Development of a new grading scale for tear ferning

Abstract

Purpose

This paper reports on the development of a new tear ferning (TF) subjective grading scale, and compares it with the Rolando scale.

Method

TF patterns obtained from tear film samples collected from normal and dry eye subjects in previous studies were collated into a large image library. From this library, 60 images were selected to represent the full range of possible TF patterns, and a further sub-set of 15 images was chosen for analysis. Twenty-five optometrists were asked to rank the images in increasing order between extreme anchors on a scale of TF patterns. Interim statistical analysis of this ranking found 7 homogeneous sub-sets, where the image rankings overlapped for a group of images. A representative image (typically the mean) from each group was then adopted as the grade standard. Using this new 7-point grading scale, 25 optometrists were asked to grade the entire 60 image library at two sessions: once using the 4-point Rolando scale and once using the new 7-point scale, applying 0.25 grade unit interpolation.

Results

Statistical analysis found that for the larger image set, the Rolando scale produced 3 homogeneous sub-sets, and the 7-point scale produced 5 homogeneous sub-sets. With this refinement, a new 5-point TF scale (Grades 0–4) was obtained.

Conclusions

The Rolando grading scale lacks discrimination between its Type I and II grades, reducing its reliability. The new 5-point grading scale is able to differentiate between TF patterns, and may provide additional support for the use of TF for both researcher and clinician.

Introduction

The chemical analysis of tear film composition is difficult due to the small volumes available, and to the transparent and dynamic nature of tears [1]. Clinicians and scientists recognise that biochemical analysis of osmolarity and other key components in a tear sample is the way forward, but the small volumes involved make biochemical analysis particularly challenging [2], [3]. Techniques available are limited by the need for expensive equipment that is difficult to use under normal clinical conditions [4]. A simple, clinical tear film test, that is quick and inexpensive to perform, and can indicate the biochemical properties of the tear film, would be very useful.

One potential and clinically suitable test involves drying a tear sample on a glass microscope slide to produce a crystallisation pattern in the form of a fern [5], [6], [7]. This phenomenon occurs with many body fluids and follows a characteristic formation process. The first discovery of tear crystallisation was reported by Fourcroy and Vauquelin in 1791 [8], but remained unnoted until 1946, when observed by Papanicolaou while studying cervical mucus [9]. Ferning patterns have been used to test different body fluids, such as vaginal and cervical mucus as an indicator of the menstrual cycle [10], oestrogen activity and ovulation [11], [12], [13], [14] and early pregnancy [13], [15]. Ferning has also been used to test saliva [16], to consider the observation of salivary ferning as a new technique for determining the fertile period [17], and to correlate salivary ferning and the fertile period [18], and using of salivary ferning in ovulation detection in family planning [19].

Crystallisation begins with the formation of a nucleus, consisting of a regularly arranged number of ions. The nucleus is formed by aggregation when the solute evaporates and dissolved ions are concentrated until super-saturation of the tear film is reached [7]. The nucleation process begins at the peripheral edges of the drop, where the solution is thinnest and super-saturation is reached rapidly [7]. Each nucleus has the ability to grow into a large crystal unit with the addition of more ions, and, so long as the sample solute is able to diffuse into areas with a lower solute concentration area, normal crystals can form. This requires a slow growth rate, low solution viscosity and low impurity levels to permit free solute diffusion.

The absence of these conditions can lead to dendritic crystal growth [20]. In this situation the stems grow longer and branch at regular intervals along the main stem. The reason for this regularity is not understood [7], but it is known that fern-like dendritic growth can be promoted by increasing the evaporation rate of the drop, by reducing atmospheric humidity, by increasing the drying temperature, or when impurities are present in low concentration, which acts as additional nuclei for crystal deposition [7].

Since tears are a complex solution, with many organic and non-organic components, the tear fern pattern produced by drying a sample depends on the composition of the tear sample [4], [7]. This variation in pattern has been suggested as a simple test for tear film quality at a gross biochemical level. This phenomenon gives tear ferning the potential, and the features, to be used as a diagnostic test in the clinic [5], [21]. Previous studies have demonstrated it to show good repeatability [22], sensitivity and specificity [21], [23], [24].

Different scales for grading tear ferning patterns have been proposed [6], [21], [25] with the Rolando scale being adopted as the main method used in previous published work in this area. However, the Rolando scale was not originally developed to produce a repeatable, standardised grading instrument, rather it arose from Rolando's observation that the Type I and II patterns were found in the majority of normal eyes, while Types III and IV were found in the majority of keratoconjuctivitis sicca (KCS) eyes [6].

The main difficulty with using the Rolando scale lies with this gross categorisation of ferning patterns, restricting sensitivity – the variance around Types I and II is particularly large – and not all types of tear ferning patterns are represented by the scale [22]. If the tear ferning test is to become part of routine clinical examination of the tear film, it is important to have a grading scale that has been developed to meet the needs of the clinician, and to address the four fundamental design requirements of a grading scale [26].

The aim of this paper is to report on the development of an improved subjective grading scale for clinicians, and the comparison of the new subjective scale with the Rolando scale.