Anda di halaman 1dari 8

Journal of Plastic, Reconstructive & Aesthetic Surgery (2017) 70, 1738e1745

Risk factors for recurrence of facial basal


cell carcinoma after surgical excision: A
follow-up analysis*
Linus T.D. Armstrong a, Mark R. Magnusson b,*,
Michelle P.B. Guppy a

a
University of New England, School of Rural Medicine, Armidale, NSW, Australia
b
Toowoomba Plastic Surgery, Toowoomba, QLD, Australia

Received 13 December 2016; accepted 14 April 2017

KEYWORDS Summary Background: Several tumour features and surgical factors have been implicated in
Basal cell carcinoma an increased risk of BCC recurrence after excision; however, there are limited data on facial
(BCC); lesions specifically. This study sought to evaluate risk factors of facial BCC, which may influ-
Nonmelanocytic skin ence future treatment and follow-up regimes.
cancer (NMSC); Methods: Facial BCCs excised from a single surgeon practice over a 2-year period were
Face; included in the study. Data pertaining to patient demographics, lesion depth of invasion, sur-
Risk factors; face area, excision margins, perineural infiltration, location, previous history of recurrence,
Recurrence; histological subtype and ulceration were extracted. A search of recurrence was conducted
Excision margins over the following 70e80 months.
Results: In total, 331 cases of facial BCC were included, and 10 lesions recurrences (3%) were
identified within the observation period. Infiltrative (p Z 0.02) and micronodular (p Z 0.04)
subtypes as well as incomplete or close (within 1 mm) peripheral (p Z 0.01) and deep excision
margins (p Z 0.04) were significantly associated with tumour recurrence. Five of the 10 recur-
rent lesions had been re-excised for a recurrence previously, placing them at much greater risk
of future recurrence (p Z 0.00).
Conclusions: Incomplete and close excision margins, infiltrative and micronodular subtypes
and previous excision are strong risk factors for facial BCC recurrence. Although depth of in-
vasion, perineural infiltration, ulceration and surface area may indicate the aggressive nature
of a lesion, the results suggest that with adequate excision margins, these factors may not in-
fluence the recurrence rate. The strongest risk factor was a lesion having already recurred

*
Part of this article has been presented at the following conference: “Characteristics of Recurrent Craniofacial BCC: Identifying the High
Risk Lesion”, Plastic Surgery Congress 2015, Brisbane, Australia (6the10th May, 2015).
* Corresponding author. Allure Cosmetic Clinic, 8 Margaret St, East Toowoomba 4350, QLD, Australia.
E-mail address: mark@toowoombaplasticsurgery.com.au (M.R. Magnusson).

http://dx.doi.org/10.1016/j.bjps.2017.04.006
1748-6815/ª 2017 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.

Downloaded for andromeda masako (andromeda72@yahoo.com) at Universitas Sumatra Utara from ClinicalKey.com by Elsevier on October 12, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
Analysis of the risk of facial BCC recurrence after excision 1739

after previous excision, and it suggested that these lesions be treated with particular caution
and a closer follow-up regime be employed.
ª 2017 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Else-
vier Ltd. All rights reserved.

