Do basal cell carcinomas recur after complete conventional surgical excision?

Transcription

1 British Journal of Plastic Surgery (2005) 58, Do basal cell carcinomas recur after complete conventional surgical excision? R.W. Griffiths*, S.K. Suvarna, J. Stone Northern General Hospital, Sheffield Teaching Hospitals NHS Trust, Sheffield, UK Received 28 October 2004; accepted 9 February 2005 KEYWORDS Basal cell carcinoma; Complete conventional excision; Recurrence rate Summary For 1378 patients treated in the 11 years by conventional excision of 1635 basal cell carcinomas, 1516 first index lesions were histologically completely excised. All patients having more than one BCC excised were identified from the data base from 1988 to 2003 to give minimum 5 years follow for last treated primary lesions in Measured clearance margins around the initial lesions at or near sites of presumptive recurrent lesions were noted and the lesions recorded photographically. All incompletely excised lesions whether or not re-excised were excluded. The median age for all patients was 70 years. Over minimum 5 years follow up, six patients developed nine subsequent lesions contiguous with the scar or graft repair of primary index lesion excision site (probable recurrences). The median interval to recurrence was 41 months (4 months 8 years 10 months), with median lateral clearance margin around the primary tumour of 2 mm ( mm). A further nine patients developed 11 new lesions near (within 1 cm of) the scar or graft of primary index lesion excision site (possible recurrences). The median interval to recurrence was 59 months (1 year 8 years 6 months). The median lateral clearance margin around the primary tumour was 4.1 mm ( mm). For the two groups combined the maximum recurrence rate expressed as a percentage of index lesions was 1.3% (20/1516). Two thirds of possible and probable recurrences occurred in the temple and forehead, although these sites represented only 22% of all lesions, which may rather suggest new lesions in an area of field change as opposed to residual disease. The measured clearance margins reported here perhaps suggest that some original lesions may well have been completely excised primarily and many recurrences were new primaries. These figures indicate there is a low order of probability for the incidence of recurrent basal cell carcinoma during minimum 5 years follow period after conventional surgical excision and conventional histological assessment of tumour resection margins. q 2005 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All rights reserved. * Corresponding author. Address: Department of Plastic Surgery, Northern General Hospital, Sheffield Teaching Hospitals NHS Trust, Herries Road, Sheffield S5 7AU, UK. Tel: C address: richard.griffiths@sth.nhs.uk (R.W. Griffiths). S /$ - see front matter q 2005 The British Association of Plastic Surgeons. Published by Elsevier Ltd. All rights reserved. doi: /j.bjps

2 796 Surgical excision may be regarded as optimal treatment of basal cell carcinomas. Complete excision should cure the condition with results for conventional complete surgical excision, in over 3500 cases giving local recurrence rates over varying periods of follow up observation in the range %. 1 5 Higher 5 years recurrence rates of 14 and 23% for 239 cases have been reported 6,7 although criteria for complete excision are not uniform, and the validity of some of these data has given rise to vigorous debate None of these studies, 1 7 however, documented measured lateral and deep clearance margins around the initial primary tumours, or illustrated presumptive recurrences with photographs. Measured clearance margins and photographic comparison of new lesions with initial lesions will assist in judging if a new lesion is likely to be a recurrence, or a new primary adjacent to a site of previous surgical scarring. Minimum 5 years follow up should detect 82% of recurrences, 12 but a further 18% of recurrences may occur 6 10 years post treatment. Because not all patients will live for 5 years after initial treatment, results can be expressed both in terms of all patients initially treated and also in terms of the observed 5 years survivors (with or without recurrence) the so called determinate patients. 13 Although advocates of Mohs micrographic surgery have suggested that recurrence rates after conventional surgical excision are higher than after Mohs surgery, Mohs surgery itself is associated with 5 years recurrence rates of % for 2960 BCCs and for 7257 BCCs. 13,14 The only 5 years UK figures show for 141 lesions, recurrences for 1.7% primary excisions and a recurrence rate of 4.8% for recurrent (previously treated) lesions. 15 It would, therefore, be valuable to carefully define the incidence of local basal cell carcinoma recurrence after conventional surgical excision. Published reviews of collected series rarely record if tumours which recur over 5 years follow up after excision by conventional surgery, were in fact initially completely excised histologically, 16,17 making it difficult to know the correct rate of local tumour recurrence after defined complete tumour excision. The importance of differentiating true recurrences from new lesions, which may arise at, or adjacent to sites of previous treatment has been emphasised. 