactions at the US Public Health Service Hospital in Seattle have been captured in a computer system which generates a profile of each patient's active and previously used drugs. We conducted a controlled trial in which patients were allocated to profile or no-profile groups while the computer continued to collect data on everyone. In all, 41,572 clinic visits made by 6,186 patients were studied. The incidence of preventable drug-drug interactions and redundancies was very low and was unaffected by profiles. For unclear reasons, prescription of two interacting drugs on the same visit was significantly more common for patients with profiles. The duration of drug-drug interaction episodes was significantly shorter for profile group patients, perhaps due to earlier detection of the error on subsequent visits. Profiles had no effect on prescribing volume or coordination of drug refill and visit schedules, but profile group patients made about 5 per cent fewer clinic visits than those in the no-profile group. In this setting, it appears that the prescribing of interacting or redundant drugs is more often due to inadequate provider knowledge than to inaccessible patient-specific drug data. Prevention of such errors would thus require a more active educational or monitoring program.

Neuropsychological recovery in head injury

Recovery of neuropsychological functions was studied in a group of adults with mild to severe head injuries. The subjects were first examined when alert and then 12 and 18 months following their first testing. The results support the following conclusions: (1) a broad range of early deficits occur representing the diversity of behavioral performances dependent on the brain; (2) improvement following losses occurs in complex as well as in simple neuropsychological functions; (3) on the basis of information available, conclusions regarding when recovery slows are premature; and most important, (4) the degree of initial deficit is a significant determinant of the subsequent amount of recovery and the residual deficits.

Clorazepate and phenobarbital as antiepileptic drugs: a double-blind study

The antiepileptic effect of clorazepate when given with phenytoin was compared, in a randomized double-blind crossover study, to the effect of the standard regimen of phenobarbital plus phenytoin in patients with partial seizures. Thirty of 42 subjects preferred the clorazepate-phenytoin regimen (p less than 0.01). The same number of subjects had fewer seizures while taking clorazepate as had fewer seizures while taking phenobarbital. However, subjects had significantly more toxicity, objective and subjective, on the phenobarbital-phenytoin regimen (p less than 0.01 in both cases). In some subjects, increased toxicity due to phenobarbital outweighed better seizure control, so that clorazepate was preferred. As an add-on antiepileptic drug, clorazepate is well tolerated, effective, and preferred by most patients to phenobarbital.

An objective method for the assessment of psychological and social problems among epileptics

Numerous investigators have identified psychological and social problems among epileptics and in many instances, these appear to be more debilitating than the seizures themselves. However, assessment of these problems has most frequently been done by subjective means and when objective tests have been used, they were almost always developed for and standardized on populations other than epileptics. The development of the Washington Psychosocial Seizure Inventory (WPSI) is presented in this paper. After pilot work, 127 adult epileptics were evaluated for psychosocial problems and they completed the 132-item Inventory. Professional assessment of difficulties was made with respect to family background, emotional adjustment, interpersonal adjustment, vocational adjustment, financial status, adjustment to seizures, and medical management. Finally, an assessment of overall psychosocial functioning was made. Through an item-by-item correlation technique, scales were empirically developed for each of these areas and a profile was produced which gives both the absolute and the relative extents of difficulties for each patient with respect to each area. Potential applications for the WPSI are presented.

An analysis for transient states with application to tumor shrinkage

The evaluation of therapies for chronic diseases is often based on the frequency and/or the duration of improvement. Treated separately, these endpoints may give contradictory impressions of the efficacy of the therapy. We propose a more unified method of summarizing improvement-related data--the probability of being in response, i.e., improved, as a function of time. Although improvement is not the only endpoint considered in most trials and this function will not always provide a clear answer to the question of which treatment has better improvement-related characteristics, it does combine the information on several endpoints usually considered separately into a single easily interpreted item. This function is estimated using the method of maximum likelihood on a distribution-free stochastic model of times to improvement and failure. Censored observations are taken into account. A detailed example using data from a cancer clinical trial is presented.