Introduction recurrence.9 It was concluded that embryologic fusion


planes are located in cosmetically sensitive areas with a
Basal cell carcinoma (BCC) is the most prevalent form of paucity of subcutaneous tissue, which are more prone to
skin cancer, comprising approximately 75% of all skin ma- inadequate or conservative excision to ensure ease of
lignancies.1 The pathogenesis of BCC is thought to be closure and cosmesis. This study is a follow-on analysis of
multifactorial with the influence of ultraviolet radiation the original participant cohort, which sought to further
(UVR) playing a large role in its development.2 The influ- evaluate facial BCC characteristics influencing recurrence.
ence of UVR in the development of BCC also supports the This study aimed to provide insight into the characteristics
anatomical distribution of lesions, with a majority found in associated with high risk of recurrence specifically for facial
the head and neck regions.3 Although BCC rarely metasta- lesions. Identifying BCC with high-risk characteristics may
sizes (<0.1%), it has the potential to cause significant affect physician treatment approaches, follow-up protocols
morbidity through the invasion and destruction of local and influence earlier referral of ‘high risk’ lesions for spe-
tissue, resulting in severe disfigurement, loss of function cialised treatment.
and, in rare cases, death.4 Surgical excision is the most
commonly used treatment modality for both low-risk and Patients and methods
high-risk BCC and is considered to have the overall highest
cure rate.5 A retrospective cohort follow-up study was conducted. All
Recurrence of BCC after primary excision occurs within lesions excised from 1 January 2006 to 31 December 2007
approximately 2e10% of cases, with several lesion charac- were extracted from a single surgeon practice in Too-
teristics and surgical variables previously identified as high woomba, Australia. Facial lesions were defined as those
risk for recurrence.6e8 Surgical excision margins have the located anterior to the ear, superior to the mandibular
greatest impact on BCC recurrence, with a higher recur- boarder and inferior to the temporal and forehead hairline
rence rate seen with narrower margin achieved.9,10 Lesions (or their equivalent in the presence of baldness). In total,
with a prior history of recurrence have also been reported 866 lesions were excised by the treating surgeon during
to have higher rates of recurrence following further exci- the study period. A total of 522 lesions were excluded for
sion than primary lesions.11 being located outside the face or not being identified as
Tumour surface area has also been implicated, with a BCC upon histological analysis. Fifteen lesions were
direct relationship existing between increasing lesion size excluded as no diagram or pathology data were available,
and recurrence rates.8,12 Invasive histological subtypes (i.e. and one patient was excluded due to unexpected death
sclerosing, ulcerative, infiltrative and micronodular) and shortly after lesion excision (unrelated to the operation).
lesions with perineural invasion have been associated with In total, 331 lesions were finally included in the original
deep subclinical extensions, higher rates of positive mar- participant cohort for statistical analysis. The original
gins after excision and, ultimately, higher rates of study and all data extracted for this follow-up analysis
recurrence.13e16 One study analysing patient demographics were approved by the University of New England Ethics
and BCC recurrence found that male gender and age over Committee on 4 June 2013.
60 years also posed significant increased risk.8 Lesion pathology data, patient consultation notes and
Despite a large number of studies assessing these vari- operative notes were obtained for the participant cohort.
ables, there are limited data on other factors such as depth Data pertaining to lesion depth of invasion, surface area,
of tissue invasion and location on the face. Additionally, excision margins, perineural infiltration, location, previous
previous studies on recurrent BCC characteristics have history of recurrence, histological subtype, post-operative
analysed data pertaining to BCC excised from all locations, treatment (adjuvant modalities used and lesions requiring
with few studies assessing recurrent lesions specifically on re-excision) and presence of ulceration were gained. Lesion
the face. There may be a variation in the incidence and location was ascertained from diagrams in patient consul-
nature of recurrent facial lesion characteristics because of tation and operation notes. Lesions were subsequently
the complex topographical anatomy and unique anatomical allocated to a specific zone on the studies anatomic dia-
structures present, a paucity of subcutaneous tissue in the gram (Figure 1). Lesion depth of invasion was ranked on the
midfacial region and high density of neural and sebaceous Clark’s scale according to the deepest structure invaded on
tissue. histological analysis. Lesion surface area was calculated
The authors previously concluded that BCC developing in using axial dimensions provided in the pathology report.
embryologic fusion planes is not more invasive nor at Standardised excision margins were used for all craniofacial
greater risk of recurrence, and that excision margins lesions excised by the treating surgeon during this period
appear to have the greatest influence on lesion (Table 1).9 Deep excision margins were guided by the

Downloaded for andromeda masako (andromeda72@yahoo.com) at Universitas Sumatra Utara from ClinicalKey.com by Elsevier on October 12, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
1740 L.T.D. Armstrong et al.

Figure 1 Anatomic diagram of the face.9 Lesions located in the grey anatomical units were considered for inclusion in the study.
Embryologic fusion planes are highlighted in green (medial canthus, paranasal, ala crease, inferior naris, philtral ridge, vertical lip
and pre-auricular).

