Ann Surg Oncol (2011) 18:497–504 DOI 10.1245/s10434-010-1294-7
ORIGINAL ARTICLE – HEAD AND NECK ONCOLOGY
Feasibility of Preservation of the Submandibular Gland During Neck Dissection in Patients With Early-Stage Oral Cancer Tseng-Cheng Chen, MD1,2,4, Pei-Jen Lou, MD, PhD1,2, Jenq-Yuh Ko, MD, PhD1,2, Tsung-Lin Yang, MD, PhD1,2,3, Wu-Chia Lo, MD5, Ya-Ling Hu, RN, MSc1,2,6, and Cheng-Ping Wang, MD1,2,3 Department of Otolaryngology, National Taiwan University Hospital, Taipei, Taiwan; 2National Taiwan University College of Medicine, Taipei, Taiwan; 3Institute of Biomedical Engineering, College of Medicine and College of Engineering, National Taiwan University, Taipei, Taiwan; 4Department of Otolaryngology, National Taiwan University Hospital Yun-Lin Branch, Taipei, Yun-Lin, Taiwan; 5Department of Otolaryngology, Far Eastern Memorial Hospital, Taipei, Taiwan; 6School of Nursing, College of Medicine, National Taiwan University, Taipei, Taiwan 1
ABSTRACT Background. The impact of submandibular gland (SMG) preservation during neck dissection on the survival of patients with early-stage oral squamous-cell carcinoma (OSCC) remains undocumented. Methods. The medical records of all patients with earlystage OSCC (stage I and II) who underwent wide excision of the primary tumor and simultaneous neck dissection between 1999 and 2006 at our facility were retrospectively reviewed. Results. We analyzed 408 patients, including 33 patients with and 375 patients without SMG preservation. The 5-year disease-free and overall survival rates were 78.8% and 90.9% for the patients with SMG preservation and 75.4% and 90.4% for the patients without SMG preservation, and these differences were not statistically significant (P = 0.79, P = 0.99, respectively). Similar survival rates between patients with and without SMG preservation were observed in those with oral tongue squamous-cell carcinoma (SCC) and with buccal SCC. Patients with T2 OSCC with SMG preservation had significantly lower 5-year disease-free survival rate than those without SMG preservation (P = 0.02), but overall survival rates were similar between these two groups. Conclusions. Preservation of the SMG during neck dissection may be oncologically safe in patients with T1 OSCC, but the feasibility of SMG preservation seems less clear for T2 OSCC. Ó Society of Surgical Oncology 2010 First Received: 20 April 2010; Published Online: 14 September 2010 C.-P. Wang, MD e-mail:
[email protected]
The status of cervical lymph node metastasis is one of the most important prognostic factors in the treatment of the patients with oral squamous-cell carcinoma (OSCC).1 Even single lymph node metastasis may diminish their survival chances.2,3 Thus, it is important to identify and eradicate possible cervical lymph node metastases when treating OSCC. Radical neck dissection is an effective surgical procedure for treating metastatic nodal disease in the neck, but it is associated with high surgical morbidity. Selective neck dissection (SND) from levels I to III/IV is an effective diagnostic procedure for clinical N0 OSCC and is a potential treatment procedure for clinical N1–N2a, with lower surgical morbidity rates as a result of the preservation of vital neck structures.4 However, SND is still not free of sequelae. SND for OSCC usually involves the removal of the submandibular gland (SMG) because it is located in the level Ib region of the neck and is close to primary oral tumors. Because the SMGs are responsible for approximately 70% to 90% of unstimulated salivary volume, especially at night, removal of one SMG as part of the neck dissection causes greatly decreased unstimulated salivary flow and an increased incidence of subjective xerostomia, even though the patient does not receive postoperative radiotherapy.5–7 Therefore, if the SMG can be preserved during neck dissection for OSCC, xerostomia might be prevented, especially when postoperative radiotherapy is not instituted. According to our previous study and the work of others, true infiltration of the SMG by OSCC is quite uncommon, especially in the early stage, indicating the SMG might be uncontaminated and thus be considered for preservation during surgical treatment of OSCC.8–10 However, there are
498
