- Email: [email protected]
Sharp Recanalization of Central Venous Occlusions
Sharp Recanalization of Central Venous Occlusionsl Tony Farrell. MD Elvira V. Lang, MD William Barnhart, RT Index terms: Central venous access Veins, stenosis or obstruction. Veins, thrombosis
JVIR 1999; 10:149-154 Abbreviations: SVC = superior vena cava, UK = urokinase
PURPOSE: To describe a sharp puncture technique for recanalization of chronic central venous occlusions that could not be traversed by a guide wire. MATERIALS AND METHODS: Five patients presented with six longstanding central venous occlusions that could not be traversed with a guide wire after thrombolysis. The occlusions occurred following radiation for lung carcinoma (n = 2) and indwelling venous catheters (n = 4). The length of venous occlusion was determined by simultaneously advancing transbrachial and transfemoral catheters to the site of occlusion. Initially, a curved guiding catheter with a Rosch-Uchida needle and, in subsequent patients, a coaxial sheathed needle with a 21-gauge stylet were used for recanalization. The recanalized veins were then balloon dilated and stents were placed. RESULTS: With use of this technique, recanalization was successful in five of the six occlusions. One occlusion was too long to traverse safely in one patient. Two patients were asymptomatic 16-18 months after the recanalization. CONCLUSION: This new technique offers an effective alternative to surgery in the treatment of central venous occlusion.
From the Department of Radiology, University of Iowa College of Medicine, Iowa City, IA. From the 1997 SCVIR annual meeting. Received May 8, lgg8; revision requested June 16; revision received August 14; accepted August 18. Address correspondence to T.F., Department of Radiology, MC2026, University of Chicages 5841 S. *ve) Chicago, IL 60637.
o SCVIR, 1999
THERAPEUTIC options for central vein (axillary, subclavian, brachiocephalic or superior vena cava [SVCI) thrombosis include anticoagulation, thrombolysis, endovascular repair, and direct surgical intervention. Chronic central venous occlusion presents a difficult management problem especially when the occlusion cannot be traversed with a guide wire, a step which is essential for endovascular treatment such as balloon angioplasty and stent placement. Many patients with an underlying malignancy present at an advanced stage in their disease process, limiting the effectiveness of traditional therapy such as radiation and chemotherapy. Furthermore, surgical treatmerit of ~ e n t r avenous l ~cclusion~ such as first rib resection, claviculectomy, jugulo-subclavian or axillojugular vein bypass involves certain morbidity and is potentially disfig-
uring (1-5). We describe a less invasive technique in which the venous occlusion is traversed with a custom-made curved sheathed needle using simultaneous transbasilic and transfemoral venous accesses.
MATERIALS AND METHODS Twenty-one patients with symptomatic central venous (axillary, subclavian, or brachiocephalic vein or SVC) stenoses or occlusions were referred to our interventional radiology service for diagnosis and treatment over a 2-year period. Their symptoms included pain, swelling and ulceration. Patients with primary axillary-subclavian vein thrombosis (Paget-von Schrotter Syndrome) were excluded from this study. Upper limb venography via a peripherally placed intravenous catheter was performed in all
149
150
Sharp Recanalization of Central Venous Occlusions
February 1999 JVIR
Figure 1. Venogram showing occlu-
sion of the right subclavian vein. patients to assess the site and extent of the venous obstruction (Fig 1). Subsequently, a 6-F vascular sheath was placed in the brachial1 basilic vein and a combination of a 5-F catheter and both the tapered and the stiff ends of a Glidewire (manufactured by Terumo; distributed by Medi-tecrnoston Scientific, Natick, MA) were used to traverse the occlusion. If thrombus was present and if the lesion was successfully traversed with a guide wire, urokinase (UK) (Abbott, North Chicago, IL) was given, through a 5-F multisidehole infusion catheter. If the lesion could not be traversed with a guide wire, then the infusion catheter was advanced just distal to the lesion. UK was infused at a rate of 90,000 U/h for a minimum of 15 hours at which time venography was repeated to assess if thrombolysis was successful. Heparin was given via the brachial sheath to keep the partial thromboplastin time between 60 and 80 seconds after a loading dose of 100 U/kg. Thrombolysis was continued if venographic improvement was apparent. If there was no improvement after thrombolysis and the lesion could not be traversed with a guide wire, an 80-cm 8-F vascular sheath (Cook, Bloomington, IN) was
placed in the common femoral vein and a 5-F catheter and Glidewire combination was used to traverse the occlusion. Using this approach, we were able to traverse the venous occlusion and perform balloon angioplasty or stent placement in all but six lesions in five patients. These latter patients form the basis for this study. Of the five patients, four were women (Table). Their ages ranged from 45 to 73 years (mean, 64 years). Four patients had prior subclavian venous access and two patients had a history of lung carcinoma, which was treated with radiation. In these five patients the length and orientation of the venous occlusion were assessed in multiple projections using simultaneously placed transbrachial and transfemoral catheters in preparation for sharp recanalization (Fig 2).