Search for recurrence


Table 1 Peripheral excision margins.9
Lesion Diameter Border acuity Margina A search of lesion recurrence was conducted over the
<1.0 cm Well defined 3 mm following 70e80 months after excision. The search was
<1.0 cm Less well defined 4 mm conducted within local patient files, private pathology
1.0e2.0 cm Well defined 4 mm laboratories operating in the area and the statewide pa-
1.0e2.0 cm Less well defined R5 mm thology service. To accurately identify recurrent lesions, a
>2.0 cm Well to poorly defined R5 mm guideline using key lesion and tissue sample characteristics
Recurrent lesions Well to poorly defined R5 mm was created in consultation with a senior pathologist. These
(all sizes) characteristics included: lesion location, microscopic find-
a ings (dermal scarring and hypertrophic changes), lesion
Standard margins of excision were employed irrespective of
lesion location. subtype and clinical history attached with pathology sam-
ples. All potentially recurrent lesions were identified and
assessed by the principle researchers by using these
appearance of tumour invasion intraoperatively, and an criteria.
unbreached fascial plane was often used as a deep margin
for lesions located in areas with a thin layer of subcu-
taneous tissue. Surgical excision with frozen section mi- Statistical analysis
croscopy was rarely used and reserved for lesions with
aggressive histological subtypes and unknown extent of Variables were coded for and statistically analysed using
invasion, where tissue sparing may have greatly influenced SPSS Version 21 (SPSS IBM, New York, USA). A p < 0.05 was
the reconstruction required and overall cosmetic outcome. considered significant for all tests. The incidence of infil-
Lesions excised with inadequate deep or superficial trative and micronodular subtypes, perineural infiltration,
margins upon histological analysis received early re- ulceration, gender and previous recurrence were compared
excision. Topical imiquimod was considered a treatment between recurrent and non-recurrent lesions by using a
option for superficial BCC with close or involved lateral Fisher’s exact test. Recurrence in the nasal region was
margins following excision and was the preferred treatment compared to that in other facial areas by using a Chi-square
for patients with normal immune status. Close margins test. The surface area and the mean patient age were
were defined as a microscopic tumour-free zone of 1 mm or compared between recurrent and non-recurrent lesions by
less on the deep and or lateral margin of the pathology using an independent samples t-test. Surface area and
sample. Invasive subtypes with close margins were also lesion re-excision due to inadequate margins was compared
routinely re-excised or treated with radiotherapy in the using an independent samples t-test. Depth of invasion for
presence of perineural infiltration. recurrent and non-recurrent lesions as well as lesions

Downloaded for andromeda masako (andromeda72@yahoo.com) at Universitas Sumatra Utara from ClinicalKey.com by Elsevier on October 12, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
Analysis of the risk of facial BCC recurrence after excision 1741

