Sequential Pattern Mining of Longitudinal Adverse Events After Left Ventricular Assist Device Implant

In search of similarity in adverse events journeys of patients with left ventricular assist devices

OBJECTIVE

The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) Event data set contains an expansive collection of longitudinal evidence of the course of adverse events (AEs) of >15,000 patients who have received a left ventricular assist device (LVAD). Buried in the huge Event data set is knowledge that can provide a deeper understanding of the patterns of the "AE journey" of patients with LVAD. Thus, the goal of this study was to examine the Event data set from a comprehensive perspective to identify unique relationships and patterns of AEs, alert potential challenges, and suggest future research directions.

METHODS

A sequential pattern mining algorithm called SPADE (ie, Sequential PAttern Discovery using Equivalence classes) was applied to 86,912 recorded AEs of 15,820 patients with a continuous-flow LVAD between 2008 and 2016, extracted from the publicly accessible INTERMACS registry. The patterns of AE journey were investigated by posing 5 descriptive research questions about most common types of AE, concomitant AEs, AE sequences, AE subsequences, and interesting relations between AEs.

RESULTS

The analysis revealed several characteristics of patterns of the AE journey of patients who received an LVAD that accounts for the types and temporal ordering of successive AEs, combinations of AEs, and their timing after surgery.

CONCLUSIONS

The high diversity and sparsity of the types and timing of AE occurrences make the AE journeys of patients dissimilar from each other, impeding the discovery of highly-patterned AE journeys among the patients. This study suggests 2 salient directions for future studies to tackle this issue using cluster analysis to cluster patients into more similar groups and translate these results into a practical clinical tool to forecast the next AE based on the history of previous AEs.

Interpretable Machine Learning Models to Predict the Resistance of Breast Cancer Patients to Doxorubicin from Their microRNA Profiles

Doxorubicin is a common treatment for breast cancer. However, not all patients respond to this drug, which sometimes causes life-threatening side effects. Accurately anticipating doxorubicin-resistant patients would therefore permit to spare them this risk while considering alternative treatments without delay. Stratifying patients based on molecular markers in their pretreatment tumors is a promising approach to advance toward this ambitious goal, but single-gene gene markers such as HER2 expression have not shown to be sufficiently predictive. The recent availability of matched doxorubicin-response and diverse molecular profiles across breast cancer patients permits now analysis at a much larger scale. 16 machine learning algorithms and 8 molecular profiles are systematically evaluated on the same cohort of patients. Only 2 of the 128 resulting models are substantially predictive, showing that they can be easily missed by a standard-scale analysis. The best model is classification and regression tree (CART) nonlinearly combining 4 selected miRNA isoforms to predict doxorubicin response (median Matthew correlation coefficient (MCC) and area under the curve (AUC) of 0.56 and 0.80, respectively). By contrast, HER2 expression is significantly less predictive (median MCC and AUC of 0.14 and 0.57, respectively). As the predictive accuracy of this CART model increases with larger training sets, its update with future data should result in even better accuracy.

Clinical and operational insights from data-driven care pathway mapping: a systematic review

Background

Accumulated electronic data from a wide variety of clinical settings has been processed using a range of informatics methods to determine the sequence of care activities experienced by patients. The “as is” or “de facto” care pathways derived can be analysed together with other data to yield clinical and operational information. It seems likely that the needs of both health systems and patients will lead to increasing application of such analyses. A comprehensive review of the literature is presented, with a focus on the study context, types of analysis undertaken, and the utility of the information gained.

Methods

A systematic review was conducted of literature abstracting sequential patient care activities (“de facto” care pathways) from care records. Broad coverage was achieved by initial screening of a Scopus search term, followed by screening of citations (forward snowball) and references (backwards snowball). Previous reviews of related topics were also considered. Studies were initially classified according to the perspective captured in the derived pathways. Concept matrices were then derived, classifying studies according to additional data used and subsequent analysis undertaken, with regard for the clinical domain examined and the knowledge gleaned.

Results

254 publications were identified. The majority (n = 217) of these studies derived care pathways from data of an administrative/clinical type. 80% (n = 173) applied further analytical techniques, while 60% (n = 131) combined care pathways with enhancing data to gain insight into care processes.

Discussion

Classification of the objectives, analyses and complementary data used in data-driven care pathway mapping illustrates areas of greater and lesser focus in the literature. The increasing tendency for these methods to find practical application in service redesign is explored across the variety of contexts and research questions identified. A limitation of our approach is that the topic is broad, limiting discussion of methodological issues.

