Using imputed pre-treatment cholesterol in a propensity score model to reduce confounding by indication: results from the multi-ethnic study of atherosclerosis

Women with familial hypercholesterolemia phenotype are undertreated and poorly controlled compared to men

Familial hypercholesterolemia (FH) is an autosomal dominant disease that has a prevalence of approximately 1/250 inhabitants and is the most frequent cause of early coronary heart disease (CHD). We included 1.343.973 women and 1.210.671 men with at least one LDL-c measurement from the Catalan primary care database. We identified 14.699 subjects with Familial hypercholesterolemia-Phenotype (FH-P) based on LDL-c cut-off points by age (7.033 and 919 women, and 5.088 and 1659 men in primary and secondary prevention, respectively). Lipid lower therapy (LLT), medication possession ratio (MPR) as an indicator of adherence, and number of patients that reached their goal on lipid levels were compared by sex. In primary and secondary prevention, 69% and 54% of women (P = 0.001) and 64% and 51% of men (P = 0.001) were on low-to-moderate-potency LLT. Adherence to LLT was reduced in women older than 55 years, especially in secondary prevention (P = 0.03), where the percentage of women and men with LDL-c > 1.81 mmol/L were 99.9% and 98.9%, respectively (P = 0.001). Women with FH-P are less often treated with high-intensity LLT, less adherent to LLT, and have a lower probability of meeting their LDL-c goals than men, especially in secondary prevention.

A methodological framework for tackling confounding by indication when assessing the treatment effects of Chinese herbal injections in the real world

AIM

In the context of integrative medicine, whether Chinese herbal injections are effective in routine practice has become a question of broad interest. However, confounding by indication (i.e., indication bias) is a prevalent and highly challenging methodological issue when using routinely collected health care data to assess the real-world effectiveness of Chinese herbal injections.

METHODS AND RESULTS

We proposed a methodological approach to tackling confounding by indication in assessing the real-world effectiveness of Chinese herbal injections, incorporating empirical experiences, a literature review and interactive discussions, and a panel of external experts to finally achieve a consensus. This approach consisted of three cohesive steps, including a full understanding of treatment patterns, construction of fair comparisons by identifying appropriate combination treatments and comparators, and using statistical methods to further control for confounding. In the investigation of treatment patterns, we proposed five domains to identify treatment patterns with Chinese herbal injections, and we offered five patterns of combination treatments to characterize how Chinese herbal injections are used in conjunction with other treatments. In constructing fair comparisons, we suggested the use of both nonuse and active comparators; given the diverse combination treatments, we developed six scenarios that may form fair comparisons. In the statistical analysis, we discussed five statistical models for controlling confounding by indication, including their pros and cons. We also included a practical example to illustrate the usefulness of the methodological approach.

CONCLUSION

The proposed approach may serve as an effective tool to guide researchers to reliably assess the effectiveness of Chinese herbal injections in the context of integrative medicine.

Hematoma Expansion and Clinical Outcomes in Patients With Factor-Xa Inhibitor-Related Atraumatic Intracerebral Hemorrhage Treated Within the ANNEXA-4 Trial Versus Real-World Usual Care

BACKGROUND AND PURPOSE

It is unestablished whether andexanet alfa, compared with guideline-based usual care including prothrombin complex concentrates, is associated with reduced hematoma expansion (HE) and mortality in patients with factor-Xa inhibitor-related intracerebral hemorrhage (ICH). We compared the occurrence of HE and clinical outcomes in patients treated either with andexanet alfa or with usual care during the acute phase of factor-Xa inhibitor-related ICH.

METHODS

Data were extracted from the multicenter, prospective, single-arm ANNEXA-4 trial (Andexanet Alfa, a Novel Antidote to the Anticoagulation Effects of Factor Xa Inhibitors) and a multicenter observational cohort study, RETRACE-II (German-Wide Multicenter Analysis of Oral Anticoagulant-Associated Intracerebral Hemorrhage - Part Two). HE was based on computed tomography scans performed within 36 hours from baseline imaging. Inverse probability of treatment weighting was performed to adjust for baseline comorbidities and ICH severity. Patients presenting with atraumatic ICH while receiving apixaban or rivaroxaban within 18 hours of admission were included. Patients with secondary ICH or not fulfilling the inclusion criteria for the ANNEXA-4 trial were excluded. We compared ANNEXA-4 patients, who received andexanet alfa for hemostatic treatment, with RETRACE-II patients who were treated with usual care, primarily administration of prothrombin complex concentrates. Primary outcome was rate of HE defined as relative increase of ≥35%. Secondary outcomes comprised mean absolute change in hematoma volume, as well as in-hospital mortality and functional outcome.

