Parotid gland masses at two South African referral hospitals: a retrospective review of fine needle aspiration cytology, histopathology and surgical management

¹Department of Otorhinolaryngology, School of Clinical Medicine, Faculty of Health Sciences, University of the Free State, Bloemfontein, South Africa, ²Department of Otorhinolaryngology, Universitas Academic Hospital, Bloemfontein, South Africa, ³Department of Otorhinolaryngology, Pelonomi Regional Hospital, Bloemfontein, South Africa, ⁴Department of Biostatistics, School of Biomedical Sciences, Faculty of Health Sciences, University of the Free State, Bloemfontein, South Africa

^&Corresponding author
Riaz Yakoob Seedat, Department of Otorhinolaryngology, School of Clinical Medicine, Faculty of Health Sciences, University of the Free State, Bloemfontein, South Africa

Introduction    

The spectrum of pathology of parotid gland masses differs in different geographic regions [1-3]. The pathology of parotid gland tumors in South Africa has only been reported in a few studies in South Africa [3,4]. Van Lierop et al. found that the most frequently occurring benign parotid gland tumor was pleomorphic adenoma, with the most frequently occurring malignant tumor being metastatic cutaneous squamous cell carcinoma [3]. Mdletshe et al. also reported that pleomorphic adenoma was the most frequently occurring benign parotid gland tumor, with mucoepidermoid carcinoma being the most frequently occurring malignant parotid gland tumor in their study [4]. Fine needle aspiration cytopathology (FNAC) can be used to determine whether parotid gland disease is benign, malignant, or non-neoplastic [3-5]. However, salivary gland tumors have broad and overlapping pathological characteristics, and there can be challenges with sampling due to hemorrhage and areas of necrosis and fibrosis in tissue being sampled, resulting in diagnostic challenges for cytopathology [6-8]. The Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) standardized the reporting of FNAC in order to improve the consistency between cytopathology and final histopathological outcome, as well as communication between pathologists and clinicians [9]. A large systematic review and meta-analysis to determine the accuracy of FNAC according to the Milan system found that FNAC had a sensitivity of 84.37% and specificity of 98.27% in differentiating a neoplasm from a non-neoplastic lesion and a sensitivity of 72.10% and specificity of 98.27% in differentiating a benign from a malignant neoplasm [10].

Wu et al. showed FNAC according to the MSRSGC had a sensitivity, specificity, positive predictive value, and negative predictive value of 89%, 99%, 98%, and 96%, respectively [11], while Viswanathan showed a FNAC sensitivity, specificity, negative predictive value, and positive predictive value of 79%, 98%, 94%, and 92%, respectively [12]. There is only one published study from South Africa on the MSRSGC that found that the concordance between salivary gland tumor FNAC reported according to the MSRSGC and final histology was 96.95% [6]. The primary standard of treatment of parotid gland tumours is parotidectomy. This procedure may be associated with several complications [13,14]. Facial nerve weakness is one of the major complications of parotidectomy, which can be temporary or permanent. Other complications of parotidectomy include hypoesthesia of the greater auricular nerve, Frey's syndrome, wound infection, haemorrhage/haematoma, sialocele, skin flap necrosis, salivary fistula, long-term scarring and deformity. This study aimed to examine the demographics and histology of parotid gland masses, compare FNAC reported according to the MSRSGC and definitive histopathology, and report on the parotid gland surgery and associated complications in patients managed at the Department of Otorhinolaryngology at Universitas Academic Hospital and Pelonomi Regional Hospital, the referral hospitals affiliated to the University of the Free State.

Methods     

Study design and setting: this was a retrospective study of patients with parotid gland masses at the Department of Otorhinolaryngology at Universitas Academic Hospital and Pelonomi Regional Hospital from 1^st July 2019 to 30^th June 2023.

Study population: the records of all patients who had parotid gland masses with definitive histopathology and who underwent parotid gland surgery were reviewed.

Data collection: the data were collected via a retrospective record review and recorded on a RedCap database [15,16]. Data collected included demographics such as age and gender, results of fine needle aspiration cytopathology reported according to the MSRSGC [9], surgical procedure, histopathology, and complications of surgery.

Definitions: surgical procedures were classified into parotid gland biopsy, partial superficial parotidectomy, superficial parotidectomy, and total parotidectomy. Parotid gland biopsy was defined as an incisional biopsy of the mass of the parotid gland; partial superficial parotidectomy was defined as surgical removal of the parotid gland mass with a cuff of normal parotid gland tissue; superficial parotidectomy was defined as removal of the mass with the superficial lobe of the parotid gland; and total parotidectomy was defined as removal of the mass with the deep and superficial lobes of the parotid gland. Cytology results were classified according to the MSRSGC [9].

