Evaluation of Lateral Temporal Bone Resection in Locally Advanced Tumours of the Parotid Gland

Abstract

This study evaluated the effects of lateral temporal bone resection (LTBR) on local tumour control in patients with locally advanced parotid gland tumours. The medical records of seven patients treated with radical parotidectomy combined with LTBR for locally advanced primary parotid tumour at the Otolaryngology Department of Dokuz Eylul University between January 1995 and December 2016 were retrospectively evaluated. Demographic variables, tumour characteristics, treatment properties, postoperative complications, follow-up durations and local, regional and distant recurrences were analysed. Before referral to our clinic, four patients had de novo primary parotid tumours, and three patients had a recurrence of primary parotid tumours. The histopathologic diagnoses were squamous cell carcinoma in two patients, and adenosquamous carcinoma, malignant myoepithelial carcinoma, adenocarcinoma, adenoid cystic carcinoma and spindle cell sarcoma in the other patients. During the follow-up period, one patient died due to postoperative pulmonary embolism in the first month, and four patients died due to distant metastasis without local or regional recurrences. LTBR combined with radical parotidectomy in locally advanced primary malignant parotid gland tumours is a feasible surgical technique for local tumour control. However, the most common cause of death in these cases is distant metastases, despite appropriate resection.

Introduction

Salivary gland tumours constitute about 3–6% of all head and neck tumours [1, 2]. The incidence was reported to be 1.2 per 100,000 individuals per year, in the USA and the 5- and 10-years survival rates were 75–80 and 65–70%, respectively [1, 2]. Curative treatment is parotidectomies, in resectable cases. If associated with high histological grade, lymph node metastasis, positive surgical margin or advanced local tumour, adjuvant radiotherapy is applied in addition to surgery [1–3]. Despite all these treatment principles, local-regional recurrence can occur in 21–33% of cases [3, 4].

In locally advanced tumours with extraglandular spread to areas, such as the mandibular, external auditory canal (EAC), temporomandibular joint (TMJ) and infratemporal fossa (ITF), standard and classic parotidectomy techniques are inadequate for complete resection and negative surgical margin. In these cases, extended surgical approaches should be performed.

Only a few studies have discussed the benefits of temporal bone resections combined with radical parotidectomy on tumour control in patients with locally advanced parotid gland tumours [1–3]. The possible reason for this limited research is that most of the cases that can be performed with extended surgery are not accepted as resectable. In this study, we investigated the results of patients with locally advanced parotid gland tumours, treated with lateral temporal bone resection (LTBR) combined with radical parotidectomy. We assessed the local and regional recurrences, distant metastasis and survival characteristics of the patients.

Methods

After the approval of the ethics committee, the medical records of seven patients treated with radical parotidectomy combined with LTBR for locally advanced primary parotid tumour at the Otolaryngology Department of Dokuz Eylul University between January 1995 and December 2016, were retrospectively evaluated. Demographic variables, like age, gender, examination findings at the time of admission, histopathologic diagnosis, imaging findings, clinical stages, surgical characteristics and resection limits, neck treatment characteristics, reconstruction types, postoperative complications, treatment outcomes and follow-up characteristics, were determined. Patients with non-glandular masses, such as skin-derived tumours, who underwent LTBR and radical parotidectomy were excluded. Patients with primary parotid gland tumours based on clinical and histopathological findings were included. The House-Brackmann grading system was implemented to establish the extent of the facial nerve functions before surgery. The 2010 American Joint Committee On Cancer (AJCC) Staging System For Parotid Tumours TNM system recommendation was used to describe the cancer stage, from stages I to IV. Statistical analysis was not performed, due to the limited number of patients.

Results

The mean age of the seven patients (four male and three female) was 53.7 years (range 32–75 years), and the mean follow-up period was 25.8 months (range 1–48 months). Three patients had grade 6 facial paralysis and one patient with grade 5 facial paralysis, at the time of admission. In the remaining three patients, facial nerve functions were normal. Combined skin and EAC invasions were observed in two patients, skin invasion alone in two patients, and EAC invasion alone in one patient. One patient had restricted mouth opening, due to invasion of the TMJ.

Three of the patients had a recurrent tumour before the consultation at our clinic. Two of whom had parotidectomy and adjuvant chemoradiotherapy, and the third had chemoradiotherapy without surgical history. The other patients had de novo tumours. Histopathological diagnoses were squamous cell carcinoma in two patients, and malignant myoepithelial carcinoma, adenocarcinoma, adenosquamous carcinoma, adenoid cystic carcinoma and spindle cell sarcoma in the remaining five patients.