RESULTS

Overall, 182 patients with factor-Xa inhibitor-related ICH (85 receiving andexanet alfa versus 97 receiving usual care) were selected for analysis. There were no relevant differences regarding demographic or clinical characteristics between both groups. HE occurred in 11 of 80 (14%) andexanet alfa patients compared with 21 of 67 (36%) usual care patients (adjusted relative risk, 0.40 [95% CI, 0.20-0.78]; P=0.005), with a reduction in mean overall hematoma volume change of 7 mL. There were no statistically significant differences among in-hospital mortality or functional outcomes. Sensitivity analysis including only usual care patients receiving prothrombin complex concentrates demonstrated consistent results.

CONCLUSIONS

As compared with usual care, andexanet alfa was associated with a lower rate of HE in atraumatic factor-Xa inhibitor-related ICH, however, without translating into significantly improved clinical outcomes. A comparative trial is needed to confirm the benefit on limiting HE and to explore clinical outcomes across patient subgroups and by time to treatment.

Survival after breast cancer in women with type 2 diabetes using antidiabetic medication and statins: a retrospective cohort study

Background: We assessed survival of breast cancer in women with type 2 diabetes (T2D) treated with metformin, other types of antidiabetic medication (ADM) and statins.Materials and Methods: The study cohort consisted of women with T2D and diagnosed with breast cancer in Finland in 1998─2011. Mortality rates from breast cancer and other causes were analysed by Cox models, and adjusted hazard ratios (HRs) with 95% confidence intervals (Cls) were estimated in relation to the use of different types of medication.Results: The final cohort consisted of 3,533 women. No clear evidence was found for breast cancer mortality being different in metformin users (HR 0.86, 95% Cl 0.63-1.17), but their other-cause mortality appeared to be lower (HR 0.73, 95% Cl 0.55-0.97) in comparison with women using other types of oral ADM. Other-cause mortality was higher among insulin users (HR 1.45, 95% Cl 1.16-1.80) compared with users of other oral ADMs, other than metformin. Prediagnostic statin use was observed to be associated with decreased mortality from both breast cancer (HR 0.76, 95% Cl 0.63-0.92) and other causes (HR 0.75, 95% Cl 0.64-0.87).Conclusions: We did not find any association between ADM use and disease-specific mortality among women with T2D diagnosed with breast cancer. However, interestingly, prediagnostic statin use was observed to predict reduced mortality from breast cancer and other causes. We hypothesise that treating treatment practices of T2D or hypercholesterolaemia of breast cancer patients might affect overall prognosis of women diagnosed with breast cancer and T2D.

Derivation and validation of SIDIAP-FHP score: A new risk model predicting cardiovascular disease in familial hypercholesterolemia phenotype

BACKGROUND AND AIMS

Assessment of individual cardiovascular risk, distinguishing primary and secondary prevention, would improve the clinical management of the population with familial hypercholesterolemia. We aimed to develop and validate two risk functions to predict incident and recurrent atherosclerotic cardiovascular disease (ASCVD) in a primary care-based population with familial hypercholesterolemia phenotype (FHP), and to compare their predictive capacity with that of the SpAnish Familial hypErcHolEsterolemiA cohoRT (SAFEHEART) risk equation (SAFEHEART-RE).

METHODS

Data from the Catalan primary care system database (SIDIAP) of patients ≥18 years old with FHP in 2006-2013 were used to develop and validate two risk functions to predict incident and recurrent ASCVD. A validation dataset was also used to compare the model predictive capacity to that of SAFEHEART-RE.

RESULTS

The new model (SIDIAP-FHP) included age, diabetes, smoking, sex (male), hypertension, and baseline low-density lipoprotein cholesterol in the primary prevention cohort and age, diabetes, smoking, and disease characteristics (progressive, recent, polyvascular, or included myocardial infarction) in the secondary prevention cohort. The models demonstrated a fair fit: C-Statistic: 0.71 (95%CI:0.68-0.75) in primary prevention and 0.65 (95%CI:0.60-0.70) in secondary prevention (higher than that of SAFEHEART-RE: 0.64 [95%CI:0.60-0.68] and 0.55 [95%CI:0.51-0.59], respectively; both p < 0.01). The Brier scores obtained with the SIDIAP-FHP score were significantly lower than that obtained with SAFEHEART-RE in both the primary and secondary prevention cohorts.