Statistical analysis: statistical analysis was performed using R version 4.3.2 [17]. The numerical variable (age) was summarized as medians and interquartile ranges (IQR). Categorical variables were analyzed as frequencies and percentages. Fisher's exact test was used to compare categorical data, and the Kruskal-Wallis test compared medians between groups. P-values of less than 0.05 were considered statistically significant. Fine needle aspiration cytopathology (FNAC) findings were compared to definitive histopathology post-surgery to determine the sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy. Fine needle aspiration cytopathology results reported as category II (non-neoplastic) and IVa (benign) according to MSRSGC were classified as benign, and category V (suspicious for malignancy) and VI (malignancy) were classified as malignant [18,19].

Ethical considerations: ethical approval to perform this study was obtained from the Health Sciences Research Ethics Committee of the University of the Free State (UFS-HSD2023/2183/3001). The Free State Department of Health provided approval to perform this study. Data were de-identified, and the RedCap database was password-protected, with two-factor authentication.

Results     

Cohort characteristics: there were 47 patients with parotid gland masses that underwent parotid gland surgery between 1^st July 2019 and 30^th June 2023. The median age of patients who had parotid gland masses was 44 years (range 6-84 years, IQR 31-61 years). There were 19 (40.4%) males and 28 (59.6%) females. Neoplasms were present in 44 (93.6%) patients, while three (6.4%) had non-neoplastic pathology. Benign tumors were present in 24 patients, while 20 had malignant tumors. The median age of patients with benign parotid gland tumors was 44 years (IQR 32-60 years), while the median age of patients with malignant parotid gland tumors was 45 years (IQR 31-62 years). Nine (37.5%) of the patients with benign parotid gland tumors and eleven (45%) of patients with malignant parotid gland tumors were male. One of the three patients with non-neoplastic disease (33.3%) was male. There was no significant difference in age (p=0.8) or sex (p=0.9) between the three groups. Fine needle aspiration cytopathology results were available for 18 patients (Table 1). Benign tumors comprised 51.1% of parotid gland masses and malignant tumors 42.6% (Table 2). The most frequently occurring parotid gland tumor was pleomorphic adenoma, followed by mucoepidermoid carcinoma.Five patients with parotid gland masses were under 18 years. Three (60%) of these patients had mucoepidermoid carcinoma. Other masses were a pleomorphic adenoma and a mucus retention cyst, each in one patient.

Comparison between fine needle aspiration cytopathology and histopathology: fine needle aspiration cytopathology results were compared to histopathology results (Table 3). The sensitivity of FNAC for detecting malignancy was 100% (95% confidence interval CI:59.0-100%), specificity 50% (95% CI: 6.8-93.2%), positive predictive value 78% (95% CI: 40.0-97.0%), and negative predictive value 100% (95% CI: 15.8-100%). The diagnostic accuracy of FNAC was 81.8% (95% CI: 48.2% to 97.7%).

Surgery: a parotidectomy was performed in 44 patients, while parotid gland biopsies were performed in three (6.4%) patients. Partial superficial parotidectomy was performed in 38.3% (18/47) of patients, followed by superficial parotidectomy in 36.2% (17/47) of patients. Total parotidectomy was performed in 19.1% (9/47) of patients. Three patients in the cohort underwent open biopsy of a parotid-region lesion. One had an ulcerated, advanced squamous cell carcinoma with prior FNAC reported as suspicious for malignancy (Milan V), for whom open biopsy was performed to confirm the diagnosis before commencing definitive radiotherapy. The second patient had a superficial Kaposi sarcoma; two prior FNACs were non-diagnostic (Milan I), and an open biopsy was therefore undertaken, after which the patient received medical management. The third patient had a secretory carcinoma of the parotid diagnosed on open biopsy without preceding FNAC; the available records did not document the rationale for choosing open biopsy over parotidectomy, and no subsequent parotidectomy was performed. Twelve patients had a perioperative complication. The most common complication was wound infection, occurring in three (6.4%) patients. Temporary facial nerve palsy, hypoesthesia of the greater auricular nerve, and keloid scarring affected two patients each. Complications of parotid gland surgery were more common in patients with benign parotid gland tumours (10/24, 41.7%) than in patients with malignant parotid gland tumours (2/20, 10%).

Discussion     

In this study, we sought to report on the pathology and management of parotid gland masses in our context at Universitas Academic Hospital and Pelonomi Regional Hospital by describing the demographics, cytology, histology, parotid gland surgery, and perioperative complications. The age and sex distribution and pathology of parotid masses were similar to those reported in other South African studies [3,4,6]. Fine needle aspiration cytopathology had high accuracy in our study, in line with previous reports [6,18-20]. The median age of patients with parotid gland masses was 44 years. This is in line with the mean ages reported in two other South African studies on parotid gland disease, which show mean ages of 47.9 years and 43.3 years [3,4]. The female-to-male ratio was 1.47:1, consistent with a South African study in Cape Town [3]. In the African context, there is a female predilection for salivary gland tumors [1,2,6], while in two other studies in an Asian context, there was a male predilection [18,19].