In the radiological examination, all the patients had extraglandular spread. ITF involvement had been detected in five patients among which, two had TMJ spread. Three patients had unilaterally pathological lymph nodes without extranodal involvement. According to the clinical and radiological examinations, the TNM stages were determined as cT4N0M0 and cT4N2bM0 in four and three patients, respectively. The clinical and radiological features, previous treatment characteristics, histopathological diagnosis and clinical TNM stages, are summarised in Table 1.

Table 1.

The treatment and follow-up characteristics of the patients

	Age	De novo/recurrent	Histopathological diagnosis	Initial treatment	Clinical and radiological features						Clinical stage	Surgery	Neck dissection	Reconstruction	Complication	Adjuvant treatment	Follow up (month)	Recurrent
					FNF	ES	SI	EACI	ITFI	TMJI
Case 1	64	Recurrent	Malignant myoepithelial carcinoma	Surgery CRT	Grade 6	Yes	Yes	Yes	Yes	None	rT4N0M0	LTBR	None	POSRF	Donor area	CT	36 ex	Lung and bone
Case 2	32	Recurrent	Spindle cell sarcoma	Surgery CRT	Normal	Yes	None	None	Yes	None	rT4N0M0	LTBR + SM	None	PFF	None	CRT	48 ex	Lung
Case 3	75	De novo	Squamous cell carcinoma	None	Normal	Yes	None	Yes	Yes	None	T4N0M0	LTBR	Elective	POSRF	Pulmonary embolism	–	1 ex	None
Case 4	58	De novo	Adenoid cystic carcinoma	None	Grade 6	Yes	None	None	Yes	Yes	T4N2bM0	LTBR + HM	Radical	PMMFF with CCTRF	Donor area	CRT	42	None
Case 5	57	Recurrent	Adenosquamous carcinoma	CRT	Normal	Yes	Yes	None	None	None	rT4N0M0	LTBR	Elective	PMMFF with CCTRF	None	CRT	32	None
Case 6	35	De novo	Adenocarcinoma	None	Grade 5	Yes	Yes	None	None	None	T4N2bM0	LTBR	Radical	PMMCF	None	CRT	12 ex	Lung
Case 7	68	De novo	Squamous cell carcinoma	None	Grade 6	Yes	Yes	Yes	Yes	Yes	T4N2bM0	LTBR	Radical	PMMCF	None	CRT	10 ex	Lung

CCTRF Cutaneous cervicothoracic rotation flap, CRT chemoradiotherapy, CT chemotherapy, EACI external auditory canal invasion, ES extraglandular spread, FNF facial nerve functions, HM hemimandibulectomy, ITFI infratemporal fossa involvement, LTBR lateral temporal bone resection, PFF parascapular free flap, PMMCF pectoralis major myocutaneous flap, PMMFF pectoralis major myofacial flap, POSRF parieto-occipital scalp rotational flap, SM segmental mandibulectomy, SI skin invasion, TMJI temporomandibular joint involvement

Radical parotidectomy and LTBR were undertaken in all patients, for treatment of the primary tumour area. Segmental mandibulectomy was additionally performed on the patient with restricted mouth openness dependent on TMJ involvement, who had a spindle cell sarcoma. In another patient, whose intraoperative frozen section analysis showed mandibular nerve involvement with adenoid cystic carcinoma, an additional hemimandibulectomy was also performed. Surgery was extended with ITF resection, in patients with ITF spread. Radical neck dissection was carried out in the cT4N2bM0 patients, and elective neck dissection was also done, in two of the cT4N0M0 patients. Neck dissection was not needed in the patients diagnosed with malignant myoepithelial carcinoma and spindle cell sarcoma. In the surgical reconstruction of the resected area, the pectoralis major myocutaneous flap was used in two patients, the pectoralis major myofacial flap with cutaneous cervicothoracic rotation flap in two patients, the parieto-occipital scalp rotational flap in two patients and the parascapular free flap in one patient.

One patient was exitus in his first month postoperatively because of multiple organ failure related to pulmonary embolism. Wound site infection and donor area epithelialisation defect developed in one patient with the parieto-occipital scalp rotational flap. Wound dehiscence, due to weighty breast volume, occurred in the donor area of the pectoralis major myocutaneous flap in one patient. Aside from these observations, no additional surgical complications occurred.

