Unplanned incomplete Mohs micrographic surgery

DERMATOLOGIC

SURGERY

Unplanned incomplete Mohs micrographic surgery Shabnam Madani, MD, Shyamala C. Huilgol, FACD, and Alastair Carruthers, MD, FRCPC Vancouver, British Columbia, Canada Background: Incomplete Mohs micrographic surgery (MMS) is the cessation of MMS while the tumor margins are known to be still positive. Objective: Our purpose was to examine the factors behind unplanned incomplete MMS and to identify means of avoiding and managing this situation when it arises. Methods: We performed a retrospective case review examining clinical presentations, histologic features, and management of incomplete MMS. Results: Fifteen of 10,346 procedures (0.15%) were identified as incomplete MMS. Complete records were available in 14 cases. The age range was 30 to 90 years; the study comprised 10 men and 4 women. The tumors included 9 basal cell and 4 squamous cell carcinomas and 1 dermatofibrosarcoma protuberans. The sites involved were nose, medial canthus, ear, scalp, and lower eyelid. Twelve cases dealt with unresectable disease, whereas two patients were unable to tolerate further surgery. Of the unresectable cases, MMS was terminated because of ongoing multifocal positive skin margins, bony invasion, or extension of tumor to other locations. Surgical defects were repaired, whereas residual disease was managed with a variety of methods. Conclusion: Incomplete MMS is a rare problem of either unresectable disease or inability of the patient to tolerate the procedure. Preoperative planning may help to identify both subgroups of patients. A multispecialty approach to managing these cases is often necessary. (J Am Acad Dermatol 2000;42:814-9.)

M

ohs micrographic surgery (MMS) is a highly effective means of ensuring complete excision of locally invasive cutaneous tumors and at the same time permits the conservation of normal surrounding tissue in cosmetically and functionally important sites. This technique relies on excision of the skin cancer in thin layers, permitting the examination of the entire tissue margins with horizontal frozen sections. Mapping of the excised tissue permits localization of residual malignancy and its sequential removal until all margins are clear.1 However, in some cases, MMS is terminated while the margins are known to be still positive. We evaluate 14 cases treated with MMS with known positive margins at the conclusion of the surgery.

From the Division of Dermatology, University of British Columbia. Accepted for publication Dec 9, 1999. Reprint requests: Alastair Carruthers, MD, Dermatologic Surgery, 835 W 10th Ave, Vancouver, BC, Canada V5Z 4E8. Copyright © 2000 by the American Academy of Dermatology, Inc. 0190-9622/2000/$12.00 + 0 16/1/104889 doi:10.1067/mjd.2000.104889

814

MATERIAL AND METHODS The records of all patients who underwent MMS in the Dermatologic Surgery Unit, Vancouver General Hospital, Vancouver, British Columbia, Canada, from Jan 1, 1984 to March 12, 1999 were examined. Cases with positive margins at the termination of the surgery were identified. The clinical presentation, histologic features, and outcomes were evaluated.

RESULTS An incomplete Mohs excision was present in 15 of 10,346 cases (0.15%). Complete records were available in 14 cases and are summarized in Table I. Of these patients, the age range was 30 to 90 years (mean, 65 years; median, 70 years); 10 were men and 4 were women. Cases 10 and 11 represent two different sites in the same man. In 11 cases, there was a history of multiple nonmelanoma skin cancers in other different sites. All patients had undergone preoperative biopsies. There were 9 basal cell carcinomas (BCCs), 4 squamous cell carcinomas (SCCs), and one dermatofibrosarcoma protuberans (DFSP). The histopatholog-