Sharp Recanalization Technique Under fluoroscopic guidance, the sheathed needle device was introduced via a transfemoral catheter and was advanced in a cephalic direction toward the tip of the transbrachially placed wirelcatheter (Fig 3a, 3b). Once the occlusion was traversed with the needle and blood aspirated after removal of the stylet, contrast material was injected to confirm its intraluminal location. A 200-cm 0.018-inch Flex T guide wire (Mallinckrodt Medical, St. Louis, MO.) was advanced distally through the needle and followed by the 5-F Teflon catheter. Both wire and needle were replaced with an exchange length Amplatz stiff or Rosen guide wire (Cook), the cephalic end of which was secured by a transbrachially placed 6-F 5-mm snare (Microvena, White Bear Lake, MN) giving through-andthrough access. The tract was dilated with an 8-mm x 4-cm BlueMax angioplasty balloon (Medi-tech/ Boston Scientific) and a Wallstent (Schneider, Minneapolis, MN) deployed (Table). One patient required placement of an overlapping Palmaz 298 stent (Johnson & John-
son Interventional Systems, Warren, NJ) in the lower SVC when the Wallstent did not expand fully due to a previously unrecognized stenosis. Repeated venography confirmed vessel patency and the absence of extravasation (Fig 4). All patients subsequently received long-term warfarin therapy. Follow-up was by patient contact and chart review.
Sharp Puncture Recanalization Devices Two types of recanalization device were used. Initially, we used an 0.038-inch trocar-stylet within a 62-cm 5-F sheath from a RoschUchida transjugular liver access set (Cook) to traverse an occlusion from a brachialhasilic vein approach in two patients. Subsequently, a custom-made sheathed needle system (Cook) consisting of a 90-cm 21gauge stylet needle, within an 82-cm 5-F tapered Teflon catheter and a 75-cm 7-F Teflon catheter was used from a femoral approach (Fig 5).
Five occlusions in five patients were successfully traversed and stents were placed with use of this technique. In one patient (patient 3), the length (5 cm) and direction of the left-sided occlusion made accurate needle placement difficult and contributed to our reluctance to proceed with recanalization in view of the risk of injuring adjacent structures such as the subclavian artery and lung. This patient had already gained some symptomatic relief after thrombolysis and because of this improvement and the relatively poor directability of our system, we believed that the risks of needle recanalization outweighed the benefits. There were no immediate complications related to thrombolysis or the recanalization technique; specifically there was no evidence of extravasation as determined by repeated venography, stable vital signs, and complete blood cell count after the procedure. Symptomatic improvement oc-
Farrell et a1
151
Volume 10 Number 2
Patient Data
Patient No./Sex/ Age (y)
Diagnosis
Prior Duration of Venous Symptoms Catheterization (wk)
Veins Occluded
Length of Occlusion* (cm)
Rt Sub
3
Rt Ax/Sub/Bc
2
Lung Ca
-
3
Lt AxlSuWBc
3
3/F/60
Lung Ca
Rt Sub
10
Bilateral SubBc
4N45
Breast Ca
Rt Sub
16
Rt BcfSvc
2
5/F/50
Porphyria
Lt Sub
14
Lt SubBc
4
W73
Renal failure
2/F/56
Note.-Ax = axillary, Bc W = Wallstent. * After thrombolysis.
