Topical Insulin for Corneal Persistent Epithelial Defects: A Multicenter Retrospective Study

Abstract

Introduction

The purpose of this study was to evaluate the clinical outcomes of insulin eye drops in a real-world cohort of eyes affected by corneal persistent epithelial defects (PEDs) refractory to conventional therapy.

Methods

This retrospective study included consecutive patients with refractory PEDs treated with topical insulin (1 IU/mL) four times daily. Clinical end points were: time to complete reepithelialization, cumulative probability of closure, longitudinal changes in defect area, postclosure complications, recurrence, and changes in best-corrected visual acuity (BCVA).

Results

A total of 45 eyes from 43 patients (mean age 64.6 ± 15.6 years) were included. The most common etiology was chronic ocular surface diseases (62.2%) followed by neurotrophic keratopathy (37.8%). Complete epithelial closure was achieved in 44 of 45 eyes (97.8%), with a median time to closure of 29 days (95%CI 24–38). The cumulative probability of complete reepithelialization was 20.0% at 14 days, 53.3% at 30 days, 90.5% at 60 days, and 95.3% at 90 days. A biphasic healing pattern was observed, with marked early reduction of the defect area progressively slowing over time (p < 0.001). Median BCVA improved from 2.0 to 1.0 logMAR at epithelial closure (p < 0.001). Two eyes (4.4%) experienced PED recurrence: one case regressed after first-line treatment while the other case progressed to corneal melting and ultimately required Gundersen flap surgery. No treatment-related adverse events were recorded.

Conclusions

Topical insulin may represent a safe, effective, and accessible therapeutic option for refractory PEDs, achieving high rates of complete reepithelialization with concomitant visual improvement and low short-term recurrence in a heterogeneous real-world cohort.

Key Summary Points

Why carry out this study?

Persistent epithelial defects (PEDs) are challenging corneal conditions that fail to heal despite standard supportive therapy, often leading to visual impairment and potentially sight-threatening complications, including stromal ulceration, melting and perforation.

Topical insulin has recently emerged as a promising regenerative option for PEDs, but the available evidence is limited by small sample sizes, heterogeneous populations, and the absence of detailed longitudinal analyses of corneal wound-healing dynamics.

This study aimed at evaluating the clinical outcomes of insulin eye drops in a real-world cohort of patients with refractory PEDs, focusing on time to complete re-epithelialization, longitudinal wound-healing dynamics, visual outcomes, and recurrence rates.

What was learned from the study?

Topical insulin achieved complete epithelial closure in 44 of 45 eyes (97.8%) with a median healing time of 29 days, a significant improvement in visual acuity, and a low short-term recurrence rate (4.4%), without any treatment-related adverse events.

Topical insulin may represent a safe, effective, and accessible therapeutic option for refractory PEDs across heterogeneous etiologies, including both ocular surface disease-related and neurotrophic forms, and warrants further validation in prospective randomized controlled trials.

