Does this patient have temporal arteritis?

Article Outline

• Rational clinical examination review source
• Objective
• Data sources
• Study selection
• Data extraction and analysis
• Main results
• Conclusions
  • Rational clinical examination author contact
• Commentary: Clinical implications
• Take home message
  • EBEM commentator contact
• Evidence-based medicine teaching point
• References

Back to Article Outline

Rational clinical examination review source 

This is a rational clinical examination abstract, a regular feature of the Annals' Evidence-Based Emergency Medicine (EBEM) series. Each features an abstract of a rational clinical examination review from the Journal of the American Medical Association and a commentary by an emergency physician knowledgeable in the subject area.

The source for this rational clinical examination review abstract is: Smetana G, Shmerling R. The rational clinical examination: does this patient have temporal arteritis? JAMA. 2002;287:92-101.

The Annals' EBEM editors assisted in the preparation of the abstract of this rational clinical examination review as well as selection of the Evidence-Based Medicine Teaching Points.

Back to Article Outline

Objective 

To review and summarize the accuracy and precision of historical features, physical examination, and erythrocyte sedimentation rate in the diagnosis of temporal arteritis.

Back to Article Outline

Data sources 

The authors performed a MEDLINE search of English-language articles from 1966 to 2000 combining topic-specific search terms with a search strategy developed for use in the rational clinical examination series. They also hand searched bibliographies of retrieved articles, previous reviews, monographs, and textbooks. Two investigators worked independently to determine eligibility, retrieve data, and assess validity.

Back to Article Outline

Study selection 

The authors selected studies that provided detailed clinical information on patients with suspected temporal arteritis referred for temporal artery biopsy. They determined 41 of 114 studies to be eligible for review. Twenty-one of these studies included both biopsy-positive and biopsy-negative patients and formed the core of the review. The majority (17 of 21 studies) were retrospective chart reviews.

Back to Article Outline

Data extraction and analysis 

The authors classified the quality of evidence in each study using 2 methods, developing their own detailed criteria, as well as using a scheme previously adapted for the rational clinical examination series. Eleven of 21 studies were of the highest quality (study quality 1) on the basis of these criteria. The authors of the review calculated sensitivity, specificity, and likelihood ratios for selected historical and physical examination features only in patients ultimately undergoing temporal artery biopsy. The 2 investigators worked independently to determine eligibility, retrieve data, and assess validity for each study.

Back to Article Outline

Main results 

The prevalence of biopsy-positive temporal arteritis was 39% in the included studies compared with less than 1% in the general population.¹ No studies evaluated the precision or interobserver variation of history and physical examination findings for the diagnosis of temporal arteritis.

Jaw claudication and diplopia had likelihood ratios of 4.2 and 3.4, respectively. They were present in 34% and 9%, respectively, of patients who were determined to have temporal arteritis. Abnormalities of the temporal artery on physical examination had likelihood ratios that ranged from 2.6 for tenderness over the artery to 4.6 for beading. The presence of any temporal artery abnormality had a likelihood ratio of 2, and the absence of any abnormality had a likelihood ratio of 0.53. See the Table for likelihood ratios and sensitivities in selected findings. Only 2 of 1,435 patients with biopsy-proven temporal arteritis were younger than 50 years.

Table. Positive and negative likelihood ratios and sensitivities for selected symptoms and signs among patients with positive temporal artery biopsy results.

LR+, Positive likelihood ratio; LR−, negative likelihood ratio; ESR, erythrocyte sedimentation rate.

Several studies examined the accuracy of erythrocyte sedimentation rate. A “normal” erythrocyte sedimentation rate, which was not consistently defined in these studies, had a likelihood ratio of 0.2. An “abnormal” erythrocyte sedimentation rate was present in 96% of patients with a positive biopsy, but had a likelihood ratio of 1.1 among patients with suspected temporal arteritis. When the cutoff of normality of erythrocyte sedimentation rate was defined as greater than 50 mm/h, the likelihood ratios for abnormal and normal sedimentation rate were 1.2 and 0.35, respectively.

Back to Article Outline

Conclusions 

The authors conclude that many of the historical and physical examination features commonly found in patients with temporal arteritis have limited value in predicting the presence of disease as defined by positive biopsy results and that the verification bias inherent in the data makes the significance of the results greater. Among the subset of patients suspected of having temporal arteritis and referred for biopsy, certain clinical features did modify the likelihood of disease, such as jaw claudication, diplopia, and abnormal artery findings on physical examination. A small number of clinical features are helpful in predicting the likelihood of a positive temporal artery biopsy among patients with a clinical suspicion of disease. The most useful finding is a normal erythrocyte sedimentation rate, which makes temporal arteritis unlikely.