=
brachiocephalic, Ca
=
cancer, Sub
=
250,000 U of UK and balloon dilation, and one of these patients (patient l)required placement of a n additional 10 x 20-mm Wallstent. Three patients died from progression of their underlying disease a t 2, 3, and 5 months, respectively. The two surviving patients were asymptomatic 16-18 months after the recanalization.
I DISCUSSION Figure 2. Simultaneous transfemoral
and transbrachial venous catheters at either end of a subclavian venous occlusion.
curred in all patients within 24 hours of the procedure. Two patients had symptom recurrence due to stent thrombosis 8 and 11 weeks after the procedure, respectively, despite systemic administration of warfarin (Table).Both patients had successful recanalization following pulse-spray thrombolysis of
Although successful thrombolysis, balloon angioplasty, and stent placement for central venous stenoses and occlusions are widely reported in the literature, to our knowledge, no one has previously described a sharp puncture recanalization technique in patients in whom the occlusion could not be traversed with a guide wire. Several published series reviewing the endovascular treatment of central venous occlusions describe a small number of patients in whom the occlusion could not be traversed
Follow-up
W 10 x 68 mm Reoccluded at 8 weeks; successful thrombolysis/stent placement with a 10 X 20 mm Wallstent; patent on venography at 11months; symptom free at 18 months. W 10 X 68 mm Reoccluded at 11 weeks; successful thrombolysis; died at 5 months; no symptom recurrence. W 10 X 42 mm Died at 2 months; no symptom recurrence. W 10 x 42 mm, Died at 3 months; no P298 symptom recurrence. W 10 x 68 mm Patent on duplex at 9 months; symptom free at 16 months.
215
subclavian, Svc
Stent
=
superior vena cava, P
=
Palmaz stent,
with a guide wire even after thrombolysis (6-8). Such technical failures present a difficult management problem as both radiation and chemotherapy are limited by dose, and improvement following these treatments is not immediate in advanced cases. Symptomatic relief of the distressing sequelae of central venous occlusion is the goal in treating these patients, many of whom have a limited life expectancy. In addition, restoration of vascular patency is advantageous in those with limited venous access. As described by Campbell et a1 and Martin et al, upper limb venography often fails to define the proximal extent of obstruction as contrast material is diluted by flow in collaterals and washout from the internal jugular and opposite brachiocephalic veins (9,lO). We measured the length of occlusion more accurately by the simultaneous placement of transbrachial and transfemoral venous catheters. Our rationale for performing catheter-directed thrombolysis was
152
Sharp Recanalization of Central Venous Occlusions
February 1999 JVIR
a.
b.
C.
