Hemangioblastomas of the CNS are capillary-rich, circumscribed, benign neoplasms. Although benign, due to their location in the neuraxis, they can cause major morbidity and mortality due to mass effect of the lesion (as seen in the patient in Fig. 1), associated cysts, or from hemorrhage. CNS hemangioblastomas are the most prevalent neoplasms in vHL, developing in 60–90 % of patients [16]. These present at a younger average age than sporadic hemangioblastomas (mean of 29 years in vHL) and more often are multifocal [17]. Both sporadic and vHL-associated hemangioblastomas typically are infratentorial (Figs. 1 and 2). A prospective evaluation of 250 vHL patients with CNS hemangioblastomas demonstrated a median of eight lesions per patient, with 1 % supratentorial, 6 % in the brainstem, 45 % in the cerebellum, and the remaining 48 % caudal to that (spinal cord, cauda equina, and nerve roots). These patients developed a median of 0.3 new lesions/patient-year [16]. Although also infratentorial, a larger proportion (> 80 %) of sporadic CNS hemangioblastomas are cerebellar [17].

Other than presentation with a symptomatic lesion, identification of CNS hemangioblastoma occurs by imaging. The majority of lesions are clinically asymptomatic and there are no available screening laboratory studies. CT is not an adequate test for this entity and contrast-enhanced magnetic resonance imaging (MRI) is the preferred modality. Current guidelines for screening in vHL patients are for ≤ 1.5 T MRI scans with and without contrast to include the brain and whole spine with thin cuts through the posterior fossa. These are to start by age 16 and be performed every 2 years. This interval is then shortened in those with demonstrated lesions.

The majority of these lesions are asymptomatic and do not require intervention. However, due to the presence of multiple lesions per patient, many vHL patients will harbor at least one symptomatic lesion. At final evaluation in a prospective series, while only 6 % of all detected lesions had associated symptoms requiring intervention, they were distributed across one-third of the patient population [16]. The rate of growth of vHL-associated CNS hemangioblastoma varies with the anatomic location of the tumor, with more rapid progression for those in the brainstem and cerebellum [16]. These lesions may develop cysts in some cases (12 %) and the presence of a cystic component is associated with faster growth and a greater likelihood of developing symptoms [16].

It is not universally agreed when to intervene on these lesions, but most believe that they should be closely observed radiographically, and correlated with history or physical exam findings for specifically associated neurologic symptoms, and that onset of symptoms should prompt treatment of the causative lesion [16]. While there are some clinical variables that are associated with lesion progression [16], these are neither sensitive nor specific. Therefore, radiographic and clinical follow up is mandatory. When treatment of CNS hemangioblastoma is deemed necessary, surgical resection is the primary modality. Due to the morbidity and difficulty in surgical access to some of these lesions, stereotactic radiosurgery has also been employed. Although this therapy was well tolerated, stereotactic radiosurgery demonstrated diminished ability to control lesions over time. Therefore it is recommended that stereotactic radiosurgery be reserved for those symptomatic lesions not amenable to surgical therapy [18]. Although the vascular nature of the lesions raised hope that antiangiogenic therapy would be effective, there does not currently exist a demonstrated effective option for medical therapy.

While the presence of multiple CNS hemangioblastomas should very strongly raise the concern for vHL, an apparently sporadic CNS hemangioblastoma may be the index presentation of a previously unknown germline vHL mutation. In one series, this occurred in 4 % of patients otherwise thought to have sporadic CNS hemangioblastomas, all of whom presented at an age less than 40 years. A further 5 % of these patients who tested negative for germline mutation subsequently developed another vHL-related lesion, which may have been the result of mosaicism [19]. Given the importance of appropriate screening for other vHL-associated lesions, this is of clear importance for those patients in whom a germline vHL mutation is detected. For this reason, it is reasonable to submit all patients with retinal or CNS hemangioblastoma for vHL testing, even if these initially are believed to be sporadic lesions.

