Primary basal cell carcinoma (BCC) is the most common periocular malignant tumor.1 Periocular tumors present therapeutic challenges due to the need for radical surgical excision while preserving eyelid function and facial aesthetics.

Although established guidelines recommend long-term follow-up for patients diagnosed with BCC due to the high risk of developing new lesions,1 Brazil lacks a national registry for skin or periocular cancers. Available data are derived from small hospital-based studies, and national guidelines do not include standardized long-term follow-up protocols, leaving monitoring decisions to individual clinicians.

To propose a tailored follow-up strategy for the Brazilian context, we retrospectively reviewed patients with primary periocular BCC who underwent surgical excision at our university hospital between March 2012 and December 2019 (Ethics approval: 39820420.1.0000.5411). All included patients who had complete excision of primary periocular BCC, confirmed histological subtype, documented margin status, and at least 12-months of follow-up. Non-periocular BCCs, recurrent tumors, prior radiotherapy or cryotherapy, immunosuppression, Gorlin-Goltz syndrome, and incomplete records were excluded.

We analyzed 135 histologically confirmed periocular BCC lesions from 116 patients, averaging 19.28 lesions treated annually. Periocular BCC predominantly affected elderly individuals, with a mean age of 69.47 ± 12.16 years old (range: 37–93) and a median age of 72. This trend likely reflects cumulative sun exposure, reduced DNA repair capacity, and age-related decline in immune response.2

Of the patients, 63 (54.31%) were female, with no significant gender difference in the occurrence of BCC. A higher incidence of periocular BCC in women has already been reported.3 The vast majority of patients (111/95.69%) had phototypes I or II, reinforcing the established link between BCC and lighter pigmentation, especially in individuals chronically exposed to sunlight. In contrast, darker skin provides greater UV protection and is associated with lower BCC risk.4

The mean tumor size was 7.23 ± 4.96 mm (range: 2–40 mm; median: 8 mm). The lower eyelid was the most affected site (110/81.48%), followed by the medial canthus (11/8.15%), upper eyelid (7/5.19%), lateral canthus (6/4.44%), and eyebrow (1/0.74%), consistent with previous studies.5,6 The lower eyelid and medial canthus are more exposed to UV radiation and inflammatory agents such as tear components. The upper eyelid, protected by the brow and orbital anatomy, showed a lower frequency. No significant laterality effect was noted, although some studies suggest a higher incidence on the side of the face more exposed during driving.7

Tumors were generally conventionally excised under local anesthesia with 2–5 mm safety margins, depending on lesion size, consistent with other reports.8 Wider excision is recommended for aggressive subtypes of tumors (sclerodermiform, mixed), though it may compromise cosmetic and functional outcomes.9

In our clinic, frozen section or surgical excision with a safety margin is reserved for lesions with indistinct borders. This differs from other approaches that use frozen sections routinely or opt for wide local excision or Mohs surgery.1

The most common histological subtype was nodular (59/43.70%), followed by mixed (52/38.52%) (Table 1). Perineural invasion occurred in only one case (1/0.74%), consistent with other studies.5,6,10

We observed nine recurrences: 6 (66.67%) occurred in lesions with clear margins, underscoring that negative margins do not eliminate recurrence risk; 3 (33.33%) recurrences were from lesions with compromised margins. Of the 14 cases with involved margins, only 3 (21.43%) recurred. Reported recurrence rates for positive margins vary widely,5,8,9 depending on histological subtype, surgical excision type, and follow-up duration.1

Recurrences in our cohort were significantly associated with histological subtype (p = 0.037), being more frequent in nodular (3/33.33%), mixed (2/22.22%), and sclerodermiform (2/22.22%) subtypes. While nodular BCC is generally well-defined and amenable to complete excision, its recurrence rate in our sample likely reflects its high prevalence rather than intrinsic aggressiveness. In mixed BCC, the lesion may be composed of either high-risk or low-risk subtypes, making it important for the histological report to specify which subtypes are present in the lesion. Conversely, aggressive subtypes such as sclerodermiform are known to pose greater risks of incomplete excision and recurrence. Most recurrences occurred in the lower eyelid (8/88.89%) or in the medial canthus (1/11.11%), which were also the most common tumor locations. No significant association was found between recurrence and tumor size or margin status. Tumor-free surgical margins were achieved in 110 (81.48%) lesions, while compromised margins were identified in 14 (10.37%) cases. Twelve (8.89%) cases had indefinite margins; a known possibility related to the quality of some specimens.

The most frequently affected margin was the deep plane (11/42.31%), like other reports.1 Although BCC typically exhibits limited depth invasion, meticulous excision of the deep portion remains essential. In the medial canthus, erroneous efforts to preserve the lacrimal drainage system may inadvertently result in incomplete excision. Nevertheless, complete removal must remain the surgical goal to prevent orbital invasion and associated morbidity.

The mean follow-up duration was 3.41 ± 2.68 years (median: 2.84). Most patients (88/75.86%) remained in follow-up. Thirty-nine (33.62%) were lost to follow-up after two years, and 4 (3.45%) died from unrelated causes. No BCC metastases were reported. Table 2 presents recurrence according to histological subtype, committed margins and time to observe lesion recurrence. Mean time to recurrence was 2.92 ± 2.37 years (range: 1.09–6.25; median: 1.39-years).

According to other studies, periocular BCC recurrence rates are influenced by tumor location, size, histological subtype (more aggressive subtypes include micronodular, infiltrative, sclerosing, and morphea), margin status, immune function, systemic diseases, prior recurrences, and adjuvant treatments.6,10

The approach to incomplete excision of BCC remains controversial. Since not all cases of low-risk tumors will recur,9 we recommend that re-excision should be evaluated on a case-by-case basis, considering factors such as high-risk anatomical location, involvement of deep or lateral margins, and the patient’s ability to adhere to follow-up.

Histologically compromised margins or incomplete lesion removal are not definitive predictors of recurrence. Recurrence may not occur even after several years, and clinical observation may be appropriate. Possible explanations for the absence of recurrence include devitalization of residual tumor cells by postoperative inflammation, containment within excision margins, differences in tumor biology, surgical technique, clinical management, or other biological mechanisms not yet fully understood. However, although re-excision is not necessary in all cases, it is strongly recommended for positive margins, particularly in high-risk histological subtypes, with meticulous excision of the affected margin.

Postoperative follow-up of excised periocular BCCs represents a significant workload for clinicians, and institutional protocols vary. For completely excised nodular BCC, recurrence risk is low, and follow-up for at least one year is recommended.

For aggressive subtypes, follow-up should extend to a minimum of five years, while recurrent tumors require indefinite surveillance. Approximately 30% of recurrences occur within the first year, and up to 82% within five years, reinforcing this period as the critical window for monitoring.

However, the purpose of follow-up is not only to detect local recurrence but also to identify new BCCs. Individuals who have already had one carry a 30%–40% risk of developing another,9 and therefore the patient requires close surveillance.