Article Text

Download PDFPDF
OC-089 Linked hospital and primary care database study of venous thromboembolism and associated mortality following general surgical procedures in england
  1. G Bouras1,
  2. EM Burns1,
  3. A-M Howell1,
  4. A Bottle2,
  5. T Athanasiou1,
  6. A Darzi1
  1. 1Department of Surgery and Cancer
  2. 2Epidemiology and Public Health, Imperial College London, London, UK


Introduction Trends towards shortening hospital stay through day case surgery and enhanced recovery mean that postoperative venous thromboembolism (VTE) may increasingly occur following hospital discharge. However, population rates of postoperative VTE cannot be measured from hospital databases alone, as they do not capture information outside of the hospital setting. The National Institute for Health and Care Excellence has suggested the use of linked hospital and primary care data to gain a better understanding of rates of hospital care-related VTE,1however, data for surgical patients is lacking.

Method Ninety-day in-hospital and post-discharge VTE and mortality were evaluated in twelve general surgical procedures performed from 1997 to 2012 using linked Clinical Practice Research Datalink (CPRD), Hospital Episodes Statistics (HES), Office of National Statistics (ONS) and National Cancer Intelligence Network (NCIN) databases. The proportion of VTE captured in each database and risk factors for VTE and mortality were evaluated.

Results 168005 operations over fifteen years were evaluated. There were 981 (0.58%) VTE events recorded after 90 days postoperatively. Different datasets recorded variable rates of VTE. CPRD recorded 56.0% (549/981), HES recorded 66.3% (650/981) and ONS recorded 8.3% (81/981) of all recorded VTE. For VTE recorded in two databases, 28.3% (278/981) were in HES and CPRD and 2.1% (21/981) in HES and ONS. Overall VTE rate ranged between 0.08% (14/17554, 3.2/1000 patient years) following haemorrhoidectomy to 2.92% (61/2091, 118.3/1000 patient years) following oesophagogastrectomy. Predictors of VTE included emergency surgery (OR = 1.91), age (OR = 1.02), body mass index (OR = 1.03), previous VTE (OR = 8.07), length of stay (OR = 1.00) and cancer stages II (OR = 1.38), III (OR = 1.50) and IV (OR = 1.63). Major organ resections had the greatest odds of VTE. The frequency of in-hospital VTE ranged from 76.4 to 288.9/1000 person years, while post-discharge VTE ranged from 2.6 to 82.1/1000 patient years. Both in-hospital (OR = 2.07) and post-discharge (OR = 4.03) VTE independently predicted 90-day mortality (p < 0.05).

Abstract OC-089 Figure 1

Overlap between databases capturing venous thromboembolism following general surgical procedures

Conclusion A large proportion of postoperative VTE were detected in the primary care setting. Evaluation of linked databases is a useful way of measuring postoperative VTE at population level. These resources identified substantial harm from post-discharge VTE following general surgical procedures.

Disclosure of interest None Declared.


  1. National Institute for Health and Care Excellence. Venous thromboembolism in patients admitted to hospital. Guideline No. 92, 2010

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.