Introduction

The corneal epithelium, as the outermost layer of the cornea, plays a crucial role in maintaining ocular surface homeostasis, acting as a mechanical and immunological barrier against external agents while preserving the smooth optical surface required for visual function [1]. Under physiological conditions, corneal epithelial repair occurs through a highly regulated process involving cell migration, proliferation and differentiation, extracellular matrix remodeling, and complex interactions among growth factors, inflammatory mediators, and corneal nerves [2, 3]. When these mechanisms become impaired, the healing process may be disrupted, resulting in loss of epithelial integrity. Corneal persistent epithelial defects (PEDs) are generally defined as epithelial defects that fail to show significant healing within 2 weeks despite standard supportive treatment [4]. These represent severe conditions potentially associated with significant visual impairment, and may develop in the setting of different conditions, most commonly neurotrophic keratopathy (NK) and severe ocular surface diseases, such as dry eye disease (DED) [5]. In particular, NK-related PEDs may arise following postherpetic corneal denervation, surgical corneal nerve injury, or trigeminal nerve impairment secondary to trauma, neoplastic disease, or neurosurgical procedures [6]. In these conditions, reduced corneal sensitivity profoundly alters the trophic mechanisms required for epithelial regeneration by impairing reflex tear secretion, blinking dynamics, and the release of neuromodulators involved in epithelial proliferation and migration [7]. Similarly, severe DED and other chronic ocular surface diseases contribute to tear film instability, increased mechanical friction, oxidative stress, and persistent production of proinflammatory cytokines, thereby perpetuating epithelial damage and delaying wound healing [8]. Despite their heterogeneous etiologies, these disorders share common pathogenic mechanisms, including impaired epithelial migration and proliferation, chronic inflammation, tear film dysfunction, and altered corneal innervation [3, 6]. Current therapeutic strategies for PEDs aim at restoring a microenvironment favorable to epithelial healing and include preservative-free lubricants, discontinuation of epitheliotoxic medications, bandage contact lenses, punctal occlusion, tarsorrhaphy, and amniotic membrane transplantation; regenerative approaches include recombinant human nerve growth factor (rhNGF, Cenegermin), blood derived eye drops, and amniotic membrane extract eye drops [5, 9]. Nevertheless, the use of these therapeutic options remain limited by various factors such as high costs, reduced accessibility, surgical invasiveness, or inconsistent efficacy in chronic and refractory cases [5, 9].

In recent years, increasing attention has been directed toward topical insulin as a potential regenerative treatment for refractory PEDs [10–13]. The rationale for the use of topical insulin in PEDs is supported by both preclinical and clinical evidence. Insulin and insulin-like growth factor-1 receptors are expressed on corneal epithelial cells, and their activation has been shown to promote epithelial cell migration, proliferation, and survival through downstream PI3K/Akt and MAPK/Erk signaling pathways [14, 15]. In addition to its direct effects on epithelial cells, insulin may contribute to corneal nerve repair and to the restoration of a trophic microenvironment favorable to epithelial regeneration [16]. These mechanisms are particularly relevant in refractory PEDs, in which impaired epithelial renewal, altered corneal innervation, chronic inflammation, and tear film dysfunction frequently coexist.

Clinical studies, case series, and recent comparative investigations have reported promising outcomes in refractory PEDs, highlighting favorable safety, tolerability, and efficacy profiles of topical insulin therapy [10–13, 17–19]. However, currently available evidence remains limited by small sample sizes, heterogeneous study populations, and the lack of detailed longitudinal analyses capable of describing corneal wound-healing dynamics over time.

Therefore, the aim of the present study was to evaluate the clinical outcomes of insulin eye drops in a retrospective real-world cohort of eyes with refractory PEDs, with particular emphasis on time to complete reepithelialization, longitudinal wound-healing dynamics, visual outcomes, and recurrence rates.

Methods

In this retrospective case series, consecutive patients with a diagnosis of PED treated with insulin-based eye drops were evaluated. The study followed the tenets of the 2013 Declaration of Helsinki and was approved by the local ethics committee (Comitato Etico Territoriale Regione Calabria—protocol no. 51/2025). A written informed consent was provided to all patients.

Eligibility criteria included the presence of PED, defined as an epithelial defect not healing for at least 2 weeks despite first line medical therapy based on preservative-free tear substitutes and nocturnal ointments. Exclusion criteria included the presence of active infection, the need for corneal procedures (e.g., amniotic membrane transplantation), known allergy or hypersensitivity to insulin or other components of the eye drops. For each patient, medical records were reviewed to collect the following variables: age, sex, systemic comorbidities, ocular history including previous surgeries and concomitant therapies, treated eye (right or left), corneal sensitivity (measured using the Cochet–Bonnet esthesiometer and classified as normal [60–50 mm], reduced [45–5 mm] or totally absent [0 mm]), duration of PED from diagnosis to study treatment initiation, lid characteristics (position and dynamics), and best-corrected visual acuity (BCVA). PED etiology was classified into chronic ocular surface diseases (severe DED and/or eyelid alterations) and neurotrophic (postherpetic, secondary to trigeminal nerve damage, and postsurgical). Adherence to therapy and any adverse events were also recorded.