requiring re-excision due to inadequate margins were The nasal region was the most common area from which
analysed using an independent samples ManneWhitney U BCC was excised, with 148 (44.7%) of all facial lesions
test. located in this region. Four of the 10 recurrent lesions were
located on the nasal tip and neighbouring ala nasi. The
remaining recurrences occurred on the temple, forehead,
Results lower eyelid (two recurrences), zygoma and medial
mandible. There was no significant difference in the rate of
In total, 331 cases of facial BCC were included in the study. recurrence between lesions excised from the nasal region
Ten lesion recurrences (3% of the cohort) were identified versus all other facial areas (p Z 0.093).
within the observation period. The mean patient age was 66 The median depth of invasion for recurrent and non-
years, with a range from 28 to 99 years of age. There was a recurrent lesions was to Clark level four (reticular dermis).
female predominance with 146 females (53.5%) and 127 There was no significant difference in the level of invasion
males (46.4%) included in the participant group. Males were distribution between the two groups (ManneWhitney
older on average than females, with a mean age of 68 years U Z 1467.5, n recurrent lesion Z 8, n non-recurrent
compared to 64 years (p Z 0.023), respectively. There was lesions Z 291, p Z 0.163). When comparing lesions
no significant difference between the mean patient age of requiring re-excision with the depth of invasion, however,
the recurrent (71 years) and non-recurrent groups (66 lesions that were re-excised due to close or inadequate
years) (p Z 0.378). There was no significant difference margins shortly after the primary excision were more deeply
between the patients’ gender and risk of recurrence invasive than lesions not requiring early re-excision (Man-
(p Z 0.353). neWhitney U Z 3714, n re-excision Z 19, n no re-
Recurrent lesions were significantly more likely to have excision Z 280, p Z 0.001). Almost all lesions requiring re-
close or incomplete peripheral and deep excision margins excision due to inadequate margins invaded to Clark level
than non-recurrent lesions on primary excision. Forty four or five. There was no significant difference in the mean
percent of recurrent lesions had a close or incomplete pe- surface area of recurrent lesions compared to non-recurrent
ripheral excision margin following primary excision lesions (103.85 mm2 and 69.17 mm2, respectively)
compared to 10.9% of the non-recurrent cohort (p Z 0.013) (p Z 0.273). There was no difference in the mean surface
(Figure 3). Thirty percent of the recurrent lesions had a area of lesions requiring early re-excision due to inadequate
close or involved deep margin upon primary excision margins and those not requiring re-excision (p Z 0.189).
compared to 12.1% of the non-recurrent cohort (p Z 0.048) A majority of lesions contained tumour of mixed histo-
(Figure 4). Twenty-three lesions required early re-excision logical types (Table 2). Recurrent lesions were more likely
due to incomplete or insufficient margins, with residual to contain infiltrative and micronodular differentiation
BCC identified within nine (39.1%) of the re-excision sam- than non-recurrent lesions, with 24.3% of non-recurrent
ples. Post-operative treatment with imiquimod and radio- lesions containing infiltrative subtype compared to 60% of
therapy was used in 13 (3.9%) and 4 (1.2%) cases, recurrent lesions (p Z 0.02) and 20.6% of non-recurrent
respectively. lesions containing micronodular subtype compared to 50%

Figure 2 Incidence of subtypes in recurrent and non-recurrent lesions. Recurrent lesions were more likely to contain infiltrative
and micronodular differentiation than non-recurrent lesions (p < 0.05). Sixty percent of recurrent lesions contained infiltrative
differentiation compared to 24.3% of non-recurrent lesions (p Z 0.02), and 50% of recurrent lesions contained micronodular dif-
ferentiation compared to 20.6% of non-recurrent lesions (p Z 0.041).

Downloaded for andromeda masako (andromeda72@yahoo.com) at Universitas Sumatra Utara from ClinicalKey.com by Elsevier on October 12, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
1742 L.T.D. Armstrong et al.

Figure 3 Peripheral margin of excision and lesion recurrence. Forty percent of recurrent lesions reported a close or incomplete
peripheral excision margin following primary excision compared to 10.9% of the non-recurrent cohort (p Z 0.013).

Figure 4 Deep margin of excision and lesion recurrence. Thirty percent of the recurrent lesions had a close or involved deep
margin upon primary excision compared to 12.1% of the non-recurrent cohort (p Z 0.048).

of recurrent lesions (p Z 0.041) (Figure 2). There was no


Table 2 Incidence of BCC subtypes.
difference in the incidence of ulceration between recurrent
Subtype Lesions (no.) % of total lesions and non-recurrent lesions (20% vs. 16.5%, respectively)
Superficial 127 38.4 (p Z 0.674).
Nodular 198 59.8 Twenty-six lesions were identified as previously having
Micronodular 71 21.5 recurred upon presentation to the treating surgeon for
Fibrosing 67 20.2 excision. These lesions were at higher risk for future
Infiltrative 84 25.4 recurrence, with 5 of the 10 recurrent lesions having been
Metatypical 5 1.5 excised and recurred previously (p Z 0.00) (Table 3).
Not specified 14 4.2 Fifteen cases of perineural invasion were identified
within the participant cohort. One case of perineural

Downloaded for andromeda masako (andromeda72@yahoo.com) at Universitas Sumatra Utara from ClinicalKey.com by Elsevier on October 12, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
Analysis of the risk of facial BCC recurrence after excision 1743

non-recurrent lesions. These findings reinforce the aggres-


Table 3 Primary or previously recurrent lesion and effect
siveness of these subtypes and the greater risk of recur-
on future recurrence.9
rence they incur.
Primary Previously Compared to primary BCC, lesions with a prior history of
recurrent recurrence have been reported to be at higher risk of
No future Count 300 21 recurrence following further excision. This is likely due to
recurrence % of all non-recurrent 93.5% 6.5% the aggressive nature of these lesions leading to recurrence
lesions in the first instance, subsequently causing the distortion of
tissue and the presence of subclinical tumour in the
Future Count 5 5 resulting scarred/hypertrophic area. Twenty-six previously
recurrence % of all recurrent 50.0% 50.0% recurrent lesions were identified in the participant cohort,
lesions five of which recurred following subsequent excision by the
Total lesions Count 305 26 study’s treating surgeon (19.23% recurrence rate). Previ-
% of total 92.1% 7.9% ously recurrent lesions were significantly more likely to
recur than primary lesions, accounting for 50% of the
recurrent lesions identified in the study follow-up period
(p Z 0.000). This strongly supports previous research
invasion was identified in a recurrent lesion sample implicating prior history of recurrence with increased risk.
(p Z 0.375). Radiotherapy was used post-operatively in Additionally, four out of five of these previously recurrent
four cases, all of which demonstrated significant perineural lesions contained infiltrative differentiation. Infiltrative
infiltration and aggressive infiltrative histological BCC with a prior history of recurrence was the group at the
differentiation. highest risk of future recurrence.
An observed increased difference in the mean surface
area of recurrent lesions compared to non-recurrent lesions
Discussion (103.85 mm2 and 69.17 mm2, respectively) did not achieve
statistical significance (p Z 0.273), which may be a prob-
The recurrence of BCC following primary excision is multi- lem of cohort size. This does not support previous research
factorial, with several lesion characteristics and treatment that suggested an increased rate of recurrence for larger
factors identified previously as high risk. Many of these lesions.12 The increased rates of recurrence for larger and
factors are suggestive of tumour aggressiveness and the more deeply invasive tumours reported previously is likely
extent of underlying tissue invasion, ultimately influencing due to the increased difficulty in ensuring adequate mar-
the ability to ensure adequate tumour clearance and thus gins with primary excision. This is reinforced by the current
lesion recurrence. It is the authors’ belief that with the study’s findings, which found that deeply invading lesions
strict use of standardised clinical excision margins and a re- (Clark levels 4 and 5) required re-excision due to inade-
treatment protocol, many of these factors can be mitigated quate margins following primary excision more often than
and may not incur the same risk reported previously. superficial lesions. The results of this study suggest that
If clear excision margins are achieved on histological ex- with strict adherence to excision margin guidelines and a
amination of the excised sample, a recurrence rate of less re-treatment protocol, lesions with larger surface areas and
than 2% can be achieved.14 However, recurrence of BCC after deeper invasion are no more likely to recur.
primary excision typically ranges between 2 and 10%, with the The incidence of perineural invasion and lesion ulcera-
difference largely depending on the re-treatment protocol tion was compared between recurrent and non-recurrent
employed and use of adjunct treatment modalities.6e8,17 Out lesions, with no significant difference found between the
of 331 lesions excised in this study, only 10 (3%) lesions were groups. This finding contradicts previous research, which
found to recur within the follow-up period, representing a has linked the presence of these two factors with greater
relatively low recurrence rate within the literature. rates of recurrence.16 Although ulceration and perineural
Previous research has suggested an increased risk for BCC invasion may indicate the aggressive nature of a lesion,
recurrence in males and with increasing age.8 Although not with adequate excision margins and the use of radiotherapy
analysed in this study, increasing age may be associated with adjunctively, the results of this study suggest that these
an increased incidence of co-morbidities and impaired factors do not influence the rate of lesion recurrence.
immunological status, ultimately influencing tumour aggres- As stated in the author’s previous study analysing the
siveness and recurrence. The results of this study do not role of embryologic fusion planes and recurrence, excision
support this, as no significant difference in recurrence was margins appear to have the greatest impact on lesion
found when comparing the mean patient age of recurrent and recurrence. Recurrent lesions were significantly more likely
non-recurrent lesions (p Z 0.353). There was also no signifi- to have close or incomplete peripheral and deep excision
cant difference in gender between the groups (p Z 0.378). margins than non-recurrent lesions (p Z 0.013 and
Sclerosing, infiltrative and micronodular lesions have p Z 0.048, respectively).
been associated with increased deep subclinical exten-
sions, higher rates of positive margins after primary exci-
sion and ultimately higher rates of recurrence (13e15). The Study limitations
results of this study support the existing literature, as a
significantly greater percentage of recurrent lesions was With only 10 lesions recurrences identified within the
found to have micronodular and infiltrative subtypes than observation period, the sample may not be sufficiently

Downloaded for andromeda masako (andromeda72@yahoo.com) at Universitas Sumatra Utara from ClinicalKey.com by Elsevier on October 12, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
1744 L.T.D. Armstrong et al.

powered to confidently compare and analyse several of the entity with any financial interest (such as honoraria;
risk factors between the two groups. The fact that a low educational grants; participation in speakers’ bureaus;
recurrence rate was achieved for the participant cohort membership, employment consultancies, stock ownership,
highlights the effectiveness of the clinical excision margins or other equity interest; and expert testimony or patent-
and re-treatment protocol. Even with larger numbers of licensing arrangements), or non-financial interest (such as
lesions containing features previously deemed high risk, a personal or professional relationships, affiliations, knowl-
low recurrence rate was still maintained. Lesions that had edge or beliefs) in the subject matter or materials dis-
not yet recurred or recurred but not been subsequently cussed in this manuscript.
treated within the observation period (70e80 months)
would not have been identified and included in the study
analysis. However, this number is expected to be quite low, STROBE statement
with research showing greater than 80% of lesions having
recurred by this time.7 It is also possible that a lesion was This study was designed and prepared in line with the
excised by a different physician and sent to a pathology STROBE statement/guideline for cohort studies. It adheres
laboratory not included in the search for recurrence. to all components of the STROBE checklist for cohort
However, the risk of this occurring was greatly mitigated, as studies; available at: http://strobestatement.org/
the treating surgeon was the only plastic surgeon present in fileadmin/Strobe/uploads/checklists/STROBE_checklist_
the town and surrounding region at the time, and a thor- v4_combined_PlosMedicine.pdf.
ough search of state-wide databases of all major pathology
laboratories operating in the region at that time was con-
ducted (Pathology Queensland, Sullivan & Nicolaides & QML References
Pathology).
1. Lozano R, Naghavi M, Foreman K, et al. Global and regional
mortality from 235 causes of death for 20 age groups in 1990
Conclusions and implications and 2010: a systematic analysis for the Global Burden of Dis-
ease Study 2010. Lancet 2012;380(9859):2095e128.
The rate of lesion recurrence for the participant cohort (3%) 2. Lear W, Dahlke E, Murray CA. Basal cell carcinoma: review of
is relatively low within the literature particularly for lesions epidemiology, pathogenesis, and associated risk factors. J
confined to the face. Peripheral and deep excision margins, Cutan Med Surg 2006;11(1):19e30.
invasive histological subtypes (infiltrative and micro- 3. Subramaniam P, Olsen CM, Thompson BS, Whiteman DC,
nodular) and previous recurrence were identified as strong Neale RE, for the QS, et al. Anatomical distributions of basal
cell carcinoma and squamous cell carcinoma in a population-
risk factors for BCC recurrence on the face. Lesions surface
based study in Queensland, Australia. JAMA dermatology 2017;
area, depth of invasion, perineural infiltration, ulceration,
153(2):175e82.
patient age and gender were not identified as significant 4. Ting PT, Kasper R, Arlette JP. Metastatic basal cell carcinoma:
risk factors. Although depth of invasion, surface area, report of two cases and literature review. J Cutan Med Surg
perineural infiltration and ulceration may indicate the 2005;9(1):10e5.
aggressive nature or extent of a lesion, the results of this 5. Bath FJ, Bong J, Perkins W, Williams HC. Interventions for basal
study suggest that with adequate excision margins and use cell carcinoma of the skin. Cochrane Database Syst Rev 2003;2:
of an effective re-treatment protocol these factors may not CD003412.
influence recurrence rate. Infiltrative and micronodular 6. Cullen FJ, Kennedy DA, Hoehn JE. Management of basal cell
subtypes are believed to be at increased risk of recurrence carcinoma: current concepts. Adv Plast Surg 1993;10(187).
7. Rowe DE, Carroll RJ, Day Jr CL. Long-term recurrence rates in
primarily due to deep invasion and spread of silent sub-
previously untreated (primary) basal cell carcinoma: implica-
clinical tumour extensions, predisposing them to inade-
tions for patient follow-up. J Dermatol Surg Oncol 1989;15(3):
quate primary excision and tumour being inadvertently 315e28.
missed in the sample preparation process. One of the 8. Silverman MK, Kopf AW, Bart RS, Grin CM, Levenstein MS.
strongest risk factors for recurrence was a lesion having Recurrence rates of treated basal cell carcinomas. Part 3:
already recurred after previous excision, and it suggested surgical excision. J Dermatol Surg Oncol 1992;18(6):471e6.
that these lesions be flagged and treated with particular 9. Armstrong LTD, Magnusson MR, Guppy MPB. The role of
caution. A more regular follow-up protocol for these lesions embryologic fusion planes in the invasiveness and recurrence
to re-assess for early recurrence may decrease time to re- of basal cell carcinoma: a classic mix-up of causation and
treatment and improve morbidity for this group. The au- correlation. Plastic Reconstr Surg e Glob Open 2015;3(12):
e582.
thors suggest biopsy-proven infiltrative and micronodular
10. Gulleth Y, Goldberg N, Silverman RP, Gastman BR. What is the
BCC subtypes be excised with more liberal peripheral and
best surgical margin for a Basal cell carcinoma: a meta-analysis
deep margins to aid complete removal of subclinical of the literature. Plast Reconstr Surg 2010;126(4):1222e31.
tumour spread and have a lower threshold for re-excision 11. Rowe DE, Carroll RJ, Day CL. Mohs surgery is the treatment of
based on pathological excision margins. choice for recurrent (previously treated) basal cell carcinoma.
J Dermatol Surg Oncol 1989;15(4):424e31.
12. Helsing P, Kramer P, Haugstvedt A, et al. [Surgical treatment of
Conflict of interest statement basal cell carcinoma]. Tidsskr Nor Laegeforen 2004;124(21):
2740e2.
All three authors of this study (Dr. Linus Armstrong, Dr. Mark 13. Hendrix Jr JD, Parlette HL. Micronodular basal cell carcinoma:
Magnusson and Dr. Michelle Guppy) certify that they have a deceptive histologic subtype with frequent clinically unde-
NO affiliations with or involvement in any organization or tected tumor extension. Arch Dermatol 1996;132(3):295.

Downloaded for andromeda masako (andromeda72@yahoo.com) at Universitas Sumatra Utara from ClinicalKey.com by Elsevier on October 12, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.
Analysis of the risk of facial BCC recurrence after excision 1745

14. Lang Jr PG, Maize JC. Histologic evolution of recurrent basal 16. Osguthorpe JD, Abel CG, Lang P, Hochman M. Neurotropic
cell carcinoma and treatment implications. J Am Acad Der- cutaneous tumors of the head and neck. Arch Otolaryngol
matol 1986;14(2):186e96. Head Neck Surg 1997;123(8):871.
15. Sexton M, Jones DB, Maloney ME. Histologic pattern analysis of 17. Smeets NWJ, Krekels GAM, Ostertag JU, et al. Surgical excision vs
basal cell carcinoma: study of a series of 1039 consecutive Mohs’ micrographic surgery for basal-cell carcinoma of the face:
neoplasms. J Am Acad Dermatol 1990;23(6):1118e26. randomised controlled trial. Lancet 2004;364(9447):1766e72.

Downloaded for andromeda masako (andromeda72@yahoo.com) at Universitas Sumatra Utara from ClinicalKey.com by Elsevier on October 12, 2018.
For personal use only. No other uses without permission. Copyright ©2018. Elsevier Inc. All rights reserved.