Diagnostic Accuracy of the History, Physical Examination, and Laboratory Testing for Giant Cell Arteritis

van der Geest KSM, Sandovici M, Brouwer E, Mackie SL. Diagnostic Accuracy of Symptoms, Physical Signs, and Laboratory Tests for Giant Cell Arteritis: A Systematic Review and Meta-analysis. JAMA Intern Med. 2020;e203050.

Study Population: 68 studies comprising 14,037 patients suspected of Giant cell arteritis

Giant cell arteritis (GCA), an inflammation of temporal arteries, can lead to multiple complications including vision loss if not diagnosed promptly.1 Unfortunately, diagnoses can often be delayed, particularly in those without classic features.2 Once the diagnosis is suspected, patients are typically started on high-dose glucocorticoids and referred for urgent temporal artery biopsy (TAB) or advanced imaging (e.g., computed tomography, magnetic resonance imaging). Unfortunately, high-dose glucocorticoids have significant side effects. Moreover, TAB is not sensitive enough to exclude the disease.3^, 4 Therefore, it is important to incorporate the pre-test probability of GCA by using the history, examination, and laboratory findings in the clinical decision-making.

The systematic review discussed here included studies that evaluated historical features, physical examinations findings, and laboratory testing to predict the risk of GCA.5 The systematic review included studies that enrolled a consecutive sample of patients suspected of having GCA. Temporal artery biopsy, imaging, or clinical diagnosis were used as the reference standard for GCA. Studies should have included at least 5 patients with GCA and at least 5 patients that did not have GCA. Risk of bias was assessed with the quality assessment of diagnostic accuracy studies (QUADAS-2) tool.

The systematic review identified 68 studies (n = 14,037 patients), of whom 4,277 (30.5%) were classified as having GCA. Temporal artery biopsy was utilized as the reference standard in 38 out of 68 studies, while clinical diagnosis was the reference standard for the remainder.

The following findings were associated with a higher pre-test probability of GCA: limb claudication (positive likelihood ratio [LR+] 6.01; 95% confidence interval [CI] 1.38-26.16), jaw claudication (LR+ 4.90; 95% CI 3.74-6.41), temporal artery thickening (LR+ 4.70; 95% CI 2.65- 8.33), temporal artery loss of pulse (LR+ 3.25; 95% CI 2.49-4.23), platelet count of greater than 400 × 10³/μL (LR+ 3.75; 95% CI 2.12-6.64), temporal tenderness (LR+ 3.14; 95% CI 1.14- 8.65), and erythrocyte sedimentation rate (ESR) >100 mm/h (LR+ 3.11; 95% CI 1.43-6.78). Findings associated with absence of GCA included an ESR <40 mm/h (negative likelihood ratio [LR-] 0.18; 95% CI 0.08-0.44), a C-reactive protein level <2.5mg/dL (LR- 0.38; 95% CI 0.25- 0.59), and age less than 70 years (LR- 0.48; 95% CI 0.27-0.86).

The included studies had clinical heterogeneity with regard to patient populations and pre-test probabilities. Most studies were conducted in academic medical centers and none were conducted in the Emergency Department (ED), which may reduce generalizability to the ED or non-academic centers. Additionally, studies differed with regard to the gold standard, which could include TAB, imaging, or clinical diagnosis per the inclusion criteria. Those receiving TAB may have been subject to selection bias because a higher index of suspicion is generally needed to order that test due to the invasive nature of it. Moreover, TAB has imperfect sensitivity, so it is possible that some patients may have been misclassified if the TAB was a false negative (particularly if the portion of the temporal artery that was biopsied was not yet involved or patients were already on glucocorticoids for an extended period of time). Alternatively, those diagnosed clinically were at risk of verification bias because the index test contributed to the diagnosis. Some of the symptoms such as jaw claudication were not clearly defined in the included studies, which may artificially increase the LR+ of this feature. Finally, the systematic review included retrospective studies, which are subject to significant limitations. However, a sensitivity analysis assessing only prospective studies demonstrated comparable likelihood ratios.

In summary, the existing evidence indicates that several findings can significantly increase the pre-test probability of GCA. The strongest predictor was limb claudication, followed by jaw claudication, temporal artery thickness or loss of pulse, platelet count >400 × 10³/μL, temporal tenderness, and ESR >100 mm/h. However, the absence of these findings does not sufficiently reduce the post-test probability to exclude further work up or testing.

February 3, 2021

LR, pretest probability and posttest (or posterior) probability are daunting terms that describe simple concepts that we all intuitively understand.

Let's start with pretest probability: that's just a fancy term for my initial impression, before we perform whatever test it is that we're going to use.

For example, a patient with prior stents comes in sweating and clutching his chest in agony, I have a pretty high suspicion that he's having an MI – let's say, 60%. That is my pretest probability.

He immediately gets an ECG (known here as the "test") showing an obvious STEMI.

Now, I know there are some STEMI mimics, so I'm not quite 100%, but based on my experience I'm 99.5% sure that he's having an MI right now. This is my posttest probability - the new impression I have that the patient has the disease after we did our test.

And likelihood ration? That's just the name for the statistical tool that converted the pretest probability to the posttest probability - it's just a mathematical description of the strength of that test.

Using an online calculator, that means the LR+ that got me from 60% to 99.5% is 145, which is about as high an LR you can get (and the actual LR for an emergency physician who thinks an ECG shows an obvious STEMI).

(Thank you to Seth Trueger, MD for this explanation!)