Laser Flare Photometry to Monitor Childhood Chronic Uveitis: A Preliminary Report of a Monocentric Italian Experience

Abstract Background: Childhood chronic non-infectious uveitis (cNIU) is a challenging disease that needs close monitoring. Slit lamp evaluation (SLE) is the cornerstone of ophthalmological evaluation for uveitis, but it is affected by interobserver variability and may be problematic in children. Laser flare photometry (LFP), a novel and objective technique, might be used in children with uveitis. Aim: The aim of this study was to attempt the use of LFP in cNIU clinical practice. Methods: Children, attending the Rheumatology Unit and who were scheduled to receive ophthalmological evaluation, were prospectively enrolled to concomitantly receive SLE and LFP. SLE was performed blind to LFP measure. Demographic, laboratory, clinical, and ophthalmology data were collected. Results: A total of 29 children (58 eyes) were enrolled, including 3 with juvenile idiopathic arthritis without uveitis (JIA-no-U), 15 with JIA-associated uveitis (JIA-U), and 11 with idiopathic chronic uveitis (ICU). We observed significantly higher LFP values in the eyes of children with uveitis compared to the others (10.1 IQR 7.1–13.6 versus 6.2 IQR 5.8–6.9, p = 0.007). Accordance between the SLE and LFP measures, at baseline (ρ.498, p < 0.001) and during the follow-up (LFP II ρ 0.460, p < 0.001, LFP III ρ 0.631, p < 0.001, LFP IV ρ 0.547, p = 0.006, LFP V ρ 0.767, p = 0.001), was detected. We evaluated significant correlation between LFP values and the presence of complications (ρ 0.538, p < 0.001), especially with cataract formation (ρ 0.542, p < 0.001). Conclusions: In this cohort, LFP measurements showed a good correlation with SLE. LFP values showed a positive correlation with the presence of complications. LFP might be considered as a reliable objective modality to monitor intraocular inflammation in cNIU.


Background
Chronic non-infectious uveitis is a challenging disease from a diagnostic and treatment perspective, due to the risk of severe complications, including blindness, if it is not promptly and correctly identified and treated [1][2][3][4].Uveitis, according to the Standardized Uveitis Nomenclature (SUN), is defined based on the anatomical location, the disease course, and the clinical symptoms [5].In childhood, idiopathic uveitis and juvenile idiopathic arthritis (JIA)-associated uveitis account for the majority of cases of chronic uveitis, with typical involvement of the anterior segment [1,2,6].
Childhood chronic non-infectious uveitis (cNIU) is frequently asymptomatic and eventually develops into severe ocular complications and visual loss.Persistent ocular inflammation and chronic use of corticosteroids can lead to the development of severe ocular complications including cataracts, glaucoma, posterior synechiae, and band keratopathy [1].In order to prevent these complications and therefore blindness, it is crucial to start proper systemic treatment and perform close monitoring.In this vein, the slit lamp examination (SLE) allows evaluation of the number of cells present in the anterior chamber and aqueous flare which are the direct consequence of exudation for inflammation.The strength of cells, and protein exudation, directly correlate with the severity of inflammation and are graded based on the SUN classification [5].In JIA, due to the well-known risk of uveitis, an established screening program with SLE leads to early interception of uveitis development and its complications [7].SLE represents, therefore, the cornerstone of ophthalmology evaluation for uveitis in JIA, and overall cNIU [1,6].
However, this is a subjective technique that is burdened by interobserver variability.Additionally, it is not able to intercept subclinical changes in the blood-aqueous barrier that can start in the earliest phase, and it may also easily miss a low number of cells, mirroring a persistent, sub-acute/chronic inflammation, thus eventually leading to complications.Additionally, in children, SLE might be challenging because of scarce collaboration.
In the last few years, increasing progress has been achieved in quantifying intraocular inflammation in an objective way, and one of them is laser flare photometry (LFP).LFP is an objective, reliable, quantitative, and non-invasive technique that measures the light scattering of a helium-neon laser beam in the anterior chamber based on the concentration of proteins [8][9][10].These measurements are directly determined by the activity of uveitis because of the integrity of the ocular blood-aqueous barrier [9].
Our aim was to compare the LFP measurements with AC cell grading by SLE according to the SUN classification.

Study Design and Setting
This was a prospective observational study of JIA and cNIU children currently followed at the rheumatology and ophthalmology units of Meyer Children's Hospital IRCCS.They underwent LFP assessment during a scheduled ophthalmology evaluation by SLE, between 15 February 2022 and 31 May 2023.

