Background: Selective amygdalohippocampectomy (SelAH) is increasingly performed in patients with mesial temporal lobe epilepsy and hippocampal sclerosis. To determine whether visual field defects are less pronounced after SelAH than after standard temporal lobectomy (StTL), we retrospectively analyzed postoperative quantitative visual fields after the 2 procedures.
Methods: Humphrey visual field analysis was obtained postoperatively in 18 patients who had undergone SelAH and in 33 patients who had undergone StTL. The SelAH was performed via a transcortical approach through the middle temporal gyrus and included the amygdala, 3 cm of the hippocampus, and the parahippocampal gyrus. The visual field pattern deviation was used for analysis. We considered a defect clinically significant if there were 3 contiguous coordinates affected at the 5% level or 2 at the 1% level.
Results: All but 2 of 18 patients who had undergone SelAH had homonymous superior quadrantic visual field defects contralateral to the side of the surgery. One patient had no defects by our criteria, and one had a mild defect that reached significance only in the ipsilateral eye. The averaged defect affected mostly coordinates close to the vertical meridian with relative sparing of points close to the horizontal meridian. All but 3 of the 33 patients who had undergone StTL had homonymous superior quadrantic visual field defects. One patient had no defects; 2 had defects that reached significance in only one eye. The averaged defect involved all points in the affected quadrant, but was also greater near the vertical meridian. Of 13 tested visual field coordinates, 4 were significantly less affected by SelAH in the ipsilateral eye and 3 in the contralateral eye. The coordinates close to the horizontal meridian were significantly spared by SelAH.
Conclusions: Visual field defects are very common after SelAH but are significantly less pronounced than after StTL. In particular, the visual field close to the horizontal meridian is relatively spared in SelAH.
Departments of Neurology (TM, MA-A, KFH, PJL, BWA-K), Neurosurgery (DAS, PEK, MP), Ophthalmology (PJL), and Biostatistics (LW, YS), Vanderbilt University Medical Center, Nashville, Tennessee.
This study was supported by an unrestricted grant from Research to Prevent Blindness, Inc.
Address correspondence to Bassel Abou-Khalil, MD, A-0118 MCN, Nashville, TN 37232; E-mail: firstname.lastname@example.org
Temporal lobe resection is a widely used and effective method for treatment of refractory temporal lobe epilepsy (1). One of the morbidities associated with standard temporal lobectomy (StTL) is a contralateral superior homonymous quadrantanopia attributed to disruption of Meyer's loop, the anterior bundle of the optic radiations that travels through the temporal lobe (2,3). Although the visual field deficit is often not noticeable to the patient (3,4), it may, depending on its density and extent, have implications for driving. In studies performed in the United Kingdom, 25%-50% of patients failed to meet driving requirements because of a postsurgical visual field deficit (5,6). Thus, minimizing the visual field deficit is an important goal in performing temporal lobectomy.
Selective amygdalohippocampectomy (SelAH) is increasingly being used as an alternative to StTL when the presurgical evaluation points to a mesial temporal epileptogenic focus, particularly with MRI evidence of hippocampal sclerosis (7). The potential advantages of SelAH include a smaller craniotomy that is associated with less discomfort and surgical morbidity and a smaller surgical resection that has less impact on language, memory, and neuropsychologic function (8-13).
We hypothesized that SelAH would be less likely than StTL to produce visual field defects. In a study using Goldmann kinetic perimetry (14), no differences were noted in the frequency of visual field defects between SelAH and StTL. In the current study, we used automated static perimetry, allowing for quantification of data for many points in the visual field and for statistical comparisons at each point to assess the severity of the visual field defect (3).
After the study was approved by the Vanderbilt University Medical Center Investigational Review Board, we recruited 18 patients who had undergone SelAH at Vanderbilt University Medical Center from 2001 to 2006. After the appropriate informed consent was obtained, all patients underwent automated static perimetry. For comparison, we studied automated static perimetry results obtained postoperatively from 33 patients who had undergone StTL from 1990 to 1995 at Vanderbilt University Medical Center. We excluded patients with glaucoma, cataracts, or other significant ocular disease, those who had had stroke as a complication of surgery, and those who were unable to follow instructions and therefore had unreliable visual fields.