18,19 Local recurrence is defined as a lesion arising in the same or contiguous site, 20 and Goldberg et al. 21 distinguished continued growth of a tumour left in the immediate area becoming clinically manifest months or years after incomplete removal, from new tumour formation in the immediate area of the previous tumour site which cannot be differentiated from the old lesion because of proximity of location to the scar. Clinical discontinuity between the new lesion and the previous scar or graft appears to be crucial to the differentiation. However, in publications to date lack of sequential photographic documentation, and absence of measured excision margins around primary index tumours has made these differentiations difficult in practice. A previous preliminary report 22 indicated probable or possible basal cell carcinoma recurrences were infrequent after conventionally assessed histologically complete excision. For an expanded period of 11 years of patient accrual we have analysed the case notes and photographic records of consecutive patients with basal cell carcinomas treated under one surgeon s care from 1988 to 1998 allowing a potential minimum 5 years follow up to the end of Specifically, deep and lateral histological clearance margins were measured for all primary lesions, at which excision sites subsequent metachronous lesions arose contiguous with or closely adjacent to (within 1 cm) scar or graft (probable or possible local recurrences). Material and methods R.W. Griffiths et al. The data base of consecutive patients having basal cell carcinomas excised under the care of one consultant for the 11 years from 1988 to 1998 was studied. The total number of patients and total number of lesions were noted. All first lesions treated in the period were designated index lesions. Second lesions identified either in the period or subsequent 5 years were studied to determine if they could reasonably be considered from their position, to represent possible or probable recurrences. All patients undergoing more than one episode of care for basal cell carcinoma, in the 16 years from 1988 to the end of 2003 were identified in order to study possible and probable tumour recurrences after complete primary lesion excision. This timescale provided a potential minimum period of 5 years to elapse after the last treated patient received their primary surgery in All patients with more than one treatment episode had their discharge summaries, notes and photographs studied in order to define those in which any subsequent treatment episode after the primary episode involved treating a new basal cell carcinoma which was contiguous with, or closely adjacent to (within 1 cm of) the previous scar or skin graft or skin flap All histological material for these patients with putative

3 Do basal cell carcinomas recur after complete conventional surgical excision? 797 recurrences was reviewed to confirm the lateral and deep clearance margins of normal tissue. Since, there was no Mohs surgery service available locally during the study period, it was assumed that patients developing further basal cell carcinomas (metachronous lesions or recurrences) after primary surgery, would be referred back to the plastic surgeon who had performed the initial surgery. 5 This assumption seemed justified because of close links with local dermatology and general practitioner colleagues both informal and formal through local multi-disciplinary team (MDT) structures and regular Skin Cancer Network Group meetings. However, the possibility that not all recurrences returned to the plastic surgery unit does exist. The referral pattern The patients were mostly tertiary referrals from dermatologist or oncologist colleagues, with less than 10% of patients being referred from general practice. Therefore, the patient group was selective for either large or awkwardly placed lesions in areas which could require skin graft or flap repair, and this gives bias to this patient population away from comparatively simple, easy to treat lesions. The histological technique The major changes that we have seen over the last 15 years include more detailed examination of the skin resection specimens. In this manner, we have now introduced inking of particular margins of specimens with careful block sampling according to the orientation of the specimen, usually in a breadslice fashion. This has taken the place of the hot cross bun and single transverse slice sections that used to be the norm. In addition, all of the resection sample is now likely to be processed compared with previously. On infrequent occasions, frozen section sampling can be taken for awkward sites such as around face/nose/eyes. Cross comparison against previous biopsies is also regularly undertaken with review of previous biopsies from other sites and review in multi-disciplinary team (MDT) format. Proforma reporting styles according to the Royal College of Pathologists currently are in place. 