The incidence of cancer following hospitalisation for a burn injury in Scotland 2009-2019: A retrospective cohort study

Abstract

Background: Studies suggest increased occurrence of cancer in persons who have experienced a burn injury with hospital admission.

Objective: To determine the incidence of cancer among those hospitalised for burn injuries in Scotland compared with a similar group without a history of burn injury hospitalisation.

Method: A retrospective cohort design was used to compare cancer (ICD10 C00-97, excluding C44) incidence in two groups: 6805 burn injury patients discharged from Scottish hospitals between 2009 and 2019, and 25,946 subjects from the general population who were matched to burn patients by sex, year of birth, and degree of social deprivation. Cancer incidence was identified from the Scottish cancer registry. Cox proportional hazard regression was used to model time to cancer incidence adjusting for age, sex, degree of deprivation and presence of a comorbidity. Cancer risk was presented as standardised incidence ratios (SIRs) and hazard ratios (HR).

Results: We found a higher prevalence of pre-existing conditions, particularly alcohol abuse among patients with burns. Pre-existing cancers were more common in the burn cohort (3.5%) than the comparison group (1.7%) and were excluded from further analysis. Over a median follow-up of 4-5 years, a total of 236 (3.5%) burn patients and 969 (3.7%) persons in the comparison group were diagnosed with cancer. At 0-6 months the cancer SIR for burn patients was 1.88 95% CI (1.40-2.52). After excluding the first six months of follow-up, the overall incidence of cancer was marginally elevated in burn patients (SIR 1.04, 95% CI 0.90-1.19, p = 0.62) and not statistically different from the incidence in comparison subjects (adjusted HR 1.03, 95% CI 0.88-1.21, p = 0.71).

Conclusions: Patients that suffer burn injury have a higher incidence of cancer than the general population and a group matched by age, sex and degree of deprivation. A higher incidence of adverse health-related behaviours such as smoking, alcohol use and pre-existing health conditions among many patients that suffer a burn most likely explain this observed increase. Any persisting inflammatory or immune dysfunction following burn injury is unlikely to account for the increase in cancers in this study.

Keywords: Burn injury; Cancer incidence; ICU; Risk factors; Socio-economic status.