Acta Oto-Laryngologica. 2014; 134: 1185–1191
ORIGINAL ARTICLE
A more appropriate clinical classification of benign parotid tumors: investigation of 425 cases
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
TAKAHIRO ICHIHARA, RYO KAWATA, MASAAKI HIGASHINO, TETSUYA TERADA & SHIN-ICHI HAGINOMORI Department of Otolaryngology, Head and Neck Surgery, Osaka Medical College, Takatsuki, Japan
Abstract Conclusions: It is appropriate to clinically classify benign parotid tumors into three groups, i.e. superficial tumors, deep tumors, and lower pole tumors. Objective: It is important to classify benign parotid tumors based on location when deciding the surgical strategy and conducting clinical research. In this study, we examined a classification of benign parotid tumors that was useful for clinical practice. Methods: We studied 425 patients with benign parotid tumors who underwent surgery at our hospital. Their age, gender, tumor histopathology, maximum tumor diameter, postoperative facial nerve paresis, operating time, and blood loss were investigated after classifying the tumors as superficial tumors, deep tumors, or lower pole tumors. We also investigated the same parameters after dividing the lower pole tumors into superficial and deep types. Results: Lower pole tumors had distinct characteristics from superficial and deep tumors. The incidence of facial nerve paresis was significantly higher and the operating time was significantly longer for deep tumors than for either superficial or lower pole tumors, while there were no significant differences between superficial and lower pole tumors. In addition, there were no significant differences in any of the parameters between the superficial and deep types of lower pole tumor.
Keywords: Lower pole tumor, facial nerve paralysis, pleomorphic adenoma, Warthin’s tumor
Introduction It is important to classify the location of parotid tumors appropriately for surgical planning as well as for clinical studies on postoperative complications and other outcomes. Traditionally, benign parotid tumors have been classified as superficial or deep tumors based on their location relative to the facial nerve [1]. The method of handling the facial nerve during surgery differs between superficial tumors and deep tumors. When resecting a superficial tumor, we only need to dissect the surface of the facial nerve in many cases. On the other hand, we need to dissect the facial nerve extensively when resecting deep tumors, even for partial parotidectomy. More invasive procedures are associated with a markedly increased risk of facial nerve damage and the incidence of postoperative facial nerve paresis is significantly higher for
patients with deep tumors than for those with superficial tumors [2–4]. Since the surgical procedure, degree of difficulty, and complications differ greatly between these two categories, it is considered reasonable to divide parotid tumors into superficial and deep tumors. When we assess tumor location, we find that parotid tumors often arise in the lower pole of the gland. If tumors arising in the region inferior to the marginal mandibular branch of the facial nerve are defined as lower pole tumors, many benign parotid tumors will be classified this way. Lower pole tumors arising in the region superficial to the marginal mandibular branch can be treated by a similar surgical procedure to that for superficial tumors. Even for lower pole tumors arising deep to the marginal mandibular branch, the surgical procedure is less difficult than that for resection of ‘true deep tumors’ arising in
Correspondence: Takahiro Ichihara, MD, Department of Otolaryngology, Head and Neck Surgery, Osaka Medical College, 2-7 Daigaku-machi, Takatsuki, 569-8686, Japan. Fax: +81 726 84 6539. E-mail: [email protected]
(Received 25 February 2014; accepted 3 April 2014) ISSN 0001-6489 print/ISSN 1651-2251 online 2014 Informa Healthcare DOI: 10.3109/00016489.2014.914246
1186
T. Ichihara et al.
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
the upper pole of the parotid gland, because we only need to protect and preserve the marginal mandibular branch. Furthermore, it is well known that Warthin’s tumor is frequent among benign tumors arising in the lower pole histopathologically [5,6]. Compared with pleomorphic adenoma, limited surgery is more acceptable for the treatment of Warthin’s tumor. In the present study, we retrospectively investigated 425 patients with benign parotid tumors who underwent surgery at our department and examined the most appropriate clinical classification of benign parotid tumors.
Material and methods Patients A total of 425 patients with benign parotid tumors underwent surgery at our department from September 1999 to February 2013. Based on the operative records, we initially classified the lesions into superficial tumors, deep tumors, and lower pole tumors according to location. We defined tumors arising in the region superficial to the main trunk of the facial nerve as superficial tumors, those arising deep to the main trunk of the facial nerve as deep tumors, and tumors that located in the region inferior to the marginal mandibular branch of the facial nerve as lower pole tumors. There were 219, 70, and 136 patients with superficial tumors, deep tumors, and lower pole tumors, respectively. In addition, we divided the lower pole tumors into two groups. Lower pole tumors arising in the region superficial to the marginal mandibular branch of the facial nerve were classified as the superficial type (n = 108) and those arising deep to the nerve were classified as the deep type (n = 28) (Figure 1). Main trunk Deep tumor Buccal branch
Superficial tumor Mandibular branch Lower pole tumor (deep type)
Lower pole tumor (superficial type)
Figure 1. Clinical classification of benign parotid tumors. We defined tumors arising superficial to the facial nerve as superficial tumors, those arising deep to the nerve as deep tumors, and those located inferior to the marginal mandibular branch as lower pole tumors.
