- Email: [email protected]
Involved surgical margins in oral and oropharyngeal carcinoma—an anatomical problem?
Available online at www.sciencedirect.com
British Journal of Oral and Maxillofacial Surgery 49 (2011) 172–175
Involved surgical margins in oral and oropharyngeal carcinoma—an anatomical problem? Jeremy D. McMahon a,∗ , John C. Devine a , Jonathan Hetherington b , Gareth Bryson a , Douglas McLellan a , Colin MacIver c , Evelyn Teasdale a , Ravi Jampana a a b c
Southern General Hospital, 1345 Govan Road, Glasgow G51 4TF, United Kingdom University of Glasgow, United Kingdom Oral and Maxillofacial Surgery, Southern General Hospital, United Kingdom
Accepted 21 February 2010 Available online 31 July 2010
Abstract A previous audit conducted in the West of Scotland (WoS) suggested that anatomical factors accounted for a substantial proportion of invaded surgical margins after resection of an oral or oropharyngeal squamous cell carcinoma (SCC). Since then a number of technical improvements have taken place, the most important of which has been advanced digital imaging that has enabled better surgical planning. In this study we compare the incidence of involved surgical margins in a recent group with those found in the earlier audit. The earlier (WoS) group comprised a consecutive series of patient operated on for a primary SCC of the oral cavity or oropharynx between November 1999 and November 2001 (n = 296). The later series comprised 178 patients operated on for oral or oropharyngeal SCC at the Southern General Hospital (SGH), Glasgow, between 2006 and 2009. A total of 245 patients in the WoS cohort had information available on the invasion of the margins of whom 68 (28%) had an invaded margin. Of 177 patients in the SGH group, 9 (5%) had an invaded margin (p = 0.001). An anatomical approach to the resection of oral and oropharyngeal SCC is appropriate, as it results in a rate of invaded margins of less than 10% irrespective of size and site of the primary lesion. © 2010 Published by Elsevier Ltd on behalf of The British Association of Oral and Maxillofacial Surgeons. Keywords: Oral cancer; Oropharynx; Squamous cell carcinoma; Surgical margins; CT imaging; Head and neck
Introduction When a histological report describes “tumour at a surgical margin” after resection of SCC from the oral cavity or oropharynx, it suggests that the patient’s prognosis will not be particularly good. The probability of local recurrence is increased by a factor of two or more in most reports.1–4 A reduction in disease-specific survival is also reported by some, but not all, investigators.5,6 An audit conducted in the West of Scotland (WoS) indicated that an appreciable proportion of patients had invaded surgical margins.7 Review of publications showed that this problem was not confined to the
∗
Corresponding author. Tel.: +44 0141 201 2842; fax: +44 0141 232 7508. E-mail address: [email protected] (J.D. McMahon).
WoS.1,8 Evaluation of those data suggested that anatomical factors accounted for a large proportion of tumour involved margins. This implied that better planned resection strategies could reduce the occurrence. In our practice an evolution in strategies for resection of primary cancers of the oral cavity and oropharynx began in the early part of this decade, and has continued. The advent of new imaging techniques, particularly those with advanced post-acquisition processing capabilities, has had a major impact. Other technical refinements have also contributed, but we think that improved anatomical understanding based on computed tomographic (CT) imaging has had the greatest influence. In this report we compare the incidence of invaded surgical margins in a recent group of patients (2006–2009) with that in an earlier group (1999–2001).
0266-4356/$ – see front matter © 2010 Published by Elsevier Ltd on behalf of The British Association of Oral and Maxillofacial Surgeons.
