Computed tomography and magnetic ... - Wiley Online Library

Computed Tomography and Magnetic Resonance Imaging in Mild to Moderate Head Injury: Early and Late Imaging Related to Outcome Joukje van der Naalt, MD,* Joffre M. Hew, MD,† Adriaan H. van Zomeren, PhD,‡ Wim J. Sluiter, PhD,§ and Jan M. Minderhoud, MD, PhD*

Serial magnetic resonance imaging (MRI) and computed tomographic (CT) studies were performed in mild to moderate head injury to evaluate whether early and late imaging have additional value in predicting outcome in this category of patients. During 1-year follow-up of a series of 67 patients, a CT scan on admission was performed together with MRI studies within 1 to 3 months and 6 to 12 months after injury. With CT, intracranial lesions were seen in 62% of patients compared with 44% with early and 19% with late MRI, located predominantly in the frontal and temporal regions. More than half of the lesions revealed with CT resulted in focal atrophy on MRI. Outcome was found to be worse in patients with edema and lesions on CT. Likewise, abnormalities detected with MRI were associated with poor outcome scores. In multiple regression analysis, only lesions in the frontal regions detected with early MRI were found to be predictive of outcome. With late MRI, only focal atrophy in the frontotemporal regions was found to be predictive of outcome. The findings in this study suggests that MRI studies may be valuable for predicting long-term outcome in patients with mild to moderate HI. van der Naalt J, Hew JM, van Zomeren AH, Sluiter WJ, Minderhoud JM. Computed tomography and magnetic resonance imaging in mild to moderate head injury: early and late imaging related to outcome. Ann Neurol 1999;46:70 –78

Head injury (HI) is one of the most common neurological causes of morbidity and mortality, especially affecting young adults. Most patients (85–90%) sustain a mild or moderate HI. Most of these patients recover within weeks to months without specific therapy. However, a subgroup of patients continues to have residual symptoms interfering with outcome and return to work.1– 4 In contrast to the well-documented studies in severe HI,5–10 few follow-up studies have been performed concerning prognostic factors and outcome in this category of patients. The most commonly used imaging technique in the acute phase of HI is computed tomography (CT) because of more accurate detection of parenchymal and subarachnoid hemorrhage. In severe HI, outcome is affected by the type of intracranial lesion demonstrated by CT on admission.11–13 Less is known, however, about the prognostic value of CT for outcome in mild to moderate HI. In approximately 10 to 40% of patients in this category, abnormalities are reported.14,15

The frequency of CT abnormalities increases significantly in patients with a lower Glasgow Coma Scale16 (GCS). However, routine early CT scanning is seldom performed in all patients, limiting conclusions concerning the relation of abnormalities with outcome.1,2,17,18 Magnetic resonance imaging (MRI) is found to be more sensitive for detecting smaller lesions or nonhemorrhagic contusions compared with CT scanning.19 –21 In severe HI more abnormalities were diagnosed by MRI in the early and late posttraumatic period.22,23 Depth of lesions revealed by MRI was found to be related to degree of unconsciousness and outcome.24,25 In mild to moderate HI, imaging abnormalities only have been related to neurobehavioral sequelae. Serial MRI scanning demonstrated resolution of lesions with simultaneous recovery of neuropsychological data.26,27 In a recent study,28 attention was drawn to good recovery of patients despite the presence of lesions on MRI scanning. It was suggested that either some lesions have minimal functional impact, or that commonly used out-

From the Departments of *Neurology, †Radiology, ‡Neuropsychology, and §Endocrinology, University Hospital, Groningen, The Netherlands.

Address correspondence to Dr van der Naalt, Department of Neurology, University Hospital Groningen, PO Box 30.001, 9700RB Groningen, The Netherlands.

Received Oct 8, 1998, and in revised form Feb 22, 1999. Accepted for publication Feb 23, 1999.

