European Journal of Haematology
ORIGINAL ARTICLE
Patterns of use and outcome of inferior vena cava filters in a tertiary care setting Amihai Rottenstreich1, Galia Spectre1, Batia Roth1, Allan I. Bloom2, Yosef Kalish1 1
Department of Hematology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel; 2Department of Radiology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel
Abstract Background: Inferior vena cava (IVC) filter placement is increasing although the evidence to justify their use is limited. Many filters are left in place indefinitely, thereby exposing patients to long-term complications. Objectives: To review indications, complications, and follow-up data of patients undergoing IVC filter placement at our center. Methods: A retrospective review of consecutive admitted patients who underwent IVC filter insertion in a large university hospital with a level I trauma center. Thrombosis specialists retrospectively assessed the appropriateness of indication for IVC filter placement as well as referral for retrieval. Results: Overall, 405 filters were inserted between 2009 and 2013. All filters were retrievable. IVC filter was placed as a primary prevention in 42% of patients. Fifty-two patients (12.8%) experienced at least one filter-related complication. The most common complication was deep vein thrombosis occurring in 6.9% of cases. Almost a third of all complications occurred in filters originally placed for prophylactic indications. Only 13.6% of filters were successfully removed. Nevertheless, according to our reviewers, an attempt for filter retrieval should have been made in 57% of all cases and in 86% of trauma patients. A significantly higher retrieval rate was found in patients followed at our thrombosis clinic (P < 0.01). During follow-up, 95 patients (23.4%) died, most of them with active cancer. Conclusions: IVC filters are placed in many cases for prophylactic indications. Their low retrieval rates together with relatively high risk of long-term complications, questions their extensive utilization. Prospective trials addressing the safety and efficacy of IVC filters are still warranted. Key words inferior vena cava filter; deep vein thrombosis; pulmonary embolism; venous thromboembolism Correspondence Yosef Kalish, MD, Department of Hematology, Hadassah-Hebrew University Medical Center, POB 12000, Jerusalem, Israel 91120. Tel: ++972 2 6779415; Fax ++972 2 6449580; e-mail: [email protected] Accepted for publication 4 March 2015
Venous thromboembolism (VTE) manifesting as deep vein thrombosis (DVT) and pulmonary embolism (PE) is a serious and potentially lethal disorder that affects 100 per 100 000 annually and results in significant morbidity and mortality (1). The standard of care is anticoagulation therapy, with inferior vena cava (IVC) filters being used as an adjunctive or alternative measure. The concept of surgical vena cava interruption was first introduced in 1865 by Trousseau (2). In 1967, surgical insertion of IVC filters was pioneered but still was infrequently used (3). Since then, the rate of IVC filter placement has increased significantly in the United States from 2000 patients in 1979 to 49 000 in 1999, with a threefold increase from 2001 to 2006 (4, 5). At least part of this dramatic rise is explained by the ease of
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
doi:10.1111/ejh.12542
percutaneous placement, pioneered in 1973 by Greenfield, with the subsequent introduction of retrievable filters in the 1990s (3, 4). Secondly and most importantly, the largest increase in use has been due to filter use in patients with no evidence of DVT or PE (i.e. prophylactic use), representing 21% of all cases (6). Although the use of IVC filters is rapidly increasing, the majority of evidence for their insertion, particularly for prophylactic indications, comes from observational studies which often lack a control group. Lack of high-quality data makes the decision to place an IVC filter controversial on many occasions. According to Spencer et al. (7), IVC filter insertion is appropriate in only 51% of cases. This is highlighted by the wide variability of indications for IVC filters insertions according to different
1
IVC filters: indications, complications, and follow-up
professional society guidelines. The latest principal guidelines were published by the American College of Chest Physicians (ACCP), American Society of Hematology (ASH) and the Society of Interventional Radiology (SIR) (8–11). While the ACCP and ASH advocate the use of IVC filters only when anticoagulation is contraindicated and refute their prophylactic use, the SIR guidelines are more permissive and have recommended a list of absolute, relative, and prophylactic indications. Moreover, while the retrieval rates of IVC filters are reported to be very low (2–20%), filters may be associated with significant long-term risks such as DVT (including IVC and filter thrombosis) IVC penetration, filter dislocation, and migration and rupture of IVC which raises concerns about their expanding utilization (12, 13). Hadassah Medical Center is a large university hospital with a level I trauma center. Given the paucity of literature regarding indications, management, and complications of IVC filters, we performed this retrospective observational study of consecutive patients undergoing IVC filter insertion in our center.
