• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

MAUDE Adverse Event Report: COOK INC UNKNOWN DTK FILTER, INTRAVASCULAR, CARDIOVASCULAR

  • Print
  • Share
  • E-mail
-
Super Search Devices@FDA
510(k) | DeNovo | Registration & Listing | Adverse Events | Recalls | PMA | HDE | Classification | Standards
CFR Title 21 | Radiation-Emitting Products | X-Ray Assembler | Medsun Reports | CLIA | TPLC
 

COOK INC UNKNOWN DTK FILTER, INTRAVASCULAR, CARDIOVASCULAR Back to Search Results
Catalog Number UNKNOWN
Device Problems Filtration Problem (2941); Appropriate Term/Code Not Available (3191)
Patient Problems Pulmonary Embolism (1498); Death (1802); No Code Available (3191)
Event Date 02/22/1998
Event Type  Death  
Manufacturer Narrative

Journal article: suprarenal vena caval filter placement: follow-up of four filter types in 22 patients, jvir 1998; vol. 9: pg. 588-593. (b)(4). The event is currently under investigation.

 
Event Description

As stated in the journal article "suprarenal vena caval filter placement: follow-up of four filter types in 22 patients, jvir 1998; vol. 9: pg. 588-593": purpose: to determine if suprarenal placement of inferior venaj cava (ivc) filters is associated with renal dysfunction or other complications. Conclusion:follow-up indicates suprarenal ivc filter placement is safe, and no evidence of permanent renal dysfunction after placement was found. Filter migration was the most frequent com­plication, but no clinical sequelae were noted with these patients. Inferior vena cava (ivc) filter placement has become routine interventional treatment in patients who have deep vein thrombosis or pulmonary embolus (pe) and who are unable to receive anticoagulation therapy or who have not re sponded to anticoagulation therapy. Normally, filters are placed caudal to the renal veins because of the theoretical risk of renal vein throm­bosis and subsequent renal dysfunc­tion should thrombosis of the ivc occur. The incidence of ivc throm­bosis associated with ivc filter placement has been difficult to establish. Reports of ivc patency for various filter types range from 50% to 100% (1-7). There are other factors that make us reluctant to place a filter above the renal veins. The suprarenal ivc is shorter and larger in diameter than the infrare­nal ivc and is often not as well de­fined on venography because of un­opacified flow from the renal veins. This causes concern for inappropri­ate positioning of the filter, which must be placed below the hepatic veins, but above the wide ostia of the renal veins. There is also a question of stability of the suprarenal filter because of proximity to the respiratory motion of the dia­phragm. Concerns for placement of a permanent filter are magnified when patients are young and healthy, as seen with pregnant patients. These factors have caused us to limit the indications for suprarenal filter placement to the following situations: (i) thrombus in the ivc to the renal veins or involving the renal veins, (ii) gonadal vein thrombosis, (iii) propagation of thrombus or pe with an infrarenal filter, and (iv) patients who are pregnant or who intend to become pregnant (based on the theory that the graviduterus would compress the infrarenal filter, and possibly injure the uterus or promote ivc thrombosis). Long-term follow-up is needed to determine the safety and effective­ ness of suprarenal ivc filter place­ment. There are few reports of su­prarenal ivc filter follow-up and, as would be predicted, the numbers of patients are small (8-10). We retrospectively analyzed our experi­ence with suprarenal ivc filter placement to add to this limited pa­tient population. We present the long-term follow-up in 22 patients with suprarenal ivc filters and specifically report evidence of renal dysfunction, ivc thrombosis, recur­ rent pe, filter migration, and ivc perforation. Materials and methods an ivc filter registry is main­tained at the university of arkan­sas for medical sciences for three institutions; university hospital, john l. Mcclellan memorial veter­an's administration hospital, and arkansas children's hospital. Registry information includes pertinent clinical information, filter place ment detail, and follow-up. All patients underwent inferior vena cavography prior to filter placement to determine the position of the renal veins, diameter of the ivc, the presence of thrombus, and identify venous anomalies. Filters were placed percutaneously via the right internal jugular vein or the right common femoral vein. Filter devices utilized in our patient population included: bird's nest type i (bn-i) (cook, bloomington, in), lg medi­cal filter type i (lgm-1) (b. Braun/ vena tech, (b)(4)), lgm type ii (lgm-ii), and the titanium greenfield filter modified hook (tgf-mh) (medi-tech/boston scien­tific, (b)(4)). After filter placement, another venogram was obtained, along with frontal and lateral radiographs of the abdomen. Copies of these examination films were maintained for each patient in our ivc filter follow-up. A regis­tered nurse practitioner managed the data entry, film file, and initi­ated patient follow-up. Patients were routinely scheduled for clinical and radiographic follow-up at 1 month and annually thereafter. At follow-up, the registered nurse prac­titioner interviewed the patient with standardized questions to de termine evidence of pe, ivc throm­bosis, or new deep vein thrombosis. If the patient described suspicious respiratory symptoms or chest pain, ventilation/perfusion scanning was performed. If there was new lower extremity edema, duplex ultra­ sonography was performed, but if the ivc was not adequately visualized a cavogram was obtained. Frontal and lateral radiographs of the abdomen were also obtained and examined by an interventional radiologist for change in filter posi­tion and device fracture. If ivc penetration was suspected, a computed tomographic (ct) scan was obtained. The current study was compiled by evaluating our registry data, fil­ter placement and follow-up radio­ graphs, hospital charts, clinical lab­ oratory files, and patient radiology files. Patient charts were examined for all 22 patients to determine whether signs or symptoms of renal vein thrombosis (ie, flank pain, hematuria) occurred in proximity to filter placement. Serum creatinine levels were obtained from the hospital charts and the clinical laboratory data. Filter migration and ivc penetration were determined from follow-up radiographs and ct ex­amination. Radiograph measurements were obtained with a standardized technique using the known leg/strut length as a reference. Radiographs were taken on inspiration. Filter migration was defined as movement greater than 2 em to account for variability with parallax and patient respiratory variability. Vena caval penetration was defined as a depth greater than 5 mm. These standards were recently pro­posed by cho et al (11) for use in evaluation of filters. Y analysis was utilized to compare the results of our previous report of infrarenal migrations to suprarenal migrations for determination of significance. Results since april 2, 1985, our interventional radiology service has placed 764 ivc filters. Twenty-two of these filters (2. 9%) were placed above the renal veins (12 in female patients and 10 in male patients). The type of filter implanted was determined by operator preference and filter availability. No vena cava measured greater than 28 mm, which would have dictated placement of a bn filter. The four filter types utilized were tgf-mh (n = 16), lgm-1 (n = 2), lgm-11 (n = 2), and bn (n =2). The mean follow-up in the 13 living patients was 50 months. Follow-up of the total 22 patients ranged from 1 to 137 months (mean follow-up, 36 months). No patient was lost to follow-up. Clinical situations that dictated suprarenal placement are noted in table 1. The most common reason for suprarenal placement was thrombus in the ivc near the level of renal veins that prevented appropriate infrarenal position, occurring in nine patients. Pregnancy, or the intent to become pregnant, was noted in four patients. Ovarian vein thrombosis was diagnosed with ct in one patient. Four patients had failure of the infrarenal filter. Two patients required a suprarenal filter after the infrarenal filter was poorly positioned and judged to be inadequate for protection from pe. One of these was a bn filter that extended into the iliac vein, another was a tgf-mh that developed a significant tilt. In two patients, the superior struts and filter wires of the bn fil­ters were inadvertently placed at or above the renal veins. These were classified as suprarenal filters. No patient developed symptoms of acute renal vein thrombosis after suprarenal ivc filter placement (ie, hematuria, flank pain). The creati­nine levels before and after filter placement were available in 11 patients. Only one patient demon­strated an increase in serum creati­nine at 4 days after filter place­ment. This patient was noted to have extensive ivc thrombosis to the renal veins prior to filter place ment. A renogram obtained 7 days after suprarenal ivc filter place ment showed normal findings and the patient's serum creatinine level returned to normal. Eleven patients had ivc throm­bus at the time of filter placement. No clinical evidence of new or increased ivc thrombosis was found in any of the 22 patients at follow­ up. Six patients reported lower extremity swelling at the time of rou­tine follow-up, but this finding had been noted in each prior to filter placement. The autopsy of one patient defined pe as the cause of death. This patient had progressive pe, diagnosed by sequential venti­lation/perfusion scans, despite anti­ coagulation and placement of a sec­ond, suprarenal filter. A cavogram at the time of suprarenal filter placement showed extensive ivc thrombus above the previously placed bn filter (fig 1). The pe seen at autopsy was thought to rep­ resent recurrent pe. This was the only patient of the 22 that demon­strated evidence of recurrent pe after suprarenal filter placement. Abdominal radiographs were ob­tained at follow-up in 18 patients. Table 2 lists the five migrations that occurred (27. 