During follow-up visits scheduled every 2 weeks, PEDs were evaluated by slit-lamp examination and pictures were taken after fluorescein staining using cobalt blue and yellow filter (Fig. 1). The area of the PED was measured in mm² using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Fig. 1.

Serial slit-lamp pictures of a representative patient (79-year-old male) affected by persistent epithelial defect treated with insulin eye drops. (A) baseline presentation, (B) follow-up image at 12 days from initiation of therapy, (C) complete reepithelialization observed at day 19 of insulin treatment

Treatment Composition and Dosing Regimen

Insulin based eye drops were prepared by Apotiga-Laboratorio Galenico (Farmacia Europea, Catanzaro, Italy) in a grade D cleanroom under laminar airflow within an ISO 4.8 (grade A) isolator, as described elsewhere [18, 20]. Briefly, 1 IU/mL of insulin in solution for subcutaneous injection (Humalog^® sc 5cart 3 mL 100 IU/mL, Eli Lilly Italia Spa, Italy) was dissolved in an ophthalmic vehicle composed of the polyethylene glycol 400 (0.4% w/v), propylene glycol (0.3% w/v), hydroxypropyl guar (0.16% w/v), boric acid (0.7% w/v), sorbitol (1.4% w/v), potassium chloride (0.12% w/v), sodium chloride (0.1% w/v), 2-amino-2-methylpropanol (0.57% w/v), and the solvent purified water. The pH modifiers sodium hydroxide and hydrochloric acid were used to adjust the final pH to 7.9. The solution was then filtered through a 0.22 µm sterile filtration system (Stericup_® Quick Release, Merck Millipore, MC2, Clermont-Ferrand, France) and introduced into 5 mL multidose low-density polyethylene dropper bottles (Acef Spa). The solution was then stored at 2–8 °C, with a maximum shelf life of 30 days from the date of preparation.

All patients instilled insulin eye drops at the same concentration and composition, following the same dosing regimen (four times daily). After complete healing, treatment regimen was continued for additional 2 weeks and then interrupted. Other supportive treatments were maintained unchanged and included preservative-free tear substitutes and nocturnal ointments as well as, when required, disease-specific medications such as antiviral tablets for patients with history of herpetic keratitis and nocturnal eye patching in eyelid alterations.

End Points

The primary end point was the time to complete reepithelialization, defined as the number of days from treatment initiation to complete healing (absence of corneal fluorescein staining except for punctate keratopathy). Secondary end points included the cumulative probability of complete healing at 14, 30, 60, and 90 days, longitudinal changes in epithelial defect areas, occurrence of corneal complications during or after the complete healing, recurrence after reepithelialization, and BCVA changes.

Statistical Analysis

All data were entered into Microsoft Office Excel 365 (Microsoft Corp., Redmond, WA, USA) and analyzed with Jamovi (version 2.7.26; The Jamovi project, 2025). BCVA was measured using a Snellen chart and was then converted into the logarithm of the minimum angle of resolution (LogMAR) for the purpose of this analysis. Descriptive statistics were computed for all variables. The Shapiro–Wilk test was used to assess the normality of data and paired t‐test or Wilcoxon signed-rank test were used for normal and non-normal continuous variables.

Time to complete reepithelialization was analyzed using the Kaplan–Meier method. Eyes that did not achieve complete epithelial closure during follow-up were censored at the time of their last available assessment. Since the event of interest was complete reepithelialization, Kaplan–Meier estimates were expressed as the cumulative probability of complete reepithelialization calculated as 1−S(t). Cumulative closure probabilities were estimated at 14, 30, 60, and 90 days.

Univariable Cox regression analyses were performed for exploratory purposes to evaluate the association between selected clinical variables and time to complete epithelial closure.

Longitudinal changes in wound area were analyzed using a linear mixed-effects model. The dependent variable was log-transformed wound area, calculated as log(area +1), given the markedly right-skewed distribution of wound area and the presence of zero values at complete closure. Time from start of therapy, expressed in days, and its quadratic term were included as fixed effects to model the wound-healing trajectory. Patient ID was included as a random intercept to account for repeated measurements within individuals.