Benign tumors were 51.1% and malignant tumors 40.4% of the parotid gland masses. The most frequently occurring tumor overall was pleomorphic adenoma, consistent with the literature [3,4,6,21]. The second most frequently occurring benign parotid gland tumor was basal cell adenoma. In other South African studies, the most frequently occurring benign parotid gland tumor was also pleomorphic adenoma, but the second most common was Warthin's tumor and lymphoepithelial cysts [3,4,6]. Warthin's tumor was found in only one case in our study and is rare in other African studies [1-3]. Mucoepidermoid carcinoma was the most frequently occurring malignant tumor in our study, in keeping with other South African studies [3,4]. Mucoepidermoid carcinoma was also the most common tumor in patients under the age of 18 years. This is consistent with a study by Murugan et al., who also found that mucoepidermoid carcinoma was the most frequently occurring malignancy of the parotid gland in pediatric patients [22].

There were two patients with rare tumours of the parotid gland: extraskeletal myxoid chondrosarcoma and Kaposi sarcoma. Extraskeletal myxoid chondrosarcoma rarely occurs in the head and neck region and has been reported in only a few patients in the literature [23]. It has a male predilection [23], but in our study, it occurred in a 37-year-old female patient. The patient with the Kaposi sarcoma of the parotid gland was a 44-year-old female who was HIV positive. Kaposi sarcoma is associated with HIV/AIDS and rarely affects the salivary glands, but should however, form part of the differential diagnosis in immunocompromised patients with parotid gland lesions [24]. In keeping with the literature, the most common Milan category was IVa (benign) followed by category II (non-diagnostic) [4,18]. Fine needle aspiration cytopathology according MSRSGC is good at risk stratification, improves communication between surgeons and pathologists and assists the surgeon in surgical planning to determine how extensive the surgery should be pre-operatively [4,18,19]. When the FNAC was compared to histopathology, the sensitivity was 100% and the specificity 50%. In other studies, sensitivity ranged from 76.5 to 92.7 and specificity from 98.1 to 99.1 [6,18,19]. Our specificity was lower than that of these studies, possibly due to the small sample size of FNAC results. We also found a relatively high diagnostic accuracy for FNAC according to the Milan system.

Wound infection was the most frequently occurring complication of surgery, constituting 6.4% of the complications for parotid gland surgery. The pooled incidence of wound infection in a systematic review and meta-analysis of minor complications of parotidectomy was 2.3%, occurring more frequently in extensive resections and surgery for malignant parotid gland tumours [14]. The second most common complications were temporary facial nerve palsy, hypoesthesia of the greater auricular nerve, and keloid scarring. The frequency of facial nerve palsy was consistent with other studies [3,25]. Two patients with benign tumours sustained temporary facial nerve palsy that was not present pre-operatively. Van Lierop et al. reported that temporary facial nerve weakness occurred in 34% of patients and permanent facial nerve weakness in 3% of patients with benign tumours of the parotid gland post-operatively [3]. The retrospective nature of this study posed an inherent limitation. Some records were incomplete, leading to fewer results for interpretation. Fine needle aspiration cytopathology was not performed uniformly in all patients with parotid gland masses, introducing selection bias into the study. This could have skewed the results of the comparison study of FNAC with histopathology, causing a challenge in the broad interpretation of the results. The study is also limited by the small number of patients who had FNAC performed.

Conclusion     

Pleomorphic adenoma and mucoepidermoid carcinoma were the most common benign and malignant tumours, respectively, in our context. FNAC has a high sensitivity and can assist with the pre-operative diagnosis of the neoplasms and surgical planning.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Ronewa Lorraine Mulaudzi: conceptualisation, methodology, data collection, data interpretation. writing-original draft. Realeboha Bothata Lesoli: methodology, data interpretation, reviewing and editing, supervision. Joseph Bukulu Sempa: methodology, data analysis, reviewing and editing. Riaz Yakoob Seedat: conceptualisation, methodology, data analysis, data interpretation, writing, review and editing, supervision. All the authors have read and agreed to the final manuscript.

Tables     

Table 1: cytology results of parotid masses according to the milan system for reporting salivary gland cytopathology (n=18)

Table 2: histopathology results of parotid masses (n=47)

Table 3: accuracy table comparing fine needle aspiration cytology (FNAC) and histopathology

References