Strategies for Enhancing Anticoagulation Adherence in Adult Patients After Cardiac Valve Replacement: An Evidence-Based Summary

Introduction

Valvular Heart Disease (VHD) is a group of organic disorders characterized by structural or functional abnormalities of the heart valves, leading to valvular stenosis or regurgitation, which subsequently affects intracardiac hemodynamics.¹ Cardiac Valve Replacement (CVR), as a primary therapeutic modality for VHD, typically involves a standard approach via median sternotomy to gain surgical access, followed by opening the pericardium to expose the heart under cardiopulmonary bypass support, and replacing the diseased valve with either a mechanical valve (fabricated from synthetic materials) or a bioprosthetic valve (derived from homologous or heterologous biological tissues).² Mechanical valves constitute 70% of valve replacement procedures³ Due to the inherent material properties of mechanical valves and the patient’s unique hemodynamic profile, their implantation is prone to trigger host coagulation cascades post-replacement, leading to severe complications such as thrombosis and embolism. Consequently, patients undergoing Mechanical Heart Valve Replacement (MHVR) require lifelong anticoagulation therapy and regular monitoring of the International Normalized Ratio (INR)⁴ (Internal Normalized Ratio). INR steadiness is a direct determinant of anticoagulation outcome and safety, and it is predicated on patients’ long-standing, meticulous compliance with their anticoagulant therapy. The World Health Organization (WHO) defines medication adherence as the extent to which a patient’s behavior in taking medication aligns with the recommendations provided by healthcare professionals, encompassing aspects such as timing, dosage, and frequency of administration during treatment.⁵ Both domestic and international studies indicate that patient adherence to anticoagulants following MHVR is a significant challenge. International research has reported that approximately 25% of patients undergoing valve replacement exhibit suboptimal medication adherence,⁶ which further declines markedly after six months of anticoagulant therapy.⁷

Patient medication adherence is influenced by a multitude of factors, including individual responses to treatment (eg, genetics, ethnicity, pathological conditions, pregnancy) and drug interactions, all of which can potentially compromise anticoagulation efficacy. The complexity of INR monitoring, due to factors such as economic burden, time cost, and geographical limitations, often prevents patients from attending timely follow-up appointments, further exacerbating the risk of variability in anticoagulation efficacy. Furthermore, studies indicate that illness perception is a predictor of adherence behavior in patients with cardiovascular disease,⁸ with those holding negative illness perceptions being more likely to be non-adherent to prescribed anticoagulants.⁹ Moreover, limited educational attainment exacerbates communication barriers between clinicians and patients, leading to diminished patient engagement in medical decision-making.¹⁰ Patients may exhibit skepticism towards oral anticoagulants, leading to deviations in medication duration, self-modification of dosage, or even unauthorized discontinuation.¹¹ Additionally, comorbidities and age-related factors, including cognitive impairment, frailty, and fear of falling, are also associated with adherence to anticoagulant therapy^12,13 Therefore, long-term postoperative management of VHD patients faces multiple challenges, including inter-individual variability in treatment response, monitoring complexity, patient cognitive limitations, and comorbidities. This consequently leads to a concerning state of clinical adherence, significantly compromising the efficacy of anticoagulant therapy.

Variations in medication adherence not only correlate with patient characteristics but are also significantly influenced by factors such as healthcare service quality, patient education, clinician-patient communication, and continuity of care. Current healthcare systems and interactive factors exhibit substantial deficiencies in supporting postoperative anticoagulation adherence among MHVR patients, forming a critical knowledge gap that demands urgent resolution. Specifically, patient education and information delivery are generally inadequate, often featuring obscure content lacking personalized design. These materials fail to adequately account for patients’ diverse health literacy levels and cultural backgrounds, resulting in poor disease understanding, frequent misconceptions, and weak self-management capabilities. Concurrently, effective clinician-patient communication models remain unestablished, with recurring issues like ambiguous information delivery, insufficient patient listening, and disregard for patient preferences. Particularly when addressing communication challenges such as educational disparities, clear guidelines for healthcare providers to proactively adjust communication strategies are notably absent, directly exacerbating patient distrust and confusion. Furthermore, healthcare accessibility and continuity suffer severe shortcomings, manifested through uneven distribution of anticoagulation clinic resources, appointment difficulties, irregular follow-up scheduling, and inefficient or absent inter-institutional information-sharing and coordination mechanisms. These deficiencies substantially increase objective barriers for patients to maintain treatment monitoring. Finally, continuity-of-care support systems remain particularly weak. Seamless transitions from hospital to home/community settings prove challenging, discharge planning often becomes superficial, convenient long-term follow-up and consultation channels remain underdeveloped, and timely INR feedback, dosage adjustment guidance, and psychosocial support for addressing treatment fatigue are universally lacking.

Therefore, to enhance medication adherence in patients post-cardiac valve replacement, alongside addressing individual patient factors, significant attention should be directed towards optimizing care processes, reinforcing patient education, and improving the overall quality of healthcare services. Evidence-based nursing (EBN) is an emerging and rapidly developing methodology in recent years. It entails nursing professionals judiciously, wisely, and accurately applying existing research findings, integrating patient needs and preferences with their own clinical expertise, to select and deliver the most effective nursing interventions for optimal patient care. However, a systematic and comprehensive body of evidence concerning the management of oral anticoagulant adherence has yet to be fully established, which consequently impedes the standardized post-operative care for VHD patients. Therefore, this study aims to synthesize the best available evidence for oral anticoagulant adherence management in VHD patients post-surgery, thereby providing a robust foundation for the development of scientific VHD anticoagulation adherence management protocols.

Materials and Methods

Establishment of the Problem

The evidence-based question was constructed using the PIPOST framework, namely: P (Population): patients undergoing cardiac valve replacement who are on oral anticoagulants; I (Intervention): adherence management; P (Professional/Performer): healthcare professionals in cardiovascular surgery; O (Outcome): medication adherence, INR monitoring status, incidence of postoperative complications, and patient satisfaction; S (Setting): postoperative rehabilitation wards and follow-up outpatient clinics; and T (Type of Evidence): guidelines, expert consensuses, systematic reviews, evidence summaries, recommended practices, and randomized controlled trials. This study has been registered with the Fudan University Evidence-Based Nursing Center, registration number ES20257742.

Literature Search Strategy

Evidence retrieval followed a top-down approach according to the “6S” Pyramid of Evidence model. The following databases were searched: Up To Date, BMJ Best Practice, UK National Institute for Health and Care Excellence (NICE), Registered Nurses’ Association of Ontario (RNAO), National Guideline Clearinghouse (NGC), Scottish Intercollegiate Guidelines Network (SIGN), Guidelines International Network (GIN), Cochrane Library, JBI Evidence Synthesis Database, PubMed, Web of Science, China Biology Medicine Disc (CBMdisc), China National Knowledge Infrastructure (CNKI), Wanfang Data, and Medlive Guidelines Network. The databases were searched using a combination of Medical Subject Headings (MeSH) and free-text keywords. The search period was limited from January 1, 2015, to May 31, 2025. The detailed steps for searching English databases, using PubMed as an example, can be found in Figure 1.

Figure 1 Literature search strategy of PubMed.

Inclusion and Exclusion Criteria

Inclusion Criteria

① Studies involving adult patients receiving mechanical ventilation in intensive care units (ICUs); ② Interventions focused on nebulized inhalation therapy and related management measures; ③ Literature types included guidelines, systematic reviews, evidence summaries, randomized controlled trials (RCTs), and expert consensuses; ④ Published in Chinese or English.

Exclusion Criteria

① Duplicate publications; ② Literature with incomplete information or unavailable full text; ③ Literature with a quality appraisal rating of C; ④ Documents solely interpreting existing guidelines.

Literature Quality Appraisal

Literature Appraisal Criteria

Guidelines were appraised using the Appraisal of Guidelines for Research and Evaluation II (AGREE II).¹⁴ Randomized controlled trials and expert consensuses were appraised using the corresponding criteria from the JBI Evidence-based Healthcare Center (2016 edition) critical appraisal tools. Systematic reviews were appraised using AMSTAR 2.¹⁵ Clinical decisions from authoritative databases such as UpToDate and JBI were considered high-quality evidence and directly included.

Literature Quality Appraisal Process

The included literature underwent independent appraisal by two researchers proficient in evidence-based medicine, followed by a rigorous cross-verification process. Any discrepancies that emerged were meticulously resolved by a third reviewer.

Evidence Integration and Grading

The principal investigators, who are also the first and second authors, conducted an exhaustive review of the articles that were finally incorporated into our research. The original literature underlying the included evidence was graded according to the JBI Levels of Evidence and Grades of Recommendation (2014), ranging from Level 1 to 5, with Level 1a being the highest and Level 5c being the lowest.

Results

Literature Inclusion Results

A total of 1807 articles were retrieved. After removing duplicates and studies that did not meet the inclusion criteria, 13 articles were finally included, comprising 5 clinical decisions, 1 evidence summary, 3 guidelines, 2 systematic reviews, 1 expert consensus, and 1 randomized controlled trial. The article screening flowchart is presented in Figure 2. General information about the articles included in our review can be found in Table 1.

Table 1 General Information of Included Studies

Figure 2 PRISMA flow chart of literature search and screening.

Results of Literature Quality Appraisal

Quality Appraisal of Guidelines

A total of 3 guidelines were included in this study. The results of their quality appraisal are presented in Table 2. All included guidelines demonstrated good appraisal consistency and were approved for inclusion.

Table 2 Results of the Evaluation of Guidelines

Quality Appraisal of Expert Consensuses

One expert consensus was included in this study. All items scored “Yes”, indicating high quality, and thus it was included.

Quality Appraisal of Systematic Reviews

Two systematic reviews were included in this study. The systematic review conducted by Anderson, et al scored “Yes” on all appraisal items. For the systematic review by Yuan et al, Item 1 (“Is a preliminary plan provided?”) was “unclear” and Item 4 (“Is the publication status considered in the inclusion criteria, such as grey articles?”) was “no”, while all other items scored “Yes”. All systematic reviews were included.

Quality Appraisal of Randomized Controlled Trials

One randomized controlled trial was included in this study. In the study by A et al, Item 4 (“Blinding of outcome assessment”) scored “No”, while all other items scored “Yes”.

Summary of Best Evidence

Ultimately, 23 pieces of evidence were extracted from 16 articles. Through extraction and synthesis, all evidence was categorized into four dimensions: symptom assessment, monitoring modalities and methods, quality control and management, and patient education. The evidence is shown in Table 3.

Table 3 Summary of Best Evidence for Anticoagulation Adherence Management in Adults After Heart Valve Replacement

Discussion

Enhancing Postoperative Anticoagulation Adherence: Early Identification and Risk Assessment

CVR stands out as a safe and efficacious clinical intervention, capable of markedly enhancing both cardiac hemodynamics and the overall quality of life.²³ However, the interaction between prosthetic valve materials and blood can readily lead to thrombus formation on the valve surface, potentially resulting in severe complications such as stroke or even death.^29,30 Anticoagulation therapy plays a crucial role in preventing post-replacement thrombus formation. By administering anticoagulant medications, the blood coagulation process can be inhibited, reducing the risk of thrombosis and ensuring the normal function of the heart valve. While warfarin is a highly effective and affordable vitamin K antagonist (VKA) for preventing thrombosis after heart valve surgery, its impact on blood coagulation exhibits significant inter-individual variability.³⁰ Patient sex, age, living arrangement, educational level, and clinical follow-up adherence can influence warfarin efficacy to varying degrees.³¹ Therefore, the consolidation of therapeutic efficacy highly depends on postoperative medication management, and enhancing patient medication adherence after surgery has become a critical factor for long-term surgical success. However, patients often exhibit knowledge deficits regarding oral anticoagulation therapy and INR monitoring, which can lead to suboptimal therapeutic outcomes.³² Equally concerning is the risk of excessive anticoagulation, evidenced by an annual intracranial hemorrhage incidence of 0.1–0.9% and major extracranial bleeding rates of 0.4–2.0%, which significantly complicates clinical management.³²

Currently, diverse direct and indirect methods are employed for assessing patient medication adherence.²² Among these approaches, patient self-reporting remains the most direct and clinically practical assessment method. Validated medication adherence scales have demonstrated efficacy in characterizing medication utilization patterns across diverse disease states. While these tools enable rapid adherence evaluation, it should be noted that all methodologies exhibit inherent limitations. For instance, direct assessment methods such as electronic pillbox monitors can record medication intake in real-time but are costly, and patients may decline their use due to privacy concerns; indirect methods, like self-report questionnaires, are susceptible to social desirability bias.³³ These methodological constraints compromise the accuracy of current adherence assessments, hindering timely identification of high-risk populations. Given the accelerated integration of artificial intelligence in contemporary healthcare, we propose conducting large-scale, multi-center cohort studies to address these limitations. These studies could leverage machine learning to precisely identify high-risk postoperative valve replacement patients with poor medication adherence and their individual risk factors, thereby constructing high-quality predictive models for medication adherence. Such models could provide a basis for healthcare teams to develop more effective intervention strategies.

Optimized Anticoagulation Monitoring Models: Precision Management and Strategic Practices

In contrast to bioprosthetic valves, mechanical valves demonstrate superior durability but require indefinite anticoagulation therapy - a therapeutic imperative that intrinsically elevates thromboembolic and hemorrhagic risks. Consequently, proactive prevention of postoperative complications and timely identification of high-risk determinants become paramount, as these measures directly influence long-term clinical outcomes and healthcare resource utilization.

As traditional anticoagulants, Vitamin K Antagonists (VKAs) such as warfarin, have their efficacy influenced by various physiological factors.³⁴ Moreover, insufficient knowledge regarding oral anticoagulants contributes to a narrow therapeutic window, thus demanding precise dose adjustment and frequent INR monitoring to avert complications arising from sub-therapeutic or supra-therapeutic states.³⁵ Although INR monitoring has become standard practice, clinical data indicate that approximately 75% of complications in patients after mechanical valve replacement are still related to thromboembolic or hemorrhagic events.³⁶ The persistent issue of insufficient warfarin adherence annually exposes millions of patients to the risk of preventable severe complications, such as hemorrhagic or ischemic strokes, thereby imposing additional strain on healthcare costs and resource allocation.³⁷

Currently, the monitoring of anticoagulant therapy can be accomplished via a multitude of approaches. These encompass the comprehensive management provided at specialized anticoagulation clinics, the in - depth laboratory - based specimen analysis conducted in hospital laboratories, and the patient - initiated self - monitoring utilizing portable coagulation devices. When juxtaposed with general outpatient clinics, specialized anticoagulation clinics exhibit significantly greater cost - effectiveness in both the monitoring and management of anticoagulation therapy. This cost-effectiveness can be further enhanced, particularly when Point-of-Care (POC) devices are integrated into patient management protocols.³⁸

Especially for patients who have undergone valve replacement, the use of bedside POC devices for INR monitoring can improve monitoring convenience, support patient self-testing (PST) and patient self-management (PSM), and has proven to be safe and effective.³⁹ Self-monitoring includes self-management, where patients perform tests and adjust their treatment according to an algorithm; in self-testing, patients conduct the test but receive treatment advice after sending results to a clinician. In the UK, a retrospective review involving 8763 patients showed that self-monitoring appears to be safe and effective compared to standard monitoring, particularly for patients with prosthetic heart valves.³⁹ Concurrently, self-monitoring of anticoagulation status, especially self-management, appears to be cost-effective compared to routine care.⁴⁰ Furthermore, patient self-management with POC devices enhances adherence self-monitoring awareness, thereby improving overall adherence.⁴¹

Concurrently, with the rapid development of digital healthcare, through the dissemination of health services and information via electronic technologies, telemedicine has gradually emerged as an effective alternative to traditional healthcare models and is widely integrated into comprehensive technological platforms within hospitals and clinics. In recent years, studies have begun to explore the use of Internet-based remote patient monitoring (RPM) combined with portable INR monitoring for assessing anticoagulation efficacy in post-cardiac surgery patients, demonstrating significant positive effects on monitoring, treatment, and patient experience.⁴²

Building Social Support and Enhancing Patient Education: Promoting Sustained Adherence

It is crucial to ensure continuous and comprehensive communication and education regarding treatment initiation, drug and dietary interactions, and temporary cessation during surgical or invasive procedures. Furthermore, continuous patient education is paramount for successful long-term treatment, requiring ongoing attention.⁴³ A cross-sectional study enrolling 383 participants revealed that patients’ cognitive level regarding anticoagulation therapy influenced anticoagulation control⁴⁴ Healthcare professionals can provide knowledge education to patients initiating warfarin therapy to enhance their understanding. Concurrently, a range of diversified educational strategies should be implemented, including face-to-face counseling, the distribution of illustrated educational materials, and the dissemination of online science communication videos. These approaches are designed to address the heterogeneous learning preferences of patients, thereby fostering accurate treatment understanding, enhancing self-management capabilities, and ultimately improving treatment adherence.

Moreover, standardizing follow-up frequency is a crucial component for ensuring treatment efficacy and reducing complication risks, making education on follow-up frequency indispensable for patients. Frequency (eg, regular INR testing) can optimize the dose adjustment of oral anticoagulants, thereby enhancing therapeutic effects and reducing the risk of complications such as bleeding and thromboembolism.⁴⁵ Healthcare professionals should, based on individual patient conditions, explain in detail the time points and frequency requirements for follow-up at different treatment stages. For instance, for patients with stable conditions, follow-up intervals can be appropriately extended; for those with significant fluctuations, the follow-up cycle should be shortened, emphasizing the importance of regular follow-ups in preventing severe complications such as bleeding and thromboembolism. However, traditional laboratory testing methods, where patients seek outpatient physicians to determine INR target values and receive dose adjustment plans after result reporting, result in complex procedures and prolonged patient waiting times, making regular patient monitoring challenging.⁴⁶ During patient education, active guidance should be provided to help patients learn how to use mobile devices. Mobile device-based monitoring, incorporating wearable technology or portable diagnostic tools, enables real-time and continuous tracking of key parameters such as INR. This approach overcomes temporal and spatial limitations, addressing the challenges associated with cumbersome traditional laboratory procedures and prolonged waiting times.

Limitations and Future Directions

Medication adherence is inherently multifactorial, with distinct determinants dominating across different populations. While this evidence summary synthesizes the highest-grade recommendations, its applicability requires contextual calibration. Crucially, the strategies were derived predominantly from studies involving adult MHVR recipients with formal education, potentially limiting generalizability to vulnerable subgroups such as: 1. Elderly patients with impaired cognitive function, reduced self-care capacity, and limited disease awareness; 2. Low-health-literacy populations facing comprehension barriers; 3. Rural communities with restricted healthcare access.These groups exhibit unique adherence determinants (eg, cognitive decline, transportation challenges) insufficiently addressed in current evidence.

Furthermore, real-world implementation gaps may arise from unvalidated feasibility in resource-constrained settings. To bridge this, we propose: conducting high-quality randomized controlled trials (RCTs) targeting specific populations—such as different age groups, genders, countries, and ethnicities—followed by optimizing the evidence through real-world applications. This approach enables dynamic evidence upgrading, ensuring continuous improvement when implementing interventions in actual clinical practice.

Conclusion

This study systematically integrates the best evidence for improving adherence to oral anticoagulation after mechanical valve replacement and provides practical measures directly applicable to clinical practice. It aims to offer actionable recommendations for clinical healthcare professionals and postoperative patients across four dimensions: symptom assessment, monitoring methods, quality control and management, and patient education, thereby reducing complication risks caused by poor adherence. Additionally, evaluating the clinical translation effects of these interventions across diverse cultural backgrounds, patient demographics, infrastructure, and healthcare environments is recommended. This will facilitate effective evidence implementation, help identify barriers to research evidence application, and ultimately maximize patient benefits.