Point-of-care technologies in heart, lung, blood and sleep disorders from the Center for Advancing Point-of-Care Technologies

Abstract

Recent advancements in point-of-care technologies have transformed care for patients with heart, lung, blood, and sleep disorders by providing rapid, cost-effective, and accessible solutions to challenges in the detection and management of many health conditions. However, major barriers exist throughout the technology development process that inhibit the actualization of many promising and potentially successful ideas. The Center for Advancing Point of Care Technologies has established a system for supporting further innovation in this field and bridging the gap between initial idea conception and implementation. We highlight current and emerging point-of-care technologies throughout the development spectrum and emphasize the need for a needs-driven model of health technology development that involve appropriate stakeholders in the process.

Introduction

Point-of-care (POC) technologies are redefining patient care by enabling bedside and at-home diagnosis and management of common and impactful disorders. Advancements in sensor design, microcircuits, health informatics, and telemedicine platforms have resulted in an explosive growth in the number of devices and systems available for the delivery of patient-centered health care at or close to the individual and outside of traditional laboratory or clinical settings. These technologies have the potential to empower patients to better monitor themselves, communicate, and connect with their health care teams, and live better and healthier lives.

Despite rapid growth in the POC technology space with respect to the number of companies developing innovative products, a relatively small number of technologies have been able to surmount the ‘valley of death’ between conceptualization and prototyping to validation and implementation in clinical practice. Despite the highly publicized clearance of several POC devices for the management of heart, lung, blood, and sleep disorders, including the AliveCor Kardia device, digital blood pressure (BP) monitors, and the Apple Watch [1–3], rates of successful development and deployment for clinical use remain low for POC innovators. Low success rates do not reflect an absence of need, significance, or innovation, but instead often relate to the highly complex and ever-changing regulatory, legal, health care, and commercial environments in the United States. Furthermore, sometimes owing to the siloed nature of the health care industry, POC technology developers often struggle to define the key factors of importance for stakeholders, including patients, clinicians, health care administrators, informaticians, regulators, and payors [4]. It is also notable that success in validation and approval for the limited number of POC approaches in use has mostly been limited to larger companies in the medical device industry, perhaps because of resources required during these final stages to achieve implementation. An additional challenge for some technology developers as well is the goal to create an approach that meets Clinical Laboratory Improvement Amendments(CLIA) waiver status as low complexity which allows use by individuals with minimal training and greatly expands market deployment. To streamline the process of POC technology development, minimize wasted potential, positively impact patient care, and improve health, key federal organizations including the National Institutes of Health (NIH) and the National Science Foundation (NSF), have invested in reimagining how novel POC technologies can be identified, supported, and nurtured through their developmental journey. In this scoping review by members of the Center for Advancing Point-of-Care Technologies (CAPCaT) in heart, lung, blood, and sleep disorders, a member of the Point-of-Care Technologies Research Network (POCTRN), we aim to review the present state of novel POC health technologies in heart, lung, blood, and sleep disorders.

The Point-of-Care Technologies Research Network

The POCTRN was created to drive the development of appropriate POC diagnostic technologies through collaborative efforts that merge scientific and technological capabilities with clinical needs. It was conceived in response to recommendations made based on a seminal workshop assessing clinical needs and opportunities for use of POC technologies in health care settings sponsored by the National Institute of Biomedical Imaging and Bioengineering (NIBIB), the National Heart, Lung, and Blood Institute (NHLBI), and the NSF. Based on these recommendations, the NIBIB released a solicitation entitled ‘Point-of-Care Technologies Research Network’ (RFA-EB-06–002) to support centers and build expertise in the development of integrated systems that address unmet clinical needs in POC testing. A major aspect of the function of each center is the creation of multidisciplinary partnerships necessary to move technologies from an early stage of development into clinical testing.

The Center for Advancing Point-of-Care Technology

The CAPCaT mission is to support testing and deployment of promising ‘late-stage’ POC technologies that can be used to enhance the diagnosis, monitoring, management, and/or treatment of heart, lung, blood, or sleep disorders (NHLBI), with an additional interest in projects that incorporate complementary and integrative health approaches (NCCIH). The University of Massachusetts’ (Lowell and Worcester campuses) CAPCaT in heart, lung, blood, and sleep disorders was established with support from both the NHLBI and NCCIH.