doi:10.1016/j.bjoms.2010.02.014
J.D. McMahon et al. / British Journal of Oral and Maxillofacial Surgery 49 (2011) 172–175
173
Table 1 Clinical and histopathological features. WoS = West of Scotland; SGH = Southern General Hospital. Characteristics of tumours
1999–2001 WoS (n = 296)No (%)
2006–2008 SGH (n = 178)No (%)
p value
T1 T2 T3 T4 Not staged Maximum depth (range, mm) Maximum depth (median, mm) Maximum depth (mean, mm) Clear margin Close margin Involved margin Unable to classify margin Perineural invasion present Lymphovascular invasion present
88 (29) 112 (37) 37 (12.5) 57 (19) 2 0.2–32 6 7.7 60 (25) 115 (47) 70 (29%) 51 (17%) 58 (19.5) 45 (15)
72 (41) 44 (25) 6 (3) 50 (29) 4 0.5–34 6.5 7.9 106 (60) 62 (35) 9 (5) 1 (<1) 48 (27) 30 (17)
0.01 0.01 0.0008 0.03 p = 0.38 (Mann-Whitney test) Exact p p = 0.61 <0.001 0.02 0.001 <0.0001 <0.0001 0.14
Table 2 Depth of tumour and number (%) with invaded resection margins. WoS = West of Scotland; SGH = Southern General Hospital. Maximum depth of tumour (mm)
1999–2001 WoS audit No (%) (n = 223)
1999–2001 WoS No (%) with involved margin (n = 113)
2006–2008 SGH No (%) (n = 178)
2006–2008 No (%) with invaded margin (n = 9)
<2 2.1–7 >7
38 (17) 85 (38) 100 (44)
4 (11) 15 (16) 49 (49)
29 (16) 65 (37) 84 (47)
0 0 9 (11)
A total of 250 patients (84%) of the 296 patients in the WoS audit had information about maximum depth available and, of these, 223 had the status of the surgical margin recorded. A further 22 patients with an involved margin did not have data recorded on maximum depth. All the patients in the SGH cohort had data about depth, and all but one about the margin, available.
Methods The Scottish National Audit of Head and Neck Cancer 1999–2001 was funded for 3 years by the Scottish Office Department of Health and has been described in a previous report.7 About a half of all the patients in the audit presented in the West of Scotland region, and about a half of that total was patients with primary oral or oropharyngeal cancer. Histopathological data entered on to the database, including the state of the margins, was quality-assured by checking them against original histological reports for all patients who were operated on as definitive treatment. The second cohort of patients comprised a consecutive series who had had primary resection of squamous cell carcinoma (SCC), of the oral cavity or oropharynx, between January 2006 and January 2009 at the Southern General
Hospital (SGH) in Glasgow by, or under the direct supervision, of two surgeons over the study period. Cases were collected from data held on the West of Scotland Cancer Audit database and checked against the operating theatre log for the study period. The histopathological data were taken from the reports independently of the surgical team and histopathologists. The surgical margin, distance from closest surgical margin (mm), depth of tumour, T stage, presence of lymphovascular invasion, and perineural invasion were recorded so that direct comparisons could be made between groups. An invaded margin was defined as one in which tumour was seen at or within 1 mm of the resection line. A depth of 7 mm was used to separate tumours into those of high and moderate risk, and 2 mm to separate low from moderate risk. A receiver operating characteristic (ROC) curve analysis of the WoS cohort
Table 3 Oral cavity: significant difference in incidence of invaded margins at p = 0.001. WoS = West of Scotland; SGH = Southern General Hospital. Site
1999–2001 WoS audit Total no (n = 224)
1999–2001 WoS audit Data available (n = 177)
Anterior tongue Floor of mouth Gum Buccal Retromolar trigone
73 81 30 12 28
55 66 24 8 24
1999–2001 WoS audit No (%) with invaded margins (n = 43) 7 (13) 19 (29) 7 (23) 2 8 (33)
2006–2008 SGH Total no (n = 152)
2006–2008 SGH No (%) with invaded margins (n = 7)
62 50 26 10 10
1 3 2 1 1
174
J.D. McMahon et al. / British Journal of Oral and Maxillofacial Surgery 49 (2011) 172–175
Table 4 Oropharynx: significant difference in incidence of invaded margins at p = 0.03. WoS = West of Scotland; SGH = Southern General Hospital. Site
1999–2001 WoS audit No operated on (n = 78)
1999–2001 WoS audit Data available (n = 68)
Posterior. pharyngeal wall Oropharynx unspecified Tonsil Base of tongue Soft palate
4 4 32 15 23
3 4 29 (91) 12 (80) 20 (87)