SelAH was performed via a transcortical approach. By use of image guidance, a 1.5-cm corticotomy was made in the superior aspect of the middle temporal gyrus followed by a corridor to the temporal horn of the lateral ventricle through the white matter. The resection included the amygdala, approximately 3 cm of the hippocampus to a point just posterior to the cerebral peduncle, and the parahippocampal gyrus. StTL involved a tailored lateral temporal resection based on intraoperative electrocorticography. Approximately 4-7 cm of the middle and inferior temporal gyri were resected. Right-sided resections included the superior temporal gyrus; left-sided resections often spared this tissue. After resection of lateral neocortical structures, the mesial structures were accessed via the temporal horn of the lateral ventricle and resected as in SelAH.
Visual Field Analysis
We used Humphrey automated static perimetry for all visual field studies. For the patients who underwent SelAH, we used the 24-2 Humphrey Swedish Interactive Threshold Algorithm (SITA) standard strategy that tests 54 points within to the central 30° around fixation. The only two points that reached 30° were around the horizontal meridian in the nasal field. They were excluded from analysis because the distribution of points had to be symmetrical. The visual fields for the StTL group were obtained with the Humphrey 30-2 Full Threshold strategy that tests 76 points in the central 30° degrees around the point of fixation. To analyze only the coordinate points located within 24° degrees from fixation, we excluded the most peripheral coordinates. For easier comparison, all visual fields were converted into a right eye format.
We used the pattern deviation (PD) thresholds for our analysis. PD is a measurement of the deviation from the age-corrected normal values, adjusted for overall sensitivity (15). Expressed in decibels, PD values are automatically provided by the program. Each value is also assigned a statistical significance level. A visual field defect was considered significant if it involved at least 3 contiguous coordinates at the 5% level or 2 contiguous coordinates at the 1% level.
The visual field for each eye was divided into 4 quadrants. We limited our analysis to the superior quadrant contralateral to the surgery. All PD values were entered into a computerized spreadsheet for analysis. We converted all fields to right eye format and recorded the pattern deviations for each coordinate in a computerized spreadsheet. The averaged defect was displayed on a grid with 13 coordinate points corresponding to the location of the test points, creating a map that showed the location and extent of the visual field defects. The affected superior quadrant was divided into two sectors, medial and lateral (Fig. 1). The points that fell on the 45° line were not included in the sector analysis.
At each coordinate, the visual pattern deviation values were compared between SelAH and StTL operations using a Wilcoxon rank-sum test. The values were also compared for the ipsilateral eye and the contralateral eye using a Wilcoxon signed-rank test. We next repeated these comparisons by sector using the generalized estimating equations (GEE) model (16). Because there were multiple observations from each patient, the GEE model was used to properly account for the within-subject correlations (16). For the comparison of SelAH and StTL, the outcome of this model was pattern deviation (in decibels), and the main effects were type of surgery (SelAH or StTL), side of eye (ipsilateral or contralateral), and sector of visual field (medial or lateral). The visual pattern deviations were next estimated for each surgery type by eye side and by sector and were compared using parameters from this model.
For these analyses, 2-sided P < 0.05 was considered statistically significant. We did not correct for multiple comparisons. Although multiple coordinates were examined and multiple hypotheses were tested, each parameter was of interest on its own, so we chose to report all individual P values and make separate statements in relation to our hypotheses. When multiple univariate test results have implications on specific responses, correction for multiple comparisons is not needed, as it is more relevant to know the strength of evidence for testing individual hypotheses on each parameter (17). Analyses were performed using SAS (version 9.1.3; SAS Institute Inc., Cary, NC) and R (version 2.3.1).
The patients who underwent SelAH included 12 women and 6 men, with a mean age of 37.4 years (range 6-56 years) at the time of surgery. Ten patients (56%) had undergone left SelAH and 8 patients (44%) had undergone right SelAH. The StTL group included 18 women and 15 men, with a mean age of 36.4 years (range 15-56 years) at the time of surgery. Thirteen patients (39%) had undergone left StTL and 20 (61%) had undergone right StTL.
