Abstract

The writing committee of the recently released "AHA/ACC Guideline for the Management of Patients with Valvular Heart Disease" used a new format to enhance the utility of the guideline in clinical practice and particularly at the point of care. The three major desired outcomes that emerged from the process included 1) creation of data-driven recommendations whenever possible, 2) development of recommendations that are most useful and accessible for clinicians, and 3) provision of recommendations to allow decision-making on an individual patient basis. By creating and storing recommendations and evidence using a systematic knowledge management system, the guideline will become more widely available for use by clinicians at the point of care, as a living document that contains only the most updated, critical knowledge required for optimal patient care.

Introduction

The recently updated "AHA/ACC Guideline for the Management of Patients with Valvular Heart Disease" were released in June 2014.^1,2 A group of 12 physicians with expertise in valvular heart disease (VHD) was convened to revise prior iterations of the guideline and included representation from general cardiology, cardiovascular imaging, interventional cardiology, surgery, and anesthesiology. At the initial meeting, the writing committee reviewed the methodology used to create the previous guidelines. It was unanimously agreed that the current committee would use a new format to enhance the utility of the guideline in clinical practice, especially at the point of care. The three major desired outcomes that emerged from the discussion included 1) creation of data-driven recommendations whenever possible which is the foundation of the ACC/AHA guideline methodology, 2) development of practice guidelines that are most useful and accessible for clinicians, and 3) provision of recommendations to allow decision-making on an individual patient basis. This manuscript outlines the methodology used to achieve these goals.

The overall purpose of the ACC/AHA clinical practice guidelines is to evaluate evidence, perform an expert analysis of data, and create recommendations that aid the practicing clinician. ³ "The medical profession should play a central role in evaluating the evidence related to drugs, devices, and procedures for the detection, management, and prevention of disease. When properly applied, expert analysis of available data on the benefits and risks of these therapies and procedures can improve the quality of care and optimize patient outcomes." ^1,2 Guidelines should be updated regularly, reviewing the continuously increasing body of new knowledge to assist physicians in providing the optimal and most up-to-date care for their patients. ^3,4

The "AHA/ACC Guideline for the Management of Patients with VHD" was first published in 1998. ⁵ A revision appeared in 2006, and a focused update was provided in 2008. ^6,7 The ACC/AHA Task Force on Clinical Practice Guidelines and members of the previous VHD guideline writing committees continually examined emerging data to determine when a new revision might be indicated. In 2012, a full revision of the VHD guideline was deemed necessary for several reasons. First, the natural history of patients with VHD has been better defined, based upon the outcomes of a relatively large number of patients meeting specific echocardiographic criteria and followed for an extended period of time. Second, advancements in noninvasive imaging and Doppler hemodynamics have provided the clinician with the tools to better image and quantitate the severity of valve disease. Third, there have been significant improvements in short- and long-term outcomes with surgery. Fourth, catheter-based interventions to treat patients with valve disease have become a reality, based in part on carefully performed randomized trials that have documented patient outcomes and resulted in regulatory approval of new devices. Thus, a better understanding of the natural history of patients with VHD, enhanced utility of noninvasive imaging, and continuously improving outcomes from an expanding array of valve treatments have lowered the threshold for intervention, both among asymptomatic patients and for older and sicker patient populations.

Data-Driven Recommendations

There has been controversy regarding the quality of the scientific evidence underlying some ACC/AHA clinical practice guidelines. ⁸ The observation has been that practice recommendations have been largely developed on the basis of lower levels of evidence (LOE) that rely primarily on expert opinion. Thus, efforts are needed to clearly define and expand the evidence base and improve the process of writing guidelines. ⁹ This issue is specifically applicable to VHD, in which there remains a paucity of randomized trials to determine the safety and efficacy of several therapeutic interventions.

Previous iterations of the VHD guideline relied on the following methodology. Each author was assigned a specific valve lesion for which text chapters were written, based upon the authors' review of the literature and their own experience. Class of recommendations (COR) were then developed by individual authors from each of their chapters, and then discussed and debated by the full writing committee. Using this approach, over 70% of the VHD guideline recommendations were LOE C and based on expert opinion only.

