Hypothalamic hamartomas: Optimal approach to clinical evaluation and diagnosis

Authors

  • Angus A. Wilfong,

    Corresponding author
    1. Division of Pediatric Neurology, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas, U.S.A
    • Address correspondence to Angus A. Wilfong, Pediatric Neurology, 6701 Fannin Street, CC1250, Houston, TX 77030, U.S.A. E-mail: awilfong@bcm.edu

    Search for more papers by this author
  • Daniel J. Curry

    1. Division ofPediatric Neurosurgery, Baylor College of Medicine, Texas Children's Hospital, Houston, Texas, U.S.A
    Search for more papers by this author

Summary

Hypothalamic hamartomas (HHs) present a difficult medical problem, manifested by gelastic seizures, which are often medically intractable. Although existing techniques offer modest surgical outcomes with the potential for significant morbidity, the relatively novel technique of magnetic resonance imaging (MRI)–guided stereotactic laser ablation (SLA) offers a potentially safer, minimally invasive method with high efficacy for the HH treatment. We report here on 14 patients with medically refractory gelastic epilepsy who underwent stereotactic frame–based placement of an MR-compatible laser catheter (1.6 mm diameter) through a 3.2-mm twist drill hole. A U.S. Food and Drug Administration (FDA)–cleared laser surgery system (Visualase, Inc.) was utilized to ablate the HH, using real-time MRI thermometry. Seizure freedom was obtained in 12 (86%) of 14 cases, with mean follow-up of 9 months. There were no permanent surgical complications, neurologic deficits, or neuroendocrine disturbances. One patient had a minor subarachnoid hemorrhage that was asymptomatic. Most patients were discharged home within 1 day. SLA was demonstrated to be a safe and effective minimally invasive tool in the ablation of epileptogenic HH. Because use of SLA for HH is being adopted by other medical centers, further data will be acquired to help treat this difficult disorder.

Approximately one third of all patients with epilepsy are refractory to pharmaceutical therapy (Brodie et al., 1997). For a substantial number of these patients, epilepsy surgery offers a potentially curative procedure by removing or destroying the primary epileptogenic focus. Improvements in surgical outcomes have been gained through development of new operative techniques, as well as better identification of cerebral anomalies underlying the epilepsy. Using electrophysiologic recordings and improved structural and functional neuroimaging, epileptogenic foci can be better identified and subsequently treated (Kwan & Brodie, 2000).

Patients with hypothalamic hamartoma (HH) have gelastic seizures that are particularly refractory to treatment. A rare, nonneoplastic developmental lesion of the inferior hypothalamus, HH is composed of cytologically normal neurons abnormally distributed within the hypothalamus (Brandberg, 2004). Gelastic epilepsy associated with HH is a well-characterized clinical syndrome consisting of gelastic seizures starting in infancy, which become refractory to medications and progress to include the development of multiple seizure types in association with severe behavioral and developmental decline (Berkovic et al., 2003; Boudreau et al., 2005). This epilepsy is particularly resistant to antiepileptic drugs, and these patients frequently develop an epileptic encephalopathy with a relentlessly deteriorating clinical course (Berkovic et al., 2003).

With greater clinical recognition and improved diagnostic capability using magnetic resonance imaging (MRI), the incidence of HH is currently reported to be about one in 200,000 (Alves et al., 2013). A variety of surgical approaches to remove HHs have been described, including skull base subfrontal, subtemporal, pterional, and frontotemporal approaches (Palmini et al., 2002; Fohlen et al., 2003), transcallosal interforniceal approaches (Rosenfeld et al., 2001; Harvey et al., 2003; Anderson, 2010), and intraventricular endoscopic techniques (Arita et al., 1998). Results of these surgical interventions for HH have shown variable outcomes, with most reporting approximately 50% seizure freedom. However, the operative morbidity is quite high, with many patients developing diabetes insipidus and memory dysfunction. In addition, there are also reports of other neuroendocrine disorders, visual impairments, and hemiparesis (Arita et al., 1998; Palmini et al., 2002; Freeman et al., 2003). Stereotactic radiosurgery, employed theoretically for its less invasive approach, has been found to be only modestly efficacious, with results delayed by months. Its applicability in the pediatric population may be limited by concerns radiation overexposure (Regis, 2007).

Current treatment options remain limited in efficacy and safety. The deep-seated location of the hamartoma and its proximity to critical structures makes it well suited for a minimally invasive technique such as stereotactic laser ablation (SLA). The Visualase MRI-guided laser ablation system was cleared by the U.S. Food and Drug Administration (FDA) for soft tissue ablation in neurosurgery; its use in the setting of failed metastatic tumor ablation (Carpentier et al., 2008) and a range of benign and malignant brain tumors has been published (Jethwa et al., 2012). In a proof of concept early clinical trial in epilepsy, SLA was shown to be safe and effective for a variety of epileptogenic lesions, including HH (Curry et al., 2012). This study reports the outcome of 14 patients with HH treated with SLA at a single center.