had identified these values as being the most sensitive and specific for separating patients into low, moderate, and high risk for the purpose of calculating disease specific survival.7 The protocol in both groups was to resect the primary tumour with a macroscopic clearance of 1 cm where feasible. In the latter group, increasing use was made of digital imaging to guide deep resection margins. Frozen sections were seldom used. The chi-square test or Fisher’s exact test, as appropriate, was used to assess the significance of differences between the groups. Calculations were aided by the InStat software package (GraphPad, San Diego, CA, USA). Results For the 1999–2001 WoS a total of 296 patients were operated on as their primary treatment. The state of the margins was recorded in 177 (79%) of those having a primary tumour of the oral cavity resected. Of these a total of 43 (24%) were recorded as being invaded. In 34 the deep margin was invaded and in 12 the mucosal margin had tumour present (Tables 1 and 2). Seventy-eight patients had primary oropharyngeal tumours resected. Of these 68 (87%) had information available about invasion of the margins. A total of 25 patients (37%) had involved surgical margins; in 21 the deep margin was affected, in 7 the mucosal margin had tumour present and in 3 had both deep and mucosal margin involvement. The distributions of tumours and involved margins are shown in Tables 3 and 4. Data on maximum depth of tumour was retrieved in 250 patients (84%), and distance from tumour to the nearest surgical margin was available in 223 (89%). Of the 68 patients with an invaded margin, a total of 49 tumours (72%) were more than 7 mm deep. For the SGH 2006–2008 group (n = 178) a total of 9 patients had invaded surgical margins. Distance from the tumour to the nearest margin was available in all but one patient (Table 1). Details of the tumours according to site are shown in Tables 3 and 4. Discussion The variations in incidence of involved margins according to site in the WoS audit had suggested that anatomical rea-
1999–2001 WoS audit No (%) with invaded margins (n = 25) 3 1 13 (45) 2 (16) 6 (30)
SGH 2006–2008 No of patients (n = 26)
SGH 2006–2008 No (%) with invaded margins
0 7 6 3 10
– 1 1 0 0
sons explained why so many patients had involved margins. This report confirms that impression, as the incidence of positive surgical margins was significantly reduced when a more anatomically based approach was used. There is nothing to suggest that the histopathological examination of resection specimens was any less assiduous in the SGH patients as The Royal College of Pathologist Guidelines for reporting were used in both series.9 Indeed, the SGH group had more patients with data on distance of tumour to nearest margin and maximum depth, indicating greater adherence to The Royal College of Pathologists Guidelines than was followed in the earlier WoS report. There was more oropharyngeal primary disease in the WoS (26% compared with 15%). However, the improvement in achieving tumour free surgical margins in the later (SGH) series was found when the oropharynx was considered in isolation. Deeper tumours were more likely to be associated with involved margins than their less invasive counterparts but depth characteristics were no different across the two groups. It seems, therefore, that the difference in rates of invasion of margins is not explained by differences in site or extent of primary disease. In the earlier (WoS) cohort eight surgeons working in five different hospitals did or supervised more than 90% of the resections whereas, in the SGH group, two surgeons in the same hospital were responsible. However, we do not think that this is a sufficient explanation for the difference in rates of invaded margins. Similar rates of involved surgical margins were reported elsewhere in the UK by an experienced surgical team, working in a single centre, with acknowledged expertise in the field.8 Both the surgeons in the SGH series were involved with the earlier audit in its late stages. The approach to reducing deep marginal involvement has come about as a consequence of anatomical approaches to affected sites. One of the surgeons (JCD) recognised the importance of including the medial pterygoid muscle in resections that involved the lateral part of the oropharynx or oropharyngeal isthmus. Any tumour that extended from the tonsillar region into the superior constrictor muscle will bring the deep aspect into close approximation, or involve the loose fibrofatty tissue of the parapharyngeal space. A tumour that arises in the retromolar region and extends medially and deeply in the region of the pterygomandibular raphe will enter this plane. Parapharyngeal fat is seldom more than a few mm wide in the coronal plane. Lateral to it lies the
J.D. McMahon et al. / British Journal of Oral and Maxillofacial Surgery 49 (2011) 172–175