After review of the prior process, the 2014 committee recommended the following steps for revision of the guidelines:

• First, construct a new taxonomy based upon the knowledge required for the clinical care of patients with VHD. This taxonomy then forms the basis of the clinical questions relevant to each condition.
• Next, systematically review the literature and summarize the evidence in detailed tables to support answers to the clinical questions, as recommended more recently by the ACC/AHA Task Force on Clinical Practice Guidelines. All members of the committee would critically review these evidence tables.
• Then, write guideline recommendations, with the COR and the LOE for each recommendation provided. Every effort is to be made to reduce the number of recommendations with LOE C, expert consensus.
• The committee would discuss each guideline recommendation to arrive at a consensus, using the evidence summarized in the detailed tables.
• Finally, link each recommendation to references supporting the LOE as included in summary evidence tables, and write a short (<500 words) text section providing further details or explanation.

By following this systematic process, the proportion of recommendations with LOE B increased from 30% to 45%, and those with LOE C decreased from >70% to <50%.

Decision-Making for the Individual Patient

The committee reviewed updated information on the natural history of the major valve lesions, and the outcome of earlier interventions for asymptomatic patients with severe valve disease as reported in observational studies. The overall result of this process was a generally lower threshold for operation for most valve lesions in order to prevent the long-term adverse consequences of VHD. However, it was discussed and emphasized that the decision to intervene on an asymptomatic patient must be individualized and dependent upon procedural availability and expertise at each institution, as well as the values and preferences of the individual patient and societal norms.

For example, in the past, the Class I indications for operation for a patient with severe degenerative mitral regurgitation were either the presence of symptoms or clear evidence of asymptomatic left ventricular systolic dysfunction, which were conservative recommendations based on the relatively high operative mortality and potential complications of implanting a valve prosthesis in patients for whom repair did not prove feasible. Long-standing volume overload due to mitral regurgitation inevitably will lead to irreversible left ventricular dysfunction, which might not be detected by either the onset of symptoms or measurements of left ventricular size and systolic function until relatively late in the natural history of the disease. In addition, mitral valve repair can now be performed with a low operative risk and excellent long-term outcome in patients with valve anatomy amenable to repair. Thus, the 2006 committee had arrived at the recommendation that it is reasonable to operate before the onset of left ventricular dysfunction or symptoms in selected patients with asymptomatic severe mitral regurgitation who could undergo valve repair with a high success rate and a low operative risk. In 2014, this evolved toward higher standards of referral of patients to centers of excellence, in which the anticipated success rate for mitral repair should exceed 95%.

Data on the outcomes of patients undergoing early operation have been generated by surgeons at institutions with a large experience and substantial expertise in mitral valve repair. There is large variability in the experience of U.S. centers for mitral valve operations, with the majority of centers performing fewer than 20 mitral valve repairs per year, with a range of 1-166, and a median of five mitral operations per surgeon per year. ¹⁰ A large proportion of patients continue to receive a valve prosthesis for primary mitral regurgitation even when valve anatomy may be amenable to surgical repair. Thus, earlier operation in the asymptomatic phase of disease can only be recommended when performed by experienced surgeons in high volume centers who can demonstrate a low operative mortality with a high rate of successful and durable mitral valve repair.

To address these concerns, the writing committee introduced the concept of "Heart Valve Centers of Excellence," criteria for which are: 1) a heart valve team consisting (at a minimum) of a multidisciplinary group of cardiologists, imaging experts, surgeons, and cardiovascular anesthesiologists, 2) a high level of surgical expertise in complex valve operations including repair of primary mitral regurgitation, 3) high volumes of patients, and 4) active participation by the center in registry data reporting and in internal continuous quality improvement processes.

Individual procedural risk assessment is also central to the decision-making process for patients with VHD. There are increasing numbers of elderly patients with advanced comorbidities being considered for either operation or transcatheter valve repair or replacement. Chronologic age alone does not necessarily predict procedural outcome or complications. ¹¹ However, the optimal method for risk stratification remains controversial, especially because many current risk scores were developed from surgical, not catheterization-based, databases and registries. Surgical risk scores are derived only from patients who have undergone surgery and do not include characteristics of the very high-risk patients in whom surgery has been denied. The writing committee recommended that risk stratification for procedural or surgical operative morbidity and mortality should be performed in higher risk patients presenting with VHD. This assessment should include a measure of frailty, particularly in older patients being considered for surgical or transcatheter aortic valve replacement. Shared decision-making with the patient and family should be implemented to achieve the optimal choice of treatment, which in turn is dependent on individual patient needs and preferences. Providing a single recommendation for all patients with VHD is not feasible or appropriate; an individualized decision based upon multiple patient, operator, and institutional factors is critical.

ACC/AHA Guidelines in the 21st Century; Providing Knowledge at the Point of Care

Physicians are currently facing an information overload. According to the National Library of Medicine, there are over 1,500 articles published in the medical literature every day and <1% of these are relevant to a physician's practice. Randomized clinical trials in cardiovascular medicine are presented at over 10 annual scientific meetings and published in over 50 cardiovascular journals. In addition, the intensity of clinical practice has increased, giving physicians little time to assimilate new knowledge. In a busy clinician's day, there are important clinical questions left unanswered due to the pace and complexity of modern medicine. Thus, a practicing physician needs the 'gist' at the point of care, defined as concise relevant bytes of knowledge that answer a specific question, synthesized by trusted experts.

The ACC/AHA clinical practice guidelines are generally recognized as the flagship of U.S. cardiovascular medicine. They are evidence-based, synthesized by experts, unbiased, and highly vetted. However, in the past, practice guidelines have been written in unstructured text format as linear documents printed on the page or viewed on a computer screen. Valuable clinically relevant knowledge is buried within the documents, which sometimes exceeded 200 pages in length, making it difficult to easily access the precise information needed for specific clinical situations. The writing committee recognized that in order to facilitate providing the clinician with relevant information at the point of care, the guideline needed to be created and organized in a structured data format, which then could be applied to knowledge management.

Instead of using a textbook format, the new guideline taxonomy was based on how clinicians think and act in a sequential timeframe. The taxonomy would serve as a useful foundation to access the knowledge needed during any part of a patient workup. For each section of the new taxonomy, knowledge bytes were created which were concise, relevant, practical, and evidence-based class recommendations. Supporting text was next developed in a concise format with a set word limit. Links to references and figures were attached to each knowledge byte. These knowledge bytes will be stored in a content management system, with proper tagging for retrieval of individual knowledge bytes at the point of care.

This formatting results in a novel process for the creation and retrieval of knowledge. Specifically, a search engine can be used to retrieve specific knowledge bytes at the point of care, when matched against semantic analysis of clinical questions. Knowledge bytes can all be grouped together to form educational programs. A future goal will be to incorporate these specific knowledge bytes into the electronic health record, converting to an "executable machine readable text".

Figure 1 Figure 1. Sequential examples of the display on a hand-held device in response to a clinical question at the point of care. Figure 2 Figure 2. An overview of the algorithmic approach to knowledge creation, storage and dissemination. After the completion of a full guideline, continuous surveillance of new knowledge, vetted by the committee members will be sustained. When there is enough evidence to affect a change in a guideline recommendation, the new knowledge is introduced as a new knowledge byte and inserted into the centralized knowledge management system. This single change will result in dissemination of this new knowledge to caregivers through 1) point of care knowledge using hand held devices 2) a full guideline document and 3) educational websites.

Figure 1 provides an example of how a clinician would access a specific knowledge byte during a patient encounter. In this instance, a young woman with a mechanical prosthetic valve presents 8 weeks pregnant, raising the question of the optimal anticoagulation regimen, which was addressed in the 2014 guideline. The clinician would type or use a voice recognition program to input "anticoagulation for mechanical valves during pregnancy" (Figure 1 left). A search engine would rapidly identify the knowledge byte that addresses this question, and then display the Class IIa recommendation which supports the use of warfarin if the current dose is less than 5 mg per day (Figure 1 middle). If the clinician wants more information, a further click can expand to display the concise and relevant supporting text, and link to appropriate references and evidence tables (Figure 1 right).

ACC/AHA Guidelines: A Living Document

The exponentially increasing knowledge base in cardiovascular medicine requires continual updates to the ACC/AHA clinical practice guidelines. With each guideline revision, approximately 20% of the Class I recommendations are changed. Some guidelines have changed over 80% of their Class I recommendations. ¹² However, each full guideline revision is labor intensive and may take up to 2-4 years from initiation of the writing committee to publication, by which time further new knowledge has often been generated. A major advantage of creating chunked and tagged knowledge bytes is that the guideline can now become a living document. (Figure 2) As new knowledge arises, the writing committee is able to search all related stored knowledge bytes. After discussion and rigorous review, only the specific knowledge bytes (i.e., recommendations and supporting text) related to new knowledge or information need to be changed—-not the entire document. Since there is overlap in specific knowledge areas from guideline to guideline (e.g., stable ischemic heart disease and percutaneous coronary intervention), a centralized database can be created for the knowledge bytes from all ACC/AHA clinical practice guidelines, with intersecting (or overlapping) content appropriately linked. When new knowledge arises that is pertinent to multiple guidelines, there will then be a single repository from which the recommended change is made, which is then incorporated into all relevant guidelines.

Conclusion

Continuous process improvement is necessary to enhance the utility of the ACC/AHA clinical practice guidelines. ^3,4 These guidelines are the flagship of the ACC, AHA, and partnering organizations and represent the vast cumulative knowledge evaluated by the content experts. They should be driven by evidence, address continued changes in practice, and support individual patient care in pragmatic and easily accessible ways. By creating and storing recommendations and evidence using a systematic knowledge management system, guidelines will become more widely available for use by clinicians at the point of care, as a living document that contains only the most updated, critical knowledge required for optimal patient care.

References

1. Nishimura RA, Otto CM, Bonow RO, et al. 2014 AHA/ACC guideline for the management of patients with VHD: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014;63:2438-88.
2. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP III, Guyton RA, O'Gara PT, Ruiz CE, Skubas NJ, Sorajja P, Sundt TM III, Thomas JD. 2014 AHA/ACC guideline for the management of patients with VHD: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014; 129:
3. Jacobs AK, Anderson JL, Halperin JL. The evolution and future of ACC/AHA clinical practice guidelines: A 30-Year Journey: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014;64:1373-84.
4. Institute of Medicine and Committee on Standards for Developing Trustworthy Clinical Practice Guidelines. Clinical Practice Guidelines We Can Trust. Washington, DC: The National Academies Press;, 2013.
5. Bonow RO, Carabello B, de Leon AC, et al. ACC/AHA guidelines for the management of patients with VHD. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients With VHD). J Am Coll Cardiol 1998;32:1486-1588.
6. Bonow RO, Carabello BA, Chatterjee K, et al. ACC/AHA 2006 guidelines for the management of patients with VHD: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing Committee to Revise the 1998 guidelines for the management of patients with VHD) developed in collaboration with the Society of Cardiovascular Anesthesiologists endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons. J Am Coll Cardiol 2006;48:e1-148.
7. Bonow RO, Carabello BA, Chatterjee K, et al. 2008 focused update incorporated into the ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Develop Guidelines for the Management of Patients With VHD). J Am Coll Cardiol 2008;52:e1–142.
8. Tricoci P, Allen JM, Kramer JM, Califf RM, Smith SC, Jr. Scientific evidence underlying the ACC/AHA clinical practice guidelines. JAMA 2009;301:831-41.
9. Kung J, Miller RR, Mackowiak PA. Failure of clinical practice guidelines to meet institute of medicine standards: Two more decades of little, if any, progress. Arch Int Med 2012;172:1628-33.
10. Bolling SF, Li S, O'Brien SM, Brennan JM, Prager RL, Gammie JS. Predictors of mitral valve repair: clinical and surgeon factors. Ann Thorac Surg 2010;90:1904-11.
11. Cleveland JC, Jr. Frailty, aging, and cardiac surgery outcomes: the stopwatch tells the story. Journal of the American College of Cardiology 2010;56:1677-8.
12. Neuman MD, Goldstein JN, Cirullo MA, Schwartz JS. Durability of class I American College of Cardiology/American Heart Association clinical practice guideline recommendations. JAMA 2014;311:2092-100.

Keywords: Anesthesiology, Cardiology, Catheterization, Comorbidity, Consensus, Electronic Health Records, Hemodynamics, Mitral Valve, Mitral Valve Insufficiency, Myocardial Ischemia, Patient Care, Percutaneous Coronary Intervention, Prostheses and Implants, Research Design, Risk, Risk Assessment, Surgeons, Transcatheter Aortic Valve Replacement, Ventricular Dysfunction, Left, Warfarin

< Back to Listings