70

Copyright © 1999 by the American Neurological Association

come scales are too insensitive to reflect the impact of MRI lesions in this category of patients.29,30 So far, few studies have investigated the prognostic value of CT and serial MRI scanning in mild and moderate HI, regarding outcome and return to work. The aim of this study was (1) to investigate whether a relation exists between abnormalities revealed by early CT and late MRI scanning in mild to moderate HI, and (2) to evaluate whether early and late imaging have additional value in predicting outcome in this category of patients. A more detailed outcome scale was used along with the Glasgow Outcome Scale (GOS). Patients and Methods Patients Sixty-seven patients (mean age, 33.2 years; SD, 14.7 years) who met criteria were included in the study. The outcome of this patient population has been reported previously.31 Inclusion criteria were age between 15 and 65 years, GCS score on admission of 9 to 14, and duration of posttraumatic amnesia (PTA) of at least 1 hour. Patients with PTA for more than 28 days were excluded. Patients with a history of previous hospitalization for HI, addiction to alcohol or drugs, known psychiatric disorder, or mental retardation were excluded. Forty-three patients sustained a mild HI (GCS on admission of 13–14) and 24 patients sustained a moderate HI (GCS on admission of 9 –12). The GCS score at admission was assessed as part of the neurological examination. Registration of PTA was done by means of a questionnaire adapted for use by nursing staff, twice daily. Once the maximum score was obtained, the patient was regarded to be out of PTA. By definition, this is equal to the period from injury to the moment when the patient has continuous memory for ongoing events.32

Imaging Studies CT scan acutely after admission was performed by using a Philips Tomoscan 350 (Eindhoven, Holland), 9-mm slices in transaxial plane. MRI studies were performed with a Philips Gyroscan 1.5 T. Images were obtained in 10-mm slices in coronal and transaxial planes, using two sequences, a T1weighted sequence (repetition time [TR], 650 msec; echo time [TE], 20 msec) and a T2-weighted sequence (TR 2,150 msec; TE 40/110 msec). The images were reviewed by an experienced neuroradiologist (J.M.H.) without knowledge of the severity of injury or outcome of the patients. Intracranial abnormalities were coded as lesions on CT scanning if a hemorrhagic lesion with or without surrounding edema was present. Lesions on MRI scanning were coded as hyperintensities on T1- and T2-weighted images. Depth of lesions was coded as absent (0), cortical (1), subcortical (2), or cortical/ subcortical (3). Size of lesions was estimated by visual inspection and coded as absent (0), less than 1 cm (1), 1 to 3 cm (2), or more than 3 cm (3). In lesions with surrounding edema, the constituent parts were summed together and the largest size was noted. Focal atrophy was judged by visual inspection of ventricles and sulci by comparison of the contralateral site and the expected size of a person of that age. In view of the predominance of focal abnormalities in the fron-

tal and temporal regions, subgroups of HI patients were formed according to sites of lesions on CT. The frontal and temporal groups included patients with lesions confined to one of these regions, whereas patients with overlapping lesions were classified as frontotemporal. A CT scan was performed in 55 patients. In 12 patients, a CT was not performed because the patient was out of PTA within several hours after admission. MRI scanning was performed in 63 patients. In 4 patients, MRI studies were not possible because of claustrophobia. (Early) MRI studies were done within 1 to 3 months (mean, 45 days) after injury. A second (late) MRI was done 6 to 12 months after injury (mean, 234 days), only if the early MRI was abnormal, that is, if intracerebral or extracerebral lesions were present.

Outcome At 1 year after injury the outcome was determined by the extended GOS,33 which comprises eight outcome categories: 8, good recovery; 7, good recovery with minor physical or mental deficits; 6, moderate disability, return to previous work with some adjustments or, 5, work at a lower level of performance; 4, severe disability, for some activities dependent on others or, 3, completely dependent on others; 2, vegetative state; and 1, death. The use of the extended GOS has been recommended because patients with a mild or moderate HI in general end up with a good recovery or moderate disability and this scale describes this upper range of outcome in more detail. To improve the assessment of outcome, a more differentiated outcome scale (Differential Outcome Scale, DOS) has been devised that specifies outcome in four subscales regarding social, behavioral, cognitive, and physical sequelae.31 In each category of outcome a 5-point scale is applied. The range of scores varies from 4 (vegetative state) to 20 (good recovery). Resumption of work or previous study (return to work, RTW) was also scored 1 year after injury, according to the following four categories: 0, previous work/study resumed; 1, previous work/study resumed, but with lower demands or part-time; 2, previous work/study not resumed, different work/study on a significantly lower level; and 3, not working/studying.

Statistical Analysis All data were analyzed with the Statistical Package for the Social Sciences (SPSS). Parametric or nonparametric tests were used for statistical analysis, when appropriate. Frequency analysis was performed by x2 analysis, with correction for continuity. Multiple regression analysis was performed by using a stepwise backward method, with the PIN (probability of F-to-enter) and the POUT (probability of F-to-remove) values set at 0.05 and 0.1, respectively.

