The risk of facial nerve palsy after benign parotidectomy. A quality project

Abstract

Background

Facial nerve palsy is a potential complication of parotidectomy for benign salivary gland tumours, necessitating a comprehensive understanding of its incidence and associated risk factors for improved patient counselling and preoperative planning.

Aim/objectives

This single-centre retrospective study aimed to assess the rate of facial nerve palsy following benign parotidectomy at a University Teaching Hospital.

Material and methods

Over a 3-year period, 160 patients undergoing parotid surgery for benign tumours were included. Data, encompassing sex, age, operation technique, tumour pathology, facial nerve function, and follow-up duration, were collected from medical records. Exclusion criteria comprised patients with prior parotid gland surgery or preoperative facial nerve palsy.

Results

The study revealed a 3.75% incidence of facial nerve palsy with no total paralysis post-parotidectomy for benign disease. Pleomorphic adenoma (50.6%) and Warthin’s tumour (44.4%) were the predominant tumour types. No significant differences were noted between groups with and without postoperative facial palsy based on obtained covariates.

Conclusion and significance

Our findings endorse partial superficial parotidectomy and extracapsular dissection as low-risk treatments for benign parotid tumours. However, prospective studies are warranted to elucidate recovery rates and long-term consequences of facial nerve palsy, contributing to refined surgical approaches and patient care in parotid surgery.

Chinese abstract

背景

面神经麻痹是良性唾液腺腮腺肿瘤切除术的潜在并发症, 需要全面了解其发病率和相关危险因素, 以获得更好的患者咨询和术前计划。

目的

这项单中心回顾性研究旨在评估在大学教学医院进行良性腮腺切除术后面神经麻痹的发生率。

材料和方法

在 3 年期间, 纳入了160 名接受良性腮腺肿瘤手术的患者。从病历中收集数据, 包括性别、年龄、手术技术、肿瘤病理、面部神经功能和随访时长。 排除标准包括既往接受过腮腺手术或术前面神经麻痹的患者。

结果

研究显示, 面神经麻痹的发生率为 3.75%, 没有良性腮腺肿瘤切除术后的完全瘫痪。 多形性腺瘤(50.6%)和沃辛瘤(44.4%)是主要的肿瘤类型。 根据获得的协变量, 有术后面瘫组和无术后面瘫组之间没有发现显著差异。

结论和意义

我们的研究结果支持腮腺浅部部分切除术和囊外切除术是良性腮腺肿瘤的低风险治疗方法。 然而, 前瞻性研究是必要的, 可以阐明面神经麻痹的恢复率及其长期后果, 有助于精炼腮腺手术的手术方法和患者护理。

Keywords:

• Benign parotid tumour
• facial nerve palsy
• parotid surgery
• parotidectomy
• postoperative complication

Introduction

The parotid gland is the largest of the salivary glands, producing about 25% of all saliva. Parotid gland tumours represent less than 3% of all head and neck tumours and around 70% of all salivary gland tumours [1]. Most of the parotid gland tumours are benign, of which pleomorphic adenomas (53–69%) and Warthin’s tumours (25–32%) are the most frequent subtypes [1].

Parotid gland resection is the recommended treatment for benign parotid gland tumours because of the risk of malignancy or malignant transformation [2]. Parotidectomy carries a risk of facial nerve injury due to the proximity of the facial nerve to the tumour. This risk is the most severe complication and a source of anxiety for both the patient and their surgeon [3].

The incidence of facial nerve palsy after parotidectomy for benign diseases varies in the literature. Recent studies have reported that temporary and permanent facial nerve palsy occurs in 18–27% and 0–6% of the patients, respectively [4–7].

Information on the frequency of facial nerve palsy is valuable when explaining the surgical procedure and its risks to patients before surgery. Therefore, this retrospective quality project aims to determine the incidence of facial nerve palsy in total and divided by the House-Brackmann (HB) grading system after parotid gland surgery for benign tumours at Odense University Hospital (OUH) in the period between January 2018 and December 2020.