medial pterygoid muscle, and resection of this provides the necessary margin to ensure complete excision. Having recognised, and successfully adopted, this approach it was obvious that tumours that arose elsewhere in the upper aeordigestive tract might be better resected if the local anatomy is carefully considered. A further step was the introduction of digital image storage and retrieval systems with the facility for surgeons to reformat axial images in coronal and sagittal planes. For such planning we make extensive use of computed tomography (CT) with 0.9 mm slices and a dual contrast bolus to maximise tumour enhancement. The typical image file for a tumour of the head and neck contains close to 1000 images. The typical sequence for a magnetic resonance image (MRI) for a similar tumour contains roughly 250 images. CT gives us the opportunity to go through a tumour’s volume, mm × mm, in all three planes, and correlate the results with those found on clinical examination. MRI is used to answer specific questions, and when CT has failed to visualise accurately the tumour seen clinically. Woolgar and Triantafyllou,8 in a report on resection specimens from 301 consecutive resections of tumours from the oral cavity or oropharynx, concluded that both anatomical factors and histological “markers” of tumour characteristics influenced the status of surgical resection margins. This study supports that conclusion. It seems that resection margins free of tumour are achievable in more than 90% of patients if a surgical strategy to resect 10 mm of macroscopically normal tissue around the tumour is followed. However, the followup data are not sufficiently mature to confirm whether or not this approach will be associated with a lower local recurrence rate and better disease specific survival. However, published reports suggest that local recurrences will be reduced compared with those in previous groups of patients treated in the West of Scotland. We agree with Yuen et al.10 that the optimal resection margin should not compromise local control by an inadequate resection, or cause unnecessary functional morbidity from too much resection of uninvolved structures. A subset of tumours will have invaded margins for biological reasons. Many such tumours are identifiable from histopathological markers, as described by Woolgar and co-workers.5,8 Increasing the width of resection beyond 1 cm seems unlikely to modify the subsequent course of the disease, but will assuredly increase functional deficits.
175
The definition of an invaded margin varies among publications.1,3,4,9 Our definition follows the standards and minimum dataset for reporting on head and cancers issued by the Royal College of Pathologists9 and has the advantages of simplicity and objectivity in interpretation; it also recognises that the requirement to see tumour at a surgical margin is not sufficient for the identification of “high risk” cancers. Using these rigorous criteria, the results reported in the group from SGH compare favourably with those published elsewhere.11,12
References 1. Slootweg PJ, Hordijk GJ, Schade Y, van Es RJJ, Koole R. Treatment failure and margin status in head and neck cancer. A critical view on the potential value of molecular pathology. Oral Oncol 2002;38:500–3. 2. Loree TR, Strong EW. Significance of positive margins in oral cavity squamous carcinoma. Am J Surg 1990;160:410–4. 3. Sutton DN, Brown JS, Rogers SN, Vaughan ED, Woolgar JA. The prognostic implications of the surgical margin in oral squamous cell carcinoma. Int J Oral Maxillofac Surg 2003;32:30–4. 4. Binahmed A, Nason RW, Abdoh AA. The clinical significance of the positive surgical margin in oral cancer. Oral Oncol 2007;43:780–4. 5. Brandwein-Gensler M, Teixera MS, Lewis CM, et al. Oral squamous cell carcinoma—histologic risk assessment, but not margin status, is strongly predictive of local disease-free and overall survival. Am J Surg Pathol 2005:167–78. 6. McMahon J, O’Brien CJ, Pathak I, et al. Influence of condition of surgical margins on local recurrence and disease-specific survival in oral and oropharyngeal cancer. Br J Oral Maxillfac Surg 2003;41:224–31. 7. McMahon J, Robertson G, Liew C, et al. Oral and oropharyngeal cancer in the West of Scotland—long-term outcome data of a prospective audit. Br J Oral Maxillofac Surg 2010; doi:10.1016/j.bjoms.2009.12.013. 8. Woolgar JA, Triantafyllou A. A histopathological appraisal of surgical margins in oral and oropharyngeal cancer resection specimens. Oral Oncol 2005;41:1034–43. 9. Standards on minimum datasets for reporting common cancers. Minimum datasets for head and neck histopathology reports.Helliwell T, Woolgar JA, editors. Royal College of Pathologists Guidelines. London: Royal College of Pathologists; 1998. 10. Yuen APW, Lam KY, Wei WI, et al. A comparison of the prognostic significance of tumour diameter, length, width, thickness, area, volume, and clinicopathological features of oral tongue carcinoma. Am J Surg 2000;180:139–43. 11. Sessions DG, Spector GJ, Lenox J, Haughey B, Chao C, Marks J. Analysis of treatment results for oral tongue cancer. Laryngoscope 2002;112:616–25. 12. Sessions DG, Lenox J, Spector GJ, Chao C, Chaudry OA. Analysis of treatment results for base of tongue cancer. Laryngoscope 2003;113:1252–61.
