A comparative study between clinical grading of anterior chamber flare and flare reading using the Kowa laser flare meter

Transcription

1 DOI /s ORIGINAL PAPER A comparative study between clinical grading of anterior chamber flare and flare reading using the Kowa laser flare meter Kallirroi Konstantopoulou Roberto Del Omo Anne M. Morley Dimitris Karagiannis Catey Bunce Carlos Pavesio Received: 29 April 2012 / Accepted: 16 July 2012 Ó Springer Science+Business Media B.V Abstract To assess the accuracy of standard clinical grading of aqueous flare in uveitis according to the Standardization of Uveitis Nomenclature consensus, and compare the results with the readings of the laser flare meter, Kowa 500. Two examiners clinically graded the flare in 110 eyes. The flare was then measured using the Kowa laser flare meter. Twentynine eyes were graded as anterior chamber flare?2; for 18 of these, the clinicians were in agreement, the rest differed by the order of one grade. The range of the laser flare meter for these eyes was photons/ ms. The median value was Seventy-four eyes were graded with flare?1. Agreement was established in 51 of these eyes. Disagreement for the rest was again by the order of 1, and the flare meter range was photons/ms, median value For the clinical measure of flare 0, the clinicians disagreed on three out of five eyes. The flare meter readings ranged from 2.5 to 14.1 photons/ms, median value 9.9. Only two eyes were graded with flare?3 and there was one step disagreement on both of them. We found little evidence of association between the flare readings and K. Konstantopoulou (&) R. Del Omo A. M. Morley D. Karagiannis C. Pavesio Medical Retina Department, Moorfields Eye Hospital, 162 City Road, London EC1V 2PD, UK kalikonsta@yahoo.gr C. Bunce Research Department, Moorfields Eye Hospital, 162 City Road, London EC1V 2PD, UK intraocular pressure or age. Our findings suggest that clinical evaluation of aqueous flare is subjective. Compared with the Kowa laser flare meter s numeric readings, the discrepancies observed indicate that clinical grading is an approximate science. The laser flare meter provides an accurate, reproducible, noninvasive assessment of aqueous flare that can prove valuable in research and clinical decisions. Keywords Introduction Uveitis Flare Laser flare meter In uveitis, there is a breakdown of the the blood aqueous barrier (BAB), allowing the influx of protein from the plasma. This produces the Tyndall effect, usually referred to as flare. Flare can be observed during the acute phase of uveitis, due to the increased concentration of immunoglobulins and plasma albumin in the aqueous humor, but it can also be observed as a consequence of chronic disruption of the BAB resulting in the leak of ordinary plasma solutes in the anterior chamber. Based on this, traditional teaching suggests that flare alone does not indicate disease activity, and that it only represents breakdown of the BAB, and therefore should not guide treatment decisions [2, 10].

2 The most common and broadly acknowledged method for determination of grading of flare is by observation of iris details by slit-lamp examination of the anterior segment, which provides a qualitative grading system. This method was initially introduced by Kimura et al. [1, 2] in 1959 and was more recently revisited by the Standardization of Uveitis Nomenclature (SUN) consensus meeting. However, this method is limited by the subjectivity of the clinician s interpretation, which will depend on his/her experience. Furthermore, several examiners may be involved in the care of the patient throughout the progression of the disease and this will potentially lead to a wide range of interpretation of the anterior chamber status. This variability in interpreting clinical findings and the lack of consistency is probably another reason why flare tends to be disregarded as a marker of disease activity. The Kowa 500 laser flare meter is an instrument that allows a non-invasive quantification of aqueous flare by measurement of scattered light [5, 17]. The instrument consists of a diode laser that scans the anterior chamber and a photomultiplier that records the light scattered from the aqueous protein. Levels of protein concentration in the anterior chamber correlate linearly with the amount of scattered light which can be quantified and measured in a numeric scale [19]. This produces an objective measurement of flare, which could offer an advantage over the current clinical method. The purpose of this study was to compare the standard qualitative assessment of anterior chamber flare, according to the SUN classification, with