Fungal interface keratitis following superficial anterior lamellar keratectomy for limbal dermoids: a case report Authors: Tuhin Chowdhury, Vipin Rana Published: 2026 ABSTRACT AbstractBackgroundInfectious interface keratitis is an uncommon but potentially vision threatening complication of lamellar corneal surgery. While it has been described after deeper lamellar procedures, its occurrence following superficial anterior lamellar keratectomy (SALK) is exceedingly rare. To the best of our knowledge, such cases following limbal dermoid excision have not been previously reported. We describe two pediatric cases of fungal interface keratitis following SALK for limbal dermoids.Case presentationTwo children, a 12-year-old girl with Goldenhar syndrome and a 6-year-old boy with isolated limbal dermoids, underwent SALK for dermoid excision. Both had an uneventful early postoperative course. At 5 weeks and 4 weeks respectively, they developed graft infiltration with interface involvement. Initial microbiological evaluation was non-contributory, and both cases showed poor response to intensive topical antibacterial therapy. Due to progressive interface infiltrates, graft removal was performed. Microbiological analysis of explanted graft and host bed samples revealed fungal filaments, with culture confirming Aspergillus niger in both cases. Targeted antifungal therapy was initiated with topical natamycin and voriconazole, resulting in resolution of infection. Final visual acuity improved from 6/36 to 6/18 in the first case and from 6/18 to 6/9 in the second case, with no recurrence during follow up.ConclusionsFungal interface keratitis can occur following SALK, although rare, and may present with delayed onset and poor response to antibacterial therapy. Early suspicion and timely graft removal are critical for diagnosis and management. Prompt initiation of antifungal therapy can lead to favorable anatomical and visual outcomes.Keywords: Fungal interface keratitis, Superficial anterior lamellar keratectomy, Limbal dermoid, Sclerocorneal dermoid, Pediatric corneal infection Background Limbal dermoids are congenital choristomas that usually present in early childhood and may occur either as isolated lesions or in association with craniofacial syndromes such as Goldenhar syndrome [1]. These lesions can induce significant corneal astigmatism, encroach upon the visual axis, and cause cosmetic deformity, potentially leading to amblyopia if left untreated [2]. Surgical intervention is most commonly indicated for cosmetic concerns or ocular irritation, and less frequently for visual rehabilitation in selected cases, particularly when there is central or paracentral involvement affecting visual development [1, 2]. Superficial anterior lamellar keratectomy (SALK) has emerged as an effective surgical option for their management [3]. This technique allows precise excision of the lesion while preserving deeper corneal layers, thereby maintaining globe integrity and reducing intraocular risk. However, lamellar corneal surgeries may be complicated by infectious interface keratitis (IIK), an uncommon but potentially vision-threatening condition [4]. Diagnosis is often delayed due to sequestration of organisms within the graft–host interface and initially negative microbiological results [4]. While IIK has been reported widely following deep anterior lamellar keratoplasty, reports following SALK are exceedingly rare [5, 6]. To the best of our knowledge, no such cases following limbal dermoid excision have been previously reported. We describe two pediatric cases of IIK following SALK performed for limbal dermoids. Case presentation Case 1 A 12-year-old girl with a known diagnosis of Goldenhar syndrome presented with a limbal dermoid involving the inferior cornea (Fig. 1A). The lesion appeared as a whitish, elevated limbal mass with partial encroachment of the visual axis. At the time of initial examination, best-corrected visual acuity was 6/36 in the affected eye. She underwent surgical excision of the limbal dermoid with SALK under general anesthesia (Fig. 1B). The margin of the limbal dermoid was marked using a 6.5 mm trephine. Lamellar dissection was performed to excise the lesion while preserving deeper sclerocorneal layers. The fatty scleral component was carefully removed. A 6.75 mm donor graft with a thickness of approximately 300 μm was prepared using automated lamellar therapeutic keratoplasty (ALTK; MORIA, Antony, France) to match the defect. The recipient bed was irrigated with balanced salt solution prior to graft placement. The graft was secured with interrupted 10 − 0 nylon sutures, and the scleral edge was covered with adjacent conjunctiva and secured using fibrin glue. The donor corneal tissue was stored in MK medium prior to surgery. Clinical photographs illustrating the course of limbal dermoid