Clinical features, treatment outcomes, and risk factors for surgery in pediatric chalazion

Abstract

Purpose:

To evaluate the clinical presentation, treatment outcomes, and baseline predictors of surgical intervention in a large cohort with pediatric chalazion.

Methods:

This retrospective observational study included children (<18 years) diagnosed with chalazion between January 2017 and December 2024 with a minimum follow-up of 21 days. Initial treatment consisted of hot fomentation and topical antibiotics. Clinical resolution was defined as complete disappearance of the lesion without the need for surgery. Survival analysis and Cox regression models were used to identify risk factors for surgery.

Results:

A total of 471 eyes from 470 children were included (median age: 6 years). Of the 414 eyes initially managed conservatively, 380 eyes (91.8%) resolved without surgery within the first 6 weeks. Surgical intervention was required in 90 eyes (19%), including 56 advised at baseline. Kaplan–Meier analysis showed that among those requiring surgery, most underwent the procedure within 60 days of presentation (cumulative risk at 2 months = 23%; 95% CI: 17–29%). Children who underwent surgery were significantly older (9.7 ± 4.9 vs. 6.9 + 4.4 years, P < 0.001) and had longer symptom duration (median 60 vs. 30 days, P < 0.001) at baseline. In multivariable Cox regression, only increasing age independently predicted surgery (HR: 1.10; 95% CI: 1.04–1.17; P = 0.0002).

Conclusion:

Most pediatric chalazia resolve within 6 weeks of conservative treatment. Increasing age at presentation was the strongest predictor of surgical intervention. These findings support a time-bound approach to conservative management and highlight the need for prospective studies exploring adjunctive therapies and systemic risk factors.

Chalazia are commonly encountered in pediatric ophthalmic practice but pose unique clinical challenges compared to adults. While almost always benign, pediatric chalazia exhibit myriad presentations, from classic painless nodules to atypical inflammatory swellings. These differ significantly from their adult counterparts due to age-specific eyelid anatomy, variations in immune response, and the potential for visual impairment through astigmatism or amblyopia in children.[1,2,3,4,5] Moreover, when surgery is required, general anesthesia is often necessary in children, adding risk and complexity to an otherwise straightforward procedure.[6,7]

Management typically follows a step-ladder approach, beginning with conservative therapy, that is, hot fomentation, lid hygiene, and topical antibiotics or steroids followed by intralesional corticosteroids or incision and curettage in nonresolving cases.[6,8,9] While this progression is widely accepted, the optimal duration for conservative treatment before surgical escalation remains unclear. Most existing studies are small case series, with limited statistical power and variable follow-up, offering little guidance on when to escalate to surgery or how to identify children at higher risk of requiring it.

Furthermore, no prior study has evaluated baseline predictors of surgical intervention using survival analysis or hazard modeling, and most lack sufficient sample sizes to permit multivariable adjustment. This limits clinicians’ ability to personalize management or counsel caregivers effectively. In this context, we conducted a large, real-world cohort study to (i) describe the clinical characteristics and treatment outcomes of pediatric chalazia, (ii) compare those who resolved versus did not resolve with conservative management, and (iii) identify baseline risk factors associated with the need for surgery using time-to-event analysis. This study aims to fill key knowledge gaps and provide a better foundation for more structured, evidence-based management strategies for pediatric chalazion.

Methods

This was a retrospective, observational study conducted at a tertiary eye care center in western India between January 2017 and December 2024. The study protocol adhered to the tenets of the Declaration of Helsinki and was approved by the institutional ethics committee. We included all children under 18 years of age diagnosed with chalazion who completed at least 21 days of follow-up after their initial presentation. Patients were excluded if they had other significant ocular pathologies such as blepharokeratoconjunctivitis, congenital eyelid malformations, dermatological conditions known to mimic chalazion, including idiopathic facial aseptic granuloma (IFAG), rosacea, or sebaceous gland disorders, or a history of systemic immunosuppression or chronic inflammatory disease.

At baseline, all children underwent a comprehensive ophthalmic examination, including uncorrected and best-corrected visual acuity, cycloplegic refraction, ocular alignment (cover–uncover and prism tests), and detailed eyelid evaluation. Eyelid findings were documented with respect to the number (single vs. multiple), size, location (upper vs. lower vs. both lids) of chalazia, and coexisting lid conditions such as internal hordeolum or styes.

