Upper vs. lower extremity deep vein thrombosis: outcome definitions of venous thromboembolism for clinical predictor rules or risk factor analyses in hospitalized patients
2009; Elsevier BV; Volume: 7; Issue: 6 Linguagem: Inglês
10.1111/j.1538-7836.2009.03351.x
ISSN1538-7933
Autores Tópico(s)Acute Myocardial Infarction Research
ResumoVenous thromboembolism (VTE) remains a major public health problem, with a substantial community-wide burden of disease in acutely ill hospitalized patients. Well-conducted randomized trials in both hospitalized surgical patients and, in the past 10 years, medical patients, have demonstrated the efficacy of thromboprophylaxis and have led to guideline recommendations based on high-quality evidence [1Geerts W.H. Bergqvist D. Pineo G.F. Heit J.A. Samama C.M. Lassen M.R. Colwell C.W. American College of Chest PhysiciansPrevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edn.).Chest. 2008; 133: S381-453Abstract Full Text Full Text PDF PubMed Scopus (3778) Google Scholar]. In addition, the development of clinical predictor rules, risk assessment models and multivariate clinical risk factor analyses for hospitalized patients is based on stringent, objectively verified or clinical outcome measures of VTE as originally defined by these same high-quality studies. The recent SWIVTER study by Kucher et al. [2Kucher N. Spirk D. Kalka C. Mazzolai L. Nobel D. Banyai M. Frauchiger B. Bounameaux H. Clinical predictors of prophylaxis use prior to the onset of acute venous thromboembolism in hospitalized patients SWIss Venous ThromboEmbolism Registry(SWIVTER).J Thromb Haemost. 2008; 6: 2082-7Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar], which identifies clinical predictors of thromboprophylaxis use prior to the onset of acute VTE in hospitalized patients, illustrates this concept of how VTE outcomes are defined for these types of analyses. The method section appropriately states that acute deep vein thrombosis (DVT) or pulmonary embolism (PE) had to be objectively confirmed, and the univariate and subsequent multivariate analyses were based on those VTE outcome definitions. All is well at this point, but one important caveat emerges: there was no explicit mention of whether DVT represented lower extremity (LE) DVT, upper extremity (UE) DVT, or both. Indeed, the fact that an indwelling central line (odds ratio 3.15) was included in the univariate analysis as a predictor leads us to hypothesize that, indeed, the analysis may have counted both UE and LE DVT. Why is the distinction between UE and LE DVT important, especially when establishing clinical prediction rules or risk assessment models based on analyses that are dependent on definitions of VTE? It is well established that the pathophysiology of UE DVT is quite different from that of LE DVT[3Martinelli I. Battaglioli T. Bucciarelli P. Passamonti S.M. Mannucci P.M. Risk factors and recurrence rate of primary deep vein thrombosis of the upper extremities.Circulation. 2004; 110: 566-70Crossref PubMed Scopus (151) Google Scholar, 4Lechner D. Wiener C. Weltermann A. Eischer L. Eichinger S. Kyrle P.A. Comparison between idiopathic deep vein thrombosis of the upper and lower extremity regarding risk factors and recurrence.J Thromb Haemost. 2008; 6: 1269-74Abstract Full Text Full Text PDF PubMed Scopus (74) Google Scholar, 5Watson H.G. Upper extremity deep vein thrombosis – not the same disease at a different site.J Thromb Haemost. 2008; 6: 1267-8Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 6Flinterman L.E. Van Hylckama Vlieg A. Rosendaal F.R. Doggen C.J. Recurrent thrombosis and survival after a first venous thrombosis of the upper extremity.Circulation. 2008; 118: 1366-72Crossref PubMed Scopus (32) Google Scholar]. UE disease may be more likely to be due to local, anatomic or mechanical factors than LE disease, which may reflect (depending upon patient type, location, and presence of bilateral disease) a systemic hypercoagulable state [4Lechner D. Wiener C. Weltermann A. Eischer L. Eichinger S. Kyrle P.A. Comparison between idiopathic deep vein thrombosis of the upper and lower extremity regarding risk factors and recurrence.J Thromb Haemost. 2008; 6: 1269-74Abstract Full Text Full Text PDF PubMed Scopus (74) Google Scholar, 6Flinterman L.E. Van Hylckama Vlieg A. Rosendaal F.R. Doggen C.J. Recurrent thrombosis and survival after a first venous thrombosis of the upper extremity.Circulation. 2008; 118: 1366-72Crossref PubMed Scopus (32) Google Scholar]. The natural history of LE DVT and risk of embolization is well established in the surgical patient, although less so in the medical patient, whereas the natural history of UE disease and risk of embolization are still matters of debate [5Watson H.G. Upper extremity deep vein thrombosis – not the same disease at a different site.J Thromb Haemost. 2008; 6: 1267-8Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar]. Lastly, when developing risk assessment scores or risk factor modeling, the use of mechanical or pharmacologic methods of thromboprophylaxis have been shown – for the vast majority of studies – to be efficacious in preventing LE DVT or PE, with unknown or uncertain effects in reducing the risk of UE disease [7Prandoni P. Prophylaxis of catheter-related thrombosis in cancer patients.Lancet. 2009; 373: 523-4Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar]. Whereas the presence of UE DVT probably represents a small fraction of the overall VTE burden in large-scale hospital-based studies assessing the efficacy of a particular prophylactic intervention across a defined group of patients, its presence (or absence) may have a greater impact when assessing the magnitude of effect when the VTE event rate is small. In addition, there is the possibility of confounding when UE events are included across smaller subgroups of patients in analyses identifying specific risk factors or when developing risk scores or clinical prediction rules. Our group conducted a subgroup analysis using the IMPROVE database [8Tapson V.F. Decousus H. Pini M. Chong B.H. Froehlich J.B. Monreal M. Spyropoulos A.C. Merli G.J. Zotz R.B. Bergmann J.F. Pavanello R. Turpie A.G. Nakamura M. Piovella F. Kakkar A.K. Spencer F.A. Fitzgerald G. Anderson Jr, F.A. IMPROVE InvestigatorsVenous thromboembolism prophylaxis in acutely ill hospitalized medical patients: findings from the International Medical Prevention Registry on Venous Thromboembolism.Chest. 2007; 132: 936-45Abstract Full Text Full Text PDF PubMed Scopus (435) Google Scholar] to elucidate VTE risk factors in developing a weighted risk score in hospitalized medical patients (A.C. Spyropoulos, F.A. Anderson, G. Fitzerald, H. Decousus, V. Tapson, for the IMPROVE Investigators, unpublished data). The final VTE risk score from the multivariate analysis based upon seven clinical variables was very different when UE DVT was (appropriately) excluded as an outcome measure, probably due to confounding. Furthermore, close inspection of the VTE outcomes used to assess reduction of VTE events using a computerized alert system based on a weighted risk score by Kucher et al. [9Kucher N. Koo S. Quiroz R. Cooper J.M. Paterno M.D. Soukonnikov B. Goldhaber S.Z. Electronic alerts to prevent venous thromboembolism among hospitalized patients.N Engl J Med. 2005; 352: 969-77Crossref PubMed Scopus (773) Google Scholar] reveals that more than half of the events (32/62) in the intervention group and nearly one-third of the events (33/103) in the control group consisted of DVT of the arms. As evidence-based thromboprophylactic measures in hospitalized patients are designed to reduce DVT of the legs and PE, this calls into question the value of that particular intervention and the VTE risk score on which that intervention was based. In addition, a validated risk score of chemotherapy-associated VTE based on clinical events was not explicit in differentiating between UE and LE DVT (as UE DVT is likely in this patient group), calling into question the usefulness of the model in identifying high-risk outpatient cancer groups for studies on thromboprophylaxis, as the authors suggested [10Khorana A.A. Kuderer N.M. Culakova E. Lyman G.H. Francis C.W. Development and validation of a predictive model for chemotherapy-associated thrombosis.Blood. 2008; 111: 4902-7Crossref PubMed Scopus (1553) Google Scholar]. The development of risk scores or clinical prediction rules based upon VTE event rates in hospitalized patients should have a common goal of identifying patient groups with specific risk factors that would place them at a high enough risk to benefit from thromboprophylaxis or in conducting future studies using specific prophylaxis interventions. As such, these risk scores/prediction rules and the possible prophylaxis interventions derived by them should be based on definitions on objectively verified or clinical VTE outcomes that explicitly exclude UE DVT, owing to the differences in the pathophysiology and natural history of disease between UE and LE DVT, and the uncertainty of thromboprophylactic measures in the setting of UE disease. Indeed, the recent Joint Commission/National Quality Forum core measures in the USA for prevention and care of VTE have excluded UE DVT as an outcome measure in potentially preventable hospital-acquired VTE [11http://www.jointcommission.org/PerformanceMeasurement/PerformanceMeasurement/VTE.htm. Accessed 4 April 2009.Google Scholar]. The author states that he has no conflict of interest.
Referência(s)