CONCLUSIONS

The SIDIAP-FHP score provides accurate ASCVD risk estimates for primary and secondary prevention in the FHP population, with better predictive capacity than that of SAFEHEART-RE in this general population, especially in persons with previous ASCVD.

Association Between Statin Use and Sex Hormone in the Multi-Ethnic Study of Atherosclerosis Cohort

PURPOSE

Based on the 2018 American College of Cardiology/American Heart Association cholesterol guidelines, the number of individuals eligible for statin therapy to reduce atherosclerotic cardiovascular disease risk has greatly expanded. Statins inhibit cholesterol biosynthesis, which can impair gonadal steroidogenesis. We evaluated the effect of statins on endogenous sex hormones in a large epidemiological study.

METHODS

A total of 6814 Multi-Ethnic Study of Atherosclerosis (MESA) participants underwent the baseline examination. Of these, 6171 had measurements of serum sex hormones available: dehydroepiandrosterone (DHEA), SHBG, estradiol, and total and bioavailable testosterone. Multivariable linear regression models were used to assess the relationship of statin use with each sex hormone.

RESULTS

A total of 345 women (17.4%) and 464 men (14.7%) were statin users (mean age, 67 years; 41% white, 29% black, 11% Chinese, and 19% Hispanic). Among the users vs nonusers of statins, the mean SHBG was 3.54 nmol/L (P < 0.01) lower in women and 3.37 nmol/L (P < 0.001) lower in men; the mean DHEA was 1.06 nmol/L (P < 0.05) lower in women and 0.70 nmol/L (P < 0.01) lower in men, after adjustment for potential confounders. With further propensity score adjustment, the mean DHEA and SHBG levels were 0.67 nmol/L (P < 0.05) and 3.49 nmol/L (P < 0.001) lower, respectively, for statin users vs nonusers. No statistically significant association was noted between estradiol, total testosterone, and bioavailable testosterone and statin use.

CONCLUSION

Statin users have lower levels of SHBG and DHEA. This is especially relevant owing to the increasing use of statin therapy.

Incidence of Cardiovascular Disease in Patients with Familial Hypercholesterolemia Phenotype: Analysis of 5 Years Follow-Up of Real-World Data from More than 1.5 Million Patients

In the statin era, the incidence of atherosclerotic cardiovascular diseases (ASCVD) in patients with familial hypercholesterolemia (FH) has not been updated. We aimed to determine the incidence of ASCVD in patients with FH-phenotype (FH-P) and to compare it with that of normal low-density lipoprotein cholesterol (LDL-C) patients. We performed a retrospective cohort study using the Database of the Catalan primary care system, including ≥18-year-old patients with an LDL-C measurement. From 1,589,264 patients available before 2009, 12,823 fulfilled FH-P criteria and 514,176 patients were normolipidemic (LDL-C < 115 mg/dL). In primary prevention, patients with FH-P had incidences of ASCVD and coronary heart disease (CHD) of 14.9/1000 and 5.8/1000 person-years, respectively, compared to 7.1/1000 and 2.1/1000 person-years in the normolipidemic group. FH-P showed hazard ratio (HR) of 7.1 and 16.7 for ASCVD and CHD, respectively, in patients younger than 35 years. In secondary prevention, patients with FH-P had incidences of ASCVD and CHD of 89.7/1000 and 34.5/1000 person-years, respectively, compared to 90.9/1000 and 28.2/1000 person-years in the normolipidemic group (HR in patients younger than 35 years: 2.4 and 6.0). In the statin era, FH-P remains associated with high cardiovascular risk, compared with the normolipidemic population. This excess of risk is markedly high in young individuals.

Number of Patients Eligible for PCSK9 Inhibitors Based on Real-world Data From 2.5 Million Patients

INTRODUCTION AND OBJECTIVES

PCSK9 inhibitors (PCSK9i) are safe and effective lipid-lowering drugs. Their main limitation is their high cost. The aim of this study was to estimate the number of patients eligible for treatment with PCSK9i according to distinct published criteria.

