Corresponding Author: Suk-Woo Yang, MD, PhD. Department of Ophthalmology and Visual Science, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 222 Banpo-daero, Seocho-gu, Seoul 06591, Korea. Eyelid malignancy has an average annual incidence of 0.46 per 100,000 people in South Korea based on the Korea Central Cancer Registry study [1]. Basal cell carcinoma (BCC) has been reported to be the most common malignant eyelid tumor followed by squamous cell carcinoma (SCC) and sebaceous gland carcinoma (SGC) in the Caucasian populations [2]. In the Asian population, SGC has been noted to be the most common or second common tumor [3]. Early detection based on clinical exams and histopathological results plays an important role to prevent potential tumor invasion to the orbit, sinuses and brain or metastasis.
The treatment is mainly surgical removal to completely excise the malignancy and to decrease the rate of recurrence [4]. Nevertheless, surgical reconstructive procedures are challenging for repair from other cutaneous sites and the main goal is to restore the aesthetic and functional purpose of the eyelid. Various reconstructive techniques have been described in literature, but their long-term outcomes and suitability for different tumor types and locations remain areas of active research [5].
This study aims to bridge the gap in our understanding of eyelid malignancies in the Asian population by presenting a comprehensive, 17-year retrospective analysis of clinical characteristics, surgical management, and long-term outcomes. By examining the relationship between tumor types, locations, reconstructive techniques, and postoperative complications, we seek to provide evidence-based guidance for optimal treatment strategies. Furthermore, we investigate prognostic factors for local recurrence and the need for orbital exenteration, offering valuable insights for risk stratification and patient counseling.
Materials and Methods
Ethics statement
This study was approved by the Institutional Review Board of Seoul St. Mary’s Hospital (No. KC25RISI0041), which served as the central review board for the study. The requirement for informed consent was waived due to the retrospective nature of the study. The study adhered to the principles outlined in the guidelines of the Declaration of Helsinki.
Study design and patients
A retrospective chart review was performed on patients diagnosed with eyelid malignant tumor in Department of Ophthalmology at Seoul St. Mary’s Hospital (Seoul, Korea) from 2004 to 2021. Data include patient age, sex (male, female), clinical data on tumor size, location, and laterality (right, left), degree of local extension as well as the presence of regional lymph node metastasis either by aspiration biopsy or imaging, or distance metastasis. Computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) scans were evaluated to assess the tumor stage accordingly. Information on surgical reconstructive procedures, additional treatments, histopathological results, and postoperative complications, outcomes, and local recurrences were also collected. The follow-up time was calculated from the date of surgical therapy. Tumor size was classified based on greatest dimension from pathological data (0-10, 11-20, 21-30, and 31-40 mm). However, to enhance statistical power and clinical relevance, we ultimately dichotomized tumor size into two groups for our main analyses: ≤10 and >10 mm.
Surgical procedure
All surgical procedures were performed by one surgeon (SWY). A complete eyelid tumor resection procedure was performed 3 mm from visible tumor edge and proceed with reconstruction after confirmed negative margin by frozen biopsy from all margin. Partial thickness excision if the tumor is more than 4 mm from the lid margin and not fixed to tarsus. Depending on the tumor location, extend of eyelid defect and lid laxity, variety of techniques were done. For small to medium defects of both the upper and lower eyelids, direct closure was performed if there was sufficient lid laxity. If not, a Tenzel semicircular flap was used for central defects, or a sliding tarsoconjunctival flap was applied for medial or lateral defects. It was combined in few cases with lateral cantholysis. or with periosteal flap to reconstruct the posterior lamella. For larger defect more than 50% modified Hughes procedure was used in the lower eyelid, and reverse modified Hughes procedure and Cutler-Beard procedure (CBP) in the upper eyelid. CBP with an acellular dermis graft (SureDerm, Hans Biomed Co), and periosteum apposition with SureDerm was performed in two cases to reconstruct posterior lamella extensive defect. Second-stage separation for flap division with lid margin repair was done at 7 to 9 weeks in modified Hughes procedure, reverse modified Hughes procedure, and CBP. Glabellar flap was performed to repair medial canthus defect. Adjacent musculocutaneous flap was done in non-full thickness anterior lamella lid defect. All of the surgical procedures were performed as previously described [5-7].
Statistical analysis
We conducted all statistical analyses using IBM SPSS ver. 27.0 (IBM Corp). For comparing group characteristics, we used chi-square or Fisher exact tests for categorical variables, and independent t-test or Mann-Whitney U-test for continuous variables, as appropriate. To identify factors associated with eyelid malignancy recurrence, we performed univariate and multivariate logistic regression analyses. We considered p < 0.05 as statistically significant throughout our analysis.