Population in Study
All children who fulfilled the following inclusion criteria were included: (A) to be ≤16 years old when LFP was performed, (B) to undergo LFP measurement the same day of the SLE, (C) and to be diagnosed for JIA according to the ILAR definition [11] or cNIU according to the SUN criteria [5].Patients were excluded from the study if (A) the medical records were incomplete for clinical information, (B) they refused to participate, or (C) if the LFP measurement available was performed within 3 months from surgery, or (D) they refused to undergo LFP.
Topical corticosteroid and cyclopentolate were permitted.Ethical approval was obtained from the Institutional Review Board of Meyer Children's Hospital (RChildUv 27/2022, date of approval 15 February 2022).The research was conducted in accordance with the tenets of the Declaration of Helsinki.

Study Procedures and Data Collection
Children enrolled in the study underwent a standard ophthalmology visit that included evaluation of visual acuity, intraocular pressure measurement, SLE, fundoscopy, complication evaluation, and LFP measurement.SLE was performed, independently, by an expert ophthalmologist (RP), while LFP was performed, blinded, in all the patients, by the same ophthalmologist (CdL) along with always the same technician (GD), as a part of the routine ophthalmological examination at each visit since the enrolment.
Flare values were measured using a Kowa FM-700 laser flare photometer (Kowa Acculas, San Jose, CA, USA), which utilizes a diode laser to measure aqueous flare through a 0.3 × 0.5 mm sampling window.The obtained values are expressed as photon counts per millisecond (ph/ms).Each patient underwent six measurements, with the highest and lowest readings being excluded from the calculation.The average (denoted as AVG) and standard deviation (SD) were automatically calculated based on the remaining four flare measurements.
The following data were collected: The treatments performed at each visit were recorded, including topical as well as systemic treatment.

Statistical Analysis
Statistical analyses were performed using SPSS v28.0 for Microsoft.Continuous variables are described using mean and standard deviation or median and range depending on their distribution.Q-Q plots were used to test the assumption of normality.For dichotomous and categorical variables, proportions were used to assess the clinical characteristics of the population.A p-value of <0.05 was considered to indicate statistical significance.Differences among categorical variables were assessed using the chi-square test or the Fisher exact test as appropriate.Specifically, the Fisher exact test was used for categorical variables with 2 categories, when 1 or more of the cell counts was less than 5. Continuous variables were compared with Student's t-test or non-parametric tests such as the Kruskal-Wallis test as appropriate.Spearman's rank correlations were performed to assess any relationship between baseline and follow-up LFP values and AC cell grades, and the correlation of these parameters with the ocular complications.
At onset, nine patients (31%) showed ocular complications, with a median number of complications of 0 (IQR 0-2.5).The most frequent reported complications at onset were optic disc swelling in 12 eyes, posterior synechiae in 11 eyes, cataracts in 4 eyes, ocular hypertension in 2 eyes, band keratopathy in 1 eye, chorioretinal scar in 1 eye, and cystoid macular oedema in 1 eye.

LFP Findings in the Whole Cohort
All the 29 patients underwent to at least one LFP measurement at a median age of 147 months (IQR 105-173), after a median duration of the disease of 51 months (IQR 16.5-101.25).The mean value of the first LFP was 14.73 ph/ms (±20.3) in the whole cohort, with significant differences among eyes with uveitis versus eyes without uveitis, according to SLE evaluation (10.1 IQR 7.1-13.6versus 6.2 IQR 5.8-6.9,p = 0.007) (Figure 2) (Table 2).

LFP Findings in Patients with Uveitis
Analysis of LFP values in children with uveitis were then stratified according to uveitis activity and type of uveitis, based on SLE evaluations.At the first LFP assessment, the LFP median values of the eyes of subjects with active uveitis (15 eyes) were significantly higher when compared to the LFP median value of subjects with inactive uveitis (35 eyes), being 12.8 IQR 11.4-44.8versus 8.7 IQR 5.8-12, p = 0.002 (Figure 3).We did not observe a significant difference in the LFP measures regarding the type of uveitis: JIA-U and idiopathic uveitis LFP values were not different (Kruskal-Wallis test, p > 0.05).At the time of the first LFP assessment, the mean AC of subjects with uveitis was 0.29 (±0.66), with 17 eyes with complications (29.3%), of whom 9 had cataracts (15.5%) and 14 posterior synechiae (Table 2).Patients with idiopathic uveitis at the moment of the LFP