Rottenstreich et al.
Results Filter placement
During the 5 yr of study, 405 patients underwent IVC filter insertion (Average: 81 filters/yr), with no significant increase during our study period. All attempts at filter insertion were successful. In the vast majority of patients (n = 388, 96%), the type of filter used was Celect (Cook Medical, Bloomington, IN, USA), an ‘optional’ filter, that may be removed or remain as a permanent implant depending of the clinical setting. In the remaining 17 patients (4%), retrievable filters of different types were inserted [ALN (Pyramed, Basingstoke Hampshire, UK), Gunther-Tulip (Cook Medical), Optease (Cordis, Fremont, CA, USA)]. All filters were placed by interventional radiologists at our medical center. The clinical services referring patients for IVC filter placement included medical/surgical intensive care unit (n = 154, 38%), neurosurgery (n = 62, 15%), internal medicine (n = 55, 13.5%), oncology (n = 48, 12%), general surgery (n = 38, 9%), orthopedics (n = 28, 7%), hematology (n = 8, 2%), and others (n = 12, 3%).
Patients and methods Patient characteristics Data collection
We conducted a retrospective review of medical records using the hospital electronic database of all patients undergoing an IVC filter insertion between January 1st, 2009 and December 31st, 2013 at a single tertiary care university hospital. Records were reviewed between June and October 2014. Data extracted included patient characteristics (age, sex, oncologic comorbidities), filter type, indication for IVC filter placement (contraindications to anticoagulation, failure of anticoagulation treatment, catheter-directed thrombolysis therapy, and prophylactically in trauma patients or peri-operatively), filter retrieval, use of anticoagulant therapy, immediate and long-term filter complications, and follow-up data. The data reviewers, three thrombosis specialists assessed the appropriateness of IVC filter placement and evaluated the appropriateness of referral for retrieval based on the FDA recommendation to remove filter as soon as mechanical protection from PE is no longer necessary (e.g. no contraindications for anticoagulants or no further apparent benefit when placed prophylactically) (14). Institutional review board approval waiving informed consent was obtained for this study. Statistical analysis
Comparison variables are described as means SD using 0.05 as a cutoff for significance. Significance between groups was assessed by the Student’s t-test and Fisher’s exact test.
2
A summary of patient’s demographics is shown in Table 1. The median age of this cohort was 57 22 yr (age range 13–103). A majority of male patients was observed with 234 (57.7%) males and 171 (42.2%) females. One hundred and thirty-nine (34.3%) patients suffered from major trauma. Among the non-trauma patients, there were 125 males (47%) and 141 women (53%), while in the trauma patients, the majority were males (n = 109, 78%) (P < 0.0001). One hundred and sixteen (28.6%) patients had active cancer. The group of patients with active cancer was significantly older (median age-63 13 yr), and those who suffered from major trauma were significantly younger (median age31 19 yr) (P < 0.0001). Indications for filter placement
Indications for IVC filter insertion (Table 2) included an acute VTE (PE or DVT) in a patient with a contraindication to full anticoagulant therapy, in 192 patients (47.4%). Twenty-three filters (5.7%) were placed prior to catheterdirected thrombolysis of proximal DVT. In 14 (3.5%) and 7 (1.7%) patients, IVC filter was used as an adjunct to full anticoagulation or due to failure of anticoagulation, respectively. Prophylactic IVC filter placement (e.g. high risk for, but no evidence of, acute VTE) was performed in 169 patients (42%): 139 patients (34.3%) who suffered from major trauma and 30 patients (7.4%) undergoing major surgery. Anticoagulation therapy was prescribed after filter insertion in 264 patients (65.2%) before discharge from hos-
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Rottenstreich et al.