J AM ACAD DERMATOL VOLUME 42, NUMBER 5, PART 1

ic subtypes of BCCs included solid (10%), solid with associated follicular hamartomas (10%), superficial multifocal (30%), infiltrative (30%), and sclerosing (10%) patterns. There were neither data nor slides available to subtype the BCC in case 14 (10%). Two SCCs revealed poor differentiation (50%), one of each grade 3 and 4, whereas the other 2 were both grade 2, well- to moderately differentiated SCCs. One of each SCC subtype (cases 10 and 11) showed basaloid differentiation. The DFSP had a myxoid pattern. In 9 of 14 cases (64%), the tumor represented recurrent disease. Previous treatment included one or more excisions (6/9), electrodesiccation and curettage (1/9), radiotherapy (4/9), and MMS (2/9). The sites involved were nose (5 cases), medial canthus (3 cases), ear (3 cases), scalp (2 cases), and lower eyelid (1 case). The preoperative size of the tumors was available in 12 of 14 cases, the smallest diameter of which was 1.0 cm or larger. At the termination of MMS, 4 patients (all with BCC) had residual multifocal cutaneous involvement; 2 cases with BCC had bony invasion; 4 (2 with BCC, and one each with SCC and DFSP) had tumor extension to orbit/external auditory meatus; and 2 (one BCC and one SCC) were unable to tolerate further surgery. In cases 10 and 11, both SCCs in this man had tumor extension into facial sinuses and bony invasion. The surgical defects in all patients were repaired with skin flaps or grafts, whereas the residual disease was managed in a variety of ways. Of the 4 patients with multifocal skin involvement, case 2 underwent aggressive electrodesiccation and curettage in the area of residual tumor with subsequent repair. Cases 5 and 12 with superficial multifocal BCC of the scalp had extensive postoperative dermabrasion 6 and 9 months, respectively, to remove the residual tumor. The fourth patient (case 7) with BCC associated with follicular hamartomas was followed up with a view to palliative excisions as necessary. Because of his age and the hamartomatous nature of the residual tumor, a decision was made not to pursue other more aggressive treatment options. Bony invasion was encountered in two patients and was managed by burring of the bone before repair (case 4) and bone resection before repair after postoperative radiotherapy (case 8) (Fig 1). Of the 4 cases with tumor extension, 3 had tumoral extension to the orbit. Case 6 demonstrated extension of a medial canthal BCC into the orbital cavity, ethmoid and maxillary sinuses, and nasolacrimal duct. He was treated with radiotherapy before repair. Case 9 (Fig 2) had a DFSP of the right medial canthus extended to the right brow, which was found to be invading his anterior orbit; it also received postoperative radiothera-

Madani, Huilgol, and Carruthers 815

py. In case 13, the BCC invaded orbital fat and muscles and was managed with complete orbital exenteration and repair. Case 1 with BCC extending into the external auditory canal was managed with postoperative radiotherapy. Cases 10 and 11 were both SCCs with extension into facial sinuses and bony invasion. On each occasion, the patient (case 11) underwent extensive resection and repair, with postoperative radiotherapy after his second surgery. MMS was continued at a later date for two elderly patients (cases 3 and 14) initially unable to tolerate the surgery. Postoperative follow-up records were available in 8 of 14 cases. The follow-up period ranged from 6 months to 13 years, with recurrence in two cases.

DISCUSSION MMS is a highly efficacious method of ensuring complete tumor excision in nonmelanoma skin cancers. Nonetheless, MMS does have limitations, and, in a minority of cases, complete excision is not attainable. The risk factors for unresectable disease are well illustrated by our 14 cases and include large tumors; recurrent tumors, especially previously irradiated tumors; unfavorable locations of lesions, aggressive tumors, and subtypes; and elderly patients. However, many of these risk factors are also indications for Mohs surgery because it is with precisely these aggressive tumors that MMS may prove most valuable. Size Large tumors present a challenge to the Mohs surgeon for several reasons. Such tumors are more likely to extend deeply as well as widely with significant subclinical spread.2 Thus there is a higher likelihood of involving deep or vital structures such as bone, sinuses, orbit, and eye. These sites of tumor extension are not amenable to removal under local anesthesia and require excision with the use of general anesthesia by the appropriate specialty surgeon. In addition, bone and other deeper tissues may not be suitable for frozen section examination. Our patients illustrate the risks of larger tumors; all were larger than 1.0 cm in the smallest diameter and demonstrated tumoral extension both widely and deeply. The larger the tumor, the more likely the chance of local recurrences. BCCs larger than 2.0 cm3,4 and SCCs larger than 1.0 cm are associated with a higher risk of recurrence.5 A higher rate of perineural invasion also occurs in larger tumors; one study has shown that SCCs larger than 2.0 cm have a higher risk of perineural invasion, making the tumor more likely to extend intracranially.6 Furthermore, two additional problems arise with