Visual Field Defects
All patients who underwent SelAH had a visual field defect that affected at least one coordinate. All but 1 SelAH patient had an identifiable contralateral superior quadrant visual field defect defined as a deficit affecting at least 3 contiguous coordinates at the 5% level or 2 contiguous coordinates at the 1% level. This defect was homonymous in all but one patient who had a small superior quadrant defect that reached significance only in the eye ipsilateral to the surgery. Three patients had fewer than 6 affected coordinate points in the ipsilateral eye and 5 had fewer than 6 affected coordinate points in the contralateral eye. The averaged defect affected mostly the medial sector abutting the vertical meridian. In general, the points closest to the horizontal meridian were relatively spared (Fig. 1). The inferior quadrant was affected in only 3 of the 18 patients, but these patients had suboptimal reliability.
All patients who underwent StTL had a visual field defect that affected at least one coordinate. One patient did have a clinically significant defect by the criteria presented above. In 30 of the remaining 32 patients, the defect was homonymous. Two patients had a defect that reached significance in one eye only, ipsilateral to surgery in one patient and contralateral to surgery in the other. The averaged defect involved all points in the affected quadrant, but was also greater in the medial sector (Fig. 1).
Comparison of the whole field for the ipsilateral and contralateral eye showed that there was relatively less visual loss with SelAH than with StTL (P = 0.0465). Of the 13 tested visual field coordinates, 4 were significantly less affected by SelAH in the ipsilateral eye and 3 were significantly less affected in the contralateral eye (Fig. 1). Points significantly less affected by SelAH were all at the y3 and y9 vertical level.
Comparison by sector showed that the lateral sector was significantly spared by SelAH in both eyes. For the ipsilateral eye, the estimated pattern deviations were 5.8 for SelAH and 11.2 for StTL (P = 0.0024). For the contralateral eye, the estimated pattern deviations were 6.4 for SelAH and 10.8 for StTL (P = 0.0158). There was no significant difference between SelAH and StTL in medial sector pattern deviation.
We have shown that SelAH produces a less severe visual field deficit than StTL. Quantitative visual field assessment using automated Humphrey perimetry revealed contralateral visual field defects in all patients after StTL and SelAH but showed marked relative sparing of the field nearest the horizontal meridian in SelAH. A previous study (14) showed no difference between transcortical SelAH and StTL in the frequency of visual field defects, as assessed by Goldmann perimetry, but the severity of the defects was not assessed. Sparing of the lateral sector relative to the medial sector of the contralateral visual field was previously shown with StTL (3,18). When transcortical SelAH is performed, our data demonstrate that the visual fibers in Meyer's loop subserving the visual fields closest to the horizontal are even less susceptible to damage.
Our investigation has limitations. The StTL and SelAH groups were studied in different time periods. We could not conduct our study using contemporaneous patient groups because we have largely abandoned StTL for temporal lobe epilepsy with hippocampal sclerosis. Also, we were not able to assess the functional significance of the relative visual field sparing with respect to function and quality of life. We suspect that there were no clinically significant differences between the two groups in the functional consequences of the visual field defects. Nevertheless, sparing any part of the visual field should be a goal. In our study, the automated perimetry program assessed more points in the StTL group than in the SelAH group, but that difference should not be of consequence as our analysis focused on the 54 points tested in both patient groups.
The visual field defects after StTL differed in the current study from those in a previous report generated by our group (3). The current study relied exclusively on field pattern deviations in postoperative visual fields. In our previous study (3), we used sensitivity values, corrected the postoperative field with the mean difference between the preoperative and postoperative unaffected hemifield for possible diffuse depression, and used the difference between preoperative and corrected postoperative field sensitivity values. Finally, the current study used the 24-2 program which has fewer coordinate points than the 30-2 program used in the previous study. Despite the methodologic differences, the finding of relative field preservation near the horizontal meridian was replicated.