IVC filters: indications, complications, and follow-up
Table 1 Patients’ demographics and clinical characteristics
Number of patients (%) Gender M/F (%) Age; median SD (range) Active cancer (%)
All patients
Trauma
405 234/171 57 22 116
139 109/39 31 19 3
(100) (58, 42) (13–103) (29)
(34) (78, 22) (15–88) (2)
Non-trauma patients
P value
266 125/141 64 18 113
ORIGINAL ARTICLE
Patterns of use and outcome of inferior vena cava filters in a tertiary care setting Amihai Rottenstreich1, Galia Spectre1, Batia Roth1, Allan I. Bloom2, Yosef Kalish1 1
Department of Hematology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel; 2Department of Radiology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel
Abstract Background: Inferior vena cava (IVC) filter placement is increasing although the evidence to justify their use is limited. Many filters are left in place indefinitely, thereby exposing patients to long-term complications. Objectives: To review indications, complications, and follow-up data of patients undergoing IVC filter placement at our center. Methods: A retrospective review of consecutive admitted patients who underwent IVC filter insertion in a large university hospital with a level I trauma center. Thrombosis specialists retrospectively assessed the appropriateness of indication for IVC filter placement as well as referral for retrieval. Results: Overall, 405 filters were inserted between 2009 and 2013. All filters were retrievable. IVC filter was placed as a primary prevention in 42% of patients. Fifty-two patients (12.8%) experienced at least one filter-related complication. The most common complication was deep vein thrombosis occurring in 6.9% of cases. Almost a third of all complications occurred in filters originally placed for prophylactic indications. Only 13.6% of filters were successfully removed. Nevertheless, according to our reviewers, an attempt for filter retrieval should have been made in 57% of all cases and in 86% of trauma patients. A significantly higher retrieval rate was found in patients followed at our thrombosis clinic (P < 0.01). During follow-up, 95 patients (23.4%) died, most of them with active cancer. Conclusions: IVC filters are placed in many cases for prophylactic indications. Their low retrieval rates together with relatively high risk of long-term complications, questions their extensive utilization. Prospective trials addressing the safety and efficacy of IVC filters are still warranted. Key words inferior vena cava filter; deep vein thrombosis; pulmonary embolism; venous thromboembolism Correspondence Yosef Kalish, MD, Department of Hematology, Hadassah-Hebrew University Medical Center, POB 12000, Jerusalem, Israel 91120. Tel: ++972 2 6779415; Fax ++972 2 6449580; e-mail: [email protected] Accepted for publication 4 March 2015
Venous thromboembolism (VTE) manifesting as deep vein thrombosis (DVT) and pulmonary embolism (PE) is a serious and potentially lethal disorder that affects 100 per 100 000 annually and results in significant morbidity and mortality (1). The standard of care is anticoagulation therapy, with inferior vena cava (IVC) filters being used as an adjunctive or alternative measure. The concept of surgical vena cava interruption was first introduced in 1865 by Trousseau (2). In 1967, surgical insertion of IVC filters was pioneered but still was infrequently used (3). Since then, the rate of IVC filter placement has increased significantly in the United States from 2000 patients in 1979 to 49 000 in 1999, with a threefold increase from 2001 to 2006 (4, 5). At least part of this dramatic rise is explained by the ease of
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
doi:10.1111/ejh.12542
percutaneous placement, pioneered in 1973 by Greenfield, with the subsequent introduction of retrievable filters in the 1990s (3, 4). Secondly and most importantly, the largest increase in use has been due to filter use in patients with no evidence of DVT or PE (i.e. prophylactic use), representing 21% of all cases (6). Although the use of IVC filters is rapidly increasing, the majority of evidence for their insertion, particularly for prophylactic indications, comes from observational studies which often lack a control group. Lack of high-quality data makes the decision to place an IVC filter controversial on many occasions. According to Spencer et al. (7), IVC filter insertion is appropriate in only 51% of cases. This is highlighted by the wide variability of indications for IVC filters insertions according to different
1
IVC filters: indications, complications, and follow-up
professional society guidelines. The latest principal guidelines were published by the American College of Chest Physicians (ACCP), American Society of Hematology (ASH) and the Society of Interventional Radiology (SIR) (8–11). While the ACCP and ASH advocate the use of IVC filters only when anticoagulation is contraindicated and refute their prophylactic use, the SIR guidelines are more permissive and have recommended a list of absolute, relative, and prophylactic indications. Moreover, while the retrieval rates of IVC filters are reported to be very low (2–20%), filters may be associated with significant long-term risks such as DVT (including IVC and filter thrombosis) IVC penetration, filter dislocation, and migration and rupture of IVC which raises concerns about their expanding utilization (12, 13). Hadassah Medical Center is a large university hospital with a level I trauma center. Given the paucity of literature regarding indications, management, and complications of IVC filters, we performed this retrospective observational study of consecutive patients undergoing IVC filter insertion in our center.
Rottenstreich et al.
Results Filter placement
During the 5 yr of study, 405 patients underwent IVC filter insertion (Average: 81 filters/yr), with no significant increase during our study period. All attempts at filter insertion were successful. In the vast majority of patients (n = 388, 96%), the type of filter used was Celect (Cook Medical, Bloomington, IN, USA), an ‘optional’ filter, that may be removed or remain as a permanent implant depending of the clinical setting. In the remaining 17 patients (4%), retrievable filters of different types were inserted [ALN (Pyramed, Basingstoke Hampshire, UK), Gunther-Tulip (Cook Medical), Optease (Cordis, Fremont, CA, USA)]. All filters were placed by interventional radiologists at our medical center. The clinical services referring patients for IVC filter placement included medical/surgical intensive care unit (n = 154, 38%), neurosurgery (n = 62, 15%), internal medicine (n = 55, 13.5%), oncology (n = 48, 12%), general surgery (n = 38, 9%), orthopedics (n = 28, 7%), hematology (n = 8, 2%), and others (n = 12, 3%).
Patients and methods Patient characteristics Data collection
We conducted a retrospective review of medical records using the hospital electronic database of all patients undergoing an IVC filter insertion between January 1st, 2009 and December 31st, 2013 at a single tertiary care university hospital. Records were reviewed between June and October 2014. Data extracted included patient characteristics (age, sex, oncologic comorbidities), filter type, indication for IVC filter placement (contraindications to anticoagulation, failure of anticoagulation treatment, catheter-directed thrombolysis therapy, and prophylactically in trauma patients or peri-operatively), filter retrieval, use of anticoagulant therapy, immediate and long-term filter complications, and follow-up data. The data reviewers, three thrombosis specialists assessed the appropriateness of IVC filter placement and evaluated the appropriateness of referral for retrieval based on the FDA recommendation to remove filter as soon as mechanical protection from PE is no longer necessary (e.g. no contraindications for anticoagulants or no further apparent benefit when placed prophylactically) (14). Institutional review board approval waiving informed consent was obtained for this study. Statistical analysis
Comparison variables are described as means SD using 0.05 as a cutoff for significance. Significance between groups was assessed by the Student’s t-test and Fisher’s exact test.
2
A summary of patient’s demographics is shown in Table 1. The median age of this cohort was 57 22 yr (age range 13–103). A majority of male patients was observed with 234 (57.7%) males and 171 (42.2%) females. One hundred and thirty-nine (34.3%) patients suffered from major trauma. Among the non-trauma patients, there were 125 males (47%) and 141 women (53%), while in the trauma patients, the majority were males (n = 109, 78%) (P < 0.0001). One hundred and sixteen (28.6%) patients had active cancer. The group of patients with active cancer was significantly older (median age-63 13 yr), and those who suffered from major trauma were significantly younger (median age31 19 yr) (P < 0.0001). Indications for filter placement
Indications for IVC filter insertion (Table 2) included an acute VTE (PE or DVT) in a patient with a contraindication to full anticoagulant therapy, in 192 patients (47.4%). Twenty-three filters (5.7%) were placed prior to catheterdirected thrombolysis of proximal DVT. In 14 (3.5%) and 7 (1.7%) patients, IVC filter was used as an adjunct to full anticoagulation or due to failure of anticoagulation, respectively. Prophylactic IVC filter placement (e.g. high risk for, but no evidence of, acute VTE) was performed in 169 patients (42%): 139 patients (34.3%) who suffered from major trauma and 30 patients (7.4%) undergoing major surgery. Anticoagulation therapy was prescribed after filter insertion in 264 patients (65.2%) before discharge from hos-
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Rottenstreich et al.
IVC filters: indications, complications, and follow-up
Table 1 Patients’ demographics and clinical characteristics
Number of patients (%) Gender M/F (%) Age; median SD (range) Active cancer (%)
All patients
Trauma
405 234/171 57 22 116
139 109/39 31 19 3
(100) (58, 42) (13–103) (29)
(34) (78, 22) (15–88) (2)
Non-trauma patients
P value
266 125/141 64 18 113
