Hallux rigidus (HR) is the second most common pathology affecting the first metatarsophalangeal joint and a primary cause of morbidity and disability. Classification of this condition helps to inform management. Over the years a number of formal HR classification systems have been devised but despite this collective experience there is no consensus on classification design, construction, application or parameters’ validity. The aim of this research was to develop an evidence-based classification framework for HR and establish its validation and reproducibility. This was achieved through four studies.
An initial study of 110 patients was used to determine the clinical parameters of HR. In addition to other pertinent findings this showed a positive relationship between second toe length and first metatarsophalangeal joint pain (P<0.001). Correlations were found between first metatarsophalangeal joint pain and pes planus (r=0.84, P=0.05) and between reduced first metatarsophalangeal joint range of motion and hallux abductus interphalangeus (r=0.92, P=0.05).
A second study examined the radiological parameters of HR (in the same population). Amongst other relevant findings comparison of joint space narrowing with either hallux abductus interphalangeus (P<0.005) or osteophyte severity (P<0.002) was established.
Intra and inter-rater reliability studies were undertaken for all parameters. Overall, inter-rater reliability was poor. Only 28% of angular inter-rater measurements fell within a 5° range.
A fourth study was used to determine ‘expert’ opinion on HR classification using semi-structured interviews. The results revealed the need for consensus agreement among clinicians and patient involvement in creation and substantiation of classification content.
This research has provided a new understanding of HR classification and informed the development of a HR classification framework based on history, clinical and radiological domains. The established framework provides more than just a measure of severity and includes other dimensions such as contributory factors and functionality. Depending on its context, other applications include use as a diagnostic tool, establishing HR prevalence, monitoring progress, and surgical decision making. An algorithmic approach can enable the classification framework to be applied in different contexts proving clinical relevance and meaning to a range of professions.
This research also highlights that classification parameters should be validated, reliable, sensitive, quantifiable and few in numbers and that there is a requirement to provide a ‘gold standard’ against which future HR research can be compared.