Clinical symptoms vary widely when comparing patients with temporal arteritis. Reports also differ in terms of the association of a particular symptom with a positive biopsy. A few studies have reported a link between jaw claudication and giant cell arteritis, such as one study from the Mayo Clinic, which showed a 78% positive predictive value [12]. In line with this association, an Odds Ratio of 8.1 for a positive biopsy result in the presence of jaw claudication was found in our study. Our opinion is that clinicians should have a high index of suspicion for patients presenting with jaw claudication. It should also be kept in mind that temporal arteritis can have many different presentations.

The effect of steroid treatment on the biopsy positivity rate is a subject of debate. A cohort study of 535 patients from the Mayo Clinic did not demonstrate any difference between the positivity rates of patients treated with steroids for 14 days or more versus those treated for less than 14 days at the time of biopsy [13]. Other studies have shown a negative impact of prednisone treatment. The positive rate ranged from 78% (< 2 weeks treatment) to 40% (> 4 weeks treatment) in another study [14]. In our study, there was no association between the use of steroid treatment and specimen positivity. This can potentially be attributed to the short delay between referral and biopsy. Almost all of the biopsies were performed in less than 2 weeks. Indeed, most were sent for identification in less than a week, as shown by the mean delay of 4.2 days. This is also highlighted by the fact that in general, steroid treatment was only initiated 4.6 days before biopsy.

A few studies have demonstrated that longer specimens are associated with higher rates of positivity [15]. However, the ideal tissue biopsy length remains unknown.

Our study did not show any significant results with regard to total length. A recent study concluded that length “is not associated with the temporal artery biopsy yield in patients with clinical suspicion of giant cell arteritis” [16]. However, an older retrospective study stated that the positivity rate was significantly higher when the size of the biopsy exceeded 0.7 cm [15]. Another study showed the mean length of biopsy to be 1.84 cm for positive biopsies and 1.29 cm for negative biopsies [17]. Surgeons should consider an average loss of 2.4 mm with specimen fixation [18].

The positivity rate of temporal artery biopsy at our institution is 23.7% (16.6–32.6%). A recent review pooling results from different studies revealed a median yield of 25% for temporal artery biopsy [9]. The interquartile range was 17–33%, suggesting that centers with yields under 17% are overperforming temporal artery biopsies, while those over the 33% limit may be underperforming the procedure.

Only one patient needed a second biopsy for a suspicion of relapse. It has been reported that bilateral biopsies increase the sensitivity of the procedure by 5% [19]. In our center, no bilateral biopsies were reported.

Treatment of patients with a negative temporal artery biopsy was associated with maintenance of corticosteroid treatment when the initial clinical suspicion of arteritis was high. Therefore, temporal artery biopsy may not be necessary for patients with a high initial clinical suspicion of giant cell arteritis.

We initially believed that the diagnostic criteria from the American Rheumatology Association, were a good outline for clinical practice. This has already been advocated in other studies, such as “The Role of Temporal Artery Biopsies in Giant Cell Arteritis”, by Davies and May [8]. In another study, the same group also noted a two-thirds reduction in the number of biopsies when using these criteria [20]. On the other hand, a recent systematic review (2019) performed a meta-regression and reported that the Rheumatology Association criteria did not improve the yield of biopsy. This study also states that these criteria were not intended for diagnostic use and that they are probably not accurate for patients with an ophthalmic subset of temporal artery vasculitis [9]. Our findings are in concordance with the previous study. This emphasizes the diagnosis of temporal arteritis as a clinical diagnosis. Moreover, the results of our study show that doctors are good at suspecting the disease without the need for a biopsy. However, there is a benefit of biopsy for patients for whom the diagnosis is uncertain. As shown in this study, when the initial clinical suspicion is high, regardless of the number of criteria and the results of the biopsy, treatment is typically pursued. This is highlighted by the fact that biopsy results only affect management in approximately 15% of patients [21]. Therefore, our opinion is that temporal artery biopsy should only be used in specific cases. Similar findings have also been published recently (2019) in the plastic surgery literature [22].

The use of Doppler examination represents an ongoing area of research, and its use is becoming more popular. Some studies have reported similar sensitivity results when comparing ultrasound to biopsy [23]. Initial studies led towards the use of ultrasound as a screening method before performing biopsy. Notable disadvantages are that the exam is operator dependent. In our study, the use of Doppler ultrasound was documented but remained scarce. This underlines the fact that ultrasound is not yet widespread in modern medical practice. In our study, both ultrasound and biopsy were ordered simultaneously when ultrasound use was reported. We believe the choice of patients requiring ultrasound should be based on the same clinical grounds as with biopsy. MR studies for temporal arteritis have also been suggested as an alternative to temporal artery biopsy. We believe MRI access and costs might limit the use of this modality in the diagnosis and treatment of giant cell arteritis [24].

Although no major complications were reported, the risks and benefits of the procedure must be balanced when considering surgery. Complications, including the risk of facial nerve injury, can be devastating for a patient. To minimize this risk, dissection should be undertaken over the parietal extension of the temporal superficial artery, which is more posterior [25]. Other complications include bleeding and infection. That said, the risk of complications is low, but surgery should only be used for patients with diagnostic uncertainty. In our study, one case of delayed bleeding occurred. However, the number of minor complications reported in our study might have been underestimated as a result of patients not reporting directly to our establishment. For example, a patient could have consulted his primary care physician outside of the hospital setting for a minor infection. Other added benefits of limiting the number of procedures include decreased costs and improved resource allocation.

Lastly, the risks of long-term corticosteroid therapy are important to consider in the management of these patients. Major risks such as cardiovascular complications, diabetes mellitus, avascular necrosis and osteoporosis are well established [26]. Consequently, we advocate in favor of the procedure if it may alter the long-term treatment of a patient. In our opinion, this is also part of optimizing the use of temporal artery biopsy.