Epilepsia, 48(8):1621–1624, 2007 Blackwell Publishing, Inc. C 2007 International League Against Epilepsy
Brief Communications
Intracarotid Amobarbital Procedure in Post-temporal Lobectomy Patients ∗ D. Bradley Burton, †Edgard Pereira, ∗ Pradeep N. Modur Departments of ∗ Neurology; and †Radiology, University of Louisville School of Medicine, Louisville, Kentucky, U.S.A.
Summary: We describe two patients who underwent intracarotid amobarbital procedure (IAP) postoperatively following temporal lobectomy (one right temporal and one left temporal lobectomy), prompted by consideration of reoperation for persistent, intractable seizures. IAP memory score, consisting of the percentage of correctly recognized dually encoded stimuli, was calculated for each hemisphere. Both patients performed well on the IAP baseline memory assessment prior to injections, and both were left hemisphere dominant for language. The IAP memory results were consistent with the postoperative neuropsychological memory evaluation in that the patient who had undergone nonlanguage-dominant temporal resection performed within normal limits, while the patient who had undergone languagedominant temporal resection performed poorly, although not
completely amnestic, on memory testing. Injection of the nonsurgical, presumably healthy, hemisphere resulted in complete failure of memory in both patients, implying that baseline memory was dependent on the functional integrity of the nonsurgical hemisphere, inactivation of which led to complete memory disruption. Secondly, the hippocampus in the nonlanguagedominant hemisphere was able to support memory to some extent, if not completely, when it functioned in isolation without the influence of the language-dominant hemisphere during the IAP. These findings are discussed in the context of functional reserve and adequacy models. Key Words: Intracarotid amobarbital procedure—Wada test—Temporal lobectomy—Epilepsy— Epilepsy surgery—Neuropsychological testing.
Intracarotid amobarbital procedure (IAP) is performed preoperatively to assess language dominance and predict the risk of postoperative memory decline in epilepsy surgery patients (Trennery and Loring, 2001). IAP consists of multiple assessments of expressive and receptive language functions, and the measurement of selective declarative memory deficits following cerebral hemianesthesia induced by a short-acting medication. In this article, we present two patients who underwent IAP postoperatively following temporal lobectomy, and discuss our findings in the context of functional reserve and functional adequacy models of postoperative memory deficits in temporal lobectomy patients. To our knowledge, such results have not been reported previously.
METHODS Patient characteristics Two patients underwent IAP and neuropsychological evaluation postoperatively after temporal lobectomy. In each case, the postoperative IAP was prompted by consideration of reoperation for persistent, intractable seizures following previous unsuccessful surgery, performed several years ago by a different epilepsy surgery team. Neither patient had preoperative IAPs, so the language and memory lateralization were unknown prior to the current IAP. The pertinent data for the patients are summarized in Table 1. Patient 1 She was a 43-year-old female with seizure onset at age 17, with no identifiable risk factors for epilepsy. She underwent right anteromedial temporal lobectomy at age 30, remained seizure free for about six months, and then began having seizures similar to her preoperative seizures. Based on partial seizures of right frontal onset recorded during postoperative video EEG monitoring, she was considered for right frontal resection. Postoperative MRI showed changes consistent with right anterior temporal
Accepted January 31, 2007. Address correspondence and reprint requests to Pradeep Modur, M.D., University of Louisville, Department of Neurology, HSC #113, 500 S. Preston Street, Louisville KY 40292. E-mail:
[email protected] doi: 10.1111/j.1528-1167.2007.01094.x
1621
1622
D. B. BURTON ET AL. TABLE 1. Patient characteristics Patient 1
Patient 2
Age at IAP, years Gender Handedness Years of education Age at seizure onset, years Age at surgery, years Surgical procedure performed Preoperative IAP Preoperative MRI Postoperative seizure localization Postoperative neuropsychological testing
43 Female Right 13 17 33 Right anteromedial temporal lobectomy Not performed Unknown Right frontal Normal memory performance after 30 minute delay
Reason for postoperative IAP Planned reoperation
Unknown language and memory lateralization Right frontal resection
41 Male Right 12 13 33 Left anteromedial temporal lobectomy Not performed Left hippocampal atrophy Left temporal Moderately impaired memory function on delay recall (regardless of material specificity) Unknown language and memory lateralization Resection of posterior left hippocampal remnant
IAP, Intracarotid amobarbital procedure.