All tests were two-sided and a p < 0.05 was considered statistically significant.

Results

Study Population

Overall, complete data from 45 eyes from 43 white patients (22 males and 21 females; 64.6 ± 15.6 years) were included in the study analysis. Two patients, one with bilateral postsurgical NK and one with chronic ocular surface disease, received study treatment in both eyes. Demographic and clinical characteristics of included patients are summarized in Table 1. The most common etiological category was chronic ocular surface disease, which accounted for 28 eyes (62.2% of the total): of these, severe DED was observed in 23 eyes (51.1%) and eye lid alterations (floppy eyelid syndrome and lagophthalmos) causing exposure keratopathy in 5 eyes (11.1%). NK accounted for 17 eyes (37.8%), all of which were of stage 2, and included postherpetic NK in 7 eyes (15.6%), postsurgical NK in 6 eyes (13.3%), and NK secondary to trigeminal nerve damage in 4 eyes (8.9%). Corneal sensitivity was totally absent in 20 eyes (44.4%), reduced in 14 eyes (31.1%) and normal in 11 eyes (24.4%).

Table 1.

Demographic and clinical characteristics of included patients

Mean age ± SD (years) Range	64.6 ± 15.6 29–88
Sex Males (%) Females (%)	22 (51.2) 21 (48.8)
Eye: right (%)	26 (57.8)
Lid characteristics (%) Normal Symblepharon Lagophthalmos Floppy eye lid syndrome	37 (82.2) 3 (6.7) 3 (6.7) 2 (4.4)
Comorbidities (%) Systemic hypertension Diabetes mellitus Trigeminal neuralgia or palsy Prostatic disease Sjögren syndrome Autoimmune thyroid disease Other autoimmune pathologies Parkinson’s disease Bipolar disorder	11 (25.6) 7 (16.3) 4 (9.4) 3 (7) 2 (4.7) 2 (4.7) 3 (7) 1 (2.3) 1 (2.3)
Previous ocular surgery (%) Phacoemulsification with intraocular lens implantation Penetrating keratoplasty Deep anterior lamellar keratoplasty Descemet stripping automated endothelial keratoplasty Vitrectomy Glaucoma surgery Strabismus surgery Eye lid surgery (tarsorrhaphy)	9 (20) 8 (17.8) 1 (2.2) 1 (2.2) 3 (6.7) 3 (6.7) 1 (2.2) 1 (2.2)

All patients administered eye drops according to the prescribed regimen. No treatment interruptions nor adverse events were recorded throughout the entire study.

Time to Reepithelialization

The median interval between PED diagnosis and treatment initiation was 16 days (interquartile range (IQR), 14–29), whereas the median follow-up duration after treatment initiation was 121 days (IQR, 112–148). At baseline, the median epithelial defect area was 8.58 mm² (IQR, 4.86–14.3) and complete epithelial closure was achieved in 44 out of 45 eyes (97.8%). The only failure occurred in a 45-year-old woman with NK secondary to trigeminal nerve damage whose epithelial defect initially decreased in size but did not achieve complete closure after 90 days of treatment. Thereafter, the patient was lost to follow-up.

The median time to complete defect closure was 29 days (95%CI 24–38), while the estimated cumulative probability of complete reepithelialization was 20.0% (95%CI 7.4–30.9) at 14 days, 53.3% (95%CI 36.2–65.9) at 30 days, 90.5% (95%CI 76.0–96.2) at 60 days, and 95.3% (95%CI 81.7–98.8) at 90 days. Fig. 2 depicts the cumulative epithelial defect closure over time.

Fig. 2.

Cumulative complete defect closure over time in days, estimated from Kaplan–Meier. Crossmark: censored observation

Cox regression analyses showed no significant association between time to complete epithelial closure and age, sex, diabetes status, presence of any comorbidity, presence of eyelid alterations, or PED etiology (all p > 0.05). In fact, median time to closure was 26 days (IQR 19.5–29.8) in eyes with chronic ocular surface diseases and 37 days (IQR 19.5–50.0) in eyes with NK (p = 0.348).