British Journal of Oral and Maxillofacial Surgery 49 (2011) 172–175
Involved surgical margins in oral and oropharyngeal carcinoma—an anatomical problem? Jeremy D. McMahon a,∗ , John C. Devine a , Jonathan Hetherington b , Gareth Bryson a , Douglas McLellan a , Colin MacIver c , Evelyn Teasdale a , Ravi Jampana a a b c
Southern General Hospital, 1345 Govan Road, Glasgow G51 4TF, United Kingdom University of Glasgow, United Kingdom Oral and Maxillofacial Surgery, Southern General Hospital, United Kingdom
Accepted 21 February 2010 Available online 31 July 2010
Abstract A previous audit conducted in the West of Scotland (WoS) suggested that anatomical factors accounted for a substantial proportion of invaded surgical margins after resection of an oral or oropharyngeal squamous cell carcinoma (SCC). Since then a number of technical improvements have taken place, the most important of which has been advanced digital imaging that has enabled better surgical planning. In this study we compare the incidence of involved surgical margins in a recent group with those found in the earlier audit. The earlier (WoS) group comprised a consecutive series of patient operated on for a primary SCC of the oral cavity or oropharynx between November 1999 and November 2001 (n = 296). The later series comprised 178 patients operated on for oral or oropharyngeal SCC at the Southern General Hospital (SGH), Glasgow, between 2006 and 2009. A total of 245 patients in the WoS cohort had information available on the invasion of the margins of whom 68 (28%) had an invaded margin. Of 177 patients in the SGH group, 9 (5%) had an invaded margin (p = 0.001). An anatomical approach to the resection of oral and oropharyngeal SCC is appropriate, as it results in a rate of invaded margins of less than 10% irrespective of size and site of the primary lesion. © 2010 Published by Elsevier Ltd on behalf of The British Association of Oral and Maxillofacial Surgeons. Keywords: Oral cancer; Oropharynx; Squamous cell carcinoma; Surgical margins; CT imaging; Head and neck
Introduction When a histological report describes “tumour at a surgical margin” after resection of SCC from the oral cavity or oropharynx, it suggests that the patient’s prognosis will not be particularly good. The probability of local recurrence is increased by a factor of two or more in most reports.1–4 A reduction in disease-specific survival is also reported by some, but not all, investigators.5,6 An audit conducted in the West of Scotland (WoS) indicated that an appreciable proportion of patients had invaded surgical margins.7 Review of publications showed that this problem was not confined to the
∗
Corresponding author. Tel.: +44 0141 201 2842; fax: +44 0141 232 7508. E-mail address: [email protected] (J.D. McMahon).
WoS.1,8 Evaluation of those data suggested that anatomical factors accounted for a large proportion of tumour involved margins. This implied that better planned resection strategies could reduce the occurrence. In our practice an evolution in strategies for resection of primary cancers of the oral cavity and oropharynx began in the early part of this decade, and has continued. The advent of new imaging techniques, particularly those with advanced post-acquisition processing capabilities, has had a major impact. Other technical refinements have also contributed, but we think that improved anatomical understanding based on computed tomographic (CT) imaging has had the greatest influence. In this report we compare the incidence of invaded surgical margins in a recent group of patients (2006–2009) with that in an earlier group (1999–2001).
0266-4356/$ – see front matter © 2010 Published by Elsevier Ltd on behalf of The British Association of Oral and Maxillofacial Surgeons.