Surgical procedure The surgical approach and technique we employed have been standardized since 1995 [4]. Surgery was performed under general anesthesia. To allow observation of movement of the facial muscles due to stimulation with the electric knife, a muscle relaxant was not used. An S-shaped skin incision was made, the operative field was exposed by the routine method, and the tragal pointer was exposed. Then the main trunk of the facial nerve was identified using the mastoid process, digastric muscle, and the tragal pointer as references. We performed partial superficial parotidectomy if the tumor was located in the lateral part of the gland (superficial tumors). After exposing the main trunk of the facial nerve, the nerve was only dissected where it passed through the part of the tumor that was removed. If the tumor was located in the medial part of the gland (deep tumors), partial deep parotidectomy was performed with dissection of the main trunk and the branches of the facial nerve. However, the part of the tumor closely attached to the nerve was sometimes removed by extracapsular dissection, which differed from enucleation. For a lower pole tumor, even if it was small, we first exposed the main trunk of the facial nerve as for a superficial tumor and dissected along the mandibular branch. Neither an operating microscope nor a magnifying glass was used and electromyographic monitoring was not done. The operating time was measured from the performance of skin incision to the last skin suture. Methods To understand the characteristics of superficial tumors, deep tumors, and lower pole tumors, we investigated the age, gender, tumor histopathology, and maximum tumor diameter (measured by ultrasonography before surgery). To evaluate surgical invasiveness and difficulty, we examined the incidence of postoperative facial nerve paresis, the operating time, and blood loss. We also divided lower pole tumors into superficial and deep types, and then investigated the same parameters (Figure 2). Moreover, we examined cases of pleomorphic adenoma, which are an indication for surgery in general. Statistical analysis For statistical procedures, we used R software (version 2.14.1). For analysis of continuous variables, multiple comparison was performed with the SteelDwass test, while non-parametric or two-group comparison was performed using the Mann–Whitney U
Clinical classification of benign parotid tumors
1187
Results
A
Histopathology Among the 425 benign parotid tumors, there were 230 pleomorphic adenomas (54%), 136 Warthin’s tumors (32%), 21 cysts (5%), 14 basal cell adenomas (3%), and 24 other tumors (6%). Among the 219 superficial tumors, there were 144 pleomorphic adenomas (66%) and 44 Warthin’s tumors (20%), while the 70 deep tumors comprised 52 pleomorphic adenomas (74%) and 8 Warthin’s tumors (11%). The 136 lower pole tumors included 34 pleomorphic adenomas (25%) and 84 Warthin’s tumors (62%) (Figure 3). Among the pleomorphic adenomas, there were 144 superficial tumors, 52 deep tumors, and 34 lower pole tumors. There were 108 superficial lower pole tumors, including 26 pleomorphic adenomas (24%) and 70 Warthin’s tumors (65%), while the 28 deep lower pole tumors comprised 8 pleomorphic adenomas (29%) and 14 Warthin’s tumors (50%). The mean age was 48.5 years for patients with pleomorphic adenoma versus 61.9 years for patients with Warthin’s tumor, and the latter group was 13.4 years older. The male:female ratio was 73:157 for pleomorphic adenoma and 115:21 for Warthin’s tumor. Since pleomorphic adenoma and Warthin’s tumor accounted for 86% of all tumors, we performed further comparison by focusing on these two histological types.
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
Tumor
B
Tumor
Figure 2. Intraoperative photograph of a lower pole tumor (right parotid gland). We defined tumors arising superficial to the marginal mandibular branch of the facial nerve as the superficial type of lower pole tumor (a) and those arising in the deeper region as the deep type of lower pole tumor (b). Arrow, mandibular branch of the facial nerve.
Comparison of superficial tumors, deep tumors, and lower pole tumors (Table I)
test. For nominal variables, the chi-squared test of independence, a nonparametric method, was employed. If there was a significant difference by multiple comparison, intergroup comparison was performed using the Ryan-Einot-Gabriel-Welsch multiple range test. For all statistical analyses, p < 0.05 was considered to indicate a significant difference. A
Based on location, we divided the tumors into three groups and compared the age, gender, histopathology, maximum diameter, postoperative facial paresis, operating time, and blood loss. The patients with lower pole tumors had a mean age of 57.4 years and were significantly older than the patients with superficial tumors (mean age of 52.2 years, p = 0.005)
B Other tumors n = 31 14%
Warthin’s tumor n = 44 20%
C Other tumors n = 10 14%
Warthin’s tumor n = 8 11%
Pleomorphic adenoma n = 144 66%
Pleomorphic adenoma n = 52 74%
Other tumors n = 18 13%
Pleomorphic adenoma n = 34 25%
Warthin’s tumor n = 84 62%
Figure 3. Histopathological diagnosis of 425 benign parotid tumors according to location. (a) Superficial tumors; (b) deep tumors; (c) lower pole tumors. Pleomorphic adenoma and Warthin’s tumor accounted for 86% of all tumors.
1188
T. Ichihara et al.
Table I. Comparison of superficial tumors, deep tumors, and lower pole tumors. Superficial tumors (n = 219)
Deep tumors (n = 70)
Lower pole tumors (n = 136)
Age (years)
52.2 ± 16.4
49.6 ± 15.7
57.4 ± 13.3
Male:female
94:125
28:42
84:52
1
34:84
0.143
< 0.001
< 0.001
0.113
0.464
0.619
< 0.001
0.621
< 0.001
Ryan’s method
< 0.001
0.036
< 0.001
Steel-Dwass test
0.036
0.978
0.039
Steel-Dwass test
Characteristic
Histopathology (P:W) Maximum diameter (mm)
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
Facial paresis (%)
144:44
52:8
25.3 ± 9.9
27.4 ± 9.4
16.4
26.5 ± 9.9
45.7
13.2
Operating time (min)
120 ± 27
161 ± 62
112 ± 32
Blood loss (ml)
24.2 ± 20.8
34.9 ± 33.8
24.5 ± 24.4
p value S-D 0.401
D-L
Statistical method
0.005
< 0.001
Steel-Dwass test
< 0.001
0.009
S-L
Ryan’s method Ryan’s method Steel-Dwass test
D, deep tumor; L, lower pole tumor; P, pleomorphic adenoma; Ryan’s method, Ryan-Einot-Gabriel-Welsch multiple range test; S, superficial tumor; W, Warthin’s tumor.
and deep tumors with regard to age, gender, and histopathology. When postoperative facial paresis was compared, the incidence was 16.4%, 45.7%, and 13.2% in patients who had superficial tumors, deep tumors, and lower pole tumors, respectively. The incidence was significantly higher for deep tumors than for superficial tumors or lower pole tumors (both p < 0.001), while there was no significant difference between superficial tumors and lower pole tumors. Paresis generally affected one branch, especially the marginal mandibular branch. In all patients, paresis was temporary and there were no permanent cases. The mean operating time was 120, 161, and 112 min for resection of superficial tumors, deep tumors, and lower pole tumors, respectively, and it was significantly longer for deep tumors than for superficial tumors or lower pole tumors (both p < 0.001). The same trend was also seen for blood loss, as it was significantly larger for deep tumors than superficial tumors or lower pole tumors (p = 0.036 and p = 0.039, respectively), but there was no significant
or those with deep tumors (mean age of 49.6 years, p < 0.001). With regard to gender, the male:female ratio was 1.6:1 for lower pole tumors, while the male: female ratio was 0.8:1 and 0.7:1 for superficial tumors and deep tumors, respectively (both p < 0.001). Age and gender showed no significant differences between patients with superficial tumors and deep tumors. With regard to histopathology, the percentage of pleomorphic adenomas was 3.3 times and 6.5 times higher than that of Warthin’s tumors for superficial tumors and deep tumors, respectively, while the percentage of Warthin’s tumors was 2.5 times higher than that of pleomorphic adenomas for lower pole tumors. There was a significant difference between superficial tumors and lower pole tumors, as well as between deep tumors and lower pole tumors (both p < 0.001). The mean maximum tumor diameter was 25.3, 27.4, and 26.5 mm for superficial tumors, deep tumors, and lower pole tumors, respectively, and no significant differences were observed among the groups. Thus, the characteristics of lower pole tumors were distinct from those of superficial
Table II. Comparison of the superficial and deep types of lower pole tumor. Characteristic Age (years)
LS (n = 108)
LD (n = 28)
p value
56.9 ± 14.1
59.1 ± 1.76
0.891
Statistical method Mann–Whitney U test
Male:female
68:40
16:12
0.572
Chi-squared test of independence
Histopathology (P:W)
26:70
08:14
0.386
Chi-squared test of independence
26.4 ± 10.4
26.7 ± 7.6
0.508
Mann–Whitney U test
12
17.9
0.418
Chi-squared test of independence
109.0 ± 28.7
126.1 ± 39.2
0.059
Mann–Whitney U test
21.9 ± 19.1
34.3 ± 37.5
0.087
Mann–Whitney U test
Maximum diameter (mm) Facial paresis (%) Operating time (min) Blood loss (ml)
LD, lower pole tumor (deep type); LS, lower pole tumor (superficial type).