Materials and methods

We examined all hospital records of patients who had undergone parotid gland surgery between January 2018 and December 2020 at the Department of Otorhinolaryngology, Head and Neck Surgery and Audiology, OUH, Denmark. Patients were chosen based on diagnostic codes, including total parotidectomy (KELB50) and partial parotidectomy (KELB40).

Data was collected from a review of the patient’s medical records, including information about sex, age, and preoperative facial nerve function. Operative information about facial nerve dissection was obtained. Furthermore, final tumour pathology using the World Health Organization’s histological classification of tumours of the salivary glands [8], postoperative facial nerve function graded using the HB scale [9], and follow-up were registered. Tumour size was obtained from the pathology description or the preoperative ultrasound scan. Postoperative facial nerve palsy was defined as the presence of at least one branch of the facial nerve with slight weakness noticeable only on close inspection, corresponding to HB = II or worse.

Study data were stored and managed using REDCap electronic data capture tools hosted at Open Patient data Explorative Network, OUH, Region of Southern Denmark [10]. REDCap (Research Electronic Data Capture) is a secure, web-based software platform designed to support data capture for research studies, providing (1) an intuitive interface for validated data capture; (2) audit trails for tracking data manipulation and export procedures; (3) automated export procedures for seamless data downloads to common statistical packages; and (4) procedures for data integration and interoperability with external sources.

Inclusion criteria

The study included all patients treated for benign parotid gland tumours having a normally functioning facial nerve preoperatively and who had not previously undergone parotid gland surgery (Figure 1).

Figure 1. Participant inclusion flowchart. *Reactive lymphadenopathy (n = 4), sialolithiasis (n = 1), parotid abscess (n = 2), inflammation (n = 1), schwannoma (n = 2), lymphangioma (n = 1), sarcoidosis (n = 2) and Langerhans’ cell histiocytosis (n = 1). Abbreviations: OUH: Odense University Hospital.

Statistical analysis

Baseline data were presented as a median and interquartile range for non-normally distributed continuous variables, mean and standard deviation for normally distributed variables, and counts and percentages for categorical variables.

Differences between the no facial nerve palsy and the postoperative facial nerve palsy groups were investigated by Pearson’s χ² test for categorical variables and a two-sample t-test or Wilcoxon rank-sum (Mann–Whitney) test for continuous variables.

Data were analysed using Stata/IC 17 (StataCorp LLC, TX), and two-sided p-values < .05 were considered statistically significant.

Ethical consideration

Ethical approval was given by the Region of Southern Denmark (j.no. 22/32469).

Results

A total of 272 patients underwent parotid surgery at OUH between January 2018 and December 2020. Of these, 80 patients had a malignant disease, 14 patients had other parotid diseases such as reactive lymphadenopathy (n = 4), sialolithiasis (n = 1), parotid abscess (n = 2), inflammation (n = 1), schwannoma (n = 2), lymphangioma (n = 1), sarcoidosis (n = 2) and Langerhans’ cell histiocytosis (n = 1), 13 patients had recurrent parotid tumours, and five patients had preoperative facial nerve palsy, all excluded (Figure 1).

Therefore, 160 patients were enrolled in the study, with a median age of 59 years and 76 (47.5%) females (Table 1). The most frequent histological types were pleomorphic adenoma (50.6%) and Warthin’s tumour (44.4%) (Table 2). The median [Q1; Q3] maximum tumour diameter was 20 [15–25] mm; in 5 cases, the diameter was not reported in the patients’ journal. Partial superficial parotidectomy (PSP), defined by facial nerve dissection, was performed in 30.0% of the surgeries (Table 1).

Table 1. Characteristics of study participants.