lobectomy with presence of a remnant of posterior hippocampus but no frontal lesion. Postoperative neuropsychological testing showed her verbal and nonverbal declarative memory scores in the average/normal range on immediate and delay recall trials (from the Wechsler Memory Scale-Revised). Patient 2 He was a 41-year-old male with seizure onset at age 14, with left mesial temporal sclerosis on preoperative MRI. He underwent left anteromedial temporal lobectomy at age 25 but never achieved seizure freedom. Based on partial seizures of left temporal onset recorded during postoperative video EEG monitoring and the presence of sclerotic left posterior hippocampal remnant on MRI, he was being evaluated for resection of the hippocampal remnant. Postoperative neuropsychological assessment revealed greater memory impairment in regards to recall of verbal as opposed to nonverbal material on immediate recall trials, along with moderately to severely impaired memory function, regardless of material specificity, on delay recall trials (from the Wechsler Memory Scale-Revised). IAP protocol Our IAP protocol is a variation of the protocol described previously (Loring et al., 1995). IAPs were performed under continuous video EEG monitoring with sodium amytal as the agent for anesthesia. Prior to catheterization, a baseline assessment of language, memory and grip strength was performed. The language and memory items for baseline testing were similar to those used for testing after injection. The stimuli for memory testing consisted of a total of 12 dually encodable objects and pictures of objects (eight objects and four pictures), presented for 4 s each in the central visual field and moved horizontally in both visual fields while emphasizing their names three times with auditory cues. After catheterization of internal carotid arteries and routine angiographic visualization, the surgical Epilepsia, Vol. 48, No. 8, 2007
and nonsurgical hemispheres were injected in that order, with at least 30 min between injections. Sodium amytal was injected over 10–15 s until flaccid paralysis of the contralateral upper extremity was objectively demonstrated. The dose of sodium amytal for each injection was variable, in the range of 80–110 mg for each injection. Speech arrest was evaluated by asking the patient to count from 1 to 20. At the completion of each injection, a brief mental status exam was performed to assess the initial impact of the injection. Memory acquisition was accomplished at the height of anesthetization and during the period of EEG slowing, usually within 60–90 s of injection. Expressive and receptive language functions, and grip strength (from a handheld dynamometer) were assessed sequentially over multiple trials until they returned to the baseline. Language dominance was determined after taking into consideration the initial impact of injection and the subsequent recovery over multiple trials. Memory recall was tested by recognition format after recovery of language functions and grip strength to baseline, or after 10 min from the end of memory acquisition phase, whichever was later. The patient was asked to respond with yes/no answers upon presentation of the targets, mixed with an equal number of foils from the same category. The total memory score, consisting of the percentage of material (both objects and pictures) correctly recognized, was calculated for each hemisphere. No corrections were made for guessing. RESULTS Both patients were left hemisphere dominant for language. For patient 1, who had undergone right temporal resection, right hemisphere injection caused no disruption of recognition memory (compared to baseline), while left hemisphere injection caused complete suppression of memory. For patient 2, who had undergone left temporal resection, left hemisphere injection caused
INTRACAROTID AMOBARBITAL PROCEDURE
1623
normal scores regardless of material specificity; patient 2 who had undergone language-dominant temporal resection showed material-specific deficit on immediate recall trials and nonmaterial specific memory impairment on delay recall trials, but was not completely amnestic for either verbal or nonverbal material. DISCUSSION
FIG. 1. Memory assessment during intracarotid amobarbital procedure showing the total memory score (% of target recognized correctly).
significant but not complete disruption of memory (compared to baseline), while right hemisphere injection caused complete suppression of recognition memory. These results are shown in Fig. 1. Data collapsed across both patients showed the following results (Fig. 2). When the language-dominant hemisphere hippocampus was functioning in isolation (dominant hippocampus), there was no disruption of memory compared to baseline. When the nonlanguagedominant hemisphere hippocampus was functioning in isolation (nondominant hippocampus), there was significant, but not complete, memory suppression. When both hippocampii were inactivated (no hippocampus), there was complete suppression of recognition memory. These findings on IAP memory assessment were also consistent with the postoperative neuropsychological assessments of memory: patient 1 who had previously undergone nonlanguage-dominant temporal resection showed
FIG. 2. Memory assessment during intracarotid amobarbital procedure showing the score collapsed across the two patients (% of targets recognized correctly).