doi:10.1016/j.bjoms.2010.02.014
J.D. McMahon et al. / British Journal of Oral and Maxillofacial Surgery 49 (2011) 172–175
173
Table 1 Clinical and histopathological features. WoS = West of Scotland; SGH = Southern General Hospital. Characteristics of tumours
1999–2001 WoS (n = 296)No (%)
2006–2008 SGH (n = 178)No (%)
p value
T1 T2 T3 T4 Not staged Maximum depth (range, mm) Maximum depth (median, mm) Maximum depth (mean, mm) Clear margin Close margin Involved margin Unable to classify margin Perineural invasion present Lymphovascular invasion present
88 (29) 112 (37) 37 (12.5) 57 (19) 2 0.2–32 6 7.7 60 (25) 115 (47) 70 (29%) 51 (17%) 58 (19.5) 45 (15)
72 (41) 44 (25) 6 (3) 50 (29) 4 0.5–34 6.5 7.9 106 (60) 62 (35) 9 (5) 1 (<1) 48 (27) 30 (17)
0.01 0.01 0.0008 0.03 p = 0.38 (Mann-Whitney test) Exact p p = 0.61 <0.001 0.02 0.001 <0.0001 <0.0001 0.14
Table 2 Depth of tumour and number (%) with invaded resection margins. WoS = West of Scotland; SGH = Southern General Hospital. Maximum depth of tumour (mm)
1999–2001 WoS audit No (%) (n = 223)
1999–2001 WoS No (%) with involved margin (n = 113)
2006–2008 SGH No (%) (n = 178)
2006–2008 No (%) with invaded margin (n = 9)
<2 2.1–7 >7
38 (17) 85 (38) 100 (44)
4 (11) 15 (16) 49 (49)
29 (16) 65 (37) 84 (47)
0 0 9 (11)
A total of 250 patients (84%) of the 296 patients in the WoS audit had information about maximum depth available and, of these, 223 had the status of the surgical margin recorded. A further 22 patients with an involved margin did not have data recorded on maximum depth. All the patients in the SGH cohort had data about depth, and all but one about the margin, available.
Methods The Scottish National Audit of Head and Neck Cancer 1999–2001 was funded for 3 years by the Scottish Office Department of Health and has been described in a previous report.7 About a half of all the patients in the audit presented in the West of Scotland region, and about a half of that total was patients with primary oral or oropharyngeal cancer. Histopathological data entered on to the database, including the state of the margins, was quality-assured by checking them against original histological reports for all patients who were operated on as definitive treatment. The second cohort of patients comprised a consecutive series who had had primary resection of squamous cell carcinoma (SCC), of the oral cavity or oropharynx, between January 2006 and January 2009 at the Southern General
Hospital (SGH) in Glasgow by, or under the direct supervision, of two surgeons over the study period. Cases were collected from data held on the West of Scotland Cancer Audit database and checked against the operating theatre log for the study period. The histopathological data were taken from the reports independently of the surgical team and histopathologists. The surgical margin, distance from closest surgical margin (mm), depth of tumour, T stage, presence of lymphovascular invasion, and perineural invasion were recorded so that direct comparisons could be made between groups. An invaded margin was defined as one in which tumour was seen at or within 1 mm of the resection line. A depth of 7 mm was used to separate tumours into those of high and moderate risk, and 2 mm to separate low from moderate risk. A receiver operating characteristic (ROC) curve analysis of the WoS cohort
Table 3 Oral cavity: significant difference in incidence of invaded margins at p = 0.001. WoS = West of Scotland; SGH = Southern General Hospital. Site
1999–2001 WoS audit Total no (n = 224)
1999–2001 WoS audit Data available (n = 177)
Anterior tongue Floor of mouth Gum Buccal Retromolar trigone
73 81 30 12 28
55 66 24 8 24
1999–2001 WoS audit No (%) with invaded margins (n = 43) 7 (13) 19 (29) 7 (23) 2 8 (33)
2006–2008 SGH Total no (n = 152)
2006–2008 SGH No (%) with invaded margins (n = 7)
62 50 26 10 10
1 3 2 1 1
174
J.D. McMahon et al. / British Journal of Oral and Maxillofacial Surgery 49 (2011) 172–175
Table 4 Oropharynx: significant difference in incidence of invaded margins at p = 0.03. WoS = West of Scotland; SGH = Southern General Hospital. Site
1999–2001 WoS audit No operated on (n = 78)
1999–2001 WoS audit Data available (n = 68)
Posterior. pharyngeal wall Oropharynx unspecified Tonsil Base of tongue Soft palate
4 4 32 15 23
3 4 29 (91) 12 (80) 20 (87)