the quantitative readings obtained by the laser flare meter during the same clinical examination, and to estimate the level of agreement of the clinical method between observers. Materials and methods A total of 110 eyes of 75 patients were examined by two clinicians, one of them being more experienced, who independently graded anterior chamber flare and cells according to the SUN classification (Table 1). Immediately after clinical assessment, the same patients had flare readings measured by an experienced technician using the laser flare meter KOWA FC 500, taking seven consecutive measurements and Table 1 SUN classification of cells and flare Grade Cells Flare 0 \1 None? ? Faint? Moderate? Marked?4 [50 Intense Slit lamp beam field size: 1 9 1mm 2 Table 2 Agreement on anterior chamber flare between observers using the SUN grading Number of eyes Senior clinician grade excluding the two extreme values from the mean value, according to the manufacturer s guidelines. All patients were assessed by the same technician (Table 2). Age, sex, ethnic origin, visual acuity, intraocular pressure and previous treatment were recorded. Exclusion criteria included any factor which could interfere with the KOWA FC 500 reading, such as the presence of corneal scars, dense cataract or extensive posterior synechiae. Children were also not included in this study. Statistical analysis Exact agreement One step disagreement Greater than one step disagreement The clinical readings of flare and cells according to the SUN consensus were compared between observers. We determined the proportions where there was: exact agreement; one-step disagreement (e.g., one examiner scored 1 and the other scored 0.5); more than one-step disagreement. Box-plots were drawn of laser flare readings against the SUN scorings for the more experienced of the two clinicians, to assess whether there was overlap

3 between the scores. A hundred eyes were chosen as the minimum figure suggested by Bland and Altman for method agreement studies. We analysed the SUN gradings as ordinal and did not linearise the gradings. Rather than attempt to linearise the ordinal measure, it was felt important to best describe the range of measures associated with each clinical grading, so that readers would have a clear understanding of how their clinical assessment related to a machine reading. Spearman s rank correlations were computed to assess any relationship between flare reading and each of intraocular pressure and age. Summary statistics were also examined (i.e., medians and interquartile ranges). Results Fifty-two (71 %) of the patients were female. The median (interquartile range) age of the patients was 42 (31 54) years. Twenty-nine of the subjects were white Caucasian, 19 described themselves as being of African origin, 22 patients were Asian and three patients were of other origin. For clinical measurement of flare 0, clinicians disagreed on three out of five eyes. Disagreement was greater than one step for one eye. The flare meter readings ranged from 2.5 to 14.1 photons/ms and the median value was 9.9. Seventy-four eyes were graded as presenting with flare?1 by the most senior clinician. Agreement between the examiners occurred for 51 of these eyes. Disagreement for the rest of the eyes was again by the order of 1, and the flare meter range was photons/ms, with a median value of Twenty-nine eyes were graded, by the more experienced ophthalmologist, as having an anterior chamber flare of?2, and the clinicians were in agreement for 18 of these, the rest differing by the order of 1 grade, according to the SUN classification. The reading range of the laser flare meter for these 29 eyes was photons/ms, median value Only two eyes were graded clinically with flare?3 and there was one step disagreement between the observers in both of them. We found little evidence of any association between the flare readings and intraocular pressure or age, as other studies have shown (Fig. 1) [3, 4]. Mean flare reading in photons/ms against IOP measurement IOP (mmhg) Fig. 1 Mean flare reading in photons/ms against intraocular pressure (IOP) measurement Table 3 Summary statistics of the laser flare readings against the more experienced clinician s clinical grades Senior clinician grading Median flare reading (range) photons/ms ( ) ( ) ( ) , a a Only two observations, median not computed log flare Log mean flare against clinical flare readings clinical flare Fig. 2 Log mean flare against clinical flare readings Table 3 and Fig. 2 show the variation of the flare readings for each of the different clinical grades, according to the SUN classification.