Results Distribution of Abnormalities With CT scanning a total of 55 intracranial lesions were identified in 34 (62%) of the 55 patients. In more than half of the patients, diffuse edema was seen. Although 1 in 4 patients had additional lesions (subarachnoidal blood, and extradural or subdural hemato-

van der Naalt et al: Imaging in Mild to Moderate HI

71

ma), only 3 patients required neurosurgical intervention (Table 1). With early MRI scanning a total of 44 intracranial lesions were identified in 28 patients (44%). Focal atrophy was seen in about one-third of patients. Less frequent abnormalities consisted of (diffuse) white matter abnormalities and extracerebral abnormalities, mainly subdural blood. On MRI and CT scanning, the number of lesions varied from one to five per patient, and were located predominantly in the frontal and temporal regions (see Table 1). On late MRI scanning, intracranial lesions were still present in 1 of 5 patients. By that time, in approximately one-half the patients focal atrophy was seen. Global Outcome Outcome assessed with the extended GOS revealed good recovery (33%), mild complaints and disturbances not interfering with daily activities or work (49%), and moderate disability (18%). Assessment of outcome with a more detailed outcome scale (DOS) revealed good outcome (31%), disturbances in only one subscale (28%), and disturbances in two or more subscales (41%). Behavioral (48%) and cognitive (40%) problems were most frequently seen. Physical disability mainly concerned cranial nerve dysfunction or posttraumatic epilepsy. None of the patients was severely disabled. One year after injury, 73% of patients had resumed previous jobs or study. Consequently, 1 in 4 patients was able to resume previous activities only partially or at a significantly lower level. CT Abnormalities and Outcome In Table 2, outcome scores are related to CT abnormalities. In patients with lesions, outcome (GOS scores) tended to be worse compared with patients without abnormalities (x2 5 5.8, df 5 2, p , 0.10). Likewise, edema tended to be associated with poor

GOS scores (x2 5 5.3, df 5 2, p , 0.10). With application of a more detailed outcome scale (DOS), the presence of both categories of abnormalities became significant (x2 5 8.2 and 7.3, respectively; both df 5 2, p , 0.05). Likewise, fewer patients were able to resume previous activities completely when edema or lesions were revealed on CT, although this was not statistically significant. Outcome was found to be related to the number of lesions, size, and depth of lesions (data not shown). As shown in Table 3, lesions localized frontotemporally were associated with poor outcome. Furthermore, a comparison was made between the outcome of subcategories of patients. Patients with a GCS of 13 to 14 with or without CT lesions were compared with patients with a GCS of 9 to 12. As shown in Table 4, the best outcome is seen in patients with a GCS of 13 to 14 without CT abnormalities. x2 analysis was significant for GOS scores (x2 5 9.7, df 5 4, p , 0.05) and DOS scores (x2 5 10.3, df 5 4, p , 0.05), but not for RTW. The outcome of patients with a GCS of 13 to 14 with CT abnormalities is comparable with patients with a GCS of 9 to 12, who all showed lesions. MRI Abnormalities Within 1 to 3 Months after Injury and Outcome Presence of lesions or focal atrophy on early MRI was weakly associated with poor outcome, scored by GOS (x2 5 5.2, df 5 2, p , 0.10). Likewise, in patients who were not able to resume previous work completely, an increased frequency of lesions and atrophy was seen (see Table 2). With application of the DOS, the presence of lesions as well as focal atrophy was significantly associated with poorer outcome (x2 5 14.0 and 8.0, respectively; df 5 2, p , 0.025). The number of lesions, and depth and size of lesions were associated with poor outcome (data not shown). As shown in Ta-

Table 1. Distribution of Abnormalities and Location of Lesions Detected with CT and MRI

Distribution of abnormalities Edema Intracerebral lesions Focal atrophy Extracerebral abnormalities Location of lesions Frontal Frontotemporal Temporal Occipital Total lesions

CT (55 Patients)

MRI (1–3 mo) (63 Patients)

MRI (6–12 mo) (63 Patientsa)

32 (58)a 34 (62) 2 (4) 14 (25)

5 (8) 28 (44) 23 (37) 7 (11)

— — 12 (19) 32 (51) 1 (2)

20 17 13 5 55

14 17 7 6 44

6 5 5 2 18

Percentages of total patients are in parentheses. In 32 patients no second MRI was done, because early MRI was normal (n 5 26) or already showed only focal atrophy (n 5 6) and no changes were expected to be seen on late MRI.

a

CT 5 computed tomography; MRI 5 magnetic resonance imaging.