The CAPCaT is making new resources available to its awardees and other POC technology researchers, including investigators and businesses that apply to be part of its network. Resources provided to awardees include access to the NIH’s POCTRN and rich technology and business development resources through the Massachusetts Medical Device Development network, which include clinical translation and validation expertise, engineering assistance (i.e. user-centered and ergonomic design), and training in implementation and dissemination using established methods (NSF’s I-Corps program). Most CAPCaT’s funds are allocated to promising clinical testing and validation awardees selected through transdisciplinary review committees based on prespecified criteria, including clinical need, innovation, and technical capabilities. The following sections highlight current and emerging applications of POC technologies in heart, lung, blood, and sleep conditions.

Cardiovascular disorders

Cardiovascular diseases are perhaps the prime bene-factor of advancements in patient-facing POC technologies because of their epidemiology, the existing telemedicine programs oriented around cardiovascular disease management, and the ubiquity of commercial sensors capable of validated predictors related to primary and secondary cardiovascular outcomes (including heart rate, activity, BP, and heart rhythm [5–7]). Many commercially available medical devices and smartphone and smartwatch compatible applications exist to address cardiovascular health needs.

Hypertension

POC technologies necessary to enable at-home BP measurement were some of the first to demonstrate the potential clinical utility of POC technology for the management of a long-term cardiovascular condition [8]. Recent advancements in biosensor technologies have enabled less burdensome, cuff-less measurements of BP. Various statistical methods have also been developed for using pulse plethysmographic (PPG) data to accurately estimate BP [9,10] and there are a multitude of PPG-based smartphone applications that are designed for measurement, dashboarding, and communication of BP values. These have also been ported into finger and wrist-worn wearable devices, including Omron’s recently FDA-cleared HeartGuide smartwatch, that offer continuous BP measurements. Continuous BP has been previously unavailable in ambulatory settings [3,11,12]. Several ongoing studies are exploring more accurate ways to determine BP, such as using two simultaneous PPG sites to determine pulse transit time or opting for a multisensor approach by combining PPG with force sensors to measure cardiac pulsation, and these developments build on a strong foundation of devices capable of BP assessment in the context of hypertension management [13–15].

Several recent meta-analyses have shown that telemonitoring improves BP control in hypertensive patients, and that BP control is better and more sustained as compared with usual care when POC-informed BP management strategies are deployed (i.e. medication titration, patient education, or lifestyle counseling) [16]. BP telemonitoring is perhaps one of the most well-studied applications of POC technologies, and recent guidelines supporting use of ambulatory POC BP technologies reflect the overwhelming level of evidence demonstrating that home-based BP monitoring can reduce hypertension-related complications, including stroke, heart attack, and congestive heart failure (HF) [17]. However, several barriers have impeded wide-spread integration of home-based BP management into US clinical practice, including the absence of wide-spread electronic health record integration and financial considerations, such as need for clearer reimbursement models for clinicians monitoring and managing hypertension outside of traditional clinical settings [18]. In addition to BP, POC technologies have also been used to explore other blood parameters indicative of chronic cardiovascular disease, such as blood lipid levels with reasonably high accuracy as well [19,20].

Heart failure

Chronic HF is common and the prevalence of HF is increasing with aging of the US population [21]. Acute decompensations of HF contribute significantly to decreased quality of life, hospitalizations, and reduced survival. Building off of an existing literature from implantable cardiac devices that has shown that intrathoracic bioimpedance can estimate lung fluid accumulation and HF worsening, novel POC technologies are using noninvasive measures of bioimpedance to determine body composition based on electrical resistance, and have been used to accurately assess fluid status in HF patients. Electrical sensors capable of measuring bioimpedance have been incorporated into wearables, including vests, patches, and undergarments, enabling POC monitoring for changes in fluid status suggestive of acute decompensation [22–24].