Hemangioblastomas also occur in the eyes of vHL patients and can be complicated by vision loss. These are frequently the earliest manifestation of vHL, and most vHL patients do not develop new retinal hemangioblastomas late in life. However, cumulatively, 70 % of vHL patients have retinal hemangioblastomas by age 60 and they frequently are bilateral. Compared to patients with sporadic retinal hemangioblastomas, they occur at an earlier age (18 vs. 36 years) and more frequently are multifocal and bilateral (about 50 %) [16, 20].

Retinal hemangioblastomas are not initially clinically apparent and are only detected by dilated ophthalmologic exam. Due to subretinal edema, traction, and hemorrhage, these lesions can result in vision loss. Monitoring and therapy aim to prevent or minimize vision loss. Treatment of most of these lesions is laser photocoagulation. There is not universal agreement and some recommend treating asymptomatic lesions while others recommend that small asymptomatic lesions may be closely monitored for growth or symptoms [16, 21]. As with CNS hemangioblastoma, investigations into antiangiogenesis therapies have demonstrated some potential role, but are not yet adequately studied.

Although not universally agreed upon, it is reasonable to offer vHL testing to patients with otherwise apparently sporadic retinal hemangioblastoma . In the presence of a known vHL germline mutation, screening for retinal hemangioblastoma is recommended annually starting at age 1 and in the hands of an ophthalmologist with expertise in these lesions.

Tumors of the endolymphatic sac (ELST) are an overgrowth of the papillary epithelium of the endolymphatic system located within the middle ear. These tumors are characterized as highly vascular and locally aggressive. While rare in the general population, they are detected in 11–16 % of vHL patients, 22 with an age at onset of 22–32 years [23]. vHL is the only condition associated with bilateral ELSTs [23]. Patients may present with acute-onset hearing loss, tinnitus, pain, vertigo, facial paresis, and aural fullness [22–24]. The mechanism for the audiovestibular dysfunction’s sudden onset is hypothesized to be acute intralabyrinthine hemorrhage. However a more gradual decline in hearing secondary to otic capsule invasion and/or hydrops formation also is reported [22, 23]. Regardless of mechanism, hearing loss usually is permanent.

Screening for ELSTs begins at birth with the routine newborn hearing screening. Periodic assessments of hearing should be performed by the child’s pediatrician. Starting at age 5, a formal audiology examination should be completed every 2–3 years, and more frequently if symptoms arise. The vHL Alliance also recommends an MRI with contrast of the internal auditory canal with thin cuts in cases of recurrent ear infections every 2 years after age 16 which is concurrent with the exam for CNS hemangioblastoma [25]. Indications for surgery include patients with radiologic evidence of a tumor on computed tomography (CT) or MRI that either have preserved hearing, or for deaf patients with persistent neurological symptoms associated with an ELST [22]. Surgical resection does not improve hearing but has been shown to prevent further damage, and improve other associated neurological symptoms. Of vHL patients who complain of audiovestibular symptoms, 60 % do not have findings on imaging and surgical intervention of these patients remains controversial [22].

Pheochromocytoma has a reported incidence of 0.1–0.5 % in the general population which correlates to 2–8 people per million per year. The “rule of 10s” previously supported the belief that approximately 10 % of pheochromocytomas occur in the setting of heritable familial syndromes. This has proven to be an underestimate as more syndromes have been recognized and testing has become more frequent [26]. Some estimates suggest that 24–30 % of sporadic pheochromocytomas occur within familial syndromes including vHL, neurofibromatosis 1, multiple endocrine neoplasias type 2, and succinate dehydrogenase mutations [27, 28]. More recently, this statistic has been contested citing referral and catchment area bias, and revised estimates are between 11.6 and 19 %. Within vHL patients, the rate of pheochromocytomas is 10–20 % [14]. However, the true risk of developing a pheochromocytoma depends on the type of vHL, varies among kindreds, and ranges from 0 % to upward of 57 %.

