Elsevier

Thrombosis Research

Volume 168, August 2018, Pages 148-155
Thrombosis Research

Full Length Article
Systematic review of risk prediction scores for venous thromboembolism following joint replacement

https://doi.org/10.1016/j.thromres.2018.06.024Get rights and content

Highlights

  • Five VTE risk prediction scores in hip or knee replacement were identified.

  • Number of component variables in a risk score ranged from 5 to 26.

  • Studies did not report calibration or discrimination statistics.

  • Two risk scores were reported to been externally validated.

  • Their impact on clinical practice has not yet been evaluated.

Abstract

Background

Venous thromboembolism (VTE) is an important cause of morbidity and a preventable cause of deaths following lower limb joint replacement. Risk prediction scores that help to predict individual VTE risk following lower limb joint replacement may inform the development of preventive strategies and guide treatment decisions. We aimed to systematically review the evidence on the development and/or validation of risk prediction scores for VTE following lower limb joint replacement.

Methods

Population-based studies that have developed and/or validated a risk prediction score for VTE following hip or knee replacement and randomised controlled trials (RCTs) that have evaluated the clinical impact of a score were searched for in MEDLINE, Embase, Web of Science, and The Cochrane Library to April 2018.

Results

Five observational cohort studies describing five risk scores were included. No RCTs were identified. The number of component variables in a single risk score ranged from 5 to 26. Three risk scores comprised 5–8 component variables. None of the studies reported calibration or discrimination statistics. Two risk scores were externally validated in single-institution cohorts and were reported to perform well. One study evaluated the general surgery Caprini risk score in primary hip and knee replacement patients and did not find it useful for VTE risk stratification.

Conclusions

Few VTE risk prediction scores in lower limb joint replacement exist and these have methodological issues, have been inadequately reported, not been sufficiently validated, and their impact on patient outcomes and decision making is unknown. Research is urgently warranted in the field.

Systematic review registration: PROSPERO 2018: CRD42018088712.

Introduction

Total joint replacement, one of the most common elective orthopaedic procedures, is a highly successful and cost-effective intervention for alleviating pain and disability associated with advanced joint disease such as osteoarthritis [1, 2]. Venous thromboembolism (VTE), which comprises pulmonary embolism (PE) and deep vein thrombosis (DVT), is a frequent complication of lower limb joint replacement. Venous thromboembolism affects several millions of people globally, it is an important cause of long-term morbidity and a preventable cause of deaths, and its management is associated with huge health costs [3]. Despite the effectiveness of anticoagulants at preventing VTE, rates ranging from 0.27% to 61.0% have been reported in patient populations undergoing lower limb replacement [4, 5].

With increasing life expectancy, there is a predicted large rise in the number of people who will be affected by joint disease and hence the number of patients undergoing primary joint replacement [6]. Despite the emergence of newer and more potent prophylactic regimens for VTE, there will also be a proportionate rise in the number of patients who will be affected by VTE. As such, the most appropriate way to tackle this is from a public health perspective – using a preventive approach. This entails identifying patients who are at high risk of developing VTE before they undergo joint replacement and developing preventative measures targeted at these high-risk groups. There are several known predisposing risk factors for VTE development following lower limb joint replacement and these include advanced age; high body mass index (BMI); smoking; and comorbidities such as cardiovascular disease, previous VTE, and cancer [[7], [8], [9]]. These established risk factors have the potential to be used to identify patients who are at high risk of VTE and can also be combined within a risk prediction score or prognostic model to predict VTE outcome risk for individuals. A risk prediction score is a statistical equation that uses multiple prognostic or risk factors in a formal combination to estimate the individualised probability or risk that a certain condition or disease will occur in the future [10]. Given that the risk of VTE development after lower limb joint replacement varies between individuals, there is an interest in developing individualised risk prediction scores for VTE risk; however, little progress has been made in the area. Much research has been focussed on the identification of risk or prognostic factors for VTE in joint replacement patients [11]. Published studies have mostly reported on measures of the strength of the association (e.g., odds ratios, risk ratios, hazard ratios), which do not address the accuracy of these factors in classifying or predicting risk of VTE in individuals following joint replacement [12].

Prevention of VTE following lower limb replacement is a high policy priority and no single risk prediction score has as yet been recommended as being optimal for VTE risk prediction in orthopaedic practice. It is also known that there is often conflicting evidence about the predictive performance of developed risk prediction scores [13]. To our knowledge, there is no summarised evidence on existing risk scores (including their component variables), their predictive performance, and whether their clinical effectiveness have been assessed in well-designed randomised controlled trials (RCTs). In this context, we aimed to systematically review all the available evidence on risk prediction scores for VTE following hip and/or knee replacement. The specific objectives were to: (i) assess clinical variables selected for model inclusion and the predictive performance of these models; (ii) assess if identified models have been externally validated and their performances compared; (iii) assess if the clinical effectiveness of these scores have been evaluated in appropriate RCTs; and (iv) to identify gaps in the existing evidence and whether further research is needed in the field. Our findings should inform clinical practice by identifying host, surgical, and laboratory characteristics that show consistent evidence of prognostic significance and should inform further research in this area.

