Technetium Tc-99m Ethyl Cysteinate Dimer Brain Single-Photon Emission CT in Mild Traumatic Brain Injury: A Prospective Study

CT Scan.

A noncontrast CT was done on a Picker 2000 scanner with 8-mm sections in all patients. The presence of any contusion(s), hematomas, or subarachnoid blood was considered as an abnormal finding.

SPECT.

SPECT was performed in all patients in a silent, dimly lit room with eyes open and ears unplugged, 45 minutes after intravenous injection of 555–925 MBq (15–25 mCi) ^99mTc-ECD (BARC, Mumbai). For patients less than 18 years of age the dose was calculated based on the patient’s weight (10 MBq /kg). Acquisition was done on a dual-headed rotating scintillation gamma camera (Elscint, Varicam) with the patient supine, headrest attached, smallest permissible radius of rotation, 128 × 128 matrix, 360°, 120 projections, 25 seconds per projection by using either low-energy ultrahigh-resolution fan beam collimator or a low-energy high-resolution parallel hole collimator.

Raw data were smoothed with Butterworth filtered Nyquist of 1.404 cycles/cm and cutoff frequency of 0.56. Chang attenuation correction was applied. Images were then reoriented in axial, coronal, and sagittal planes. Final data were displayed on a computer monitor and analyzed by using 10-graded color scale. The cerebellum was used as reference site (100% maximum value). Any decrease in cerebral perfusion in the cortex or basal ganglia <70% or <50% in the medial temporal lobe compared with the cerebellum was considered abnormal. This method has been described elsewhere.8

The studies were analyzed independently by 2 experienced nuclear physicians who were blinded to the patients’ clinical symptoms as well as the CT findings. Only if there was concordance between them was the study considered positive. SPECT data were then compared with the CT scan and clinical findings by the neurosurgeons.

There is normal variation of the radiotracer uptake pattern on SPECT throughout the entire brain, which depends on the radiopharmaceutical used. We therefore carried out SPECT by using Tc99m- ECD on healthy subjects to further validate our results. Brain SPECT was performed on 40 healthy volunteers to develop an institutional normal data base. There were 30 men and 10 women, ranging in age from 18 to 65 years. None of these patients showed any abnormality with the criteria used above. The consensus rate between 2 observers for SPECT study of normal controls was 100%. For the patients with MTBI, the consensus rate was 93%.

Statistical Analysis.

Pearson χ² statistic was used to compare SPECT and CT findings in patients who had PTA, LOC, or symptoms of PCS, with patients who did not have these symptoms. McNemar test was used to compare the SPECT and CT findings within each of these groups separately. A χ² test for trend analysis was carried out to examine whether positive SPECT and CT findings were related to duration since head injury and GCS.

“No head injury is too trivial to ignore” (Hippocrates, 460–377 BC, in Ingebrigsten)11

Head trauma has been one of the major causes of morbidity and mortality in all ages. The incidence is particularly high in the productive age group.1 Although accounting for more than 70% cases of head injury, MTBI has lacked attention by neurosurgeons and neurologists alike. The situation, fortunately, has improved dramatically in recent years, and the long-term effects of concussion are now being recognized so much so that MTBI is now considered a silent epidemic.5 Of all the imaging modalities, CT has revolutionized the management of head injury and is considered the investigation of choice.6, 12 It can be easily performed, is less expensive, and bony injuries and hematomas are better visualized.13 On the other hand, studies comparing CT with MR imaging have shown that CT can be relatively insensitive and that MR imaging is more sensitive than CT particularly for nonhemorrhagic lesions.14, 15 MR imaging, however has its own limitations, mainly due to the expenses involved, associated motion artifacts in agitated patients, and the deterioration in image quality in the presence of ferromagnetic monitoring equipment.16 Even if one manages to overcome these limitations, not all lesions following MTBI may be detected.17 Although positron-emission tomography provides information on cerebral metabolism and has shown promise in MTBI,12 it is not widely available, not regularly reimbursed, and is more expensive than SPECT. The discovery of significant changes in relative cerebral blood flow (rCBF) in patients with MTBI by various researchers18–20 makes SPECT a promising tool in evaluating patients with MTBI. Previous studies have shown that SPECT may be a viable technique in evaluating MTBI.8, 21, 22 Most of these studies, however, are retrospective by design8, 20 or involve small study populations.17, 20

In their prospective study, Jacobs et al evaluated 136 patients with MTBI,22 who underwent initial SPECT within 4 weeks after trauma. The patients were followed at 3, 6, and 12 months after injury. They concluded that normal SPECT findings are a reliable tool to exclude clinical sequelae of mild injury. Also at 12 months postinjury, a positive SPECT study is also a reliable predictor for clinical outcome.23 It has been suggested that, the earlier the SPECT is done after MTBI, the greater the number of lesions that will be detected.8 In our study, SPECT demonstrated perfusion abnormalities in 63% patients, whereas CT showed brain parenchymal lesions in only 34% of patients. This confirms earlier studies, which have shown that SPECT is more sensitive than CT in detecting underperfused areas of the brain in MTBI.8, 23–29 Lesions were mainly localized to the frontal and temporal lobes followed by the basal ganglia and thalamus. Similar results were seen by Abdel-Dayem et al.8 Frontal and temporal lobes are commonly affected in head injury mainly due to the gliding effect of the brain over the underlying skull.6 All patients in our study had hypoperfusion in the affected regions. Although this hypoperfusion is generally attributed to the cerebral edema that surrounds the damaged brain and limits cerebral blood flow,20 it may also result from vasospasm, direct vascular injury, and perfusion changes due to alterations in remote neuronal activity (diaschisis).17, 27, 28 There is evidence to show that the brain is more vulnerable to ischemic injury after minor head injury, and it has been hypothesized that SPECT findings representing hypoperfusion may in fact lead to secondary ischemic injury.17