72

Annals of Neurology

Vol 46

No 1

July 1999

Table 2. Abnormalities on CT and MRI in Relation to Outcome (Expressed as Percentages of Patients in Subcategories of Outcome)

CT First MRI Second MRI

GOS

DOS

CT or MRI Abnormality

8

7

#6

20

19

#18

RTW

Lesion Edema Normal Lesion Atrophy Normal Lesion Atrophy Normal

24 25 56 21 30 48 25 25 45

53 50 38 50 44 41 50 47 45

24 25 6 29 26 11 25 28 10

21 22 56 21 26 48 25 22 45

21 22 25 14 22 37 — 25 35

58a 56a 19 65a 52a 15 75a 53a 20

64 65 81 52a 65 89 42a 61 86

p , 0.05, by x2 test.

a

CT 5 computed tomography; MRI 5 magnetic resonance imaging; GOS 5 Glasgow Outcome Scale; DOS 5 Differential Outcome Scale; RTW 5 return to work (percentages of patients who resumed work completely).

Table 3. Mean Outcome Scores in Patients With (1) or Without (2) Abnormalities on CT and MRI a Mean GOS Abnormalities CT (lesions) First MRI (lesions) Second MRI (atrophy)

F FT T F FT T F FT T

Mean DOS

Mean RTW

1

2

1

2

1

2

7.12 6.80b 7.00 6.75b 6.93 6.86 7.11 6.74b 7.10

7.14 7.23 7.17 7.22 7.19 7.17 7.14 7.29 7.14

18.24 17.47b 18.23 17.50b 17.93 17.57 18.44 17.47b 18.40

18.54 18.75 18.52 18.67 18.61 18.57 18.46 18.85 18.47

0.32 0.60b 0.38 0.75b 0.47 0.57 0.33 0.58b 0.30

0.29 0.23 0.30 0.22 0.27 0.28 0.31 0.21 0.32

a

Only lesions on early MRI and atrophy on late MRI are presented. p , 0.05, t test.

b

CT 5 computed tomography; MRI 5 magnetic resonance imaging; F 5 frontal; FT 5 frontotemporal; T 5 temporal; GOS 5 Glasgow Outcome Scale; DOS 5 Differential Outcome Scale; RTW 5 return to work.

ble 3, lesions localized frontally were associated with poor outcome. MRI Abnormalities within 6 to 12 Months after Injury and Outcome Presence of lesions or focal atrophy on late MRI was not significantly related with poor outcome as scored by GOS. With application of the DOS, both the presence of lesions and focal atrophy were significantly associated with poor outcome scores (x2 5 9.7, df 5 2 [p , 0.01] and x2 5 7.2, df 5 2 [p , 0.05], respectively). Twice as many patients were moderately disabled (according to the GOS scores) if either lesions or atrophy were present on MRI. Only 42% of patients with lesions were able to resume previous work completely compared with 86% of patients with a normal MRI (x2 5 5.8, df 5 1, p , 0.025). As shown in Table 3, focal atrophy localized frontotemporally was associated with poor outcome.

Comparison of Detection of Lesions with CT and MRI In 51 patients both CT and MRI scans were obtained, allowing comparison of detected abnormalities. With CT scanning, a total of 50 lesions were detected in 31 of these patients (61%). With early MRI scanning, lesions were diagnosed in three-fourths of patients with CT abnormalities. In 5 patients with normal CT on admission, lesions were detected with MRI (25%). These “new” lesions were all situated in the frontotemporal region, at the base of the skull. When comparing neuroanatomical localization of lesions on CT and MRI, 91% agreement was found between localization of lesions (Table 5). With late MRI, in 25% of patients with lesions on CT, still abnormalities were demonstrated. Likewise, a comparison was made between lesions detected with CT on admission and focal atrophy detected with early and late MRI. Almost half the lesions


73

Table 4. Comparison of Outcome in Subcategories of Patients Based on GCS and Presence of CT Lesions (Expressed as Percentages of Patients) Subcategories of CT Lesions Outcome

Characteristics

13–14 CT2 (n 5 14)

13–14 CT1 (n 5 18)

9–12 (n 5 24)

GOSa

8 7 6 20 19 #18

64 36 0 64 21 15 86

22 56 22 22 28 50 67

25 46 29 21 21 58 64

DOSa RTW

p , 0.05, by x2 test.

a

GCS 5 Glasgow Coma Scale; CT 5 computed tomography; GOS 5 Glasgow Outcome Scale; RTW 5 return to work (percentages of patients who resumed work completely); DOS 5 Differential Outcome Scale; CT1 5 CT with lesions; CT2 5 CT without lesions.