Rational clinical examination author contact 

Gerald Smetana, MD

Division of General Medicine and Primary Care

Beth Israel Deaconess Medical Center

Boston, MA

Back to Article Outline

Commentary: Clinical implications 

The emergency physician faces a difficult challenge in approaching the patient with suspected temporal arteritis. Traditional teaching states that temporal arteritis should be suspected in patients aged older than 50 years presenting with co ncerning features on the history and physical examination such as fever, new headache, visual complaints, or jaw/tongue claudication. Prompt identification, referral for biopsy, and institution of treatment are paramount in preventing the most severe complication of the disease, irreversible blindness. Therefore, the challenge lies in determining which features significantly increase the probability of temporal arteritis.

This review presents useful data for emergency physicians, with the caveat that the data are largely derived from retrospective reviews and the population is one of very high disease prevalence (those being referred for temporal artery biopsy). In the patient history, jaw claudication and diplopia are important historical elements. Although their absence does not make the diagnosis of temporal arteritis less likely, their presence is significant in predicting presence of disease with sufficient power to be clinically useful. On physical examination, abnormalities of the temporal artery examination should be sought (ie, beading, prominence, tenderness, absence of pulse), because the likelihood of disease is substantially increased in their presence. Among laboratory tests, whereas a “normal” erythrocyte sedimentation rate (or erythrocyte sedimentation rate <50 mm/h) rendered temporal arteritis significantly less likely, a high erythrocyte sedimentation rate was less predictive of disease.

Finally, the diagnosis of temporal arteritis is only the first in a line of clinical decisions required of emergency physicians in the course of evaluating patients with a chief complaint of headache. When to begin treatment is another important clinical decision, and no clear therapeutic strategy is advocated by the authors of the review. Many clinicians initiate corticosteroid therapy in the patient with suspected temporal arteritis when the referral for biopsy is made. The authors of this study acknowledge that no constellation of signs and symptoms has been derived that provides sufficient predictive value to justify or safely preclude immediate treatment. Therefore, although the study does offer worthwhile data on how to refine the evaluation of a patient with suspected temporal arteritis, the decision to treat, and if so, when to treat, is still left to the clinician.

Back to Article Outline

Take home message 

Emergency physicians considering a diagnosis of temporal arteritis should seek a history of jaw claudication and/or diplopia and carefully examine the temporal artery for abnormality, because these features significantly increase the likelihood of the diagnosis.

Although a normal erythrocyte sedimentation rate makes the diagnosis far less likely, an abnormal erythrocyte sedimentation rate is a sensitive but nonspecific finding. Virtually all individual features demonstrated poor sensitivity, and the lack of any feature on history or physical examination does not effectively rule out temporal arteritis in patients older than 50 years. Temporal artery biopsy remains the criterion standard for diagnosis and should be promptly arranged and corticosteroid treatment initiated in patients suspected of the disease.

EBEM commentator contact 

Corinne R. Widico, MD

Alameda County Medical Center–Highland Hospital

Oakland, CA

E-mail corinnewidico@yahoo.com

Back to Article Outline

Evidence-based medicine teaching point 

Verification bias. The data from this review represent an excellent example of workup bias, also called verification bias. Verification bias exists when the decision to perform the reference standard (also referred to as the “gold standard,” in this case the temporal artery biopsy) is influenced by the results of other tests. In this case, physical examination, historical findings, and erythrocyte sedimentation rate all acted as screening tests and led to referral of patients for temporal artery biopsy. This biases the results and should prompt the reader to examine a study to see if other means of measuring outcome were used in those patients who did not receive the invasive test. For example, in invasive testing such as for pulmonary embolism, the reference standard of pulmonary angiogram is often not performed on patients who have a negative workup. In high-quality diagnostic studies, patients should be followed up for several months afterward to ensure that the negative assessment was valid, particularly in those patients who did not receive the reference standard test. The clinical outcome then acts as an appropriate “proxy measure” of outcome.

Back to Article Outline

References 

1. Gonzalez-Gay MA. Epidemiology of the vasculidities. Rheum Dis Clin North Am. 2001;27:729–749
   • View In Article
   • CrossRef
2. Smetana G, Shmerling R. The rational clinical examination: does this patient have temporal arteritis?. JAMA. 2002;287:92–101
   • View In Article
   • MEDLINE
   • CrossRef