23 The surgical practice The surgical practice in this period was to perform as many procedures as possible as day cases under local anaesthetic. Lesions were photographed and loupe magnification (!3.5) was routinely used The excision margin was marked around the assessed dotted outline of the tumour before the infiltration of 0.5% lignocaine/1: adrenaline. Margins of 2 3 mm were marked around tumours with a distinct margin, with 5 mm used around less distinct bordered lesions. A marker suture was placed at the 12 O clock position on the specimen to assist in orientation during histopathological assessment. If there was concern about completeness of excision, peripheral and/or deep biopsies beyond the main resection specimen were taken and either submitted to frozen section or paraffin section analysis. The policy was to reexcise the site of any lesion reported as incompletely excised. Only if the patient declined would irradiation or long term observation be considered. Determinate patients Determinate patients (those surviving 5 years or more after surgery or dying within 5 years of BCC) are not likely to be less than 60% of the total (we have previously shown for squamous cell carcinoma (SCC) that 40% will died within 5 years of unrelated disease). 27 Thus results are expressed in terms of the whole cohort and then the likely number of determinate patients, (assuming that for BCC patients, at least 60% would be determinate as they tend to be younger than patients with SCC). Consent Written permission to use clinical photographs was obtained from all patients, and in one instance from the next of kin for a patient who died of unrelated disease in Results In the period (11 years) 1378 patients (median age 70 years) were treated with conventional surgical excision for 1635 first index basal cell carcinomas. One hundred and nineteen (7%) were reported as incompletely excised and the policy was to re-excise these. In practice 91/119 (76%) were re-excised with residual tumour found in 48/91 (53%). Probable recurrences Median age was 69 years Of the 1516 lesions that were judged primarily

4 798 completely excised by conventional histological techniques, six patients subsequently developed nine new lesions contiguous with the primary treatment site (scar or graft), Table 1. Patients 1 and 3 had previous treatment before being referred for plastic surgery. Patient 1 had the upper lip BCC first excised in 1980 by a dermatologist and the first recurrence was treated in 1983 by irradiation. His first plastic surgery excision of the next recurrence was before this review period in 1986 with a deep resection clearance margin of 8.2 mm and lateral resection margin clearance of 0.4 mm. Patient 3 had an excision of a right forehead BCC in 1987 and BCC right temple and right forehead in 1988 by the dermatologist. In 1989, she was referred for plastic surgery when an index BCC of right temple was excised completely. Six of the nine were on the temple or forehead. Seven developed in the period and two between 1999 and These were designated probable recurrences, shown in Figs The mean interval to probable recurrences was 41 months (4 months 8 years 10 months). For probable recurrences the median lateral resection clearance margin was 2 mm (range ), and the median deep resection margin clearance was 2.4 mm (range ). For probable recurrences only two lateral clearance resection margins were!1 mm, and only one deep clearance resection margin was less than 1 mm. Possible recurrences Median age was 68 years A further 11 new lesions developed in nine patients near (within 1 cm of) the original scar or graft and were designated possible recurrences, Table 2. Seven of eleven were sited in the forehead or temple. Six developed between and five in The mean interval to possible recurrences was 59 months (12 months 8 years 6 months). For possible recurrences the median lateral resection margin clearance was 4.1 mm (range ), and median deep resection clearance margin was 3.1 mm (range ). Only one lateral clearance margin was!1 mm. There were no deep clearance margins!1 mm. Probable and possible combined Patient ages, sex, site of lesions and the intervals to second tumour appearance are given together with the deep and lateral nontumour tissue clearances around the primary lesion are given in Tables 1 and 2. Thus as percentages of all lesions initially completely excised there were 0.6% probable local recurrences, and 0.72% possible recurrences. Combining these figures produces a maximum recurrence rate of 20/1516 (1.3%). Determinate patients R.W. Griffiths et al. Mohs 13 defined determinate patients as those who survived 5 years after initial treatment differentiating these from patients dying within 5 years of