Table 5. Location of Lesions Detected With CT Compared with Lesions on MRI and Local Atrophy on MRI CT Lesions Compared with lesions on MRIa Frontal Frontotemporal Temporal Occipital Total Compared with local atrophy on MRIb Frontal Frontotemporal Temporal Occipital Total

Frontal

Frontotemporal

Temporal

Occipital

No Lesions

Total

10 1 — — 11

2 11 — — 13

— — 6 — 6

— — — 3 3

6 4 5 2 17

18 16 11 5 50

5 — 1 — 6

4 12 — — 16

2 2 6 — 10

— — — 1 1

7 2 4 4 17

18 16 11 5 50

a,b

Performed a1 to 3 months or b6 to 12 months after injury (in 31 patients).

CT 5 computed tomography; MRI 5 magnetic resonance imaging.

detected with CT on admission did result in focal atrophy detected with early MRI. With late MRI scanning, focal atrophy was seen in three-fourths of these patients (see Table 5). Agreement between localization of lesions and atrophy was increased (73%) compared with early MRI (28%). A tendency of frontal lesions to evolve into frontotemporal atrophy was seen. Thus, 6 to 12 months after injury, in patients with an abnormal CT on admission, MRI lesions still were seen in 25% of patients whereas in 75% of those patients focal atrophy was present. CT and MRI Abnormalities and Outcome In Table 6, the outcome of subgroups of patients is presented. atients were categorized whether lesions were detected with CT or first MRI. As a result, the following four groups of patients were formed: Group 1, CT and MRI normal (15 patients); Group 2, CT abnormal/MRI normal (8 patients); Group 3, CT normal/MRI abnormal (6 patients); and Group 4, CT and MRI abnormal (23 patients). In patients without imaging abnormali-

74

Annals of Neurology

Vol 46

No 1

July 1999

ties, the best prognosis was seen (GOS scores: x2 5 11.5, df 5 6, p , 0.10; and DOS scores: x2 5 20.8, df 5 6, p , 0.01). Most resumed work, and few were moderately disabled. In patients with abnormalities on both CT and MRI scanning, worst outcome was seen with moderate disability in one-third of patients. In this subgroup half the patients were not able to resume work completely. Although the numbers of patients in some categories were small, there was a tendency that with MRI abnormalities poorer outcome was seen, irrespective of the presence of CT abnormalities (Groups 3 and 4). Multiple Regression Analysis A stepwise multiple regression analysis was performed to determine factors of importance in predicting outcome. In univariate analysis, duration of PTA explained 21% of the variance in GOS scores and 30% of variance in the DOS scores. The presence of edema, lesions, or atrophy explained approximately 10% of the variance of outcome scores. Other variables like GCS

Table 6. Outcome Related to the Presence of CT and MRI Abnormalities (Expressed as Percentages of Patients)

CT MRI GOS DOS RTW

8 7 6 20a 19 #18

Group 1 (n 5 15)

Group 2 (n 5 8)

Group 3 (n 5 6)

Group 4 (n 5 23)

Normal Normal

Abnormal Normal

Normal Abnormal

Abnormal Abnormal

67 27 7 67 27 7 93

25 63 13 13 50 37 88

17 67 17 17 33 50 50

22 48 30 22 9 69 50

p , 0.05, by x2 test.

a

CT 5 computed tomography; MRI 5 magnetic resonance imaging; GOS 5 Glasgow Outcome Scale; DOS 5 Differential Outcome Scale; RTW 5 return to work (percentages of patients who resumed work completely).

on admission, age, sex, or education did not account for more than 1% of the remaining variance, and none reached statistical significance. Despite the association of the presence of imaging abnormalities with outcome, these variables did not independently contribute to outcome (GOS or DOS scores) as long as PTA was part of the regression model. This applied to abnormalities on CT as well as on MRI. When localization of these imaging abnormalities was entered as a separate variable, however, frontal lesions on early MRI were a significant prognostic factor independent from PTA for DOS scores (r 2 5 0.41), but not for GOS scores. Frontotemporal atrophy on late MRI was also found to be an independent factor of outcome for DOS scores (r 2 5 0.40), but not for the GOS scores (Fig). Discussion In this study, the results of early imaging with CT were related to late imaging with MRI and both methods were examined as predictors of outcome in mild to moderate HI. Although these two imaging methods are not comparable, it seems that each of them can provide valuable information regarding outcome in this category of patients. With CT scanning on admission in 62% of patients, intracranial lesions were diagnosed. This incidence is comparable with findings of other studies.27,28 MRI studies performed several days after injury revealed almost twice as many abnormalities compared with CT.20,26 In the present study, MRI obtained within 1 to 3 months after injury revealed lesions in 44% of patients, mainly confined to the frontal and temporal regions. Furthermore, in one-third of patients focal atrophy was seen. The frequency of focal atrophy in this category of HI patients was surprisingly high. Focal atrophy has not been described separately in previous studies, allowing no comparison with other data. Only in severe HI, frontal and temporal tissue loss has been