The initial institutional treatment protocol consisted of hot fomentation applied at least twice daily for a minimum of 21 days, combined with topical antibiotic and/or steroid eye drops, as determined by the clinician. Some children with a history of a long-standing chalazion (typically >6–8 weeks) were offered early surgical intervention (incision and curettage) at baseline. At the first follow-up visit (after 21 days), treatment was adjusted based on clinical response. Subsequent decisions to either continue conservative therapy or escalate to surgical intervention were guided by lesion nonresolution beyond 3–6 weeks. The timing of surgery varied based on the persistence of the lesion, parental preference, and logistical considerations, such as the availability of general anesthesia.

Outcomes were tracked across follow-up visits, with special attention to whether the lesion had resolved, persisted, or required surgical management. For the purposes of this study, resolution of chalazion was defined as the complete clinical disappearance of the lesion, that is, no visible or palpable eyelid nodule, with or without residual pigmentation or lid thickening, and without the need for surgical intervention, as noted in case files. This assessment was made at or after the first follow-up visit (minimum 21 days post-presentation) by the treating ophthalmologist. Lesions that showed partial improvement but remained clinically evident or symptomatic were classified as nonresolved.

Statistical analysis

Continuous variables were summarized as mean ± standard deviation (SD) or median with interquartile range (IQR), as appropriate based on distribution. Categorical variables were presented as frequencies and percentages. Group comparisons between (a) patients who resolved versus did not resolve with conservative treatment and (b) those who underwent surgery versus those who did not were performed using the independent samples t-test or Mann–Whitney U test for continuous variables and the Chi-square test or Fisher’s exact test for categorical variables.

To identify baseline factors associated with spontaneous resolution of chalazion, univariate logistic regression analyses were conducted for each candidate predictor variable. Variables with P < 0.10 in univariate analysis were included in a multivariable logistic regression model to determine independent predictors of resolution. Odds ratios (ORs) with 95% confidence intervals (CIs) and corresponding P values were reported.

Time-to-event analysis was performed to assess risk factors for surgical intervention. The primary outcome was time from presentation to surgery; patients who did not undergo surgery were censored at their last recorded follow-up. Kaplan–Meier survival curves were generated to estimate the probability of remaining surgery-free over time.

Cox proportional hazards regression models were used to identify factors associated with time to surgery. Univariate Cox models were initially fitted for each covariate. Variables with P < 0.10 in univariate analysis were considered for inclusion in the multivariable Cox model. Hazard ratios (HRs) with 95% CIs and P values were reported. Model diagnostics showed no issues with multicollinearity (variance inflation), interaction, or proportional hazards assumption (Schoenfeld residuals). Influence diagnostics were performed to identify potential outliers or influential observations.

All analyses were performed using R (version 4.3.2) with the Julius AI interface. All statistical tests were two-sided, and a P value <0.05 was considered statistically significant.

Results

We included a total of 471 eyes from 470 children in the study. The median age of presentation was 6 years (IQR = 3.5–11 years), and 52% were boys. At presentation [Table 1], right eye involvement was observed in 42%, left eye involvement in 45.9%, and bilateral involvement in 12.1% of cases [Fig. 1]. The upper eyelid was affected in the majority (56.9%), followed by the lower lid (33.3%), with involvement of both lids in 5.9%.

Table 1.

Baseline demographics and clinical characteristics of 471 children with chalazion

Variable		Value
Age		7.47±4.64; 6 [3.45, 11]
Gender		Boys: 246 (52.2); Girls: 225 (47.8)
Eye		Right Eye: 198 (42); Left Eye: 216 (45.9); Both Eyes: 57 (12.1)
Lid		Upper Lid: 268 (56.9); Lower Lid: 157 (33.3); Both Lids: 28 (5.9)
Chief complaints		Boil/s: 163 (34.6); Lid Swelling: 113 (24); Nodular Swelling: 169 (35.9); Mass On Lid: 2 (0.4)
No of Lesions		Single: 443 (95.8); Multiple: 20 (4.2)
Durn of chief complaint (Days)		86.45±155.52; 30 [9.5, 90]
Eye pain		38 (8.1)
Recurrent Chalazion at baseline		44 (9.3)
Spectacle use		45 (9.6)
Previous Ophthalmologist Consult		90 (19.1)
Uncorrected Distance Visual Acuity		0.12±0.15; 0 [0, 0.2]
Uncorrected Near Visual Acuity		0.1±0.05; 0.1 [0.1, 0.1]
Spherical Power		-0.08±0.96; 0 [0, 0]
Cylindrical Power		-0.48±0.83; 0 [-0.56, 0]
Corrected Distance Visual Acuity		0.08±0.1; 0 [0, 0.2]
Corrected Near Visual Acuity		0.1±0.05; 0.1 [0.1, 0.1]
Lids Findings		Blepharitis: 1 (0.2); Burst Chalazion: 3 (0.6) Chalazion: 455 (96.6); Internal Hordeolum: 6 (0.9); Lid Edema: 1 (0.2); Localised Swelling: 1 (0.2); Nodular Swelling: 1 (0.2); Stye: 4 (0.7)
Hot Fomentation advised		395 (83.9)
Glass Advised at baseline		90 (19.1)
Surgery at baseline		I & C: 56 (11.9)

Continuous variables: mean+SD; median (Interquartile range), Categorical variables: Category n (%)

Figure 1.