few studies focusing on the feasibility and oncological safety of removing the lymph nodes around the SMG and preserving the gland in patients with early-stage OSCC. In this article, we have retrospectively reviewed our OSCC patients and report on the feasibility of SMG preservation during neck dissection in early-stage OSCC patients by comparing the survival rates between the patients who underwent SMG-preserving neck dissection and those who underwent SMG-removing neck dissection. PATIENTS AND METHODS We retrospectively reviewed the medical and pathological records of all patients with newly diagnosed stage I and stage II OSCC, who underwent curative wide excision of the primary tumor and simultaneous neck dissection at the Department of Otolaryngology of the National Taiwan University Hospital between January 1999 and December 2006. The exclusion criteria included patients with previously treated OSCC, primary head and neck cancer other than OSCC, other malignancies outside the head and neck regions, the presence of a simultaneous second primary cancer, and previous radiotherapy involving the head and neck regions to treat other diseases. The patients were divided into two groups according to whether or not the SMG had been preserved. The decision of whether to preserve the SMG depended on the primary location and the size of OSCC, the preoperative neck status, the intraoperative findings regarding gross cancer infiltration of the SMG, the concerns of xerostomia, and cosmetics. When the patients were concerned about xerostomia and deformation of the neck postoperatively, and when the tumor was small and far away from the SMG, such as small buccal or lateral tongue cancer, with no grossly important lymph nodes, which were removed with preservation of the SMG, preservation of the SMG during neck dissection was considered for the patient. The tumor, node, metastasis system status of each tumor was classified according to the 2002 criteria of the American Joint Committee on Cancer.11 All statistical analyses were performed by SPSS software, version 15.0 (SPSS, Chicago, IL). The Fisher exact, v2, and t-tests were used to determine differences between clinical parameters (i.e., sex, age, T and N classification, extent of neck dissection, pathological differentiation of the tumor, extracapsular spread, perineural invasion, lymphovascular invasion, margin status, and postoperative radiotherapy) for patients with and without preservation of the SMG, as appropriate. The starting point of the follow-up period was defined as the time since the completion of the initial treatment for each patient. The end point of the follow-up period was defined as the time when the patient died or June 2009.
T.-C. Chen et al.
The rates of disease-free survival and overall survival were calculated by the Kaplan-Meier product limit method. All sites of persistent, residual, or recurrent tumors were recorded as failures in terms of the disease-free survival metric, and all deaths were recorded against the overall survival parameter. Significance levels among curves were determined by the log rank test. Potential risk factors and pathological characteristics were further analyzed by a multivariate Cox regression model. Corresponding P values \ 0.05 were interpreted as being statistically significant. RESULTS A total of 408 eligible patients with T1–2N0 OSCC were enrolled onto this study, including 348 male and 60 female patients. Their ages ranged from 22 to 90 years, with a mean of 51 years. All patients underwent wide excision of the primary OSCC with a safe margin of 1.5 to 2 cm and simultaneous neck treatment by means of modified radical neck dissection or SND. Thirty-three patients were recorded with preservation of the SMG during neck dissection. The other 375 patients underwent neck dissection without preservation of the SMG. All clinical characteristics and potential risk factors for OSCC between the groups with and without preservation of the SMG are listed in Table 1. The data indicate no statistical differences between these two groups, except for the fact that a higher percentage of female patients received SMG-preserving neck dissection (P = 0.017). The follow-up period for these 408 patients was from 1 to 115 months, with a mean of 51 months. Among the 33 patients who underwent SMG-preserving neck dissection, 4 experienced local recurrence, and 3 experienced regional recurrence. Of the 3 patients with regional recurrence, 2 experienced recurrence in the ipsilateral neck and 1 in the level Ia region. Of the 7 patients with recurrent disease with preservation of the SMG, 3 patients died of OSCC. The clinical characteristics of all patients with recurrent disease in the SMG-preserving group are listed in Table 2. The 5-year disease-free survival rates of the SMG-preserving and SMG-removing groups were 78.8% and 75.4%, respectively (P = 0.79, Fig. 1a), and the 5-year overall survival rates of these two groups were 90.9% and 90.4%, respectively (P = 0.99, Fig. 1b). We found no statistical differences between the survival rates of these two groups with and without SMG preservation. The multivariate analyses of potential risk factors by the Cox regression model suggested that male sex, the presence of perineural invasion, and close/positive section margin status were adverse factors for disease-free survival, and close/positive section margin
Feasible to Preserve Submandibular Gland TABLE 1 Characteristics of patients with and without preservation of the submandibular gland
499
Characteristics
With preservation (n = 33)
Without preservation (n = 375)
Female sex
10/33 (30.3%)
50/375 (13.33%)
0.017
Age (years)
50.9 ± 12.36
52.11 ± 11.40
0.56
Tongue
20/33 (60.6%)
188/375 (50.13%)
Buccal
8/33 (24.24%)
129/375 (34.4%)
Mouth floor
0/33 (0%)
16/375 (4.27%)
Gingiva
2/33 (6.06%)
22/375 (5.87%)
Retromolar trigone
0/33 (0%)
5/375 (1.33%)
Hard palate
1/33 (3.03%)
10/375 (2.67%)
Lip
2/33 (6.06%)
5/375 (1.33%)
Tl
23/33 (69.70%)
236/375 (62.9%)
T2
10/33 (30.30%)
139/375 (37.1%)
SOHND MRND
30/33 (90.9%) 3/33 (9.09%)
283/375 (75.47%) 79/375 (21.06%)
Bilateral SOHND
0/33
13/375 (3.47%)
28/33 (84.85%)
324/375 (86.4%)
Primary tumor location
0.27
Clinical T classification
0.46
Neck dissection
0.18
Tumor differentiation Well
MRND modified radical neck dissection, SOHND supraomohyoid neck dissection Age: mean ± SD v2 test is for ‘‘female sex’’, primary tumor location, clinical T classification Test for recurrence T1 is Fisher’s exact test Test for age is student’s t test Others: Fisher’s exact test
P value
0.73
Moderate
4/33 (12.12%)
42/375 (11.2%)
Poor
1/33 (3.03%)
9/375 (2.4%)
Extracapsular spread
1/33 (3.03%)
9/375 (2.4%)
0.57
Perineural invasion
6/33 (18.18%)
46/375 (12.26%)
0.41
Lymphovascular
1/33 (3.03%)
27/375 (7.2%)
Margin status
0.72 0.16
Involved
1/33 (3.03%)
14/375 (3.73%)
Close
6/33 (18.18%)
30/375 (8%)
4/33 (12.12%)
39/375 (10.4%)
0.77
Tl
2/23 (8.70%)
56/236 (23.73%)
0.12
T2
5/10 (50%)
31/139 (22.30%)
0.06
Postoperative radiotherapy Recurrence
status was the only adverse factor for overall survival. Preservation of the SMG during the neck dissection did not affect either disease-free or overall survival in the early-stage OSCC patients (P = 0.98, and P = 0.88, respectively; Table 3). Oral tongue and buccal cancers accounted for 85% of all OSCCs. The 5-year disease-free survival rates of the patients with oral tongue squamous-cell carcinoma (SCC) with and without preservation of the SMG were 80.0% and 80.6%, respectively (P = 0.99, Fig. 2a); the 5-year overall survival rates of these two groups were 90.0% and 87.8%, respectively (P = 0.95, Fig. 2b). Among patients with buccal SCC, the 5-year disease-free survival rates of those with and without preservation of the SMG were 75.0% and 69.3%, respectively (P = 0.83, Fig. 2c); the 5-year overall survival rates of these two groups were 87.5% and 95.6%,
respectively (P = 0.54, Fig. 2d). There were no statistical differences in survival between these two groups in both oral tongue and buccal SCC patients. We also considered the T classification of OSCC in these patients. The 5-year disease-free survival rates of the T1 OSCC patients with and without preservation of the SMG were 91.3% and 75.1%, respectively (P = 0.10, Fig. 3a); the 5-year overall survival rates of these two groups of T1 OSCC patients were 95.7% and 91.6%, respectively (P = 0.41, Fig. 3b). For T2 OSCC patients, the 5-year disease-free survival rates for those with and without preservation of the SMG were 50.0% and 75.8%, respectively, a statistically significant difference (P = 0.02, Fig. 3c). The 5-year overall survival rates of these two T2 OSCC groups were 80.0% and 88.2%, respectively (P = 0.22, Fig. 3d).
