• 
• 

Three Part Question

In [a child with history of ankle injury] are [the Ottawa ankle rules] reliable [in eliminating the need for x ray examination in some patients without the risk of missing fractures]?

Clinical Scenario

A 5-year-old boy attends the A&E department after sustaining a twisting injury to his left ankle. On examination there is swelling and tenderness over the lateral malleolus. You know that the Ottawa ankle rules are applicable in adult patients and you wonder whether they are applicable in children too.

Search Strategy

Medline 1966-08/01 using the OVID interface.
Updated search: Pubmed
2000 search:[exp ANKLE/ or ankle.mp. or exp ANKLE INJURIES/ or exp ANKLE JOINT/ or exp LATERAL LIGAMENT, ANKLE/] AND[clinical decision.mp. or exp Decision Support Systems, Clinical/ or exp Decision Support Techniques/ or ottawa.mp.] AND [pediatr$.mp. or paed$.mp. or exp Age Factors/ or age factors.mp. or Child/] LIMIT to human and english
New Pubmed search: Clinical queries - Ottawa ankle rules AND child

Search Outcome

8 relevant papers found (4 previously)

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Chande VT, 1995	68 patients aged 2-18 years	Prospective survey with 24 variables obtained by physicians; x rays were taken of all study participants with blinding of investigator as to results of x rays when applying OAR to evaluate for qualification of x ray	Inclusion All types of fractures Exclusion Open fractures, patients without follow-up	Sensitivity 100% (95% CI. 77% to 100%). Specificity 32% (95% CI. 21% to 43%)	Small sample size
McBride KL 1997	318 adults and children (37 children) presenting with ankle injury to a community ED	Prospective survey looking at the ability of OAR to decrease need for x ray after instructing family practitioners in the ER setting on the use of these rules	Inclusion Fracture <3mm Exclusion Pregnancy, open injury, presentation >1 week after injury, enrolment one time per patient	100% sensitivity (95% CI=0.87-1.0). Specificity 28% (95% CI=0.14-0.39)	Small study, no children <9 years old and only five were younger than 12 years old. This limited the issue of growth plate fractures
Plint AC et al, 1999	670 patients aged 2-16 years	Patients were evaluated by staff and fellows trained in OAR at two hospital Eds; x rays were obtained based on each hospital's practices. Data forms with physical exam findings were filled out prior to viewing the x ray. The principal investigator reviewed the data forms and made a decision regarding positive or negative OAR	Inclusion Present with injury within 48 hours, fractures =>3mm Exclusion Salter-Harris I, nonsignificant fractures defined as <3 mm, <2 years old, multiple injuries, obvious open fractures, neurovascular compromise, diseases predisposing to fractures (OI), underlying disease with sensory/neural abnormalities (spina bifida), isolated injuries of the skin, patients returning for reassessment of the same injury, patients referred to the ED with x rays, intoxication	Sensitivity 100% (95% CI = 0.58–1.0) Specificity 27% (95% CI = 0.11–0.42)	119 Salter-Harris I fractures, 32 nonsignificant fractures [*when calculating the 2x2 table, 96 patients were counted twice (once for ankle fractures and a second time for foot fracture) therefore the N in this study was 766]. Mid-foot injuries were included in this study Inter-rater reliability was not assessed
Libetta C et al, 1999	761 patients aged 1-15 years	A historical control group was included prior to the implementation of OAR in this prospective evaluation as a comparison to predict need for x ray	Inclusion Patients that had ability to walk prior to injury Exclusion Patients were excluded in August in order to give the staff one month to learn and implement the Ottawa ankle rules	Sensitivity 98% (95% CI = 0.95–1.0) Specificity 46% (95% CI = 0.43–0.51)	Small number of children <5 years old. Total of 57 children out of 761 patients. Mid-foot injuries were included in this study
Karpas et al, 2002	190 patients evaluated ages 5–19 years	Blinded cross-sectional study that implemented OAR after two nurse training sessions	Inclusion Patients who presented within 48 hours of injury and all fractures Exclusion Open fracture, multiple traumas, developmental delay, referral with x ray, recurrent visits for the same injury in the last 2 weeks	Sensitivity 96% (95% CI = 0.82–0.99) Specificity 27% (95% CI = 0.18–0.32)	Study included one patient with Salter-Harris I and negative rules
Cuello-Garcia et al, 2004	111 patients evaluated ages 3–18 years	Prospective evaluation by paediatric nurses, third year residents, and attendings in the ER. OAR was applied, and x rays obtained at physician discretion. Radiology was blinded to OAR results	Inclusion Salter-Harris II–V Exclusion Multiple trauma, >7 days from event, changes in consciousness, bony disease, patients who came for reevaluation, Salter-Harris I fractures	Sensitivity 100% (95% CI = 0.95–1.0) Specificity 6% (95% CI = 0.01–0.11)	Salter-Harris I fractures were not included; there were 18 of these total. Patients were followed up at one month with telephone calls, and none of the patients showed later complications or changes in the diagnosis
Clarke and Tanner, 2003	160 patients evaluated ages 0–18 years old	Prospective survey with 22 variables; x rays were obtained on all patients with radiologists being blinded to survey results	Inclusion All types of fractures Exclusion Age >18, intoxication, previous films, pregnancy, suspected physical abuse, open fractures, OI, metabolic disease, patient's without phone contact, neurologic impairment	Sensitivity 83% (95% CI = 0.65–0.94) Specificity 50% (95% CI = 0.41–0.58)	There was only case in a child <5 years that was a true negative for rules and fracture, and no true positives
Boutis et al, 2001	607 patients evaluated ages 3–16 years old	Blinded prospective study in 2 similar urban emergency departments with fellows and attending staff as participants. Instruction on the use of OAR was given by orthopaedic surgeons prior to start of study	Inclusion Isolated ankle trauma within 72 hours of injury Exclusion Age <3 years and >16 years, preexisting musculoskeletal disease, coagulopathy, developmental delay, previous history of surgery or recent 3 months injury of affected ankle or multi-system trauma	Sensitivity 100%(95% CI = 0.96–1.0) Specificity 13%(95% CI = 0.11–0.16)	Patients were divided into low risk and high risk groups. Low risk consisted of isolated pain, tenderness, or both with or without oedema or ecchymosis of the distal fibula below the level of the joint line of the ankle. All other findings were classified as high risk. They also assessed the potential for reduction in radiographs when comparing the low risk clinical findings with those obtained by combining the Ottawa ankle rules