	Total n = 160	No facial nerve palsy n = 154	Postoperative facial nerve palsy n = 6	p-Value
Age, years	59 [48–69]	59 [48–69]	58 [35–66]	.587
Sex				.338
Female	76 (47.5 %)	72 (46.8 %)	4 (66.7 %)
Male	84 (52.5 %)	82 (53.3 %)	2 (33.3 %)
Facial nerve dissection				.856
Yes	48 (30.0 %)	46 (29.9 %)	2 (33.3 %)
No	112 (70.0 %)	108 (70.1 %)	4 (66.7 %)
Tumour type				.258
Pleomorphic adenoma	81 (50.6 %)	76 (49.4 %)	5 (83.3 %)
Warthin’s tumour	71 (44.4 %)	70 (45.5 %)	1 (16.7 %)
Other	8 (5.0 %)	8 (5.2 %)	0 (0 %)
Tumours max diameter (n = 155), mm	20 [15–25]	20 [15–25]	18.5 [15–21]	.471

Differences between the no facial nerve palsy and the postoperative facial nerve palsy groups were tested using the two-sample Wilcoxon rank sum (Mann–Whitney) test for comparing medians of non-normally distributed variables. Pearson’s χ² test was used for comparing distributions of categorical variables. Values are median [Q1–Q3] for continuous variables and count (proportion) for categorical variables.

Table 2. Parotid tumour histopathology.

Tumour type	Patients (n (%))
Pleomorphic adenoma	81 (50.6 %)
Warthin’s tumour	71 (44.4 %)
Oncocytoma	3 (1.9 %)
Epithelial cyst	3 (1.9 %)
Canalicular adenoma	1 (0.6 %)
Castleman disease	1 (0.6 %)

In all, 154 patients had normal facial nerve function after surgery; postoperative facial nerve palsy incidence was 3.75% (n = 6). Three of these patients were diagnosed with facial nerve palsy immediately after surgery, and three were diagnosed with facial nerve palsy after 13, 17, and 21 days postoperative.

Three patients had a follow-up at 33, 39, and 201 days after being diagnosed with facial nerve palsy following surgery. Additionally, three patients did not have any follow-up. As described by the HB scale, three cases had a moderately/severe nerve dysfunction (HB = III) when determining facial nerve palsy, and three cases had a moderate nerve dysfunction (HB = II). At the last examination, all patients had a slight nerve dysfunction of one branch of the facial nerve (HB = II) [9]. The most frequent branch involved was the marginal mandibular nerve, which was affected in five cases. The frontal and zygomatic branches were affected in one case each.

There was no significant difference between the no facial nerve palsy and the postoperative facial nerve palsy groups according to sex, age, facial nerve dissection, tumour type, and size (Table 1).

Discussion

This study was done since facial nerve palsy is the most severe complication of parotidectomy, and therefore, it is essential to improve patient counselling and preoperative planning.

In this single-centre retrospective study, we found an incidence of facial nerve palsy of 3.75% (HB = II) after parotidectomy for benign diseases at OUH between January 2018 and December 2020, including no facial nerve paralysis. This rate is lower than the previously reported incidences of temporary facial nerve palsy of 18–27% but comparable with permanent facial nerve palsy of 0–6% after parotidectomy for benign diseases [4–7]. A recent systematic review and meta-analysis showed a corresponding pooled incidence of temporary (23.4%) and permanent (5.7%) facial nerve palsy using intraoperative facial nerve monitoring [11]. Unfortunately, this study could not distinguish between malignant and benign disease, which lowers the comparability since the risk of facial nerve palsy depends on tumour histology, size, and location [12]. In the present study, we found no difference in tumour type or size between the no-facial nerve palsy and the postoperative facial nerve palsy groups.

In our study, it was impossible to distinguish reversible and irreversible facial nerve palsy since none of the patients had a long-term follow-up. The treatment at the hospital ended when the patient still had a slight nerve dysfunction. The most frequently injured branch was the marginal mandibular nerve, which aligns with previous studies [13].