The demographics and manifestations of pheochromocytoma in vHL differ significantly compared to patients with sporadic disease, as well as when compared to other familial pheochromocytomas . The mean age of presentation is 30 years but has been reported prior to the age of 10 [14]. Pheochromocytomas in vHL patients are less likely to present with the classic signs and symptoms (palpitations, headache, diaphoresis, hypertension, tachycardia, pallor, and nausea) than sporadic pheochromocytoma. Over one-third of pheochromocytoma in vHL are clinically asymptomatic. Most pheochromocytomas in vHL are within the adrenal gland and up to 50 % of vHL patients have bilateral intra-adrenal disease. However, extra-adrenal paragangliomas also are more prevalent compared to sporadic disease, with 12 % of vHL patients having extra-adrenal disease, and some patients having bilateral intra-adrenal as well as extra-adrenal disease (Fig. 2). Paragangliomas can arise in the glomus jugulare, carotid body, and the periaortic tissues. Pheochromocytoma in vHL is less likely to have malignant transformation than pheochromocytoma in sporadic disease, with a rate of 3.3 %.

Patients with a diagnosis of vHL or at-risk family members with a known carrier status for a germline mutation should undergo routine screening for pheochromocytoma regardless of their disease subtype. Screening for pheochromocytoma should include both biochemical and radiographic testing. Biochemical screening is recommended to begin annually no later than 5 years of age with some studies suggesting initiating screening even earlier [14]. Serum chromogranin A is a nonspecific marker with poor sensitivity and is not a standard part of screening. Plasma-free metanephrines and normetanephrines are used as the first-line screening tool in vHL as in sporadic disease with a sensitivity of 97–100 % and specificity of 85–89 %. As with testing in sporadic disease, attention must be given to the medication list and testing environment to limit the rate of false-positive results. If the results from the serum testing are equivocal, the patient should conduct a 24-h urine collection for catecholamines. When performed correctly, urine biochemical studies have a sensitivity and specificity of 98 and 98 %, respectively, for pheochromocytoma; however, they are cumbersome for the patient to collect and are prone to errors in collection.

Regardless of the method of testing, pheochromocytoma in vHL has a predictable biochemical signature, with elevations in norepinephrine and normetanephrine, without elevations in epinephrine and metanephrine levels (as demonstrated by both patients in Figs. 1 and 2). In contrast, sporadic disease and other familial syndromes such as multiple endocrine neoplasia (MEN) have concurrent elevations of epinephrine and metanephrine [29]. The reason for primary production of norepinephrine in vHL patients is the relative lack of phenylethanolamine-N-methyltransferase (PNMT) responsible for the conversion of norepinephrine to epinephrine found exclusively within the adrenal medulla. vHL patients also express less tyrosine hydroxylase (the rate-limiting enzyme in catecholamine synthesis) than do patients with MEN2.

The age for initiating radiographic imaging for intra-abdominal manifestations of vHL is no later than 8 years of age. Family history and physical examination may suggest the need for earlier imaging as well as biochemical testing on a case-by-case basis and left to the discretion of the individual’s physician. The imaging modality of choice for screening of at-risk (known mutation carriers with no previous intra-abdominal lesions and negative biochemical studies) and asymptomatic established vHL patients is abdominal ultrasonography. If biochemical abnormalities are found, the most appropriate follow-up imaging study is an abdominal MRI in an effort to minimize the cumulative ionizing radiation exposure over a lifetime of repeated imaging studies. This is in contrast with sporadic disease, which can be evaluated with a dedicated adrenal protocol CT of the abdomen and pelvis with intravenous and oral contrast if there are no allergic or renal contraindications to intravenous contrast. CT and MRI have been shown to be equally efficacious.

In cases where CT/MRI are negative but there is high clinical suspicion or abnormalities in biochemical studies, a metaiodobenzylguanidine (MIBG) scan can be performed for localization with sensitivity of 95 % [30]. Recently, using CT/positron emission tomography (PET) with 18F-Dopa to localize pheochromocytomas has been studied. Small studies have shown a nearly 100 % sensitivity for pheochromocytomas. Unfortunately, the availability and acceptance of the test have been inhibited in part due to high cost and difficulty of synthesizing the 18F-Dopa [31]. All imaging modalities will have a lower sensitivity for small-volume disease.

