In this issue of Blood, Cannegieter et al use the Danish National Patient Registry to report on the high incidence of venous thromboembolism (VTE), mortality, and arterial thrombosis following a diagnosis of superficial vein thrombosis (SVT). Remarkably, individuals with SVT had nearly a 14% risk of VTE over 10 years, with a 3.3% risk of VTE in the 3 months following the SVT.1 Although during the 10 years of follow-up this translated into an 8-fold increased risk of VTE, the risk was much higher within the 3 months following an SVT (71-fold increased risk of VTE). The hazard ratios of arterial events and death over the entire follow-up were more modest (around 1.2-1.3), however, with noticeably increased risk within 3 months of the SVT event with a 1.5- to 3.5-fold increased risk.
Thrombosis is a family of diseases, with common manifestations such as VTE and SVT and rarer ones such as cerebral sinus thrombosis and splanchnic vein thrombosis. Although there are common themes and risk factors for each site of thrombosis, risks and benefits of treatment as well as individual risk factors may differ. In the case of SVT, the traditional paradigm of it being a benign, self-limiting disease treated with anti-inflammatories and warmth is quickly giving way to the recognition of significant morbidity and even mortality.2 Recent trials of anticoagulants to treat SVT demonstrate reduced progression to VTE and improved patient comfort, both on anticoagulants and after anticoagulants were stopped.3 In fact, this has led to a recommendation for anticoagulant treatment of SVT by the American College of Chest Physicians,4 with some experts recommending that SVT be treated similarly to VTE.2
Cannegieter et al elegantly demonstrate the power of administrative databases to study either rare diseases or common diseases with no centralized care. Because care for SVT, and thrombosis in general, is spread across different provider types and treatment locations, the epidemiology and consequences of SVT are infrequently reported in the literature.5 The Danish National Patient Registry was established in 1977 and over the years has evolved to record more components of health and illness in the population of Denmark.6 Despite multiple insights into the health of the Danish population, this registry was never designed for research, but as a monitoring instrument for hospital activities and a basis for payment for public and private hospitals. The major weakness of this registry is that it is not linked to clinical databases such as electronic health records (EHRs) and so further conformation about the diagnoses and treatments cannot be obtained.1,6
Imagine the potential of harnessing all the information of an EHR in addition to an administrative database, linking billing codes, inpatient and outpatient diagnoses, vital signs, anthropomorphic measures, family and social history, laboratory studies, physician notes, and results from imaging studies.7 At best, EHRs are a means of streamlining health care to improve communication and efficiency. The reality has been a mishmash of incompatible systems designed to improve billing and comply with government mandates in which a “simple” question such as the date of an initial or recurrent VTE event requires human abstraction of the medical record. The potential to assess the health of a population with the EHR, to study rare diseases, and to study common diseases with no centralized care is not realized.7 Hospitals, acute care facilities, laboratories, imaging facilities, and outpatient physician offices all use different records that do not communicate with each other—making it impossible to capture the health of an individual over time.
In the case of SVT, many questions remain unanswered. Despite the recommendation that most SVT be treated with anticoagulants, how commonly is this done in clinical practice? Does anticoagulation change the frequency of VTE or mortality following an episode of SVT? What is the role of the newer oral anticoagulants in treating SVT? How often does bleeding complicate treatment of SVT? Harnessing the power of the EHR could help us efficiently gain insight into the answers to these questions. But first, we need to ensure the EHR accurately captures important outcomes in a manner that improves clinical care and can be used for research.
Overall, this study conclusively demonstrates that SVT is not a benign disease and has consequences for patients. This study also demonstrates the power of population registries in improving health care. With the advent of EHRs in the United States and other nations, we must focus on how we can accurately record the health and illness of populations with the goal of providing cost-conscious, evidence-based care for everyone.
Conflict-of-interest disclosure: The author declares no competing financial interests.