Figure 3. (a-c) Sharp recanalization technique with a transfernorally placed coaxial needle system, using a transbrachially
placed guide wire as a target. twofold. First, to recanalize the occluded vein and second to minimize the risk of pulmonary embolism in the event of successful recanalization. Campbell et a1 reported two deaths due to pulmonary embolism in a group of 22 patients treated for central venous stenoses while anticoagulated with heparin (9). None of our patients developed clinically significant pulmonary emboli after recanalization. Our standard practice was to continue UK infusion until either complete thrombolysis occurred or no further thrombolysis was noted on venography. We cannot predict if further thrombolysis would have made the occlusion easier to traverse with a guide wire instead of resorting to recanalization with the sharp puncture method. To increase the success of thrombolysis, we opted to infuse UK via a multisidehole catheter placed a s closely as possible to the occlusion rather than give the thrombolytic agent through a peripherally
placed intravenous catheter. Even catheter-directed thrombolysis is sometimes limited in its effectiveness, a s there is preferential flow via collaterals bypassing the venous occlusion. The dosage of UK (90,000 U/h) was dictated by our hospital protocol, a s patients receiving a higher dosage intravenously would have required admission to the intensive care unit for monitoring. We also opted to infuse UK for 15-18 hours (usually overnight) initially rather than giving it in a pulse-spray fashion. Sharp needle recanalization in our first two patients was performed with a modified RoschUchida transjugular liver set using a high basilic vein approach because the shortness of the needle prevented a femoral vein approach. We subsequently used a custommade coaxial system composed of a 21-gauge needle and tapered Teflon catheters, which were long enough
for use via a transfemoral venous access. The sheathed needle is flexible enough to follow the Teflon catheter around such angles a s the SVChrachiocephalic vein junction. Putting a curve on the distal needle prior to its introduction may facilitate its passage. Also withdrawing the trocar portion of the needle a centimeter so it lies inside the obturator tip helps advance i t within the Teflon catheter. Our only technical failure was in a patient with a long (5 cm) left brachiocephalic occlusion in whom we had successfully recanalized and placed a stent for a shorter right subclavian occlusion. The length and direction of the left-sided occlusion made accurate needle placement difficult and contributed to our reluctance to proceed with recanalization since this patient had already improved after thrombolysis. Two patients underwent computed tomography of the thorax, a s part of their work up for lung can-
Farrell et a1
153
Volume 10 Number 2
Figure 4. Venogram following thrombolysis, sharp puncture, balloon dilation, and stent placement showing a patent right subclavian and brachiocephalic veins.
cer. One patient had a right suprahilar mass, which compressed the ipsilateral brachiocephalic vein. No mass lesion was seen in the second patient, and it is thought that the occlusion was mainly due to radiation. Although no significant extravasation occurred after sharp recanalization, such a complication could have a serious consequence with hemorrhage into the thoracic cavity leading to exsanguination. For this reason, we recommend passage of the recanalization needle in a cephalic rather than a caudal direction because of the lower venous pressure below the occlusion. If extravasation were to occur, prompt inflation of an angioplasty balloon or placement of a coated stent across the tract should prove effective in controlling the hemorrhage. Ferral et a1 describe a technique of accessing occluded veins with use of a transfemorally placed loop snare as a target at which a 21gauge needle is directed percutaneously (11).Once correct placement of the needle within the snare is confirmed, an 0.018inch guide wire is passed through the snare and
Figure 5. Coaxial covered needle system. A 21-gauge needle is within a 5-F tapered Teflon catheter. The 7-F Teflon catheter is not shown.
through-and-through access is obtained to facilitate the placement of central venous catheters. However, one of their patients developed moderate bleeding at the insertion site requiring vessel ligation due to the presence of extensive collaterals at the site of insertion. Although not essential, placement of a through-and-through guide wire facilitates subsequent balloon angioplasty and stent placement in the presence of a highgrade or long stenosis. A simultaneous cubito-femoral approach with a through-and-through guide wire has been described to deploy Wallstents in four patients with central venous occlusions and stenoses (12). As an alternative to the cubito-femoral route, Dondelinger et a1 recommend a simultaneous right and left axillary vein approach for lesions not involving the SVC (13). Thrombolysis followed by balloon dilation alone was associated with a
poor outcome in a series of patients with subclavian thrombosis reported by Begyui et al (14). We advocate balloon dilation to facilitate stent placement particularly for stents requiring a large introducing sheath. Several series report high technical success rates and good longterm patency following deployment of metallic stents including the Wallstent, the Gianturco Z stent, and the Palmaz stent for venous stenoses and occlusions (6,15-19). We placed Wallstents because of their flexibility and their relatively small delivery system (7-8 F). We considered the optimal Wallstent diameter to range from 6-10 mm in size for subclavian and brachiocephalic veins to 14-16 mm for the svc. In one patient, we also deployed a Palmaz 298 stent below a 10- x 68-mm Wallstent to treat a short SVC stenosis, which had prevented complete opening of the Wallstent despite attempted maximal dilation with an angioplasty balloon. Kee et a1 reported a similar experience following the placement of kissing Wallstents for stenotic brachiocephalic veins and SVC, as the lower ends of these stents were kept open after Palmaz stents were deployed (8).
