The Foot 22 (2012) 156–162 Contents lists available at SciVerse ScienceDirect The Foot jou rn al h om epage: www.elsevier.com/locate/foot The reliability of measurements taken from radiographs in the assessment of paediatric flat foot deformity Stuart A. Metcalfe ∗ , Frank L. Bowling, V. Baltzopoulos, C. Maganaris, Neil D. Reeves Institute for Biomedical Research into Human Movement & Health, School of Healthcare Science, Manchester Metropolitan University, John Dalton Building, Oxford Road, Manchester, UK a r t i c l e i n f o Article history: Received 22 November 2011 Received in revised form 15 February 2012 Accepted 18 February 2012 Keywords: Reliability Reproducibility Radiographic measures a b s t r a c t Background: Radiographic charted values are widely applied in the assessment of paediatric flat foot and as quantitative markers of intervention success. Nearly all literature pertaining to the reliability of these measures relates to the adult foot. In view of the importance placed on these charted values we assessed the inter and intra-rater reliability of 10 key radiographic measures in children aged 7–14 years. Methods: We present the reliability of these measures in terms of intra class correlation and standard error based on repeated measures of 50 radiographs across five independent trained observers. Two new detailed methods are presented for the calculation of lateral and dorso-plantar talo-calcaneal overlap in an attempt to better define the talo-calcaneal inter-relationship. Results: The results demonstrate wide variation of measurement reliability, with some measures being highly reproducible and others showing poor reproducibility. A strong negative correlation between reliability and the number of steps required for the measurement was demonstrated. Consistent with other studies intra-rater reliability tended to be higher than that of inter-rater reliability. Conclusion: In conclusion, we have provided evidence as to which radiographic charted measures may be reliably used in evaluating the paediatric flat foot in children aged 7–14 years, and which measures are not reliable and we advise against their application. © 2012 Elsevier Ltd. All rights reserved. 1. Introduction Structural deformity of the foot is common and frequently necessitates surgical intervention [1–4]. Procedure selection is based on clinical and radiological examination, which determines the severity of the deformity and thus aids and establishes the sur- gical plan. In addition radiographic measures are used to assess outcome after surgery [5–8]. With respect to flat foot the most commonly charted measures are derived from weight-bearing dorso-plantar and lateral radiographs. These include but are not limited to foot length, calcaneal inclination, lateral talo-calcaneal, talo-1st metatarsal angle, navicular height, talar-calcaneal overlap (lateral and dorso-plantar), calcaneo-cuboid angle, talo-navicular angle, Kite’s angle and talo-navicular coverage [1,9,10]. Barwell first proposed the use of radiographs for the evalua- tion of club foot [11]. Subsequently attempts have been made to define normative data for a range of skeletal alignments. Wisburn was the first to report on one such measure within the foot, the ∗ Corresponding author at: Solent NHS Trust, Department of Podiatric Surgery, St James Hospital, Locksway Road, Southsea, UK. Tel.: +44 7931535823; fax: +44 122701577. E-mail address: footconsultant@gmail.com (S.A. Metcalfe). talo-calcaneal angle in 1932 [12]. Early reports of normative values were taken from the unaffected side in infants and children with unilateral club foot deformity although the weight-bearing status of the presumed normal foot was not made clear [13,14]. Adding to the limitations of early published normative values is the lack of standardisation for exposures and positioning, it is not clear in many circumstances whether subjects were weight- bearing for the exposures. This is of critical importance given that the inter-relationships of the tarsus and metatarsus are changed with loading. In 1965 the first data set suggesting a development trend within the tarsus was published by Templeton [15]. Measur- ing three angles on the radiographs of 160 normal feet Templeton published the first important set of data demonstrating a decrease of the anterior–posterior (AP) talo-calcaneal angle with increasing age [15]. This trend was later confirmed by Main in 1977 [16] and has remained to this day an important radiographic measure. Application of radiographic measures in the pre and postopera- tive evaluation of various foot pathologies requires measures to be reliable and reproducible. Sources of radiographic measurement error can be categorised as: 1. Patient a. Position (standardised) b. Movement 0958-2592/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.foot.2012.02.006