Thoracic epidural analgesia in a patient with von Hippel‐Lindau disease

Abstract von Hippel‐Lindau disease (VHLD) is an autosomal dominant disorder characterized by central nervous system hemangioblastomas and renal tumors. Here, we report a case of thoracic epidural placement in a 35‐year‐old woman with VHLD presenting for left open heminephrectomy for renal masses. We also reviewed the literature on this topic.


| INTRODUCTION
von Hippel-Lindau disease (VHLD) has a reported incidence of 1:36,000 live births. 1 Although it is an uncommon condition, it is not unusual for an anesthesiologist to encounter these patients because they often have multiple surgeries or may be seen in the peripartum period. The hallmark features of VHLD are renal cysts and carcinomas, pheochromocytoma, and hemangioblastomas in the central nervous system and retina.
We report a case of a thoracic epidural placement in a young woman with VHLD who presented for heminephrectomy for a renal tumor. We also conducted a literature review to identify the considerations and practices pertaining to neuraxial anesthesia techniques in these patients.

| CASE PRESENTATION
A 35-year-old woman presented for left open heminephrectomy for enlarging renal masses suspicious for carcinoma. The patient's medical and surgical history included VHLD, panhypopituitarism after craniotomy for suprasellar hemangioblastoma resection, left open partial nephrectomy with epidural analgesia, right laparoscopic nephrectomy, left temporal craniotomies for seizures, and suboccipital craniotomy for fourth ventricular hemangioblastoma resection. She had no spine imaging studies performed prior to placement of her last epidural catheter. She had no preoperative neurological or metabolic symptoms. Her medications included acetaminophen, desmopressin, dexamethasone, hydrocortisone, lamotrigine, levothyroxine, lisinopril, and ethinyl estradiol norethindrone. Her most recent electrolytes and complete blood count results were normal with a platelet count of 260,000 mm 3 . Recent abdominal magnetic resonance imaging (MRI) and computed tomography (CT) scans did not demonstrate large spinal hemangioblastomas but were inadequate to detect smaller spinal hemangioblastomas. No formal spine MRI was previously or recently performed. For this surgery, an epidural was requested for postoperative analgesia by the surgical team, which is a common practice for major abdominal surgeries at our institute.
Our primary analgesic plan was for a paravertebral catheter with alternatives, including a transversus abdominis plane or quadratus lumborum blocks. These options, including epidural analgesia, were discussed with the patient. We did not favor epidural catheter placement because of a possible increased risk of spinal hematoma with the potential presence of spinal hemangioblastomas. The patient expressed a strong preference for thoracic epidural placement despite the risks because she had a perception of low pain tolerance and had had an uncomplicated epidural placement in the past. An epidural catheter was placed after considering her clinical history and our knowledge of the temporospatial nature of these hemangioblastomas. We were especially careful to avoid a dural puncture. Using anatomical landmarks, a 20G epidural catheter was sterilely placed at T7-8 via an 18G, 90 mm Hustead needle in one attempt, and without complications.
Her intraoperative course was uncomplicated. Intraoperatively, a 0.05% bupivacaine infusion at 8 ml/h was started one hour prior to extubation. Postoperatively, she was given patient-controlled analgesia of hydromorphone and the epidural infusion was increased to 10 ml/h. She was closely monitored with neurological assessment of her lower extremities every two hours for the first day after placement. We increased her epidural infusion rate to 14 ml/h on postoperative day (POD) 1, and eventually removed the epidural catheter on POD 3. She had no focal neurological deficits immediately after the epidural placement, on daily assessment with the epidural in place, at epidural removal, and up to 2 weeks postoperatively.