LFP Findings in Patients with Uveitis
Analysis of LFP values in children with uveitis were then stratified according to uveitis activity and type of uveitis, based on SLE evaluations.At the first LFP assessment, the LFP median values of the eyes of subjects with active uveitis (15 eyes) were significantly higher when compared to the LFP median value of subjects with inactive uveitis (35 eyes), being 12.8 IQR 11.4-44.8versus 8.7 IQR 5.8-12, p = 0.002 (Figure 3).We did not observe a significant difference in the LFP measures regarding the type of uveitis: JIA-U and idiopathic uveitis LFP values were not different (Kruskal-Wallis test, p > 0.05).
Analysis of LFP values in children with uveitis were then stratified according to uveitis activity and type of uveitis, based on SLE evaluations.At the first LFP assessment, the LFP median values of the eyes of subjects with active uveitis (15 eyes) were significantly higher when compared to the LFP median value of subjects with inactive uveitis (35 eyes), being 12.8 IQR 11.4-44.8versus 8.7 IQR 5.8-12, p = 0.002 (Figure 3).We did not observe a significant difference in the LFP measures regarding the type of uveitis: JIA-U and idiopathic uveitis LFP values were not different (Kruskal-Wallis test, p > 0.05).At the time of the first LFP assessment, the mean AC of subjects with uveitis was 0.29 (±0.66), with 17 eyes with complications (29.3%), of whom 9 had cataracts (15.5%) and 14 posterior synechiae (Table 2).Patients with idiopathic uveitis at the moment of the LFP At the time of the first LFP assessment, the mean AC of subjects with uveitis was 0.29 (±0.66), with 17 eyes with complications (29.3%), of whom 9 had cataracts (15.5%) and 14 posterior synechiae (Table 2).Patients with idiopathic uveitis at the moment of the LFP measurement had a higher prevalence of ocular complication compared to the others (χ 2 3.84, p = 0.05).

Discussion
This prospective study is one of the few studies that assesses the role of LFP in routine clinical practice to evaluate ocular inflammation in children with uveitis in comparison with SLE.We identified a consistent positive correlation between the LFP values and SLE performed by an expert ophthalmologist.We demonstrated the applicability of this novel technique in a cohort of children, where it was able to distinguish children with uveitis from children without uveitis, and we used this technique to monitor ocular inflammation in an objective way.Additionally, we highlighted a positive relation between higher LFP values and the presence of structural complications.

Discussion
This prospective study is one of the few studies that assesses the role of LFP in routine clinical practice to evaluate ocular inflammation in children with uveitis in comparison with SLE.We identified a consistent positive correlation between the LFP values and SLE performed by an expert ophthalmologist.We demonstrated the applicability of this novel technique in a cohort of children, where it was able to distinguish children with uveitis from children without uveitis, and we used this technique to monitor ocular inflammation in an objective way.Additionally, we highlighted a positive relation between higher LFP values and the presence of structural complications.
Detection and monitoring of disease activity in uveitis is crucial for customizing the treatment of our children and preventing complications.The SUN grading system, which is the currently used referred system to evaluate ocular inflammation, shows several limitations.First of all, there is the subjective estimation of inflammation by an observer, even though there are clear rules to count the cells, and additionally the use of noncontinuous variables does not allow estimation of minimal changes [5].Novel technologies such as LFP and the anterior segment optical coherence tomography (AS-OCT) showed a good acceptability profile for children, but more importantly an objective ability to quantify ocular inflammation in the absence of an expert ophthalmologist [16][17][18][19].As we evaluated in our study, the LFP values have a strong correlation with SLE performed by an expert ophthalmologist, and the technique is able to distinguish between patients with uveitis and without uveitis, and patients with active and inactive uveitis.
Indeed, uveitis leads to an alteration of the blood-aqueous barrier that changes the protein composition, and the LFP technique also offers the possibility to detect minimal changes in the composition of humor aqueous in chronic uveitis, even though there are not evident changes in the number of cells in the anterior chamber.
The first report of LFP use in this indication was published in 1972 by Herbort et al. [20], where they transformed flare from a qualitative to a quantitative measure of intraocular inflammation.Several clinical studies of uveitis patients have shown that flare measurements by LFP allow precise monitoring of uveitis, a response to therapy, and prediction of disease relapse or exacerbation.Correlations of LFP values with complications of uveitis and visual loss indicate that flare measurement by LFP is a valuable tool in a follow-up in patients with uveitis.However, there is limited evidence about the use of LFP, as far as we know, in childhood.
Thus, LFP offers an extremely objective instrument also able to detect minimal changes in the aqueous composition, with minimal collaboration from young patients compared to common SLE that, in in ambiguous situations, requires prolonged collaboration from young patients.
The use of this novel technique in routine clinical practice has already been reported in limited pediatric studies [8,10,13,15,21,22].It was also recently used in association with SLE to monitor ocular inflammation in a randomized controlled trial conducted by Quartier et colleagues that contributed to the approval of adalimumab for the treatment of childhood chronic anterior uveitis [22].Additionally, as recently reported by Yalcidang et al., it might be a useful tool for the monitoring of patients with idiopathic uveitis, as confirmed in our report where we assessed eleven patients with idiopathic uveitis that showed LFP parameters similar to JIA-associated uveitis and good concordance with AC grades over the time [13].
In accordance with recent studies, our study showed significantly increased LFP values in patients who showed structural complications such as cataracts and posterior synechiae, in a consistent way, regardless of inflammatory status [9,10,14,15,21,23,24].In our cohort, we further showed a persistent correlation between LFP values and AC cells, even though there were complications.This seems in accordance with previous data reported not only in a large adult cohort such as Gonzales et al., but also in pediatric cohorts by Holland et al., Davis et al.,and Yalcidang et al. [8,9,13,15].

