Factors predicting death and cancer in patients with giant cell arteritis in Western Norway 1972–2012: a retrospective observational cohort study

OBJECTIVES: Evidence as to whether or not giant cell arteritis (GCA) confers added risk of cancer or death is conflicting. Our aim was to identify factors predicting death or cancer in a large Norwegian GCA-cohort. METHODS: This is a retrospective observational cohort study including patients diagnosed with GCA in Western Norway during 1972–2012. Patients were identified through computerized hospital records using the International Classification of Diseases coding. Medical records were reviewed and data about registered deaths and cancer occurrences were extracted from the Norwegian Cause of Death Registry and the Cancer Registry of Norway. We investigated predicting factors using Cox proportional hazards regression. RESULTS: We identified 881 cases with a validated diagnosis of GCA (60% biopsy-verified). 490 patients (56%) died during the study period. Among 767 patients with no registered cancer prior to GCA diagnosis, 120 (16%) were diagnosed with cancer during the study period. Traditional risk factors were the main predictors of death; age at time of GCA-diagnosis [hazard ratio (HR) 2.81], smoking (HR 1.61), hypertension (HR 1.48) and previous cardiovascular disease (HR 1.26). Hemoglobin (Hb) level was also associated with risk of death with increasing Hb-levels at time of GCA-diagnosis indicating decreased risk of death (HR 0.91). Other GCA-related factors were not predictive of death. We did not identify any predictors of cancer risk. CONCLUSION: In our cohort of GCA-patients, the risk of death was predominantly predicted by age and traditional risk factors. We found no significant associations with regards to the risk of incident cancer.