7%). This is significantly different (p <. 002) from the migra­tion rate noted with infrarenal filter placement, as defined by our previous report of 320 infrarenal filters (1). All migrations occurred in a caudal di­rection and included four tgf-mh migrations and one lgm type i filter migration (fig 2). Act scan obtained for an indication other than filter follow-up, depicted a tgf-mh penetration to a depth of 1. 8 em in a patient also noted to have filter mi­gration (fig 3). No clinical sequelae were noted with filter migration or ivc penetration. Our data will not show whether one filter type has a higher incidence of complications because of the small numbers of lgm and bn devices placed compared to thetgf-mh. Discussion the main concern with suprare­nal ivc filter placement is the theoretical risk of renal vein thrombosis or venous hypertension and subsequent renal dysfunction. There are few reports of suprarenal filter fol­low-up. Greenfield et al (8) reported on 71 patients that included the 12 patients previously reported by stewart et al(10), and orsini and jarrell reported follow-up of only 11 patients (9). Table 3 compares our results with the previous filter fol­low-up studies. There was no evidence of renal dysfunction in any of the 82 patients described in these reports (8-10). In our study, a sin­gle patient was found to have a transient increase in serum creati­nine level. This patient's serum creatinine and renogram were normal within a week. We found no other evidence of renal dysfunction, de­ spite the large number of patients with extensive ivc thrombosis (11 patients). A report by brenner et al in 1992, however, provides a note of caution conceming the risk of renal dysfunction (12). In brenner's study, six patients with renal cell carcinoma who had extensive ivc thrombus were treated with supra­ renal ivc filter placement prior to nephrectomy and thrombectomy. The mean follow-up in this study was only 12 months and only the postoperative serum creatinine levels were reported. There was some evidence of mild renal dysfunction with serum creatinine levels rang­ing from 1. 3 to 2. 1 mg/dl (four of the six patients had a serum creati­nine level greater than 1. 5 mg/dl). In this same report, brenner also describes performing unilateral nephrectomy and suprarenal ivc ligation in 10 dogs to investigate the risk to renal function. Six of the dogs had persistent loss of renal function and three of the six died of uremia. Brenner's study is in con­trast to a experiment reported by peyton et al (13), in which the ivc was ligated in 12 dogs who did not undergo nephrectomy. There was a transient renal dysfunction, but this quickly resolved in 11 of the 12 living dogs. These reports suggest that there may be a greater risk of renal dysfunction in patients with a sin­gle kidney. Hypothetically, there may be transient renal venous hypertension and even subclinical suprarenal ivc thrombosis in patients who receive a suprarenal filter, but there is either resolution of this thrombus or development of ade­quate collateral outflow veins. Logi­cally, the patient might not compen­ate as well with only one kidney. Despite the lack of evidence for permanent renal dysfunction in the current study or in the previous su­prarenal filter follow-up reports, these were small retrospective stud­ies that were not specifically designed to assess renal function. We depended on incidental biochemical measurements (obtained in only 11 patients), chart review, and clinical follow-up to assess for renal dysfunction. Other reports have the same or greater limitations in assessing renal dysfunction (8-10,12). Overall, however, there does not appear to be a significant risk of permanent renal dysfunction when placing an ivc filter above the re­nal veins. The report by brenner did indicate that the risk of renal injury may be greater in a patient with a single kidney. Our study found rates of recur­ rent pe (4. 5%), ivc thrombosis (0%), and penetration of the ivc by filter struts (4. 