The organization of the fiber tracts in Meyer's loop has been a topic of considerable debate, as the presence and extent of visual field loss after StTL has a high degree of intersubject variability. It has been suggested that the fibers subserving the field closest to the horizontal meridian loop more posteriorly than the fibers subserving the fields closest to the vertical meridian (3). A recent study using Goldmann perimetry and a revised retinotopic model with central magnification suggested that the most anterior fibers of Meyer's loop represent the superior field rather than the field near the vertical meridian (18). Our data showing that transcortical SelAH relatively spares the field near the horizontal meridian provides additional information about the retinotopic anatomy of the temporal lobe optic radiations after they complete their anterior loop. With the transcortical surgical approach used in this patient population, the anterior temporal lobe was left intact, and the optic radiations were most likely interrupted in their course along the lateral wall of the temporal horn. In this setting, the relatively greater involvement of the medial-superior field suggests that the corresponding optic radiation fibers are located in the lateral wall of the temporal horn, with fibers subserving the relatively spared field near the horizontal meridian likely to be coursing more superiorly along the lateral wall and roof of the temporal horn. There was a large degree of variability in the visual field deficit across patients in the SelAH group, suggesting significant variability in the organization of the optic radiations in this segment of the pathway.
The surgical approach of SelAH is likely to play an important role in the degree of disruption of Meyer's loop. SelAH can be performed through several approaches (19), including transsylvian (20), subtemporal (10), and transcortical (21). The transcortical approach itself is also variable (19) as the corticotomy can be located in any of the gyri or sulci. Because the superior temporal gyrus contains the primary auditory cortex and may play a greater role in language function in the dominant hemisphere, the site of entry is typically through the middle temporal gyrus (21), as in the present study, or through the inferior temporal sulcus (22). Regardless of approach, resection of the mesial temporal structures routinely involves the temporal horn of the lateral ventricle as a corridor. Therefore, the relationship of the optic radiations to the temporal horn is paramount in predicting the extent of visual field defects. The medial wall and floor of the temporal horn are free of optic radiations anterior to the lateral geniculate body. In contrast, the roof and lateral wall of the temporal horn are covered by optic radiations (23). Thus, it should be expected that the visual outcomes will vary with different approaches. The subtemporal approach in conjunction with a stereotactic navigation system is reported to spare visual fields as the corridor into the temporal horn is through its floor (24). In contrast, the transsylvian-transinsular and transcortical approaches enter the temporal horn through the roof and lateral wall, respectively, and will probably violate the optic radiations (25). Direct violation of the optic radiations is not the only potential mechanism of injury to these white matter tracts during surgery. Interference with arterial perfusion and venous drainage due to vasospasm, undue brain retraction, or direct disruption can also injure the optic radiations.
As each surgical approach to the temporal horn has advantages and disadvantages (25), visual field sparing cannot be the sole element of morbidity considered in devising the surgical approach. Factors such as seizure outcome, neuropsychologic outcome, overall quality of life score, and technical difficulty of the procedure must all be considered in choosing the best approach to SelAH.
Advances in brain imaging offer the potential to guide surgery to limit disruption of optic radiation pathways. Diffusion tensor imaging allows for visualization of white matter tracts. This technique has been used to characterize the pattern of postsurgical impairment (26). More recently, it was demonstrated as a means of visualizing the course of Meyer's loop through the temporal lobe (27) and of confirming a high degree of intersubject variability in its course, particularly regarding its anterior extent (28). Additional investigation of the visual complications of various SelAH approaches, combined with the radiologic method of preoperative mapping of Meyer's loop, may be helpful to further limit the visual field defects resulting from SelAH.
1. Wiebe S, Blume WT, Girvin JP, et al. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med
2. Guenot M, Krolak-Salmon P, Mertens P, et al. MRI assessment of the anatomy of optic radiations after temporal lobe epilepsy surgery. Stereotact Funct Neurosurg
3. Hughes TS, Abou-Khalil B, Lavin PJ, et al. Visual field defects after temporal lobe resection: a prospective quantitative analysis. Neurology
4. Marino R Jr, Rasmussen T. Visual field changes after temporal lobectomy in man. Neurology
5. Manji H, Plant GT. Epilepsy surgery, visual fields, and driving: a study of the visual field criteria for driving in patients after temporal lobe epilepsy surgery with a comparison of Goldmann and Esterman perimetry. J Neurol Neurosurg Psychiatry
6. Pathak-Ray V, Ray A, Walters R, et al. Detection of visual field defects in patients after anterior temporal lobectomy for mesial temporal sclerosis-establishing eligibility to drive. Eye
7. Wieser HG, Ortega M, Friedman A, et al. Long-term seizure outcomes following amygdalohippocampectomy. J Neurosurg
8. Morino M, Uda T, Naito K, et al. Comparison of neuropsychological outcomes after selective amygdalohippocampectomy versus anterior temporal lobectomy. Epilepsy Behav
9. Paglioli E, Palmini A, Portuguez M, et al. Seizure and memory outcome following temporal lobe surgery: selective compared with nonselective approaches for hippocampal sclerosis. J Neurosurg
10. Hori T, Yamane F, Ochiai T, et al. Selective subtemporal amygdalohippocampectomy for refractory temporal lobe epilepsy: operative and neuropsychological outcomes. J Neurosurg
11. Lacruz ME, Alarcon G, Akanuma N, et al. Neuropsychological effects associated with temporal lobectomy and amygdalohippocampectomy depending on Wada test failure. J Neurol Neurosurg Psychiatry
12. Hori T, Yamane F, Ochiai T, et al. Subtemporal amygdalohippocampectomy prevents verbal memory impairment in the language-dominant hemisphere. Stereotact Funct Neurosurg
13. Wieser HG, Yasargil MG. Selective amygdalohippocampectomy as a surgical treatment of mesiobasal limbic epilepsy. Surg Neurol
14. Egan RA, Shults WT, So N, et al. Visual field deficits in conventional anterior temporal lobectomy versus amygdalohippocampectomy. Neurology
15. Haley MJ, ed. The Field Analyzer Primer
. 2nd ed. San Leandro, CA: Allergan Humphrey; 1987.
16. Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika
17. Cook RJ, Farewell VT. Multiplicity considerations in the design and analysis of clinical trials. J R Stat Soc Ser A
18. Barton JJ, Hefter R, Chang B, et al. The field defects of anterior temporal lobectomy: a quantitative reassessment of Meyer's loop. Brain
19. Goncalves-Ferreira A, Miguens J, Farias JP, et al. Selective amygdalohippocampectomy: which route is the best? An experimental study in 80 human cerebral hemispheres. Stereotact Funct Neurosurg
20. Yasargil MG, Teddy PJ, Roth P. Selective amygdalo-hippocampectomy. Operative anatomy and surgical technique. Adv Tech Stand Neurosurg
21. Olivier A. Transcortical selective amygdalohippocampectomy in temporal lobe epilepsy. Can J Neurol Sci
2000;27(Suppl 1):S68-S76; discussion S92-S166.
22. Miyagi Y, Shima F, Ishido K, et al. Inferior temporal sulcus approach for amygdalohippocampectomy guided by a laser beam of stereotactic navigator. Neurosurgery
23. Sincoff EH, Tan Y, Abdulrauf SI. White matter fiber dissection of the optic radiations of the temporal lobe and implications for surgical approaches to the temporal horn. J Neurosurg
24. Hori T, Tabuchi S, Kurosaki M, et al. Subtemporal amygdalohippocampectomy for treating medically intractable temporal lobe epilepsy. Neurosurgery
25. Campero A, Troccoli G, Martins C, et al. Microsurgical approaches to the medial temporal region: an anatomical study. Neurosurgery
26. Taoka T, Sakamoto M, Iwasaki S, et al. Diffusion tensor imaging in cases with visual field defect after anterior temporal lobectomy. AJNR Am J Neuroradiol
27. Yamamoto A, Miki Y, Urayama S, et al. Diffusion tensor fiber tractography of the optic radiation: analysis with 6-, 12-, 40-, and 81-directional motion-probing gradients, a preliminary study. AJNR Am J Neuroradiol
28. Nilsson D, Starck G, Ljungberg M, et al. Intersubject variability in the anterior extent of the optic radiation assessed by tractography. Epilepsy Res