SND selective neck dissection, PNI perineural invasion, OP operation, CCRT concurrent chemoradiotherapy, DOD dead of disease, MRND modified radical neck dissection, AW alive and well
86 AW OP Regional, ipsilateral, level II 21 No PNI T2N0 SND T2N0 39/M 7
Tongue
50
22 AW OP
OP ? CCRT AW 5
6 Regional, ipsilateral, level lb
Regional level la No Margin close
PNI, margin close Yes T2N0
T2N0 SND
SND T2N0
T2N0
47/M 6
Tongue
48/M 5
Buccal
3
9 OP ? CCRT DOD 2 Primary No Margin involved T2N0 SND T2N0 56/F 4
Tongue
20 AW
DOD –
OP 7
2 Primary
Primary No
No Margin involved
Margin close T1N0
T2N0 MRND
SND T1N0
T2N0 56/M 3
Buccal
44/M 2
Lower lip
OP ? CCRT DOD 3 Primary PNI, margin close Yes T1N0 SND T1N0 Tongue 66/F 1
Time to Salvage recurrence treatment (mo) Postoperative Recurrent site radiotherapy Patient Age/sex Primary tumor Clinical Neck Pathologic Other risk No. location stage management stage factors
TABLE 2 Characteristics of patients with recurrent disease in the group of preservation of submandibular gland
8
T.-C. Chen et al. Outcome Length of follow-up (mo)
500
DISCUSSION Because OSCC is associated with an increased risk of cervical lymph node metastasis, neck dissection is a routine surgical procedure that accompanies wide excision of the primary tumor, even for the patients with early-stage OSCC. With recent advances in our understanding of the pattern of neck metastasis and improvements in surgical techniques, neck dissection procedures have evolved to limit to the neck region with the risk of lymph node metastasis and to preserve nonlymphatic structures to the greatest extent as possible, such as the spinal accessory nerve, internal jugular vein, and sternocleidomastoid muscle, to reduce the risk of postoperative sequelae. SND for OSCC is an effective procedure to collect or remove the lymphatic structure in levels I to III/IV of the neck for accurate N classification. Despite lower morbidity, conventional SND of levels I to III/IV still removes a nonlymphatic organ, namely the SMG, which has many important physiological functions. First, saliva produced by the SMG is not only important for lubrication of the oral cavity but is also vital for antimicrobial activity in the mouth, maintenance of oral mucosal immunity, and preparation of the bolus of food during mastication. Jacob et al. found that a third of patients with early-stage OSCC who underwent neck dissection with concomitant removal of the SMG experienced xerostomia with an increased incidence of dental caries compared with control groups.6 Although pharmaceuticals have recently been developed that may maximize the functional capacity of the remaining gland, most of these have failed to greatly improve the postoperative quality of life and have resulted in troubling adverse effects.12 To date, no effective treatment has been developed to fully restore saliva flow once the SMG has been sacrificed.13,14 Jaguar et al. reported that the compensatory salivary mechanism of the residual SMG seems not to be satisfactory.15 Preserving the cancer-free SMG during neck dissection seems to be the best way to prevent any associated complications. Second, unilateral SMG excision can lead to the asymmetric depression of the ipsilateral neck below the mandible, although this disfiguration has no adverse physiological impact on the patients. This may explain, at least in part, why the rate of SMG-preserving procedure was higher in female patients in this series. Third, despite attempts to avoid it, the removal of a SMG still risks injuring the hypoglossal (2.9%) and lingual nerves (1.4%).16 For early-stage OSCC patients, injury of the hypoglossal or lingual nerve could result in swallowing dysfunction, dysgeusia, and tongue paraesthesia to some degree. If the SMG is infiltrated by the OSCC, either from the oral tumor directly or from the metastatic lymph nodes in or around the gland, the SMG must be removed during
Feasible to Preserve Submandibular Gland
501
a
b
FIG. 1 Disease-free survival curves (a) and overall survival curves (b) between the patients with and without preservation of the submandibular gland
TABLE 3 Multivariate analyses of risk factors regarding diseasefree and overall survival by the Cox regression model
Characteristics
Disease-free survival 5-year HR survival (%) (95% CI)
Overall survival P value 5-year HR survival (%) (95% CI)
P value
Gender Male
73.60
Female
88.10
3.17 (1.36–7.37) 0.007
90.70 88.90
1.52 (0.52–4.43)
0.45
1.10 (0.33–3.61)
0.88
Preservation of submandibular gland With
78.80
Without
75.40
1.00 (0.46–2.19) 0.98
90.90 90.40
Perineural invasion Positive
58.10
Negative
78.10
3.21 (1.70–6.06) \0.001
83.20 91.60
2.28 (0.83–6028) 0.11
Lymphovascular invasion Positive
76.80
Negative
75.70
0.49 (0.19–1.22) 0.12
76.80 91.40
1.26 (0.43–3.75)
0.67
Margin involved Positive
42.40
2.89 (1.04–8.07) 0.04
69.60
2.76(0.67–11.28) 0.16
Close
55.70
2.72 (1.49–4.98) 0.001
76.30
3.19 (1.26–8.08)
0.01
Negative
78.50
1.46 (0.26–8.23)
0.67
0.94 (0.32–2.79)
0.91
91.60
Extracapsular spread Positive
57.90
Negative
76.20
1.13 (0.28–4.60) 0.87
88.90 90.50
Postoperative radiotherapy HR hazard ratio; 95% CI 95% confidence interval
With
69
Without
76.40
neck dissection even though the removal may lead to many sequelae, as mentioned above. However, as shown in our previous study and in the studies of others, the incidence of
0.48 (0.22–1.08) 0.08
85.90 91.00
OSCC metastasis to the SMG is extremely low, even at the late stage.8–10 If the OSCC metastasizes to the SMG, the primary tumor is very close to the gland such as SCC in the
502
T.-C. Chen et al.