Clinical classification of benign parotid tumors difference between superficial tumors and lower pole tumors. Thus, deep tumors have different characteristics from superficial tumors and lower pole tumors with regard to surgical difficulty and complications.
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
Comparison of the superficial and deep types of lower pole tumors (Table II) We divided lower pole tumors into two groups, which were the superficial type (LS) and the deep type (LD), and then investigated the same parameters. In terms of clinicopathological characteristics (age, gender, histopathology, and maximum diameter), there were no differences between LS and LD. Also, the parameters of surgical invasiveness and operative difficulty (postoperative facial nerve paresis, operating time, and blood loss) showed no significant differences between the LS and LD groups. The operating time was longer and blood loss was somewhat larger in the LD group than LS group but there was no significant difference (p = 0.059 and p = 0.087, respectively). Investigation of pleomorphic adenoma (Table III) Pleomorphic adenomas are generally accepted as an indication for surgery due to the possibility for malignant change. Therefore, we divided pleomorphic adenomas into three groups and compared the age, gender, maximum diameter, postoperative facial paresis, operating time, and blood loss. Age and gender showed no significant differences among the three groups. However, the maximum diameter of deep tumors was larger than that of superficial tumors and lower pole tumors. Also, the incidence of postoperative facial paresis and operating time were significantly higher for deep tumors than for superficial tumors and lower pole tumors, which concurred with the results of the study on whole benign tumors.
1189
Discussion In patients with parotid tumors, the location determines the surgical approach and operative difficulty, as well as the incidence of postoperative complications including facial nerve paresis. Pleomorphic adenoma arising in the lateral part of the parotid gland is the most common type of parotid tumor, but there is controversy regarding the best surgical approach for it. The goal of surgery is to completely resect the tumor and avoid recurrence without causing postoperative facial nerve paresis [7,8]. Recently, it has been proposed that limited superficial parotidectomy, i.e. partial superficial parotidectomy (PSP), is a useful surgical procedure for pleomorphic adenoma in the lateral lobe [9]. More recently, the very limited procedure of extracapsular dissection (ECD) has been reported [10–12]. ECD involves removing the tumor without identifying the facial nerve and it is considered to cause fewer postoperative complications compared with PSP. However, this method can only be used for small superficial tumors located in the lateral lobe. Dell’Aversana Orabona et al. [13] studied 232 patients and reported that the mean diameter of the tumors treated by PSP was 3.49 cm, while it was 1.89 cm for those treated by ECD. Since tumors with a diameter of more than 2 cm are usually adjacent to a branch of the facial nerve, the standard surgical procedure for parotid tumors may be PSP apart from small tumors. Even if we perform PSP, it is difficult to remove the tumor with an appropriate margin of normal tissue all around, and similar resection to ECD is required at any site where the tumor is adjacent to a nerve. When deciding which surgical approach to employ for the treatment of parotid tumors, we clearly need to classify the tumors by location. It is also important to classify tumors appropriately by location to predict postoperative complications as well as to investigate operative outcomes accurately.
Table III. Investigation of pleomorphic adenoma. p value Characteristic Age (years) Male:female Maximum diameter (mm) Facial paresis (%)
Superficial tumors (n = 144)
Deep tumors (n = 52)
Lower pole tumors (n = 34)
48.6 ± 16.5
46.6 ± 14.4
50.2 ± 16.4
46:98 24.0 ± 9.6 14.6
18:34 28.5 ± 9.5 40.4
09:25 22.7 ± 7.6 14.7
Operating time (min)
118 ± 26
163 ± 68
113 ± 41
Blood loss (ml)
20.8 ± 18.1
34.7 ± 35.0
21.3 ± 20.5
S-D
S-L
Statistical method
D-L
0.654
0.794
0.465
Steel-Dwass test
1
0.802
1
Ryan’s method
0.004
0.876
0.008
Steel-Dwass test
0.017
Ryan’s method
< 0.001
1
< 0.001
0.347
< 0.001
Steel-Dwass test
0.005
0.999
0.065
Steel-Dwass test
D, deep tumor; L, lower pole tumor; Ryan’s method, Ryan-Einot-Gabriel-Welsch multiple range test; S, superficial tumor.
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
1190
T. Ichihara et al.
In many reports on the clinical features of benign parotid tumors, the tumors are classified into two groups (superficial tumors and deep tumors) according to their location relative to the facial nerve. Since deep tumors are located under the facial nerve, surgical resection is more difficult than for superficial tumors. Indeed, the incidence of postoperative facial nerve paresis, the most critical surgical complication, shows a marked difference between superficial tumors and deep tumors [2–4]. In this study, there was a significant difference in the incidence of postoperative facial paresis between tumor locations (superficial lobe vs deep lobe), but no difference was observed with respect to age, gender, histopathology, and maximum diameter. Bron and O’Brien [2] also investigated the incidence of postoperative facial nerve paresis and concluded that tumor location is the most important factor apart from recurrence and inflammation. Moreover, Harney et al. [14] compared histological findings between pleomorphic adenomas in the lateral lobe and the medial lobe, and reported that superficial tumors tended to have a thinner capsule and a higher rate of extracapsular growth. Therefore, from the viewpoints of surgical difficulty, complications, and tumor characteristics, it is reasonable to classify parotid tumors into two groups, which are superficial tumors and deep tumors. Parotid tumors often occur in the lower pole of the gland, and there were 136 lower pole tumors (32%) among a total of 425 parotid tumors in the present study. We compared lower pole tumors with superficial and deep tumors, identifying significant differences of age, gender, and histopathology that may be explained by the higher prevalence of Warthin’s tumors among lower pole tumors [4,5]. The differences of age and gender were probably observed because Warthin’s tumors are frequently seen in elderly men. Since lower pole tumors have distinct characteristics from superficial tumors and deep tumors, it was considered appropriate to separate lower pole tumors in our classification. The incidence of facial paresis was similar for lower pole tumors and superficial tumors. In surgery for lower pole tumors, we should mainly pay attention to identifying and preserving the marginal mandibular branch of the facial nerve and it is not necessary to perform wide dissection of the nerve. Even if a lower pole tumor is located in the deep region of the gland, surgery is relatively simple compared with resection of a ‘true deep tumor’ arising in the deep region of the upper pole of the parotid gland, which requires identification and preservation of several branches of the facial nerve. Some reports of a low incidence of facial nerve paresis associated with deep tumors may well
have included patients with deep lower pole tumors. When we compared the operating time and blood loss as parameters of surgical invasiveness, there was a significant difference between lower pole tumors and deep tumors, but no difference was observed between lower pole tumors and superficial tumors. Thus, in terms of operative difficulty and complications, lower pole tumors and superficial tumors are comparable, while resection of deep tumors is associated with greater difficulty and a high incidence of postoperative facial nerve paresis. Lower pole tumors can arise superficial to the marginal mandibular branch of the facial nerve (LS) or in the deeper region of the lower pole (LD). To determine whether we should separate these two types, we compared the LS and LD tumor types. As a result, there were no differences of tumor characteristics (age, gender, histopathology, and maximum diameter) between the two types. In addition, there was no difference in surgical difficulty and complications (postoperative facial nerve paresis, operating time, and blood loss) between the two types. Therefore, we consider that there is little clinical value in classifying lower pole tumors according to depth and it appears to be appropriate to combine the LS and LD types in one group. Based on the above results, it seems appropriate to clinically classify benign parotid tumors into three groups, i.e. superficial tumors, deep tumors, and lower pole tumors. Our study on pleomorphic adenomas, which are generally accepted for indication of surgery, yielded nearly the same results. Therefore, it seems appropriate to clinically classify pleomorphic tumors into three groups. Even though we feel it appropriate to classify pleomorphic tumors into three groups, it is useful for us to diagnose them before operation because of operative methods and explanation for the patient. We have recently made a diagnostic criterion of differential diagnosis for superficial and deep tumors by using ultrasonopraphy [15]. The minimum thickness of normal parotid gland tissue between the parotideomasseteric fascia and tumor showed significant differences between superficial and deep tumors. On the other hand, mandibular branch and the location of the tumor were key points in the diagnosis of lower pole tumors. Therefore, it may be possible to diagnose lower pole tumors using CT/MRI. Conclusion We examined the clinical classification of benign parotid tumors according to anatomic location. As a result, we found that lower pole tumors have distinct characteristics (including differences of age, gender,
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
Clinical classification of benign parotid tumors and histopathology) compared with superficial tumors and deep tumors. In addition, we examined postoperative facial nerve paresis, the operating time, and operative blood loss as parameters of surgical difficulty and complications, and found that resection of deep tumors was associated with significantly greater difficulty and a higher incidence of complications compared with resection of superficial tumors and lower pole tumors. When we compared lower pole tumors arising superficial to the marginal mandibular branch of the facial nerve with those arising in the deeper region, no significant differences were observed, so we concluded that both types should be combined in one group as lower pole tumors. Therefore, when benign parotid tumors are classified according to location, it may be appropriate to separate them into three groups: superficial tumors, deep tumors, and lower pole tumors.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Imaizumi A, Kuribayashi A, Okochi K, Ishii J, Sumi Y, Yoshino N, et al. Differentiation between superficial and deep lobe parotid tumors by magnetic resonance imaging: usefulness of the parotid duct criterion. Acta Radiol 2009;50: 806–11. [2] Bron LP, O’Brien CJ. Facial nerve function after parotidectomy. Arch Otolaryngol Head Neck Surg 1997; 123:1091–6. [3] Gaillard C, Perie S, Susini B, St Guily JL. Facial nerve dysfunction after parotidectomy: the role of local factors. Laryngoscope 2005;115:287–91.
1191
[4] Kawata R, Lee K, Araki M, Takenaka H. Safety and usefulness of an electric knife during surgery for parotid benign tumor: postoperative facial paresis and its risk factors. Acta Otolaryngol 2007;127:966–9. [5] Yamashita T, Tomoda K, Kumazawa T. The usefulness of partial parotidectomy for benign parotid gland tumors. A retrospective study of 306 cases. Acta Otolaryngol Suppl 1993;500:113–16. [6] Przewozny T, Stankiewicz C, Narozny W, Kuczkowski J. Warthin’s tumor of the parotid gland. Epidemiological and clinical analysis of 127 cases. Otolaryngol Pol 2004;58:583–92. [7] Witt RL. The significance of the margin in parotid surgery for pleomorphic adenoma. Laryngoscope 2002;112:2141–54. [8] Emodi O, El-Naaj IA, Gordin A, Akrish S, Peled M. Superficial parotidectomy versus retrograde partial superficial parotidectomy in treating benign salivary gland tumor (pleomorphic adenoma). J Oral Maxillofac Surg 2010;68:2092–8. [9] O’Brien CJ. Current management of benign parotid tumors— the role of limited superficial parotidectomy. Head Neck 2003; 25:946–52. [10] Witt RL, Rejto L. Pleomorphic adenoma: extracapsular dissection versus partial superficial parotidectomy with facial nerve dissection. Del Med J 2009;81:119–25. [11] Albergotti WG, Nguyen SA, Zenk J, Gillespie MB. Extracapsular dissection for benign parotid tumors: a metaanalysis. Laryngoscope 2012;122:1954–60. [12] Zbären P, Vander Poorten V, Witt RL, Woolgar JA, Shaha AR, Triantafyllou A, et al. Pleomorphic adenoma of the parotid: formal parotidectomy or limited surgery? Am J Surg 2013;205:109–18. [13] Dell’Aversana Orabona G, Bonavolontà P, Iaconetta G, Forte R, Califano L. Surgical management of benign tumors of the parotid gland: extracapsular dissection versus superficial parotidectomy – our experience in 232 cases. J Oral Maxillofac Surg 2013;71:410–13. [14] Harney MS, Murphy C, Hone S, Toner M, Timon CV. A histological comparison of deep and superficial lobe pleomorphic adenomas of the parotid gland. Head Neck 2003;25: 649–53. [15] Higashino M, Kawata R, Haginomori S, Lee K, Yoshimura K, Nishikawa S. A novel differential diagnostic method for superficial/deep tumor of the parotid gland using ultrasonography. Head Neck 2013;35:1153–7.
