Interdisciplinary approach allows minimally invasive, nerve-sparing removal of retroperitoneal peripheral nerve sheath tumors

Neurosurgeons have developed and refined microsurgical, nerve-sparing techniques to remove benign peripheral nerve sheath tumors (PNST) [1]. Typically, PNST surgery involves the following steps: (1) identification of the tumor pseudocapsule consisting of the epineurium and viable nerve fascicles which are stretched and thinned out around the tumor; (2) incision of the pseudocapsule, avoiding viable nerve fascicles in the pseudocapsule; and (3) extirpation of the tumor after dissection of the plane between the pseudocapsule and the true tumor capsule.

These techniques have enabled neurosurgeons to perform gross total tumor removal of PNST in almost every localization while improving preoperative symptoms, preserving neurological function, and obtaining excellent oncological results [2].

Retroperitoneal PNST, originating from lumbosacral nerve roots or the lumbar plexus, are a notable exception. Given that PNST are just one of the many potential tumor varieties in the retroperitoneal space [3], retroperitoneal PNST are typically treated by surgeons from other disciplines such as visceral surgery and oncological surgery.

There are two main strategies in dealing with histology-proven PNST, specifically, en bloc resection with or without margin resection and radiological monitoring. A recent retrospective multicenter study studied growth dynamics of 248 retroperitoneal PNST, which were initially subjected to radiological monitoring [4]. The average annual growth rate was 10.5%, prompting the authors to recommend early surgery in cases of symptomatic tumor, diagnostic uncertainty, and existing evidence of rapid expansion or patient preference after adequate counseling.

Complete excision is the therapy of choice but considerable controversy exists over timing of surgery and negative soft tissue margins [4, 5]. Aggressive en bloc resection to avoid local tumor recurrence and/or malignant transformation is recommended by national guidelines, such as the German guidelines for malignant soft tissue tumors. In addition, guidelines [6] recommend a one-specimen en bloc resection as the surgery of choice for histology-proven retroperitoneal malignant peripheral nerve sheath tumors (MPNST). Ideally, this includes resection of adherent structures, even if not evidently infiltrated. Preservation of specific organs/neurovascular structures should be considered on an individualized basis. En bloc resection of the tumor, however, inevitably leads to loss of function of viable nerve fascicles in the tumor pseudocapsule. Even though some authors advocate a less radical retroperitoneal PNST resection to limit neurological morbidity [5], this approach typically still involves resection of nerve fascicles trapped in the tumor pseudocapsule and can be associated with severe impairment of neurological function.

There is a void of information regarding neurological outcome after retroperitoneal PNST resection. It is likely, however, that neurological morbidity after radical retroperitoneal PNST resection is substantially higher than neurological morbidity after neurosurgical, nerve-sparing PNST removal.

In our department, retroperitoneal PNST are routinely treated by an interdisciplinary team consisting of a visceral surgeon and a neurosurgeon, i.e., combining the general surgeon’s skills to navigate the complex anatomy of the retroperitoneal space and the neurosurgeon’s familiarity with microsurgical, nerve-sparing tumor resections. We hypothesize that such an interdisciplinary approach is best suited to improve preoperative symptoms and maintain neurological function [7] and offers the possibility to move from a microsurgical, nerve-sparing resection to a more aggressive, oncological resection in rare cases were MPNST can be diagnosed with intraoperative histology. In addition, our study is motivated by three considerations: (1) Does retroperitoneal PNST resection achieve a similar oncological outcome compared with radical tumor resection?; (2) Is our minimally invasive tumor resection, associated with less surgical morbidity, e.g., less intraoperative blood loss? ; (3) Does our interdisciplinary approach prolong surgery compared with a monodisciplinary tumor resection?

Here, we conduct a retrospective evaluation of 18 patients suffering from retroperitoneal tumors, including 16 PNST and 2 MPNST, who were treated at our department between 2009 and 2018 by an interdisciplinary surgical team. Specifically, we evaluate neurological as well as oncological outcome, and study surgical morbidity and operative times.

PNST are benign tumors that rarely recur after tumor removal and have no impact on a patient’s life expectancy or quality of life. As such, guidelines generally recommend nerve-sparing PNST removal. Retroperitoneal PNST, however, are rare, difficult to diagnose preoperatively, and challenging to operate [3, 8]. In addition, retroperitoneal PNST are treated by surgeons from a variety of disciplines such as visceral surgery, oncological surgery, urology, and gynecology, which reduces the average number of cases that any given surgeon will see in a lifetime to just a handful of tumors.

Given the rareness of retroperitoneal PNST, their proximity to vital visceral organs and blood vessels as well as potential differential diagnoses such as sarcomas, there is no standard treatment for retroperitoneal PNST.

As such, it is difficult to conceive a prospective study comparing radical and nerve-sparing retroperitoneal PNST resection in terms of neurological outcome, surgical complications, and oncological outcome. Therefore, our study and previous reports are limited to retrospective case series and any comparison between studies should bear this limitation in mind.

We further considered, whether our minimally invasive tumor resection was associated with less surgical morbidity, e.g., less intraoperative blood loss, which has been consistently identified as a major intraoperative problem in previous studies [5, 9].

