Deep Venous Thrombosis (DVT) in Trauma: a Literature Review

Vena Cava Filters

First Author	Year	Reference Title	Class	Conclusions
Carabasi RA III	1987	Complications encountered with the use of the Greenfield filter. Am J Surg 154:163-8	II	200 filters in 193 patients, unspecified long-term follow-up. Complications: venous anomalies 2.5%; insertion compilations 9.2%; Postoperative: 2.5% minor, 4.1% major; 0.5% mortality secondary to distal migration. Stress importance of preop venography, visualization of thrombus, marking of renal veins, diagnosing venous anomalies, knowing accurate size of vena cava.
Greenfield LJ	1988	Twelve-year clinical experience with the Greenfield vena cava filter. Surgery 104:706-12	III	Long-term follow-up of 469 patients with mean follow-up of 43 months (0.3-138) from 1974-1986. 81 filters placed for "extended" indications (17%), 40 trauma patients included in follow-up. 96% IVC patency, 98% filter patency rate, 4% misplacement rate, 3% recurrent PE rate.
Greenfield LJ	1992	Late results of suprarenal Greenfield vena cava filter placement. Arch Surg 127:969-73	III	Review of 71 patients who had suprarenal placement of Greenfield filter. 60 available for follow-up, mean=53 months (18 months-16 yrs). 24 deaths, none secondary to recurrent embolism or renal failure. Recurrent embolism rate was 4% which is identical to infrarenal experience. Duplex exam (n=22) showed all filters were patent. 16 patients (41%) had lower extremity edema that predated filter insertion. Filter fracture in 2 patients and distal migration in 2 patients with no clinical symptoms. Suprarenal placement of Greenfield filter is safe and effective for thrombus extending above renal veins and for pregnant patients or women of childbearing age.
Ferris EJ	1993	Percutaneous inferior vena cava filters: Follow-up of seven designs in 320 patients. Radiology 188:851-6	III	324 filters placed over 7 yrs. No placement-related mortality or morbidity. Average follow-up=404 days (1-2392). 19% caval thrombosis; 9% delayed penetration through IVC wall; 6% migration more than 1 cm, 2% fracture strut. Insertion site DVT was 2%. Long-term radiologic follow-up recommended for IVC filters.

Vena Cava Filters - "Traditional" Indications

First Author	Year	Reference Title	Class	Conclusions
Jarrell BE	1983	A new method of management using the Kim-Ray Greenfield filter for deep venous thrombosis and pulmonary embolism in spinal cord injury. Surg Gynecol Obstet 157:316-20	III	21 SCI patients with filter placed for "traditional" indications. 1 death secondary to PE in filter patients secondary to misplacement in right iliac vein. 2 thrombosed IVCs. Overall DVT rate in SCI population 62%. Emphasis on knowing exact location of DVT, anatomy of IVC, that filter must protect from all sources of emboli in lower extremity, and that there is a risk of thrombosis through large collateral vessels.
Leach TA	1994	Surgical prophylaxis for pulmonary embolism. Am Surg 16:292-5	II	205 vena cava filters placed for indications that were outlined prospective-ly, although many were inserted for "traditional" indications. No PEs in these filter patients and minimal insertion complications.
"Extended" Indications: Ortho
Golueke PJ	1988	Interruption of the vena cava by means of the Greenfield filter: Expanding the indications. Surgery 103:111-7	III	16 filters inserted prophylactically before joint replacement surgery in patients with history of VTE. 72 filters inserted for "traditional" indications. Mean follow-up=16.4 months (1-60) in 65 patients. Complications: 3% recurrent PE, 9% leg edema, 7.5% caval occlusion, 92.5% patency. No PEs in prophylactic group that received anti-platelet and sequential compression therapy. Indications should be extended for vena cava filter to help reduce preventable deaths secondary to PE.
Webb LX	1992	Greenfield filter prophylaxis of pulmonary embolism in patients undergoing surgery for acetabular fracture. J Orthop Trauma 6:139-45	II	Outlined predisposing factors for VTE. In patients undergoing acetabular fracture repair with 2 or more risk factors, prophylactic filter was placed (24/51). No insertion complications and no PEs. 4 patients had leg edema and 1 had phlegmasia. 27 patients did not receive preop filter; 2 PEs in this group, 1 fatal. All patients had SQ heparin and aspirin.
Rogers FB	1993	Prophylactic vena cava filter insertion in severely injured trauma patients: Indications and preliminary results. J Trauma 35:637-42	II	Prospective criteria for prophylactic filter insertion after retrospective review of trauma registry. Prophylactic filters placed in patients who could not receive anticoagulation and grouped:1) age>55 with long bone fracture; 2) severe closed head injury and coma; 3) multiple long bone fractures and pelvic fractures; 4) spinal cord injury. 34 patients had prophylactic filters placed. No PEs, 17.6% DVT rate. 30 day patency 100%, 1-year patency 89% (n=17).