The mixed effects model showed a significant association between time and reduction of the epithelial defect (conditional R² 0.761, marginal R² 0.487; p < 0.001). The linear time term was negative (β = −0.0736, 95%CI −0.0848 to −0.0623, p < 0.001), whereas the quadratic time term was positive (β = 0.000441, 95%CI 0.000301–0.000582, p < 0.001), indicating that defect area reduction was more pronounced during the early follow-up period and progressively slowed over time (Fig. 3).

Fig. 3.

Scatter plot of longitudinal change in log-transformed epithelial defect area over time

Outcomes after Epithelial Closure

Median BCVA varied from 2.0 logMAR (IQR 0.7–2.3) to 1.0 logMAR (IQR 0.5–2.0) at the time of epithelial closure (p < 0.001). Following reepithelialization, the median observation period was 95 days (IQR, 73.5–116). At the last recorded visit 12 eyes (26.7%) presented with stromal scarring, 6 (13.3%) had corneal neovascularization, 1 (2.2%) had both scarring and corneal neovascularization and 1 (2.2%) presented with band keratopathy. Corneal sensitivity remained unchanged after treatment (absent in 20 eyes, reduced in 14 eyes and normal in 11 eyes). PED recurrence occurred in two eyes (4.4%), respectively 44 days and 56 days after epithelial healing. Of these, one patient developed corneal melting and ultimately required Gundersen flap surgery for tectonic purposes, while the second patient was treated conservatively (tear substitutes and bandage contact lens) with success.

Discussion

In this retrospective real-world study, complete reepithelialization was achieved in 44 of 45 eyes (97.8%) treated with topical insulin for refractory PEDs, with a median time to closure of 29 days. The cases were characterized by a heterogeneous etiologic background, predominantly comprising severe ocular surface diseases and neurotrophic forms. The longitudinal analysis demonstrated a biphasic healing pattern, characterized by rapid early reduction of the epithelial defect followed by progressive deceleration over time. Reepithelialization was accompanied by a significant improvement in BCVA, while structural sequelae and recurrence rates remained very low. Treatment was uniformly well tolerated, and no treatment-related adverse events were recorded.

The clinical efficacy observed across distinct phenotypes is biologically supported by the multimodal mechanism of action of topical insulin. In fact, insulin and insulin-like growth factor-1 (IGF-1) receptors are expressed on corneal epithelial cells, and their activation triggers downstream PI3K/Akt and MAPK/Erk signaling pathways involved in cell migration, proliferation, and survival [14, 15]. Beyond its direct effect on epithelial cells, insulin has been shown to support limbal stem cell function, contributing to the long-term maintenance of ocular surface homeostasis, and to promote corneal nerve regeneration, with potential benefits on the trophic component of healing in eyes with reduced corneal sensitivity [14, 16].

The proportion of eyes achieving complete epithelial closure in this cohort matches the rates reported in previously published series. In the prospective studies by Diaz-Valle et al. and Abdi et al., complete healing was achieved in 81% of 21 eyes and in 69.6% of 23 eyes, respectively [10, 11]. Comparable findings were reported in the retrospective series by Almeida et al. and Wang et al., with closure rates ranging from 77.3% in 15 eyes to 100% in 6 eyes [12, 21]. When directly compared with other regenerative therapies, topical insulin has emerged as a possible alternative, with recovery rates of 84% versus 48% against autologous serum eye drops and 100% versus 44.4% against amniotic membrane extract eye drops [18, 22]. Importantly, the median time to complete reepithelialization found in this study (29 days) is in line with prior reports, as reported by Almeida et al. (31.7 ± 21.4 days) [12] and Diaz-Valle et al. (34.8 ± 29.9 days) [10], likely supporting the reproducibility of the underlying healing kinetics across independent cohorts.

Interestingly, the reduction of the defect area was most pronounced in the early phase after treatment initiation and progressively attenuated as healing proceeded. This course is consistent with the established biology of corneal epithelial repair, in which migratory and proliferative phases dominate the early stages of healing, whereas later phases involve slower remodeling and consolidation of the regenerated epithelium [2, 3].

