Introduction Some UK studies have suggested higher case-fatality rates following acute myocardial infarction (AMI) in British South Asian (SA), compared to white European (WE) people, driven by higher prevalence of diabetes in the SA ethnic group. However other studies have suggested similar or even better adjusted overall post-AMI survival for these ethnic groups. In patients with AMI, both prior diagnosis of diabetes as well as acutely elevated blood glucose regardless of diabetes status are associated with adverse outcomes. The aim of this study was to compare survival rates following AMI in SA and WE patients drawn from a contemporary, multi-ethnic UK population.
Methods: We conducted a retrospective cohort study of total 4111 (SA 18%) consecutive patients with AMI admitted between October 2002 and September 2008. Baseline differences between the ethnic subgroups were examined using independent two-sample t tests for continuous and χ2 tests for categorical variables. Cox regression models were constructed to identify determinants of 30-days and 1-year mortality, entering ethnicity, random admission blood glucose and antecedent diabetes individually and together along with other relevant variables.
Results: SA patients were younger (62 vs 67 years, p<0.005) and less likely to have smoked (16% vs 40%, p<0.005) but more likely to have hypertension (55% vs 49%, p=0.004) or diabetes (40% vs 16%, p<0.005) at presentation compared to WE patients. All cause 30-day and 1-year mortality proportions were 10.0 % and 15.2% in SA compared to respectively 9.9 % and 16.7 % in WE patients. For SA ethnicity, the univariate HR of 30-day mortality was 1.01 (95% CI 0.79 to 1.30) compared to WE ethnicity. On multivariate analysis (excluding antecedent diabetes and admission blood glucose) this association of SA ethnicity and mortality became significant (HR 1.56, CI 1.10 to 2.23) and remained so when antecedent diabetes was added to the analysis (HR 1.48, CI 1.03 to 2.13). However when admission blood glucose was added to the model, association of ethnicity with mortality became non-significant (HR 1.31, CI 0.86 to 1.99). Conversely each unit (mmol/l) increase in admission blood glucose was associated with 7% increase in mortality (HR 1.07, CI 1.04 to 1.10) in this model, after adjusting for all the covariates. Furthermore exclusion of ethnicity and antecedent diabetes from the model did not alter the predictive value of admission blood glucose (HR 1.08, CI 1.05 to 1.10). Similar associations were observed for 1-year mortality.
Conclusions Despite higher prevalence of diabetes in SA patients, their mortality post AMI was similar to WE patients. Furthermore, admission hyperglycaemia more so than antecedent diabetes was an important predictor of increased mortality post AMI. To improve survival, active management of admission hyperglycaemia should be considered in patients admitted with AMI, regardless of their diabetes status or ethnicity.
- acute myocardial infarction