Section snippets

Data sources and search strategy

We conducted this review in accordance with the CHARMS checklist [14] and PRISMA guidelines [15], (Appendix A) and using a predefined protocol, which has been registered in the PROSPERO prospective register of systematic reviews (CRD42018088712). We searched for eligible studies in MEDLINE, Embase, Web of Science, and The Cochrane Library electronic databases from inception up to 20 April 2018. The computer-based search strategy combined free and MeSH search terms and combination of key words

Study identification and selection

The flow of studies through the screening and selection process is shown in Fig. 1. The literature search strategy identified 603 potentially relevant articles. After an initial screen of titles and abstracts, 11 articles were selected for full text evaluation. Following detailed evaluation, six articles were excluded because (i) they did not report development and/or validation of a specific risk prediction score (n = 4); (ii) the population was not relevant (n = 1); and (iii) the outcome was

Key findings

This systematic review of available risk prediction scores for VTE following hip and knee replacement identified five studies, all published in the last six years. Three studies reported the development of three independent risk scores [[20], [21], [22]], whereas two studies evaluated existing scores originally developed for different populations or outcomes [23, 24]. Only two studies reported on the definitions used for VTE outcomes [20, 22]. The number of component variables for the five risk

Conclusions

Only a small number of risk scores to predict VTE in hip and knee joint replacement patients have been developed and these have several limitations. The existing risk scores have been developed using inadequate methodology, have been inadequately reported, not been sufficiently validated, and their impact on patient outcomes and decision making is unknown. The ACS-NSQIP-derived risk stratification system may have some potential for use in clinical practice; however, inadequate methodology was

Funding sources

This study was supported by the NIHR Biomedical Research Centre at University Hospitals Bristol NHS Foundation Trust and the University of Bristol. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health and Social Care.

Declarations of interest

None.

References (50)

  • J.A. Caprini et al.

    Effective risk stratification of surgical and nonsurgical patients for venous thromboembolic disease

    Semin. Hematol.

    (2001)
  • C.J. Pannucci et al.

    Validation of the Caprini risk assessment model in plastic and reconstructive surgery patients

    J. Am. Coll. Surg.

    (2011)
  • N.A. Zakai et al.

    Risk factors for venous thrombosis in medical inpatients: validation of a thrombosis risk score

    J. Thromb. Haemost.

    (2004)
  • P.R. Fortin et al.

    Outcomes of total hip and knee replacement: preoperative functional status predicts outcomes at six months after surgery

    Arthritis Rheum.

    (1999)
  • L. Li et al.

    Statins for primary prevention of venous thromboembolism

    Cochrane Database Syst. Rev.

    (2014)
  • B.F. Gage et al.

    Effect of genotype-guided warfarin dosing on clinical events and anticoagulation control among patients undergoing hip or knee arthroplasty: the GIFT randomized clinical trial

    JAMA

    (2017)
  • H. Zhang et al.

    Incidence and risk factors of deep vein thrombosis (DVT) after total hip or knee arthroplasty: a retrospective study with routinely applied venography

    Blood Coagul. Fibrinolysis

    (2017)
  • S.M. Kurtz et al.

    Future young patient demand for primary and revision joint replacement: national projections from 2010 to 2030

    Clin. Orthop.

    (2009)
  • N.F. Soohoo et al.

    Factors that predict short-term complication rates after total hip arthroplasty

    Clin. Orthop.

    (2010)
  • S.G. Memtsoudis et al.

    Risk factors for pulmonary embolism after hip and knee arthroplasty: a population-based study

    Int. Orthop.

    (2009)
  • R.H. White et al.

    Risk factors for venous thromboembolism after total hip and knee replacement surgery

    Curr. Opin. Pulm. Med.

    (2002)
  • E.W. Steyerberg et al.

    Prognosis Research Strategy (PROGRESS) 3: prognostic model research

    PLoS Med.

    (2013)
  • J. Zhang et al.

    Risk factors for venous thromboembolism after total hip and total knee arthroplasty: a meta-analysis

    Arch. Orthop. Trauma Surg.

    (2015)
  • M.J. Pencina et al.

    Statistical methods for assessment of added usefulness of new biomarkers

    Clin. Chem. Lab. Med.

    (2010)
  • T.P. Debray et al.

    A guide to systematic review and meta-analysis of prediction model performance

    BMJ

    (2017)
  • Cited by (13)

    • Prostate Cancer History and Total Hip Arthroplasty: A Matched Cohort Analysis Investigating Venous Thromboembolism and Anticoagulation

      2022, Arthroplasty Today
      Citation Excerpt :

      However, patients who undergo “major lower extremity arthroplasty” are automatically classified as “high risk,” and therefore, the clinical utility of this risk-stratifying tool for THA is limited. One systematic review by Kunutsor et al. evaluated several risk assessment tools including the Caprini score and found limited validation, inadequate reporting, and unknown impact on patient outcomes with all currently available VTE risk-stratifying tools for total joint arthroplasty [15]. Therefore, more research is needed to determine the impact of VTE risk factors, such as PCa history, on total joint arthroplasty outcomes to improve clinical guidelines on minimizing VTE risk.

    • Host-related factors for venous thromboembolism following total joint replacement: A meta-analysis of 89 observational studies involving over 14 million hip and knee replacements

      2020, Journal of Orthopaedic Science
      Citation Excerpt :

      The current findings highlight potentially host modifiable VTE risk factors as well as factors that can be easily assessed prior to joint replacement surgery. It has recently been shown that very few VTE risk prediction scores that aid in VTE risk stratification in lower limb joint replacement patients exist for use in clinical practice and these have not been well validated [32]. In addition to established VTE risk factors, some of the factors we have identified have the potential to be used to identify patients who are at high risk of VTE and can also be combined within risk prediction scores or prognostic models to predict VTE outcome risk for individuals.

    • Identification of hypercoagulability with thrombelastography in patients with hip fracture receiving thromboprophylaxis

      2021, Canadian Journal of Surgery
      Citation Excerpt :

      This would allow for potential risk modification by adjusting thromboprophylaxis dosages or using additional antithrombotic medications known to affect TEG parameters.24,25 Increased age and body mass index and female sex, which are factors known to be associated with increased risk for VTE,26,27 were not found to be associated with hypercoagulable TEG values in our cohort at 2 and 6 weeks. This was likely due to the small sample size.

    View all citing articles on Scopus
    View full text