Abstract
Conventional total superficial parotidectomy (TP) has commonly been used, but partial superficial parotidectomy (PP) offers the possibility of better preserving glandular function and avoiding palsy of the facial nerves. In this study, the extent to which saliva secretion and facial nerve function were conserved in patients who received TP vs. PP was compared. Data were collected from patients who received a PP ( n = 163) or a TP ( n = 105) for benign primary tumours in the superficial lobe of the parotid glands between 1995 and 2009 at a single hospital. The incidence of transient facial paralysis was significantly lower in patients who received PP than in those who received TP. Secretory function was preserved for patients with a conserved Stensen’s duct, whereas patients in whom the duct had been ligated lost secretory function. Partial superficial parotidectomy reduces the incidence of postoperative facial nerve dysfunction and is conducive to preserving Stensen’s duct and saliva secretion.
Approximately 75% of parotid gland tumours are benign. About 80% of the time, they are located in the superficial lobe of the parotid gland, lateral to the facial nerve. A conventional total superficial parotidectomy (TP) is commonly performed to surgically remove benign parotid tumours that are restricted to the superficial lobe. In this procedure, the superficial lobe is completely resected and the integrity of the facial nerve anatomy is maintained. The goal of surgical treatment is the complete excision of the lesion with complete anatomical and functional preservation of the facial nerve. Most of the time, a considerable amount of non-tumourous parotid tissue is also resected and the intraparotid facial nerve is fully dissected to separate it from this tissue, which may disrupt Stensen’s duct, leading to a high incidence of secretory hypofunction and facial nerve paralysis or weakness. TP has recently come to be considered unnecessary for preventing recurrence in the majority of patients, depending on the histology, size, and location of the tumour. Therefore, the surgical practice for benign parotid tumours has evolved such that TP is now less common than partial superficial parotidectomy (PP). In a PP, the parotid tumour is resected with a surrounding 0.5–1 cm cuff of normal parotid tissue, or the tail of the parotid gland is resected when the tumour is located in the posterio-inferior portion. This method removes only the tumour-bearing area and obviates the need for more extensive facial nerve dissection to preserve healthy tissues. The functional parotidectomy (FP) became established on the basis of PP. An FP includes the management of the PP and the lifting incision, and conservation of the superficial musculo-aponeurotic system (SMAS) flap, the great auricular nerve, and Stensen’s duct. This preserves more of the function of the parotid gland and the related tissues, improves cosmetic results, and minimizes the incidence of facial nerve paralysis and other side effects. The FP is now the treatment of choice for most patients with benign tumours in the superficial lobe of the parotid gland.
Damage to the facial nerve is one of the most serious complications of parotid gland surgery. As many as 30–65% of patients experience transient weakness from facial nerve paralysis, and 3–6% experience permanent dysfunction of the facial nerve following TP. In addition, xerostomia can reduce a patient’s quality of life after parotidectomy. In the present study, the outcomes of the use of PP compared with TP performed during the same time-interval at a single institution were reviewed. Specifically, the conservation of salivary secretion and facial nerve function was compared between patients who received PP and TP.
Subjects and methods
A total of 268 patients with previously untreated benign parotid tumours within the superficial lobe, who underwent a primary parotidectomy between 1995 and 2009, were included in this study. Of these patients, 105 received a TP and 163 received a PP. For all patients, the diagnosis was confirmed by pathological examination of the excised tumours. The median age of the patients was 51 years (range 6–81 years); 118 were female and 150 were male (male to female ratio 1.27:1) ( Table 1 ).
