Outcomes of Levator Resection for Pediatric Ptosis: A Nine-Year Retrospective Review: An Abstract from Akron Children’s Hospital Resident and Fellow Academic Day

Pediatric blepharoptosis is a condition characterized by abnormal drooping of the upper eyelid. Primarily resulting from dysfunction of the levator palpebrae superioris muscle in congenital forms. It is commonly congenital but may also arise from acquired etiologies. Untreated ptosis can result in visual impairment, including amblyopia, induced astigmatism, and compensatory head positioning, emphasizing the importance of timely intervention. Margin Reflex Distance (MRD) is an essential clinical measurement in the evaluation of ptosis and its postoperative assessment, defined as the distance in millimeters between the corneal light reflex and the eyelid margin, measured with the patient in primary gaze. MRD1 refers to the distance from the corneal reflex to the upper eyelid margin and is the most clinically relevant parameter, while MRD2 measures the distance to the lower eyelid margin. Normally, MRD1 is approximately 4–5 mm, and a reduction in this value directly correlates with the severity of ptosis, allowing it to be classified as mild, moderate, or severe based on how low the upper lid sits relative to the pupil. MRD1 is particularly important in determining whether the visual axis is compromised and in distinguishing true ptosis from pseudoptosis when interpreted alongside other clinical findings. In the postoperative setting, MRD1 serves as the primary objective indicator of surgical success, with an ideal outcome typically around 3–4 mm; values below this suggest undercorrection, and higher values may indicate overcorrection, potentially leading to complications such as lagophthalmos. Additionally, MRD measurements are used to assess eyelid symmetry between both eyes, monitor healing over time, and identify early complications, making MRD a cornerstone metric in both the preoperative planning and follow-up of patients undergoing ptosis correction. Surgical correction depends largely on the degree of levator function (LF). Frontalis sling procedures are commonly employed in patients with poor LF (mostly < 5 mm), whereas levator resection (LR) is preferred when function is moderate to good. Despite being an established technique, LR is less frequently performed in pediatric populations due to anatomical limitations, poor LF, intraoperative challenges under general anesthesia, difficulty predicting postoperative eyelid height, and surgeon experience. Previous studies have reported variable success rates with LR in children, and there remains limited consensus regarding predictors of surgical outcomes. This study aims to contribute to the existing literature by evaluating a relatively large cohort of pediatric patients undergoing external levator resection and identifying factors associated with postoperative outcomes. Institutional review board approval was obtained (2025-164). A retrospective chart review was conducted of pediatric patients who underwent external levator resection between January 2017 and January 2025 at a quaternary care pediatric hospital, pediatric ophthalmology clinic, with data extracted from electronic medical records. Patients with more than 3 months of follow-up were included. Demographic data, including age at presentation, age at surgery, gender, laterality, etiology of ptosis, preoperative and postoperative eyelid examination findings, including LF, interocular eyelid height difference, MRD1, amount of LR, postoperative outcomes, including eyelid height, symmetry, contour, eyelid crease scar, complications, and follow-up time, were recorded. Postoperative interocular eyelid height difference (PoEHΔ) was categorized as: good: ≤1 mm, satisfactory: 1–2 mm, and poor: ≥2 mm. Surgical success (SS) was defined as: a clear visual axis, PoEHΔ ≤1 mm, good eyelid contour, and cosmetically non-significant eyelid crease scar. Undercorrection and overcorrection were defined as a postoperative eyelid position ≥1 mm lower or higher than the fellow eye, respectively. Descriptive statistics were used to summarize patient demographics, clinical characteristics, and surgical outcomes. Continuous variables were reported as mean ± standard deviation (SD). Correlation analyses using Pearson and point-biserial coefficients were performed to evaluate associations between preoperative variables, demographic data, and postoperative outcomes (PoMRD1, PoEHΔ, and SS). Correlation coefficients (r) were interpreted based on effect size as negligible (|r| < 0.10), small (0.10–0.29), moderate (0.30–0.49), or large (≥ 0.50). Results were visualized using a heat map, in which color intensity reflected the magnitude and direction of correlation (positive correlations shown in red, negative correlations in blue, and negligible associations in neutral tones). Statistical significance was defined as p < 0.05. All statistical analyses were performed using R statistical software (R Foundation for Statistical Computing, Vienna, Austria). A total of 60 eyelids from 55 patients were analyzed. There were 17 female (2 bilateral surgery), and 38 male (3 bilateral surgery) patients. A total of 23 right and 27 left eyelids underwent surgery. The mean postoperative follow-up duration was 20.5 ± 25.1 months (range 3 – 84.4). The majority of cases (n=52 eyelids, 86.6%) were congenital ptosis. Three eyelids had ptosis secondary to Marcus Gunn jaw-winking syndrome, two eyelids from a single patient were due to myasthenia gravis, and one eyelid each was attributed to Duane syndrome, congenital third nerve palsy, and traumatic ptosis. The mean age at surgery was 75.8 ± 61.7 months (range 13 months to 18 years). 