Kidney replacement therapy: trends in incidence, treatment, and outcomes of myocardial infarction and stroke in a nationwide Scottish study

Abstract Background and Aims Patients with kidney failure have a higher risk of cardiovascular disease compared with the general population. Whilst temporal trends of myocardial infarction and stroke are declining in the general population, these have not been evaluated in patients with kidney failure. This study aimed to describe national trends in the incidence, treatment, and outcomes of myocardial infarction and stroke in patients with kidney failure (i.e. on dialysis or with a kidney transplant) over a 20-year period, stratified by age and sex. Methods In this retrospective national data linkage study, all patients with kidney failure in Scotland (UK) receiving kidney replacement therapy between January 1996 and December 2016 were linked to national hospitalization, prescribing, and death records. The primary outcomes were the incidence of myocardial infarction and stroke, and subsequent cardiovascular death. Generalized additive models were constructed to estimate age-standardized, sex-stratified incidence rates and trends in cardiovascular and all-cause death. Results Amongst 16 050 patients with kidney failure [52 (SD 15) years; 41.5% women], there were 1992 [66 (SD 12) years; 34.8% women] and 996 [65 (SD 13) years; 45.1% women] incident myocardial infarctions and strokes, respectively, between January 1996 and December 2016. During this period, the age-standardized incidence of myocardial infarction per 100 000 decreased in men {from 4376 [95% confidence interval (CI) 3998–4785] to 1835 (95% CI 1692–1988)} and women [from 3268 (95% CI 2982–3593) to 1369 (95% CI 1257–1491)]. Similarly, the age-standardized incidence of stroke per 100 000 also decreased in men [from 1978 (95% CI 1795–2175) to 799 (95% CI 729–875)] and women [from 2234 (95% CI 2031–2468) to 903 (95% CI 824–990)]. Compared with the general population, the incidence of myocardial infarction was four- to eight-fold higher in patients with kidney failure, whilst for stroke it was two- to four-fold higher. The use of evidence-based cardioprotective treatment increased over the study period, and the predicted probability of cardiovascular death within 1 year of myocardial infarction for a 66-year-old patient with kidney failure (mean age of the cohort) fell in men (76.6% to 38.6%) and women (76.8% to 38.8%), and also decreased in both sexes following stroke (men, from 63.5% to 41.4%; women, from 67.6% to 45.8%). Conclusions The incidence of myocardial infarction and stroke has halved in patients with kidney failure over the past 20 years but remains significantly higher than in the general population. Despite improvements in treatment and outcomes, the prognosis of these patients following myocardial infarction and stroke remains poor.


Prescribing Information Service
Prescriptions Information Service holds data regarding every prescription dispensed in the community from 2009. 4 The dataset includes information about the dispensed drug, the patient, the prescriber, and the dispenser.Prescriptions Information Service does not hold information regarding prescriptions dispensed from in-hospital pharmacies or regarding prescriptions that are not collected by patients in the community.The completeness and accuracy of this dataset are expected to be high because dispensers do not receive payment for prescriptions until this information is submitted.

Supplementary table 1.
List of ICD-9/-10 codes employed in study.Abbreviations: BNF: British National Formulary; ICD: international classification of diseases; OPCS-4: Office of Population, Censuses and Surveys 4. *For chronic respiratory disease, the presence of a single ICD-10 code in SMR01 data OR a single BNF code in community prescription data in the 365 days preceding the incident event was sufficient to identify this comorbidity.

Supplementary table 5.
Crude outcomes at 1 and 3 years for patients with kidney failure and incident myocardial infarction (left panel) and stroke (right panel) between 1996 and 2016, grouped by KRT modality at time of event.

.
IRRs for stroke for patients aged 40-69 (left panel) and ≥70 (right panel) years old between 1996 and 2014.Predicted probability of cardiovascular mortality at 1 year following incident stroke with kidney failure between 1996 and 2016, according to KRT modality (i.e., hemodialysis, peritoneal dialysis, kidney transplant).Predicted probability of all-cause mortality at 1 year following incident myocardial infarction with kidney failure between 1996 and 2016, according to KRT modality (i.e., hemodialysis, peritoneal dialysis, kidney transplant).

table 2 .
Baseline characteristics of patients with kidney failure and incident myocardial infarction (left panel) and stroke (right panel) between 1996 and 2016, grouped by KRT modality at time of event.SIMD) was missing in 0.4% and 0.3%, primary kidney disease was missing in 0.1% and 0.1%, and KRT modality was missing in 1.0% and 0.7% of all patients with myocardial infarction and stroke, respectively.

table 3 .
Count data and age-standardized incidence rate per 100,000 of myocardial infarction (left panel) and stroke (right panel) in men and women with kidney failure between 1996 and 2016.

table 4 .
Baseline characteristics of patients with kidney failure and incident myocardial infarction (left panel) and stroke (right panel) between 2009 and 2016, in 2-year groups.

table 6 .
Values are median [interquartile range] or n (%).*The denominator for these percentages is derived from the number of non-fatal events in that KRT category.NAs represent redacted data that might be considered potentially identifiable, including count data ≤5.Predicted probability of cardiovascular mortality rate at 1 year (from generalized linear models) following incident myocardial infarction (left panel) and stroke (right panel) between 1996 and 2016, grouped by sex.Incidence of myocardial infarction in patients with kidney failure between 1996 and 2016, according to KRT modality (i.e., hemodialysis, peritoneal dialysis, and kidney transplant) at the time of incident event.IRRs for myocardial infarction for patients aged 40-69 (left panel) and ≥70 (right panel) years old between 1996 and 2014.