Introduction

PGLs are extremely rare, the incidence is less than 1 in 30000. Even though typically benign and curable by surgical resection, up to 6% can be malignant¹. Scoring systems based on morphology and clinical features to assess malignant potential have been proposed but have yet to be fully validated². Approximately 50% occur in the head and neck region, most frequently as hypervascularized carotid body. Patients may thus notice a slow growing and painless lateral neck mass, pulse-like sensations or voice changes³.

Carotid Body PGLs have been traditionally subdivided in three different types: Sporadic, Familial and Hyperplastic. Representing around 85% of Carotid Body PGLs, the Sporadic form is the most common type.

To this day, three genes associated with Familial PGLs have been identified. All three encode subunits (D, B and C) of the enzyme succinate dehydrogenase complex (SDH), which plays a pivotal role in both the Kreb’s cycle and electron transport chain. It has been postulated that a defective succinate dehydrogenase can cause an intracellular increase of vascular endothelial growth factor (VEGF) and hypoxic molecular mediators, leading to hyperplasia, angiogenesis and neoplasia⁴.

It’s estimated that up to 70% of familial PGL cases carry germline mutations in one of the three SDH genes. In reality, it’s likely that this percentage is even higher because most studies don’t test for large deletions.

Surgical resection is recommended for most CBTs in healthy patients because of the risk of local complications related to tumor size and a small but definite risk of malignancy. The surgical technique has remained unchanged throughout the years. The most common complication associated with surgical resection is cranial nerve injury which can occur in 19% to 49% of patients. The larger and more adherent the tumor, the higher the risk of cranial nerve injury. Fortunately most cranial nerve deficits are usually temporary⁵.

Methods

Forty six patients diagnosed with CBT in the Angiology and Vascular Surgery Departments of 9 different hospitals in Portugal between 1989 and 2017, were retrospectively evaluated. All patients underwent ultrasound, magnetic resonance or computerized angiotomography imaging preoperatively and the confirmation of the diagnosis of CBTs was made by histopathologic examination after surgical resection. All patients had unilateral tumors except for two who had bilateral carotid tumors. Data on demographics, imaging diagnostic, family history, clinical presentation, surgical treatment, complications, and the patients outcome were retrieved from the clinical files. In addition, the tumour classification according to the Shamblin type was also recorded. For the present study, all these identified patients were contacted and, after their informed consent, a blood sample was taken for genetic testing on SDHx genes.

Statistical analysis was performed using the IBM Statistical Package for the Social Sciences (SPSS) 22. Χ2 test was used to assess differences between categorical variables. Normality of continuous variables was first assessed by the Kolmogorov-Smirnov test and statistical differences between groups were analyzed either through the t-Student test or the Mann-Whitney U test.

Results

A total of forty six patients 21 male and 25 female ; mean age of 53,1±16,6 years (15-84) were retrospectively evaluated. There were no differences in gender or age of onset between patients carrying SDHx mutations and non-mutation carriers.

The genetic analysis identified seven different heterozygous mutations in SDHD gene in eleven patients (24%). Of these, four were missense mutations, previously reported in Paraganglioma patients, (NP002993.1:p.Met1Val, p.Met1Ile, p.Gly12Ser, p.Pro53Leu), one was a novel frameshift mutation, (NP002993.1:p.Leu139Phefs*29), not previously identified and two were splicing mutations (NM_003002.3:c.314+5G>A and c.315-2A>C), one of which was also a novel mutation, as presented in Table 2. The mutation search on SDHB gene revealed a frameshift mutation in one patient (NM_003000.2:c.591delC; NP_002991.2:p.Ser198Alafs*22) previously associated with the Paraganglioma phenotype.

Preoperative embolization of the tumour was performed in six patients. The choice for this technique was due to the size of the tumor, two patients had a Shamblin II and four had a Shamblin III tumor. No strokes or other major complications occurred after preoperative embolization with polyvinyl alcohol particles 1 day before surgery.

The incidence of neurological complications was 8.7 % for TIA and 6.5 % for stroke.

The incidence of cranial nerve injury (especially the Hypoglossal nerve) was 21.71%. One of the patients developed a Horner's syndrome after resection of the tumor. Patients with larger tumors had a higher lesion rate.

Vascular resection and reconstruction was performed in 5 tumors (10.4%).

Discussion

For the present study, the 46 identified CBT patients were contacted and after their informed consent, they underwent genetic testing for SDHD gene, since mutations in this gene are the leading cause of HNPGLs worldwide. The genetic analysis revealed seven different heterozygous variants in SDHD gene in eleven patients with CBT, of whom eight had a positive family history. These variants consisted of four missense mutations, one frameshift and two splicing mutations, of which, two were not reported previously.

The prevalence of the familial CBT cases varies between reported studies from as low as 5% to as high as 30%, in our study they represent 8/46 (17%) of the sample. As expected, the percentage of cases with positive family history was significantly higher in patients with SDHx mutations than in patients with no mutations identified (p < 0.001).

In contrast to previous studies that reported the occurrence of SDHx mutant tumors at a younger age, in the present study there was no difference in age at diagnosis between patients carrying SDHx mutations and non-mutation carriers. This discrepancy might be explained in part by the fact that the mutation analysis for SDHB gene was not performed for all the individuals.

The rate of complications varies greatly from one study to another. In our study the incidence was 8.7 % for TIA and 6.5 % for stroke. Stroke occurred in two patients with tumors classified as Shamblin III and one as Shamblin II. Only one patient had a mutation in SDHD. Despite the complexity of the tumors, 2/3 were Shamblin III with an average diameter of 48 mm, we presented a high morbidity.

Vascular resection and reconstruction was performed in 5 tumors (10.4%) and 2 of these patients had a stroke. These complications may be related to the clamping time which in these patients is less well tolerated. In all cases a great saphenous vein was harvested from the leg to be used as a conduit. Early detection of smaller tumors should reduce the morbidity of the procedure.

Carotid body PGLs are mostly benign, slow-growing tumors. Only 6 % of carotid body disclose uncertain malignant potential. Therefore, performing genetic tests in all CBT patients may be an interesting strategy so that only those with a higher risk of metastization can be selected for surgery. The watchful waiting approach is an option that is currently used in bilateral tumors because of the risk of baroreceptor reflex syndrome and may be an option in patients with unilateral paragangliomas and a low risk of malignancy.

The establishment of a genotype / phenotype relationship may allow to select the best candidates for this strategy and thereby reduce the morbidity in this disease⁶.

In conclusion, genetic screening allows identifying familial cases and improves clinical decision-making and adequate management of patients and their relatives. The present data contributes to a broader characterization of the clinical and molecular profile of European patients with CBT