| DISCUSSION
The primary concern with neuraxial anesthesia in patients with VHLD is the potential risk of rupturing a spinal hemangioblastoma, which is a common feature of VHLD. Accordingly, we considered other alternatives to epidural analgesia such as a transversus abdominis plane (TAP), quadratus lumborum (QL), and paravertebral catheter placement. However, TAP 2 and QL catheters would require frequent boluses to achieve an adequate level of analgesia and, in this patient, would have been within the surgical field. Thus, these options were not viable. Because spinal hemangioblastomas can also occur at dorsal nerve roots in 0.3% of cases with VHLD, performing a paravertebral block does not eliminate the risk of hemangioblastoma puncture. 3,4 A prospective, randomized control trial by Schreiber et al. in patients undergoing liver surgery suggests that epidural analgesia provides a modest but significant improvement in pain control compared to paravertebral block catheters. 5 Therefore, an epidural was likely to be the most effective technique for postoperative analgesia in the presented case. However, the risk accompanied with epidural was difficult to estimate in the setting of no pre-procedural spine imaging.
An observational histopathological study suggested that in the transverse plane 60% of hemangioblastomas are intermedullary, 11% are intramedullary and extramedullary, 21% are intradural and extramedullary, and only 8% are extradural in location. 6 In a radiological observational study, 88% of 24 intermedullary tumors were located in the posterior aspect of the spinal cord. 3 Thus, considering the distribution, the risk might be small with an epidural catheter placement with the needle outside the dura mater. However, we also need to take into account the risk of dural puncture during epidural technique, and the incidence of unintended dural puncture has been reported to be 0.19%-3.6%. 7 In a prospective observational study of 1278 VHLD-associated craniospinal hemangioblastomas, 51% remained stable in size whereas 49% exhibited growth, and male sex was also found to be associated with a larger tumor burden and growth. 4 Based on: (1) the less distribution of spinal hemangioblastomas located in the extradural space 6 ; (2) the association of smaller hemangioblastomas being asymptomatic 3 ; and (3) our patient's gender, we perceived that the risk of epidural catheter placement in our patient would be acceptably low even without further imaging studies.
Despite an uncomplicated epidural placement in our patient, we remained inquisitive as to the information that might exist in the literature pertaining to neuraxial anesthesia techniques in patients in VHLD. We therefore performed a literature search using the EMBASE and MEDLINE databases for case reports or series in the English language whereby neuraxial anesthesia techniques were used or discussed in patients with VHLD. We used a combination of the keywords "anesthesia" or "epidural" in combination with "von Hippel Lindau," "von Hippel Lindau disease," or "hemangioblastoma." Our search yielded 413 articles of which 259 were duplicates. The abstracts or texts of the remaining 154 articles were reviewed and 22 articles were included for this literature review ( Figure 1) to answer the following questions: (1) Did practitioners obtain pre-procedural neuraxial imaging study(ies)? (2) Based on their experience, what recommendations had been made regarding performing neuraxial anesthesia techniques in patients with VHLD? (3) What were the outcomes in patients with VHLD who received neuraxial anesthesia techniques?
Did practitioners obtain pre-procedural neuraxial imaging study(ies)? Of the 15 patients who received neuraxial anesthesia, 14 had epidurals [8][9][10]12,13,[15][16][17][18][19][20]24,26,27 while 1 patient had spinal anesthetic for a cesarean section. 23 Of these 15 patients, 7 had an MRI of the spine reported, 13,16,17,20,23,24,27 while 5 had MRI/CT of the brain 9,10,15,18,19 with no details of spine imaging reported in the article. There were 3 patients without any imaging studies who received neuraxial anesthesia. 8,12,26 Thus, of the 15 patients who had a neuraxial anesthesia, only 7 (47%) had pre-procedural imaging of the spine (Figure 2). Expert opinion presented at the 2013 European Society of Regional Anesthesia Congress recommends that patients with VHLD should have an MRI performed as closely to the planned neuraxial anesthetic technique as possible and that neuraxial anesthesia should be avoided if imaging is unavailable. 29 Nevertheless, fewer than half of the patients (7/15) included in this literature review had imaging studies of the spinal cord prior to administration of neuraxial anesthesia.
What recommendations had been made regarding performing neuraxial anesthesia techniques in patients with VHLD? 5 groups did not provide any opinion or recommendations with regard to neuraxial anesthesia techniques in patients with VHLD. 8,10,19,24,28 10 groups suggested that neuraxial anesthesia can be considered for use in VHLD patients in the absence of contraindications . 1,12,[16][17][18]20,22,[25][26][27] 9 groups advised reviewing neurological imaging prior to performing neuraxial anesthesia techniques 9,13,16,18,20,22,[25][26][27] with 1 group specifically stating that updated neurological imaging should be acquired. 22 1 group stated that neuraxial anesthesia is an absolute contraindication in VHLD patients due to the potential presence of spinal hemangioblastomas 11 but most others stated that the diagnosis of VHLD should not completely exclude the use of neuraxial anesthesia.
What were the outcomes in patients with VHLD who received neuraxial anesthesia techniques? We found no reports of complications following neuraxial anesthesia techniques in the 15 VHLD patients included in this review.

| CONCLUSION
Spinal hemangioblastomas in patients with VHLD may be ruptured by neuraxial instrumentation. However, in the absence of spinal hemangioblastoma close to the site of needle entry, neuraxial anesthesia can be used safely. Nevertheless, there are no specific guidelines for neuraxial anesthesia, and recommendations and opinions differ among the reported literatures. The current evidence is insufficient to determine if neuraxial anesthesia is safe or contraindicated in VHLD in the absence of spine imaging. Therefore, the decision should be made on a case-by-case basis with the risks and benefits in mind.

T A B L E 1 Case reports on recommendations regarding neuraxial anesthetic techniques in von Hippel-Lindau disease
Author