other disease unrelated to basal cell carcinoma. He expressed his results in relation to determinate patients rather than the whole initial cohort. We can apply such an adjustment to the present series, and assume that 40% of patients died within 5 years of other disease (as we have found for cutaneous squamous cell carcinoma patients 27 with median age 76). It is likely that as BCC patients were Table 1 Patient number Data for six patients with probable recurrences Sex Age at first operation Interval between first and subsequent operations Site Index lesions deep resection margin clearance Index lesions lateral resection margin clearance 1 M 78 5 years 9 months Upper lip F 67 4 years 8 months; Forehead; forehead 3.4; ; year 7 months 3 F 53 1 year; 8 years 10 R temple; R 1.9; 2.4; 2.5 2; 0.3; 1.5 months; 2 years 2 months temple; L temple 4 F 78 4 months Forehead F 74 2 years 3 months Glabella M 66 4 years 3 months Postauricular Median months Range (53 78) (4 months 8 years 10 months) ( ) ( )

5 Do basal cell carcinomas recur after complete conventional surgical excision? 799 Figure 1 Patient 1: (A) left upper lip index lesion, (B) left upper lip second lesion 5 years 9 months later. younger, less than 40% would die of other causes in the 5 years, but the 60% determinate figure has been retained as it would if anything overestimate the likely recurrence rate for these patients. [For the present series in which median age for all patients was 70 years, and for those with probable or possible recurrences median ages were 68 and 69 years, if it is assumed that 60% of the original patients remain alive for 5 years follow up, these figures for determinate surviving patients adjust to Table 2 Patient number Data for nine patients with possible recurrences Sex Age at first operation Interval between first and subsequent operations Site Index lesions deep resection margin clearance Index lesions lateral resection margin clearance 1 F 77 8 years 6 months Scalp M 68; 70 2 years 6 months; Scalp; scalp 3.3; ; 4 5 years 3 F 61 5 years Forehead M 63 5 years 8 months Temple F 70 8 years Temple F 52 7 years Nose F 85 5 years Temple M 54 5 years Temple M 75; 75 1 year; 1 year Temple; temple 4; ; 3.6 Median months Range (52 85) (1 year 8 years 6 months) ( ) ( )

6 800 Figure 2 Patient 2: (A) left forehead index lesion, (B) left forehead second lesion at lower limit of scar 4 years 8 months later, (C) left forehead third lesion at right lower limit of skin graft 1 year 7 months after second lesion appearance. probable recurrence 9/910 (0.98%), possible 11/910 (1.2%), and probable and possible combined 20/910 (2%)]. Discussion Despite the theoretical advantage of more extensive histological assessment of excision margins for BCC with Mohs micrographic surgery, the 5 years R.W. Griffiths et al. recurrence rates for BCC after complete excision by either Mohs surgery or conventional surgery are similar in the range %. 1 5,13,14 These publications contain little information on the nature or definition of the local recurrences, however. Specifically, local recurrence is not consistently or uniformly defined, nor are index lesion photographs compared with the appearance of new lesions, nor are initial histological clearance margins around index lesions quantified. The present series represents a minimum 5 years follow up study attempting to identify all patients returning to one surgeon with probable or possible recurrent basal cell carcinoma after conventional surgical excision, relating probability of recurrence to histological clearance measurements and colour photographs. Some of the earlier treated patients (from 1988 onwards) had much longer potential follow up maximum 15 years and indeed some recurrences (two probable and four possible) in this series occurred O5 years. It is recognised that some 18 20% of recurrences will occur beyond 5 years, 12,28,29 but the minimum 5 years follow up chosen here is a useful time scale for comparison between series. An assumption of the study was that recurrences would be referred back to the plastic surgeon rather than elsewhere. Because of our formal network skin cancer links and relationships with general practitioner and dermatology and oncology colleagues we consider this a valid assumption. Also in the study period there was no local Mohs surgeon to whom the patients might have alternatively been referred. Finally, we have not been made aware of any patients being treated elsewhere, and have not had requests for notes nor been subsequently informed of patients being treated for recurrences in other units. However, the possibility that some recurrences have not returned to the plastic surgery unit is acknowledged. The present study indicates that for the whole cohort the combined possible/probable recurrence rate, over minimum 5 years follow up is 1.3 and 2% for determinate cases. Two-thirds of probable and possible recurrences occurred in the forehead and temple area, although these sites account for only 22% of basal cell carcinomas in our practice. 30 These two sites seem, therefore, particularly prone to either true recurrences, or perhaps more likely to field change leading to multiple adjacent separate primary (metachronous) tumours. This study appears to be the first to relate putative recurrences (determined from contemporary colour photographs), to measured clearance margins around the primary tumour, allowing an attempt to assess the likelihood that a further