Results
A total of 152 patients diagnosed with eyelid malignancies between 2004 and 2021 were included in this study. Initially 143 patients underwent frozen section-guided tumor excision with simultaneous eyelid reconstructive surgery, and the other remaining 9 patients were subjected to exenteration. The clinical characteristics of patients with eyelid carcinoma based on histopathologic type and its categorization according to the American Joint Committee on Cancer (AJCC) Staging Manual, 7th edition, and number of recurrences are presented on (Table 1). The mean age of the patients was 69.17 years (range, 29-90 years), with a female predominance (68.7%). The most common eyelid malignancy was BCC, accounting for 52.6% of cases, followed by SGC at 32.2% and SCC at 13.8%. Notably, BCC predominantly affected the lower eyelid (71.3%), while SGC and SCC were more frequently located on the upper eyelid (63.3% and 52.4%, respectively). The tumor was located on the left side in 78 cases (51.3%) and on the right in 74 (48.7%). In this study there was one patient with melanoma and another one with mucinous eccrine carcinoma involving the upper eyelid. Actinic keratosis was noted in four patients that were associated with BCC. Bowen disease and Bowenoid actinic keratosis were noted in two patients with SCC. No exposure to carcinogenic substances or radiotherapy is noted. Metastasis was noted to parotid gland, lung, and kidney.
In terms of tumor size, the majority of carcinomas were less than 20 mm in diameter (91.4%), while 13 patients (8.6%) presented with larger tumors (>20 mm). Lymph node metastasis was detected in four patients (8.2%) with SGC and one patient (4.8%) with SCC, while distant metastasis was less common, observed only in a few cases. Tumor recurrence occurred in 13 patients (8.6%) after initial excision, with a median time to recurrence of 16.9 months. Notably, a total of 13 patients underwent orbital exenteration due to extensive tumor invasion, including four cases of BCC, five cases of SGC, three cases of SCC, and one case of malignant melanoma. Further surgical intervention was required in nine of these recurrent cases, while the remaining four patients were treated with exenteration. When analyzing prognostic factors for local recurrence, distant metastasis, lymph node involvement, and positive intraoperative frozen section margins were significant predictors in univariate analysis. In multivariable logistic analysis, patients with lymph node metastasis had a markedly higher risk of recurrence (odds ratio [OR], 21.291; 95% confidence interval [CI], 2.593-174.817; p = 0.004), as did those with positive frozen section margins (OR, 7.083; 95% CI, 1.390-36.087; p = 0.018). Conversely, factors such as patient age, sex, tumor size, and the type of reconstructive procedure did not significantly affect recurrence risk (Table 2).
Exenteration due to presence of orbital invasion was more likely associated with tumor location in the medial canthus area (p < 0.001), lymph node metastasis (p = 0.041), and larger tumor size (>10 mm, p = 0.007; all Fisher exact test) as compared with patients who underwent surgical excision (Table 3).
Adjuvant surgical procedures were needed in few cases to reconstruct the defect. The number of combined reconstructive surgery to reconstruct the eyelid after tumor excision are presented in Table 4. The mean postoperative follow-up for tumor removal and lid reconstruction was 42.8 months.
Postoperative complications were relatively infrequent but did occur in certain cases. For example, 12 patients developed mild trichiasis, requiring only minor intervention, and 14 patients experienced lagophthalmos, which was managed conservatively. More serious complications, such as canalicular obstruction and punctum ectropion, required additional surgical correction in a few instances. Importantly, the majority of patients achieved stable eyelid function and satisfactory aesthetic outcomes following their reconstructive procedures (Table 5).
All patients had good eyelid position and the functional effect of surgery on eyelid remained stable. Till date, the follow-up of other patients included in this study not shown any evidence of recurrence. But further follow-up for a longer period is important to detect any recurrence at its earliest stage.
In summary, our findings highlight that while the overall outcomes of eyelid reconstruction following malignancy excision are favorable, careful attention must be paid to risk factors such as distant metastasis, lymph node involvement and positive intraoperative margins at diagnosis, which are strongly associated with local recurrence. Additionally, tumor location in the medial canthus, lymph node metastasis, and larger tumor size (>10 mm) were significantly more frequent in cases requiring orbital exenteration, highlighting the need for vigilant monitoring and early intervention to manage potential tumor invasion.
Discussion
BCC is by far the most common locally invasive malignant eyelid tumor reported in literature, it does account for 90% in Europe. This is followed in incidence by tumor with potential metastasis SCC (5%-10%), then SGC (<5%) and melanoma (<1%) in Europe [2]. But in the Asian population, eyelid SGC has been reported to be the most common or second common tumor [3]. Our study is similar in incident to other Asian descent study. Melanoma and mucinous eccrine carcinoma are rare tumors that account for less than 1% of lid malignancy [8]. The most common tumor location in BCC was the lower eyelid, followed by the inner canthus, and less frequently the upper eyelid and lateral canthus [9]. For SCC, lower lids have been reported as the most frequently involved [10] but in our study was the upper eyelid. In SGC, upper eyelid is the predominant site where most of meibomian gland are located.
In our study the mean age of onset was in the seventh decade of life. All type of tumors were more predominant in women. Advanced age is an important risk factor.