described with late MRI in the absence of focal brain lesions.22,35 Presumably, cortical contusions cause eventually more focal atrophy in mild to moderate HI as opposed to the diffuse atrophy seen in severe HI, which results from diffuse axonal injury. In this study, it was decided not to evaluate the effect of diffuse atrophy. Besides the methodological difficulties, diffuse atrophy is expected to be rare in this category of HI patients. The presence of lesions and focal atrophy was found to be related to outcome. Patients with lesions on CT scanning were more often moderately disabled or had problems with resumption of work (RTW). Even in patients with mild HI, the presence of lesions resulted in worse outcome comparable with moderate HI, in accordance with other studies.26,34 Likewise, patients with lesions or focal atrophy on MRI more often were moderately disabled. Not only the presence of lesions or atrophy, but also the localization of these abnormalities was found to be related to outcome. In the present study, with CT scanning, a tendency to poorer outcome was seen in patients with lesions in the frontotemporal regions. However, with MRI performed 1 to 3 months after injury, frontal regions were more often affected in patients with poorer outcome. This apparent change of position of abnormalities is in accordance with earlier studies. In serial MRI scanning, hyperintensities became more localized and seemed to change position to adjacent areas, for example, frontotemporal lesions changed to frontal or temporal lesions.27 Apparently, focal lesions on MRI result in more extensive (frontotemporal) atrophy fully apparent on late MRI, as demonstrated in the present study. With multiple regression analysis, the contribution of different variables on outcome was analyzed. Although lesions and edema on CT scanning were found to be of influence on outcome in univariate analysis, significance was lost in combination with PTA. This


75

Fig. Imaging studies of a 49-year-old patient after a motor vehicle accident (Glasgow Coma Scale 5 14; posttraumatic amnesia 5 18 days). One year after injury the patient was moderately disabled (Glasgow Outcome Scale 5 6; differentiated outcome scale 5 16). (A) Computed tomographic scanning on admission revealed diffuse edema. Magnetic resonance imaging (MRI) studies 2 months after injury revealed a lesion of low intensity on T1 image (arrow in B) and a hyperintense lesion in the right temporal lobe with a hypointense center on T2 image (arrow in C ), suggestive of temporal lobe contusion with edema. In the same area slight focal atrophy is seen. MRI studies 9 months after injury showed increased focal atrophy on T1 image (D) and disappearance of the focal lesion in the right temporal lobe on T2 image (E).

confirms an earlier study36 where type of brain lesion on CT scanning was found to be an important prognostic factor for early outcome, but not for late outcome. Furthermore, in the present study, a tendency to poorer outcome was seen when MRI lesions were diagnosed, irrespective of the presence of CT lesions. However, with CT scanning, the frequency of lesions was underestimated, as suggested by the fact that MRI revealed lesions in 25% of patients with normal CT on admission. No relation with outcome was found for level of consciousness, as reflected by the GCS, in accordance with other studies.8,10 These findings support the idea that GCS and PTA reflect different aspects of brain damage. In severe HI, the relation of the GCS to depth of lesions is regarded as compatible with more widespread diffuse damage.37 In mild to moderate HI, more focal contusions are seen. As a result, in these less severe injuries, increased duration of PTA could be a signal of hemispheric damage more than a lower level of consciousness.38 As stated in a previous study,31 in severe HI outcome is determined particularly by the

76

Annals of Neurology

Vol 46

No 1

July 1999

M-score of the GCS. In mild to moderate HI, the V-score in general reaches its maximum later than the M-score. Therefore, it seems plausible that outcome in this category of patients is determined by the V-score, and thus by the PTA instead of the GCS as opposed to severe HI. In the present study, only frontal lesions disclosed by early MRI in combination with duration of PTA were found to be predictive of outcome. With late MRI scanning, only frontotemporal atrophy in combination with duration of PTA was found to be predictive of outcome. It is suggested that a link is present between the presence of imaging abnormalities and PTA in head trauma outcome. To explain the relationship of injury severity and PTA it could be assumed that PTA is related to damage of specific areas, such as the temporal or frontal regions. So far, the precise relation remains unclear. However, it does not seem accurate to use the presence of imaging abnormalities as a surrogate marker for PTA. A clinical marker like PTA cannot be replaced by imaging techniques. First, MRI