Although recommended by most authors after placement of intravascular stents for venous stenoses and occlusions, anticoagulation with warfarin does not guarantee patency. Both patients in our study who developed stent occlusion were therapeutically anticoagulated with warfarin. One of these patients developed an interval stenosis adjacent to the existing stent, underlining the importance of surveillance preferably using noninvasive techniques such as Doppler ultrasonography. Both Vesely et a1 and Mickley et a1 describe good long-term patency rates following the use of Wallstents for venous stenoses in hemodialysis patients, and they emphasize the need for close clinical surveillance and multiple re-interventions to maintain patency (20,211. Patients with primary axillary-
154
Sharp Recanalization of Central Venous Occlusions
February 1999 JVIR
subclavian vein thrombosis (Pagetvon Schrotter disease) were not included in this study because these patients are generally younger and in good health, and the long-term results of stent placement in this population are unknown. Furthermore, there is evidence that these patients respond well to catheterdirected thrombolysis followed by anticoagulation and first rib resection (22,23). We realize that a limitation of our study was that patients were not randomized to receive sharp recanalization, as the technique was performed only if less invasive methods of recanalization had failed. We believe that this new technique is a minimally invasive option in the management of central venous occlusion and it results in prompt relief of symptoms and is associated with a low morbidity. References 1. Bhatia DS, Money SR, Ochsner JL, et al. Comparison of surgical bypass and percutaneous balloon dilatation with primary stent placement in the treatment of central venous obstruction in the dialysis patient: one-year follow-up. Ann Vasc Surg 1996; 10:452-455. 2. Doty DB, Doty JR, Jones KW. Bypass of superior vena cava. Fifteen years' experience with spiral vein graft for obstruction of superior vena cava caused by benign disease. J Thorac Cardiovasc Surg 1990; 99: 889-995; discussion 895-896. 3. Meier GH, Pollak JS, Rosenblatt M, Dickey KW, Gusberg RJ. Initial experience with venous stents in exertional axillary-subclavian vein thrombosis. J Vasc Surg 1996; 24: 974-981; discussion 981-983. 4. Sanders RJ, Cooper MA. Surgical management of subclavian vein obstruction, including six cases of subclavian vein bypass. Surgery 1995; 118:856-863.
5. Wisselink W, Money SR, Becker MO, et al. Comparison of operative reconstruction and percutaneous balloon dilatation for central venous obstruction. Am J Surg 1993; 166: 200-204; discussion 204-205. 6. Stock KW, Jacob AL, Proske M, Bolliger CT, Rochlitz C. Steinbrich treatment of malignant obstruction of the superior vena cava with the self-expanding Wallstent. Thorax 1995; 50:1151-1156. 7. Crowe MT, Davies CH, Gaines PA. Percutaneous management of superior vena cava occlusions. Cardiovasc I n t e n Radiol 1995; 18:367372. 8. Kee ST, Kinoshita L, Razavi MK, Nyman UR, Semba CP, Dake MD. Superior vena cava syndrome: treatment with catheter-directed thrombolysis and endovascular stent placement. Radiology 1998; 206:187-193. 9. Campbell CB, Chandler JG, Tegtmeyer CJ, Bernstein EF. Axillary, subclavian, and brachiocephalic vein obstruction. Surgery 1977; 82:816826. 10. Martin EC, Koser M, Gordon DH. Venography in axillary-subclavian vein thrombosis. Cardiovasc Radiol 1979; 2:261-266. 11. Ferral H, Bjarnason H, Wholey M, Lopera J , Maynar M, CastanedaZuniga W. Recanalization of occluded veins to provide access for central catheter placement. JVIR 1996; 7~681-685. 12. Link J, Brossmann J, Muller-Hulsbeck S, Heller M. Venous stent application with a simultaneous cubitofemoral approach. [German] ROFO 1995; 163:81-83. 13. Dondelinger RF, Gofette P, Kurdziel JC, Roche A. Expandable metal stents for stenoses of the vena cava and large veins. Semin Interv Radiol 1991; 8:252-263. 14. Beygui RE, Olcott C, Dalman RL. Subclavian vein thrombosis: outcome analysis based on etiology and modality of treatment. Ann Vasc Surg 1997; 11:247-255.
15. Dyet JF, Nicholson AA, Cook AM. The use of the Wallstent endovascular prosthesis in the treatment of malignant obstruction of the superior vena cava. Clin Radiol 1993; 48:381-385. 16. Hennequin LM, Fade 0 , Fays JG, et al. Superior vena cava stent placement: results with the Wallstent endoprosthesis. Radiology 1995; 196:353-361. 17. Rosch J , Bedell JE, Putnam J, Antonovic R, Uchida B. Gianturco expandable wire stents in the treatment of superior vena cava syndrome recurring after maximumtolerance radiation. Cancer 1987; 60:1243-1246. 18. Gaines PA, Belli AM, Anderson PB, McBride K, Hemingway AP. Superior vena caval obstruction managed by the Gianturco Z stent. Clin Radiol 1994; 49:202-206; discussion 207-208. 19. Elson JD, Becker GJ, Wholey MH, Ehrman KO. Vena caval and central venous stenoses: management with Palmaz balloon-expandable intraluminal stents. JVIR 1991; 2:215-223. 20. Vesely TM, Hovsepian DM, Pilgram TK, Coyne DW, Shenoy S. Upper extremity central venous obstruction in hemodialysis patients: treatment with Wallstents. Radiology 1997; 204:343-348. 21. Mickley V, Gorich J , Rilinger N, Storck M, Abendroth D. Stenting of central venous stenoses in hemodialysis patients: long-term results. Kidney Int 1997; 51:277-280. 22. Machleder HI. Thrombolytic therapy and surgery for primary axillosubclavian vein thrombosis: current approach. Semin Vasc Surg 1996; 9:46-49. 23. Sheeran SR, Hallisey MJ, Murphy TP, Faberman RS, Sherman S. Local thrombolytic therapy as part of a multidisciplinary approach to acute axillosubclavian vein thrombosis (Paget-von Schroetter syndrome). JVIR 1997; 8:253-260.
JVIR 1999; 10:149-154 Abbreviations: SVC = superior vena cava, UK = urokinase
PURPOSE: To describe a sharp puncture technique for recanalization of chronic central venous occlusions that could not be traversed by a guide wire. MATERIALS AND METHODS: Five patients presented with six longstanding central venous occlusions that could not be traversed with a guide wire after thrombolysis. The occlusions occurred following radiation for lung carcinoma (n = 2) and indwelling venous catheters (n = 4). The length of venous occlusion was determined by simultaneously advancing transbrachial and transfemoral catheters to the site of occlusion. Initially, a curved guiding catheter with a Rosch-Uchida needle and, in subsequent patients, a coaxial sheathed needle with a 21-gauge stylet were used for recanalization. The recanalized veins were then balloon dilated and stents were placed. RESULTS: With use of this technique, recanalization was successful in five of the six occlusions. One occlusion was too long to traverse safely in one patient. Two patients were asymptomatic 16-18 months after the recanalization. CONCLUSION: This new technique offers an effective alternative to surgery in the treatment of central venous occlusion.
From the Department of Radiology, University of Iowa College of Medicine, Iowa City, IA. From the 1997 SCVIR annual meeting. Received May 8, lgg8; revision requested June 16; revision received August 14; accepted August 18. Address correspondence to T.F., Department of Radiology, MC2026, University of Chicages 5841 S. *ve) Chicago, IL 60637.
o SCVIR, 1999
THERAPEUTIC options for central vein (axillary, subclavian, brachiocephalic or superior vena cava [SVCI) thrombosis include anticoagulation, thrombolysis, endovascular repair, and direct surgical intervention. Chronic central venous occlusion presents a difficult management problem especially when the occlusion cannot be traversed with a guide wire, a step which is essential for endovascular treatment such as balloon angioplasty and stent placement. Many patients with an underlying malignancy present at an advanced stage in their disease process, limiting the effectiveness of traditional therapy such as radiation and chemotherapy. Furthermore, surgical treatmerit of ~ e n t r avenous l ~cclusion~ such as first rib resection, claviculectomy, jugulo-subclavian or axillojugular vein bypass involves certain morbidity and is potentially disfig-
uring (1-5). We describe a less invasive technique in which the venous occlusion is traversed with a custom-made curved sheathed needle using simultaneous transbasilic and transfemoral venous accesses.
MATERIALS AND METHODS Twenty-one patients with symptomatic central venous (axillary, subclavian, or brachiocephalic vein or SVC) stenoses or occlusions were referred to our interventional radiology service for diagnosis and treatment over a 2-year period. Their symptoms included pain, swelling and ulceration. Patients with primary axillary-subclavian vein thrombosis (Paget-von Schrotter Syndrome) were excluded from this study. Upper limb venography via a peripherally placed intravenous catheter was performed in all
149
150
Sharp Recanalization of Central Venous Occlusions
February 1999 JVIR
Figure 1. Venogram showing occlu-
sion of the right subclavian vein. patients to assess the site and extent of the venous obstruction (Fig 1). Subsequently, a 6-F vascular sheath was placed in the brachial1 basilic vein and a combination of a 5-F catheter and both the tapered and the stiff ends of a Glidewire (manufactured by Terumo; distributed by Medi-tecrnoston Scientific, Natick, MA) were used to traverse the occlusion. If thrombus was present and if the lesion was successfully traversed with a guide wire, urokinase (UK) (Abbott, North Chicago, IL) was given, through a 5-F multisidehole infusion catheter. If the lesion could not be traversed with a guide wire, then the infusion catheter was advanced just distal to the lesion. UK was infused at a rate of 90,000 U/h for a minimum of 15 hours at which time venography was repeated to assess if thrombolysis was successful. Heparin was given via the brachial sheath to keep the partial thromboplastin time between 60 and 80 seconds after a loading dose of 100 U/kg. Thrombolysis was continued if venographic improvement was apparent. If there was no improvement after thrombolysis and the lesion could not be traversed with a guide wire, an 80-cm 8-F vascular sheath (Cook, Bloomington, IN) was
placed in the common femoral vein and a 5-F catheter and Glidewire combination was used to traverse the occlusion. Using this approach, we were able to traverse the venous occlusion and perform balloon angioplasty or stent placement in all but six lesions in five patients. These latter patients form the basis for this study. Of the five patients, four were women (Table). Their ages ranged from 45 to 73 years (mean, 64 years). Four patients had prior subclavian venous access and two patients had a history of lung carcinoma, which was treated with radiation. In these five patients the length and orientation of the venous occlusion were assessed in multiple projections using simultaneously placed transbrachial and transfemoral catheters in preparation for sharp recanalization (Fig 2).