Comment(s)

The physical examination findings for the Ottawa ankle rules are as follows: tenderness over the lateral malleoli, inability to bear weight, and tenderness over the posterior distal tibia and fibula. A patient that exhibits one of these characteristics is deemed in need of x ray examination. The OAR have been validated for use as a screening tool in adults who have sustained ankle or mid-foot injuries (Bachman). Three considerations render the applicability of OAR to children less certain. Children may not be as reliable with regard to verbal history. Because Salter-Harris type I fractures, defined as a separation of bone >3 mm through the physis, more commonly accompany trauma in infants and children, point tenderness will generally be present. Further, a child must be able to walk freely prior to injury, in order for the OAR to be applied. Thus the OAR criteria will be positive and unnecessary radiographs may be obtained for an injury that will ultimately be treated the same as a sprain. Data analysis: We computed a random effects meta-analysis model directly on the proportions with weights based on the variance of a binomial distribution. We used a pooled estimate of sensitivity/specificity, instead of individual sensitivities/specificities for each study. Statistical calculations were made using the meta library, version 0.5, with the R software package, version 2.01 (R Foundation for Statistical Computing, Vienna, Austria). Formulas from Evidence-Based Medicine text by Sackett were used to calculate prevalence, likelihood ratios, post-test odds, PPV, and NPV. Main results: The overall sensitivity was calculated to be 97% with confidence limits of 93%–100%. The overall specificity was calculated to be 29% with confidence limits of 18%–40%. An estimated prevalence of 12% was calculated based on the number of fractures in the studies divided by the total number of patients. The prevalence and likelihood ratio were then used to derive the PPV and NPV. There was one article (Clarke) that showed five patients with negative results when applying the rules who ultimately had a fracture. All other articles had zero or 1 in this category. Using the Ottawa ankle rules has relevance in the clinical setting; as it is a tool that can be used to aid the clinician in decreasing unnecessary x ray examinations. This may very well decrease patient care costs, as well as patient time spent in the acute care setting. A small percentage of patients that are excluded from receiving x ray evaluation based on the Ottawa ankle rules, will actually have a fracture. It is a low percentage of patients at 1.4%. These missed fractures will often be of little clinical significance, as many of them will represent the Salter-Harris I classification. While there may be no long term consequences to these missed fractures, each clinician must decide their comfort level in applying the rules to individual patients.