During our interdisciplinary approach, dissection of major blood vessels was kept at a minimum, especially since PNST often adhere to major blood vessels. Once the tumor pseudocapsule had been identified, the neurosurgeon continued the operation, incised the pseudocapsule, and extirpated the tumor, thus reducing the risk of intraoperative hemorrhage. It is important to mention that in n = 3/18 cases, tumors were in direct contact with iliac vessels and required careful mobilization of vessels to expose the tumor. Once the tumor pseudocapsule was incised, tumors could be removed in safe distance from the iliac vessels, i.e., the pseudocapsule served as plane separating the tumor site from blood vessels. Still, in two of these cases, ruptured branches of the inner iliac vein, which were draining the tumor, had to be sewed by the general surgeon. In one of these cases, major blood loss (3800 ml) led to blood transfusion. Assistance of vascular surgeons, however, was not required. To our knowledge, other studies have not systematically studied blood loss after retroperitoneal PNST removal, but incidents with massive, sudden blood loss [5, 9] and tearing of adjacent arteries have been reported 8]. In our series, mean intraoperative blood loss was 680.6 ml (95% CI, 194.3–1167.0 ml) and was comparable to blood loss reported after open resection of other benign retroperitoneal tumors [10].

We did not record any intraoperative surgical or anesthesiologic complications and our low postoperative surgical morbidity was in line with complication rates reported in other studies on retroperitoneal PNST resection [8, 9, 11] and in studies on resection of other benign retroperitoneal tumor entities [10].

Additionally, mean operative time for our interdisciplinary approach was 162.5 min (95% CI, 121.6–203.4 min). While other studies have not systematically studied surgery durations, our mean operative times are comparable to that of other surgical studies reporting on resection of benign retroperitoneal tumor entities. Liu et al., for instance, reported mean operative times of 152.7 min ± SD 72.4 min for open resection of benign retroperitoneal tumors, including PNST [10]. We thus hypothesize that our interdisciplinary approach is, at least, comparable to “classical” retroperitoneal PNST resection in terms of surgical morbidity and that our interdisciplinary approach is not more time-consuming than traditional, monodisciplinary surgery.

At the same time, gross total tumor removal could be achieved in 88% of our PNST patients (n = 14/16 patients), which is well in line with two other studies, which reported complete tumor removal rates of 73% (n = 60/82 patients) [9] and 85% (n = 17/20 patients) [11]. In the former study, tumor enucleation was followed by tumor (pseudo) capsule removal. In the latter study, the surgical method was not explicitly mentioned. Most likely though, “classical” tumor resection was performed, as one patient underwent resection of a contiguous organ and one patient received only subtotal resection due to tumor adherence to the iliac vein. Finally, a third study reported a complete resection rate of 100% after en bloc tumor resection but included 7 patients only [8].

In terms of follow-up after complete PNST removal, all totally resected patients were tumor-free at follow-up and patients that had been subjected to partial tumor resection were diagnosed with static tumor remnants. With a mean follow-up of 24.9 months, our study is within the follow-up range of other clinical series, and oncological results are in line with other studies on retroperitoneal PNST. Reported rates of recurrence after total resection have been as low as 1.2% (n = 1/81 patients, mean follow-up 63 months) [9] and 0% (0/7 patients, mean follow-up 17 months) [8]. In the former study, stable tumor burden after partial tumor resection was reported for 8/8 patients after a mean follow-up of 9.5 years.

Finally, and in contrast to another report [8], preoperative CT-guided needle biopsy in our study was accurate in 4/4 cases, i.e., biopsy results were confirmed by intraoperative pathology. We will thus continue to employ CT-guided needle biopsies in cases with unclear retroperitoneal pathology. In addition, we will expand their use in the light of new and/or updated guidelines for the treatment of sarcoma [6] and especially, in order to avoid cases of unplanned MPNST resections.

We must point out, though, that biopsy recommendations in current sarcoma guidelines are primarily based on studies focusing on tumors of the trunk and of the extremities [12‐14]. Some of these studies did not include retroperitoneal tumors or excluded patients with impalpable, deep-seated lesions from their final analysis [13], and others included a small number of retroperitoneal tumors only [12]. At the same time, Dupuy et al. reported that biopsy accuracy depended on tumor localization and reported a mismatch rate of 20% in the spinal and paraspinal region [15]. In these cases, biopsies had retrieved unrepresentative tissue and radiologists arguably felt less comfortable using longer and larger needles and performing multiple passes [15].

Altogether, we believe that there is little knowledge on the diagnostic accuracy of core-needle biopsies in cases of retroperitoneal tumors, and guideline recommendations are based primarily on experiences with sarcomas of the trunk and of the extremities.

Finally, a main advantage of our interdisciplinary approach for PNST resection is the possibility to move from a microsurgical, nerve-sparing resection to a more aggressive, oncological resection in those rare cases were MPNST can be diagnosed with intraoperative histology. And even in cases with unplanned MPNST resection, re-resection seems to be feasible without influencing overall survival [16].

Interdisciplinary, nerve-sparing enucleation of retroperitoneal PNST is well suited to improve preoperative symptoms and maintain neurological function, while achieving results similar to other series on radical retroperitoneal PNST resection, in terms of oncological outcome, surgical morbidity, and operative times.