The participants in our study were broadly comparable with previous studies. Other studies have found a mean age of 51–58 years at diagnosis, and a gender ratio of 49–56% females and 44–51% males [5,7,13,14], compared to our mean age of 59 years and 47.5% females 52.5% males. This study’s recent data collection may explain the age difference compared to the other studies. A large Danish study from 2015 found that the median age at diagnosis of pleomorphic adenoma significantly increased between the 1980s and the 2000s [15]. Our study found no age difference between the no facial nerve palsy and the postoperative facial nerve palsy groups. Advanced patient age has previously been found to be a risk factor for temporary facial nerve palsy [4], but the results are inconsistent [7,14].

As anticipated, our study’s most prevalent tumour type was pleomorphic adenoma (50.6%), followed by Warthin’s tumour (44.4%). We had a considerable occurrence of Warthin’s tumours, in accordance with recent research, which has indicated an increase in the prevalence of Warthin’s tumours, mainly in northern and central Europe [16]. The reason could be due to population changes, changes in the histological classification, histological analyses, or the widespread use of improved and more suitable imaging techniques, potentially affecting epidemiological trends [8,16]. In addition, recent findings indicate that those with Warthin’s tumours have a higher body mass index than those with other benign parotid gland tumours, suggesting a correlation between Warthin’s tumours and obesity [17].

Earlier, superficial or total parotidectomy was the gold standard operation technique for parotid tumours, but today, a more conservative approach is used to reduce parotid surgery’s sequelae [18]. Consequently, PSP or extracapsular dissection (ECD) has almost replaced superficial or total parotidectomy for benign salivary gland tumours. In the present study, PSP was performed in 30.0% of the surgeries and 70.0% was an ECD. The difference in operation technique complicates the comparability between studies. Some studies include all operation techniques [6] or excluded ECD [7]. A meta-analysis showed that the recurrence rate was higher in patients treated with superficial parotidectomy compared to ECD and that the incidence of facial nerve palsy was lower in the ECD group. However, the study further describes that ECD should not be considered in large tumours (> 4 cm) with poor mobility or deep lobe involvement [18]. Another recent study concludes that PSD and ECD are safe surgical options in the treatment of benign tumours in the parotid gland. They found similar recurrence rates and post-surgical complications between PSP and ECD [19].

In this study, all the surgeons used intraoperative facial nerve monitoring as a standard of care. Some previous studies used no electromyographic facial nerve monitoring [4,5], others added facial nerve monitoring as a standard throughout the data collection period [7], and some studies compared the risk of facial nerve palsy with the use or not [6]. A recent systematic review and meta-analysis has shown a lower risk of immediate and permanent postoperative facial nerve palsy after parotidectomy with the use of intraoperative electromyographic facial nerve monitoring [11]. In contrast, a previous meta-analysis only found a lower risk for temporary facial nerve palsy with the use of electromyographic facial nerve monitoring compared to the unmonitored group [20].

Limitations

The main limitation of this study is its retrospective nature, and a prospective study should confirm the results. A further limitation is the non-consequently and relatively short period of follow-up. Another limitation is that a specific HB scoring was not consistently reported in the patient’s medical records, resulting in subsequent interpretations. Furthermore, tumour size was obtained in two different ways. Despite these limitations, our findings support previous studies supporting PSP and ECD as effective treatment methods for benign parotid tumours with a low risk of facial nerve palsy and no paralysis.

Conclusion

We found an incidence of facial nerve palsy of 3.75% (HB = II) after parotidectomy for benign disease. As expected, our study’s most prevalent tumour type was pleomorphic adenoma, followed by Warthin’s tumour. Further prospective studies are essential to describe recovery rates and long-term consequences of facial nerve palsy after benign parotidectomy.

Acknowledgements

We thank OPEN, Open Patient data Explorative Network, Odense University Hospital, Region of Southern Denmark, for facilitating the REDCap database and OPEN Analyse to store and process data.

Disclosure statement

No potential conflict of interest was reported by the author(s).