This article presents a protocol based on Hydra — a web-based system for clinical decision support that integrates a full and detailed set of functionalities and services required by physicians for complete cardiovascular analysis, risk assessment, early diagnosis, treatment, and monitoring over time.
Cardiovascular diseases (CVDs) are the leading cause of death throughout the world. The total risk of developing CVD is determined by the combined effect of different cardiovascular risk factors (e.g., diabetes, raised blood pressure, unhealthy diet, tobacco use, stress, etc.) that commonly coexist and act multiplicatively. Most CVDs can be prevented by an early identification of the highest risk factors and an appropriate treatment. The stratification of cardiovascular risk factors involves a wide range of parameters and tests that specialists use in their clinical practice. In addition to cardiovascular (CV) risk stratification, ambulatory blood pressure monitoring (ABPM) also provides relevant information for diagnostic and treatment purposes. This work presents a list of protocols based on the Hydra platform, a web-based system for clinical decision support which incorporates a set of functionalities and services that are required for complete cardiovascular analysis, risk assessment, early diagnosis, treatment and monitoring of patients over time. The program includes tools for inputting and managing comprehensive patient data, organized into different checkups to track the evolution over time. It also has a risk stratification tool to compute a CV risk factor based upon several risk stratification tables of reference. Additionally, the program includes a tool that incorporates ABPM analysis and allows the extraction of valuable information by monitoring blood pressure over a specific period of time. Finally, the reporting service summarizes the most relevant information in a set of reports that aid clinicians in their clinical decision-making process.
Cardiovascular diseases (CVDs) are a group of disorders of the circulatory system that constitute the leading cause of disability and premature death throughout the world1,2. According to the World Health Organization (WHO), an estimated 17.7 million people died from CVDs in 2015, representing 31% of all global deaths1,2. There are many risk factors for CVDs, including behavioral factors such as tobacco use, an unhealthy diet, harmful use of alcohol and inadequate physical activity as well as physiological factors, including raised blood pressure (hypertension), high cholesterol or elevated blood glucose, among others2,3. Hypertension represents a major risk factor for premature cardiovascular disease, being responsible for a high level of cardiovascular morbidity and mortality4,5. Furthermore, it is estimated that the incidence of hypertension among adults in developed countries is almost 40%6,7,8. However, it remains widely undetected, undertreated and poorly controlled3,4.
CVD is a major public health problem which imposes a significant economic burden on any given health-care system6. Early identification of the highest cardiovascular risks and appropriate treatment can prevent clinical events and premature deaths4,5. Hence, there are noticeable health and economic gains attached to comprehensively and thoroughly tracking all these factors. The total risk of developing a CVD is determined by the combined effect of cardiovascular risk factors2,4,5, which commonly coexist and act multiplicatively. Therefore, a total-risk approach is advisable for early detection, as well as for clinical decision-making on the intensity of preventive interventions. Thus, morbidity, early mortality and disability could be reduced and the quality of life could be improved in individuals with an elevated total CVD risk2.
The diagnosis of CVDs is determined by the analysis of a wide range of parameters that are gathered by different procedures used by physicians in their clinical practice. The assessment of these parameters allows the computation of a total CV risk factor which is useful for diagnostic and treatment purposes2,4,5. In addition to the stratification of CV risks, ambulatory blood pressure monitoring (ABPM)9 also provides valuable information. The ABPM test allows the tracking of the patient's blood pressure (BP) during their daily routine, avoiding the influence of the clinical setting (white coat syndrome). Thus, a reliable set of measurements is obtained, allowing the extraction of additional information that supports the clinical decision-making process.
Therefore, the analysis of the cardiovascular system involves a large amount of data, entailing a tedious and time-consuming task that complicates diagnosis and treatment prescription. In this regard, the availability of a patient's full profile that gathers all the required data together with a set of automated services to extract the necessary information would be a significant improvement to guide clinicians in their decision-making process. Apart from this, the availability of an accessible platform that centralizes all patient information not only enables collaboration among different specialists from different locations but also allows discussion of debatable cases and provides reliable diagnoses.
In recent years, the use of computer-based applications and telemedicine has increased considerably, playing an important role in improving public health and welfare in all sectors of the population. This is due to their ability to extract relevant and useful information for the early diagnosis and treatment of several diseases10. The use of these tools improves the quality of health-care services, thus conveniently and reliably satisfying patient demand as well as reducing costs11. As a reference, the number of global imaging-based procedures has risen considerably, given the increasing availability of medical equipment and more sophisticated capture devices. Therefore, Lundberg et al.12 proposed a telemedicine tool to assess digital image quality and agreement between examiners in the field of the otorhinolaryngology. Ortega et al.13 developed SIRIUS, a computer-aided diagnosis framework for the analysis of retinal images. Novo et al.14 also presented their platform for the analysis of retinal microcirculation in combination with carotid macrocirculation.
With regard to CV assessment, there has been a steady increase in the number of tools available throughout the years. Some of the utilities are designed to predict cardiovascular disease risk — such as the tool proposed by Paredes et al.15 — or to calculate risk online by implementing the algorithm proposed by Goff et al.16 according to a guideline on the assessment of cardiovascular risk to calculate the 10-year risk of heart disease. Other systems are designed to be used with mobile phones, such as the proposal of Sufi et al.17 that identifies diseases from body sensors, the device designed by Lin et al.18 for tracking the electrocardiogram in order to detect the presence of abnormal rhythms and send an alarm, the app from Lee et al.19 for monitoring breathing and heart rate values while a person exercising or the application implemented by Kang and Park20 to manage raised blood pressure on the basis of clinical guidelines.
The available utilities are mainly designed to satisfy patient demand in specific scenarios. On the other hand, this article describes a protocol based on Hydra21, a platform focused on the analysis of the cardiovascular system, that is designed entirely to support specialists in their clinical decision-making process. This tool incorporates a set of functionalities and services that physicians require for reliable cardiovascular analysis including risk assessment, early diagnosis, treatment prescription and the monitoring of patients over time. Therefore, there is a tool for the input and management of patient data recorded in different checkups. Then, a risk stratification tool automatically provides a CV risk factor based on different risk stratification tables of reference. In addition to this, the ABPM analysis tool allows the extraction of valuable information from the analysis of blood pressure recordings over a specific period of time. Finally, the most relevant information is summarized in a set of reports that guide clinicians in diagnosis and proper treatment prescription. In this way, the described protocol leads to an improvement in complete cardiovascular analysis supporting a reliable diagnosis and proper treatment. Furthermore, the presented platform allows collaboration among experts, thereby promoting clinical research.
All procedures were conducted under institutionally approved protocols with patient consent.
1. Patient and Checkup Registration
Note: See Figure 1.
2. Risk Stratification Tables
Note: The risk stratification service provides an automatic computation of the CV risk factor based upon various risk stratification tables that are recommended in the guidelines of the European Society of Hypertension/European Society of Cardiology (ESH/ESC)22. For each of the tables, the CV risk factor is computed and recorded based upon various parameters that are uploaded in the patient profile throughout the steps of the checkup data input. The higher or lower importance of each of the tables in the analysis is provided by the specialist while ensuring that each designed stratification table pays special attention to the specific conditions of the patient.
3. ABPM Analysis
Note: ABPM is a common test that allows the monitoring of the patient’s blood pressure throughout their daytime/nocturnal routine9. The device selected for recording ABPM measurements (see the Table of Materials) is among the few BP monitors that are officially validated by international organizations such as the British Hypertension Society (BHS) or the ESH.
4. Clinical Reports
Note: The report service provides a set of reports that gather all the relevant information to support the clinical decision-making process, helping physicians in their clinical practice and promoting collaboration among experts.
The patient registration described in step 1 is carried out by filling in the form presented in Figure 1. Once the user registers a new patient, the application moves forward to introduce the first checkup, which allows the input of comprehensive patient data. Figure 2 shows a screenshot of the first form of the checkup information. Once the Next button is clicked, the application moves forward to the second checkup form presented in Figure 3. After clicking on the End button, the checkup is recorded by the system (assigned to the patient). Hydra (referred to as the 'platform') moves forward to the register checkup page, including all the introduced data. From this page, the user can edit the introduced data or access the implemented treatment form shown in Figure 4, in order to prescribe the patient a specific treatment. Once the checkup registration process is complete, the platform moves forward to the patient page shown in Figure 5, including general data and a list of the submitted checkups.
Besides the centralized management of all the patient data, the platform also provides an automatic computation of the total cardiovascular risk factor based on different risk stratification tables recommended in the standard guidelines of the ESC/ESH. Figure 6 shows an example of CV risk calculation on the basis of the antihypertensive treatment decision table. In this case, the computed risk appears highlighted in the table, and below, the recommendations and treatment related to this risk are shown. Moreover, the different factors that have contributed to the computation of the risk are listed below. An example of the metabolic syndrome (MS) table is shown in Figure 7 including the risk that is obtained using two different criteria of reference and the factors involved in these computations. Figure 8 shows an example of the systematic coronary risk evaluation (SCORE) table indicating the 10-year risk of suffering a coronary event and the list of relevant parameters. Finally, Figure 9 shows an example of the Framingham table that calculates the risk of severe CVD or a hard event and the contribution of each category to the final risk. This way, the service of risk stratification allows an automatic computation of the CV risk on the basis of different tables of reference as well as the involvement of the different parameters that have contributed to reaching the related risk, for a more detailed analysis by the expert clinician.
In addition to risk stratification, the ABPM also provides valuable information to support the clinical decision-making process. Therefore, given an ABPM file containing recordings over a period of time, the tool can provide automatic computation of additional relevant parameters such as the mean and standard deviation of the different measurements (SBP, DBP and pulse), the area of the records over and under the thresholds that represent the maximum normal values, the circadian profile, etc. Figure 10 shows a graphical representation of the ABPM map and a table containing the information automatically computed by the ABPM tool.
Finally, the reporting service provides summarized reports that gather all the relevant, available information to help clinicians in their decision-making process and promotes collaboration among experts. An example of some representative parts of a full report is shown in Figure 11. Similarly, Figure 12 and Figure 13 show examples of a smart report and an ABPM report respectively. All the services the platform has to offer result in improved quality of health-care, while helping physicians perform complete cardiovascular analysis.
Figure 1. Patient registration form. The form is used to register a new patient and includes various global parameters related to patient enrolment. The block of family precedents of premature cardiovascular illness can be hidden. Please click here to view a larger version of this figure.
Figure 2. First form for checkup registration. This includes information about habits, pathologies and previous treatments grouped into different blocks. All the different blocks have the option to keep them hidden or visible. If all the information of a block is unknown, the user should use the hide the block. Please click here to view a larger version of this figure.
Figure 3. Second form for checkup registration. This covers the physical and clinical analyses, grouped in different blocks. All the different blocks include the option to be hidden or visible. If all the information of a block is unknown, the user should use the hidden option. Please click here to view a larger version of this figure.
Figure 4. Implemented treatment form to prescribe the patient any specific treatment. This includes blocks for anti-hypertensive treatment, treatments that can affect blood pressure and other treatments. Please click here to view a larger version of this figure.
Figure 5. The patient profile page. This includes general data and a list of submitted checkups. From this list, it is possible to access the different reports of each checkup. Please click here to view a larger version of this figure.
Figure 6. Example of an antihypertensive treatment decision table. The highlighted cell represents the computed CV risk and the "Risk/treatment" field details the recommendations related to this risk. Moreover, the contribution of the different factors to the final result is listed below. Please click here to view a larger version of this figure.
Figure 7. Example of an MS risk stratification table on the basis of two different criteria. The conditions that are true for each criterion are highlighted in red. The results for each criterion and the factors involved in these computations are shown on the right-hand side. Please click here to view a larger version of this figure.
Figure 8. Example of a SCORE risk stratification table. The highlighted cell corresponds to the 10-year risk of CVD and the list of risk factors summarizing the parameters that lead to this result. Please click here to view a larger version of this figure.
Figure 9. Example of a Framingham risk stratification table. In each table, the contribution of each category is highlighted in red. The computed risk of severe CV or hard events is shown below the tables as well as the risk factors involved in the calculation. Please click here to view a larger version of this figure.
Figure 10. Example of an ABPM map. This includes the graphical representation and complementary measurements of a 48 h monitor register. Green dots represent manual measurements. The red and blue lines are related to the maximum levels for systolic and diastolic blood pressure, respectively. The filled areas correspond to the intervals that exceed these maximum levels during the day and night. Please click here to view a larger version of this figure.
Figure 11. Example of a full report. This report lists all the introduced data for a specific checkup. Some representative portions are included. Please click here to view a larger version of this figure.
Figure 12. Example of a smart report. This report includes essential information to support the clinical decision-making process. It includes the results of the risk stratification tables, the ABPM map and a list of relevant parameters for diagnosis and treatment. Please click here to view a larger version of this figure.
Figure 13. Example of an ABPM report. It includes the ABPM map and all the information extracted from the ABPM service. Please click here to view a larger version of this figure.
The early identification and monitoring of various cardiovascular risk factors together with an appropriate treatment are critical for the prevention of cardiovascular diseases and premature deaths. In the daily clinical routine, clinicians have to handle large amounts of diverse information to check all the different variables and parameters that affect the circulatory system. Hence, it is a tedious and time-consuming task that complicates diagnosis and treatment prescription.
The proposed protocols allow a complete analysis of the cardiovascular system. These protocols include the input of all data related to cardiovascular analysis which are recorded in a full patient profile and organized into different checkups throughout time. The centralized management of these data together with the various services provided by the platform facilitates the clinical decision-making process as well as information interchange between experts. The various services included on the platform were designed and implemented considering the needs and preferences of expert clinicians in order to incorporate all the necessary tools in the best possible way for a comfortable professional use. This way, the checkup service allows the recording of comprehensive patient data, organized into different checkups to track the evolution over time. From the raw data, the platform automatically analyzes and extracts all the properties that are needed for diagnostic and treatment purposes resulting in considerable reduction in time and effort. Here, various risk stratification tables of reference are incorporated into the platform for automatic CV risk computation. Furthermore, the ABPM service allows the tracking of blood pressure over a period of time, allowing the extraction of additional, valuable information. Finally, the report services allow an efficient review of the summarized relevant data.
Therefore, the proposed platform collects a large quantity of diverse, relevant parameters and gathers them using different standard protocols according to ESH/ESC guidelines22 in order to support the decision-making process. The limitation of this protocol is the availability of the large amount of data involved since it comprises an exhaustive anamnesis, a physical examination, recordings of several measurements, biological data extracted from the blood test, knowledge about family precedents, etc. Each of the tools/services combines multiple medical parameters to compute the cardiovascular risk factors in such a way that these calculations cannot be performed when no data are available. However, even if the patient profile is not complete, the availability of partial data allows the computation of some of the risk factors providing relevant results to support the clinical decision-making process. Each service details the data involved in its computation and the results are incremental on the basis of the available data.
There are a number of tools available for CV assessment which are mainly focused on satisfying patient demand in specific scenarios. However, the proposed protocol is fully oriented to medical specialists, covering all the services to support the decision-making process in their daily routine. Regarding BP monitoring, there are several commercial systems that are mostly focused on the performance of the measurements, the compatibility with other operative systems, the ease of use, etc. These devices do not analyze recorded data whereas the ABPM service, given a recording file, will analyze all the information from the measurements and automatically extract valuable parameters relevant for clinical practice. Moreover, it provides a graphical representation incorporating additional data that facilitates visualization and analysis. Finally, the reporting service allows an efficient review of the summarized data containing all the relevant information with the aim of helping clinicians in their clinical practice. Therefore, the proposed protocol allows a complete and reliable analysis of the cardiovascular system to support the clinical decision-making process via a set of functionalities and services that are required by physicians for risk evaluation, early diagnosis, treatment prescription, and tracking over time. This leads to a qualitative improvement in health-care services and a reduction in time and effort, facilitating the work of clinicians in their daily practice.
The large amount of medical data involved together with the possibility of discussions among experts provides an adequate environment for clinical research. Future work in this field will include the analysis of the impact of the different CV risk factors and the correlation between various medical parameters in order to extract additional information relevant to clinical practice. The gathering and storage of significant volumes of clinical data can also serve as a basis for computational analysis of big data with the objective of data dimensionality reduction; this can also serve as a complementary source of information for the clinical users of the platform. Furthermore, future work will involve the inclusion of specific questionnaires — for a more exhaustive analysis of some factors (e.g., stress, diet or exercise) — and internalization in the form of support for more languages and reference units. Graphical improvements are also planned, e.g., the integration of cosinor analysis for blood pressure times series which can facilitate the inspection of blood pressure characteristics and tendencies.
The authors have nothing to disclose.
This work is supported by the Instituto de Salud Carlos III of the Spanish Government and the European Regional Development Fund (ERDF) through the PI14/02161 and the DTS15/00153 research projects and Xunta de Galicia, Centro singular de investigación de Galicia accreditation 2016-2019 Ref. ED431G/01; and Grupos de Referencia Competitiva, Ref. ED431C 2016-047.
Computer with color screen | N/A | N/A | |
Internet connection | N/A | N/A | |
Modern web broser | N/A | N/A | Google Chrome, Internet Explorer, Safari, Fierfox, etc. |
Blood pressure monitor | Spacelabs | N/A | Spacelabs 90217 |