Stefan Gustafsson’s research while affiliated with Uppsala University and other places

Background and Aims Ambulatory patients presenting with signs or symptoms of heart failure (HF) should undergo natriuretic peptide testing. Rates of death, HF hospitalization, and healthcare costs were examined in patients thus identified with suspected de novo HF. Methods This population-based study (REVOLUTION HF) encompassing two large healthcare regions in Sweden examined patients who presented to outpatient care for the first time between 1 January 2015 and 31 December 2020, who had a recorded sign (peripheral oedema) or symptom (dyspnoea) of HF, and whose N-terminal pro-B-type natriuretic peptide (NT-proBNP) measured >300 ng/L within ±30 days of that sign or symptom. Characteristics, outcomes, healthcare patterns, and healthcare costs for these patients were followed for 1 year. Comparisons were made with matched controls without history of HF, its signs, its symptoms, or elevated NT-proBNP. Results Overall, 5942 patients (median age 78.7 years; 54% women) presented with suspected de novo HF. Within 1 year, 29% had received a HF diagnosis. Patients with suspected de novo HF had higher rates of all-cause death (11.7 vs. 6.5 events/100 person-years) and HF hospitalizations (12.5 vs. 2.2 events/100 person-years) than matched controls (n = 2048), with the highest event rates in the weeks after presentation. Rates were higher with higher NT-proBNP levels. Although some patients already used HF guideline-directed medical therapies for other indications, initiation of new medications was variable. Healthcare costs were higher in patients with suspected de novo HF than in matched controls, driven mostly by HF and chronic kidney disease. Conclusions Patients with suspected HF and elevated NT-proBNP had high mortality and morbidity in the weeks after presentation, and accrued substantial healthcare costs, highlighting an urgent need for prompt identification, evaluation, and treatment of HF.

Background: We previously reported significant correlations between a direct measure of insulin sensitivity (IS) and blood levels of proteins measured using the Proximity Extension Assay (PEA) in two European cohorts. However, protein correlations with IS within non-European populations, in response to short-term interventions that improve IS, and any causal associations with IS have not yet been established. Methods: We measured 1,470 proteins using the PEA in the plasma of 1,015 research participants at Stanford University who underwent one or more direct measures of IS. Association analyses were carried out with multivariable linear regression within and across Stanford subgroups and within each of the two European cohorts. Association statistics were also meta-analyzed after transformation and harmonization of the two direct measures of IS. Lastly, we performed genome-wide association studies of IS and used genetic instruments of plasma proteins from the UK Biobank to identify candidate causal proteins for IS through Mendelian Randomization (MR) analysis. Results: In age and sex adjusted model, 810 proteins were associated with baseline IS among 652 self-reported European participants in the Stanford cohort at a false discovery rate (FDR) < 0.05. Effect sizes for these proteins were highly correlated with those observed in 122 South Asian, 92 East Asian, 85 Hispanic, and 52 Black/African American persons (r= 0.68 to 0.83, all P<=4.3x10^-113). Meta-analysis of the full Stanford cohort with the two European cohorts (N=2,945) yielded 247 significant protein associations (FDR < 0.05), with 75 remaining significant after further adjustment for body mass index. In a subset of Stanford participants undergoing insulin sensitizing interventions (N=53 taking thiazolidinediones, N=66 with weight loss), 79.6% of protein level changes were directionally consistent with the respective baseline association (observed/expected p=6.7x10^-16). MR analyses identified eight candidate causal proteins for IS, among which were SELE and ASGR1, proteins with established drug targets currently under investigation. Conclusion: Plasma proteins measured using the PEA provide a robust signature for IS across diverse populations and after short-term insulin sensitizing interventions highlighting their potential value as universal biomarkers of insulin resistance. A small subset of markers provided insights into potential causal molecular mechanisms and therapeutic targets.

