Comparison of Unsupervised Machine Learning Approaches for Cluster Analysis to Define Subgroups of Heart Failure with Preserved Ejection Fraction with Different Outcomes.

Heart failure with preserved ejection (HFpEF) is a heterogenous condition affecting nearly half of all patients with heart failure (HF). Artificial intelligence methodologies can be useful to identify patient subclassifications with important clinical implications. We sought a comparison of different machine learning (ML) techniques and clustering capabilities in defining meaningful subsets of patients with HFpEF. Three unsupervised clustering strategies, hierarchical clustering, K-prototype, and partitioning around medoids (PAM), were used to identify distinct clusters in patients with HFpEF, based on a wide range of demographic, laboratory, and clinical parameters. The study population had a median age of 77 years, with a female majority, and moderate diastolic dysfunction. Hierarchical clustering produced six groups but two were too small (two and seven cases) to be clinically meaningful. The K-prototype methods produced clusters in which several clinical and biochemical features did not show statistically significant differences and there was significant overlap between the clusters. The PAM methodology provided the best group separations and identified six mutually exclusive groups (HFpEF1-6) with statistically significant differences in patient characteristics and outcomes. Comparison of three different unsupervised ML clustering strategies, hierarchical clustering, K-prototype, and partitioning around medoids (PAM), was performed on a mixed dataset of patients with HFpEF containing clinical and numerical data. The PAM method identified six distinct subsets of patients with HFpEF with different long-term outcomes or mortality. By comparison, the two other clustering algorithms, the hierarchical clustering and K-prototype, were less optimal.

Test-retest reliability of the Valsalva maneuver in spinal cord injury.

Objective: To determine the test-retest reliability of quantitative and qualitative baroreflex sensitivity (BRS) parameters derived from the Valsalva maneuver (VM) in individuals with traumatic cervical SCI.Design: Test-retest reliability.Setting: Tertiary rehabilitation center.Participants: Fourteen participants with cervical SCI (ranging from C3-C8 neurological level).Outcome Measurements: Beat-to-beat systolic blood pressure (SBP) traces (finger photoplethysmography) were obtained during a 15-second forced expiration at two time points (7.6 ± 2.9 days between sessions) to assess VM reliability. Test-retest reliability of BRS metrics from derived from the VM (Valsalva ratio; VR, pressure recovery time; PRT, vagal baroreflex sensitivity; BRSv, adrenergic baroreflex sensitivity; BRSa1, and total recovery; TR) were assessed by intra-class correlation coefficient (ICC, with 95% confidence interval; CI) and by qualitative reproducibility (V, N, or M pattern).Results: ICCs for quantitative parameters were (CI): VR = 0.894 (0.703-0.965), TR = 0.927 (0.789-0.976), BRSa1 = 0.561 (0.149-0.911), PRT = 0.728 (0.343-0.904), BRSv = 0.243 (-0.309-0.673). Qualitatively, 12 subjects (85.7%) demonstrated reproducible VM patterns at both time points (3 "M" pattern, 8 "V" pattern and one "N" pattern).Conclusion: VR (a measure of cardiovagal function) and TR (a measure of sympathetic adrenergic function) are reliable quantitative parameters that can be derived from SBP response to VM in participants with SCI. Qualitative waveform analysis was reproducible in 12/14 participants. This provides the foundational evidence required to pursue further validity testing to establish a role for VM in the assessment of autonomic functions in SCI.

A Case of Lymphocytic Myocarditis With Complete Heart Block Requiring Permanent Pacemaker Implantation.

Lymphocytic myocarditis (LM) is a rare condition with a broad spectrum of clinical presentations. Complete heart block (CHB) is an uncommon sequela. We present a case of a young man with biopsy-proven LM who presented with cardiogenic shock and went on to develop nonresolving CHB, requiring permanent pacemaker implantation.

Glycemic Control and Cardiovascular Risk Factor Management in Adults With Type 2 Diabetes With and Without Chronic Kidney Disease Before Sodium-Glucose Cotransporter Protein 2 Inhibitors: Insights From the Diabetes Mellitus Status in Canada Survey.

OBJECTIVES: Optimal control of cardiovascular risk factors in adults with type 2 diabetes (T2D) and chronic kidney disease (CKD) is challenging. Limited data are available from the primary care setting on achievement of guideline-recommended targets in this population before the use of sodium-glucose cotransporter protein 2 (SGLT2) inhibitors and glucagon-like peptide-1 (GLP-1) receptor agonists. METHODS: The Diabetes Mellitus Status in Canada survey included 5,172 patients with T2D seen by primary care physicians (PCPs) in November 2012. We compared treatment targets and therapeutic interventions in patients with and without CKD. RESULTS: Compared with those without CKD (n=3,804), patients with CKD (n=1,368) were older, more likely to be female, had a longer duration of diabetes and had more vascular complications. Patients with CKD more frequently had a less stringent glycated hemoglobin (A1C) target of ≤8.0% set by PCPs (10.3% vs 20%, p<0.001), and fewer patients with CKD met the A1C target of ≤7.0% (50.9% vs 47.1%, p=0.016) than those without CKD. Both groups had a similar likelihood of achieving the blood pressure (BP) target of ≤130/80 mmHg (36.8% vs 34.8%, p=0.20), whereas patients with CKD more frequently achieved a low-density lipoprotein cholesterol target of ≤2.0 mmol/L (54.8% vs 61.3%, p<0.001). Overall, only 12.5% in both groups achieved all 3 targets (12.3% vs 13.3%, p=0.33). CONCLUSIONS: Only 1 of 8 patients with T2D achieved optimal glycemic, BP and cholesterol targets, regardless of the presence or absence of CKD. Although more medical interventions were used in patients with CKD, a lower proportion achieved guideline-recommended targets for A1C. These findings provide a benchmark for future comparison.

A new approach to the clinical subclassification of heart failure with preserved ejection fraction.

OBJECTIVE: Heart failure with preserved ejection (HFpEF) represents nearly half of all patients with heart failure (HF). The objective of this study was to determine whether patient characteristics identify discrete kinds of HFpEF. METHODS: Data were collected on 196 patients with HFpEF in a non-hospitalized setting. Clinical and laboratory variables were collected, and 47 candidate variables were examined by the unsupervised clustering strategy partitioning around medoids. The Meta-analysis Global Group in Chronic Heart Failure (MAGGIC) risk score was calculated. Follow-up data on all-cause mortality, cardiovascular mortality, and HF exacerbation, were collected and were not part of the data used to identify subgroups. RESULTS: Six significantly different groups or clusters were found. There were three groups of women (i) individuals with a low proportion of vascular risk factors (HFpEF1) (ii) individuals with a high proportion of hypertension and diabetes, but lower proportion of kidney disease and diastolic dysfunction (HFpEF3) (iii) older individuals with high rates of atrial fibrillation (AF), chronic kidney disease. They had the worst long-term outcomes (HFpEF4). There were three groups of men (i) individuals with a high proportion of coronary artery disease (CAD), dyslipidemia, higher serum creatinine, and diastolic dysfunction (HFpEF2)(ii) individuals with highest BMI, and high proportion of CAD, obstructive sleep apnea, and poorly controlled diabetes (HFpEF5) (iii) individuals with high rates of AF, elevated BNP, biventricular remodeling (HFpEF6). They had a high cardiovascular mortality. CONCLUSIONS: HFpEF consists of a heterogenous group of individuals with six distinct clinical subsets that have different long-term outcomes.

The effect of exercise on the ECG criteria for early repolarization pattern.

OBJECTIVE: To determine the effect of exercise and heart rate on the early repolarization (ER) pattern; focusing on the new criteria for identification of ER. METHODS: The ECG measurements on the terminal QRS notch or slur found in early repolarization were quantitated before, during and after exercise; specifically: (i) the amplitude at the onset of the notch (Jo) (ii) the amplitude at the peak of the notch (Jp), (iii) the amplitude at the end of the notch (Jt), (iv) the duration from Jo to Jp (D1) and (v) the duration from Jo to Jt (D2). RESULTS: All individuals (N = 21) fulfilling the criteria for ER showed complete disappearance of ER after 3 min of exercise. After 5 min of recovery, 29% of subjects showed return of the ER. The return of ER was dynamic with QRS notching of varying extent, without ST elevation, being evident first. The relationship between heart rate and ER was significant and nonlinear, best fit by a second-order polynomial, suggesting that changes in heart rate with exercise was a factor influencing the presence of the ER pattern and the parameters that define ER. CONCLUSION: Each of the newly defined characteristics of the ER are modified and eventually disappear with exercise. The return of ER was dynamic with QRS notching of varying extent being evident first. The changes correlated with variations in heart rate, during both exercise and recovery, suggesting, in part, a role in the underlying mechanism of ER.

Value of the New Spline QTc Formula in Adjusting for Pacing-Induced Changes in Heart Rate.

AIMS: To determine whether a new QTc calculation based on a Spline fit model derived and validated from a large population remained stable in the same individual across a range of heart rates (HRs). Second, to determine whether this formula incorporating QRS duration can be of value in QT measurement, compared to direct measurement of the JT interval, during ventricular pacing. METHODS: Individuals (N=30; 14 males) aged 51.9 ± 14.3 years were paced with decremental atrial followed by decremental ventricular pacing. RESULTS: The new QTc changed minimally with shorter RR intervals, poorly fit even a linear relationship, and did not fit a second-order polynomial. In contrast, the Bazett formula (QTcBZT) showed a steep and marked increase in QTc with shorter RR intervals. For atrial pacing data, QTcBZT was fit best by a second-order polynomial and demonstrated a dramatic increase in QTc with progressively shorter RR intervals. For ventricular pacing, the new QTc minus QRS duration did not meaningfully change with HR in contrast to the HR dependency of QTcBZT and JT interval. CONCLUSION: The new QT correction formula is minimally impacted by HR acceleration induced by atrial or ventricular pacing. The Spline QTc minus QRS duration is an excellent method to estimate QTc in ventricular paced complexes.

Multi-Domain Assessment of Autonomic Function in Spinal Cord Injury Using a Modified Autonomic Reflex Screen.

The aim of this study was to characterize autonomic lesions in participants with spinal cord injury (SCI; n = 10) using an autonomic reflex screen, incorporating sudomotor, cardiovagal, and sympathetic adrenergic tests, as well as hemodynamic responses to head-up tilt (HUT). Hemodynamic responses were compared to healthy controls (n = 20) and previously published normative cutoffs in order better identify autonomic impairments. Sympathetic skin responses (SSRs), heart rate response to deep breathing (HRDB), and heart rate and beat-to-beat blood pressure responses to Valsalva maneuver (VM) and HUT were measured. SCI participants demonstrated impairment in at least one domain, with 7 of 10 demonstrating autonomic impairment across all domains. No single test was concordant with orthostatic hypotension on HUT, in all participants. Measures of cardiovagal function, including HRDB (SCI = 7.7 ± 3.8 beats/min vs. controls = 17.6 ± 8.1 beats/min) and Valsalva ratio (SCI = 1.53 ± 0.29 vs. controls = 1.85 ± 0.37), were significantly reduced in SCI participants, compared to controls (p < 0.05). These findings suggest that an autonomic reflex screen, which includes standardized testing protocol and normative data for comparison, is useful for determining the autonomic domains affected by the neurological injury in SCI. We also demonstrated significant cardiovagal impairment in SCI participants compared to controls, which warrants further investigation to determine whether cardiovagal dysfunction is associated with the negative cardiovascular outcomes, which are known to occur in SCI.

Long-term functional and mobility outcomes for individuals with arthrogryposis multiplex congenita.

Arthrogryposis multiplex congenita (AMC) is a birth defect that involves congenital joint contractures in two or more joints including the limbs, spine, and jaw. The purpose of our study was to identify long-term outcomes of adults with AMC. We recruited 177 participants from over 15 countries, making this the largest international study of adults with AMC. Participants provided demographic information including living situation and mobility and completed two standardized outcome measures, of quality of life and physical activity, using an online survey format. The data were compiled and descriptive analyses were performed. The study group consisted of 72% females and a mean age of 39 years. Over 90% of participants had upper and lower limb involvement, 35% had scoliosis or lordosis while 16% had jaw problems. Participants had an average of nine (0-70) surgeries at the time of the study. The majority (75%) of respondents lived independently of family members (on their own or with a partner). Participants were nearly three times more likely to have a graduate degree than the general US population. Participants reported lower physical function scores than the general US population; however, they reported similar or higher scores for the other quality of life domains of the SF-36. They were considerably less physically active than able-bodied individuals. Half of participants experienced chronic back pain and 60% reported joint pain. Additionally, almost half of the participants took regular pain medications.

Advanced signal analysis for the detection of periodic limb movements from bilateral ankle actigraphy.

Actigraphy can assist in the detection of periodic limb movements in sleep. Although several actigraphs have been previously reported to accurately detect periodic limb movements, many are no longer available; of the existing actigraphs, most sample too infrequently to accurately detect periodic limb movements. The purpose of this study was to use advanced signal analysis to validate a readily available actigraph that has the capability of sampling at relatively high frequencies. We simultaneously recorded polysomnography and bilateral ankle actigraphy in 96 consecutive patients presenting to our sleep laboratory. After pre-processing and conditioning, the bilateral ankle actigraphy signals were then analysed for 14 simple time, frequency and morphology-based features. These features reduced the signal dimensionality and aided in better representation of the periodic limb movement activity in the actigraph signals. These features were then processed by a Naïve-Bayes binary classifier for distinguishing between normal and abnormal periodic limb movement indices. We trained the Naïve-Bayes classifier using a training set, and subsequently tested its classification accuracy using a testing set. From our experiments, using a periodic limb movement index cut-off of 5, we found that the Naïve-Bayes classifier had a correct classification rate of 78.9%, with a sensitivity of 80.3% and a specificity of 73.7%. The algorithm developed in this study has the potential of facilitating identification of periodic limb movements across a wide spectrum of patient populations via the use of bilateral ankle actigraphy.