7 Do basal cell carcinomas recur after complete conventional surgical excision? 801 Figure 3 Patient 3: (A) right temple second lesion at lateral limit of scar of index lesion excision 1 year earlier, (B) left temple index lesion, (C) right temple third lesion just above upper edge of skin graft 8 years 10 months after second lesion excision, (D) left temple second lesion 2 years 2 months after excision of index lesion. lesion is a recurrence or new lesion. For probable recurrences, the median lateral clearance around the initial lesion was 2 mm with only two lesions out of nine having margins!1 mm (both 0.3 mm), and two having clearance O5 mm (5.3 and 6.8 mm). For the possible recurrences, the median lateral Figure 4 Patient 4 index lesion left forehead, the second lesion developed 4 months later as a crust at the 6 O clock position on the edge of the graft (recorded diagramatically in the notes but not photographed). clearance around the initial lesion was 4.1 mm, with only 1/11 having a margin!1 mm (0.8 mm). If not all these lesions were true recurrences, how many were new primary lesions? It is reasonable to question if, for the second lesions with initial primary index lesion lateral resection clearance margins of O1 mm and certainly O2 mm, the second lesion is more likely than not (on the balance of probabilities), to be a second primary lesion in an area of field change, rather than a recurrence of the original either from undetected tumour at the resection edge, or from discontinuous tumour. If the true recurrences were regarded as only those with lateral resection margins of clearance!1mm around the primary tumour, then only 3/1516 (0.2%) would be regarded as likely recurrences. It may well be that some of the possible/probable recurrences here were indeed new primary lesions. If the concerns of Mohs surgery advocates were legitimate, such limited margin sampling would result in conventional histological assessment missing many margins involved by tumour. The consequence should then be a higher recurrence rate than seen with Mohs micrographic surgery. This has not been seen, however, and the local

8 802 R.W. Griffiths et al. Figure 5 Patient 5: (A) index lesion of glabella skin, (B) the second lesion 2 years 3 months later. recurrence rate after conventional surgery does not exceed that associated with the much more extensive tissue examination of micrographic surgery. 13,14 No excess of local recurrence (probable/ possible) was seen in our series, compared with other series of conventional surgery and micrographic surgery. Indeed despite its more extensive examination of margins, Mohs therapy has been associated with an 11% incidence of residual basal cell carcinoma found at wider excision, 31 and clinical recurrence rates ranging % ,32 The definition of local recurrence can be difficult, but the differentiation of recurrence after apparently complete excision, from new tumour adjacent to the primary index lesion excision site is useful. 21 Recurrence after apparent complete excision could arise from discontinuous tumour, 33 although this problem of discontinuous tumour, 34 seems likely to apply equally to micrographic surgery and conventional surgical excision. Whether with relatively large clearance margins of conventional surgery or the lesser 1 3 mm margins for Mohs therapy, (stopping once a tumour free plane was reached), if the main bulk of a basal cell carcinoma has any element of tumour in discontinuity then clinical recurrence of this discontinuous element could appear at the edge of scar or graft. This may explain the findings of Lang et al. 31 Alternatively, new tumour at or beyond the edge of graft or scar could be a further primary unconnected with the original lesion, with the calculated risk of developing one or more new primary metachronous basal cell carcinomas ranging 19 83%, and averaging around 50% over 5 years after index lesion treatment A marked 2 mm surgical margin may be in error by 45%, 47 and measured histological margins may be 20 30% less than the marked surgical margins These figures indicate the corrections required to all reported histological margins to relate to the original surgical margins, and give pause for thought on the practical implementation of advice such as give it a 2 mm margin. All techniques for assessing histological margins have limitations, which need to be acknowledged in interpreting studies of clearance margins. 48,51 53 The histological methods used in this study have been defined in order that findings are put in that context.

