The Oxford Finger Nail Appearance Score - a new scoring system for fingernail deformity following paediatric finger tip trauma

https://doi.org/10.1016/j.bjps.2020.08.002Get rights and content

Summary

Fingernail deformity is common, yet current methods used to define cosmetic appearance following trauma are mainly descriptive. In order to quantify the cosmetic appearance of the fingernail, we developed the Oxford Fingernail Appearance Score using a three stage iterative process. The score has five cosmetic components marked as binary outcomes composed of nail shape, nail adherence, eponychial appearance, nail surface appearance and presence of a split. In the first stage, two assessors independently assessed 25 photographs of fingernails taken at a minimum of four months following paediatric nail bed repair and compared them to the corresponding contralateral uninjured finger. Following refinement in the score, ten different assessors scored a further 62 photographs of fingernails taken after paediatric nail bed repair. Assessors completed each of the five components, and the overall component score was calculated by statisticians post-hoc, taking the ideal appearance of each component as 1 (“identical to opposite” for nail shape, eponychium and surface, “complete” for adherence, “absent” for split) and all the non-ideal appearances as 0. Assessors effectively scored the photographs’ integer values between 0 (least optimal appearance) and 5 (most optimal appearance). Refinements in the scoring system resulted in an improvement in a weighted kappa statistic of 0.36 (95% CI:0.09,0.68) in the initial score to 0.52 (95% CI: 0.42, 0.61). The Oxford Fingernail Appearance Score is a user-friendly and reliable scoring system which has application in a clinical trial setting.

Introduction

Nail deformity can be associated with systemic conditions and commonly occurs following infection and trauma1. The description and classification of fingernail deformity is subjective1,2. However, a quantifiable measure of cosmesis allows differences in appearance between different fingernails and changes within the same fingernail following treatment to be measured and compared. This is important as it allows an objective measure of appearance. In order to quantify improvement in appearance following surgery, Zook et al3 developed a comparative scoring system. This was based on five domains (with between two and five subdomains) resulting in a summation score of major and minor variations. In 2017, we published the results of a pilot randomised control trial of paediatric nail bed injury (NINJA-P) in which 60 children were randomised to having the nail replaced or discarded following nail bed repair4. Cosmetic appearance was an outcome measure, and the Zook score3 was used to assess the cosmetic appearance of the injured nail four months after the injury/surgery had occurred.

In the original NINJA-P study4, two independent assessors used the Zook score3 to assess 25 clinical photographs of the operated fingers of 25 children compared to the corresponding uninjured contralateral finger taken at the four-month time-point (Figure 1). Agreement between assessors was measured by way of Cohen's kappa and percentage agreement on each of the score's domains and the overall score. The overall score measure was created by taking a positive outcome for each component as a 1 and a negative outcome as a 0. The outcome for the components were added together to create a total score of between 0 and 5 (where 0 was the least optimal and 5 was the most optimal nail appearance). Percentage agreement in the nail shape and surface components was 36% and 48%, respectively. Adherence, eponychium and split scored higher at 72%, 88% and 100%, respectively. The agreement for the overall score was 40%, and a weighted kappa for the overall score was 0.36 (95% CI 0.09 to 0.68).

The NINJA-P study identified agreement was poor between assessors for the Zook based score4. It was clear that the original Zook score developed over 35 years ago was not reliable for use in clinical trials to assess the outcome of nail appearance following trauma. As part of a large randomised control trial assessing cosmetic outcomes following nail bed repair5, we modified the score to achieve greater consistency between assessors and added training to reduce inter-rater variability. The aim of the work was to design a score based on the Zook score to improve the consistency and reliability of nail assessments in both clinical practice and clinical trial settings.

Section snippets

Methods

In order to improve the reliability of the score, we simplified the design of the Zook score (Figure 2A) by reducing the dimensionality of the score's components (Figure 2B). This was performed in two stages, as it was thought that the results obtained from the first set of modifications could be improved upon. The reduction of dimensionality of the score involved making all five components of the Zook score binary by comparing the affected nail to the contralateral corresponding non-injured

Results

The first round of modifications led to some improvement in the Zook score. However, the levels of percentage agreement and the kappa statistic for the overall score were relatively low. This prompted the second round of modifications described above, the results of which are reported in Table 1.

Of the 62 photographs, 58 were given a valid assessment by all 10 of the assessors. Three of the photographs were deemed suitable for valid assessment by nine assessors, and one of the photographs was

Discussion

Nail deformity is commonly seen following infection and trauma, yet there are no validated scoring systems by which to quantify the deformity. Current methods use descriptive terms for the appearance of the nail. However, these make comparison difficult when determining outcomes of treatment. Zook et al3 attempted to quantify fingernail appearance using a score utilising five important components of nail appearance, namely: nail shape, adherence, appearance of eponychium, nail surface

Funding

AJ, MDG, CC, AS, JC, DB and AVHG obtained grant funding for this project. This paper presents independent research funded by the National Institute for Health Research (NIHR) under its Research for Patient Benefit (RfPB) Programme (Grant Reference Number PB-PG-1215-20041) and was supported by the NIHR Biomedical Research Centre. The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.

Competing Interests

None to declare.

Ethics approval

The National Research Ethics Committee approved this study on 2nd February 2018 (18/SC/0024).

References (9)

There are more references available in the full text version of this article.

Cited by (2)

View full text