| Total superficial parotidectomy (TP) ( n = 105) | Partial superficial parotidectomy (PP) ( n = 163) | Total ( n = 268) | ||||
|---|---|---|---|---|---|---|
| No. | % | No. | % | No. | % | |
| Sex | ||||||
| Male | 59 | 56.2 | 91 | 55.8 | 150 | 56.0 |
| Female | 46 | 43.8 | 72 | 44.2 | 118 | 44.0 |
| Surgery position | ||||||
| Left | 55 | 52.4 | 81 | 49.7 | 136 | 50.7 |
| Right | 50 | 47.6 | 82 | 50.3 | 132 | 49.3 |
| P athologic diagnosis | ||||||
| Pleomorphic adenomas | 71 | 67.6 | 86 | 52.8 | 157 | 58.6 |
| Warthin’s tumour | 26 | 24.8 | 52 | 31.9 | 78 | 29.1 |
| Basal cell adenoma | 4 | 3.8 | 16 | 9.8 | 20 | 7.5 |
| Other types | 4 | 3.8 | 9 | 5.5 | 13 | 4.9 |
| The management of the Stensen’s duct | ||||||
| Preserving the duct | 35 | 33.3 | 130 | 79.8 | 165 | 61.6 |
| Ligating the duct | 70 | 66.7 | 33 | 20.2 | 103 | 38.4 |
| Recurrence of the tumour | 0 | 0 | 1 | 0.6 | 1 | 0.37 |
The medical records were reviewed to obtain the following information: patient demographics, preoperative clinical assessment, histological findings, and postoperative transient facial paralysis. Intraoperative details were recorded by the surgeons at the time of surgery and included the location and size of each tumour and its relationship to the facial nerve, the extent of surgery, which branches of the facial nerve were dissected, and the management of the Stensen’s duct.
In this study, TP constituted removal of the entire lateral lobe of the parotid gland and complete facial nerve dissection. FP included the management of the PP and the lifting incision, and conservation of the SMAS flap, the great auricular nerve, and Stensen’s duct. PP was defined as any procedure in which less than a superficial lobectomy was performed; all 163 patients who received a PP received an FP, so these terms are used interchangeably. In a TP, the intraparotid facial nerve can be dissected using either an anterograde or retrograde technique. Retrograde dissection of the facial nerve has been more popular in China and this practice was followed in the current study. Using this method, the peripheral branches were identified first, and then dissected proximally to the bifurcation or main trunk. After the skin flap was raised, the resection of the parotid gland began from the anterior border, where Stensen’s duct emanates from the gland onto the masseter muscle, and the facial nerve was separated away from the tumour. Whenever the facial nerve was found to lie across the duct, efforts were made to preserve the duct; if, however, the intraparotid facial nerve was below Stensen’s duct, the duct was transected and ligated. When the bifurcation and main trunk of the facial nerve became exposed, the superior parotid gland was resected at its posterior border with the tumour ( Fig. 1 ). If the tumour was less than 2 cm in diameter or if the tumour was located in the tail of the parotid gland, a PP was carried out.
In the PP, only a portion of the facial nerve, i.e. the peripheral branches beyond the tumour site, was meticulously dissected, and the parotid tumour was resected with a surrounding 0.5–1 cm cuff of normal parotid tissue ( Fig. 2 ), except when the tumour abutted the plane of the facial nerve. The Stensen’s duct was usually preserved in the PP unless it hindered the procedure of tumour resection.
Transient facial palsy at 1 week after surgery was recorded in the case notes. All patients were followed for at least 6 months after surgery; follow-up ranged from 6 to 159 months, and the median follow-up time was 18 months. The patients were asked to give a history and were given a clinical examination. Every patient was examined by palpation for recurrence of the tumour. When a recurrence was suspected, ultrasound, computed tomography, or magnetic resonance imaging was performed to confirm or exclude this diagnosis. They were also given a questionnaire on postoperative complications, and formal tests for facial movement and secretion function. The objective clinical evaluation of postoperative complications in all patients was performed by a single clinician (Dr. Zhang) who was blinded to the details of the surgery.