No significant association was detected between time to complete epithelial closure and age, sex, diabetes, lid position abnormalities, or PEDs etiology. These findings are consistent with those reported by Abdi et al., who similarly found no significant correlation between healing time and age, sex, and diabetes [11]. In particular, the absence of a detrimental effect of diabetes is in agreement with the hypothesis that exogenous insulin can restore epithelial healing capacity even when endogenous insulin/IGF-1 signaling is altered [15, 23]. In addition, since complete epithelial closure was observed across a clinically heterogeneous cohort, including both ocular surface disease-related and neurotrophic forms of PED, the lack of an etiology-related effect may further suggest that the therapeutic effect of topical insulin operates broadly across phenotypically distinct PED subtypes, regardless of whether the dominant pathogenic driver is tear film instability, chronic inflammation, exposure-related trauma, or impaired corneal trophism. However, given the limited number of events in some subgroups, these analyses should be regarded as exploratory and require confirmation in larger cohorts.

In the present study, corneal sensitivity was impaired in most eyes, being totally absent in 44.4% of cases and reduced in 31.1%. The high prevalence of damage to corneal nerves, exceeding the proportion of cases formally classified as NK, suggests that subclinical or coexisting neurotrophic impairment is frequent in patients with longstanding ocular surface diseases, in line with previous reports describing reduced corneal innervation and sensitivity in severe or chronic DED [24, 25].

The improvement in BCVA from a median of 2.0 logMAR at baseline to 1.0 logMAR at the time of epithelial closure underscores the visual relevance of restoring epithelial integrity. Although a residual visual deficit persisted in most patients, this functional gain is clinically meaningful in a population in which severe visual impairment was the rule at presentation. The proportion of eyes presenting with stromal scarring (26.7%) and corneal neovascularization (13.3%) at the last follow-up likely reflects the long-term sequelae of the chronic epithelial breakdown that preceded treatment, rather than complications of insulin therapy itself. Indeed, no patient experienced treatment intolerance, ocular discomfort upon instillation, or other adverse events directly attributable to the eye drops. Adherence to the prescribed regimen was complete throughout the follow-up period, similar to the favorable safety profile already reported [11–13, 18, 19, 22].

Interestingly, a low recurrence rate (4.4%) was observed, with only two eyes experiencing PED recurrence after a median postclosure observation period of 95 days. This is consistent with the absence of recurrences reported by Buzzi et al. and Esmail et al. over a comparable follow-up period [18, 19]. Conversely, higher recurrence rates (13.3%) were reported by Almeida et al. over a longer follow-up (17.9 ± 7.63 months) [12]. One of the recurrence cases of this cohort followed an aggressive course, complicated by corneal melting and ultimately requiring Gundersen flap surgery, while the other healed without further complication after conservative treatment. This may highlight that, despite the generally favorable prognosis, a small subset of refractory cases may follow an unpredictable trajectory and warrants close surveillance even after apparently complete epithelial closure.

Beyond clinical efficacy, topical insulin can offer practical advantages in real-world settings and may influence its positioning within the therapeutic algorithm of refractory PEDs. Compared with biological alternatives such as autologous serum eye drops, amniotic membrane extract eye drops, or rhNGF, insulin eye drops present a favorable accessibility profile [5, 18, 26]. Insulin is universally available as an inexpensive molecule, can be compounded by hospital pharmacies into stable ophthalmic formulations, and does not require donor-derived material, blood sampling, or complex regulatory pathways for tissue procurement. This translates into shorter time intervals between prescription and treatment initiation, improved continuity of supply, and a markedly reduced economic burden on both patients and healthcare systems [5, 10, 18]. Conversely, autologous serum eye drops require periodic blood draws and dedicated processing, whereas amniotic membrane-derived products depend on tissue banks, donor screening, cold-chain logistics, and short post-thaw shelf life [5, 27]. Taking together all these aspects, topical insulin can be reasonably considered as an accessible option for refractory PEDs, particularly in clinical settings where other biological or regenerative therapies are unavailable, unaffordable, or contraindicated.

