Key points

Introduction

The history of lateral skull base tumor management is best understood through the history of acoustic neuroma (AN) surgery. Like most lateral skull base lesions, ANs were initially thought to be unresectable. Surgical resection of lateral skull base tumors had a tumultuous course before arriving at the extremely low mortalities seen with modern surgery. As understanding of these lesions continues to advance so does our treatment algorithm. This article describes the major turning points in the treatment of lateral skull base lesions and how treatment strategies continue to evolve.

Introduction

The history of lateral skull base tumor management is best understood through the history of acoustic neuroma (AN) surgery. Like most lateral skull base lesions, ANs were initially thought to be unresectable. Surgical resection of lateral skull base tumors had a tumultuous course before arriving at the extremely low mortalities seen with modern surgery. As understanding of these lesions continues to advance so does our treatment algorithm. This article describes the major turning points in the treatment of lateral skull base lesions and how treatment strategies continue to evolve.

Early history of acoustic neuroma surgery

ANs were among the first lesions to be anatomically localized based on symptomatology alone. ANs therefore played a major role in neurosurgical history. The earliest descriptions of cerebellopontine angle lesions causing deafness and facial numbness date back to the mid–nineteenth century. However, it was not until the later part of that century that the first attempts at AN resection were made.

The first reported surgical attempt was by Charles McBurney in 1891, who opened a suboccipital plate with a mallet and chisel. Because of excessive cerebellar swelling, no tumor could be removed and the patient died 12 days later. Soon thereafter (1894), an account from Charles Ballance described the first successful complete removal of an AN. Exposure was performed through a suboccipital approach and tumor was removed using blunt finger dissection. During this earliest era of AN surgery, the surgical mortality approached 80%, with high operative morbidity affecting the few patients who survived. It is clear that if posterior fossa surgery were to continue, significant advances would be needed.

More modern surgical techniques were ushered in during the Harvey Cushing era (early 1900s). His technique included using an extended bilateral suboccipital approach and a belief that only the core of the tumor could be safely resected. He postulated that, by leaving the tumor capsule intact, cranial nerve function would be preserved and the brainstem vasculature would be left undisturbed. Cushing also advocated for the use of delicate and meticulous surgical technique, decreased cerebellar retraction, and the importance of relieving cerebrospinal fluid pressure to improve operative space. Along with the use of bone wax and electrocautery for hemostasis, this allowed him to decrease the surgical mortality to 20% in his lifetime.

Although there was significant overlap with Cushing, the Walter Dandy era saw the increased use of diagnostic imaging to guide surgical technique. In 1918, Dandy invented ventriculography, which allowed neurosurgeons to identify the approximate location and size of brain tumors for the first time. This ability was further advanced 1 year later by his development of pneumoencephalography. In addition, Dandy, developed the more modern unilateral suboccipital approach using a smaller bone flap. He also created a schism between himself and Cushing by advocating and successfully performing total tumor excision. Improving on Cushing’s advances, Dandy was able to advance the modern surgical era.

The introduction of the operative microscope by William House in 1961 changed neurosurgery forever. This microscope allowed improved viewing of the cranial nerves and blood vessels, permitting more complete tumor resection with decreased morbidity. It made facial nerve preservation with tumor resection possible for the first time. This improved vision paved the way for House to fully develop the translabyrinthine approach; unknown to House, this had been described by Panse (1904) and Quix (1911) decades earlier. The microscope also allowed the development of the transcochlear approach as well as hearing preservation surgery in the form of the middle cranial fossa and the modern retrosigmoid approach.

Before William House’s era, there was no cooperation between otologists and neurosurgeons in posterior fossa surgery. They were often vehemently opposed to their counterparts performing these surgeries. This lack of cooperation changed when William House invited William Hitselberger to join forces in treating posterior fossa lesions. The two spent countless hours together in the temporal bone laboratory cross-training to gain from each other’s expertise. Over time, they created the modern collaborative neurosurgeon/neuro-otologist relationship that is now used across the world. It is easy to take this working relationship for granted in the modern era, but it is important to remember that it arose from the hard work and dedication of these two men.

The surgical treatment of meningiomas, glomus tumors, facial nerve schwannomas, petrous apex lesions, and other lateral skull base lesions was also similarly affected by the surgical advances discussed earlier. The modernization and adaptation of these surgical approaches, along with use of the operative microscope, allowed surgeons to surgically address these lesions with improved outcomes.

Imaging

Without radiological advances, our current understanding of the size and growth characteristics of lateral skull base lesions would be deficient. It became possible to identify moderate-sized tumors with the use of positive contrast cisternography. Later, the development of the Polytome Pantopaque technique allowed the diagnosis of smaller lesions confined to the internal auditory canal. Although all these advances were helpful at their time, the widespread use of computer tomography and eventually MRI in the 1980s greatly improved the characterization of these lesions and ultimately became the primary diagnostic imaging study. The improved identification of AN, along with improved MRI resolution, allowed physicians to better understand the natural history of skull base lesions and intervene at an earlier stage than was previously possible.

Natural history of skull base lesions

Perhaps the greatest breakthrough in the management of lateral skull base tumors has been a better understanding of the natural growth rates of these lesions. Before this, nearly all lesions were surgically resected on diagnosis regardless of symptoms. It is now known that meningiomas, facial nerve schwannomas, and glomus tumors follow similar slow growth patterns to AN.

Multiple studies have reported the overall growth rates of ANs to vary widely, between 6% and 73%. However, when data have been collected in meta-analyses, 51% of ANs show no growth, 43% to 54% grow, and 4% to 8% regress in size.