All patients eligible for the study were evaluated by an index score created specifically to evaluate facial motor function and salivation. The severity of the patient’s loss of salivation was rated using a scale from 1 (fully satisfied with current status without any discomfort) to 4 (extremely uncomfortable). Facial nerve dysfunction was evaluated using the House–Brackmann grading system, which includes six grades ranging from grade I (normal function) to grade VI (complete loss of facial motor function), and the branches of the facial nerve that appeared to be paralyzed were recorded.
Saliva flow rates in the operated and contralateral parotid glands were measured by sampling the secreted saliva using a Lashley cup. First, the unstimulated saliva flow was collected for 5 min; then, after resting for 10 min, salivation was stimulated by application of 2% citric acid to the dorsal surface of the anterior tongue with a saturated cotton swab, five times at 15-s intervals. The stimulated saliva flow was also collected for 5 min and compared between the operated and contralateral sides.
The study had institutional review board approval and written informed consent was obtained from each patient. All data collected were entered onto code sheets which were then captured in a database. Data were presented either as means with standard deviations or as percentages when appropriate. All statistical analyses were performed using SPSS 13.0 for Windows. All categorical variables were analyzed in a univariate analysis using χ 2 tests; t -tests were employed when continuous variables were encountered. Saliva flow rates were compared between TP and PP by independent samples t -test, and saliva flow rates of the operated and contralateral sides for both TP and PP were compared by paired samples t -test. Incidences of recurrence in the two groups were analyzed by Fisher’s exact test. P -values of less than 0.05 were interpreted as statistically significant.
Results
Information was obtained from the patient case notes and a contemporaneous database; patients presenting symptoms of complications following total or partial superficial parotidectomy were investigated by Dr. Zhang. During the study period, a total of 268 patients with a benign tumour in the superficial lobe of the parotid gland were treated with parotid surgery, including 163 (60.8%) who received a PP and 105 (39.2%) who received a conventional TP. Among them, 136 patients (50.7%) had a tumour located on the left side of the face, and 132 (49.3%) had a tumour on the right side. Of all 268 patients in this series, 75 were followed up for 6–12 months, 161 for 13–24 months, 16 for 25–36 months, and 11 for more than 37 months after the operation. Histologically, the tumours were pleomorphic adenomas in 157 patients (58.6%), Warthin’s tumour in 78 patients (29.1%), and basal cell adenomas in 20 patients (7.5%); the rest were benign lesions of other types ( n = 13, 4.9%). One patient with a Warthin’s tumour had recurrence after PP. There was no statistically significant difference in the recurrence rate between the TP and PP groups, which were 0% and 0.6%, respectively ( P = 0.608).
Facial nerve function
The most common surgical complication was transient facial palsy. The overall transient palsy rate was 24%. According to the surgical records, an average of 3.55 ± 1.152 branches of the intraparotid facial nerve were dissected for each patient in the TP group, and an average of 1.58 ± 0.974 branches for each patient in the PP group ( P = 0.006). Facial nerve function was assessed at 1 week postoperatively, and palsy rates were compared among patients with TP vs. PP. The incidence of temporary facial nerve dysfunction was higher in patients who had TP (34/105, 32.4%), compared with PP (29/163, 17.8%) ( P = 0.005). There were proportionally more patients with multi-branch nerve palsy in the TP group (6/105, 5.7%) than in the PP group (6/163, 3.7%; Table 2 ), however there was no statistically significant difference. A permanent palsy was more common in the TP group (4/105, 3.8%) compared with the PP group (3/163, 1.8%), however the difference failed to reach statistical significance ( P = 0.324) ( Table 2 ). All seven patients with permanent dysfunction of the nerve branches had mild dysfunction, defined as House–Brackmann grade II. In most of them (5/7, 71.4%), the nerve affected was the marginal mandibular branch. The marginal mandibular facial nerve was involved during the surgery of 83 of 105 (79.0%) patients in the TP group and 94 of 163 (57.7%) patients in the PP group. More details regarding postoperative facial nerve function are provided in Tables 2 and 3 .