Intracarotid amobarbital procedure is not commonly performed postoperatively in patients who have undergone epilepsy surgery. Nevertheless, when performed postoperatively as in our cases, IAP memory scores were well correlated with hippocampal dominance, and the existence of the dominant versus the nondominant hippocampus predicted the facility of memory performance on neuropsychological assessments. We present this as new information not previously reported in the literature. Regardless of their postoperative memory performance on neuropsychological testing, both patients had high (83% correct) memory scores on baseline testing immediately before the IAP injections. This difference between the baseline IAP memory scores and the neuropsychological memory scores in the two patients can be attributed to the methodological differences between the two types of measures, i.e., recognition format with IAP versus free recall format with neuropsychological testing. Upon injection of the nonsurgical, presumably healthy hemispheres (left hemisphere injection in patient 1 and right hemisphere injection in patient 2), both patients showed complete failure of memory (0% correct). This implies that their baseline memory, regardless of the normatively derived level of performance, was heavily dependent on the functional integrity of the nonsurgical hemisphere, inactivation of which led to complete memory disruption. Thus, the ability of the nonsurgical hemisphere to support memory at baseline (without injection) or in isolation during the IAP (upon injection of the surgical hemisphere) provides strong evidence in favor of functional reserve theory (Chelune, 1995), which states that postoperative memory deficits are related to the ability of the hippocampus contralateral to the surgical hemisphere to support memory functions. Our findings are consistent with recent studies supporting the functional reserve model in left temporal lobectomy patients (Bell et al., 2000; Chiaravalloti and Glosser, 2001). We also found that the hippocampus in the nonlanguage-dominant hemisphere (i.e., nondominant hippocampus) can support memory to some extent, although not at the level of the hippocampus located in the language-dominant hemisphere (i.e., dominant hippocampus), when it functions in isolation. This is evident from the 33% total recognition memory score (compared to the baseline) upon left hemisphere injection in patient 2, who had undergone left anterior temporal lobectomy, Epilepsia, Vol. 48, No. 8, 2007
1624
D. B. BURTON ET AL.
ipsilateral to the language-dominant hemisphere. This finding of partial memory failure on postoperative IAP memory testing is similar to the previous studies that have reported memory decline postoperatively on neuropsychological testing in left temporal lobectomy patients (Bell et al., 2000; Sabsevitz et al., 2001; Stroup et al., 2003). This finding is not consistent with the functional adequacy model, which states that the preoperative integrity of the resected tissue predicts postoperative functioning (Chelune, 1995), since this patient had hippocampal atrophy on MRI preoperatively. Rather, it clearly suggests that recognition memory as evaluated during the baseline and injection phases of the IAP is dependent upon the memory capabilities of the remaining hippocampus. Based on the results of postoperative IAP and neuropsychological testing, additional language-dominant hemisphere resection in patient 2 who had undergone prior resection in the same area would not predict further decline in memory function taking a functional reserve approach. Since we do not have the preoperative IAP and neuropsychological data, we can speculate that if this patient’s preoperative level of functioning fell within normal limits, then the dominant hemisphere resection already dramatically reduced his functional memory skills beyond the point where additional resection could cause further impact; on the other hand, if this patient had poor preoperative memory, perhaps correlated with the finding of hippocampal atrophy, then again additional resection would be unlikely to result in further memory decline. Thus, regardless of this patient’s preoperative level of performance, the important factor predicting postoperative memory performance would be the ability of the remaining hippocampus to subserve memory function, consistent with the functional reserve model.
Epilepsia, Vol. 48, No. 8, 2007
Finally, the ability of the nonlanguage-dominant hippocampus to support memory seems reassuring in the context of recommending dominant temporal lobectomy for those patients with intractable epilepsy who might otherwise be denied surgery based on failure on IAP memory testing. For these patients, the extent to which the nondominant hemisphere hippocampus can support memory may lie on a continuum that perhaps can be assessed preoperatively, allowing better prediction of the potential risks of postoperative memory decline.
REFERENCES Bell BD, Davies KG, Haltiner AM, Walters GL. (2000) Intracarotid amobarbital procedure and prediction of postoperative memory in patients with left temporal lobe epilepsy and hippocampal sclerosis. Epilepsia 41:992–997. Chelune GJ. (1995) Hippocampal adequacy versus functional reserve: predicting memory functions following temporal lobectomy. Archives of Clinical Neuropsychology10:413–432. Chiaravalloti ND, Glosser G. (2001) Material-specific memory changes after anterior temporal lobectomy as predicted by the intracarotid amobarbital test. Epilepsia 42:902–911. Loring DW, Meador KJ, Lee GP, King DW, Nichols ME, Park YD, Murro AM, Gallagher BB, Smith JR. (1995) Wada memory asymmetries predict verbal memory decline after anterior temporal lobectomy. Neurology 45:1329–1333. Sabsevitz DS, Swanson SJ, Morris GL, Mueller WM, Seidenberg M. (2001) Memory outcome after left anterior temporal lobectomy in patients with expected and reversed Wada memory asymmetry scores. Epilepsia 42:1408–1415. Stroup E, Langfitt J, Berg M, McDermott M, Pilcher W, Como P. (2003) Predicting verbal memory decline following anterior temporal lobectomy (ATL). Neurology 60:1266–1273. Trenerry M, Loring DW. (2001) The intracarotid amobarbital procedure. In Wyllie E (Ed) The treatment of epilepsy: principles and practice. Lippincott, Williams, & Wilkins, Philadelphia, pp.1043–1052.