Data from the Denmark Prospective Observational Study show that 17% of intrameatal and 28.9% of extrameatal ANs showed growth ( P <.001). There was also a clear predilection toward growth in the first years of observation. Of those tumors that grew, most (62%–64%) grew during the first year, 23% to 28.9% grew during the second year, 5% to 10% grew during the third, and 0% to 2% grew during the fourth year of observation. When quantified, the mean growth rate for all tumors is 1 to 2 mm/y. For only those tumors that grow, this rate increases to 2 to 4 mm/y. When analyzing these data, it is important to realize that there is significant selection bias. Only those tumors that could be observed without intervention were included. Therefore, any tumor that was deemed too large or symptomatic for observation (and presumptively had grown before time of presentation) was not included in the study. Therefore, the true AN growth rate may be higher than reported.

However, an improved understanding of the natural history of AN has greatly altered the treatment paradigm. Surgery was no longer the only option because a watch-and-wait approach was seen to be safe and advisable in a select group of patients. This same concept was eventually applied to other lateral skull base lesions with slow growth patterns.

Radiotherapy

The next major advancement in treatment of lateral skull base lesions was the invention of stereotactic radiosurgery by Dr Lars Leksell. The first Gamma Knife (GK) unit was installed in Stockholm, Sweden, in 1968 and the first in the United States was at the University of Pittsburgh in 1987. With the publication of their seminal New England Journal of Medicine article, Kondziolka and colleagues showed GK to be a viable treatment option for ANs. Although initial control rates were excellent, the tumor margin radiation doses ranged from 14 to 20 Gy. With these large treatment doses, there was significant associated morbidity. Some degree of facial nerve neuropathy was seen in 21% of patients and 27% developed trigeminal nerve symptoms that were statistically associated with higher doses of radiation ( P <.003).

However, over time, the marginal tumor doses have be deescalated to 12 to 13 Gy with consistent tumor control rates greater than 90%. Compared with studies with larger radiation doses, it seems that there is a trend toward decreased tumor shrinkage but persistent tumor stability. In contrast, there has also been a dramatic decrease in the frequency of cranial neuropathy.

Understanding of hearing loss after radiosurgery continues to evolve. For years the literature was troubled by inconsistent methods of measuring hearing preservation in addition to inadequate follow-up periods. It was initially thought that hearing loss stabilized 2 to 3 years after treatment. Numerous early studies showed that after this time period the functional hearing can be preserved in 50% to 100% of patients. It is now apparent that those with serviceable hearing before treatment continue to show deterioration over time. Recent studies have shown that serviceable hearing preservation rates 8 to 10 years after treatment are 23% to 34%.

Stereotactic radiosurgery has now been used to treat numerous lateral skull base lesions. However, perhaps the greatest impact has been on glomus jugulare tumors. Stereotactic radiosurgery has become the first line of treatment at many centers for growing symptomatic tumors because of decreased morbidity compared with surgery and comparable control rates.

As understanding of the long-term outcomes of stereotactic radiosurgery become more complete, the treatment algorithm may continue to change. Radiosurgery is indicated in a select group of patient and has shown excellent control rates. However, the risks and expected outcomes should be fully understood by the recommending physician and patient before treatment.

Surgery

With application of radiosurgery and a better understanding of the natural history of lateral skull base lesions, the indications for surgery have greatly changed over time. Before this, nearly any lesion of the lateral skull base was resected on diagnosis. Since the 1960s, the 3 main approaches to the lateral skull base (middle fossa, retrosigmoid, and translabyrinthine approach) have increasingly been refined. The creation of accurate and real-time cranial nerve monitoring, specifically facial nerve and auditory brainstem response, have improved safety and surgical proficiency. In addition, continued advances in microscope technology have improved microscopic vision, and enhanced instrumentation such as the laser and ultrasonic aspirators have greatly aided tumor resection. Collectively, these improvements have resulted in a reduction in overall mortality to 0.5%.

As technology and surgical technique have improved, there has been an increased emphasis on functional preservation rather than complete resection. It has been clearly reported that postoperative facial nerve function is likely to be worse with large tumors. This finding has led many institutions, including our own, to prefer near-total or subtotal resections to complete resection with facial nerve injury.

In our institution’s data for patients with ANs greater than 2.5 cm, there is a clear trend for patients to have worse facial nerve function when receiving gross total resection. Comparing facial nerve outcomes in those undergoing a near total resection, subtotal resection, and gross total resection, the percentage of those having House-Brackmann grades I and II were 97%, 96%, and 77% respectively ( P <.001). Patients undergoing near-total resections and subtotal resections had higher rates of tumor regrowth than gross total resection but both groups had a low need for further treatment (2% and 10% vs 0%; P <.001). Although some clinicians advocate planned stereotactic radiation for incompletely resected large tumors, we do not. Given the low need for retreatment seen earlier, it is our belief that, when a true subtotal or near-total resection is performed, the decreased vascularity of the tumor results in a low likelihood of tumor growth.

Similar less aggressive surgical approaches have been applied to other lateral skull base lesions. For example, surgery for facial nerve schwannomas is currently limited to decompression when patients have good facial nerve function and resection with nerve grafting when facial nerve function is poor. This approach is used because of poor facial nerve outcomes when partial resections are performed. Glomus jugulare tumors have had a similar treatment progression. Rather than risk significant lower cranial nerve morbidity, surgical resection is limited to preservation of the lower cranial nerves when normal function exists. Observation versus postoperative stereotactic radiation is often used after surgery in these cases to minimize lower cranial nerve injury.