Background Heart failure treatment patterns towards the end of life have rarely been studied at large scale. Yet, this is important to assess, since heart failure is a chronic condition with patients on treatment for potentially many years. Purpose To investigate treatment patterns in the last years of life in patients with heart failure. Methods In a register-based study covering the whole Swedish population between 2007 and 2020, we investigated treatment patterns in the last years of life in 364,480 individuals who died with heart failure. We combined binomial models (to estimate proportion on treatment at death) and Poisson models (to estimate discontinuation rate), to estimate the proportion on each heart failure treatment up to 5 years prior to death, using penalized splines to allow for non-linearities. We also assessed comorbidity patterns over time. Results All heart failure treatments saw a gradual discontinuation closer to death, but the discontinuation rate of beta blockers decreased over time, resulting in an increasing proportion of patients on treatment at the time of death (Fig). Also, the pattern of comorbidities changed over time, with an increase in e.g., hypertension and atrial fibrillation, but a decline in ischemic heart disease (63% in 2007 gradually declining to 54% in 2020). Conclusions In this nationwide study, we observed a changing pattern of medical treatment of heart failure during the last years of life, most notably with an increasing use of beta blockers. Comorbidities and the shares of heart failure etiologies also changed over time, with an increase in several conditions but a decline in ischemic heart disease. Taken together, this may have important implications for planning and optimizing care for these patients.

Background The beneficial effects of physical activity on cardiovascular risk are well documented. Yet, the underlying mechanisms by which physical activity prevents cardiovascular disease remain unclear (1). Advancements of omics technology enables us to explore novel biomarkers of this complex molecular relationship and potentially unravel underlying molecular mechanisms involved in the progression of cardiovascular events. Purpose To investigate associations between long term physical activity and plasma proteins. As a second step investigate associations between identified plasma proteins and future myocardial infarction and mortality. Methods In the population based prospective cohort Uppsala Longitudinal Study of Adult Men (ULSAM), men were repeatedly investigated with a validated questionnaire on physical activity at ages 50, 60 and 70. 720 plasma proteins were analyzed at age 70 (n=782) with follow up on mortality data for 30 years. In the case-cohort MIMI (Markers of Imminent Myocardial Infarction study) the same plasma proteins were measured in disease-free individuals from six European cohorts. Individuals with acute myocardial infarction within 6 months from baseline were defined as cases (n=420) and up tto four cohort representatives per case were defined as controls (n=1598). Results Higher level of physical activity during 20 years was significantly associated with 12 plasma proteins, in a linear regression model adjusted for age, education and smoking after Bonferroni correction for multiple testing (p<0.000069). After additional adjustment for known cardiovascular risk factors (education, smoking status, BMI, blood pressure, low-density lipoprotein, diabetes diagnosis, and hypertensive-, lipid- or diabetes-treatment), all 12 proteins remained negatively associated with higher level of physical activity in the ULSAM cohort (p<0,05). Two proteins of the 12 proteins associated with physical activity were also significantly associated with future myocardial infarction in the MIMI study (FGF21 and IL6, p<0.05 for both) and 10 proteins were associated with increased mortality risk in the ULSAM cohort (p<0.05 for all). Moreover, proteins that were more negatively associated with physical activity were generally more positively associated with the risk of myocardial infarction in MIMI (Figure). Conclusion Multiple novel associations were found between long-term physical activity and plasma proteins. Some proteins were also associated with future myocardial infarction in an independent cohort which could suggest that the cardioprotective effects of physical activity to some degree may be mediated via the circulating proteome. Our findings encourage additional studies in order to understand the underlying causal mechanisms of these associations.