4 Discussion Since flare was first observed in uveitis, many articles have been published in an effort to establish its importance in the disease process. Some of these have demonstrated a connection between flare and relapses of uveitis; others have shown that in children, increased flare at presentation indicated a poorer visual prognosis [8, 9, 16, 17]. In 2005 an international group of specialists decided that anterior chamber activity should only be defined by the presence of inflammatory cells [2]. Their view was that flare did not necessarily indicate active disease. The measurement of flare has traditionally been conducted by a qualitative clinical method that relies on the skill and experience of the examiner and as such is a subjective finding [33]. Even though clinicians grade and document flare, they tend to rely on cell count to make decisions regarding disease activity and treatment. The introduction of a flare meter device many years ago has made a significant change in the ability of recording flare, since it allows for a quantitative reading of flare. In our study, we used different examiners in order to imitate daily clinical practice, and we showed that there can be considerable disagreement between observers when clinically grading anterior chamber flare. What we found surprising was the wide variation in the readings of the laser flare meter for each clinical grade. This suggests that clinicians may be far from accurate when assessing flare. As shown in our data, even when very high readings are obtained by the quantitative method, the clinical grade can be low. For example, a change from a reading of about 800 to as low as 5 photons/ms may go completely unnoticed by our standard clinical method, but it may actually indicate a significant change in disease activity or BAB recovery. The random selection of patients and eyes in our study did not allow us to have sufficient comparative data regarding higher degrees of clinical flare. The overall incidence of?3 or?4 flare was lower than grades?2 and?1, as expected. Nevertheless, the variations for each grade of clinical reading encountered in this study indicate that there are probably different degrees of barrier damage and, more importantly, there is possibly, in some cases, an ongoing inflammatory process for which therapy might be beneficial, as some studies have shown [14, 15]. Many authors in the past have recognized the limitations of clinical grading and the need for a universal reproducible grading system that would allow for a more reliable method of flare measurement [7, 9, 16, 25, 27]. Numerous clinical studies have incorporated the laser flare meter for interpretation of anterior chamber flare in order to achieve accuracy, objective reproducibility and comprehensible results [4, 6, 8, 10, 17, 18, 21, 25]. Herbort et al. demonstrated that observing flare and keeping a record of consecutive laser flare meter readings identified those eyes that responded slowly to treatment and required prolongation or even alteration in the management and those for which it was safe to taper therapy in order to avoid unnecessary prolonged exposure to treatment and its side effects [12, 13, 22 24]. The same group has looked into the use of flare readings for the identification of eyes with posterior uveitis which were about to develop a relapse of the disease. This offers the possibility of early intervention and, potentially, prevention of a new event [11]. Articles on paediatric uveitis state that increased values of flare correlate with a higher incident of ocular complications and a poorer visual prognosis [8, 9, 20, 26, 28]. Although all these reports and our findings are far from suggesting that flare alone implies active uveitis, they strongly indicate that the value of anterior chamber flare as an inflammatory marker is underestimated. Our findings also show that an instrument like the laser flare meter can provide with the necessary information that can actually be used for clinical decisions as well as for clinical trials with precise and universally understood results. In conclusion, according to our study the use of a clinical grading system to measure flare is quite subjective and does not allow for accurate measurement of flare values. Even though incorporating another test in routine clinical practice can be costly and time-consuming, the laser flare meter may represent a valuable asset for clinical decisions and clinical trials, where accuracy is essential. Acknowledgments This research has received a proportion of its funding from the Department of Health s NIHR Biomedical