ORIGINAL ARTICLE
A more appropriate clinical classification of benign parotid tumors: investigation of 425 cases
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
TAKAHIRO ICHIHARA, RYO KAWATA, MASAAKI HIGASHINO, TETSUYA TERADA & SHIN-ICHI HAGINOMORI Department of Otolaryngology, Head and Neck Surgery, Osaka Medical College, Takatsuki, Japan
Abstract Conclusions: It is appropriate to clinically classify benign parotid tumors into three groups, i.e. superficial tumors, deep tumors, and lower pole tumors. Objective: It is important to classify benign parotid tumors based on location when deciding the surgical strategy and conducting clinical research. In this study, we examined a classification of benign parotid tumors that was useful for clinical practice. Methods: We studied 425 patients with benign parotid tumors who underwent surgery at our hospital. Their age, gender, tumor histopathology, maximum tumor diameter, postoperative facial nerve paresis, operating time, and blood loss were investigated after classifying the tumors as superficial tumors, deep tumors, or lower pole tumors. We also investigated the same parameters after dividing the lower pole tumors into superficial and deep types. Results: Lower pole tumors had distinct characteristics from superficial and deep tumors. The incidence of facial nerve paresis was significantly higher and the operating time was significantly longer for deep tumors than for either superficial or lower pole tumors, while there were no significant differences between superficial and lower pole tumors. In addition, there were no significant differences in any of the parameters between the superficial and deep types of lower pole tumor.
Keywords: Lower pole tumor, facial nerve paralysis, pleomorphic adenoma, Warthin’s tumor
Introduction It is important to classify the location of parotid tumors appropriately for surgical planning as well as for clinical studies on postoperative complications and other outcomes. Traditionally, benign parotid tumors have been classified as superficial or deep tumors based on their location relative to the facial nerve [1]. The method of handling the facial nerve during surgery differs between superficial tumors and deep tumors. When resecting a superficial tumor, we only need to dissect the surface of the facial nerve in many cases. On the other hand, we need to dissect the facial nerve extensively when resecting deep tumors, even for partial parotidectomy. More invasive procedures are associated with a markedly increased risk of facial nerve damage and the incidence of postoperative facial nerve paresis is significantly higher for
patients with deep tumors than for those with superficial tumors [2–4]. Since the surgical procedure, degree of difficulty, and complications differ greatly between these two categories, it is considered reasonable to divide parotid tumors into superficial and deep tumors. When we assess tumor location, we find that parotid tumors often arise in the lower pole of the gland. If tumors arising in the region inferior to the marginal mandibular branch of the facial nerve are defined as lower pole tumors, many benign parotid tumors will be classified this way. Lower pole tumors arising in the region superficial to the marginal mandibular branch can be treated by a similar surgical procedure to that for superficial tumors. Even for lower pole tumors arising deep to the marginal mandibular branch, the surgical procedure is less difficult than that for resection of ‘true deep tumors’ arising in
Correspondence: Takahiro Ichihara, MD, Department of Otolaryngology, Head and Neck Surgery, Osaka Medical College, 2-7 Daigaku-machi, Takatsuki, 569-8686, Japan. Fax: +81 726 84 6539. E-mail: [email protected]
(Received 25 February 2014; accepted 3 April 2014) ISSN 0001-6489 print/ISSN 1651-2251 online 2014 Informa Healthcare DOI: 10.3109/00016489.2014.914246
1186
T. Ichihara et al.
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
the upper pole of the parotid gland, because we only need to protect and preserve the marginal mandibular branch. Furthermore, it is well known that Warthin’s tumor is frequent among benign tumors arising in the lower pole histopathologically [5,6]. Compared with pleomorphic adenoma, limited surgery is more acceptable for the treatment of Warthin’s tumor. In the present study, we retrospectively investigated 425 patients with benign parotid tumors who underwent surgery at our department and examined the most appropriate clinical classification of benign parotid tumors.
Material and methods Patients A total of 425 patients with benign parotid tumors underwent surgery at our department from September 1999 to February 2013. Based on the operative records, we initially classified the lesions into superficial tumors, deep tumors, and lower pole tumors according to location. We defined tumors arising in the region superficial to the main trunk of the facial nerve as superficial tumors, those arising deep to the main trunk of the facial nerve as deep tumors, and tumors that located in the region inferior to the marginal mandibular branch of the facial nerve as lower pole tumors. There were 219, 70, and 136 patients with superficial tumors, deep tumors, and lower pole tumors, respectively. In addition, we divided the lower pole tumors into two groups. Lower pole tumors arising in the region superficial to the marginal mandibular branch of the facial nerve were classified as the superficial type (n = 108) and those arising deep to the nerve were classified as the deep type (n = 28) (Figure 1). Main trunk Deep tumor Buccal branch
Superficial tumor Mandibular branch Lower pole tumor (deep type)
Lower pole tumor (superficial type)
Figure 1. Clinical classification of benign parotid tumors. We defined tumors arising superficial to the facial nerve as superficial tumors, those arising deep to the nerve as deep tumors, and those located inferior to the marginal mandibular branch as lower pole tumors.
