Diagnostic delay for giant cell arteritis (GCA): a systematic review and meta-analysis
Giant cell arteritis (GCA) is the commonest form of large-vessel vasculitis. The diagnosis of GCA remains difficult for general practitioners (GP), with patients often presenting with non-specific and atypical symptoms. Such ambiguity can lead to delays in the diagnosis of GCA, which in-turn may result in the patient experiencing preventable and life-altering outcomes, including blindness. Furthermore, instances of missed GCA diagnosis are a major reason for medical litigation. Despite the seriousness of such diagnostic delay, the extent of this has seldom been the primary focus of GCA research. Without a clear understanding of the extent to which GCA diagnosis is delayed, it is difficult to quantify the current problem in primary care. Our objective was to determine the average time-period between the onset of symptoms and receiving a GCA diagnosis.
We conducted a systematic review and meta-analysis to identify research literature which has examined diagnostic delay of GCA and to determine the extent of this delay. Literature searches were conducted in the following bibliometric databases; MEDLINE, EMBASE, CINAHL, PsycInfo and ISI web of knowledge. A single reviewer initially performed a title screen; abstracts and then full text articles were subsequently reviewed by two reviewers. Final article selection was based on pre-specified inclusion criteria and from these, data on a multitude of factors was extracted. The primary outcome of interest was the ‘average number of weeks between onset of GCA symptoms and GCA diagnosis’. Where diagnostic delay was reported as ‘days’ or ‘months’, data were converted to ‘weeks’ to provide a standardised dataset for analysis. Random-effects meta-analysis was used to report the pooled mean number of weeks (95% confidence interval (CI)) between symptom onset and GCA diagnosis.
4,128 articles were initially identified, 185 were reviewed in full and 34 articles were included in the final systematic review. Of these, the average age ranged from 65.2 to 81.6 years and GCA samples from 31 articles were recruited from secondary care. Delay was determined by the article reporting ‘how many days, weeks or months had occurred between GCA symptom onset and receiving a diagnosis of GCA’. 16 articles were included in the meta-analysis, resulting in a mean number of weeks between symptom onset and GCA diagnosis of 8.87 (95% CI 6.4 to 11.3) (I2 = 95.8%, p<0.001).
On average, patients experience approximately a 9-week delay between the onset of their symptoms and receiving a diagnosis of GCA. The reasons for this are yet to be understood, but could provide important insight and inform future strategies to improve outcomes for patients. Our research provides the current benchmark for diagnostic delay of GCA for which future efforts to reduce this problem can be measured against.