816 Madani, Huilgol, and Carruthers

J AM ACAD DERMATOL MAY 2000

Table I. Incomplete MMS: Details of patients, tumors, management, and outcome Patient No.

MMS date

Age (y)/ Sex

Location

Tumor & defect sizes (cm)

MMS levels

1 2 3

1999 1998 1998

53/M 90/F 87/M

Ear: R concha & EAC Nose: R nasal ala Ear: R antihelix

1.5 × 1.7; 3.2 × 3.9 1.0 × 1.2; 3.4 × 2.3 1.3 × 0.9; 7.2 × 4.2

6 6 7

4 5

1997 1996

79/M 47/F

Nose: L ala Scalp: R centroparietal Eye: R MC

2.0 × 1.4; 6.5 × 6.5 6.0 × 2.5; 7.5 × 7.5

4 4

6

1995

46/M

3.0 × 1.0; 3.2 × 1.5

3

7

1994

74/M

Nose: R dorsum

1.0 × 1.0; N/A

6

8 9

1994 1991

82/F 73/M

Eye: L MC Eye: R brow & MC

2.0 × 2.0; 4.8 × 3.5 2.0 × 1.5; 7.2 × 4.7

4 3

10

1989

67/M

Nose: L ala, tip & lower glabella Nose: R nasal side wall, R cheek Eye: R lower eyelid

7.0 × 4.5; 8.0 × 6.0

5

N/A; 7.0 × 6.8

3

40/M

SCC: poor differentiation w/ basaloid features, RC SCC: good to moderate differentiation, RC BCC: infiltrative, RC

11

1989

67/M

12

1988

N/A; N/A

3

13 14

1984 1984

30/F 81/M

BCC: multifocal superficial, P BCC: RC

Scalp: R frontal Ear: R EAC

2.5 × 3.0; 7 × 7 N/A; N/A

7 8

Tumor type

BCC: infiltrative, P BCC: solid, P SCC: poor differentiation, RC BCC: morpheic, RC BCC: multifocal superficial, RC SCC: good to moderate differentiation, RC BCC: solid, associated w/ FH, P BCC: infiltrative, P DFSP: myxoid, RC

EAC, External auditory canal; FH, follicular hamartomas; L, left; MC, medial canthus; N/A, not available; P, primary; R, right; RC, recurrent; RT, radiotherapy; tx, therapy.

deeper tumors: the Mohs surgeon is generally less familiar with the anatomy, and maintaining the correct orientation of specimens for mapping becomes more difficult. It has been suggested that the best approach for such tumors is to adopt a team approach in a standard operating theater, the Mohs surgeon removing superficial tissue and assisting with mapping and histopathologic examination of deeper tissues.7 In addition, larger tumors are likely to involve several cosmetic units, making repairs more difficult and cosmetically less appealing. Previous treatments Recurrent tumors also pose several problems. They are well known to have a greater incidence of further recurrence after treatment by any modality and this includes Mohs surgery. Although 5-year recurrence rates for primary BCCs after Mohs surgery and other modalities are 1% and 7.5% to 10.1%, respectively,8 recurrent BCCs have 5-year cure rates of 60% to 90% after other treatment modalities and 96% after MMS.1 Long-term (>5 years) follow-up of recurrent SCCs after surgical excision or MMS revealed local recurrence rates of 23.3% and 10.0%, respectively.5 Not only do recurrent cancers appear to be more clinically aggressive, previous therapy produces tis-