Materials and Methods

The Visualase Thermal Therapy System (Visualase, Inc., Houston, TX, U.S.A.) was employed to perform MRI-guided laser ablation of the HH in 14 patients at Texas Children's Hospital, Houston, Texas, U.S.A. The system comprises a computer workstation, a 15 W, 980 nm diode laser, a cooling pump, and a disposable laser applicator set composed of 400 mm core silica fiber optic with a cylindrical diffusing tip, housed within a 1.65 mm diameter saline-cooled polycarbonate cooling catheter.

Protocol

An institutional review board–approved protocol (Baylor College of Medicine, Houston, TX, U.S.A.) was used to enroll patients in this study at Texas Children's Hospital. The goal was to evaluate feasibility and safety of MR-guided SLA of hypothalamic hamartoma in a medically refractory patient population and study the clinical outcomes.

An MRI of the brain was acquired in all patients, diagnosing a hypothalamic hamartoma (Fig. 1). Video-electroencephalography (VEEG) monitoring of gelastic seizures was also acquired, along with detailed neuroendocrine and neuroophthalmologic evaluations. After detailed discussion of risks, benefits, and alternatives for treatment in these medically refractory patients, informed consent was acquired and the operation was undertaken. Overall, the procedure was divided into two portions: (1) placement of the laser probe into the HH using the CRW frame; and (2) MR-based confirmation of the probe location, with subsequent ablation.

Figure 1.

MRI of the hamartoma. (A) Sagittal T1 image shows its connection at the third ventricle. (B) Coronal FLAIR image shows that the hamartoma has more hyperintense signal, and shows greater attachment to one side.

The Visualase Thermal Therapy System (Visualase, Inc.) comprises a computer workstation, a 15 W 980 nm diode laser, a cooling pump, and a disposable laser applicator set composed of 400 mm core silica fiber optic with a cylindrical diffusing tip housed within a 1.65 mm diameter saline-cooled polycarbonate cooling catheter (Kangasniemi et al., 2004; McNichols et al., 2004a,b). The workstation uses an Ethernet connection to connect to the clinical MR scanner and retrieve reconstructed images from the scanner as soon as they are available. Extracted thermal data produce color-coded “thermal” and “damage” images based on an Arrhenius rate process model (Svaasand, 1995), which are displayed on the workstation. The damage image accounts for the cumulative effects of the time-temperature history of each voxel in the image. In addition to this visualization, the user interface allows the association of prescribed “limit temperatures” to specific target points on the image. If during treatment the computed temperature at one of these targets exceeds the associated limit temperature, a signal is sent to automatically deactivate the laser. These “limit points” are established by “clicking” on the images and may be moved in real-time during the therapy. The monitoring could be done in single plane or multiple parallel or orthogonal planes as required. An image from the display, showing the catheter in position, is shown in Fig. 2 prior to commencing ablation, and in Fig. 3 during ablation. Figure 4 shows the postablation-enhanced MRI.

Figure 2.

(A) The entire length of laser catheter and target is captured with an oblique coronal image. (B) Then, a test dose of 3 W for 30 s is used to confirm location of fiber.

Figure 3.

(A) Real-time temperature monitoring is shown with a laser exposure of 6 W for 40 s. (B) Arrhenius-based damage estimation in orange enables the surgeon to evaluate coverage of targeted hamartoma.

Figure 4.

Acute confirmation of ablation coverage is done with postcontrast T1 images, as seen in the coronal (A) and sagittal (B) images.

Postablation follow-up

All patients underwent a follow-up MRI 3–6 months after ablation and were examined by the surgeon and neurologist. Data were collected with respect to seizure recurrence, immediate versus delayed side effects or complications, and overall status, noting any new problems otherwise. Patients who had recurrent, incomplete, or no response to the treatment were offered re-treatment.

Results

The patient demographics, in terms of age, sex, seizure frequency, and lesion appearance with any prior treatments tried are listed in the Table 1. Posttreatment data, including hospital days and outcome, are also listed.

Table 1. Demographic and treatment response data for the HH patients treated in this series
AgeSexLesion (with prior treatment)Seizure frequencyHospital daysOutcomeFollow-up (mo.)
  1. *HH, hypothalamic hamartoma; M, male; F, female; TC, transcallosal; XRT, radiation therapy.

  2. Note that the initially listed patient was re-treated for suboptimal seizure control, and has since been seizure-free.

8 years (2×)aMHH3/day2, 11/day/Sz-free12, 6
9 yearsMHH2/h8Engel I24
15 yearsMHH3/week5Engel I20
5 yearsFHH5/day1Engel I18
7 yearsMHH6/day2Engel I13
8 yearsMHH4/day2Seizure-free8
4 yearsMHH4/day2Seizure-free7
22 monthsMHH (redo pending)4/day150% decrease7
9 yearsMHH (s/p prior TC resection, XRT)7/day1Seizure-free7
8 yearsMHH3/day1Seizure-free6
20 yearsFHH (s/p Gamma knife)4/day1Seizure-free5
3 yearsMHH (redo pending)15/day150% decrease3
3 yearsFHH10/day1Seizure-free2
9 yearsMHH (s/p prior TC resection)15/day1Seizure-free1

Overall, ages ranged from 22 months to 20 years, with a median of 8 years. The median and mode for length of stay were each 1 day. The initial success rate in terms of seizure freedom after a single treatment was 79%; one patient underwent a second ablation and is seizure-free at 6-month follow-up, increasing the overall seizure-free rate to 86% based on current follow-up varying from 1 to 24 months. If only patients with 6-month follow-up are considered, the seizure-free rate was 80% following their first ablation procedure, and 90% following both procedures (n = 10). Average follow-up was approximately 9 months.