Surgical procedure The surgical approach and technique we employed have been standardized since 1995 [4]. Surgery was performed under general anesthesia. To allow observation of movement of the facial muscles due to stimulation with the electric knife, a muscle relaxant was not used. An S-shaped skin incision was made, the operative field was exposed by the routine method, and the tragal pointer was exposed. Then the main trunk of the facial nerve was identified using the mastoid process, digastric muscle, and the tragal pointer as references. We performed partial superficial parotidectomy if the tumor was located in the lateral part of the gland (superficial tumors). After exposing the main trunk of the facial nerve, the nerve was only dissected where it passed through the part of the tumor that was removed. If the tumor was located in the medial part of the gland (deep tumors), partial deep parotidectomy was performed with dissection of the main trunk and the branches of the facial nerve. However, the part of the tumor closely attached to the nerve was sometimes removed by extracapsular dissection, which differed from enucleation. For a lower pole tumor, even if it was small, we first exposed the main trunk of the facial nerve as for a superficial tumor and dissected along the mandibular branch. Neither an operating microscope nor a magnifying glass was used and electromyographic monitoring was not done. The operating time was measured from the performance of skin incision to the last skin suture. Methods To understand the characteristics of superficial tumors, deep tumors, and lower pole tumors, we investigated the age, gender, tumor histopathology, and maximum tumor diameter (measured by ultrasonography before surgery). To evaluate surgical invasiveness and difficulty, we examined the incidence of postoperative facial nerve paresis, the operating time, and blood loss. We also divided lower pole tumors into superficial and deep types, and then investigated the same parameters (Figure 2). Moreover, we examined cases of pleomorphic adenoma, which are an indication for surgery in general. Statistical analysis For statistical procedures, we used R software (version 2.14.1). For analysis of continuous variables, multiple comparison was performed with the SteelDwass test, while non-parametric or two-group comparison was performed using the Mann–Whitney U
Clinical classification of benign parotid tumors
1187
Results
A
Histopathology Among the 425 benign parotid tumors, there were 230 pleomorphic adenomas (54%), 136 Warthin’s tumors (32%), 21 cysts (5%), 14 basal cell adenomas (3%), and 24 other tumors (6%). Among the 219 superficial tumors, there were 144 pleomorphic adenomas (66%) and 44 Warthin’s tumors (20%), while the 70 deep tumors comprised 52 pleomorphic adenomas (74%) and 8 Warthin’s tumors (11%). The 136 lower pole tumors included 34 pleomorphic adenomas (25%) and 84 Warthin’s tumors (62%) (Figure 3). Among the pleomorphic adenomas, there were 144 superficial tumors, 52 deep tumors, and 34 lower pole tumors. There were 108 superficial lower pole tumors, including 26 pleomorphic adenomas (24%) and 70 Warthin’s tumors (65%), while the 28 deep lower pole tumors comprised 8 pleomorphic adenomas (29%) and 14 Warthin’s tumors (50%). The mean age was 48.5 years for patients with pleomorphic adenoma versus 61.9 years for patients with Warthin’s tumor, and the latter group was 13.4 years older. The male:female ratio was 73:157 for pleomorphic adenoma and 115:21 for Warthin’s tumor. Since pleomorphic adenoma and Warthin’s tumor accounted for 86% of all tumors, we performed further comparison by focusing on these two histological types.
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
Tumor
B
Tumor
Figure 2. Intraoperative photograph of a lower pole tumor (right parotid gland). We defined tumors arising superficial to the marginal mandibular branch of the facial nerve as the superficial type of lower pole tumor (a) and those arising in the deeper region as the deep type of lower pole tumor (b). Arrow, mandibular branch of the facial nerve.
Comparison of superficial tumors, deep tumors, and lower pole tumors (Table I)
test. For nominal variables, the chi-squared test of independence, a nonparametric method, was employed. If there was a significant difference by multiple comparison, intergroup comparison was performed using the Ryan-Einot-Gabriel-Welsch multiple range test. For all statistical analyses, p < 0.05 was considered to indicate a significant difference. A
Based on location, we divided the tumors into three groups and compared the age, gender, histopathology, maximum diameter, postoperative facial paresis, operating time, and blood loss. The patients with lower pole tumors had a mean age of 57.4 years and were significantly older than the patients with superficial tumors (mean age of 52.2 years, p = 0.005)
B Other tumors n = 31 14%
Warthin’s tumor n = 44 20%
C Other tumors n = 10 14%
Warthin’s tumor n = 8 11%
Pleomorphic adenoma n = 144 66%
Pleomorphic adenoma n = 52 74%
Other tumors n = 18 13%
Pleomorphic adenoma n = 34 25%
Warthin’s tumor n = 84 62%
Figure 3. Histopathological diagnosis of 425 benign parotid tumors according to location. (a) Superficial tumors; (b) deep tumors; (c) lower pole tumors. Pleomorphic adenoma and Warthin’s tumor accounted for 86% of all tumors.
1188
T. Ichihara et al.
Table I. Comparison of superficial tumors, deep tumors, and lower pole tumors. Superficial tumors (n = 219)
Deep tumors (n = 70)
Lower pole tumors (n = 136)
Age (years)
52.2 ± 16.4
49.6 ± 15.7
57.4 ± 13.3
Male:female
94:125
28:42
84:52
1
34:84
0.143
< 0.001
< 0.001
0.113
0.464
0.619
< 0.001
0.621
< 0.001
Ryan’s method
< 0.001
0.036
< 0.001
Steel-Dwass test
0.036
0.978
0.039
Steel-Dwass test
Characteristic
Histopathology (P:W) Maximum diameter (mm)
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
Facial paresis (%)
144:44
52:8
25.3 ± 9.9
27.4 ± 9.4
16.4
26.5 ± 9.9
45.7
13.2
Operating time (min)
120 ± 27
161 ± 62
112 ± 32
Blood loss (ml)
24.2 ± 20.8
34.9 ± 33.8
24.5 ± 24.4
p value S-D 0.401
D-L
Statistical method
0.005
< 0.001
Steel-Dwass test
< 0.001
0.009
S-L
Ryan’s method Ryan’s method Steel-Dwass test
D, deep tumor; L, lower pole tumor; P, pleomorphic adenoma; Ryan’s method, Ryan-Einot-Gabriel-Welsch multiple range test; S, superficial tumor; W, Warthin’s tumor.
and deep tumors with regard to age, gender, and histopathology. When postoperative facial paresis was compared, the incidence was 16.4%, 45.7%, and 13.2% in patients who had superficial tumors, deep tumors, and lower pole tumors, respectively. The incidence was significantly higher for deep tumors than for superficial tumors or lower pole tumors (both p < 0.001), while there was no significant difference between superficial tumors and lower pole tumors. Paresis generally affected one branch, especially the marginal mandibular branch. In all patients, paresis was temporary and there were no permanent cases. The mean operating time was 120, 161, and 112 min for resection of superficial tumors, deep tumors, and lower pole tumors, respectively, and it was significantly longer for deep tumors than for superficial tumors or lower pole tumors (both p < 0.001). The same trend was also seen for blood loss, as it was significantly larger for deep tumors than superficial tumors or lower pole tumors (p = 0.036 and p = 0.039, respectively), but there was no significant
or those with deep tumors (mean age of 49.6 years, p < 0.001). With regard to gender, the male:female ratio was 1.6:1 for lower pole tumors, while the male: female ratio was 0.8:1 and 0.7:1 for superficial tumors and deep tumors, respectively (both p < 0.001). Age and gender showed no significant differences between patients with superficial tumors and deep tumors. With regard to histopathology, the percentage of pleomorphic adenomas was 3.3 times and 6.5 times higher than that of Warthin’s tumors for superficial tumors and deep tumors, respectively, while the percentage of Warthin’s tumors was 2.5 times higher than that of pleomorphic adenomas for lower pole tumors. There was a significant difference between superficial tumors and lower pole tumors, as well as between deep tumors and lower pole tumors (both p < 0.001). The mean maximum tumor diameter was 25.3, 27.4, and 26.5 mm for superficial tumors, deep tumors, and lower pole tumors, respectively, and no significant differences were observed among the groups. Thus, the characteristics of lower pole tumors were distinct from those of superficial
Table II. Comparison of the superficial and deep types of lower pole tumor. Characteristic Age (years)
LS (n = 108)
LD (n = 28)
p value
56.9 ± 14.1
59.1 ± 1.76
0.891
Statistical method Mann–Whitney U test
Male:female
68:40
16:12
0.572
Chi-squared test of independence
Histopathology (P:W)
26:70
08:14
0.386
Chi-squared test of independence
26.4 ± 10.4
26.7 ± 7.6
0.508
Mann–Whitney U test
12
17.9
0.418
Chi-squared test of independence
109.0 ± 28.7
126.1 ± 39.2
0.059
Mann–Whitney U test
21.9 ± 19.1
34.3 ± 37.5
0.087
Mann–Whitney U test
Maximum diameter (mm) Facial paresis (%) Operating time (min) Blood loss (ml)
LD, lower pole tumor (deep type); LS, lower pole tumor (superficial type).
