Are the Ottawa knee rules reliable at identifying patients who require radiography?
Date First Published:
July 8, 2011
Last Updated:
July 15, 2011
Report by:
Bryony Patrick, Medical student (University of Manchester)
Search checked by:
Bryony Patrick, University of Manchester
Three-Part Question:
In [adult patients presenting with traumatic knee pain] are the [Ottawa knee rules] reliable at [identifying patients who require radiography]?
Clinical Scenario:
A 30 year old man presents to the ED with a painful knee due to a traumatic sporting injury. You suspect a soft tissue injury and wonder if he needs an x-ray?
Search Strategy:
Medline 1948 - July 2011 using the Pubmed interface
Updated search using OVID Medline 1948 – June Week 4 2011
Embase 1980 – Week 26 2011
CINAHL 1937 – July 2011
Cochrane database
Updated search using OVID Medline 1948 – June Week 4 2011
Embase 1980 – Week 26 2011
CINAHL 1937 – July 2011
Cochrane database
Search Details:
Medline 1948 – July 2011 using the Pubmed interface
Search terms:
(clinical[All Fields] AND decision[All Fields] AND rules[All Fields] AND ("knee"[MeSH Terms] OR "knee"[All Fields] OR "knee joint"[MeSH Terms] OR ("knee"[All Fields] AND "joint"[All Fields]) OR "knee joint"[All Fields])) AND ("humans"[MeSH Terms] AND English[lang] AND "adult"[MeSH Terms]). Found 11 results of which 9 were relevant.
(ottawa[All Fields] AND ("knee"[MeSH Terms] OR "knee"[All Fields] OR "knee joint"[MeSH Terms] OR ("knee"[All Fields] AND "joint"[All Fields]) OR "knee joint"[All Fields]) AND rule[All Fields]) AND ("humans"[MeSH Terms] AND English[lang] AND "adult"[MeSH Terms]) Found 12 results (all relevant) of which 8 were duplicates of other searches.
Updated search using OVID Medline 1948 – June Week 4 2011
Search terms:
‘clinical decision rules knee’ limited to English, humans, full text and related terms. ound 1516 results of which 2 relevant, both duplicates of previous search results.
‘ottawa knee rule’ limited to English, humans, full text, adults 19+ years and related terms. Found 218 results of which 3 relevant, 2 duplicates of previous searches.
Advanced search strategy:
1.tClinical decision rule.mpt(178)
2.tClinical decision rules.mpt(132)
3.tClinical prediction rule.mpt(270)
4.tClinical prediction rules.mpt(185)
5.tClinical decision rul$.mpt(285)
6.tClinical prediction rul$.mpt(426)
7.tOr/1-6tttt(702)
8.tExp Knee/ttt(9233)
9.tKnee.mpttt(86091)
10.tExp Knee Injuries/tt(13496)
11.tOr/8-10tttt(86222)
12.t7 and 11ttt(30)
Found 30 results of which 14 relevant, 10 found in previous searches.
Embase 1980 – Week 26 2011
Search terms:
‘Ottawa knee rule’ limited to related terms, humans, English, full text and adult <18 to 64 years> Found 423 results of which 5 relevant, 3 being duplicates of previous searches.
‘Pittsburgh knee rule’ limited to related terms, humans, English, full text and adult <18 to 64 years> Found 1260 results of which 4 relevant, 2 duplicates of previous searches.
CINAHL 1937 – July 2011
Search terms:
‘clinical decision rules’(title) AND ‘knee’(title) Found 4 results of which 2 were relevant, 1 being a duplicate of previous searches.
‘ottawa knee rule’(title) Found 13 results of which 6 relevant, 2 being duplicates of previous searches.
Cochrane database
‘clinical decision rule knee’ Found 5 results of which 3 relevant, 2 being duplicates of previous searches.
‘ottawa knee rule’ Found 0 results.
Search terms:
(clinical[All Fields] AND decision[All Fields] AND rules[All Fields] AND ("knee"[MeSH Terms] OR "knee"[All Fields] OR "knee joint"[MeSH Terms] OR ("knee"[All Fields] AND "joint"[All Fields]) OR "knee joint"[All Fields])) AND ("humans"[MeSH Terms] AND English[lang] AND "adult"[MeSH Terms]). Found 11 results of which 9 were relevant.
(ottawa[All Fields] AND ("knee"[MeSH Terms] OR "knee"[All Fields] OR "knee joint"[MeSH Terms] OR ("knee"[All Fields] AND "joint"[All Fields]) OR "knee joint"[All Fields]) AND rule[All Fields]) AND ("humans"[MeSH Terms] AND English[lang] AND "adult"[MeSH Terms]) Found 12 results (all relevant) of which 8 were duplicates of other searches.
Updated search using OVID Medline 1948 – June Week 4 2011
Search terms:
‘clinical decision rules knee’ limited to English, humans, full text and related terms. ound 1516 results of which 2 relevant, both duplicates of previous search results.
‘ottawa knee rule’ limited to English, humans, full text, adults 19+ years and related terms. Found 218 results of which 3 relevant, 2 duplicates of previous searches.
Advanced search strategy:
1.tClinical decision rule.mpt(178)
2.tClinical decision rules.mpt(132)
3.tClinical prediction rule.mpt(270)
4.tClinical prediction rules.mpt(185)
5.tClinical decision rul$.mpt(285)
6.tClinical prediction rul$.mpt(426)
7.tOr/1-6tttt(702)
8.tExp Knee/ttt(9233)
9.tKnee.mpttt(86091)
10.tExp Knee Injuries/tt(13496)
11.tOr/8-10tttt(86222)
12.t7 and 11ttt(30)
Found 30 results of which 14 relevant, 10 found in previous searches.
Embase 1980 – Week 26 2011
Search terms:
‘Ottawa knee rule’ limited to related terms, humans, English, full text and adult <18 to 64 years> Found 423 results of which 5 relevant, 3 being duplicates of previous searches.
‘Pittsburgh knee rule’ limited to related terms, humans, English, full text and adult <18 to 64 years> Found 1260 results of which 4 relevant, 2 duplicates of previous searches.
CINAHL 1937 – July 2011
Search terms:
‘clinical decision rules’(title) AND ‘knee’(title) Found 4 results of which 2 were relevant, 1 being a duplicate of previous searches.
‘ottawa knee rule’(title) Found 13 results of which 6 relevant, 2 being duplicates of previous searches.
Cochrane database
‘clinical decision rule knee’ Found 5 results of which 3 relevant, 2 being duplicates of previous searches.
‘ottawa knee rule’ Found 0 results.
Outcome:
A total of 6 papers were of sufficient quality to be included for critical appraisal.
Relevant Paper(s):
| Study Title | Patient Group | Study type (level of evidence) | Outcomes | Key results | Study Weaknesses |
|---|---|---|---|---|---|
| The accuracy of the Ottawa knee rule to rule out knee fractures Bachmann L.M., Haberzeth S., Steurer J., ter Riet G. 2004 USA | Systematic review of 11 papers on validation of the Ottawa rule. 5 excluded. Total of 6 papers included involving 4249 patients analysed. | Systematic review | Mean sensitivity | 98.50% | No testing for heterogeneity. |
| Mean specificity | 49.00% | ||||
| Mean negative likelihood ratio | 0.05 | ||||
| Probability of fracture after negative rule result | 0.37% (assuming 7% fracture prevalence rate) | ||||
| Implementation of the Ottawa knee rule for the use of radiography in acute knee injuries Stiell I.G., Wells G.A., Hoag R.H., Sivilotti M.L. et al 1997 Canada | Convenience sample totalling 3907 patients presenting to 4 hospitals. Two hospitals assigned controls and two implemented intervention (use of Ottawa rule) after 12 months. Most underwent radiology to determine sensitivity and specificity of rule, others followed up to check for missed fractures. Level of inter-observer agreement calculated. | Cohort | Sensitivity | 1 | No power calculation stated. Rule could only be interpreted for 987 patients as documents incomplete. Inter-observer agreement carried out on only 69 cases. One of intervention hospitals involved in creation and validation of rule previously, so staff may be more experienced using it. Interpretation of x-rays by radiologists not blinded. |
| Specificity | 0.48 | ||||
| Negative predictive value | 1 | ||||
| Positive predictive value | 0.11 | ||||
| Inter-observer agreement | K=0.91 | ||||
| Reduction in radiography: Intervention group | 26.40% | ||||
| Control group | 1.30% | ||||