After the histopathological examinations of surgical specimens, a positive surgical margin was detected microscopically in two patients with adenoid cystic carcinoma and epidermoid carcinoma respectively. Metastatic lymph nodes were also discovered in pathologic examinations of three patients, who were treated with radical neck dissection due to clinically positive metastatic lymph nodes. Metastatic lymph nodes were not detected in the neck dissection specimens of two patients, who underwent elective neck dissection.

Five of the patients received chemoradiotherapy as an adjuvant treatment. The patient who had previous surgery and adjuvant chemoradiotherapy history due to malignant myoepithelial carcinoma and, additionally, had a donor area epithelialisation defect, received chemotherapy only, without adjuvant radiotherapy. The other patient could not receive additional treatment, because he had multiple organ failure and eventually he was exitus.

Three of the patients were exitus at 10, 12 and 48 months after the initial treatment, respectively, due to lung metastasis. Another was exitus at 36 months, due to lung and bone metastasis. The one who could not complete postoperative additional therapy was exitus in his first month postoperatively because of multiple organ failure related to pulmonary embolism. Two patients, at the time of this report, remain disease-free at 32 and 42 months of follow-up. None of the patients developed local or regional recurrence. The treatment and follow-up characteristics of the patients are given in Table 1.

Discussion

Malign tumours of the parotid gland are rare entities, with wide histopathological variability [1–5]. Surgical resection is the priority treatment of these tumours because of their chemo-/radioresistance [1–6]. Total or superficial parotidectomy without adjuvant therapy is sufficient for low-grade and early-stage intraglandular tumours. In the presence of risk factors, such as high histologic grade, advanced-stage, nodal metastasis or surgical border positivity, adjuvant radiotherapy should be applied in addition to surgery [1–5]. Postoperative adjuvant radiotherapy has been found to achieve a 20–50% reduction in local recurrence [6–8]. Surgery with adjuvant radiotherapy has been shown to attain a success rate of over 70% in local tumour control [6–12].

Treatment options are conversely different and complicated, in locally advanced tumours with extraglandular spread. Routine parotidectomy techniques are inadequate to achieve negative surgical margins in the presence of skin, mandibula, EAC, TMJ, ITF or temporal bone extension. In such cases, the resection borders must be extended to include all structures and regions with involvement, and LTBR should be applied in combination with parotidectomy.

LTBR is frequently the preferred approach for locally advanced skin tumours of the temporal bone and/or auricle [13, 14]. This procedure was first described by Conley and Novack in 1960 [15–17]. The LTBR boundaries are an enlarged facial recess with mastoidectomy posteriorly, incudostapedial joint medially, zygomatic cells superiorly and TMJ anteriorly. Although this procedure is commonly used in the treatment of locally advanced temporal bone tumours, there is limited literature data on its use for locally advanced parotid tumours. Moreover, skin tumours that show temporal bone with parotid gland invasion, and parotid gland tumours with skin extension, are clinically intertwined or can mimic each other [13]. Skin-derived epidermoid carcinoma may cause parotid gland involvement, with local invasion or intraparotid lymph node metastasis. Conversely, a tumour that originates from the parotid gland may mimic a primary skin-derived tumour, particularly in advanced stages, by creating peripheral soft tissue spread, involving, for instance, the skin or EAC [14]. Therefore, especially high-grade mucoepidermoid carcinoma and skin-induced metastatic epidermoid carcinoma should be excluded, to diagnose primary parotid epidermoid carcinoma. In common practice, a parotid mass with the presence of a primary epidermoid carcinoma locus in the skin region, which may constitute parotid metastases, indicates metastatic involvement of the parotid [18–21]. Despite these conflicting conditions, parotid tumours originate from glandular cells and constitute a histopathologically distinct group. Similarities in the surgical treatment principles of these different tumour types, diagnoses, treatments and follow-up protocols should be evaluated separately.