a
b
c
d
FIG. 2 Disease-free survival curves (a) and overall survival curves (b) between the tongue cancer patients with and without preservation of the submandibular gland (SMG). Disease-free survival curves
(c) and overall survival curves (d) between the buccal cancer patients with and without preservation of the SMG
floor of mouth, and it may be very locally advanced or have marked nodal metastases, especially in level I of the neck.8–10 Therefore, oncologically safe removal of the lymph nodes in the submandibular region with preservation of the SMG may be possible if no events as mentioned above are present. Several years ago, we began to perform lymph node dissection along the capsule of the SMG, leaving the SMG in situ for some patients with stage I or II OSCC if the primary tumor was far from the SMG, and if salivary function and cosmetics were concerned. The lymphoareolar tissue in front of the gland and medial to the mylohyoid muscle was also excised with care. By means of this procedure, four of the six lymph node groups in the submandibular region, preglandular, postglandular, prevascular, and postvascular groups, which are the more
consistent and draining nodes of OSCC, are easily dissected from the gland. According to our retrospective survival analyses, the statistically significant adverse prognostic factors of stage I and II OSCC in this series were male sex, the presence of perineural invasion, and close/positive section margin status for disease-free survival; and close/positive section margin for overall survival. Preservation of the SMG or not during the neck dissection did not affect disease-free or overall survival of the early-stage OSCC patients by either univariate or multivariate analyses. Furthermore, there were no statistically significant differences of the survivals between these two groups in both patients with oral tongue and buccal SCC, which constituted most of the OSCC patients in this series. The patients with OSCC over the
Feasible to Preserve Submandibular Gland
503
a
b
c
d
FIG. 3 Disease-free survival curves (a) and overall survival curves (b) between T1 patients with and without preservation of the submandibular gland (SMG). Disease-free survival curves (c) and overall survival curves (d) between T2 patients with and without preservation of the SMG
floor of the mouth, which is closest to the SMG and exhibits the highest probability of SMG metastasis, all underwent SMG-removing neck dissection. Therefore, this study cannot provide any insights into the feasibility of SMG-preserving neck dissection in mouth-floor SCC patients. Although the largest diameter of a T2 OSCC approached 4 cm according to the American Joint Committee on Cancer definition, previous studies failed to find any SMG invasion in patients with stage II OSCC.8–10 However, locoregional recurrence rates were increased in T2 OSCC patients who underwent the SMG-preserving procedure, which resulted in statistically lower disease-free survival rate for the SMG-preserving group than for the SMGremoving group. Analyses of the recurrence pattern in the
SMG-preserving group showed that 3 T2 OSCC patients exhibited regional recurrence. Of these three individuals, one had ipsilateral level II, another had ipsilateral level Ib, and the other had level Ia regional recurrence, which may be related to the SMG-preserving procedure. All of these three patients received salvage radical neck dissection and were again free of disease. Despite this, SMG-preserving neck dissection for T2 OSCC remains questionable, and further study is required to verify its feasibility because of the statistical limitation that there were only 10 T2 OSCC patients with preservation of the SMG, compared to 139 patients without preservation of the SMG. No patient with T1 OSCC who underwent the SMG-preserving procedure experienced regional recurrence, either in the ipsilateral or the contralateral neck, except for two patients who
504
T.-C. Chen et al.
