Article Text
Abstract
Background Cancer survival rates are poorer in England than in many other European nations, resulting in an estimated 6000 premature deaths per year in Britain. One factor associated with poorer survival for some cancers in England is emergency presentation, which may occur due to unmanageable symptoms and the complications of advanced disease. The frequency with which the emergency presentation occurs at diagnosis of aggressive lymphoma has not been examined. This disease is of interest, not only because it is often difficult to diagnose due to vague intermittent symptoms, but also because it is curable with early treatment. This study investigated the relationship between emergency presentation and survival in aggressive lymphoma using data from a specialist UK population-based registry (www.hmrn.org) linked to Hospital Episode Statistics (HES).
Methods All patients in the Haematological Malignancy Research Network are identified at diagnosis with core data (demographic, prognostic including performance status and disease stage, treatment) routinely abstracted from their hospital records. Adults with aggressive lymphoma (diffuse large B-cell lymphoma) diagnosed between September 2004 and March 2011 were linked to HES Admitted Patient Care to ascertain whether they were admitted as an emergency at lymphoma presentation. Patients were followed-up for three years, and survival for patients who presented as an emergency was compared to those that did not, adjusting for sex, age, performance status, disease stage and treatment. Analyses were conducted using Cox proportional hazards regression with a varying effect of emergency presentation on survival over time.
Results Of 1659 newly diagnosed patients, 39% presented as an emergency admission at diagnosis. This group had poorer survival than patients who did not present as an emergency, after one month (74.9% versus 95.9%), six months (54.7% vs 83.7%), a year (45.6% vs 76.4%) and three years (38.4% vs 65.6%). Emergency presentation remained related to poorer three-year survival after adjusting for outcome-related factors including age, poorer performance status, advanced disease, and treatment without curative intent (hazard ratio, HR = 1.59, 95% confidence interval, CI 1.36–1.89). Outcome was worse in the shorter term, mostly in the first month (HR = 3.74, 95% CI 2.61–5.36), but also over one to three (HR = 1.49, 95% CI 1.04–2.12), and three to six months (HR = 1.52, 95% CI 1.07–2.16), but not thereafter.
Conclusion For patients with aggressive lymphoma, emergency presentation is related to poorer survival, even after considering outcome-related variables. The effect is manifest close to diagnosis, and strategies to promote early detection and avoid emergency presentation should be considered.