From a complication analysis, one patient had a subclinical subarachnoid hemorrhage noted on MRI that was clinically asymptomatic, did not require intervention, and was self-limiting. No patients experienced postablation diabetes insipidus, memory impairment, other hormonal changes, hemiparesis, visual changes, or other significant complication.

Discussion

The incidence of hypothalamic hamartoma (HH) is rare, with about one in 200,000 affected (Alves et al., 2013). Common presentations include precocious puberty and gelastic seizures (Faizah et al., 2012). Left untreated, HH can result in a catastrophic epileptic encephalopathy with early onset gelastic seizures and cognitive regression; secondary refractory epilepsy can also result (Striano et al., 2012).

Open and endoscopic procedures have been described as resulting in seizure resolution in about 50–60% of patients (Pati et al., 2011). Various techniques have been employed, including stereotactic radiosurgery (SRS), endoscopy, and a variety of open surgical approaches (including orbitozygomatic and transcallosal techniques). At a single large institution, use of these techniques revealed that >90% of seizure reduction was noted in 19%, 50–90% seizure reduction in 48%, and 24% unchanged in seizure frequency upon reoperation. Reoperation for patients failing primary surgical treatment (including SRS) resulted in related complications for several, including hemiparesis, thalamic stroke (asymptomatic and symptomatic), hyperphagia, and pan-hypopituitarism. This study found that, in 157 patients analyzed, reoperation should be considered, as more than half the patients had significant reductions in seizure. In a series of HH in 165 patients, a study found that 16% required more than one treatment for their HH (Wait et al., 2011). Another study (Polkey, 2003), suggested that the transcallosal interforniceal approach is the most successful among open approaches, with 69% of patients seizure-free and an associated complication rate of 24%. In a series of 10 patients treated with SRS, 40% were seizure-free following only SRS, with an additional 20% seizure-free following additional therapy (Abla, 2010). However, there was a significant lag period of usually several months between treatment and observation of efficacy.

Stereotactic disconnection, where the entire hamartoma need not be removed, has been shown to be successful (de Almeida et al., 2008). The technique of SLA used in our patients similarly targets disconnection of larger HHs, by ablating the connection presumably responsible for propagation of the seizures. Smaller HHs are able to be completed ablated. This precision of stereotactic lesioning precludes other complications as seen with open surgery, such as hemiparesis, memory changes, and hormonal dysfunction. The only complication seen in this series, the presence of a traumatic subarachnoid hemorrhage related to catheter placement, is a known risk of stereotactic surgery.

Reoperation for HH due to recurrence of gelastic seizures has been described in a substantial number of patients (Wait et al., 2011). Up to 40% would potentially qualify, since initial treatment with open or endoscopic surgery and SRS succeeds in only 40–60% of patients. However, the actual number of patients or their families willing to consent to a second procedure is 19% (Wait et al., 2011). Our study found that of the two patients with incomplete seizure resolution, one is seizure-free without complication upon re-treatment and the second is scheduled and awaiting reablation.

The Visualase stereotactic ablation procedure has been used by various neurosurgeons to treat different diagnoses, ranging from primary and metastatic tumors to mesial temporal lobe epilepsy. Multiple centers have used this procedure, which has resulted in >300 patients being treated for various diagnoses. The device is FDA-cleared for soft tissue ablation, and surgeons have placed the laser probe stereotactically, using both frameless and frame-based techniques. Although HH remains a rare disorder, the sequelae in untreated patients can be devastating. As such, multiple centers have now adopted Visualase SLA for HH, and further series of patients will help acquire more data regarding proper use of this device for the future.

Conclusions

The evaluation of HH relies on careful history and documentation of symptoms. Gelastic seizures and precocious puberty are common presenting features. Uncontrolled gelastic seizures are commonly associated with a progressive and devastating epileptic encephalopathy.

Open surgical treatment has met with modest success, with a modest complication rate. Less invasive approaches, such as endoscopic and SRS techniques, have shown similar efficacy and improved complication rates. This analysis of 14 patients with intractable gelastic epilepsy treated for HH using Visualase laser ablation suggests an effective role, with 90% of the 10 patients with at least 6-month follow-up being seizure-free postablation. The lower risk profile of laser ablation, coupled with a typical length of stay of one hospital day, suggests further consideration of using this technique for patients with HH.

Disclosure

The authors have no conflicts of interest to disclose. The authors confirm that they have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Ancillary