5%) similar to those reported previously with our infra­renal filter study (1) and green­ field's suprarenal filter follow-up (8). These complication rates are limited by reliance on clinical fol­low-up. To define the true incidence of pe and ivc thrombosis, expen­sive imaging and, sometimes, invasive examinations would have to be performed routinely in all patients. Because this is not possible, most filter follow-up studies, including this one, are limited by finding these complications only in patients who present with symptoms that initiate definitive imaging studies. This differs from filter migration and device fracture that can be readily determined from routinely obtained radiographs, which we had in 18 of 22 patients. In our current study, five of 18 filters (27. 7%) migrated 2 em or more, all inferiorly. This is signifi­cantly different (p <. 002) from the 6% infrarenal migration rate re­ported on follow-up of 320 filters (1). This finding is also different from greenfield's 3% migration rate in 60 suprarenal placements (8). Greenfield's report did not define the migration distance, but both were inferior migrations. Our filter migrations are detailed in table 2. Three of the five migrations produced a splayed or tilted appear­ance to the filters, apparently be­ cause of migration into the renal veins (figs 2, 3). No clinical sequelae associated with these migrations was seen. This 27. 7% migration rate raises concern that the ivc filter may be unstable in the suprarenal position. Greenfield has speculated that the mechanical forces of the diaphragm and liver in the suprarenal ivc may be greater than in the infrarenal ivc (8). An­ other factor may be the increased diameter of the ivc above the renal veins. Also, the device hooks associated with the tgf-mh and lgm filters are directed to impede superior migration and may possibly al­ low inferior migration. If mechani­cal forces are greater in the supra­ renal ivc, they did not cause a greater number of filter fractures when compared to infrarenal filters. A single filter (lgm-ii) fracture (5. 5%) was found in our study. This is not significantly higher than the 2% rate noted in the infrarenal data (1). Greenfield found two filter frac­tures in 60 suprarenal filter patients for a 3% fracture rate (8). The types of filters utilized in our study depended on operator preference. The tgf-mh was chosen most often by our interventional radiologists, not only because of fa­miliarity, but also because the deployment of this filter allows for precise positioning. Recommending a filter type based on the results of this study, however, would be inappropriate because of the small numbers of different filters. There are several considerations when choosing which filter to place above the renal veins. As with any ivc filter placement, familiarity with the de­vice is important. Precise position­ing, when placing a filter above the renal veins, is important to avoid the renal and hepatic veins. The height of the filter is also a consideration because the length of the suprarenal ivc is less than that of the infrarenal segment. The deployed heights of the devices currently available in the united states are as follows: tgf-mh =4. 7 em; stainless-steel greenfield, over-the-wire, alternating hook (ssg-ah) = 4. 4 em; stainless-steel greenfield = 4. 6 em; bn = > 5 em; lgm ii = 3. 8 em; and simon-niti­nol = 3. 8-4. 0 em. The current bn filter is difficult to place in the suprarenal position due to its length. The simon-nitinol device has a known tendency to 'jump" caudally, which is unfavorable. The lgm ii filter height is more compatible with this region of the cava, but compared with other filters, there may be a higher rate of ivc occlusion (4). The optimum filter choice for the suprarenal ivc would ap­pear to be the ssg-ah because of its shorter height and more exact positioning, as it is deployed over a guide wire. The alternating hooks might also prevent the problem with inferior migration. In conclusion, we found no evi­dence of permanent renal dysfunc­tion in follow-up of 22 patients re­ceiving suprarenal placement of ivc filters. Our review did find a significantly higher rate of filter migration compared to previous in­frarenal data, yet we found no clinical sequelae associated with these migrations. We continue to recommend placement of suprarenal ivc filters in a limited number of clini­cal situations, as outlined in the introduction. Further follow-up studies of ivc filters are needed to define the efficacy and safety of de­ vices placed above and below the renal veins. We recommend maintaining an ivc filter follow-up registry, such as ours, with additional renal function evaluation in patients who receive a suprarenal filter.

 
Manufacturer Narrative

Additional information: upon further review of the article, the autopsy of one of the patients noted a pulmonary embolism (pe) as the cause of death. The article further states: "this patient had progressive pe, diagnosed by sequential ventilation/perfusion scans, despite anti-coagulation and placement of a second, suprarenal filter. A cavogram at the time of suprarenal filter placement showed extensive ivc thrombus above the previously placed bn [birds nest] filter. The pe seen at autopsy was though to represent recurrent pe. This was the only patient of the 22 that demonstrated evidence of recurrent pe after suprarenal filter placement. ¿. A cavogram was obtained at a follow-up of a bn filter because of lower extremity edema. Extensive thrombus is seen in the ivc extending above the bn filter, requiring placement of another manufacturer's modified hook above the renal veins. Journal article: suprarenal vena caval filter placement: follow-up of four filter types in 22 patients, jvir 1998; vol. 9: pg. 588-593. (b)(4). The event is currently under investigation.

 
Manufacturer Narrative

A review of the complaint history, documentation, drawings, instructions for use (ifu), manufacturing instructions, quality control and specifications was conducted. The complaint device was not returned for investigation. There are no images within the article that are relevant to this particular event. The complaint was initiated based on a report contained in a journal article. The article was published in 1998 and is a retrospective analysis of patients treated with suprarenal ivc filters. The data was pulled from an ivc filter registry maintained at the (b)(6) for medical sciences and consisted of twenty-two patients, two of whom received the bird's nest type 1 filter. A clinical review of the article summarizes it as follows: it was reported that "three failures of the infrarenal filter. One filter was poorly positioned and judged to be inadequate for protection for protection from pe. In two patients, the superior struts and filter wires of the bn filters were inadvertently placed at or above the renal veins. One additional patient deceased, autopsy defined pe as the cause of death. This patient had progressive pe, diagnosed by sequential ventilation/perfusion scans, despite anti-coagulation and placement of a second, suprarenal filter. A cavogram at the time of the suprarenal filter placement showed extensive ivc thrombus above the previously placed bnf filter. The pe seen at autopsy was thought to represent recurrent pe. ". "in conclusion, we found no evidence of permanent renal dysfunction in follow-up of 22 patients receiving suprarenal placement of ivc filters. Our review did find a significantly higher rate of filter migration compared to previous in­frarenal data, yet we found no clinical sequelae associated with these migrations. We continue to recommend placement of suprarenal ivc filters in a limited number of clini­cal situations, as outlined in the introduction. Further follow-up studies of ivc filters are needed to define the efficacy and safety of devices placed above and below the renal veins. We recommend maintaining an ivc filter follow-up registry, such as ours, with additional renal function evaluation in patients who receive a suprarenal filter". The device is supplied with ifu which gives device description and intended use as well as contraindications, warnings, precautions and instructions for device placement and use. The ifu includes the following warning: ¿no technique will completely eliminate the possibility of recurrent pte. (with the bird¿s nest vena cava filter, the observed incidence of recurrent pte clinically acceptable. ¿ the ifu also includes a list of potential adverse events. Pulmonary embolism and death are listed among the potential adverse events. Without imaging, specific event details or product lot information we are unable to draw a conclusion about the cause of this event. Based on the minimal information available, a definitive root cause cannot be determined.

 
Search Alerts/Recalls

  New Search  |  Submit an Adverse Event Report

Brand NameUNKNOWN
Type of DeviceDTK FILTER, INTRAVASCULAR, CARDIOVASCULAR
Manufacturer (Section D)
COOK INC
750 daniels way
bloomington IN 47404
Manufacturer Contact
larry pool
750 daniels way
bloomington, IN 47404
8123392235
MDR Report Key5723881
MDR Text Key47362746
Report Number1820334-2016-00568
Device Sequence Number1
Product Code DTK
Combination Product (Y/N)N
Reporter Country CodeUS
PMA/PMN NumberP850049
Number of Events Reported1
Summary Report (Y/N)N
Report Source Manufacturer
Source Type LITERATURE
Reporter Occupation OTHER HEALTH CARE PROFESSIONAL
Type of Report Initial,Followup,Followup
Report Date 06/12/2017
1 Device Was Involved in the Event
1 Patient Was Involved in the Event
Date FDA Received06/14/2016
Is This An Adverse Event Report? Yes
Is This A Product Problem Report? No
Device Operator HEALTH PROFESSIONAL
Device Catalogue NumberUNKNOWN
Was Device Available For Evaluation? No
Is The Reporter A Health Professional? Yes
Was the Report Sent to FDA? No
Distributor Facility Aware Date02/22/1998
Event Location Hospital
Date Manufacturer Received04/05/2017
Was Device Evaluated By Manufacturer? No Answer Provided
Is The Device Single Use? Yes
Is this a Reprocessed and Reused Single-Use Device? No
Type of Device Usage Initial

Patient TREATMENT DATA
Date Received: 06/14/2016 Patient Sequence Number: 1
-
-