59 eyelids had a primary LR. Mean LF was 10.7 ± 2.7 mm (range 5 to 17). Mean MRD1 was 0.75 ± 0.78 mm (range (-)1 to (+)3). The amount of levator resection was adjusted intraoperatively based on the observed eyelid lift. The mean LR performed was 13.6 ± 3.5 mm (range 7 to 20). Mean postoperative MRD1 was 2.7 ± 0.7 mm (range 1 to 4.25). Mean postoperative MRD1 was significantly greater than preoperative MRD1 in the operated eyes (p < 0.05). Mean PoEHΔ was 0.7 ± 0.7 mm (range 0 to 2.5). A good PoEHΔ (≤1 mm) was achieved in 82.4% of eyelids (n = 47), while a satisfactory PoEHΔ (<2 mm) was achieved in 92.9% of eyelids (n = 53). 3 patients were excluded from this assessment due to a lack of data. 45 (75%) eyelids achieved SS. While 15 eyelids failed to achieve SS. The most common reasons for failure were undercorrection (n=8 eyelids) and visually significant eyelid crease scar (n=3 eyelids). Mean postoperative follow-up duration was 22.1 ± 25.2 months (range 3 to 84.4). The most common complications included: undercorrection 13.3% (n= 8 eyelids), infection 6% (n=4 eyelids), and visually significant eyelid crease scar 5% (n=3 eyelids). Overcorrection 3% (n=2 eyelids), reoperation for ptosis 3% (n=2 eyelids), mild nighttime lagophthalmos 3% (n=2 eyelids), entropion 1.6% (n=1 eyelid), trace poor eyelid contour 1.6% (n=1 eyelid), and double eyelid crease 1.6% (n=1 eyelid) were observed. All patients who developed infection responded well to oral antibiotics; one patient required admission with intravenous antibiotics. One patient underwent frontalis sling with a silicone rod, and one required repeat LR. No sight-threatening complications were observed. Greater LR dose correlated with higher PoMRD1 (p = 0.041). Male gender was associated with slightly lower PoMRD1 compared to female (p = 0.025). There was no association between the preoperative LF, preoperative MRD1, age at surgery, laterality, or gender and PoMRD1 (p>0.05). Correlation analysis revealed uniformly weak associations between preoperative factors and postoperative outcomes. Laterality and gender showed small negative correlations with PoEHΔ, while preoperative LF, MRD1, and age at surgery demonstrated small positive correlations. LR dose exhibited the strongest relationship, with a moderate positive correlation with PoEHΔ and a small association with SS. Notably, no variables demonstrated meaningful associations with SS, and none reached statistical significance (p > 0.05), underscoring the limited predictive value of preoperative factors. This study demonstrates that external levator resection provides favorable outcomes in pediatric patients with ptosis and adequate LF. A high proportion of patients achieved satisfactory eyelid symmetry, supporting LR as a viable alternative to frontalis sling procedures in selected cases. The absence of strong correlations between preoperative parameters and postoperative symmetry highlights the complexity of pediatric ptosis surgery. Intraoperative adjustments appear to play a critical role, likely reflecting variability in levator muscle integrity and tissue response. These findings are consistent with previous studies suggesting that standardized surgical tables may serve as guidelines but cannot fully predict outcomes. The observed rate of undercorrection is comparable to published literature and may reflect the conservative intraoperative approach often adopted to minimize overcorrection. More importantly, most of the patients who had undercorrection had a clear visual axis and did not require a secondary intervention. Over or undercorrection is reported as a complication of currently used surgical techniques in ptosis correction. This may be due to the degree of cooperation during the preoperative examination, resulting in inaccurate results. Infection rates were low and manageable with no permanent functional or cosmetic adverse events. Interestingly, increased resection correlated with improved MRD1, supporting the dose-response relationship. The association between male gender and slightly lower postoperative MRD1 requires further investigation and may be incidental. Limitations of this study include its retrospective design, variability in follow-up duration, and the potential for measurement error in eyelid assessment due to the challenges of examining a complex pediatric population, particularly infants