The 2018 updated National Institute for Health and Care Excellence (NICE) guidance on venous thromboembolism (VTE) (NICE 2018)includes
significant changes and extensions to the 2010 guidance on VTE risk
assessment and thromboprophylaxis for hospital patients (NICE 2010).
The purpose of this guidance was to support further reduction in the
incidence of hospital‐associated thrombosis (HAT), which is estimated to
cause 50–60% of VTE. Over the last decade, pioneering work to reduce
HAT has been led through the UK VTE Exemplar Network of hospitals
(Department of Health 2010; Roberts et al, 2013; Catterick & Hunt, 2014; Rowswell & Nokes, 2017).
One aspect of this work was to encourage hospitals and commissioners of
healthcare to enter into contractual arrangements that ensured root
cause analysis (RCA) of all hospital‐associated VTE events to support
system learning and inform future practice.
Salisbury NHS Foundation Trust became a member of the VTE
Exemplar Network in 2008 and has been collecting data on incidence of
HAT for over 10 years. All new patients diagnosed with deep vein
thrombosis (DVT) or pulmonary embolism (PE) in our community are
referred through the direct access specialist VTE service and data is
also collected on deaths with VTE in hospital. Whole leg scanning is
performed for DVT through a specialist vascular service. The coroner and
local GPs will generally advise us of deaths from VTE in the community
due to a strong engagement culture, but it is possible small numbers of
deaths from VTE in the community may be missed.
HAT data from Salisbury initially included VTE events within 6 weeks of hospital admission. Early intervention with risk assessment and prophylaxis brought significant improvement in HAT incidence. In 2011 the standard for HAT was adjusted to include events within 12 weeks of admission. Gradual but steady improvement in HAT incidence has been delivered with subsequent sequential quality improvements (Fig 1). HAT currently stands at 0·19% hospital admissions, which is lower than any other published standard. Within the last 2 years Salisbury have reviewed a total of 1090 symptomatic VTE events including 902 community acquired events and 188 HAT events, giving a relative HAT incidence of 17·4%. Of the 188 HAT events, 82 were distal DVT (44%), 39 proximal DVT (21%) and 67 PE (35%).
Figure 1 - Salisbury NHS Trust hospital‐associated thrombosis (HAT) data over a 10‐year‐period (2008–2018) showing HAT as a percentage of all hospital admissions (1) = within 6 weeks of admission (2) = within 12 weeks of admission.
The 2018 NICE clinical guideline (NICE 2018) states: “Offer pharmacological VTE prophylaxis for a minimum of 7 days to acutely ill medical patients whose risk of VTE outweighs their risk of bleeding.” This recommendation is also made for patients having abdominal surgery. For other non‐orthopaedic surgical patients, the recommendation is downgraded to “consider” rather than “offer” but in normal practice clinicians find it difficult not to offer prophylaxis to individual patients if NICE recommends consideration of this.
During 2017 Salisbury admitted 42 508 patients and the median length of stay for admissions was 3·37 days for medicine and 1·02 days for surgery. During this same period, 91·6% of our inpatients were deemed at high risk of thrombosis and 78·8% received some form of chemical thromboprophylaxis (including patients normally on oral anticoagulation). In accordance with NICE (2010), thromboprophylaxis was only continued at hospital discharge in a small minority of patients with specific surgical indications (Total knee or hip replacement, neck of femur fracture and some abdomino‐pelvic surgeries for cancer). A coarse calculation shows that, had we applied 2018 guidance on 7 days of treatment for all medical and surgical patients during this period we would have given between 120–200 000 additional days of prophylaxis at a cost of up to £180 000.
Seven of the 65 HAT cases diagnosed in 2016 and 2017 received suboptimal prophylaxis (6 were inpatients at the time of HAT diagnosis). Of these 65 cases, 36 had a prior medical admission and 29 had a prior surgical admission. At the time of VTE diagnosis, 16 medical and 10 surgical patients were still in hospital and receiving thromboprophylaxis. Of the 39 patients that developed VTE after hospital discharge, 8 [2 medical (1 distal DVT, 1 PE); 6 surgical (3 distal, 2 proximal DVT, 1 PE)] developed their HAT within 7 days of hospital discharge from, with 31 occurring at a later date (mean 21/29 days for medicine/surgery; range 8–63 days). Of the 31 later events, a further 8 occurred within 14 days (1 distal, 1 proximal DVT; 5 PE) and therefore might possibly have been prevented by 7 days of post‐discharge prophylaxis. The approximate £180 000 additional cost might therefore have prevented a maximum of 15 post‐discharge VTE events.
The 2018 guidance (NICE 2018) has extended the patient age to include 16‐ to 18‐year‐olds; 1584 patients in this age group were admitted to Salisbury during 2016 and 2017. No cases of VTE at all were diagnosed in this age group during this period where VTE risk assessment was not performed and thromboprophylaxis not given. Had the 2018 standard been applied to this age group many of these young people may have been given off license or off label thromboprophylaxis with no benefit.
The 2018 guidance includes a recommendation for thromboprophylaxis for up to 14 days following arthroscopy with a combined anaesthetic and surgical time of >90 min (NICE 2018). Salisbury does not retrospectively collect data on precise operating times but in 2016 and 2017, 319 arthroscopies were performed without chemical thromboprophylaxis. In 2016 and 2017, only 3 patients developed HAT following arthroscopy, all 3 of whom had distal DVT. It therefore seems that thromboprophylaxis following our local practice in arthroscopy will only be appropriate in a very small minority of patients.
The 2018 guidance recommends consideration of thromboprophylaxis following miscarriage and termination of pregnancy (TOP) (NICE 2018). Salisbury recorded 216 admissions in this group in 2016 and 2017, none of whom received thromboprophylaxis and no HAT events were recorded in this group. It therefore seems that thromboprophylaxis will rarely be a cost‐effective intervention following miscarriage or TOP with our local practice.
In conclusion, evidence from our centre suggests that the additional practical and financial NHS investment in chemical thromboprophylaxis that would result from implementation of the 2018 update of the NICE VTE guidance (NICE 2018) may be out of step with the potential clinical gains in terms of thrombosis prevention. We would like to see published evidence from HAT RCA at other centres before we embark on a complex and costly change in practice.
Conflicts of interest
The authors declare no conflict of interest.