sue distortion and scarring, concealing the lesion and any multifocal disease. Previous surgery may also open new planes of growth (eg, in planes of blunt dissection for increasing tissue mobility).9 Tumors recurring after radiotherapy appear to represent a subset of particularly difficult recurrences caused by biologic alterations with increased aggressiveness combined with the changes of postradiation scarring.10 Our patients had a difficult group of tumors; 9 were recurrent disease. Five of these had received radiotherapy, and 2 had been previously treated with MMS, both on two separate occasions. Location Certain anatomic locations are more likely to have high rates of recurrence. These include the embryonic fusion planes (inner canthus, nasolabial fold, preauricular area, retroauricular sulcus, mid to lower lip and chin, and philtrum), as well as the nasal ala, periocular area, ear, scalp, and temple.2 Tumors in these locations may spread along tissue planes, perichondrium, or bone; invade deeply in anatomic folds; and skip along nerves.1,11 Tumors located in the scalp show a tendency for distant outgrowths beyond the clinically apparent borders of the tumor.12 In our study, the nose, ears, medial canthi, and scalp were sites of unresectable disease, suggesting

Madani, Huilgol, and Carruthers 817

J AM ACAD DERMATOL VOLUME 42, NUMBER 5, PART 1

Past tx

Reason for MMS termination

N/A N/A ED&C, MMS × 2

Extension down EAC Superficial multifocal tumor Intolerance

Repair & RT ED&C, repair MMS later

N/A 6 mo: BCC, RC 9 mo: no RC

Excisions Excision & RT

Bony adhesion Superficial multifocal tumor

Burred bone & repair Repair & DA 6 mo postop

— 3 y: no RC

Excisions

Tumor extension into orbital wall, facial sinuses Superficial multifocal tumor

RT & repair

18 mo: no RC

Repair & observation

—

Bone resection, repair & RT Repair & RT

— 8 y: radiation; optic neuropathy

N/A N/A Excisions Excisions & RT Excisions & RT Excisions × 2, RT & MMS × 2 N/A RT

Management

Follow-up

Bony invasion Tumor extension into orbit: trochlea Bony invasion & tumor extension to facial sinuses Bony invasion & tumor extension to facial sinuses Extension to orbit

Extensive resection & repair

—

Extensive bone & soft tissue resection, repair & RT Orbital exenteration & repair

—

Superficial multifocal tumor Intolerance

Repair & DA 9 mo postop MMS at later date, complete

13 y: RC at 5 y —

that the same sites of high risk for recurrence are also more likely to be at risk for incomplete MMS. Unresectable disease falls into two groups: cutaneous and deep disease. Very extensive cutaneous disease was present in our cases on the scalp and nose, whereas deep disease occurred on the nose, around the eye, and on the ear. Therefore we suggest careful preoperative evaluation of tumors, particularly in the mid face (eyelids, particularly medial canthus, and nose), periauricular areas, and scalp. The medial canthus seems to be a particularly unfavorable location, as noted in other articles on this subject. Boynton et al,7 addressing the issue of incomplete Mohs surgery, had 4 cases involving the periocular area. Mordick, Hamilton, and Dzubow13 reported 3 cases of incomplete Mohs surgery in their large series, two of which were in the periocular area. Preoperative imaging should be strongly considered for cutaneous malignancies of the medial canthus because of their propensity for intraorbital spread. Alternatively, ophthalmologic review may assist in identifying clinical signs of intraorbital disease such as restricted extraocular movements. Histologic type In evaluation of patients, the tumor type and subtype are also highly relevant. Certain histo-