does not detect all lesions. Patients could be in PTA without imaging abnormalities. Second, lesions and edema reflect different aspects of brain damage and are inconsistently related to duration of PTA. From the clinician’s point of view, an essential question would be, does the assessment of CT and MRI have additional value over PTA, in providing prognostic information in patients with mild to moderate HI? In general, neurologists are often consulted by patients several months after injury when problems with resumption of work (RTW) are encountered. Therefore, results of imaging studies obtained during this period could give valuable information. First, knowledge of duration of PTA is important. MRI will have additional value either by disclosing focal lesions or atrophy. Outcome, however, will have to be determined with a more detailed outcome scale to measure the more subtle influence of various parameters on outcome. In this study, only a few of the various factors determining outcome have been studied. It should be emphasized that these data cannot provide predictions of outcome for individual patients. Group data, however, can be used to formulate probabilities of outcome. Further research must reveal whether data from neuropsychological tests or results from new imaging procedures (diffusion-weighted MRI,39 FLAIR-MRI,40 and positron emission tomography41) will have additional value in predicting outcome in mild to moderate HI patients. References 1. Rimel RW, Giordani B, Barth JT, et al. Disability caused by minor head injury. Neurosurgery 1981;9:221–228 2. Rimel RW, Giordani B, Barth JT, Jane JA. Moderate head injury: completing the clinical spectrum of brain trauma. Neurosurgery 1982;11:344 –351 3. Englander J, Hall K, Stimpsons T, et al. Mild traumatic injury in an insured population: subjective complaints and return to employment. Brain Injury 1992;6:161–166 4. Stambrook M, Moore AD, Peters LC, et al. Effects of mild, moderate and severe closed head injury on long-term vocational status. Brain Injury 1990;4:183–190 5. Jennett B, Teasdale G, Galbraith S, et al. Severe head injury in three countries. J Neurol Neurosurg Psychiatry 1977;40:291– 298 6. Brooks DN, Aughton ME, Bond MR, et al. Cognitive sequelae in relationship to early indices of severity of brain damage after severe blunt head injury. J Neurol Neurosurg Psychiatry 1980; 43:529 –534 7. Marshall LF, Gautille T, Klauber MR, et al. The outcome of severe closed head injury. J Neurosurg 1991;75:S28 –S37 8. Bishara SN, Partridge FM, Godfrey HPD, Knight RG. Posttraumatic amnesia and Glasgow Coma Scale related to outcome in survivors in a consecutive series of patients with severe closed head injury. Brain Injury 1992;6:373–380 9. Karzmark P. Prediction of long-term cognitive outcome of brain injury with neuropsychological, severity of injury, and demographic data. Brain Injury 1992;6:213–217 10. Katz DI, Alexander MP. Traumatic brain injury: predicting course of recovery and outcome for patients admitted to rehabilitation. Arch Neurol 1994;51:661– 670