Ballistocardiographic sensors embedded in stationary fixtures such as scales, chairs, beds, under-mattress sensor pads, or even toilet seats can accurately measure recoil force of the body in response to each heart-beat, allowing for the estimation of clinical parameters important in HF patients, including cardiac output and left ventricular systolic function [25–27]. The POC assessment of cardiac output can provide an incredibly valuable and previously inaccessible data to inform the home-based management of HF [27]. Heart rate variability (HRV) is another parameter that can be measured from most wearable devices capable of pulse or ECG assessment, and HRV is a robust predictor of HF decompensation [28]. Finally, several POC devices for measuring B-type natriuretic peptide (BNP) or N-terminal-proBNP (NTproBNP) concentrations in serum to diagnose acute decompensated heart failure(ADHF) have been explored and sensitivity for acute HF decompensation is generally high [29,30].

POC technologies also enable telemedicine solutions for home management of HF. The general consensus is that although certain studies have shown that telemedicine management may lead to improvements in HF-related readmission rates [31,32], further research is needed on whether these approaches are cost-effective or improve long-term outcomes like survival among HF patients. Many mobile applications also exist for self-management of HF. However, these commercially available applications are extremely heterogeneous in quality and self-care components, and this inconsistency is also reflected in the level of evidence generated by studies assessing them [33]. Although many of these studies are underpowered, on the whole they trend towards the conclusion that mHealth interventions may offer a cost-effective method to improve HF patient outcomes. Furthermore, usability studies indicate that social influence as well as perceived ease of use and usefulness of these applications were associated with intention to use mHealth interventions in HF patients. Telemedicine and mHealth strategies for remote management of HF patients certainly show considerable promise for changing the scope of the health care in this field.

Activity tracking and fitness

Perhaps the most well-recognized application of POC technologies in health are pulse-based activity tracker for fitness assessment. Calculation of heart rate is the most common application of pulse measurements. Many commercially available heart rate measurement applications use PPG technologies (both fingertip and facial), which have been shown to be highly accurate in resting and post-exercise states [34,35] when compared with clinical standards, although high-intensity exercise is still associated with increased measurement error [36,37]. These activity trackers are extremely well studied and various meta-analyses have shown them to be effective for increasing physical activity and exercise maintenance [38], even in older adults [39]. This may be due to the incorporation of evidence-based behavior change techniques into watch devices by most commercially available wearable activity trackers [40]. There also exists a large collection of physical activity and weight loss smartphone applications that use a myriad of motivational strategies and tracking functionalities, although recent reviews suggest that the effectiveness of such applications in improving cardiorespiratory fitness and long-term physical activity are likely modest at best [41,42], especially given the wide range of quality in these applications and studies.

Finally, diminished HRV has been long associated with adverse cardiac events, including sudden cardiac death, and long-term survival [43,44], and consequently, many applications and mHealth devices also include the functionality to calculate HRV using PPG data [45]. In the context of fitness tracking, there has been a recent surge of interest in adapting athletic training with this metric [46]. Smartphone-derived measurements of resting HRV have been used to monitor response to and guide training in athletes thus allowing for individualization of training regimens [46,47].

Sleep

As mHealth technologies have evolved, sleep tracking has become an integral component to many health devices currently on the market. Sleep offers a unique window of high-quality data collection that is relatively unabated by motion noise or external disturbances. However, sleep medicine is a relatively underexplored application of sensor-based technologies, and the accuracy of commercial sleep tracking devices is questionable [48,49].

Sleep quality

Many applications and devices advertise themselves as being able to provide accurate data on the quality and duration of sleep for their users. Most POC devices focused on sleep assessment generally use actigraphy from wearable device accelerometers, such as on a smartwatch or wrist-based wearable, to estimate wakefulness, movement, and sleep quality. However, although wrist actigraphy has high sensitivity compared with polysomnography, specificity remains poor owing to frequent miscategorization of wakeful states as sleep [50]. As such, activity-based estimates of sleep duration and quality from smartphones, wearable devices, or motion sensor equipped smart beds may not capture sleep episodes appropriately [48]. Developments in sleep tracking technology integrating environmental cues such as ambient light or noise may serve to increase accuracy of sleep tracking.

One area of sleep quality and tracking that serves as an ideal target for POC technology is in facilitation of sleep hygiene self-management. Many smartphone applications that pair with POC technologies include sleep diaries, audio monitoring for snoring (a marker of sleep apnea), as well as data visualization and analytic tools for the user to better understand their sleep patterns [51,52]. Some applications further facilitate desired sleeping behaviors or modify patterns of sleep through functionalities such as audio interventions, alarms timed in concordance with sleep cycles, or facilitation of mindfulness and relaxation techniques [53].

Sleep apnea

Although advancements in telemedicine have largely streamlined the process of sleep apnea diagnosis through home testing kits, the process itself is still cumbersome and confers high patient burden. Smart-phones have been used to fill this health care need in several ways. Existing smartphone applications have used the smartphone’s ability to emit radio frequency waves, and using the reflected waves like a sonar system, the data are processed through an algorithm that is able to distinguish breathing patterns in wakefulness, normal sleep breathing, and sleep apnea with high accuracy [54]. Other groups have used combinations of parameters such as analyzing motion data from the phone accelerometer, respiratory effort as determined by audio data, or an external oximeter attachment for the smartphone to also accurately classify patients with sleep apnea [55].

Telemedicine and mHealth technologies also present a cost-effective and accurate method for monitoring and encouraging adherence to sleep apnea treatment. The automated transmission of data from newer continuous positive airway pressure devices allows for surveillance and targeted counseling and patient education of patients [56]. In addition, web-based support systems and automated adherence feedback may increase patient adherence, especially early in the treatment course [56].

Lung

Chronic obstructive pulmonary disease

Chronic obstructive pulmonary disease (COPD) management benefits from many different facets of emerging POC technologies and the largest category are biomarker analyzing sensors. Most POC technologies analyze blood or sputum samples, although newer approaches using saliva or breath samples, have also been researched [57]. Most devices measure inflammatory markers, including C-reactive protein and procalcitonin, that are associated with acute COPD exacerbations. In addition, specific pathogen identification from sputum samples (e.g. tuberculosis, pseudomonas, respiratory syncytial virus) have also been proposed and tested in patients with long-term respiratory conditions to guide management with appropriate antibodies during episodes of exacerbation [58,59].

Although spirometry alone may not adequately detect COPD exacerbations, these measures are helpful in detecting COPD or asthma and categorizing disease severity [57]. Accurate and relatively accessible hand-held or smartphone attached spirometers have begun to emerge in the market that may enable more large-scale screening efforts for pulmonary disease [60–62]. Spirometry technology has also been adapted as a smartphone application using the phone’s built-in microphone to measure respiratory effort within reasonable clinical accuracy, and even more recently, this technology has been transformed further to measure spirometry parameters through any phone (including landlines and older cellphone models) through a 1–800 number call service [63].

Smartwatches and textiles are also being used to monitor lung function. For example, although several health-specific smartwatches such as Oxitone’s FDA-cleared wrist blood oxygen sensor and Aseptika’s BuddyWOTCH focus on measuring blood oxygenation, this functionality is by no means exclusive to these devices. Current generation smartwatches from both Garmin (Fenix 5X Plus) and Fitbit (both Ionic and Versa) all have continuous oxygen monitoring capabilities that may help detect lung-related illness. Wearable garments, including vests and shirts [64], have also been developed to monitor lung function in COPD patients. Vibration response imaging technology records acoustic lung vibrations and translates them into numerical lung function parameters, and has been embedded in devices such as chest straps [65], allowing for continuous long-term monitoring. Most smart textiles are also able to accurately measure respiratory rate and some like the Hexoskin vest are additionally able to track lung volume and airflow, although the accuracy may be questionable [66].

Finally, like many other long-term health conditions, telemedicine and remote care platforms may also be effective for management of COPD. Trials evaluating telemedicine approaches for COPD have shown inconsistent results with regard to improved patient outcomes, but there are data to suggest that these technologies may reduce the number of emergency department visits and hospitalization rates [67,68]. Main barriers of use identified in several studies of telemedicine for COPD management included the increased workload for physicians, unclear cost-effectiveness, and the limited quantity or quality of collected data. As POC technologies develop further, medical infrastructures reorganize to accommodate them.

Reactive airway diseases

The diagnosis and management of asthma overlaps in some regards with that of COPD, and as such, several of the technologies described above that measure pulmonary function, oxygen saturation, or respiratory rate can conceivably also be applied to asthma as well. In addition, there is a commercially available fingerstick-based blood test for IgE to aid in the diagnosis and guide medical treatment of asthma [69]. However, the major difference between asthma and COPD management with respect to POC technologies lies in self-management telemedicine options. Studies on the subject are variable in quality, design, and outcomes, but evidence suggests that telemedicine tools for asthma self-management may increase treatment adherence and asthma control [70]. Specifically, technologies that incorporated interactive features and real-time feedback provided more improvements in asthma control [71].

Blood

Sepsis

Sepsis is an area in critical need for novel improvements in diagnostic and monitoring techniques, as timing of diagnosis is critical and every hour of delay in diagnosis and medical intervention can worsen patient outcomes significantly. Several commercially available sensor pads designed to be inserted under a patient’s mattress or chair cushion allow for continuous monitoring of heart rate, respiratory rate, and motion, which can aid in monitoring patients for a change in vital status that may be indicative of sepsis [72]. In addition, traditional blood cultures used to aid in the diagnosis of sepsis generally take many hours to days for results to develop, and almost certainly after the initiation of reflexive broad-spectrum antibiotics rather than targeted therapy, thereby contributing to drug-resistant pathogens. However, advancements in modern molecular diagnostic techniques have increased the accessibility of intensive diagnostic methods that have previously required a central laboratory and enable the measurement of both pathogen information and host response in the setting of sepsis. Several commercially available, FDA-approved products using nucleic amplification techniques are sensitive enough to forego initial microbial growth and provide pathogen information generally within one to 2 h [73]. In addition, nanoparticle-based electro-chemical as well as optical sensors are clinically available for detection of host inflammatory biomarkers suggestive of sepsis (i.e. c-reactive protein, interleukins, procalcitonin) and a significant research focus in recent years has been incorporating these into microfluidic biochips that can be used in POC diagnostic devices. The accessibility provided by these devices in combination with clinical judgment and traditional sepsis risk stratification (i.e. acute physiology and chronic health evaluation(APACHE) or sequential organ failure assessment (SOFA) scores) can significantly aid in the timely diagnosis and management of sepsis patients.

In addition to sensor-based technologies, advancements in telemedicine and machine learning algorithms also have the potential to drastically change the face of sepsis care. As patient complexity has been steadily rising and intensivists remain in short supply, the intensive care unit (ICU) setting has been probably one of the biggest beneficiaries of telemedicine and remote care. Meta-analyses of these tele-ICU systems continually demonstrate improved patient outcomes [74], and specifically in the setting of sepsis or septic shock [75]. These remote systems particularly benefit smaller hospital facilities in rural settings that often face challenges in funding and staffing of their ICU units. In addition, powerful machine learning algorithms for sepsis identification may provide more accurate detection than traditional risk scores based on a limited number of patient parameters, and have prompted the development of various risk stratification and early notification tools that are integrated into various electronic health record systems across the country [76]. Implementation of mobile alerts for sepsis screenings has been shown to cut sepsis mortality by more than half, and drastically decrease ICU length of stay as well as 30-day readmission rates, saving more than $8000 per patient [76,77].

Finally, POC technologies have also been used as an implementation strategy to facilitate sepsis prevention protocols such as hand hygiene. Health care systems have reported clinician reminders and tracking via smartwatch alerts for sepsis protocol adherence positively impacts infection rates in-hospital [78,79].

Bleeding and thrombosis

Coagulation analysis is a robust field in which POC technologies can potentially have significant impact. Oral anticoagulation therapy, especially with vitamin K antagonists like warfarin, is frequently prescribed to patients at high risk of thrombotic events, and although highly effective and cost-efficient, these agents confer considerable patient burden. They have a narrow therapeutic window that can be impacted by factors such as vitamin K content in the patient’s diet, necessitating frequent and accurate testing of the patient’s pro-thrombin time/international normalized ratio (PT/INR) to ensure the regimen falls within proper therapeutic ranges. POC measurement for PT/INR often takes the form of a fingerstick blood test for which multiple large biotech manufacturers, such as Alere, Roche, Philips, and Siemens, have all received FDA approval or clearance. Multiple systematic reviews and meta-analyses have shown POC PT/INR testing to be not only accurate across many settings but also cost-effective compared with standard of care [80,81]. Patient self-management of their oral anticoagulation therapy resulted in more time in the therapeutic window, lower rates of both hemorrhagic and thromboembolic events, and all-cause mortality [80]. These devices can be used not only in the outpatient setting to expedite clinic visits and improve medication adherence [82], but also in hospital settings if timing is imperative such as in the emergency department, as they can provide an instantaneous result.

POC technologies can also be used to help identify thromboembolic events. A handful of POC fingerstick D-dimer assays have either been cleared or approved by the FDA for aiding in or excluding diagnosis of throm-boembolism [83]. This is likely due to the fact that although sensitivity of D-dimer for thromboembolic events is quite high, the specificity is poor, especially in populations with low pre-test probability [84,85]. Even so, these POC D-dimer assays can be used in conjunction with a patient’s pre-test probability to quickly rule out deep vein thrombosis or pulmonary embolism and streamline workflow for providers.

Hemoglobinopathies and anemia

Traditional laboratory-based diagnostic methods for the detection of anemia and hemoglobinopathies cost thousands of dollars in equipment and labor, high-lighting the need for cheap and accessible diagnostic options. POC testing using immunoassays, solubility tests, and density separation are able to distinguish sickle cell disease, and in many cases sickle cell trait, from normal hemoglobin [86]. These options all require minimal blood volume obtainable from a fingerstick and provide instant and accurate results [86]. Further efforts are also underway to extend similar technologies into cost-effective smartphone attachments are promising and cost-effective as well [87]. Several popular POC systems also allow for accurate measurement of complete blood count parameters (i.e. red blood cell count, hemoglobin, mean corpuscular volume, mean corpuscular hemoglobin, platelets, etc), which can inform a potential diagnosis of anemia [88,89]. Furthermore, various microfluidic chip models have been developed to measure ferritin levels to specifically diagnose iron deficiency anemia [90].

Clinical needs and future directions

Although considerable progress has been made in developing promising POC technologies to address chronic disease states across heart, lung, blood, and sleep domains, important gaps still remain. New models are emerging in both academia and industry to implement cross-functional development teams to speed the translational process [91–93]. A key element is the development of novel training programs to expose POC technology developers and users to the clinical and process issues related to device development, validation, and commercialization. The CAPCaT will develop and lead innovative teaching programs and leverage complementary educational opportunities to provide a comprehensive training and dissemination plan for the POCT development community. Highlights of this program will include a short-course for pilot grant awardees modeled on the NSF/NIH I-Corps program. Trainees will learn first-hand through the customer discovery process about the clinical and user needs for POC technologies. This information will be used to inform further development of their own products.

The CAPCaT will also conduct a series of needs assessments for heart, lung, blood, and sleep disorders. Clinical needs assessment strategies encompass both quantitative and qualitative methods and focus on collecting data to identify gaps in health care practices and delivery [94]. A thorough needs assessment allows both the development and continuous monitoring of programs, insuring that activities remain on track, and goals and objectives are met [95]. This is an important step in the research and development process, and given that one of the leading contributors to the valley of death has been suggested to be a lack of nuanced understanding of clinical problems [96], a thorough identification of clinical needs is necessary to maximize likelihood of success in developing POC technologies. Studies show that clinicians are not only largely enthusiastic about POC technologies [97] but also are often able to pinpoint areas of need for their implementation and specify their priorities in functionalities [94,95]. In addition to clinician surveys, focus groups are invaluable in needs assessment, providing in-depth insight into the familiarity, understanding, competency, and confidence of individuals developing these technologies. Results of focus groups afford planners the opportunity to develop effective programming from the outset and assess the effectiveness of the training once completed [98–100]. This need-driven model serves as a strong foundation for POC technology research and development moving forward, and the field as a whole will certainly benefit from a more structured approach in translating promising ideas to clinical reality.

Conclusions

Many of the technologies described in this review showcase innovations in health care that have been proven to improve patient outcomes and quality of care, but the path to the end point of mass adoption is by no means easy. The major goal of CAPCaT, and by extension POCTRN as a whole, is to further enable the realization of ideas such as these by catapulting them past the so-called valley of death and reach actualization in health care settings by providing crucial resources for promising POC technology applications. POC technologies have already fundamentally changed how we approach disease diagnosis and management, and will undoubtedly continue to shape our understanding of health care in years to come, thus it is imperative that organizations such as CAPCaT and POCTRN exist to nurture this growth and evolution.