managed with superficial anterior lamellar keratectomy and subsequent postoperative complications. A) Preoperative slit-lamp photograph showing a limbal dermoid involving the inferotemporal cornea. B) Early postoperative appearance following excision of the limbal dermoid with superficial anterior lamellar keratectomy, showing a well-apposed lamellar graft. C) Development of graft infiltration with interface haze and surrounding conjunctival congestion noted several weeks after surgery, suggestive of interface keratitis. D) Final clinical outcome after graft removal and appropriate antifungal therapy, showing resolution of infection with residual peripheral corneal scarring and a clear visual axis The immediate postoperative period was uneventful, and the patient remained asymptomatic for approximately five weeks. Postoperatively, she was started on topical Moxifloxacin 0.5% eye drops four times daily, Prednisolone acetate 1% eye drops four times daily, Homatropine 2% eye drops twice daily for two weeks and 0.5% Carboxymethylcellulose four times daily. Five weeks after surgery, she presented with redness and pain in the operated eye. Slit-lamp examination revealed graft infiltration with interface involvement (Fig. 1C). Initial microbiological evaluation, with corneal scraping performed from the edge of the infiltrate at the graft–host junction, was non-conclusive. She was treated with intensive topical fortified antibiotics, including Cefazolin 5% and Tobramycin 1.3%, along with Atropine 1% eyedrops. Graft explantation was performed due to progressive interface infiltrates despite nearly two weeks of medical therapy. The explanted graft, along with samples from the host bed, was sent for microbiological evaluation, including direct microscopy with 10% potassium hydroxide and Gram staining, and culture on blood agar, chocolate agar, and Sabouraud dextrose agar, which revealed fungal filaments on potassium hydroxide staining with culture positivity for Aspergillus niger. Topical antifungal therapy with 5% Natamycin and 1% Voriconazole was initiated hourly, followed by gradual tapering to every 2 h, then 4–6 times daily, and subsequently once daily over a total duration of 6–8 weeks, guided by clinical response, following which the corneal infiltrate healed with minimal peripheral scarring. No repeat corneal grafting was required (Fig. 1D). Best-corrected visual acuity improved to 6/18 at final follow-up. The patient was followed for approximately 4 years after resolution of infection, with no evidence of recurrence. Case 2 A 6-year-old boy presented with a limbal dermoid involving the inferior limbus with extension into the adjacent paracentral cornea and sclera in the left eye. (Fig. 2A). At presentation, best-corrected visual acuity was 6/18 in the affected eye. Clinical photographs illustrating the postoperative course of a limbal dermoid following superficial anterior lamellar keratectomy. A) Baseline slit-lamp image showing a limbal dermoid involving the inferior cornea. B) Early postoperative photograph after limbal dermoid excision with superficial anterior lamellar keratectomy, showing an intact lamellar graft in situ. C) Onset of interface infiltration with associated corneal haze and surrounding conjunctival hyperemia noted four weeks postoperatively. D) Appearance after graft explantation and institution of antifungal therapy, revealing resolution of the infectious process with mild residual peripheral corneal opacity He underwent dermoid excision with SALK under general anesthesia (Fig. 2B). The surgical steps were comparable; however, a larger recipient bed of 8.25 mm was fashioned, and a correspondingly oversized donor graft of 8.50 mm with a thickness of approximately 300 μm was prepared. The graft was secured with interrupted 10 − 0 nylon sutures. The early postoperative course was uneventful. Postoperatively, he received topical Moxifloxacin 0.5% four times daily, Prednisolone acetate 1% four times daily, Atropine 1% once daily for two weeks, and 0.5% Carboxymethylcellulose four times daily. At four weeks post-surgery, earlier than in Case 1, he developed localized graft infiltration associated with interface haze (Fig. 2C). Initial management consisted of intensive topical fortified antibiotics similar to Case 1; however, the clinical response remained suboptimal with progression of interface involvement. In view of worsening features, graft removal was performed. Microbiological analysis of explanted graft and samples obtained from the host bed demonstrated fungal filaments on 10% potassium hydroxide staining, with culture confirming Aspergillus niger, consistent with Case 1. Following confirmation, targeted antifungal therapy was initiated using topical 5% Natamycin and 1% Voriconazole in a tapering regimen guided by clinical response. The corneal infiltrate resolved with minimal peripheral scarring (Fig. 2D). In contrast to Case 1, visual recovery was better, with best-corrected visual acuity improving from 6/18 preoperatively to 6/9 at final follow-up. Discussion and conclusion IIK is an uncommon but potentially vision-threatening complication of lamellar keratoplasty [4]. It is characterized by infection confined to the graft–host interface and is often associated with delayed diagnosis and poor response to medical therapy [4]. Although IIK has been well described following deep anterior lamellar keratoplasty and endothelial keratoplasty, reports following SALK are exceedingly rare [5, 6]. Large reviews have consistently demonstrated that IIK predominantly occurs after deeper lamellar procedures. Fontana et al., in a review of 42 cases of IIK, reported that 31 cases occurred after endothelial keratoplasty and 12 cases after deep anterior lamellar keratoplasty, with no cases reported following SALK [7]. Fungal organisms, predominantly Candida species, were the most common causative pathogens, with donor-to-host transmission documented in the majority of cases. Yan Gao et al. further demonstrated fungal predominance in IIK, reporting fungal etiology in 46 of 62 cases (75.4%) following lamellar keratoplasty, with a mean onset of symptoms of 51.5 ± 102.4 days and Candida albicans as the most frequently isolated organism. In our report, Case 1 and Case 2 developed IIK at 35 and 28 days, respectively, following SALK for limbal dermoid excision, falling within the delayed postoperative window described by Yan Gao et al. Both cases were filamentous fungal in etiology and required graft removal for resolution, paralleling the clinical course reported in these large series, while differing in that IIK occurred after SALK, a procedure not previously implicated in the literature. Initial microbiological evaluation from the edge of the infiltrates at the graft–host junction, was non-contributory in both cases, and empirical treatment with intensive fortified antibacterial therapy failed to produce clinical improvement. Fungal infection was confirmed only after graft removal and repeat microbiological evaluation from the host bed and explanted graft. These findings underscore the limited diagnostic yield of initial corneal scrapings in interface infections and reinforce the need for a high index of suspicion for fungal IIK in cases of delayed graft infiltration unresponsive to antibacterial therapy. Similar observations were reported by Sharma et al., who emphasized that IIK is characterized by deep-seated infiltrates, delayed diagnosis, poor response to medical therapy, and a frequent requirement for surgical intervention [8]. Graft removal remains a critical component in the management of fungal IIK, particularly in cases refractory to medical therapy. The graft–host interface in lamellar procedures can serve as a relatively sequestered space, limiting the penetration and efficacy of topical antimicrobial agents and allowing persistence of microorganisms. The lamellar interface may act as a sequestered nidus of infection, further reducing drug effectiveness. In addition, factors such as irregular or thinned residual stromal beds following dermoid excision, interface fluid, and postoperative corticosteroid use may increase susceptibility to infection [9, 10]. Surgical explantation eliminates this reservoir and facilitates improved drug access to the host cornea. Although limbal dermoids themselves are not infectious lesions, factors related to limbal dermoid excision such as peripheral limbal location, irregular or thinned residual stromal beds, postoperative steroid use, and pediatric age may indirectly predispose to interface infection following SALK. In both cases, graft removal followed by targeted antifungal therapy resulted in resolution of infection with preservation of useful visual acuity. In conclusion, delayed-onset graft infiltration following SALK should prompt a high index of suspicion for fungal IIK, particularly in cases unresponsive to antibacterial therapy. Early consideration of graft removal and prompt initiation of antifungal treatment are critical for achieving favorable anatomical and visual outcomes. Acknowledgements Nil. Abbreviations Infectious interface keratitis Superficial anterior lamellar keratectomy Automated lamellar therapeutic keratoplasty Potassium hydroxide Best corrected visual acuity Author contributions VR drafted the manuscript and collected patient information, edited the photos, TC critically revised the manuscript and completed the project. All authors read and approved the final manuscript. Funding None. Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the parents of both the patients (for publication of this case report and accompanying images). Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.