Aims In patients with heart failure, treatment patterns in the last years of life have not been assessed at large scale. We aimed to assess whether heart failure treatment patterns up to 5 years prior to death changed over time. Methods and results In a cohort study covering the whole Swedish population, we assessed all heart failure patients who died between 1 July 2007 and 31 December 2020 for evidence‐based treatments. The proportion on the respective treatment at the time of death was examined by year of death using binomial regression. Looking back in time, treatment discontinuation rates were estimated using Poisson regression on time‐split data. Combining these models, the proportion on each medication was estimated up to 5 years prior to death. A total of 364 480 patients died with heart failure during the study period. Half were women, and the median (interquartile range) age at death was 86 (79–90). The use of all heart failure treatments decreased gradually closer to death, but the discontinuation rate of beta blockers decreased over time, resulting in an increasing proportion of patients on treatment at the time of death. Conclusion In patients with heart failure, a changing pattern of medical treatment during the last years of life was observed, most notably with an increasing use of beta blockers. This may in part be due to a changing pattern of comorbidities over time, with an increase in e.g. hypertension and atrial fibrillation, but a decline in ischaemic heart disease.

Imbalances in electrolyte concentrations can have severe consequences, but accurate and accessible measurements could improve patient outcomes. The current measurement method based on blood tests is accurate but invasive and time-consuming and is often unavailable for example in remote locations or an ambulance setting. In this paper, we explore the use of deep neural networks (DNNs) for regression tasks to accurately predict continuous electrolyte concentrations from electrocardiograms (ECGs), a quick and widely adopted tool. We analyze our DNN models on a novel dataset of over 290,000 ECGs across four major electrolytes and compare their performance with traditional machine learning models. For improved understanding, we also study the full spectrum from continuous predictions to a binary classification of extreme concentration levels. Finally, we investigate probabilistic regression approaches and explore uncertainty estimates for enhanced clinical usefulness. Our results show that DNNs outperform traditional models but model performance varies significantly across different electrolytes. While discretization leads to good classification performance, it does not address the original problem of continuous concentration level prediction. Probabilistic regression has practical potential, but our uncertainty estimates are not perfectly calibrated. Our study is therefore a first step towards developing an accurate and reliable ECG-based method for electrolyte concentration level prediction—a method with high potential impact within multiple clinical scenarios.

Background Lipid-lowering treatment (LLT) is one of the cornerstones in secondary prevention in patients with chronic coronary syndrome (CCS). Despite having the same recommendations for target levels of low-density lipoprotein cholesterol (LDL-C), studies have shown that women have worse control of LDL-C than men. Older age has been suggested as a significant explanatory factor. Purpose To describe and compare the treatment pattern and target achievement of LDL-C levels in men and women managed for CCS in a primary care setting. Furthermore, to elucidate factors contributing to sex differences by examining the frequency of LDL-C assessment and LLT intensity by age. Methods A retrospective observational study of patients with CCS was performed in a region of Sweden comprising 390 000 inhabitants. The study population included men and women with a CCS first-time diagnosed between 2012 and 2020 and without previous myocardial infarction. The CCS diagnosis was based on the combination of an international classification diagnosis code (ICD-10) of angina and coronary artery disease confirmed by either a non-invasive or an invasive investigation. Data on clinical measurements after inclusion were collected from the electronic health records. Target levels were LDL-C <1.8 mmol/L (2012-2015) and LDL-C <1.4 mmol/L (2015-2020) and systolic blood pressure (SBP) <140 mmHg. Data were stratified by sex and age categories (<60, 60-<70, 70-<80, ≥80 years). Measurement of blood pressure (BP) was used as a proxy for a healthcare visit. Results Median age was 68 (interquartile range [IQR], 62-74) and 70 (IQR, 63-76) years in men (N=1037) and women (N=415), respectively. Compared with men, women were less often treated with LLT and less likely to receive combination with a statin and ezetimibe (Figure 1). A larger proportion of women was not treated with any LLT. In all age subgroups, a decline in LLT use in women compared to men was observed over time. Assessment of LDL-C decreased markedly after the first year of follow-up irrespective of sex (assessment of LDL-C was 67.7% and 50.9% at year one and three, respectively). In both men and women, target achievement of LDL-C was low and decreased during follow-up (Figure 2). Regardless of sex, measurements and target achievement of BP were monitored more frequently than those of LDL-C levels. Conclusion Women were less often and less aggressively treated with LLT than men, on average. Although most patients with CCS are in contact with healthcare professionals for BP monitoring, little or no actions seem to be taken to improve lipid management.