11. Genarelli TA, Spielman GM, Langfitt TW, et al. Influence of the type of intracranial lesion on outcome from severe head injury: a multicenter study using a new classification system. J Neurosurg 1982;56:26 –32 12. Lobato RD, Cordobes F, Rivas JJ, et al. Outcome from severe head injury related to the type of intracranial lesion: a computerized tomography study. J Neurosurg 1983;59:762–774 13. Marshall LF, Bowers Marshall S, Klauber MR, et al. A new classification of head injury based on computerized tomography. J Neurosurg 1991;75:S14 –S20 14. Stein SC, Ross SE. Moderate head injury: a guide to initial management. J Neurosurg 1992;77:562–564 15. Ingebrigtsen T, Romner B. Routine early CT-scan is cost saving after minor head injury. Acta Neurol Scand 1996;93:207–210 16. Hsiang JNK, Yeung T, Yu ALM, Poon WS. High-risk mild head injury. J Neurosurg 1997;87:234 –238 17. Feuerman T, Wackym PA, Gade GF, Becker DP. Value of skull radiography, head computed tomographic scanning, and admission for observation in cases of minor head injury. Neurosurgery 1988;22:449 – 453 18. Gomez PA, Lobato RD, Ortega JM, De La Cruz J. Mild head injury: differences in prognosis among patients with a Glasgow Coma Scale score of 13 to 15 and analysis of factors associated with abnormal CT findings. Br J Neurosurg 1996;10:453– 460 19. Gandy SE, Snow RB, Zimmerman RD, Deck MDF. Cranial nuclear magnetic resonance imaging in head trauma. Ann Neurol 1984;16:254 –257 20. Jenkins A, Teasdale G, Hadley MDM, et al. Brain lesions detected by magnetic resonance imaging in mild and severe head injuries. Lancet 1986;2:445– 446 21. Snow RB, Zimmerman RD, Gandy SE, Deck MDF. Comparison of magnetic resonance imaging and computed tomography in the evaluation of head injury. Neurosurgery 1986;18:45–52 22. Groswasser Z, Reider-Groswasser I, Soroker N, Machtey M. Magnetic resonance imaging in head injured patients with normal late computed tomography scans. Surg Neurol 1987;27: 331–337 23. Wilberger JE, Deeb Z, Rothfus W. Magnetic resonance imaging in cases of severe head injury. Neurosurgery 1987;20:571– 579 24. Levin HS, Williams D, Crofford MJ, et al. Relationship of depth of brain lesions to consciousness and outcome after closed head injury. J Neurosurg 1988;69:861– 866 25. Wilson JTL, Wiedmann KD, Hadley DM, et al. Early and late magnetic resonance imaging and neuropsychological outcome after head injury. J Neurol Neurosurg Psychiatry 1988;51:391– 396 26. Levin HS, Amparo E, Eisenberg HM, et al. Magnetic resonance imaging and computerized tomography in relation to the neurobehavioral sequelae of mild and moderate head injuries. J Neurosurg 1987;66:706 –713 27. Levin HS, Williams DH, Eisenberg HM, et al. Serial MRI and neurobehavioral findings after mild to moderate closed head injury. J Neurol Neurosurg Psychiatry 1992;55:255–262 28. Mitchener A, Wyper DJ, Patterson J, et al. SPECT, CT, and MRI in head injury: acute abnormalities followed up at six months. J Neurol Neurosurg Psychiatry 1997;62:633– 636 29. Brooks DN, Hosie J, Bond MR, et al. Cognitive sequelae of severe head injury in relation to the Glasgow Outcome Scale. J Neurol Neurosurg Psychiatry 1986;49:549 –553 30. Anderson SI, Housley AM, Jones PA, et al. Glasgow Outcome Scale: an inter-rater reliability study. Brain Injury 1993;7:309 – 317 31. Naalt van der J, Zomeren van AH, Sluiter WJ, Minderhoud JM. One year outcome in mild to moderate head injury: the predictive value of the acute injury characteristics related to


77

32. 33.

34. 35.

36.

78

complaints and return to work. J Neurol Neurosurg Psychiatry 1999;66:207–214 Russell WR, Smith A. Posttraumatic amnesia in closed head injury. Arch Neurol 1961;5:16 –29 Jennett B, Snoek J, Bond MR, Brooks N. Disability after severe head injury: observations on the use of the Glasgow Outcome Scale. J Neurol Neurosurg Psychiatry 1981;44:285–293 Williams DH, Levin HS, Eisenberg HM. Mild head injury classification. Neurosurgery 1990;27:422– 428 Berryhill P, Lilly MA, Levin HS, et al. Frontal lobe changes after severe diffuse closed head injury in children: a volumetric study of magnetic resonance imaging. Neurosurgery 1995;37: 392– 400 Eide PK, Tysnes OB. Early and late outcome in head injury patients with radiological evidence of brain damage. Acta Neurol Scand 1992;86:194 –198

Annals of Neurology

Vol 46

No 1

July 1999

37. Adams JH, Graham DI, Murray LS, Scott G. Diffuse axonal injury due to nonmissile head injury in humans: an analysis of 45 cases. Ann Neurol 1982;12:557–563 38. Wilson JTL, Teasdale GM, Hadley DM, et al. Post-traumatic amnesia: still a valuable yardstick. J Neurol Neurosurg Psychiatry 1993;56:198 –201 39. Assaf Y, Beit-Yanai I, Shohami E, et al. Diffusion- and T2weighted MRI of closed-head injury in rats: a time course study and correlation with histology. Magn Reson Imaging 1997;15: 77– 85 40. Ashikaga R, Araki I, Ishida O. MRI of head injury using FLAIR. Neuroradiology 1997;39:239 –242 41. Jansen HML, van der Naalt J, van Zomeren AH, et al. Cobalt-55 positron emission tomography in traumatic brain injury: a pilot study. J Neurol Neurosurg Psychiatry 1996;60: 221–224