9 Do basal cell carcinomas recur after complete conventional surgical excision? 803 Figure 6 Patient 6: (A) index lesion of left post-auricular skin, (B) second lesion at the lower graft edge 4 years 3 months later. Regardless of treatment modality, basal cell carcinomas have been described not only to spontaneously progress but also regress, 54 and specifically, may regress and disappear after incomplete diagnostic incision biopsy All studies of all BCC treatments should acknowledge the potential ability of residual tumour to regress. If conducted with care it appears that complete excision of BCC by conventional surgery and Mohs therapy can be associated with a low order of probability for recurrence over 5 years follow up. To facilitate future comparative studies of different treatment modalities in BCC all clinicians treating basal cell carcinoma should maintain a data base which allows for such studies. Postscript Since, submitting this manuscript for review the results have been published of a prospective randomised trial of surgical excision versus Mohs surgery for basal cell carcinoma of the face with 30 months follow up, which detected no statistically significant differences between the recurrence rates after either intervention. 58 Acknowledgements We are delighted to thank Robert Salthouse for his hard work in preparing the colour illustrations to such a high standard. References 1. Pascal RR, Hobby LW, Lattes R, Crikelair GF. Prognosis of incompletely excised versus completely excised basal cell carcinoma. Plast Reconstr Surg 1968;41: Emmett AJJ, Broadbent GG. Basal cell carcinoma in Queensland. Aust N Z J Surg 1981;51: Breuninger H, Schaumburg-Lever G. Control of excisional margins by conventional histopathological techniques in the treatment of skin tumours. An alternative to Mohs technique. J Pathol 1988;154: Soutar DS, Tiwari R. The skin. In: Soutar DS, Tiwari R,

10 804 editors. Excision and reconstruction in head and neck cancer. Edinburgh: Churchill Livingstone; p [chapter 24]. 5. Park AJ, Strick M, Watson JD. Basal cell carcinomas: do they need to be followed up? J R Coll Surg Edinb 1994;39: Hauben DJ, Zirkin H, Mahler D, Sacks M. The biologic behavior of basal cell carcinoma; analysis of recurrence in excised basal cell carcinoma: part II. Plast Reconstr Surg 1982;69: Nagore E, Grau C, Molinero J, Fortea JM. Positive margins in basal cell carcinoma; relationship to clinical features and recurrence risk. A retrospective study of 248 patients. J Eur Acad Dermatol Venereol 2003;17: Dzubow LE. Recurrent basal cell carcinoma. Plast Reconstr Surg 1982;70: Olshansky K. Recurrent basal cell carcinoma. Plast Reconstr Surg 1982;70: Ley RD. Recurrent basal cell carcinoma. Plast Reconstr Surg 1982;70: Koss N. Recurrent basal cell carcinoma, reply. Plast Reconstr Surg 1982;70: Rowe DE. Comparison of treatment modalities for basal cell carcinoma. Clin Dermatol 1995;13: Mohs FE. Chemosurgery: microscopically controlled surgery for skin cancer past, present and future. J Dermatol Surg Oncol 1978;4: Rigel DS, Robins P, Friedman RJ. Predicting recurrence of basal cell carcinomas treated by microscopically controlled excision. A recurrence index score. J Dermatol Surg Oncol 1981;7: Julian