Clinical classification of benign parotid tumors difference between superficial tumors and lower pole tumors. Thus, deep tumors have different characteristics from superficial tumors and lower pole tumors with regard to surgical difficulty and complications.
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
Comparison of the superficial and deep types of lower pole tumors (Table II) We divided lower pole tumors into two groups, which were the superficial type (LS) and the deep type (LD), and then investigated the same parameters. In terms of clinicopathological characteristics (age, gender, histopathology, and maximum diameter), there were no differences between LS and LD. Also, the parameters of surgical invasiveness and operative difficulty (postoperative facial nerve paresis, operating time, and blood loss) showed no significant differences between the LS and LD groups. The operating time was longer and blood loss was somewhat larger in the LD group than LS group but there was no significant difference (p = 0.059 and p = 0.087, respectively). Investigation of pleomorphic adenoma (Table III) Pleomorphic adenomas are generally accepted as an indication for surgery due to the possibility for malignant change. Therefore, we divided pleomorphic adenomas into three groups and compared the age, gender, maximum diameter, postoperative facial paresis, operating time, and blood loss. Age and gender showed no significant differences among the three groups. However, the maximum diameter of deep tumors was larger than that of superficial tumors and lower pole tumors. Also, the incidence of postoperative facial paresis and operating time were significantly higher for deep tumors than for superficial tumors and lower pole tumors, which concurred with the results of the study on whole benign tumors.
1189
Discussion In patients with parotid tumors, the location determines the surgical approach and operative difficulty, as well as the incidence of postoperative complications including facial nerve paresis. Pleomorphic adenoma arising in the lateral part of the parotid gland is the most common type of parotid tumor, but there is controversy regarding the best surgical approach for it. The goal of surgery is to completely resect the tumor and avoid recurrence without causing postoperative facial nerve paresis [7,8]. Recently, it has been proposed that limited superficial parotidectomy, i.e. partial superficial parotidectomy (PSP), is a useful surgical procedure for pleomorphic adenoma in the lateral lobe [9]. More recently, the very limited procedure of extracapsular dissection (ECD) has been reported [10–12]. ECD involves removing the tumor without identifying the facial nerve and it is considered to cause fewer postoperative complications compared with PSP. However, this method can only be used for small superficial tumors located in the lateral lobe. Dell’Aversana Orabona et al. [13] studied 232 patients and reported that the mean diameter of the tumors treated by PSP was 3.49 cm, while it was 1.89 cm for those treated by ECD. Since tumors with a diameter of more than 2 cm are usually adjacent to a branch of the facial nerve, the standard surgical procedure for parotid tumors may be PSP apart from small tumors. Even if we perform PSP, it is difficult to remove the tumor with an appropriate margin of normal tissue all around, and similar resection to ECD is required at any site where the tumor is adjacent to a nerve. When deciding which surgical approach to employ for the treatment of parotid tumors, we clearly need to classify the tumors by location. It is also important to classify tumors appropriately by location to predict postoperative complications as well as to investigate operative outcomes accurately.
Table III. Investigation of pleomorphic adenoma. p value Characteristic Age (years) Male:female Maximum diameter (mm) Facial paresis (%)
Superficial tumors (n = 144)
Deep tumors (n = 52)
Lower pole tumors (n = 34)
48.6 ± 16.5
46.6 ± 14.4
50.2 ± 16.4
46:98 24.0 ± 9.6 14.6
18:34 28.5 ± 9.5 40.4
09:25 22.7 ± 7.6 14.7
Operating time (min)
118 ± 26
163 ± 68
113 ± 41
Blood loss (ml)
20.8 ± 18.1
34.7 ± 35.0
21.3 ± 20.5
S-D
S-L
Statistical method
D-L
0.654
0.794
0.465
Steel-Dwass test
1
0.802
1
Ryan’s method
0.004
0.876
0.008
Steel-Dwass test
0.017
Ryan’s method
< 0.001
1
< 0.001
0.347
< 0.001
Steel-Dwass test
0.005
0.999
0.065
Steel-Dwass test
D, deep tumor; L, lower pole tumor; Ryan’s method, Ryan-Einot-Gabriel-Welsch multiple range test; S, superficial tumor.