Editor Comment

Yuen Man-Cheuk. The Ottawa ankle rules in children. Emerg Med J. 2001;18:466-7. 25/11/05 - The Ottawa ankle rules (OAR) are a set of guidelines to help the physician as to decision making regarding need for x ray examination after ankle and mid-foot injury. This best evidence topic report (previously completed by Dy Yeun and checked by Dr Saunders ) examined whether these rules could be applied to children. At that time there was insufficient evidence to make a determination. This appraisal updates that topic

Clinical Bottom Line

These rules are meant to be applied to those patients who have the ability to walk prior to their injury, and can localise pain with verbal communication. (grade A) Negative results when applying the rules should help the physician to decrease x ray usage without an increase in missed fractures. (grade A) For every 1000 patients that exhibit negative Ottawa ankle rules, 14 will actually have fractures. (grade A)

References

1. Chande VT. Decision rules for roentgenography of children with acute ankle injuries. Arch Pediatr Adolesc Med 1995;149:255-8.
2. McBride KL. Validation of the Ottawa ankle rules. Experience at a community hospital. Can Fam Physician 1997;43:459-465
3. Plint AC, Bulloch B, Osmond MH, et al. Validation of the Ottawa ankle rules in children with ankle injuries. Acad Emerg Med 1999;6(10):1005-9.
4. Libetta C, Burke D, Brennan P, et al. Validation of the Ottawa ankle rules in children. J Accid Emerg Med 1999;16(5):342-4.
5. Karpas A, Hennes H, Walsh-Kelly CM. et al. Utilization of Ottawa ankle rules by nurses in a pediatric emergency department. Acad Emerg Med 2002;9:130–3.
6. Cuello-Garcia C, Ruiz-Flores A, Ramos-Gomez L. et al. The Ottawa ankle rules: evaluation in an emergency department in Mexico [in Spanish]. Pediatr (Barc) 2004;60:454–8.
7. Clarke KD. Tanner S. Evaluation of Ottawa ankle rules in children. Pediatr Emerg Care 2003;19:73–8.
8. Boutis K, Komar L, Jaramillo D. et al. Sensitivity of clinical examination to predict need for radiography in children with ankle injuries: a prospective study. Lancet 2001;358:2118–21.
9. Laupicis A. Sekar N, Stiell AG. et al. Clinical prediction rules: a review and suggested modifications of methodological standards. JAMA 1997;277:488–94.
10. Sackett D. Straus S, Richardson WS. et al. Evidence-based medicine: how to practice and teach EBM, 2nd edn. Edinburgh: Churchill Livingstone 2000:247.
11. Bachman LM, Kolb E, Koller MT. et al. Accuracy of Ottawa ankle rules to exclude fractures of the ankle and mid-foot: systematic review. BMJ 2003;326:417.