Does preoperative mechanical prophylaxis have additional effectiveness in preventing postoperative venous thromboembolism in elderly patients with hip fracture? – Retrospective case controlled study. PLoS ONE 12(11):e0187337 doi:10.1371/journal.prone.0187337

Presented by: Dr. Sebastian Willis          


Elderly patients who undergo surgery following a hip fracture are at high risk of postoperative venous thromboembolism. This study set out to establish if pre-operative mechanical thromboprophlaxis conferred any additional benefit over an existing regimen of post-operative mechanical and chemical thromboprophlaxis in the prevention of symptomatic venous thromboembolism.

Design & Setting

Retrospective review of records for 614 consecutive patients undergoing surgery for femoral neck or intertrochanteric fracture between January 2012 and June 2016 at a university hospital in the Republic of Korea. The incidence of symptomatic, CT angiogram confirmed venous thromboembolism was recorded.


614 consecutive patients over the age of 70 with femoral neck or intertrochanteric fracture. 75 patients were excluded. The remaining 539 patients fell into two cohorts;

  1. Those undergoing surgery from January 2012 to December 2014 received standard treatment (control group)
  2. Those undergoing surgery from Jan 2015 to June 2016 received study intervention (study group).

There was no process for matching subjects.


  • Patients in the control group received a standard protocol of post operative mechanical and chemical thromboprophylaxis with tinzaprain.
  • Those in the intervention group received graduated compression stockings and intermittent pneumatic compression device pre-operatively and the same standardised protocol for VTE prophylaxis post operatively.


The incidence of symptomatic, CT angiogram confirmed DVT and PE occurring within 30 days of surgery.


  • After multivariate analysis and adjustment for ASA grade, cardiovascular disease, neurological disease, and anticoagulant medication the incidence of symptomatic DVT was lower in the study group (7.4% vs 2.2%, OR 0.28 [0.08-0.95], P = 0.042).
  • The incidence of symptomatic PE was lower in the study group, though this difference was not statistically significant (3.7% vs 1.5%, OR 0.39 [0.09-1.77], P = 0.223).
  • The mean length of stay for the study group was significantly shorter than the control group (17.4 +/-8.5 vs 22.7+/-12.4, P = 0.043).
  • The study group had a significantly higher proportion of ASA III-IV patients (94% vs 84%, P=0.025).
  • The study group were significantly more likely to take anticoagulant or anti-platelet medication prior to admission (39% vs 28%, P=0.025)


Preoperative mechanical prophylaxis may confer additional benefits in preventing symptomatic VTE. More evidence in the form of randomised controlled trials is needed to confirm benefit and exclude confounding variables as the cause for the observed effect.


  • This is the first study to suggest possible benefits of preoperative mechanical prophylaxis in hip fracture patients.
  • The study population was representative of the typical demographic of patients presenting with hip fractures as it retrospectively recruited all patients over 70 undergoing surgery for hip fractures.
  • There were relatively few exclusion criteria.
  • The study only presented data on symptomatic and therefore clinically significant VTE.


  • This is an observational study, therefore there was no randomisation.
  • The authors described this study as a case control study. In fact it is an observational cohort study.
  • There was no patient matching between cohorts.
  • The two cohorts have significant demographic differences as described in the results, though some of these were adjusted for.
  • A further potential confounding factor not described in the paper is the influence of changing practice over time.
  • The two groups were temporally consecutive cohorts, so changes in management not recorded might affect outcomes. For example, recent emphasis on early mobilisation and rehab could influence VTE rates and was not accounted for in the study.
  • There was no mention of adverse events or undesirable side effects associated with mechanical prophylaxis, E.g. patient discomfort, barrier to mobility.
  • Exclusion criteria are not justified in the paper and seem to have been arbitrarily chosen. Example: Patients on warfarin are excluded but those on LMWH are not. Patients who are bed ridden and those who have a history of VTE are excluded despite these patients being at high risk of VTE.


The weaknesses of this study greatly limit the conclusions that can be drawn as confounding variables could explain the difference in rates of VTE.


At this stage application is limited due to weakness of the study. Changes in practice will require further evidence of efficacy and safety before firm recommendations can be made.

Potential for impact

VTE is a significant complication of hip fractures and if further evidence supported the use of mechanical prophylaxis preoperatively there is potential to reduce a significant burden of disease.