METHODS

Data were obtained from the Information System for the Development of Research in Primary Care. Included patients were equal to or older than 18 years and had at least 1 low-density lipoprotein cholesterol measurement recorded between 2006 and 2014 (n = 2 500 907). An indication for treatment with PCSK9i was assigned according to the following guidelines: National Health System, Spanish Society of Arteriosclerosis, Spanish Society of Cardiology, National Institute for Health and Care Excellence, and the European Society of Cardiology/European Atherosclerosis Society Task Force. Lipid-lowering treatment was defined as optimized if it reduced low-density lipoprotein levels by ≥ 50% and adherence was > 80%.

RESULTS

Among the Spanish population aged 18 years or older, the number of possible candidates to receive PCSK9i in an optimal lipid-lowering treatment scenario ranged from 0.1% to 1.7%, depending on the guideline considered. The subgroup of patients with the highest proportion of potential candidates consisted of patients with familial hypercholesterolemia, and the subgroup with the highest absolute number consisted of patients in secondary cardiovascular prevention.

CONCLUSIONS

The number of candidates to receive PCSK9i in conditions of real-world clinical practice is high and varies widely depending on the recommendations of distinct scientific societies.

Methods to estimate underlying blood pressure: The Atherosclerosis Risk in Communities (ARIC) Study

Antihypertensive medications complicate studies of blood pressure (BP) natural history; BP if untreated (“underlying BP”) needs to be estimated. Our objectives were to compare validity of five missing data imputation methods to estimate underlying BP and longitudinal associations of underlying BP and age. We simulated BP treatment in untreated hypertensive participants from Atherosclerosis Risk in Communities (ARIC) in visits 1–5 (1987–2013) using matched treated hypertensive participants. The underlying BP was imputed: #1, set as missing; #2, add 10 mmHg for systolic, 5 mmHg for diastolic; #3, add medication class-specific constant; #4, truncated normal regression; and #5, truncated normal regression including prior visit data. Longitudinal associations were estimated using linear mixed models of imputed underlying BP for simulated treated and measured BP for untreated participants. Method 3 was the best-performing for systolic BP; lowest relative bias (5.3% for intercept at age 50, 0% for age coefficient) and average deviation from expected (0.04 to -1.79). Method 2 performed best for diastolic BP; lowest relative bias (0.6% intercept at age 50, 33.3% age <60, 9.1% age 60+) and average deviation (-1.25 to -1.68). Methods 4 and 5 were comparable or slightly inferior. In conclusion, constant addition methods yielded valid and precise underlying BP and longitudinal associations.

Familial hypercholesterolemia in a European Mediterranean population-Prevalence and clinical data from 2.5 million primary care patients

BACKGROUND

Familial hypercholesterolemia (FH), the most frequent hereditary cause of premature coronary heart disease (CHD), is underdiagnosed and insufficiently treated.

OBJECTIVES

The objectives of the study were to estimate the prevalence of the FH phenotype (FH-P) and to describe its clinical characteristics in a Mediterranean population.

METHODS

Data were obtained from the Catalan primary care system's clinical records database (Catalan acronym: SIDIAP). Patients aged >7 years with at least 1 low-density lipoprotein cholesterol measurement recorded between 2006 and 2014 (n = 2,554,644) were included. Heterozygous FH-P and homozygous FH-P were defined by untreated low-density lipoprotein cholesterol plasma concentrations. The presence of cardiovascular diseases and risk factors was defined by coded medical records from primary care and hospital discharge databases.

RESULTS

The age- and sex-standardized prevalence of heterozygous FH-P and homozygous FH-P were 1/192 individuals and 1/425,774 individuals, respectively. In the group aged 8 to 18 years, 0.46% (95% confidence interval: 0.41-0.52) had FH-P; overall prevalence was 0.58% (95% confidence interval: 0.58-0.60). Among patients with FH-P aged >18 years, cardiovascular disease prevalence was 3.5 times higher than in general population, and CHD prevalence in those aged 35 to 59 years was 4.5 times higher than in those without FH-P. Lipid-lowering therapy was lacking in 13.5% of patients with FH-P, and only 31.6% of men and 22.7 of women were receiving high or very high-intensity lipid-lowering therapy.

CONCLUSION

Prevalence of FH-P was higher than expected, but underdiagnosed and suboptimally treated, especially in women. Moreover, treatment started late considering the high CHD incidence associated with this condition.