had identified these values as being the most sensitive and specific for separating patients into low, moderate, and high risk for the purpose of calculating disease specific survival.7 The protocol in both groups was to resect the primary tumour with a macroscopic clearance of 1 cm where feasible. In the latter group, increasing use was made of digital imaging to guide deep resection margins. Frozen sections were seldom used. The chi-square test or Fisher’s exact test, as appropriate, was used to assess the significance of differences between the groups. Calculations were aided by the InStat software package (GraphPad, San Diego, CA, USA). Results For the 1999–2001 WoS a total of 296 patients were operated on as their primary treatment. The state of the margins was recorded in 177 (79%) of those having a primary tumour of the oral cavity resected. Of these a total of 43 (24%) were recorded as being invaded. In 34 the deep margin was invaded and in 12 the mucosal margin had tumour present (Tables 1 and 2). Seventy-eight patients had primary oropharyngeal tumours resected. Of these 68 (87%) had information available about invasion of the margins. A total of 25 patients (37%) had involved surgical margins; in 21 the deep margin was affected, in 7 the mucosal margin had tumour present and in 3 had both deep and mucosal margin involvement. The distributions of tumours and involved margins are shown in Tables 3 and 4. Data on maximum depth of tumour was retrieved in 250 patients (84%), and distance from tumour to the nearest surgical margin was available in 223 (89%). Of the 68 patients with an invaded margin, a total of 49 tumours (72%) were more than 7 mm deep. For the SGH 2006–2008 group (n = 178) a total of 9 patients had invaded surgical margins. Distance from the tumour to the nearest margin was available in all but one patient (Table 1). Details of the tumours according to site are shown in Tables 3 and 4. Discussion The variations in incidence of involved margins according to site in the WoS audit had suggested that anatomical rea-
1999–2001 WoS audit No (%) with invaded margins (n = 25) 3 1 13 (45) 2 (16) 6 (30)
SGH 2006–2008 No of patients (n = 26)
SGH 2006–2008 No (%) with invaded margins
0 7 6 3 10
– 1 1 0 0
sons explained why so many patients had involved margins. This report confirms that impression, as the incidence of positive surgical margins was significantly reduced when a more anatomically based approach was used. There is nothing to suggest that the histopathological examination of resection specimens was any less assiduous in the SGH patients as The Royal College of Pathologist Guidelines for reporting were used in both series.9 Indeed, the SGH group had more patients with data on distance of tumour to nearest margin and maximum depth, indicating greater adherence to The Royal College of Pathologists Guidelines than was followed in the earlier WoS report. There was more oropharyngeal primary disease in the WoS (26% compared with 15%). However, the improvement in achieving tumour free surgical margins in the later (SGH) series was found when the oropharynx was considered in isolation. Deeper tumours were more likely to be associated with involved margins than their less invasive counterparts but depth characteristics were no different across the two groups. It seems, therefore, that the difference in rates of invasion of margins is not explained by differences in site or extent of primary disease. In the earlier (WoS) cohort eight surgeons working in five different hospitals did or supervised more than 90% of the resections whereas, in the SGH group, two surgeons in the same hospital were responsible. However, we do not think that this is a sufficient explanation for the difference in rates of invaded margins. Similar rates of involved surgical margins were reported elsewhere in the UK by an experienced surgical team, working in a single centre, with acknowledged expertise in the field.8 Both the surgeons in the SGH series were involved with the earlier audit in its late stages. The approach to reducing deep marginal involvement has come about as a consequence of anatomical approaches to affected sites. One of the surgeons (JCD) recognised the importance of including the medial pterygoid muscle in resections that involved the lateral part of the oropharynx or oropharyngeal isthmus. Any tumour that extended from the tonsillar region into the superior constrictor muscle will bring the deep aspect into close approximation, or involve the loose fibrofatty tissue of the parapharyngeal space. A tumour that arises in the retromolar region and extends medially and deeply in the region of the pterygomandibular raphe will enter this plane. Parapharyngeal fat is seldom more than a few mm wide in the coronal plane. Lateral to it lies the
J.D. McMahon et al. / British Journal of Oral and Maxillofacial Surgery 49 (2011) 172–175
