Ottawa ankle rules in children: age limits and accuracy

Do the Ottawa ankle rules work in children?

The Ottawa ankle rules work well in children as a tool to rule fractures out, according to the largest summary of the pediatric evidence. Dowling and colleagues published a meta-analysis in Academic Emergency Medicine in 2009 that pooled 12 studies with 3,130 children and 671 fractures, a fracture prevalence of 21.4%. The pooled sensitivity was 98.5% (95% CI 97.3% to 99.2%), which the authors read as showing that the rules can be used to rule out a fracture. Ten fractures were missed across all the studies. An earlier systematic review in the BMJ by Bachmann and colleagues, which covered adults and children, found a pooled negative likelihood ratio of 0.07 (95% CI 0.03 to 0.18) for the ankle and midfoot together in children, close to the 0.08 it found for the ankle and for the midfoot across all patients. Applied to a 15% prevalence of fracture, Bachmann's ratios left a probability of actual fracture below 1.4%.

From what age can you use the Ottawa ankle rules?

The age question has a direct answer in the pediatric meta-analysis: Dowling and colleagues concluded that the Ottawa ankle rules appear to be a reliable tool to exclude fractures in children greater than 5 years of age presenting with ankle and midfoot injuries. That conclusion matters because children played no part in building the rules. Stiell and colleagues derived them from 750 adults in 1992 and refined and validated them in 1,032 and then 453 adults in 1993. The first 1992 version even included age 55 or over as a criterion, which the refined rules dropped. Plint and colleagues tested the rules in children aged 2 to 16 at two children's hospitals in 1999, so younger children have been studied, but the meta-analysis drew its conclusion for children over 5. Below that age, a child may not be able to point to where it hurts or take part in the 4-step test, so the rule depends more heavily on clinical judgement.

The Ottawa ankle rules calculator has an age setting: it gives the same result at any age but adds a caution when the child is 5 or under.

Growth plates and Salter-Harris fractures

Growth-plate injuries are the main concern when the Ottawa ankle rules are used in children, and the pediatric studies dealt with them in different ways, which shapes how their results read. Salter-Harris fractures involve the growth plate, and the type I pattern runs through it. In Dowling's meta-analysis, 10 of the 12 studies reported Salter-Harris type I fractures. Four of the 10 missed fractures were described: one Salter-Harris type I, one Salter-Harris type IV, and two that the authors called insignificant, meaning either a Salter-Harris type I or an avulsion fracture smaller than 3 mm. Plint's 1999 validation counted only clinically significant fractures, defined as fragments of 3 mm or more. Children who were not X-rayed were followed up, and those without symptoms 5 to 7 days after the injury were counted as having no significant fracture. A negative rule in a child therefore lowers the chance of a significant fracture, but it was not designed to find every small growth-plate injury.

How many X-rays the rules save in children

The Ottawa ankle rules save fewer X-rays in children than in adults, and the saving varies a great deal between hospitals. Dowling's meta-analysis pooled a reduction in X-ray use of 24.8% (95% CI 23.3% to 26.3%), with a range of 5% to 44% across the studies. In Plint's study of 670 children, following the rules would have cut ankle X-rays by 16% and foot X-rays by 29% without missing any clinically significant fracture, with a specificity of 24% for the ankle and 36% for the midfoot. Yet when the two hospitals were analysed separately, applying the rules would have reduced X-rays at one and increased them at the other. For comparison, Stiell's 1993 study estimated a possible reduction of 34% for the ankle series and 30% for the foot series in adults, and Bachmann's review put the expected reduction at 30% to 40%. A separate pediatric rule, the Low Risk Ankle Rule, has also been tested in children.

Boutis and colleagues introduced the Low Risk Ankle Rule in three of six Canadian emergency departments, studying 2,151 children aged 3 to 16. In the main implementation phase, ankle radiography fell by 21.9 percentage points at the intervention sites relative to the control sites, and during implementation that rule had a sensitivity of 100% (95% CI 85.4% to 100%) and a specificity of 53.1%.

Worked examples in children

Three children show how the Ottawa ankle rules decide each zone on its own. A 12-year-old rolls her ankle at football. She has pain around the lateral malleolus and tenderness over the soft tissue in front of it, but no bone tenderness at the posterior edge or tip of either malleolus, and she walks four steps in the department with a limp, which still counts as bearing weight. She has no midfoot pain. The rules are negative, so no X-ray series is called for. A 9-year-old boy jumps from a wall and lands awkwardly. He has malleolar zone pain and is tender at the tip of the lateral malleolus, though he can walk. One finding in the malleolar zone is enough, so he needs an ankle X-ray series. A 14-year-old inverts her foot on a step. Her ankle is not painful, but she has midfoot pain and is tender at the base of the fifth metatarsal, so she needs a foot X-ray series.

Enter each child in the Ottawa ankle rules calculator, with the age set to child over 5, to see the same results and the findings behind them.