This study has several strengths, including a relatively large sample size, a pragmatic real-world design that enhances external validity, and the detailed longitudinal modeling of wound-healing dynamics. Nevertheless, several limitations should be acknowledged. Firstly, the retrospective design is inherently subject to selection and information bias. Secondly, the absence of a control group precludes any direct comparison with alternative treatments or with the natural history of refractory PEDs and limits causal inference regarding the magnitude of the treatment effect. The very high rate of epithelial closure observed in this cohort is encouraging but should be confirmed in prospective controlled studies. Thirdly, follow-up duration after epithelial closure was relatively limited, and longer observation periods will be needed to fully characterize the long-term recurrence pattern.

Conclusions

Topical insulin may represent a safe, effective, and accessible therapeutic option for refractory PEDs, achieving high rates of complete reepithelialization with concomitant visual improvement and low short-term recurrence in a heterogeneous real-world cohort. Prospective, randomized, controlled studies will be necessary to definitively establish the role of topical insulin in the therapeutic algorithm of PEDs, including its dosing schedule, treatment duration, and the potential indications for maintenance therapy.

Author Contributions

Conceptualization: Filippo Lixi, Alessandra Mancini, Costanza Rossi, Vincenzo Scorcia, Giuseppe Giannaccare; methodology: Filippo Lixi, Alessandra Mancini, Costanza Rossi, Claudia Corda, Benedetta Pintus, Andrea Lucisano, Federica Bianca, Maura Mancini, Mario Verdiglione, Giulia Coco, Mara Tomi, Alina Gabriela Gheorghe, Adriano Carnevali, Vincenzo Scorcia, Giuseppe Giannaccare; formal analysis and investigation: Filippo Lixi, Alessandra Mancini, Costanza Rossi, Claudia Corda, Benedetta Pintus, Andrea Lucisano, Federica Bianca, Maura Mancini, Giuseppe Giannaccare; resources: Filippo Lixi, Alessandra Mancini, Costanza Rossi, Claudia Corda, Benedetta Pintus, Andrea Lucisano, Federica Bianca, Maura Mancini, Mario Verdiglione, Giulia Coco, Mara Tomi, Alina Gabriela Gheorghe, Adriano Carnevali, Vincenzo Scorcia, Giuseppe Giannaccare; writing—original draft preparation: Filippo Lixi, Alessandra Mancini, Costanza Rossi, Claudia Corda, Benedetta Pintus, Andrea Lucisano, Federica Bianca, Maura Mancini, Mario Verdiglione, Giulia Coco, Mara Tomi, Alina Gabriela Gheorghe, Adriano Carnevali, Vincenzo Scorcia, Giuseppe Giannaccare; writing—review and editing: Filippo Lixi, Alessandra Mancini, Costanza Rossi, Claudia Corda, Benedetta Pintus, Andrea Lucisano, Federica Bianca, Maura Mancini, Mario Verdiglione, Giulia Coco, Mara Tomi, Alina Gabriela Gheorghe, Adriano Carnevali, Vincenzo Scorcia, Giuseppe Giannaccare; supervision: Filippo Lixi, Alessandra Mancini, Costanza Rossi, Giulia Coco, Alina Gabriela Gheorghe, Adriano Carnevali, Vincenzo Scorcia, Giuseppe Giannaccare.

Funding

No funding or sponsorship was received for this study or publication of this article.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethical Approval

This study followed the tenets of the 2013 Declaration of Helsinki and its later amendments and was approved by the ethics committee Comitato Etico Territoriale Regione Calabria—protocol no. 51/2025. A written informed consent was provided to all patients.

Conflict of Interest

Filippo Lixi, Alessandra Mancini, Costanza Rossi, Claudia Corda, Benedetta Pintus, Andrea Lucisano, Federica Bianca, Maura Mancini, Mario Verdiglione, Giulia Coco, Mara Tomi, Alina Gabriela Gheorghe, Adriano Carnevali, Vincenzo Scorcia, and Giuseppe Giannaccare have nothing to disclose.