CG, Bowers PW. A prospective study of Mohs micrographic surgery in two English centres. Br J Dermatol 1997;136: Rowe DE, Carroll RJ, Day CL. Long term recurrence rates in previously untreated (primary) basal cell carcinoma: implications for patient follow-up. J Dermatol Surg Oncol 1989; 15: Rowe DE, Carroll RJ, Day CL. Mohs surgery is the treatment of choice for recurrent (previously treated) basal cell carcinoma. J Dermatol Surg Oncol 1989;15: Hirsch P. False negative margins. J Dermatol Surg Oncol 1989;15: Wagner RF. New primary basal cell carcinomas arising in skin flaps following Mohs micrographic surgery for primary and recurrent basal cell carcinoma. J Dermatol Surg Oncol 1990; 16: Robins P. Chemosurgery; my 15 years of experience. J Dermatol Surg Oncol 1981;7: Goldberg L, Stal S, Spira M. Recognition and treatment of recurrent basal cell carcinoma. Ann Plast Surg 1983;11: Griffiths RW. The case for other surgical options. In: Ross D, Spittle M, editors. Key advances in the clinical management of skin cancer. London: Royal Society of Medicine Press; Slater DN, McKee PH. Minimum dataset for the histopathological reporting of common skin cancers. Standards and minimum datasets for reporting cancers. London: Royal College of Pathologists; 2002 p Bentkover SH, Grande DM, Soto H, Kozlicak BA, Guillaume D, Girouard S. Excision of head and neck basal cell carcinoma with a rapid, cross sectional, frozen section technique. Arch Facial Plast Surg 2002;4: Netscher DT, Spira M. Basal cell carcinoma: an overview of tumor biology and treatment. Plast Reconstr Surg 2004;113: 74e 94e. R.W. Griffiths et al. 26. Jallali N. Loupe magnification reduces the incidence of incomplete excision of basal cell carcinoma. Plast Reconstr Surg 2004;113: Griffiths RW, Feeley K, Suvarna SK. Audit of clinical and histological prognostic factors in primary invasive squamous cell carcinoma of the skin: assessment in a minimum 5 year follow up study after conventional excisional surgery. Br J Plast Surg 2002;55: Hruza GJ. Mohs micrographic surgery local recurrences. J Dermatol Surg Oncol 1994;20: Randle HW. Basal cell carcinoma. Identification and treatment of the high risk patient. Dermatol Surg 1996;22: Griffiths RW. Audit of histologically incompletely excised basal cell carcinomas: recommendations for management by re-excision. Br J Plast Surg 1999;52: Lang PG, Duncan IM, Hochman M. Occurrence of subclinical tumor in excised facial subunits. Arch Facial Plast Surg 2004; 6: Wennberg A-M, Larko O, Stenquist B. Five year results of Mohs micrographic surgery for aggressive facial basal cell carcinoma in Sweden. Acta Derm Venereol 1999;79: Wagner RF, Cottel WI. Multifocal recurrent basal cell carcinoma following primary treatment by electrodesiccation and curettage. J Am Acad Dermatol 1987;17: Seidman JD, Berman JJ, Moore GW. Basal cell carcinoma: importance of histologic discontinuities in the evaluation of resection margins. Mod Pathol 1991;4: Beirne GA, Beirne CG. Observations on the critical margin for the complete excision of carcinoma of the skin. Arch Dermatol 1959;80: Downes RN, Walker NPJ, Collin JRO. Micrographic (MOHS ) surgery in the management of peri-ocular basal cell epitheliomas. Eye 1990;4: Telfer NR. Mohs micrographic