medial pterygoid muscle, and resection of this provides the necessary margin to ensure complete excision. Having recognised, and successfully adopted, this approach it was obvious that tumours that arose elsewhere in the upper aeordigestive tract might be better resected if the local anatomy is carefully considered. A further step was the introduction of digital image storage and retrieval systems with the facility for surgeons to reformat axial images in coronal and sagittal planes. For such planning we make extensive use of computed tomography (CT) with 0.9 mm slices and a dual contrast bolus to maximise tumour enhancement. The typical image file for a tumour of the head and neck contains close to 1000 images. The typical sequence for a magnetic resonance image (MRI) for a similar tumour contains roughly 250 images. CT gives us the opportunity to go through a tumour’s volume, mm × mm, in all three planes, and correlate the results with those found on clinical examination. MRI is used to answer specific questions, and when CT has failed to visualise accurately the tumour seen clinically. Woolgar and Triantafyllou,8 in a report on resection specimens from 301 consecutive resections of tumours from the oral cavity or oropharynx, concluded that both anatomical factors and histological “markers” of tumour characteristics influenced the status of surgical resection margins. This study supports that conclusion. It seems that resection margins free of tumour are achievable in more than 90% of patients if a surgical strategy to resect 10 mm of macroscopically normal tissue around the tumour is followed. However, the followup data are not sufficiently mature to confirm whether or not this approach will be associated with a lower local recurrence rate and better disease specific survival. However, published reports suggest that local recurrences will be reduced compared with those in previous groups of patients treated in the West of Scotland. We agree with Yuen et al.10 that the optimal resection margin should not compromise local control by an inadequate resection, or cause unnecessary functional morbidity from too much resection of uninvolved structures. A subset of tumours will have invaded margins for biological reasons. Many such tumours are identifiable from histopathological markers, as described by Woolgar and co-workers.5,8 Increasing the width of resection beyond 1 cm seems unlikely to modify the subsequent course of the disease, but will assuredly increase functional deficits.
175
The definition of an invaded margin varies among publications.1,3,4,9 Our definition follows the standards and minimum dataset for reporting on head and cancers issued by the Royal College of Pathologists9 and has the advantages of simplicity and objectivity in interpretation; it also recognises that the requirement to see tumour at a surgical margin is not sufficient for the identification of “high risk” cancers. Using these rigorous criteria, the results reported in the group from SGH compare favourably with those published elsewhere.11,12
References 1. Slootweg PJ, Hordijk GJ, Schade Y, van Es RJJ, Koole R. Treatment failure and margin status in head and neck cancer. A critical view on the potential value of molecular pathology. Oral Oncol 2002;38:500–3. 2. Loree TR, Strong EW. Significance of positive margins in oral cavity squamous carcinoma. Am J Surg 1990;160:410–4. 3. Sutton DN, Brown JS, Rogers SN, Vaughan ED, Woolgar JA. The prognostic implications of the surgical margin in oral squamous cell carcinoma. Int J Oral Maxillofac Surg 2003;32:30–4. 4. Binahmed A, Nason RW, Abdoh AA. The clinical significance of the positive surgical margin in oral cancer. Oral Oncol 2007;43:780–4. 5. Brandwein-Gensler M, Teixera MS, Lewis CM, et al. Oral squamous cell carcinoma—histologic risk assessment, but not margin status, is strongly predictive of local disease-free and overall survival. Am J Surg Pathol 2005:167–78. 6. McMahon J, O’Brien CJ, Pathak I, et al. Influence of condition of surgical margins on local recurrence and disease-specific survival in oral and oropharyngeal cancer. Br J Oral Maxillfac Surg 2003;41:224–31. 7. McMahon J, Robertson G, Liew C, et al. Oral and oropharyngeal cancer in the West of Scotland—long-term outcome data of a prospective audit. Br J Oral Maxillofac Surg 2010; doi:10.1016/j.bjoms.2009.12.013. 8. Woolgar JA, Triantafyllou A. A histopathological appraisal of surgical margins in oral and oropharyngeal cancer resection specimens. Oral Oncol 2005;41:1034–43. 9. Standards on minimum datasets for reporting common cancers. Minimum datasets for head and neck histopathology reports.Helliwell T, Woolgar JA, editors. Royal College of Pathologists Guidelines. London: Royal College of Pathologists; 1998. 10. Yuen APW, Lam KY, Wei WI, et al. A comparison of the prognostic significance of tumour diameter, length, width, thickness, area, volume, and clinicopathological features of oral tongue carcinoma. Am J Surg 2000;180:139–43. 11. Sessions DG, Spector GJ, Lenox J, Haughey B, Chao C, Marks J. Analysis of treatment results for oral tongue cancer. Laryngoscope 2002;112:616–25. 12. Sessions DG, Lenox J, Spector GJ, Chao C, Chaudry OA. Analysis of treatment results for base of tongue cancer. Laryngoscope 2003;113:1252–61.