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
1190
T. Ichihara et al.
In many reports on the clinical features of benign parotid tumors, the tumors are classified into two groups (superficial tumors and deep tumors) according to their location relative to the facial nerve. Since deep tumors are located under the facial nerve, surgical resection is more difficult than for superficial tumors. Indeed, the incidence of postoperative facial nerve paresis, the most critical surgical complication, shows a marked difference between superficial tumors and deep tumors [2–4]. In this study, there was a significant difference in the incidence of postoperative facial paresis between tumor locations (superficial lobe vs deep lobe), but no difference was observed with respect to age, gender, histopathology, and maximum diameter. Bron and O’Brien [2] also investigated the incidence of postoperative facial nerve paresis and concluded that tumor location is the most important factor apart from recurrence and inflammation. Moreover, Harney et al. [14] compared histological findings between pleomorphic adenomas in the lateral lobe and the medial lobe, and reported that superficial tumors tended to have a thinner capsule and a higher rate of extracapsular growth. Therefore, from the viewpoints of surgical difficulty, complications, and tumor characteristics, it is reasonable to classify parotid tumors into two groups, which are superficial tumors and deep tumors. Parotid tumors often occur in the lower pole of the gland, and there were 136 lower pole tumors (32%) among a total of 425 parotid tumors in the present study. We compared lower pole tumors with superficial and deep tumors, identifying significant differences of age, gender, and histopathology that may be explained by the higher prevalence of Warthin’s tumors among lower pole tumors [4,5]. The differences of age and gender were probably observed because Warthin’s tumors are frequently seen in elderly men. Since lower pole tumors have distinct characteristics from superficial tumors and deep tumors, it was considered appropriate to separate lower pole tumors in our classification. The incidence of facial paresis was similar for lower pole tumors and superficial tumors. In surgery for lower pole tumors, we should mainly pay attention to identifying and preserving the marginal mandibular branch of the facial nerve and it is not necessary to perform wide dissection of the nerve. Even if a lower pole tumor is located in the deep region of the gland, surgery is relatively simple compared with resection of a ‘true deep tumor’ arising in the deep region of the upper pole of the parotid gland, which requires identification and preservation of several branches of the facial nerve. Some reports of a low incidence of facial nerve paresis associated with deep tumors may well
have included patients with deep lower pole tumors. When we compared the operating time and blood loss as parameters of surgical invasiveness, there was a significant difference between lower pole tumors and deep tumors, but no difference was observed between lower pole tumors and superficial tumors. Thus, in terms of operative difficulty and complications, lower pole tumors and superficial tumors are comparable, while resection of deep tumors is associated with greater difficulty and a high incidence of postoperative facial nerve paresis. Lower pole tumors can arise superficial to the marginal mandibular branch of the facial nerve (LS) or in the deeper region of the lower pole (LD). To determine whether we should separate these two types, we compared the LS and LD tumor types. As a result, there were no differences of tumor characteristics (age, gender, histopathology, and maximum diameter) between the two types. In addition, there was no difference in surgical difficulty and complications (postoperative facial nerve paresis, operating time, and blood loss) between the two types. Therefore, we consider that there is little clinical value in classifying lower pole tumors according to depth and it appears to be appropriate to combine the LS and LD types in one group. Based on the above results, it seems appropriate to clinically classify benign parotid tumors into three groups, i.e. superficial tumors, deep tumors, and lower pole tumors. Our study on pleomorphic adenomas, which are generally accepted for indication of surgery, yielded nearly the same results. Therefore, it seems appropriate to clinically classify pleomorphic tumors into three groups. Even though we feel it appropriate to classify pleomorphic tumors into three groups, it is useful for us to diagnose them before operation because of operative methods and explanation for the patient. We have recently made a diagnostic criterion of differential diagnosis for superficial and deep tumors by using ultrasonopraphy [15]. The minimum thickness of normal parotid gland tissue between the parotideomasseteric fascia and tumor showed significant differences between superficial and deep tumors. On the other hand, mandibular branch and the location of the tumor were key points in the diagnosis of lower pole tumors. Therefore, it may be possible to diagnose lower pole tumors using CT/MRI. Conclusion We examined the clinical classification of benign parotid tumors according to anatomic location. As a result, we found that lower pole tumors have distinct characteristics (including differences of age, gender,
Acta Otolaryngol Downloaded from informahealthcare.com by Selcuk Universitesi on 01/11/15 For personal use only.
Clinical classification of benign parotid tumors and histopathology) compared with superficial tumors and deep tumors. In addition, we examined postoperative facial nerve paresis, the operating time, and operative blood loss as parameters of surgical difficulty and complications, and found that resection of deep tumors was associated with significantly greater difficulty and a higher incidence of complications compared with resection of superficial tumors and lower pole tumors. When we compared lower pole tumors arising superficial to the marginal mandibular branch of the facial nerve with those arising in the deeper region, no significant differences were observed, so we concluded that both types should be combined in one group as lower pole tumors. Therefore, when benign parotid tumors are classified according to location, it may be appropriate to separate them into three groups: superficial tumors, deep tumors, and lower pole tumors.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Imaizumi A, Kuribayashi A, Okochi K, Ishii J, Sumi Y, Yoshino N, et al. Differentiation between superficial and deep lobe parotid tumors by magnetic resonance imaging: usefulness of the parotid duct criterion. Acta Radiol 2009;50: 806–11. [2] Bron LP, O’Brien CJ. Facial nerve function after parotidectomy. Arch Otolaryngol Head Neck Surg 1997; 123:1091–6. [3] Gaillard C, Perie S, Susini B, St Guily JL. Facial nerve dysfunction after parotidectomy: the role of local factors. Laryngoscope 2005;115:287–91.
1191
[4] Kawata R, Lee K, Araki M, Takenaka H. Safety and usefulness of an electric knife during surgery for parotid benign tumor: postoperative facial paresis and its risk factors. Acta Otolaryngol 2007;127:966–9. [5] Yamashita T, Tomoda K, Kumazawa T. The usefulness of partial parotidectomy for benign parotid gland tumors. A retrospective study of 306 cases. Acta Otolaryngol Suppl 1993;500:113–16. [6] Przewozny T, Stankiewicz C, Narozny W, Kuczkowski J. Warthin’s tumor of the parotid gland. Epidemiological and clinical analysis of 127 cases. Otolaryngol Pol 2004;58:583–92. [7] Witt RL. The significance of the margin in parotid surgery for pleomorphic adenoma. Laryngoscope 2002;112:2141–54. [8] Emodi O, El-Naaj IA, Gordin A, Akrish S, Peled M. Superficial parotidectomy versus retrograde partial superficial parotidectomy in treating benign salivary gland tumor (pleomorphic adenoma). J Oral Maxillofac Surg 2010;68:2092–8. [9] O’Brien CJ. Current management of benign parotid tumors— the role of limited superficial parotidectomy. Head Neck 2003; 25:946–52. [10] Witt RL, Rejto L. Pleomorphic adenoma: extracapsular dissection versus partial superficial parotidectomy with facial nerve dissection. Del Med J 2009;81:119–25. [11] Albergotti WG, Nguyen SA, Zenk J, Gillespie MB. Extracapsular dissection for benign parotid tumors: a metaanalysis. Laryngoscope 2012;122:1954–60. [12] Zbären P, Vander Poorten V, Witt RL, Woolgar JA, Shaha AR, Triantafyllou A, et al. Pleomorphic adenoma of the parotid: formal parotidectomy or limited surgery? Am J Surg 2013;205:109–18. [13] Dell’Aversana Orabona G, Bonavolontà P, Iaconetta G, Forte R, Califano L. Surgical management of benign tumors of the parotid gland: extracapsular dissection versus superficial parotidectomy – our experience in 232 cases. J Oral Maxillofac Surg 2013;71:410–13. [14] Harney MS, Murphy C, Hone S, Toner M, Timon CV. A histological comparison of deep and superficial lobe pleomorphic adenomas of the parotid gland. Head Neck 2003;25: 649–53. [15] Higashino M, Kawata R, Haginomori S, Lee K, Yoshimura K, Nishikawa S. A novel differential diagnostic method for superficial/deep tumor of the parotid gland using ultrasonography. Head Neck 2013;35:1153–7.