Some image-detected pheochromocytomas in vHL do not result in elevated biochemical levels of catecholamines and there is no consensus on intervention or observation in that scenario. While some will intervene on any detected pheochromocytoma, others advocate withholding surgical intervention until the 24-h urine catecholamine excretion is double the upper limit of normal, or to remove nonfunctional pheochromocytoma if the tumor has grown to greater than 3.5 cm in diameter, the patient is to undergo an abdominal operation for another pathology, or if the patient desires to become pregnant or becomes pregnant [32].

Traditionally, the treatment of intra-adrenal pheochromocytoma is ipsilateral adrenalectomy, which classically was performed using an open technique by either a transabdominal or a posterior approach. With the advent of laparoscopy, today’s standard is a laparoscopic approach by either the lateral transabdominal or the posterior retroperitoneal approach, depending on the preference and expertise of the surgeon and center. There also are experiences with single port, robotically assisted, and single port robotic approaches to this procedure through both the transabdominal and retroperitoneal approaches. It is the opinion of these authors that there is at most marginal superiority of any one of these variations on minimally invasive adrenalectomy, beyond the significant benefit that resulted with dissemination of minimally invasive adrenalectomy techniques in general. Regarding the transabdominal versus the retroperitoneal approach, we select the approach best suited to the patient’s anatomy and prior operative history.

Traditionally, adrenalectomy is performed as a complete extirpation of the gland in question. Either sequential or simultaneous total adrenalectomy renders the patient surgically an-adrenal, and thus completely dependent on exogenous steroids, with associated quality-of-life impairment and risk for Addisonian crisis. This approach mandates thorough patient counseling and education. More recently, multiple large case series have demonstrated partial adrenalectomy (i.e., cortical-sparing adrenalectomy) via open, lateral or posterior laparoscopy, and robotic-assisted approaches to be safe and efficacious [33–35]. This involves removing the target lesion from the adrenal gland and leaving behind the grossly and radiographically normal portion of the gland. The main adrenal vein may be spared or sacrificed depending on the location of the tumor. The venae comitantes running with the arterial supply are adequate to maintain drainage even if the main adrenal vein is sacrificed, and as little as 10 % of a viable single adrenal gland has been shown to maintain adequate function to handle stress without requiring exogenous steroids [36]. Cortical-sparing adrenalectomy has been validated in the pediatric population with hereditary forms of pheochromocytoma [37]. The ability to leave a viable cortical remnant allows patients the opportunity to retain adrenocortical function, without dependence on exogenous steroids, while achieving comparable biochemical responses. Rates of Addisonian crises and long-term steroid replacement are significantly lower compared to complete adrenalectomy [33, 36, 38, 39]. As expected, recurrence rates are higher and time to recurrence is shorter with partial adrenalectomy. Outcomes data suggest these rates to be within an acceptable range and successfully treated with surveillance screening and additional surgery. The National Cancer Institute published their experience on hereditary pheochromocytoma as having a 6 % recurrence rate, which has been reproduced at other high-patient volume centers [38, 40]. In patients with vHL, the recurrence rate after partial adrenalectomy has been reported to be from 10 to 20 % after at least a 5-year follow up [39, 41]. The recent acceptance of cortical-sparing adrenalectomies makes it impossible to comment on the rates of patients that subsequently develop malignant pheochromocytoma and underscores the ongoing need for lifelong surveillance after surgical resection. In instances where additional pheochromocytomas develop, the benefit of adrenal-sparing surgery lies in delaying the development of adrenal insufficiency and subsequent lifetime duration of exposure to exogenous steroid. Therefore, the decision to perform cortical-sparing adrenalectomy versus total adrenalectomy must be made with consideration of the patient’s willingness and ability to tolerate adrenal insufficiency versus recurrence of disease with subsequent reoperation. The ultimate success of both approaches demands ongoing patient follow up.