8 y: no RC

pathologic types of tumors, such as some subtypes of BCC (morpheaform, infiltrative, micronodular),14 SCCs with histologic grade 2 or higher,15 and DFSP show aggressive behavior in tissue invasion. In our study, there were 9 BCCs with histopathologic subtypes of solid (10%), solid with associated follicular hamartomas (10%), multifocal superficial (30%), infiltrative (30%), and sclerosing (10%) patterns. There were 4 SCCs, all with grade 2 and higher, two of which were poorly differentiated, and one myxoid type DFSP. Other factors Patient comfort is another factor that was relevant in our cases of incomplete MMS. Two patients, aged 80 and 87 years, one with recurrent SSC of the antihelix and the other with recurrent BCC of the external auditory meatus could not tolerate further surgery at the point when MMS was terminated. They did, however, have subsequent MMS to successfully remove residual tumor. MMS may be difficult for patients to endure for reasons of comfort in a prolonged procedure or because of inadequate analgesia. Deeper tissues on the head and neck are often more difficult to anesthetize with local anesthesia and may produce diffuse headaches. Scar tissue is also, at times, more difficult to infiltrate.

818 Madani, Huilgol, and Carruthers

J AM ACAD DERMATOL MAY 2000

Fig 1. Case 8. Tumor invasion of the bone.

Consideration should be given to these matters by attempting to provide a restful place between levels and perhaps utilizing regional blocks, longer acting anesthetics, and oral analgesia. Furthermore, even though the use of MMS in patients older than 90 years has been shown to be safe and effective,16 the general health and expectations of the patients and their families should be considered in the management of skin tumors in the elderly patient. In the very elderly, perhaps the simplest procedure to eradicate the tumor should be considered, even if the cure rate is considerably lower than that of MMS. Although Mohs surgery was planned in all these cases with the intent of achieving a tumor-free plane, some of the cases had built-in provisos. Case 9 was referred from an orbital surgeon with the proviso that if the tumor was close to the trochlea, Mohs surgery would be terminated, leaving residual disease in that area which only could be treated by radiation therapy after surgical repair. Case 12 was assessed before the third Mohs surgery and the decision was made that if this aggressive tumor involved the orbit in this young man, then, when other margins were clear, orbital exenteration would be performed. Cases 11 and 12 demonstrate the important principle of using en bloc for section technique at appropriate times. When the tumor is in orbital fat or parotid gland, especially an aggressive tumor or in a young patient, it is often preferable to plan orbital exenteration or parotidectomy rather than continuing with MMS. These and other cases demonstrate the importance of multispecialty consultation in these challenging cases. Presurgical assessment using appropriate biopsy, imaging techniques, and referral can help the judgment of the Mohs surgeon. In all of these cases the primary surgeon was an experienced Mohs surgeon, and in the majority of them the

Fig 2. Case 9. Tumor extension into the orbit.

extensive nature of the tumor was not at all apparent before MMS. Patient 13, who had a small red scaly patch in the right parietal scalp anteriorly, had superficial multifocal BCC throughout the scalp, later treated by extensive dermabrasion. Planned incomplete MMS is performed more frequently. Examples include situations in which the bone, orbit, and other sites are known to be involved, and it is the task of the Mohs surgeon to achieve tumor-free soft tissue or extraorbital margins before the second phase of surgery.

CONCLUSIONS Our cases demonstrate a variety of ways of managing the residual disease after the termination of MMS. For superficial disease, we used subsequent dermabrasion and observation. Other resurfacing techniques (eg, carbon dioxide laser) or radiotherapy would be alternatives, although it might be difficult to decide where the radiation field should cut off. In deep disease, wherever possible, further resection was arranged. Radiotherapy was given in several cases either after resection or as a sole thera-