5 Research Centre for Ophthalmology at Moorfields Eye Hospital and UCL Institute of Ophthalmology. The views expressed in the publication are those of the authors and not necessarily those of the Department of Health. The study has been approved by the appropriate ethics committee and all persons gave their informed consent prior to their inclusion in the study. It has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. References 1. Kimura SJ, Thygeson P, Hogan AJ (1959) Grading of anterior chamber flare, signs and symptoms of uveitis. I. Anterior uveitis. Am J Ophthalmol 47: The Standardization of Uveitis Nomenclature (SUN) Working Group (2005) Standardization of uveitis nomenclature for reporting clinical data. Results of the first international workshop. Am J Ophthalmol 140(3): Shah SM, Spalton DJ, Allen RJ, Smith SE (1993) A comparison of the laser flare meter and fluorophotomety in assessment of the blood aqueous barrier. Invest Ophthalmol Vis Sci 34(11): Kuchle M (1994) Laser Tyndallometry in anterior segment diseases. Curr Opin Ophthalmol 5(4): Shah SM, Spalton DJ, Taylor JC (1992) Correlation between laser flare measurements and anterior chamber protein concentrations. Invest Ophthalmol Vis Sci 33(10): El-Harazi SherifM, Ruiz RichardS, Feldman RobertM, Chuang AliceZ, Villanueva Guillermina (2002) Quantitative assessment of aqueous flare: the effect of age and pupillary dilation. Ophthalmic Surg Lasers 33(5): Onodera T, Gimbel HV, DeBroff BM (1993) Aqueous flare and cell number in healthy eyes of Caucasians. Jpn J Ophthalmol 37: Davis JL, Dacanay LM, Holland GN, Berrocal AM, Giese PhD MJ, Feuer WJ (2003) Laser flare photometry and complications of chronic uveitis in children. Am J Ophthalmol 135(6): Gonzales CA, Ladas JG, Davis JL, Feuer WJ, Holland GN (2001) Relationships between laser flare photometry values and complications of uveitis. Arch Ophthalmol 119: Ladas G, Wheeler NC, Morhun PJ, Rimmer SO, Holland GN (2005) Laser flare-cell photometry: methodology and clinical applications. Surv Ophthalmol 50(1): Norsell K, Holmer A-K, Jacobson H (1998) Aqueous flare in patients with monocular iris atrophy and uveitis. A laser flare and iris angiography study. Acta Ophthalmol Scand 76: Guignard L, Gossec C, Salliot A, Ruyssen-Witrand M, Luc M, Duclos M, Dougados M (2006) Efficacy of tumour necrosis factor blockers in reducing uveitis flare in patients with spondyloarthropathy: a retrospective study. Ann Rheum Dis 65: Cellini M, Caramazza R, Bonsanto D, Bernabini B, Campos EC (2004) Prostaglandin analogs and blood aqueous barrier integrity: a flare cell meter study. Ophthalmologica 218: Hayasaka Y, Hayasaka S, Zhang X-Y, Nagaki Y (2003) Effects of topical mydriatics and vasoconstrictors on prostaglandin-e2-induced aqueous flare elevation in pigmented rabbits. Ophthalmic Res 35: El-Maghraby A, Marzouki A, Matheen TM, Souchek J, Van Der Karr M (1993) Reproducibility and validity of laser flare/cell meter measurements of intraocular inflammation. J Cataract Refract Surg 19: de Ancos E, Pittet N, Herbort CP (1994) Mesure quantitative de l inflammation dans la uveite anterieure aique HLA- B27 a l aide du laser flare-cell meter Kowa FC Klin Monbl Augenheilkd 204: Guex-Crosier Y, Pittet N, Herbort CP (1994) Evaluation of laser flare-cell photometry in the appraisal and management of intraocular inflammation in uveitis. Ophthalmology 101(4): Guex-Crosier Y, Pittet N, Herbort CP (1995) Sensitivity of laser flare photometry to monitor inflammation in uveitis of the posterior segment. Ophthalmology 102(4): Ladas JG, Yu F, Loo R, Davis JL, Coleman AL, Levinson RD, Holland GN (2001) Relationship between aqueous humour protein level and outflow facility in patients with uveitis. Invest Ophthalmol Vis Sci 42(11): Tappeiner C, Heinz C, Roesel M, Heiligenhaus A (2011) Elevated laser flare values correlate with complicated course of anterior uveitis in patients with juvenile idiopathic arthritis. Acta Ophthalmol 89(6):e521 e527. doi: / j x Epub 2011 May Heinz C, Zurek-Imhoff B, Koch J, Roesel M, Heiligenhaus A (2011) Long-term reduction of laser flare values after trabeculectomy but not after cyclodestructive procedures in uveitis patients. Int Ophthalmol 31(3): Wakefield D, Herbort CP, Tugal-Tutkun I, Zierhut M (2010) Controversies in ocular inflammation and immunology laser flare photometry. Ocul Immunol Inflamm 18(5): Bernasconi O, Papadia M, Herbort CP (2010) Sensitivity of laser flare photometry compared to slit-lamp cell evaluation in monitoring anterior chamber inflammation in uveitis. Int Ophthalmol 30(5): Herbort CP, Tugal-Tutkun I (2010) Editorial: laser flare (cell) photometry: 20 years already. Int Ophthalmol 30(5): Tugal-Tutkun I, Herbort CP (2010) Laser flare photometry: a non-invasive, objective, and quantitative method to measure intraocular inflammation. Int Ophthalmol 30(5): Holland GN, Denove CS, Yu F (2009) Chronic anterior uveitis in children: clinical characteristics and complications. Am J Ophthalmol 147(4): e5 27. Kempen JH, Ganesh SK, Sangwan VS, Rathinam SR (2008) Interobserver agreement in grading activity and site of inflammation in eyes of patients with uveitis. Am J Ophthalmol 146(6):813 8.e1 28. Holland GN (2007) A reconsideration of anterior chamber flare and its clinical relevance for children with chronic anterior uveitis (an American Ophthalmological Society Thesis). Trans Am Ophthalmol Soc 105: