Advertisement

Impact of Vena Cava Filters on In-hospital Case Fatality Rate from Pulmonary Embolism

Published:February 06, 2012DOI:https://doi.org/10.1016/j.amjmed.2011.05.025

      Abstract

      Background

      The effects of vena cava filters on case fatality rate are not clear, although they are used increasingly in patients with pulmonary embolism. The purpose of this investigation is to determine categories of patients with pulmonary embolism in whom vena cava filters reduce in-hospital case fatality rate.

      Methods

      In-hospital all-cause case fatality rate according to the use of vena cava filters was determined in patients with pulmonary embolism discharged from short-stay hospitals throughout the United States using data from the Nationwide Inpatient Sample.

      Results

      In-hospital case fatality rate was marginally lower in stable patients who received a vena cava filter: 21,420 of 297,700 (7.2%) versus 135,240 of 1,712,800 (7.9%) (P<.0001). Filters did not improve in-hospital case fatality rate if deep venous thrombosis was diagnosed in stable patients. A few stable patients (1.4%) received thrombolytic therapy. Such patients who received a vena cava filter had a lower case fatality rate than those who did not: 550 of 8550 (6.4%) versus 2950 of 19,050 (15%) (P<.0001). Unstable patients who received thrombolytic therapy had a lower in-hospital case fatality rate with vena cava filters than those who did not: 505 of 6630 (7.6%) versus 2600 of 14,760 (18%) (P<.0001). Unstable patients who did not receive thrombolytic therapy also had a lower in-hospital case fatality rate with a vena cava filter: 4260 of 12,850 (33%) versus 19,560 of 38,000 (51%) (P<.0001).

      Conclusion

      At present, it seems prudent to consider a vena cava filter in patients with pulmonary embolism who are receiving thrombolytic therapy and in unstable patients who may not be candidates for thrombolytic therapy. Future prospective study is warranted to better define in which patients a filter is appropriate.

      Keywords

      SEE RELATED EDITORIAL AND ARTICLES pp. 429, 465, and 471
      There is an extensive and increasing use of vena cava filters in the United States.
      • Stein P.D.
      • Matta F.
      • Hull R.D.
      Increasing use of vena cava filters for prevention of pulmonary embolism.
      The proportion of patients with pulmonary embolism who received vena cava filters increased linearly from 1991 to 2006, and by 2006, inferior vena cava filters were inserted in 92,000 patients in short-stay hospitals throughout the United States.
      • Stein P.D.
      • Matta F.
      • Hull R.D.
      Increasing use of vena cava filters for prevention of pulmonary embolism.
      Notwithstanding the increased use of vena cava filters, their effects on mortality from pulmonary embolism are not clear, and categories of patients with pulmonary embolism who would benefit most from vena cava filters are not known. The purpose of this investigation is to assess in-hospital case fatality rate among patients with pulmonary embolism to determine categories of patients who might benefit from vena cava filters. This will be a step toward advocating use where shown to be beneficial and addressing the challenge of preventing unnecessary insertion of inferior vena cava filters through improved methods of risk stratification.
      • Greenfield L.J.
      • Proctor M.C.
      • Michaels A.J.
      • Taheri P.A.
      Prophylactic vena caval filters in trauma: the rest of the story.
      • Vena cava filters were strongly associated with reduced in-hospital case fatality rate in unstable patients, whether or not they received thrombolytic therapy.
      • Vena cava filters were strongly associated with reduced in-hospital case fatality rate in stable patients who received thrombolytic therapy.
      • Vena cava filters were not associated with a clinically important decreased case fatality rate in stable patients with pulmonary embolism who did not receive thrombolytic therapy.

      Materials and Methods

      Patients discharged from short-stay hospitals throughout the United States from 1999 to 2008 with pulmonary embolism, and their in-hospital case fatality rate according to the use of vena cava filters, were identified from the Nationwide Inpatient Sample, Healthcare Cost and Utilization Project, Agency for Healthcare Research and Quality.
      HCUP Nationwide Inpatient Sample (NIS) Healthcare Cost and Utilization Project (HCUP). 1998-2008.
      Patients with pulmonary embolism were stratified according to whether they were stable or unstable and further stratified according to whether they received thrombolytic therapy. Additional stratification included whether deep venous thrombosis was diagnosed. Within each group, in-hospital all-cause case fatality rate was assessed according to whether a vena cava filter was inserted.
      The Nationwide Inpatient Sample contains data from 5 to 8 million hospital stays from approximately 1000 hospitals. It is designed to approximate a 20% sample of US nonfederal, short-term hospitals as defined by the American Hospital Association and is stratified according to geographic region, ownership, location, teaching status, and bed size.
      HCUP Nationwide Inpatient Sample (NIS) Healthcare Cost and Utilization Project (HCUP). 1998-2008.
      Weights are provided to calculate national estimates. The Nationwide Inpatient Sample is drawn from those states participating in the Healthcare Cost and Utilization Project. The Nationwide Inpatient Sample contains uniform inpatient stay data collected from existing hospital discharge databases maintained by state agencies, hospital associations, and other private data organizations.
      HCUP Nationwide Inpatient Sample (NIS) Healthcare Cost and Utilization Project (HCUP). 1998-2008.

      Identification of Pulmonary Embolism

      The International Classification of Diseases, 9th Edition, Clinical Modification (ICD-9-CM) codes used for identification of patients with pulmonary embolism were 415.1, 634.6, 635.6, 636.6, 637.6, 638.6, and 673.2.

      Identification of Deep Venous Thrombosis

      The ICD-9-CM codes used for identification of patients with deep venous thrombosis were 451.1, 451.2, 451.8, 451.9, 453.2, 453.4, 453.8, 453.9, 671.3, 671.4, and 671.9. Five-digit codes, such as 451.11 (included under the code 451.1), were not listed, because they were included under the corresponding 4-digit codes.

      Identification of Vena Cava Filter Insertion

      The ICD-9-CM code used for insertion of a vena cava filter was 38.7, “Interruption of the vena cava, insertion of implant or sieve in vena cava, ligation of vena cava (inferior, superior), plication of vena cava.” Since 1979 this code applies nearly entirely to vena cava filters. From 1979 to 1985, the use of surgical caval interruption decreased to virtually zero.
      • Athanasoulis C.A.
      • Kaufman J.A.
      • Halpern E.F.
      • et al.
      Inferior vena caval filters: review of a 26-year single-center clinical experience.
      In 1994, the use of vena caval clips was “considered by most to belong to the history of medicine.”
      • Bergqvist D.
      The role of vena caval interruption in patients with venous thromboembolism.

      Thrombolytic Therapy

      Thrombolytic therapy was identified as ICD-9-CM procedure code 99.10.

      Definition of Unstable

      Unstable was defined as having a listed code for shock (ICD-9-CM code 785.5) or ventilator dependence (ICD-9-CM code V461).

      Statistical Methods

      Because hospital discharge data incompletely capture many diagnoses and procedures, we focused on trends and relative differences of outcome. Patients were matched according to stability, thrombolytic therapy, and diagnosis of deep venous thrombosis. Propensity scores were not done because of the matching. Propensity scores are useful when there is a need to control for multiple confounders simultaneously.
      • Arbogast P.G.
      GCRC Research Skills Workshop.
      Differences in mortality rates were assessed by chi-square using software from GraphPad Software, Inc. (San Diego, Calif). Age was presented as mean±standard deviation, and comparisons were made using Student t test. Relative risk, absolute risk reduction, number to treat to save 1 life, and 95% confidence intervals were calculated using a calculator for confidence intervals of relative risk (www.sign.ac.uk/methodology/risk.xls).

      Results

      From 1999 to 2008, 2,110,320 patients were discharged from short-stay hospitals in the United States with a diagnosis of pulmonary embolism. Demographic characteristics are listed Table 1 according to the various stratified groups.
      Table 1Demographic Characteristics
      Race
      Age (Mean Years ± SD)Gender (% Male)VC FilterNo VC Filter
      VC FilterNo VC FilterP ValueVC FilterNo VC FilterWhiteBlackWhiteBlack
      Stable
       No thrombolytic therapy
        DVT67±1664±17<0.000148%47%77%14%79%13%
        No DVT67±1663±18<0.000146%41%75%16%76%15%
        All67±1663±18<0.000147%43%76%15%77%14%
       Thrombolytic therapy
        DVT56±1854±17<0.000147%53%79%13%79%13%
        No DVT61±1658±18<0.000150%43%76%16%75%18%
        All57±1756±18<0.000148%48%78%14%77%16%
      Unstable
       No thrombolytic therapy
        DVT66±1765±170.000348%47%71%17%72%16%
        No DVT66±1564±17<0.000152%47%72%17%70%17%
        All66±1564±17<0.000150%47%72%17%71%17%
       Thrombolytic therapy
        DVT58±1756±16<0.000149%52%78%13%81%14%
        No DVT62±1658±17<0.000147%52%78%14%75%18%
        All59±1657±17<0.000148%46%78%13%78%16%
      VC=vena cava; PE=pulmonary embolism; DVT=deep venous thrombosis; SD=standard deviation.

      Stable Patients Who Did Not Receive Thrombolytic Therapy

      The majority of patients with pulmonary embolism, 2,010,490 of 2,110,320 (95%), were stable and did not receive thrombolytic therapy. In-hospital case fatality rate was marginally lower in those who received a vena cava filter: 21,420 of 297,700 (7.2%) versus 135,240 of 1,712,800 (7.9%) (P<.0001) (Table 2, Figure 1) . Vena cava filters were not associated with a lower case fatality rate in stable patients diagnosed with deep venous thrombosis. To the contrary, vena cava filters were associated with slightly higher in-hospital case fatality rate if deep venous thrombosis was diagnosed (Table 2). Because stable patients with deep venous thrombosis who had a vena cava filter showed a higher case fatality rate than those who did not have a vena cava filter, we matched these patients by decade of age to be certain that the higher case fatality rate was not due to age. In each decade of age except those aged more than 80 years, in-hospital case fatality rate was higher in those with a vena cava filter. Relative risk, confidence intervals, and number needed to treat are shown in Table 2.
      Table 2Case Fatality Rate, Risk and Number Needed To Treat
      Case Fatality Rate VC Filter n/N (%)Case Fatality Rate No VC Filter n/N (%)P ValueRelative Risk (95% CI)Absolute Risk Reduction (95% CI)NNT (95% CI)
      Stable
       No thrombolytic therapy
        DVT11,590/173,600 (6.7)29,220/547,670 (5.3)<0.00011.25 (1.23-1.28)
        No DVT9,830/124,140 (7.9)106,020/1,165,120 (9.1)<0.00010.87 (0.85-0.89)0.13 (0.11-0.15)85 (75-98)
        All21,420/297,700 (7.2)135,240/1,712,800 (7.9)<0.00010.91 (0.90-0.92)0.09 (0.08-0.10)143 (125-167)
       Thrombolytic therapy
        DVT305/6,120 (5.0)740/9,290 (8.0)<0.00010.63 (0.55-0.71)0.37 (0.29-0.45)34 (27-45)
        No DVT245/2,430 (10)2,210/9,760 (23)<0.00010.45 (0.39-0.50)0.54 (0.55-0.61)8 (7-9)
        All550/8,550 (6.4)2,950/19,050 (15)<0.00010.42 (0.38-0.45)0.58 (0.55-0.62)11 (10-12)
      Unstable
       No thrombolytic therapy
        DVT2,040/6,340 (32)3,940/8,800 (45)<0.00010.72 (0.69-0.75)0.28 (0.25-0.31)8 (7-9)
        No DVT2,220/6,510 (34)15,620/29,200 (53)<0.00010.64 (0.62-0.66)0.36 (0.34-0.38)5 (5-6)
        All4,260/12,850 (33)19,560/38,000 (51)<0.00010.64 (0.63-0.66)0.36 (0.34-0.37)5 (5-6)
       Thrombolytic therapy
        DVT260/4,470 (5.8)710/6,810 (10)<0.00010.56 (0.49-0.64)0.44 (0.36-0.51)22 (18-28)
        No DVT245/2,150 (11)1,890/7,950 (24)<0.00010.48 (0.42-0.54)0.52 (0.46-0.58)8 (7-9)
        All505/6,630 (7.6)2,600/14,760 (18)<0.00010.43 (0.39-0.47)0.57 (0.53-0.61)10 (9-11)
      VC=vena cava; CI=Confidence Interval; NNT=number needed to treat; PE=pulmonary embolism; DVT= deep venous thrombosis.
      Figure thumbnail gr1
      Figure 1In-hospital all-cause case fatality rate in patients with pulmonary embolism who received a vena cava filter and those who did not. Patients are shown according to whether they were stable or unstable, and whether they received thrombolytic (lytic) therapy. Case fatality rate was lower with a vena cava filter in each group (P<.0001). PE=pulmonary embolism; VC = vena cava.

      Stable Patients Who Received Thrombolytic Therapy

      Among stable patients, 27,600 of 2,038,090 (1.4%) received thrombolytic therapy. Those who received a vena cava filter had a lower in-hospital case fatality rate than those who did not: 550 of 8550 (6.4%) versus 2950 of 19,050 (15%) (P<.0001) (Table 2) (Figure 1). Case fatality rate was lower with a vena cava filter irrespective of whether deep venous thrombosis was diagnosed (Table 2).

      Unstable Patients Who Did Not Receive Thrombolytic Therapy

      Only a small proportion of patients with pulmonary embolism were unstable, 72,230 of 2,110,320 (3.4%). The majority of unstable patients, 50,840 of 72,230 (70%), did not receive thrombolytic therapy. Those who received a vena cava filter had a lower in-hospital-case fatality rate than those who did not: 4260 of 12,850 (33%) versus 19,560 of 38,000 (51%) (P<.0001) (Table 2, Figure 1). Case fatality rate was lower with a vena cava filter irrespective of whether deep venous thrombosis was diagnosed (Table 2).

      Unstable Patients Who Received Thrombolytic Therapy

      During the 10 years of study, 21,390 of 72,230 unstable patients (30%) received thrombolytic therapy. Those who received a vena cava filter had lower in-hospital case fatality rate than those who did not: 505 of 6630 (7.6%) versus 2600 of 14,760 (18%) (P<.0001) (Table 2, Figure 1). Case fatality rate was lower with a vena cava filter irrespective of whether deep venous thrombosis was diagnosed (Table 2).

      Discussion

      An impact on case fatality rate with vena cava filters was shown in unstable patients, whether or not they received thrombolytic therapy, and in stable patients who received thrombolytic therapy. Vena cava filters were associated with a lower all-cause in-hospital case fatality rate among unstable patients who received thrombolytic therapy (7.6% vs 18%) and lower all-cause case fatality rate in unstable patients who did not receive thrombolytic therapy (33% vs 51%). The majority of patients, however, were stable. Although a statistically significantly lower in-hospital case fatality rate was shown in stable patients, the difference was not clinically important and there may have been patient selection bias. It is possible that patients with pulmonary embolism and a high likelihood of dying from advanced comorbidities were less likely to receive a filter than those with less severe comorbidities. If filters were inserted more often in healthier patients, this could explain the findings. The reason for inserting or not inserting vena cava filters was not possible to determine.
      Case fatality rate of patients who received vena cava filters was compared with those who did not receive vena cava filters in groups matched according to stability, treatment with thrombolytic agents, and diagnosis of deep venous thrombosis. Patients in each group who received vena cava filters were either the same age or older than those who did not receive filters. Because case fatality rate from pulmonary embolism is directly related to age,
      • Stein P.D.
      • Kayali F.
      • Olson R.E.
      Estimated case fatality rate from pulmonary embolism, 1979-1998.
      the lower case fatality rate in those with vena cava filters was not a result of differences of age. If no beneficial effect of filters was shown, patients were matched according to decades of age to be certain that any lack of benefit was not due to older age.
      In greater Worcester residents, all-cause case fatality rate was higher in patients who received vena cava filters, although patients with vena cava filters were older and had more comorbidities.
      • Spencer F.A.
      • Bates S.M.
      • Goldberg R.J.
      • et al.
      A population-based study of inferior vena cava filters in patients with acute venous thromboembolism.
      Likewise, Kucher et al
      • Kucher N.
      • Rossi E.
      • De Rosa M.
      • Goldhaber S.Z.
      Massive pulmonary embolism.
      reported all-cause case fatality rate was higher at 90 days among patients with non-massive pulmonary embolism who received vena cava filters (21% vs 14%). However, in 11 patients with massive pulmonary embolism, 90-day all-cause case fatality rate was lower with filters.
      • Kucher N.
      • Rossi E.
      • De Rosa M.
      • Goldhaber S.Z.
      Massive pulmonary embolism.
      Among patients at Veterans Affairs Medical Centers, after adjustment for age and illness severity, those with pulmonary embolism who underwent vena cava filter insertion showed a reduced in-hospital all-cause case fatality rate (0.48 odds ratio).
      • Kazmers A.
      • Jacobs L.A.
      • Perkins A.J.
      Pulmonary embolism in Veterans Affairs Medical Centers: is vena cava interruption underutilized?.
      A reduced 30-day case fatality rate from pulmonary embolism was shown among patients who received a vena cava filter in the Japanese Society of Pulmonary Embolism Research registry (odds ratio 0.22).
      • Sakuma M.
      • Nakamura M.
      • Nakanishi N.
      • et al.
      Inferior vena cava filter is a new additional therapeutic option to reduce mortality from acute pulmonary embolism.
      Proctor and Greenfield
      • Proctor M.C.
      • Greenfield L.J.
      Pulmonary embolism: diagnosis, incidence and implications.
      also showed reduced in-hospital case fatality rate among patients with pulmonary embolism who received a vena cava filter (18% vs 44%).
      Permanent vena cava filters were evaluated in the Prévention du Risque d'Embolie Pulmonaire par Interruption Cave study, in which patients with proximal deep venous thrombosis were randomly selected to receive a permanent vena cava filter.
      • Decousus H.
      • Leizorovicz A.
      • Parent F.
      • et al.
      A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis Prévention du Risque d'Embolie Pulmonaire par Interruption Cave Study Group.
      PREPIC Study Group
      Eight-year follow-up of patients with permanent vena cava filters in the prevention of pulmonary embolism: the PREPIC (Prevention du Risque d'Embolie Pulmonaire par Interruption Cave) randomized study.
      All patients received anticoagulants. There was no difference in the death rate at any time during follow-up, comparing those with and without vena filters.
      • Decousus H.
      • Leizorovicz A.
      • Parent F.
      • et al.
      A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis Prévention du Risque d'Embolie Pulmonaire par Interruption Cave Study Group.
      PREPIC Study Group
      Eight-year follow-up of patients with permanent vena cava filters in the prevention of pulmonary embolism: the PREPIC (Prevention du Risque d'Embolie Pulmonaire par Interruption Cave) randomized study.
      Permanent filters reduced the risk of pulmonary embolism but increased the risk of deep venous thrombosis.
      PREPIC Study Group
      Eight-year follow-up of patients with permanent vena cava filters in the prevention of pulmonary embolism: the PREPIC (Prevention du Risque d'Embolie Pulmonaire par Interruption Cave) randomized study.
      In a population-based study of the effectiveness of inferior vena cava filters among patients with venous thromboembolism, in-hospital all-cause deaths were more frequent among those who received filters.
      • White R.H.
      • Zhou H.
      • Kim J.
      • Romano P.S.
      A population-based study of the effectiveness of inferior vena cava filter use among patients with venous thromboembolism.
      However, patients treated with inferior vena cava filters had a significantly greater incidence of prior pulmonary embolism, recent major hemorrhages, malignant neoplasm, and stroke.
      • White R.H.
      • Zhou H.
      • Kim J.
      • Romano P.S.
      A population-based study of the effectiveness of inferior vena cava filter use among patients with venous thromboembolism.
      Retrievable inferior vena cava filters
      • Janjua M.
      • Younas F.
      • Moinuddin I.
      • et al.
      Outcome with retrievable inferior vena cava filters.
      • Stein P.D.
      • Alnas M.
      • Skaf E.
      • et al.
      Outcome and complications of retrievable inferior vena cava filters.
      and permanent inferior vena cava filters
      • Arnold T.E.
      • Karabinis V.D.
      • Mehta V.
      • et al.
      Potential of overuse of the inferior vena cava filter.
      carry risk. Although retrievable vena cava filters can be removed when the risk of recurrent pulmonary embolism is no longer present, pooled data showed that no attempt was made to remove the filters in most patients.
      • Janjua M.
      • Younas F.
      • Moinuddin I.
      • et al.
      Outcome with retrievable inferior vena cava filters.
      The nonselective use of inferior vena cava filters is associated with unacceptable morbidity and mortality.
      • Arnold T.E.
      • Karabinis V.D.
      • Mehta V.
      • et al.
      Potential of overuse of the inferior vena cava filter.
      Some estimated that only a few patients among those surveyed would have benefited from an inferior vena cava filter.
      • Thomsen M.B.
      • Lindblad B.
      • Bergqvist D.
      Fatal pulmonary embolism in an unselected series: the possible role of caval filters in prevention.
      In the Worcester study, use of inferior vena cava filters was considered appropriate in only 51% of patients.
      • Spencer F.A.
      • Bates S.M.
      • Goldberg R.J.
      • et al.
      A population-based study of inferior vena cava filters in patients with acute venous thromboembolism.
      Complications of permanent inferior vena cava filters include improper anatomic placement (7%), migration (2%-3%), angulation of the filter (2%), caval stenosis (2%), caval occlusion (2%-9%), air embolism (1%), penetration of the caval wall (1%), lower-extremity edema (13%-26%), and sequelae of venous stasis (27%).
      • Stein P.D.
      • Kayali F.
      • Olson R.E.
      Trends over 21 years in the use of vena cava filters.
      Deep venous thrombosis at the puncture site has been reported in 8% to 41%.
      • Stein P.D.
      • Kayali F.
      • Olson R.E.
      Trends over 21 years in the use of vena cava filters.
      Additional complications include filter deformation, filter fracture, insufficient opening of the filter, and erosion of the caval wall.
      • Stein P.D.
      • Kayali F.
      • Olson R.E.
      Trends over 21 years in the use of vena cava filters.
      With most types of retrievable filters, thrombi trapped in the filter occurred in 10% to 22%.
      • Stein P.D.
      • Alnas M.
      • Skaf E.
      • et al.
      Outcome and complications of retrievable inferior vena cava filters.
      Anticoagulants generally were not used routinely.
      • Stein P.D.
      • Alnas M.
      • Skaf E.
      • et al.
      Outcome and complications of retrievable inferior vena cava filters.
      It is widely accepted that vena cava filter insertion is indicated in patients with pulmonary embolism if anticoagulants are contraindicated,
      • Kearon C.
      • Kahn S.R.
      • Agnelli G.
      • et al.
      American College of Chest Physicians
      Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition).
      pulmonary embolism has recurred while on adequate anticoagulant therapy,
      ACCP Consensus Committee on Pulmonary Embolism Opinions regarding the diagnosis and management of venous thromboembolism.
      or pulmonary embolism is so severe that any recurrent pulmonary embolism may be fatal.
      ACCP Consensus Committee on Pulmonary Embolism Opinions regarding the diagnosis and management of venous thromboembolism.
      Insertion of an inferior vena cava filter also has been recommended in patients after pulmonary embolectomy, patients with chronic recurrent pulmonary embolism, and patients with pulmonary hypertension.
      • Hyers T.M.
      • Agnelli G.
      • Hull R.D.
      • et al.
      Antithrombotic therapy for venous thromboembolic disease.
      • Jones T.K.
      • Barnes R.W.
      • Greenfield L.J.
      Greenfield vena caval filter: rationale and current indications.
      Inferior vena cava filters also are advocated for patients who show a free-floating thrombus in the inferior vena cava.
      • Ballew K.A.
      • Philbrick J.T.
      • Becker D.M.
      Vena cava filter devices.
      More liberal recommendations by some include prophylaxis in patients with cancer,
      • Cohen J.R.
      • Grella L.
      • Citron M.
      Greenfield filter instead of Heparin as primary treatment for deep venous thrombosis or pulmonary embolism in patients with cancer.
      trauma,
      • Greenfield L.J.
      • Proctor M.C.
      • Michaels A.J.
      • Taheri P.A.
      Prophylactic vena caval filters in trauma: the rest of the story.
      • Carlin A.M.
      • Tyburski J.G.
      • Wilson R.F.
      • Steffes C.
      Prophylactic and therapeutic inferior vena cava filters to prevent pulmonary emboli in trauma patients.
      burns,
      • Still J.
      • Friedman B.
      • Furman S.
      • Cohen H.
      • et al.
      Experience with the insertion of vena caval filters in acutely burned patients.
      or acetabular fracture,
      • Webb L.X.
      • Rush P.T.
      • Fuller S.B.
      • Meredith J.W.
      Greenfield filter prophylaxis of pulmonary embolism in patients undergoing surgery for acetabular fracture.
      hip or knee replacement in patients with a history of thromboembolism,
      • Golueke P.J.
      • Garrett W.V.
      • Thompson J.E.
      • et al.
      Interruption of the vena cava by means of the Greenfield filter: expanding the indications.
      or prophylaxis in all patients with pulmonary embolism, especially if more than 65 years of age.
      • Fink J.A.
      • Jones B.T.
      The Greenfield filter as the primary means of therapy in venous thromboembolic disease.
      Many of the indications for insertion of inferior vena cava filters are a matter of opinion.
      • Girard P.
      • Stern J.-B.
      • Parent F.
      Medical literature and vena cava filters.
      According to a Scientific Statement from the American Heart Association, an inferior vena cava filter should not be used routinely as an adjuvant to anticoagulation and systemic fibrinolysis in the treatment of acute pulmonary embolism.
      • Jaff M.R.
      • McMurtry M.S.
      • Archer S.L.
      • et al.
      Management of massive and submassive pulmonary embolism, iliofemoral deep vein thrombosis, and chronic thromboembolic pulmonary hypertension: a scientific statement from the American Heart Association.
      This recommendation was based on a low quality evidence that included only limited populations, only consensus opinion of experts, case studies, or standard of care. Results of the present investigation support the recommendation that vena cava filters should not be used routinely as an adjuvant to anticoagulation. However, the data suggest that patients who receive systemic fibrinolytic therapy and unstable patients who do not receive systemic fibrinolytic therapy may benefit from routine insertion of a vena cava filter. Likewise, the American College of Chest Physicians Evidenced-Based Clinical Practice Guidelines, 8th edition, recommended against an inferior vena cava filter in addition to anticoagulant therapy for most patients with pulmonary embolism.
      • Kearon C.
      • Kahn S.R.
      • Agnelli G.
      • et al.
      American College of Chest Physicians
      Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition).
      Our data are concordant with this recommendation for most patients with pulmonary embolism. However, among the 3% of patients who were unstable and the 1% who were stable but received thrombolytic therapy, vena cava filters were associated with a lower in-hospital mortality. The data suggest that vena cava filters should be considered in such patients.

      Assumptions

      We assumed that any inaccuracies of coding would be similar among patients with and without vena cava filters in each of the matched categories.
      We assumed that patients who received thrombolytic therapy also received treatment with anticoagulants. It is inconceivable that patients who received thrombolytic therapy would not have been treated with anticoagulants, because a contraindication to anticoagulants also would be a contraindication to thrombolytic therapy. Therefore, vena cava filters in these patients were used in addition to anticoagulants. For patients who did not receive thrombolytic therapy, however, it is possible that some may have received vena cava filters instead of anticoagulants.
      The specificity of ICD-9-CM coding is high. Thus, the majority of procedures that were coded in discharge abstracts actually occurred. The frequencies, however, may be underreported because of the imperfect sensitivity of coding for capturing diagnostic procedures.
      • Romano P.S.
      • Mark D.H.
      Bias in the coding of hospital discharge data and its implications for quality assessment.
      In regard to coding for pulmonary embolism, review and reabstraction of a sample of Medicare hospitalizations from late 1986 and early 1989 showed that 92% of codable cases for pulmonary embolism were on the abstract.
      • Kniffin Jr, W.D.
      • Baron J.A.
      • Barrett J.
      • et al.
      The epidemiology of diagnosed pulmonary embolism and deep venous thrombosis in the elderly.
      A 0.4% prevalence of pulmonary embolism in hospitalized whites and African Americans aged 20 years or more throughout the United States, based on the National Hospital Discharge Survey's use of ICD-9 codes,
      • Stein P.D.
      • Kayali F.
      • Olson R.E.
      • et al.
      Pulmonary thromboembolism in Asian/Pacific Islanders in the United States: analysis of data from the National Hospital Discharge Survey and the United States Bureau of the Census.
      was remarkably close to the incidences of pulmonary embolism in a university hospital (0.4%), a non-university tertiary care center (excluding estimates of undiagnosed pulmonary embolism in patients who died) (0.4%), and a community/teaching hospital (0.3%).
      • Proctor M.C.
      • Greenfield L.J.
      Pulmonary embolism: diagnosis, incidence, and implications.
      • Stein P.D.
      • Henry J.W.
      Prevalence of acute pulmonary embolism in a general hospital and at autopsy.
      • Stein P.D.
      • Patel K.C.
      • Kalra N.J.
      • et al.
      Estimated incidence of acute pulmonary embolism in a community/teaching general hospital.
      These incidences were based on review of multiple data sources, including radiographic reports and autopsies.
      • Proctor M.C.
      • Greenfield L.J.
      Pulmonary embolism: diagnosis, incidence, and implications.
      • Stein P.D.
      • Henry J.W.
      Prevalence of acute pulmonary embolism in a general hospital and at autopsy.
      • Stein P.D.
      • Patel K.C.
      • Kalra N.J.
      • et al.
      Estimated incidence of acute pulmonary embolism in a community/teaching general hospital.
      In regard to discharge codes for deep venous thrombosis, White et al
      • White R.H.
      • Zhou H.
      • Romano P.S.
      Incidence of idiopathic deep venous thrombosis and secondary thromboembolism among ethnic groups in California.
      showed that 92% of 198 coded cases of idiopathic deep venous thrombosis were valid.
      In regard to the robustness of codes for mortality, it would be extraordinary to imagine that some patients were coded as dying when they survived and vice versa. We do not have data on the sensitivity of coding for shock.
      In the early 1980s, the Health Care Financing Administration introduced Diagnosis-Related Groups as a mechanism for reimbursing hospitals providing care to Medicare recipients. Hospital administrations recognize that reimbursement is directly linked to the extent of coding for individual patients. It is likely, therefore, that the sensitivity of coding for the use of vena cava filters, thrombolytic therapy, and ventilatory support was high.

      Conclusions

      An impact on in-hospital case fatality rate with vena cava filters was shown in unstable patients, whether or not they received thrombolytic therapy, and in stable patients who received thrombolytic therapy. The majority of hospitalized patients with pulmonary embolism, however, were stable and did not receive thrombolytic therapy. Future prospective study with assessment of the absolute risks of filter placement in various subgroups of patients is warranted to better define in which patients a filter is appropriate. For now, it seems prudent to consider a vena cava filter in patients with pulmonary embolism who are receiving thrombolytic therapy and in unstable patients who may not be candidates for thrombolytic therapy.

      References

        • Stein P.D.
        • Matta F.
        • Hull R.D.
        Increasing use of vena cava filters for prevention of pulmonary embolism.
        Am J Med. 2011; 124: 655-661
        • Greenfield L.J.
        • Proctor M.C.
        • Michaels A.J.
        • Taheri P.A.
        Prophylactic vena caval filters in trauma: the rest of the story.
        J Vasc Surg. 2000; 32: 490-497
      1. HCUP Nationwide Inpatient Sample (NIS).
        (Agency for Healthcare Research and Quality, Rockville, MD) (Accessed July 12, 2011)
        • Athanasoulis C.A.
        • Kaufman J.A.
        • Halpern E.F.
        • et al.
        Inferior vena caval filters: review of a 26-year single-center clinical experience.
        Radiology. 2000; 216: 54-66
        • Bergqvist D.
        The role of vena caval interruption in patients with venous thromboembolism.
        Prog Cardiovasc Dis. 1994; 37: 25-37
        • Arbogast P.G.
        GCRC Research Skills Workshop.
        (Accessed July 12, 2011)
        • Stein P.D.
        • Kayali F.
        • Olson R.E.
        Estimated case fatality rate from pulmonary embolism, 1979-1998.
        Am J Cardiol. 2004; 93: 1197-1199
        • Spencer F.A.
        • Bates S.M.
        • Goldberg R.J.
        • et al.
        A population-based study of inferior vena cava filters in patients with acute venous thromboembolism.
        Arch Intern Med. 2010; 170: 1456-1462
        • Kucher N.
        • Rossi E.
        • De Rosa M.
        • Goldhaber S.Z.
        Massive pulmonary embolism.
        Circulation. 2006; 113: 577-582
        • Kazmers A.
        • Jacobs L.A.
        • Perkins A.J.
        Pulmonary embolism in Veterans Affairs Medical Centers: is vena cava interruption underutilized?.
        Am Surg. 1999; 65: 1171-1175
        • Sakuma M.
        • Nakamura M.
        • Nakanishi N.
        • et al.
        Inferior vena cava filter is a new additional therapeutic option to reduce mortality from acute pulmonary embolism.
        Circ J. 2004; 68: 816-821
        • Proctor M.C.
        • Greenfield L.J.
        Pulmonary embolism: diagnosis, incidence and implications.
        Cardiovasc Surg. 1997; 5: 77-81
        • Decousus H.
        • Leizorovicz A.
        • Parent F.
        • et al.
        A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis.
        N Engl J Med. 1998; 338: 409-415
        • PREPIC Study Group
        Eight-year follow-up of patients with permanent vena cava filters in the prevention of pulmonary embolism: the PREPIC (Prevention du Risque d'Embolie Pulmonaire par Interruption Cave) randomized study.
        Circulation. 2005; 112: 416-422
        • White R.H.
        • Zhou H.
        • Kim J.
        • Romano P.S.
        A population-based study of the effectiveness of inferior vena cava filter use among patients with venous thromboembolism.
        Arch Intern Med. 2000; 160: 2033-2041
        • Janjua M.
        • Younas F.
        • Moinuddin I.
        • et al.
        Outcome with retrievable inferior vena cava filters.
        J Invas Cardiol. 2010; 22: 235-239
        • Stein P.D.
        • Alnas M.
        • Skaf E.
        • et al.
        Outcome and complications of retrievable inferior vena cava filters.
        Am J Cardiol. 2004; 94: 1090-1093
        • Arnold T.E.
        • Karabinis V.D.
        • Mehta V.
        • et al.
        Potential of overuse of the inferior vena cava filter.
        Surg Gynecol Obstet. 1993; 177: 463-467
        • Thomsen M.B.
        • Lindblad B.
        • Bergqvist D.
        Fatal pulmonary embolism in an unselected series: the possible role of caval filters in prevention.
        Eur J Surg. 1994; 160: 553-559
        • Stein P.D.
        • Kayali F.
        • Olson R.E.
        Trends over 21 years in the use of vena cava filters.
        Arch Intern Med. 2004; 164: 1541-1545
        • Kearon C.
        • Kahn S.R.
        • Agnelli G.
        • et al.
        • American College of Chest Physicians
        Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition).
        Chest. 2008; 133: 454S-545S
      2. ACCP Consensus Committee on Pulmonary Embolism Opinions regarding the diagnosis and management of venous thromboembolism.
        Chest. 1996; 109: 233-237
        • Hyers T.M.
        • Agnelli G.
        • Hull R.D.
        • et al.
        Antithrombotic therapy for venous thromboembolic disease.
        Chest. 2001; 119: 176S-193S
        • Jones T.K.
        • Barnes R.W.
        • Greenfield L.J.
        Greenfield vena caval filter: rationale and current indications.
        Ann Thorac Surg. 1986; 42: S48-S55
        • Ballew K.A.
        • Philbrick J.T.
        • Becker D.M.
        Vena cava filter devices.
        Clin Chest Med. 1995; 16: 295-305
        • Cohen J.R.
        • Grella L.
        • Citron M.
        Greenfield filter instead of Heparin as primary treatment for deep venous thrombosis or pulmonary embolism in patients with cancer.
        Cancer. 1992; 70: 1993-1996
        • Carlin A.M.
        • Tyburski J.G.
        • Wilson R.F.
        • Steffes C.
        Prophylactic and therapeutic inferior vena cava filters to prevent pulmonary emboli in trauma patients.
        Arch Surg. 2002; 137: 521-527
        • Still J.
        • Friedman B.
        • Furman S.
        • Cohen H.
        • et al.
        Experience with the insertion of vena caval filters in acutely burned patients.
        Am Surg. 2000; 66: 277-279
        • Webb L.X.
        • Rush P.T.
        • Fuller S.B.
        • Meredith J.W.
        Greenfield filter prophylaxis of pulmonary embolism in patients undergoing surgery for acetabular fracture.
        J Orthop Trauma. 1992; 6: 139-145
        • Golueke P.J.
        • Garrett W.V.
        • Thompson J.E.
        • et al.
        Interruption of the vena cava by means of the Greenfield filter: expanding the indications.
        Surgery. 1988; 103: 111-117
        • Fink J.A.
        • Jones B.T.
        The Greenfield filter as the primary means of therapy in venous thromboembolic disease.
        Surg Gyenecol Obstet. 1991; 172: 253-256
        • Girard P.
        • Stern J.-B.
        • Parent F.
        Medical literature and vena cava filters.
        Chest. 2002; 122: 963-967
        • Jaff M.R.
        • McMurtry M.S.
        • Archer S.L.
        • et al.
        Management of massive and submassive pulmonary embolism, iliofemoral deep vein thrombosis, and chronic thromboembolic pulmonary hypertension: a scientific statement from the American Heart Association.
        Circulation. 2011; 123: 1788-1830
        • Romano P.S.
        • Mark D.H.
        Bias in the coding of hospital discharge data and its implications for quality assessment.
        Med Care. 1994; 32: 81-90
        • Kniffin Jr, W.D.
        • Baron J.A.
        • Barrett J.
        • et al.
        The epidemiology of diagnosed pulmonary embolism and deep venous thrombosis in the elderly.
        Arch Intern Med. 1994; 154: 861-866
        • Stein P.D.
        • Kayali F.
        • Olson R.E.
        • et al.
        Pulmonary thromboembolism in Asian/Pacific Islanders in the United States: analysis of data from the National Hospital Discharge Survey and the United States Bureau of the Census.
        Am J Med. 2004; 116: 435-442
        • Proctor M.C.
        • Greenfield L.J.
        Pulmonary embolism: diagnosis, incidence, and implications.
        Cardiovasc Surg. 1997; 5: 77-81
        • Stein P.D.
        • Henry J.W.
        Prevalence of acute pulmonary embolism in a general hospital and at autopsy.
        Chest. 1995; 108: 978-981
        • Stein P.D.
        • Patel K.C.
        • Kalra N.J.
        • et al.
        Estimated incidence of acute pulmonary embolism in a community/teaching general hospital.
        Chest. 2002; 121: 802-805
        • White R.H.
        • Zhou H.
        • Romano P.S.
        Incidence of idiopathic deep venous thrombosis and secondary thromboembolism among ethnic groups in California.
        Ann Intern Med. 1998; 128: 737-740

      Linked Article

      • Impact of Vena Cava Filters on In-hospital Case Fatality Rate from Pulmonary Embolism
        The American Journal of MedicineVol. 126Issue 1
        • Preview
          We read with great interest the article by Stein et al1 about the impact of vena cava filters on in-hospital case fatality rate from pulmonary embolism. Utilizing data from the Nationwide Inpatient Sample, the authors concluded that insertion of a vena cava filter is prudent in patients with pulmonary embolism who are receiving thrombolytic therapy and in unstable patients who may not be candidates for thrombolytic therapy. As the authors alluded to in the paper, one potential confounding factor is that patients who are most ill may not receive vena cava filters due to their poor prognosis or general futility.
        • Full-Text
        • PDF