surgery for nonmelanoma skin cancer. Clin Dermatol 1995;13: Robinson JK. Risk of developing another basal cell carcinoma. A 5 year prospective study. Cancer 1987;60: Schreiber MM, Moon TE, Fox SH, Davidson J. The risk of developing subsequent nonmelanoma skin cancers. J Am Acad Dermatol 1990;23: Karagas MR, Stukel TA, Greenberg ER, Baron JA, Mott LA, Stern RS. Risk of subsequent basal cell carcinoma and squamous cell carcinoma of the skin among patients with prior skin cancer. JAMA 1992;267: Marghoob A, Kopf AW, Bart RS, Sanfilippo L, Silverman MK, Lee P, et al. Risk of another basal cell carcinoma developing after treatment of a basal cell carcinoma. J Am Acad Dermatol 1993;28: Marcil I, Stern RS. Risk of developing a subsequent nonmelanoma skin cancer in patients with a history of nonmelanoma skin cancer. Arch Dermatol 2000;136: Veien K, Veien NK. Risk of developing subsequent nonmelanoma skin cancers. Arch Dermatol 2001;137: Schinstine M, Goldman GD. Risk of synchronous and metachronous second nonmelanoma skin cancer when referred for Mohs micrographic surgery. J Am Acad Dermatol 2001;44: Graells J. The risk and risk factors of a second non-melanoma skin cancer: a study in a Mediterranean population. J Eur Acad Dermatol Venereol 2004;18: Revenga F, Paricio JF, Vazquez MM, Del Villar V. Risk of subsequent non-melanoma skin cancer in a cohort of patients with primary basal cell carcinoma. J Eur Acad Dermatol Venereol 2004;18:514 5.

11 Do basal cell carcinomas recur after complete conventional surgical excision? Lalla R, Brown TL, Griffiths RW. Where to draw the line: the error in marking surgical excision margins defined. Br J Plast Surg 2003;56: Bennett RG. The meaning and significance of tissue margins. Adv Dermatol 1989;4: Blasdale C, Charlton F, Lawrence CM. Factors influencing tumour-free margins in excised basal cell carcinoma. British Association of Dermatologists. Abstr Br J Dermatol 2002; 147(Suppl 62);62:DS Gregory N, Mulvaney M, Pattison T, Hill J, Carlson JA, Goncharuk V. Shrinkage of skin excision specimens and downcoding. Arch Dermatol 2003;139: Rapini RP. Comparison of methods for checking surgical margins. J Am Acad Dermatol 1990;23: Abide JM, Nahai F, Bennett RG. The meaning of surgical margins. Plast Reconstr Surg 1984;73: Paterson DA, Davies JD, McLaren KM. Failure to demonstrate the true resection margins of excised skin tumours: a case for routine marking. Br J Dermatol 1992;127: Franchimont C, Pierard GE, Van Cauwenberge D, Damseaux M, Lapiere ChM. Episodic progression and regression of basal cell carcinomas. Br J Dermatol 1982; 106: Goldwyn RM, Kasdon EJ. The disappearance of residual basal cell carcinoma of the skin. Ann Plast Surg 1978;1: Swetter SM, Boldrick JC, Pierre P, Wong P, Egbert BM. Effects of biopsy-induced wound healing on residual basal cell and squamous cell carcinomas: rate of tumor regression in excisional specimens. J Cutan Pathol 2003;30: Gupta M, Puri P, Kamal A, Nelson ME. Complete spontaneous regression of a basal cell carcinoma. Eye 2003;17: Smeets NWJ, Krekels GAM, Ostertag JU, Essers BAB, Dirksen CD, Nieman FHM, et al. Surgical excision vs Mohs micrographic surgery for basal-cell carcinoma of the face: randomised controlled trial. Lancet 2004;364: