A known unknown? Pharmacological prevention of venous thromboembolism in nursing home residents

Venous thromboembolisms (VTEs), including pulmonary embolism (PE) and deep vein thrombosis (DVT) are a growing healthcare problem strongly associated with mortality and functional deterioration, with a rising prevalence due to aging of the population and age-associated risk-factors.1 Preventing VTE lowers the incidence of morbidity and mortality and improves health-related quality of life.2 Therefore, primary pharmacological prophylaxis is generally recommended in surgical and medical in-hospital wards.3 However, in nursing homes, where residents frequently are at similar or even higher VTE risk than hospitalized medical patients, it is still unknown whether primary pharmacological prophylaxis is an effective and safe approach to prevent VTE.4, 5 On top of the increasing VTE incidence with age,6 nursing home residents are known to be at an additional high risk of first-time VTE compared with community dwelling older persons due to distinct risk factors, such as immobility, ischemic stroke, congestive heart failure, malignancy, and (recurrent) episodes of infection.7, 8 The incidence of acute VTE among nursing home residents varies between 0.7 and 1.3 events per 100 person-years, increasing even further to 2.4 events per 100 person-years when also asymptomatic VTEs are taken into account.9 In an autopsy study of institutionalized older individuals, PE was found to be the clinical cause of death with the lowest rate of detection and highest diagnostic error rate.10 As such, these data suggest that there is a serious underestimation of this potentially lethal, yet preventable condition as most nursing home residents do not show typical signs of VTE due to their diminished mobility, nor will they routinely undergo diagnostic imaging. Furthermore, little is known about the safety and efficacy of thromboprophylaxis to prevent VTE in this population. Indeed, current guidelines provide no specific recommendations for thromboprophylaxis in the nursing home setting.3 This is likely the case because in-hospital studies cannot be readily used to extrapolate to the nursing home population, due to distinct risk factors for VTE in the nursing home population such as dependency in activities of daily living (ADL), wheelchair use, and frequent/prolonged immobilization.5, 8 Alternatively, there exists an incremental increase of bleeding risk in this vulnerable population based on individual risk factors such as falls, dementia, use of non-steroidal anti-inflammatory drugs, and/or renal failure to make a balanced decision on the use of thromboprophylaxis. When we systematically assessed the currently available literature on primary VTE-prevention in nursing home residents (see Appendix S1 for methods [Tables S1–S3] and results [Figure S1]), we identified a vital discrepancy between the high prevalence of VTE in nursing home residents and a complete lack of clinical trials investigating whether pharmacological interventions can safely mitigate this risk. To the best of our abilities, we could not identify any study that formally evaluated the efficacy and safety of thromboprophylaxis in this large, high-risk, often overlooked, population. Nevertheless, critically reviewing the currently available evidence surrounding thromboprophylaxis in nursing home residents may help to expose the exact knowledge gaps and serve as a guide for future research. A summary of the current literature (Table 1) shows that most available evidence primarily addressed the prescribing pattern of primary thromboprophylaxis in a nursing home population, serving at best as circumstantial evidence for the use of pharmacological VTE prevention. In multiple studies11, 12 evaluation of the adequacy of thromboprophylaxis prescription decisions was based on consensus or clinical practice guidelines aimed at hospitalized patients at risk for VTE.13 This highlights the problem at hand; lack of awareness concerning the fact that nursing home residents cannot simply be compared to hospitalized patients in general. In two other studies,14, 15 questionnaires were sent out to geriatricians to survey the rate and duration of thromboprophylaxis in long-term care patients, while two additional studies evaluated thromboprophylaxis prescription in hospice inpatients.16, 17 Unfortunately, none of these studies addressed the safety and efficacy of thromboprophylaxis Nevertheless, all these studies highlight the fact that thromboprophylaxis is deemed important in clinical practice even though evidence supporting its use is lacking. One prospective study in older post-acute care patients did address the efficacy of thromboprophylaxis as well as the lack of an applicable guideline in this specific population of hospitalized older patients.18 A practice guideline (developed using group consensus of expert panelists and a systematic review of literature) for the prescription of thromboprophylaxis was developed after which a preintervention–postintervention study was conducted. The prophylactic intervention