J AM ACAD DERMATOL VOLUME 42, NUMBER 5, PART 1

peutic maneuver for residual disease. MMS was recommenced at a later date in those patients initially unable to tolerate further surgery. The Mohs surgeon needs to take great care when referring cases of incomplete MMS to other specialists to ensure accurate identification of the residual areas of tumor positivity. We suggest the use of Polaroid photographs, combined with diagrams and marking of the patient with nicks, sutures, or dyes such as gentian violet. Alternatively, the Mohs surgeon could accompany the patient and communicate in person. In conclusion, careful assessment of high-risk tumors is required preoperatively. Factors to consider include tumor size, histology and subtype, location, previous treatment, and the patient’s age and general health. Consideration should be given to imaging investigations and other specialist consultations in cases in which it is suspected that complete resection with MMS will not be possible. Depending on the site of involvement, consultations with ophthalmologists, ear/nose/throat and plastic surgeons, and surgical and radiation oncologists may be necessary. It may be possible to arrange joint operations or schedule sequential surgeries. Prearrangements with the other physicians will provide the best management for the patient, improve the quality of care, and help to minimize the possibility for miscommunication between the physicians. When a decision is made to terminate MMS, the appropriate treatment of residual tumor should be arranged. Finally, prolonged follow-up is desirable in these difficult management cases. Drs L. M. Warshawski and D. M. Zloty are sincerely thanked for permission to include their patients in this study as is Dr J. Shapiro for his valuable advice and encouragement.

Madani, Huilgol, and Carruthers 819

REFERENCES 1. Shriner DL, McCoy DK, Goldberg DJ, Wagner RF. Mohs micrographic surgery. J Am Acad Dermatol 1998;39:79-97. 2. Lang PG, Osguthorpe JD. Indications and limitations of Mohs micrographic surgery. Dermatol Clin 1989;7:627-44. 3. Mohs FE. Microscopic surgery for the microscopically controlled excision of eyelid cancer: history and development. Adv Ophthalmol Plast Reconstr Surg 1986;5:381-408. 4. Robins P. Chemosurgery: my 15 years of experience. J Dermatol Surg Oncol 1981;7:779-89. 5. Rowe DE, Carroll RJ, Day CL. Prognostic factors for local recurrence, metastasis, and survival rates in squamous cell carcinoma of the skin, ear, and lip. J Am Acad Dermatol 1992;26:976-90. 6. Lawrence N, Cottel WI. Squamous cell carcinoma of skin with perineural invasion. J Am Acad Dermatol 1994;31:30-3. 7. Boynton JR, Rounds MF, Quatela VC, Brown MD.The significance of positive margins (known and unknown) at the conclusion of Mohs surgery in the orbital region. Ophthalmol Plast Reconstr Surg 1995;12:51-7. 8. 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;15:315-28. 9. Bennett RG. Mohs’ surgery: new concepts and applications. In: Bailin PL, Ratz JL, Wheeland RG, editors. Dermatologic clinics. Philadelphia: WB Saunders; 1987. p. 409-28. 10. Smith SP, Grande DJ. Basal cell carcinoma recurring after radiotherapy: a unique, difficult treatment subclass of recurrent basal cell carcinoma. J Dermatol Surg Oncol 1991;17:574-8. 11. Dufresne RG. Mohs surgery: fresh tissue technique. In: Wheeland RG, editor. Cutaneous surgery. Philadelphia: WB Saunders; 1994. p. 880-91. 12. Mohs FE, Zitelli JA. Microscopically controlled surgery in the treatment of carcinoma of the scalp. Arch Dermatol 1981;117:764-9. 13. Mordick TG, Hamilton R, Dzubow LM. Delayed reconstruction following Mohs’ chemosurgery for skin cancers of the head and neck. Am J Surg 1990;160:447-9. 14. Sexton M, Jones DB, Maloney ME. Histologic pattern analysis of basal cell carcinoma. J Am Acad Dermatol 1990;23:1118-26. 15. Brodland DG, Zitelli JA. Surgical margins for excision of primary cutaneous squamous cell carcinoma. J Am Acad Dermatol 1992;27:241-8. 16. Farlane DF, Pustelny BL, Goldberg LH. An assessment of the suitability of Mohs Micrographic Surgery in patients aged 90 years and older. Dermatol Surg 1997;23:389-93.