The most common presentation in our study by AJCC Staging Manual, 7th edition among BCC was stage IA (51%), and stage IB for SCC (42.9%) and SGC (30.6%). Local recurrence was seen in patients who had lymph node involvement and distant metastasis. However, there was no correlation found between the tumor size with local recurrence. Another significant risk factor for recurrence was intraoperative initially positive frozen section margin. The intraoperative positive frozen section was mainly cases of SGC. Positive margin was resected until clear margin was achieved and clear tumor resection margin was confirmed on permanent section. The recurrence rate is higher if surgical excision is done without intraoperative clear margin compared to the control frozen section or Mohs micrographic surgery and in cases of SGC, it can be difficult to interpret due to intraepithelial pagetoid spread [11].
In our knowledge, this study is the first to attempt to correlate the frozen section margin in eyelid carcinoma result with risk of recurrence. Given these results, particularly in cases of SGC with Pagetoid spread, even when clear margins are achieved after additional excision following a positive frozen section margin, the possibility of undetected tumor seeding should be considered. Further research is needed to explore this issue. Large tumor size, location in the medial canthal area, and lymph node metastasis were important predictor of orbital invasion. In our study, medial canthal tumor was almost all BCC, and cases with lymph node metastasis was SGC. The overall rate of local recurrence and exenteration in this study were the same (8.6%), and are in line with rates reported previously which range 6% to 24% and 6.3% to 36%, respectively [12-14], being more in SGC.
The initial diagnostic excision and biopsy allows us to know the type of tumor and to plan for therapeutic management according to the results. If the tumor is capable of metastasizing, staging should be carried out. Orbital CT and MRI are the best way to determine if there is tumor invasion to the orbit, bones, sinuses, or the brain. CT of the neck and thorax or MRI can be done to evaluate metastasis. CT is the best to visualize bony involvement. On the other hand, MRI is best to visualize soft tissue. Regional lymph node sonographic guided aspiration biopsy can be done in suspicious lymphatic metastasis [15]. On the other hand, hematogenous metastasis can be detected by PET scan.
Surgical excision is the main standard for management of eyelid malignancy despite the newly developed therapeutic options. Depending on defect size, thickness, location, laxity of the eyelid, and availability of donor graft, different surgical techniques are performed case-by-case to completely remove the tumor with free (controlled) margin and to restore the eyelid position and function for cornea protection and to achieve good aesthetic result [5]. We report different reconstructive techniques used to manage eyelid defect after malignancy removal and their complications. Surgical techniques can be combined with other adjunctive techniques to close large anterior or posterior lamella defect. A second surgical correction procedure can be performed to manage related complications.
Wound defect is assessed in surgery for direct defect closure as it is the optimal procedure to perform if the size of the wound defect less than one-third of the eyelid length. For larger defects, extended more than 50% of the eyelid length, nonmarginal lid sharing technique can be performed to reconstruct the lower lid defect by Hughes procedure and to reconstruct the upper lid defect by CBP or reverse Hughes procedure. One disadvantage of these procedure, however, is the need to occlude the eye for period of time before flap division. Previously reported early flap division at less than 3 weeks showed to be safe in most of the patients [16,17]. Maximal Hughes procedure can be used to reconstruct maximal eyelid defect with excellent result [18]. For moderated defect, no more than 75% of eyelid length, Tenzel semicircular flap can be performed on either upper lid or lower lid. Sliding tarsoconjunctival flap is used to repair medial or lateral defect [19]. The creation of lateral periosteal flap can be used for large defect or involving lateral canthus. SureDerm is an acellular dermal matrix used as substitute graft to reconstruct the posterior lamellae. It was used in two cases because of larger defect size and the need to give more support to decrease risk for eyelid malposition. Other substitute for posterior lamellae such as cartilage, sclera, free tarsus, and commercial dermal matrix can be used [17,20-22].
A total of 11 patients had complications related to eyelid reconstruction in which second correction surgery is needed. Many of the lower eyelid reconstruction procedures has more incidence for lower ectropion due to cicatricial or mechanical skin tension. While upper eyelid reconstructive procedures have more incidence of upper lid entropion. Delaying second-stage division in nonmarginal lid sharing technique 7 to 9 weeks can supplement counteract traction to prevent eyelid malposition especially in less lid laxity and younger patient.
Eyelid defects that involves canaliculus are repaired with conventional silicon tube insertion. If the entire canaliculus and punctum, second surgery may be needed using conjunctivodacryocystorhinostomy with Jones tube insertion. Other minor sequalae did not require surgical revisions. No necrosis was seen in any of the cases due to intrinsic form of vascular supply in the flap. Majority of the patients had good aesthetic outcome with no major irreversible complication.
In conclusion, eyelid carcinoma management starts with the prevention of risk factors and early diagnosis with excision biopsy confirmed with histopathologic result make it possible to avoid further tumor progression. While CT and MRI can be used to evaluate tumor extension, surgical treatment should be individualized to ensure free-margin excision. The surgical approach is determined by the tumor location, extent, and the size of the eyelid defect to restore both function and cosmesis. Complications are infrequent and can be managed if needed by second surgery. There were strong correlations of positive intraoperative frozen section margin and lymph node metastasis with recurrence. There were significant correlations of tumor size >10 mm, lymph node metastasis, and medially located tumor with orbital invasion.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