resulted in a significant decrease in the rate of any DVT within the study population. Although promising, these results cannot be extrapolated to the nursing home population since it does not account for risk factors specific to the high-risk nursing home population even though the population studied shows great resemblance to the high-risk nursing home population. To provide a uniform prescription guideline specifically for high-risk nursing home residents an American study group developed an assessment tool to assess the risk for VTE among long-term care residents.19 This model also included recommendations on the use of thromboprophylaxis. Although this model could be used to select patients at high risk of thrombosis in the nursing home population, this model has yet to be validated in a prospective clinical trial whether its use is safe and effective. Descriptive cohort study Phase 1 of a 3-phase educational intervention study. It describes current VTE-P practices Altogether, the key issue at hand seems to be that data on VTE prophylaxis from in-hospital or post-acute care settings cannot be extrapolated to the nursing home population, as these populations differ in several key areas. First, compared with hospitalized community dwelling individuals, nursing home residents may carry an increased risk of VTE due to advanced age, higher proportion of comorbidities, ADL-dependency, and immobility. Furthermore, while hospitalized patients only experience a transiently increased risk of VTE, nursing home residents experience a persistently elevated risk of VTE, which may increase even further when common (recurrent) acute medical illnesses occur in this population. Second, while the benefit of preventing fatal VTE may be obvious in both settings, preventing post-VTE morbidity (such as post-thrombotic syndrome or chronic thromboembolic pulmonary hypertension) may be less beneficial in the nursing home population that already experiences a severely reduced exercise intolerance. Third, the nursing home population consists of a clinically heterogeneous group of individuals, while institutional characteristics may further contribute to heterogeneity of potential benefit from VTE prophylaxis. It is likely that traits, such as reason for admission, cognitive status and life expectancy play an important role in whether physicians consider diagnostic investigations to diagnose VTE, let alone to initiate thromboprophylaxis. The main issue to be addressed remains whether primary pharmacological prophylaxis is safe and efficacious in the nursing home population. Therefore, a trial comparing thromboprophylaxis to usual care (or placebo) among high-risk nursing home patients has the main priority. Primary outcomes of interest are in our view the incidence of symptomatic VTE and (major) bleeding. Nevertheless, such a trial alone would not be sufficient to write evidence-based guidelines for the NH population, as additional studies are needed to address (i) the epidemiology and risk factors for VTE in the nursing home, to achieve a validated risk stratification between low-risk and high-risk nursing home residents; (ii) the development of a reliable screening test and subsequent diagnostic trajectory in the nursing home; (iii) implementation strategies of VTE prophylaxis in the nursing homes; and (iv) feasibility, acceptability, and costs of the proposed diagnostic trajectories and treatments. Finally, also health-related quality of life should be integrated as outcome. In conclusion, as Western societies show a demography of more and more older people, the percentage of older individuals residing in long-term care facilities is also increasing. Moreover, this unique, but overall frail population with multiple comorbidities, may not only become severely affected by VTE, but is also underrepresented in the current literature about prevention of VTE. It is therefore of vital importance to determine the impact of thromboprophylaxis in this population. We have identified several important knowledge gaps concerning thromboprophylaxis that cannot be filled in by simply extrapolating hospital practices to the nursing home setting. Therefore, additional studies are needed and we suggest a randomized clinical trial that includes clear risk stratification between low risk and high risk nursing home residents to address whether thromboprophylaxis is safe and efficacious in this population. The authors have declared no conflicts of interest. Study conception and design: Brüggemann, Alnima, ten Cate, Spaetgens, ten Cate-Hoek. Acquisition of data: Brüggemann, Alnima, Spaetgens. Analysis and interpretation of data: Brüggemann, Alnima, Brouns, Spaetgens, ten Cate-Hoek. Drafting of the manuscript: Brüggemann, Alnima, Spaetgens, ten Cate-Hoek. Critical revision of the manuscript for important intellectual content: Brüggemann, Alnima, Brouns, Hanssen, Schols, ten Cate, Spaetgens, ten Cate-Hoek. All authors approved the final version to be submitted for publication. No funding was received for this work. Appendix S1. Supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.