Association between diabetes mellitus and incidence and case-fatality of spontaneous primary intracerebral haemorrhage
MetadataShow full item record
Background–Whether diabetes mellitus (DM) is a risk factor for spontaneous (non-traumatic) intracerebral haemorrhage (ICH) and for death after ICH remains unclear. The aims of this dissertation are firstly to examine the associations between DM and the occurrence of ICH and the outcome after ICH in a systematic review and a meta-analysis; and secondly to assess the associations between DM and the incidence of ICH and the 30-day case fatality after ICH in a Scottish nation-wide population-based cohort study. Methods – For this systematic review, studies investigating the associations between DM and ICH occurrence (incident and recurrent) or DM and ICH outcome were sought from 1980-2014. Each study’s eligibility, assessment of risk of bias, and data extraction were undertaken independently by two reviewers. Studies were combined using random effects meta-analysis. For this retrospective cohort study using linked pseudonymised national population-based data from a diabetes register, hospital and death records, we included people aged 40 to 89 years between 2004 and 2013. Incidence and relative risks (RRs) for incidence of ICH defined using the I61 code from the tenth revision of the International Classification of diseases were calculated in people with type 1 and type 2 DM compared to people without DM using standardisation and quasi-Poisson regression adjusting for age, sex and an area-based measure of socio-economic status (SES). Adjusted RRs for 30-day case-fatality after hospital admission were calculated using logistic regression and subsequent correction. Results –19 case-control studies (3,397 people with ICH and 5,747 people without ICH) found an association between DM and ICH occurrence (unadjusted OR 1.23, 95% CI 1.04 to 1.45), which did not differ between hospital-based and populationbased studies (pdiff=0.70), and was similar in the 16 studies that controlled for age and sex (unadjusted OR 1.15, 95% CI 0.95 to 1.40; I2=14%). This association was not identified in three population-based cohort studies in which ICH occurred in 38 (0.66%) of 5,724 people with DM and 448 (0.57%) of 78,702 people without DM (unadjusted RR 1.27, 95% CI 0.68 to 2.36). There was an association between DM and death by 30 days or hospital discharge in 18 cohort studies involving 813 people with DM and 3,714 people without DM (unadjusted RR 1.52, 95% CI 1.28 to 1.81). In Scotland, total ICH incidence per 100,000 person-years was 67.9 for type 1 DM, 46.8 for type 2 DM, and 41.5 for no DM. Age, sex and SES adjusted RRs for incident ICH were 1.74 (95% CI 1.36 to 2.19) for type 1 DM and 1.06 (95% CI 0.99 to 1.12) for type 2 DM. There was no interaction by DM status and sex for either type of diabetes (Pint=0.89 for type 1 DM and 0.77 for type 2 DM). Case fatality within 30 days after hospital admission for ICH was 44 % in people with type 1 DM, 38% in people with type 2 DM and 36% in people without DM. Age, sex and SES adjusted RRs for death were 1.35 (95% CI 1.01 to 1.70) for type 1 DM and 1.04 (95% CI 0.96 to 1.13) for type 2 DM. There was no interaction by DM status and sex for either type of diabetes (Pint=0.26 for type 1 DM and 0.82 for type 2 DM). Conclusion –Type 1 DM is associated with increased risk of ICH and case fatality after ICH. Further research is needed to identify whether the association is confounded by hypertension, use of anti-thrombotic drugs or other risk factors for ICH and to identify effective interventions for reducing the excess risk.