Clinical photographs of a pediatric patient showing multiple eyelid lesions. (Top) Primary gaze demonstrating chalazion on the right upper and left lower lids and bilateral lower-lid hordeolum. (Bottom) Downgaze view of the same child, highlighting the extent of chalazion on the right upper lid

Regarding clinical symptoms, boil-like lesions were the most common complaint (34.6%), followed by lid swelling (24%). Other complaints included pain, foreign body sensation, or discharge, each seen in a smaller subset of patients (<1%). Single lesions were observed in 93.2% of patients, with multiple chalazia in only 6%. The median duration of symptoms before presentation was 30 days, with a wide interquartile range (9.5 to 90 days), and eye pain was reported in a minority of patients. A history of recurrent chalazia was noted in a limited proportion at baseline.

Notably, 56 eyes were advised and underwent I and C surgery during their first visit. These eyes had a significantly longer history of lid swelling (median = 75 days (IQR = 12) vs 30 (IQR = 52), P < 0.001) compared to those who were treated conservatively at the first visit. Initial conservative management consisted primarily of hot fomentation and antibiotic eye drops, which was recommended in nearly all children [Table 1]. Steroid eye drops were prescribed in 8% (n = 38) cases. The minimum prescribed duration of conservative therapy was 21 days.

Comparison of resolved versus nonresolved eyes

Among eyes managed conservatively, baseline characteristics between those that resolved (n = 302) and did not resolve (n = 113, 27%) showed no significant differences across most parameters, including age, gender, presenting symptoms (boils, lid swelling, nodular swelling), laterality, lid involvement, or recurrence status [Table 2]. While children in the nonresolved group had a slightly higher mean age (7.44 vs. 6.99 years) and a greater proportion had multiple lesions, neither difference was statistically significant [Table 2]. However, the mean spherical power, though minimal, was slightly more myopic in the nonresolved group (−0.25 vs. 0.01 D; P = 0.004). Univariate logistic regression showed nonsignificant trends toward nonresolution with increasing age and spectacle use (proxy for refractive error), but no factor reached statistical significance.

Table 2.

Comparison between those resolved versus not resolved with conservative management at baseline

Variable		Category		Resolved (n=302)		Nonresolved (n=113)		P
Age (Years)				6.99 (4.49)		7.44 (4.38)		0.23
Chief complaints Duration (days)				80.09 (158.91)		72.87 (117.71)		0.66
Chief complaints		Nodular Swelling		104 (36.2%)		42 (40.4%)		0.84
		Lid Swelling		80 (27.9%)		27 (26%)
		Boil/s		102 (35.5%)		35 (33.7%)
		Mass On Lid		1 (0.3%)		0 (0%)
Eye		Right Eye		146 (48.3%)		48 (42.5%)		0.45
		Left Eye		121 (40.1%)		53 (46.9%)
		Both Eyes		35 (11.6%)		12 (10.6%)
Eye pain at baseline		Yes		27 (9%)		10 (8.8%)		0.99
Gender		Male		155 (51.3%)		63 (55.8%)		0.49
Hot Fomentation done (Days)				17.67 (7.09)		16.70 (5.57)		0.30
Lid		Lower Lid		99 (34%)		35 (33%)		0.67
		Upper Lid		177 (60.8%)		63 (59.4%)
		Both Lids		15 (5.2%)		8 (7.5%)
No of Lesions		Single		280 (94%)		106 (97.2%)		0.22
		Multiple		18 (6%)		3 (2.8%)
H/o Recurrent chalazion at baseline		Yes		23 (7.6%)		12 (10.6%)		0.44
Spherical refraction (D)				0.01 (0.64)		-0.25 (1.54)		0.004
Spherical Equivalent (D)				-0.25 (0.82)		-0.53 (1.92)		0.73
Time Interval to first follow-up (days)				26.00 (23-29)		26.00 (21-78)		0.57

Time-to-surgery analysis (Kaplan–Meier curve and Cox regression)

The Kaplan–Meier survival analysis revealed that the majority of surgeries occurred within the first 60 days of presentation. A sharp early drop in the survival curve corresponded to patients advised to undergo surgery at baseline, followed by a gradual decline. The curve plateaued beyond 2 months (Cumulative risk of surgery at 2 months = 23%, 95% CI = 17–29%), suggesting a limited benefit from further conservative management beyond this point [Fig. 2].

Figure 2.

Kaplan–Meier Survival Curve showing the cumulative probability of remaining surgery-free over time since initial presentation

Comparison of surgery versus no surgery

Children who eventually underwent surgery (n = 90), including those advised at baseline, were significantly older (mean age 9.68 vs. 6.95 years; P < 0.001) and had longer duration of symptoms at baseline (median 60 vs. 30 days; P < 0.001) [Table 3]. They also reported performing marginally shorter duration of hot fomentation (mean 14.7 vs. 17.7 days; P < 0.001) and were more likely to have mild myopia (spherical power −0.21 vs. −0.06 D; P = 0.01). Presenting symptoms and laterality trends were different across groups but did not reach statistical significance. Eye pain, recurrence, and spectacle use were more common among surgical eyes but not statistically significant.

Table 3.

Comparison of those who did and did not undergo surgery as well as baseline factors predicting risk of surgery during follow-up

Variable		Surgery (n=90)		No Surgery (n=381)		P		Univariate regression				Multivariable regression
								HR		95% CI		HR		95% CI
Age (years)		9.68 (4.89)		6.95 (4.43)		<0.001		1.09**		1.04–1.13**		1.1		1.04–1.17
Duration of CC (days)		60 (30-150)		30 (7-90)		<0.001		3.08**		1.53–6.21		1.23		0.59–2.5
HF Duration (days)		14.65 (5.57)		17.69 (6.85)		<0.001		0.95*		0.91-1		0.97		0.92-1
Spherical Power		-0.21 (0.76)		-0.06 (0.99)		0.01		0.98		0.85–1.14
Spherical Equivalent		-0.45 (1.04)		-0.33 (1.27)		0.48		1.02		0.88–1.18
Right Eye		41 (45.56)		157 (41.21)		0.15		0.73		0.47–1.16
Eye Pain		3 (3.33)		35 (9.21)		0.08		0.44		0.14–1.38
Boys		48 (53.33)		198 (51.97)		0.91		1.2		0.79–1.83
Upper Lid		51 (56.67)		217 (59.78)		0.09		0.8		0.51–1.27
Both Lids		10 (11.11)		18 (4.96)
No. of Lesions (single)		86 (96.63)		357 (95.45)		0.78		1.22		0.38–3.89
Recurrent at baseline		12 (13.33)		32 (8.42)		0.16		1.39		0.76–2.56
Glasses Use		11 (12.22)		34 (8.95)		0.33		1.18		0.63–0.63

*P<0.1, **P<0.05

In univariate Cox regression [Table 3], increasing age (HR: 1.09; 95% CI: 1.04–1.13; P < 0.001) and longer duration of symptoms (median 60 vs. 30 days; P < 0.001) were significantly associated with higher risk of surgery. Duration of hot fomentation was inversely related to surgery risk (P = 0.03). Multivariable Cox regression confirmed that only advancing age remained an independent predictor of surgery (HR: 1.10; 95% CI: 1.04–1.17; P = 0.0002), while duration of chalazion (every 30-day increment) was not statistically significant in the adjusted model (HR: 1.23; 95% CI: 0.59–2.50; P = 0.57).

Discussion

In this large, retrospective observational study of 471 eyes from 470 children with chalazion, we found that most lesions resolved with conservative management, particularly within the first 4 to 8 weeks. A notable subset of patients, those with prolonged symptom duration at presentation, required surgical intervention, often advised at the first visit itself. The probability of remaining surgery-free declined steeply within the first 60 days and plateaued thereafter, suggesting limited additional benefit from extended conservative therapy beyond this time window. Importantly, children who eventually underwent surgery were significantly older, had a longer median duration of symptoms, and more commonly had mild myopia. Advancing age emerged as the most consistent independent predictor of surgical intervention across multivariable analyses. These findings underscore the value of a structured, time-bound trial of conservative treatment in pediatric chalazion and may help guide clinical decision-making, especially in settings where general anesthesia poses additional risks.

Chalazia are among the most frequently encountered eyelid lesions in the pediatric age group, though actual incidence may vary widely between practice settings. In children, these lesions can appear as firm, painless nodules that are often self-limiting but occasionally persistent or recurrent, leading to visual complications such as induced astigmatism and amblyopia if they obstruct the visual axis.[2,3] Current pediatric management paradigms involve conservative treatment, typically starting with warm compresses, lid hygiene, and topical antibiotics with or without steroids, with surgical intervention reserved for cases refractory to initial therapy.[6,8]

Our study adds substantially to the existing literature by offering one of the largest evaluations of pediatric chalazia to date and by applying time-to-event analysis (Kaplan–Meier) and hazard modeling to determine predictors of surgical intervention. To our knowledge, no previous study has quantified the temporal likelihood of surgery following presentation, nor identified baseline variables associated with failure of conservative treatment using survival methods or regression modeling. We found that the majority of children responded well to conservative management, particularly within the first 30–60 days. Eyes that eventually required surgery had a significantly longer median symptom duration and were more likely to be older, aligning with previous observations from 101 eyes of 91 children <15 years of age, that younger children may mount a more robust resolution response to treatment.[8] Notably, age emerged as the only independent predictor of surgery in multivariable Cox regression analysis (10% increased risk with every year increment in age), highlighting the role of delayed presentation or decreased efficacy of conservative measures in older children. This finding has not been previously reported and may guide clinicians in tailoring management strategies based on age at presentation.

Our Kaplan–Meier survival analysis showed that extending conservative management beyond 8 weeks is unlikely to offer additional benefits, a finding that could inform time-bound treatment algorithms. Given that chalazion surgery in children often requires general anesthesia, which carries added risk, this observation supports a structured 6–8 week trial of conservative therapy before escalating to surgical options, especially in younger children.

While prior studies have demonstrated high success rates for intralesional corticosteroid injections,[6,10] this modality was not part of our treatment protocol due to institutional preference and parental concerns regarding injectable steroids in young children. However, its reported efficacy, comparable to surgery after 1–3 injections, positions it as a valuable intermediate step in future treatment protocols, particularly in children with moderate disease or recurrent chalazia.[9,11] Topical steroid drops were prescribed in a small subset of children, mostly in cases where there was significant surrounding inflammation or inadequate response to initial conservative therapy. Although not routinely used in pediatric chalazion, topical steroids have been reported in adult literature as adjuncts to hasten resolution or reduce inflammation, especially in recalcitrant lesions.[12]

Our comparison of eyes that resolved versus those that did not with conservative treatment at 1 month is novel. While no strong clinical predictors emerged, nonresolving cases showed slightly higher degrees of myopia. Whether this reflects a predisposition to chalazion from increased eye rubbing in myopic children or a visual disturbance-induced refractive shift remains unclear. Although spectacle use and refractive status were not independently predictive in regression models, they may still contribute to disease behavior. The rising incidence of chalazia in very young children, where evidence for topical treatment safety is scarce, and the potential role of increasing digital screen exposure highlight emerging patterns that merit deeper etiological exploration.

Emerging literature also hints at broader systemic contributors to chalazion development, including vitamin A deficiency and gut microbiome alterations, particularly in recurrent cases.[13,14,15] While our study did not explore these associations, they represent important directions for future research, particularly in children with persistent or atypical lesions. Last, our findings reinforce the need for standardized protocols for chalazion management in the pediatric age group. Despite high rates of spontaneous resolution, variability in presentation, potential for amblyopia, and the surgical burden of general anesthesia call for clearer guidelines on when to observe, when to intervene medically, and when to escalate surgically.

This study has several limitations, including its retrospective design and reliance on variable follow-up. Surgical decisions were clinician-dependent, lesion size was not reliably documented in files and not used for surgical decisions, and treatments such as intralesional steroids were not employed, limiting generalizability. Functional outcomes like astigmatism or visual acuity changes were not assessed longitudinally. Despite these limitations, this represents the largest pediatric chalazion cohort to date and is the first to use survival analysis and hazard modeling to evaluate predictors of surgery. The detailed comparison between resolved and nonresolved cases, along with granular baseline data and pragmatic outcome measures, strengthens the study’s relevance and real-world applicability.

Conclusion

In conclusion, most pediatric chalazia resolve within 6–8 weeks of conservative management, and age at presentation is the strongest predictor of surgical intervention. A structured, time-bound approach may help avoid unnecessary procedures, especially in younger children. Future studies should explore the role of adjunctive therapies and investigate potential etiological factors, including environmental exposures, vitamin deficiencies, and systemic health, to guide more personalized, preventive care strategies.

Author contributions

Concept and design: MS, SH

1. Data acquisition: AM, SD

2. Statistical analysis: SS

3. Manuscript preparation: AM, SS

4. Final manuscript review: All co-authors.

Conflicts of interest

Dr Sengupta reports receiving speaker honoraria from Novartis India, Roche India, Lupin, and Appasamy Associates. The remaining authors declare no conflicts of interest.

Funding Statement

Nil.