Conclusion

This review indicates that methods utilising data-driven determination of de facto patient care pathways can provide empirical information relevant to healthcare planning, management, and practice. It is clear that despite the number of publications found the topic reviewed is still in its infancy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12911-022-01756-2.

Long-term survival on LVAD support: Device complications and end-organ dysfunction limit long-term success

BACKGROUND

Preoperative variables can predict short term left ventricular assist device (LVAD) survival, but predictors of extended survival remain insufficiently characterized.

METHOD

Patients undergoing LVAD implant (2012-2018) in the Intermacs registry were grouped according to time on support: short-term (<1 year, n = 7,483), mid-term (MT, 1-3 years, n = 5,976) and long-term (LT, ≥3 years, n = 3,015). Landmarked hazard analyses (adjusted hazard ratio, HR) were performed to identify correlates of survival after 1 and 3 years of support.

RESULTS

After surviving 1 year of support, additional LVAD survival was less likely in older (HR 1.15 per decade), Caucasian (HR 1.22) and unmarried (HR 1.16) patients (p < 0.05). After 3 years of support, only 3 preoperative characteristics (age, race, and history of bypass surgery, p < 0.05) correlated with extended survival. Postoperative events most negatively influenced achieving LT survival. In those alive at 1 year or 3 years, the occurrence of postoperative renal (creatinine HR MT = 1.09; LT HR = 1.10 per mg/dl) and hepatic dysfunction (AST HR MT = 1.29; LT HR = 1.34 per 100 IU), stroke (MT HR = 1.24; LT HR = 1.42), infection (MT HR = 1.13; LT HR = 1.10), and/or device malfunction (MT HR = 1.22; LT HR = 1.46) reduced extended survival (all p ≤ 0.03).

CONCLUSIONS

Success with LVAD therapy hinges on achieving long term survival in more recipients. After 1 year, extended survival is heavily constrained by the occurrence of adverse events and postoperative end-organ dysfunction. The growth of destination therapy intent mandates that future LVAD studies be designed with follow up sufficient for capturing outcomes beyond 24 months.

A data mining-based cross-industry process for predicting major bleeding in mechanical circulatory support

Aims

Over a third of patients, treated with mechanical circulatory support (MCS) for end-stage heart failure, experience major bleeding. Currently, the prediction of a major bleeding in the near future is difficult because of many contributing factors. Predictive analytics using data mining could help calculating the risk of bleeding; however, its application is generally reserved for experienced researchers on this subject. We propose an easily applicable data mining tool to predict major bleeding in MCS patients.

Methods and results

All data of electronic health records of MCS patients in the University Medical Centre Utrecht were included. Based on the cross-industry standard process for data mining (CRISP-DM) methodology, an application named Auto-Crisp was developed. Auto-Crisp was used to evaluate the predictive models for a major bleeding in the next 3, 7, and 30 days after the first 30 days post-operatively following MCS implantation. The performance of the predictive models is investigated by the area under the curve (AUC) evaluation measure. In 25.6% of 273 patients, a total of 142 major bleedings occurred during a median follow-up period of 542 [interquartile range (IQR) 205-1044] days. The best predictive models assessed by Auto-Crisp had AUC values of 0.792, 0.788, and 0.776 for bleedings in the next 3, 7, and 30 days, respectively.

Conclusion

The Auto-Crisp-based predictive model created in this study had an acceptable performance to predict major bleeding in MCS patients in the near future. However, further validation of the application is needed to evaluate Auto-Crisp in other research projects.

Identifying urban built environment factors in pregnancy care and maternal mental health outcomes

Backgrounds

Risk factors related to the built environment have been associated with women’s mental health and preventive care. This study sought to identify built environment factors that are associated with variations in prenatal care and subsequent pregnancy-related outcomes in an urban setting.

Methods

In a retrospective observational study, we characterized the types and frequency of prenatal care events that are associated with the various built environment factors of the patients’ residing neighborhoods. In comparison to women living in higher-quality built environments, we hypothesize that women who reside in lower-quality built environments experience different patterns of clinical events that may increase the risk for adverse outcomes. Using machine learning, we performed pattern detection to characterize the variability in prenatal care concerning encounter types, clinical problems, and medication prescriptions. Structural equation modeling was used to test the associations among built environment, prenatal care variation, and pregnancy outcome. The main outcome is postpartum depression (PPD) diagnosis within 1 year following childbirth. The exposures were the quality of the built environment in the patients’ residing neighborhoods. Electronic health records (EHR) data of pregnant women (n = 8,949) who had live delivery at an urban academic medical center from 2015 to 2017 were included in the study.

Results

We discovered prenatal care patterns that were summarized into three common types. Women who experienced the prenatal care pattern with the highest rates of PPD were more likely to reside in neighborhoods with homogeneous land use, lower walkability, lower air pollutant concentration, and lower retail floor ratios after adjusting for age, neighborhood average education level, marital status, and income inequality.

Conclusions

In an urban setting, multi-purpose and walkable communities were found to be associated with a lower risk of PPD. Findings may inform urban design policies and provide awareness for care providers on the association of patients’ residing neighborhoods and healthy pregnancy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-021-04056-1.

Machine learning, artificial intelligence and mechanical circulatory support: A primer for clinicians

Artificial intelligence (AI) refers to the ability of machines to perform intelligent tasks, and machine learning (ML) is a subset of AI describing the ability of machines to learn independently and make accurate predictions. The application of AI combined with "big data" from the electronic health records, is poised to impact how we take care of patients. In recent years, an expanding body of literature has been published using ML in cardiovascular health care, including mechanical circulatory support (MCS). This primer article provides an overview for clinicians on relevant concepts of ML and AI, reviews predictive modeling concepts in ML and provides contextual reference to how AI is being adapted in the field of MCS. Lastly, it explains how these methods could be incorporated in the practices of medicine to improve patient outcomes.

Left Ventricular Assist Device-Related Complications

Left ventricular assist device (LVAD) has emerged as a safe, durable, and revolutionary therapy for end-stage heart failure patients. Despite the appearance of newer-generation devices that have improved patient outcomes, the burden of adverse events remains significant. Although the survival rate for patients with LVAD is appreciated to be 81% at 1 year and 70% at 2 years, the incidence of adverse events is also high. Over time, both early and late postimplant complications have diminished in terms of prevalence and impact; however, complications, such as infections, bleeding, right heart failure, pump thrombosis, aortic insufficiency, or stroke, continue to represent a challenge for the practitioner. Therefore, the aim of this review is to highlight the most recent data regarding the current use of LVAD in the treatment of end-stage heart failure, with a specific focus on LVAD-related complications, in order to improve device-related outcomes. It will also revise how to mitigate the risk and how to approach specific adverse events. Withal, understanding the predisposing risk factors associated with postimplant complications, early recognition and appropriate treatment help to significantly improve the prognosis for patients with end-stage heart failure.

Delineating Pathways to Death by Multisystem Organ Failure in Patients With a Left Ventricular Assist Device

BACKGROUND

This study delineates the sequences of adverse events (AEs) preceding mortality attributed to multisystem organ failure (MSOF) in patients with a left ventricular assist device (LVAD).

METHODS

We analyzed 3765 AEs after 536 LVAD implants recorded in The Society of Thoracic Surgeons Intermacs data registry between 2006 and 2015 that resulted in MSOF death. Hierarchical clustering identified and visualized quantitatively unique clusters of patients with similar AE profiles. Markov modeling was used to illustrate the AE sequences that led to MSOF death within the clusters. Cox proportional hazard models determined the risk-adjusted, preimplant predictors of MSOF.

RESULTS

We identified 2 distinct MSOF clusters based on their proportion of AE types and survival time. The early-death cluster (418 patients, 2304 AEs) had a median survival of 1 month (interquartile range, 3-6 months), whereas the late-death cluster (118 patients, 1,461 AEs) had a median survival of 11 months (interquartile range, 6-22 months). The predominant AE sequences in the early-death and late-death clusters were renal failure, to respiratory failure, to death (62%) and bleeding, to infection, to respiratory failure, to death (45%), respectively. Significant risk-adjusted preimplant predictors of MSOF included line sepsis (hazard ratio [HR] 3.0; 95% confidence interval [CI], 1.1-8.2), extracorporeal membrane oxygenation (HR, 2.2; 95% CI, 1.2-3.9), and dialysis or ultrafiltration (HR, 2.1; 95% CI, 1.5-3.0).

CONCLUSIONS

This analysis identified 2 AE clusters and the predominant sequences that result in MSOF-associated mortality. MSOF develops in 1 cluster of patients after chronic bleeding and repeated infections but has prolonged survival, while another group dies early after renal and respiratory complications.