In patients showing lesions on CT, hypoperfusion on SPECT corresponded to the CT lesions, and in most of the cases hypoperfused lesions were much larger than the lesions seen on CT. In some of these patients SPECT showed hypoperfusion in regions that appeared normal on CT. In 29 patients with normal CT, SPECT demonstrated areas of hypoperfusion. This may be due to relative changes in the rCBF not detected by CT. Similar findings have been observed in other studies as well.8, 21, 23 Two of the patients had positive CT findings but negative SPECT findings. Both had subarachnoid hemorrhages. The possible reasons are due to the poorer resolution of SPECT and the absence of significant cerebral blood flow changes in the underlying brain in these patients.8

Some authors have observed hyperperfusion with Tc99m HMPAO SPECT in regions corresponding to the CT lesions.7 We did not observe hyperperfusion in any case, and this could be because of the use of Tc99m-ECD in all cases. Hyperperfusion is generally reported in moderate to severe brain injury patients, in whom there may be dissociation between cerebral blood flow and reduced metabolic demand in the lesions.7 Tc99m-ECD shows predominant cellular-metabolic uptake, and hexamethyl propylene amineoxine (HMPAO) reflects blood flow arrival to cerebral regions.30 This accounts for the slight differences in the SPECT perfusion patterns. For cerebral perfusion-metabolic uncoupling, increased HMPAO uptake may occur, thus reflecting the luxury perfusion phenomenon whereas ECD uptake remains low, reflecting hypometabolism at the site of ischemia. Image interpretation is therefore more straightforward with ECD. Moreover image quality obtained by using ECD is relatively better than images obtained with HMPAO.30

When comparing SPECT and CT findings with the presenting complaints such as LOC, PTA, and the symptoms associated with PCS, SPECT was found to be more sensitive in detecting brain lesions in all the groups. SPECT positivity was greater in patients having these symptoms compared with patients without these symptoms. In the 37 patients who showed LOC and normal CT findings, SPECT was positive in 16 patients (43.2%). In patients with PTA and normal CT (n = 17), SPECT was positive in 12 patients (70.1%). Similar findings have been reported by Lorberboym et al.31 SPECT was positive in 84.6% of patients showing PCS and the findings are similar to the study by Kant et al.32 Unlike previous studies in which frontal lobe lesions were most common, however, we found temporal lobes to be most commonly involved in patients having PCS. Ten of the 28(35.7%) children showed PCS, whereas 16 of 64 adults (25%) had PCS 1 week following injury. Of these 26 patients, SPECT was abnormal in 22 (84.6%) and revealed temporal hypoperfusion in 18 (69.2%), frontal hypoperfusion in 12 (46.2%), and hypoperfusion in the basal ganglia and thalamic regions in 12 (46.2%; Table 2).

Previous studies have suggested that a significant proportion of children admitted with MTBI may have moderate disability at follow-up.33 In a recently published study, posttraumatic symptoms in children did not correlate with somatic, neurologic, or electroencephalographic findings observed immediately after the injury or at the follow-up investigation after 6 weeks.34 Our study shows that SPECT with its higher sensitivity may therefore help in predicting persistent post-traumatic symptoms in children with MTBI. Our data also support the contention that functional brain imaging with SPECT could be a useful investigational tool in patients having PTA, LOC, or PCS, particularly in patients who have normal anatomic imaging.

In the absence of a gold standard to detect brain lesions following MTBI, however, one cannot calculate the exact sensitivity and specificity of SPECT and CT. Nevertheless, with the available data, we can conclude that SPECT is much more sensitive in detecting lesions than CT. There was a significant increase in the number of brain lesions detected by SPECT as the GCS decreased from 15 to 13 (Table 4), reiterating the importance of GCS in evaluation of MTBI; however, no similar association was found when SPECT findings were compared with duration of head injury. This disparity could be because relatively less symptomatic patients were imaged within the initial 48 hours and discharged from hospital, whereas more symptomatic patients were imaged on the third day and were observed for longer duration.

A meta-analysis by the World Health Organization has suggested that financial compensation was a strong risk factor for long-term disability, symptoms, and objective findings after MTBI.35 SPECT being more sensitive can be useful in solving legal aspects in MTBI better than CT by ruling out any underlying brain lesion in a symptomatic case. According to the current standard of practice, the treatment of patients with MTBI with a normal head CT is purely based on symptoms, irrespective of SPECT findings. We believe that this mode of management should be reassessed and investigated based on the findings of our study. Serial SPECT imaging may possibly serve as a platform to test the efficacy of various drugs and neurobehavioral interventions in symptomatic patients with MTBI.