In our literature review, only two studies described the use of temporal bone resection or extended surgery on locally advanced-stage parotid gland tumours. In the first of these studies, Mehra et al. [3] presented the data of twelve patients, who underwent LTBR. The study comprised three mucoepidermoid carcinomas, three myoepithelial carcinomas, two adenoid cystic carcinomas, two squamous/adenosquamous carcinomas, one adenoid carcinoma and one ductal carcinoma. Facial nerve involvement was clinically detected in eight patients, EAC/temporal bone involvement in six patients, skin involvement in four patients, ITF involvement in two patients and jugular foramen involvement in one patient. Eleven patients underwent neck dissection and postoperative adjuvant radiotherapy. Postoperative respiratory distress developed in one patient, haematoma in one patient, and myocardial infarction and deep vein thrombosis in one patient. A recurrent tumour was detected in eleven of these cases, but local recurrence rates were low, with only three. Local recurrence was found in one patient, local and distant recurrence in one patient, and regional and distant recurrence in another. Although the overall recurrence rate was 92%, the authors argued that extended surgery (LTBR) has benefits regarding local and regional tumour control.

In the second study, Gidley et al. [1] reported 49 patients, who underwent parotidectomy plus mastoidectomy in 33 cases and temporal bone resection in 16 cases. All the patients had mastoid bone, EAC or mandibular involvement. Although the study population included 14 different types of pathology, squamous cell carcinoma was the most common in 14 patients. The other frequent pathologic diagnoses were acinic cell carcinoma in eight patients, ductal carcinoma in six patients, adenoid cystic carcinoma in five patients and adenocarcinoma in four patients, respectively. Forty-five of the cases underwent neck dissection, and 38 received adjuvant radiotherapy. Postoperative complications included mental status impairment in two patients, hyponatremia in one patient, pulmonary oedema in one patient, respiratory distress due to mucus plug in one patient, wound infection in one patient and vertigo in another patient. In definitive histopathological examinations, extraglandular extension was detected in 26.5% of the patients, in the form of bone, muscle or cartilage tissue involvement. At the end of the 24.9 months (range 2.9–66.2 months) mean follow-up period, local recurrence was detected in seven patients, distant recurrence in five patients, regional recurrence in two patients, local and distant recurrence in two patients, and regional and distant recurrence in one patient. The authors stated that local, regional and distant recurrence rates are high, despite extended surgery and adjuvant radiotherapy in these types of cases and so chemotherapy is likely to increase its role in the future, although, currently, chemotherapy has low efficacy in parotid tumours.

In our study, there were no local, regional or distant metastases in the two cases whose follow-up periods are ongoing. Three patients were exitus with lung metastasis, and one patient was also exitus with lung and bone metastases. Local or regional recurrences were not detected during follow-up in these patients with distant metastases. According to the results of our cases, we found that large resections can prevent the development of local-regional recurrence. Conversely, similar to the findings of other studies, distant metastases are currently a significant cause of mortality in these cases [1–3]. When patients are examined for complication development, it has been observed in our series that systemic complications may occur after such elongated surgeries, as in other studies [1–3]. In these patients, mortal complications, such as pulmonary oedema, thrombosis, embolism or myocardial infarction can be observed in the early postoperative period. Also, in our series, one of the patients was lost due to pulmonary embolism and postoperative multiorgan failure in the first postoperative month. Soft tissue infection in one patient and donor site problems in the form of suture dehiscence in one patient developed as other minor complications.

It is known that palliative or radical radiotherapy/chemoradiotherapy options are used in the treatment of nonresectable tumours of the parotid gland [22]. It has been reported that radical radiotherapy at 60–70 and > 70 Gy dose can provide 25% and 40% local control success, respectively [6–23]. More acceptable oncological results can be achieved with concomitant chemoradiotherapy rather than radiotherapy, only because of the additive effects of chemotherapy on radiotherapy sensitivity [22]. The most controversial subject in these studies is patient selection. In cases in which surgical border safety can be provided with LTBR are considered to be non-resectable, negative results may be encountered, especially regarding local and regional disease control. Prospective randomised trials, comparing the outcomes of radical chemoradiotherapy and widened surgeries, such as LTBR, are performed on locally advanced-stage tumours.

Conclusion

LTBR combined with parotidectomy can reduce local recurrence development in locally advanced malignant parotid tumours. Currently, despite adequate results with expanded surgeries on local recurrences, the most frequent cause of failure and death in these cases is distant metastases.

Funding

There is not any funding organisations to be disclosed for the present study.

Compliance with Ethical Standards

Conflict of interest

The authors have no funding, financial relationships or conflict of interest to disclose.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent

Informed consent was obtained from all individual participants included in the study.