experienced local primary tumor recurrence. This may be related to risk factors other than the preservation of the SMG. We note several limitations to this study. First, this study is a retrospective review. Because preservation of the SMG during neck dissection is not the standard of care for OSCC, unavoidable or unnoticed selection bias must be present in this study despite the distribution of parameters between these two groups was equal, except for the higher percentage of female patients in the SMG-preserving group. Second, the number of patients in the SMG-preserving group was small. Consequently, the effects of the primary tumor factors and other possible adverse pathologic factors on the feasibility of the SMG-preserving neck dissection were not evaluated with adequate statistical power. Therefore, well-designed randomized prospective studies with more patients are necessary to clarify the feasibility of SMG-preserving neck dissection in earlystage OSCC patients. This retrospective study suggests that SMG-preserving neck dissection may be feasible in patients with T1 OSCC, especially oral tongue or buccal SCC, though its feasibility remains questionable in patients with T2 OSCC. ACKNOWLEDGMENT This work was supported in part by the National Science Council of the Republic of China (NSC 98-2314-B002-046-MY3) and in part by the National Taiwan University Hospital (NTUH.98-M1238). We thank the staff of the Eighth Core Lab, Department of Medical Research, National Taiwan University Hospital for technical support during the study. CONFLICTS OF INTEREST
None.
REFERENCES 1. Shingaki S, Takada M, Sasai K, et al. Impact of lymph node metastasis on the pattern of failure and survival in oral carcinomas. Am J Surg. 2003;185:278–84. 2. Shah JP. Patterns of cervical lymph node metastasis from squamous carcinomas of the upper aerodigestive tract. Am J Surg. 1990;160:405–9.
3. Chen TC, Wang CT, Ko JY, et al. Postoperative radiotherapy for primary early oral tongue cancer with pathologic N1 neck. Head Neck. 2010;32:555–61. 4. Iype EM, Sebastian P, Mathew A, et al. The role of selective neck dissection (I–III) in the treatment of node negative (N0) neck in oral cancer. Oral Oncol. 2008;44:1134–8. 5. Cunning DM, Lipke N, Wax MK. Significance of unilateral submandibular gland excision on salivary flow in noncancer patients. Laryngoscope. 1998;108:812–5. 6. Jacob RF, Weber RS, King GE. Whole salivary flow rates following submandibular gland resection. Head Neck. 1996;18: 242–7. 7. Saarilahti K, Kouri M, Collan J, et al. Sparing of the submandibular glands by intensity modulated radiotherapy in the treatment of head and neck cancer. Radiother Oncol. 2006;78: 270–5. 8. Chen TC, Lo WC, Ko JY, et al. Rare involvement of submandibular gland by oral squamous cell carcinoma. Head Neck. 2009;31:877–81. 9. Byeon HK, Lim YC, Koo BS, Choi EC. Metastasis to the submandibular gland in oral cavity squamous cell carcinomas: pathologic analysis. Acta Otolaryngol. 2009;129:96–100. 10. Razfar A, Walvekar RR, Melkane A, Johnson JT, Myers EN. Incidence and patterns of regional metastasis in early oral squamous cell cancers: feasibility of submandibular gland preservation. Head Neck. 2009;31:1619–23. 11. American Joint Committee on Cancer (AJCC). Manual for Staging of Cancer. New York: Springer-Verlag; 2002. 12. Fox PC, van der Ven PF, Sonies BC, Weiffenbach JM, Baum BJ. Xerostomia: evaluation of a symptom with increasing significance. J Am Dent Assoc. 1985;110:519–25. 13. Murdoch-Kinch CA, Kim HM, Vineberg KA, Ship JA, Eisbruch A. Dose-effect relationships for the submandibular salivary glands and implications for their sparing by intensity modulated radiotherapy. Int J Radiat Oncol Biol Phys. 2008;72:373–82. 14. Jha N, Seikaly H, Harris J, et al. Phase III randomized study: oral pilocarpine versus submandibular salivary gland transfer protocol for the management of radiation-induced xerostomia. Head Neck. 2009;31:234–43. 15. Jaguar GC, Lima EN, Kowalski LP, et al. Impact of submandibular gland excision on salivary gland function in head and neck cancer patients. Oral Oncol. 2010;46:349–54. 16. Berini-Aytes L, Gay-Escoda C. Morbidity associated with removal of the submandibular gland. J Craniomaxillofac Surg. 1992;20:216–9.