Figure 1 .
Figure 1.Selection process of the population included.

Figure 2 .
Figure 2. Comparison of laser flare photometry (LFP) values between patients with uveitis and without uveitis with the Kruskal-Wallis test (p = 0.007) that shows higher value of LFP values in patients with uveitis.The central line represents the distribution median, boxes span from the 25th to the 75th percentile, and error bars extend from the 10th to the 90th percentile.Asterixis are values higher than the 90th percentile.

Figure 3 .
Figure 3.Comparison of laser flare photometry values between patients with active and inactive uveitis with the Kruskal-Wallis test (p = 0.002).The central line represents the distribution median, boxes span from the 25th to the 75th percentile, and error bars extend from the 10th to the 90th percentile.Dots are values higher than the 90th percentile.

Figure 2 .
Figure 2. Comparison of laser flare photometry (LFP) values between patients with uveitis and without uveitis with the Kruskal-Wallis test (p = 0.007) that shows higher value of LFP values in patients with uveitis.The central line represents the distribution median, boxes span from the 25th to the 75th percentile, and error bars extend from the 10th to the 90th percentile.Asterixis are values higher than the 90th percentile.

Table 2 .
Comparison of clinical and ocular characteristics in patients with uveitis and without uveitis at the time of the first laser flare photometry measurement. of abbreviations: U uveitis, JIA juvenile idiopathic arthritis, IQR interquartile range, LFP laser flare photometry, SD standard deviation, AC anterior chamber, n number, pts patients, E eyes, vs. versus.

Figure 3 .
Figure 3.Comparison of laser flare photometry values between patients with active and inactive uveitis with the Kruskal-Wallis test (p = 0.002).The central line represents the distribution median, boxes span from the 25th to the 75th percentile, and error bars extend from the 10th to the 90th percentile.Dots are values higher than the 90th percentile.

Figure 3 .
Figure 3.Comparison of laser flare photometry values between patients with active and inactive uveitis with the Kruskal-Wallis test (p = 0.002).The central line represents the distribution median, boxes span from the 25th to the 75th percentile, and error bars extend from the 10th to the 90th percentile.Dots are values higher than the 90th percentile.

Diagnostics 2023 , 5 Figure 4 .
Figure 4. Chromatic covariation matrix of the correlations performed.The red color represents higher coefficient of correlation, while blue represents lower coefficient of correlation.List of abbreviations: SLE slit lamp evaluation, AC: anterior chamber cells, LFP laser flare photometry, N number, CCS Corticosteroid, VG LFP: value of LFP, Visual Acuity: visual acuity expressed in LogMar as a continuous variable, BCVA: best corrected visual acuity stratified as normal < 0.4, impaired 0.4-1, blindness > 1.The commas reported in the present figure should be considered as separator of decimal numbers.

Figure 4 . 5 Figure 5 .
Figure 4. Chromatic covariation matrix of the correlations performed.The red color represents higher coefficient of correlation, while blue represents lower coefficient of correlation.List of abbreviations: SLE slit lamp evaluation, AC: anterior chamber cells, LFP laser flare photometry, N number, CCS Corticosteroid, VG LFP: value of LFP, Visual Acuity: visual acuity expressed in LogMar as a continuous variable, BCVA: best corrected visual acuity stratified as normal < 0.4, impaired 0.4-1, blindness > 1.The commas reported in the present figure should be considered as separator of decimal numbers.Diagnostics 2023, 13, x FOR PEER REVIEW 4 of 5

Figure 5 .
Figure 5.Comparison of laser flare photometry values in patients with uveitis with and without ocular complications (p < 0.001) that shows higher LFP values in patients with complicatios.The central line represents the distribution median, boxes span from the 25th to the 75th percentile, and error bars extend from the 10th to the 90th percentile.Asterixis are values higher than the 90th percentile.

Table 1 .
Clinical, laboratory, and ocular characteristics of the population at onset.

onset, mg/dl, median (IQR)
Figure 1.Selection process of the population included.

Table 1 .
Clinical, laboratory, and ocular characteristics of the population at onset.

Table 3 .
This table reports the evolution of the mean of LFP values and mean of slit lamp at the different time points considered, with the different complications evaluated.