Vena Cava Filters - "Extended" Indications: Trauma

First Author	Year	Reference Title	Class	Conclusions
Rogers FB	1995	Routine prophylactic vena cava filter insertion in severely injured trauma patients decreases the incidence of pulmonary embolism. J Am Coll Surg 180:641-7	II	Continued follow-up from J Trauma í93. 63 prophylactic vena cava filters placed in high risk patients as previously outlined. DVT rate:30%; 1 PE (fatal). No insertion complications, 3.5% insertion related thromboses. 30-day patency-100% (n=36), 1 year-96% (n=34), 2 year-96% (n=16).
Wilson JT	1994	Prophylactic vena cava filter insertion in patients with traumatic spinal cord injury: Preliminary results. Neurosurgery 35:234-9	II	Retrospective analysis of 111 SCI patients showed 7 PEs (6.3%) accounting for 31% of trauma PEs. 6 PEs occurred after patient discharge, mean time 78 days (9-5993). 15 prophylactic filters placed in SCI patients. No insertion problems or PEs. 30-day patency rate 100% (n=14), 1-year 82% (n=9).
Winchell RJ	1994	Risk factors associated with pulmonary embolism despite routine prophylaxis: Implications for improved protection. J Trauma 37:600-6	III	8-year retrospective registry review at Level 1 trauma center (9721 patients). Overall PE rate=37%. 29 prophylactic vena cava filters placed with no PEs or short-term complications. Average time to PE in this group was 14.5 days. High risk categories: head+spinal cord injury (4.5%); head+long bone fracture (8.8%); severe pelvis plus long bone fracture (12%); multiple long bone fracture (10%). Patients with estimated risk of PE, despite prophylaxis of > 2-5%, are reasonable candidates for prophylactic vena cava filter placement, especially if conventional measures cannot be used.
Rosenthal D	1994	Use of the Greenfield filter in patients with major trauma. Cardiovasc Surg 2:52-5	II	Control group 1984-88, 94 patients with 22 PEs (23%) and 5% fatal PE rate. 1988-92, after adoption of protocol to place prophylactic filters, 67 patients with only 1 PE and no fatal PEs. Minimal insertion morbidity. No long-term follow-up reported.
Zolfaghari D	1995	Expanded use of inferior vena cava filters in the trauma population. Surgery Annual 27:99-105	III	Retrospective analysis of 45 filters placed in 3005 patients. 38/45 had extend-ed indications for filter placement as they were placed for no DVT or in patients with DVT or PE but no contraindication to anticoagulation. No PEs after filter placement, and there was 1 death secondary to closed head injury.
Cipolle M	1995	Prophylactic vena caval filters reduce pulmonary embolism in trauma patients [Abstract]. Critical Care Medicine 23:A93.	III	Review of 43 high risk trauma patients who had vena cava filters placed, 16 for "traditional" indications and 27 for "extended" indications. 0 PEs in prophylactic group and 5 PEs in "traditional" indications group. Overall PE rate was 11.6%.

Vena Cava Filters - "Extended" Indications: Trauma

First Author	Year	Reference Title	Class	Conclusions
Patton JH Jr	1996	Prophylactic Greenfield filter: Acute complications and long-term follow-up. J Trauma 41:231-7	II	Follow-up of prophylactic filters placed between 1991-1994. 69 filters with 9% insertion rate. 15 patients died. 30 patients were located and 19 returned for follow-up evaluation (35%). Average follow-up was 770 days (246-1255). No caval thrombosis. 14 patients had chronic DVT. 11/14 had chronic venous insufficiency. No long-term caval thromboses. Not clear, however, if filter caused chronic venous insufficiency because there was no nonfilter group.
Rodriguez JL	1996	Early placement of prophylactic vena cava filters in injured patients at high risk for pulmonary embolism. J Trauma 40:797-804	II	40 vena cava filters placed in consecutive patients with 3 or more risk factors compared to 80 matched historic controls. 1 PE in VCF group, 14 PEs in non-VCF group. PE related mortality and overall mortality was the same in each group as was the incidence of DVT, 15% in VCF group and 19% in no VCF group.
Greenfield LJ	1996	Posttrauma thromboembolism prophylaxis. 8th Annual American Venous Forum	I	Pilot study for large, multicenter trial. 53 patients randomized to receive SCD, LMWH, or unfractionated heparin and 1/2 randomized to receive vena cava filter. Inclusion criteria were ISS>9 and VTE risk factor score developed by investigators. 26 patients got VCF. No complications of filter placement or evidence of vena caval occlusion. No PEs in either groups. 12 DVTs in nonfilter patients and 11 DVTs in filter patients.

Vena Cava Filters - "Extended" Indications: Medical/Surgical

First Author	Year	Reference Title	Class	Conclusions
Rohrer MJ	1989	Extended indications for placement of inferior vena cava filters. J Vasc Surg 10:44-50	III	264 filters placed in all types of patients. 66 placed prophylactically. "Extended" indications: 1) no documented DVT but high risk; 2) small PE would be fatal due to poor cardiopulmonary reserve; 3) large ileofemoral thrombus; 4) procedure in conjunction with venous thrombectomy; 5) thrombus above previously placed IVC filter. No deaths in either group. Prophylactic group had minimal morbidity. 3 PEs (4.5%) despite filter, 1 mortality, and 4.5% occlusion. Recommend liberalizing indications for insertion of Greenfield filter since they had an insertion mortality rate of 0% and fatal PE rate of 1.5% in high risk prophylactic group.