| Validation of the Ottawa knee rules Emparanza J.I. and Aginaga J.R. 2001 Spain | 1522 patients assessed in 11 hospitals. All underwent radiology to determine sensitivity and specificity of rule. | Cohort | Sensitivity | 1 | Inter-observer agreement for interpretation of criteria not assessed. X-rays interpreted by physicians – no blinding. Participating physicians internists / family physicians who have unofficial training in Emergency Medicine. |
| Specificity | 0.52 | ||||
| Negative predictive value | 1 | ||||
| Positive predictive value | 0.11 | ||||
| Potential reduction in radiography | 49% | ||||
| X-ray requesting patterns before and after introduction of the Ottawa knee rules in a UK emergency department Atkinson P., Boyle A., Chisholm E. 2004 UK | 58 patients assessed by junior doctors with criteria completed after examination. Further 72 patients assessed for positive signs outlined in clinical decision rule after formal teaching. Most underwent radiography to assess sensitivity and specificity of rule. All followed up to check for missed fracture. | Cohort | Before rule knowledge: Sensitivity | 80% | Small sample size. Low initial rate and small reduction in radiography rates. Only included junior doctors. |
| Specificity | 43.40% | ||||
| Negative predictive value | 95.80% | ||||
| Positive predictive value | 11.80% | ||||
| No. of missed fractures | 1 of 5 | ||||
| After rule introduction: Sensitivity | 100% | ||||
| Specificity | 53.80% | ||||
| Negative predictive value | 100% | ||||
| Positive predictive value | 18.90% | ||||
| No. of missed fractures | 0 of 7 | ||||
| Reduction in radiography | 3% | ||||
| External validation of the Ottawa knee rules in an urban trauma center in the United States Tigges S., Pitts S., Mukundan Jr S., Morrison D., Olson M., Shahriara A. 1999 USA | Convenience sample of 378 patients assessed for rule criteria and radiology findings. | Cohort | Sensitivity | 0.98 | Selection and work-up bias for patient inclusion. Poor follow-up to assess for missed fractures. |
| Specificity | 0.19 | ||||
| Positive predictive value | 0.13 | ||||
| Negative predictive value | 0.98 | ||||
| Positive likelihood ratio | 1.21 | ||||
| Negative likelihood ratio | 0.12 | ||||
| Validation of the Ottawa knee rule in Iran: a prospective study Jalili M. and Gharebaghi H. 2010 Iran | Convenience sample of 283 patients assessed for rule criteria. All received radiography to calculate sensitivity and specificity rates of rule. | Cohort | Sensitivity | 95.40% | Inter-observer compliance not assessed. X-rays assessed by orthopaedic surgeon and all fractures considered significant (even if not clinically significant). |
| Specificity | 44.00% | ||||
| Negative predictive value | 99.10% | ||||
| Positive predictive value | 12.50% | ||||
| Negative likelihood ratio | 0.09% | ||||
| No. of missed fractures | 1 of 22 | ||||
| Reduction in radiography | 41% |
Author Commentary:
The Ottawa rule has been widely validated and its generalisability confirmed. It is a very reliable tool for accurately assessing whether a patient requires radiography and can be used by different levels of medically-trained staff effectively.
Bottom Line:
The Ottawa knee rule is an excellent clinical decision tool and should be used by triage nurses and physicians in emergency departments alongside clinical judgement and experience to ensure a high level of specificity and sensitivity is maintained.
References:
- Bachmann L.M., Haberzeth S., Steurer J., ter Riet G.. The accuracy of the Ottawa knee rule to rule out knee fractures
- Stiell I.G., Wells G.A., Hoag R.H., Sivilotti M.L. et al. Implementation of the Ottawa knee rule for the use of radiography in acute knee injuries
- Emparanza J.I. and Aginaga J.R.. Validation of the Ottawa knee rules
- Atkinson P., Boyle A., Chisholm E.. X-ray requesting patterns before and after introduction of the Ottawa knee rules in a UK emergency department
- Tigges S., Pitts S., Mukundan Jr S., Morrison D., Olson M., Shahriara A.. External validation of the Ottawa knee rules in an urban trauma center in the United States
- Jalili M. and Gharebaghi H.. Validation of the Ottawa knee rule in Iran: a prospective study