Sharp Recanalization Technique Under fluoroscopic guidance, the sheathed needle device was introduced via a transfemoral catheter and was advanced in a cephalic direction toward the tip of the transbrachially placed wirelcatheter (Fig 3a, 3b). Once the occlusion was traversed with the needle and blood aspirated after removal of the stylet, contrast material was injected to confirm its intraluminal location. A 200-cm 0.018-inch Flex T guide wire (Mallinckrodt Medical, St. Louis, MO.) was advanced distally through the needle and followed by the 5-F Teflon catheter. Both wire and needle were replaced with an exchange length Amplatz stiff or Rosen guide wire (Cook), the cephalic end of which was secured by a transbrachially placed 6-F 5-mm snare (Microvena, White Bear Lake, MN) giving through-andthrough access. The tract was dilated with an 8-mm x 4-cm BlueMax angioplasty balloon (Medi-tech/ Boston Scientific) and a Wallstent (Schneider, Minneapolis, MN) deployed (Table). One patient required placement of an overlapping Palmaz 298 stent (Johnson & John-
son Interventional Systems, Warren, NJ) in the lower SVC when the Wallstent did not expand fully due to a previously unrecognized stenosis. Repeated venography confirmed vessel patency and the absence of extravasation (Fig 4). All patients subsequently received long-term warfarin therapy. Follow-up was by patient contact and chart review.
Sharp Puncture Recanalization Devices Two types of recanalization device were used. Initially, we used an 0.038-inch trocar-stylet within a 62-cm 5-F sheath from a RoschUchida transjugular liver access set (Cook) to traverse an occlusion from a brachialhasilic vein approach in two patients. Subsequently, a custom-made sheathed needle system (Cook) consisting of a 90-cm 21gauge stylet needle, within an 82-cm 5-F tapered Teflon catheter and a 75-cm 7-F Teflon catheter was used from a femoral approach (Fig 5).
Five occlusions in five patients were successfully traversed and stents were placed with use of this technique. In one patient (patient 3), the length (5 cm) and direction of the left-sided occlusion made accurate needle placement difficult and contributed to our reluctance to proceed with recanalization in view of the risk of injuring adjacent structures such as the subclavian artery and lung. This patient had already gained some symptomatic relief after thrombolysis and because of this improvement and the relatively poor directability of our system, we believed that the risks of needle recanalization outweighed the benefits. There were no immediate complications related to thrombolysis or the recanalization technique; specifically there was no evidence of extravasation as determined by repeated venography, stable vital signs, and complete blood cell count after the procedure. Symptomatic improvement oc-
Farrell et a1
151
Volume 10 Number 2
Patient Data
Patient No./Sex/ Age (y)
Diagnosis
Prior Duration of Venous Symptoms Catheterization (wk)
Veins Occluded
Length of Occlusion* (cm)
Rt Sub
3
Rt Ax/Sub/Bc
2
Lung Ca
-
3
Lt AxlSuWBc
3
3/F/60
Lung Ca
Rt Sub
10
Bilateral SubBc
4N45
Breast Ca
Rt Sub
16
Rt BcfSvc
2
5/F/50
Porphyria
Lt Sub
14
Lt SubBc
4
W73
Renal failure
2/F/56
Note.-Ax = axillary, Bc W = Wallstent. * After thrombolysis.
=
brachiocephalic, Ca
=
cancer, Sub
=
250,000 U of UK and balloon dilation, and one of these patients (patient l)required placement of a n additional 10 x 20-mm Wallstent. Three patients died from progression of their underlying disease a t 2, 3, and 5 months, respectively. The two surviving patients were asymptomatic 16-18 months after the recanalization.
I DISCUSSION Figure 2. Simultaneous transfemoral
and transbrachial venous catheters at either end of a subclavian venous occlusion.
curred in all patients within 24 hours of the procedure. Two patients had symptom recurrence due to stent thrombosis 8 and 11 weeks after the procedure, respectively, despite systemic administration of warfarin (Table).Both patients had successful recanalization following pulse-spray thrombolysis of
Although successful thrombolysis, balloon angioplasty, and stent placement for central venous stenoses and occlusions are widely reported in the literature, to our knowledge, no one has previously described a sharp puncture recanalization technique in patients in whom the occlusion could not be traversed with a guide wire. Several published series reviewing the endovascular treatment of central venous occlusions describe a small number of patients in whom the occlusion could not be traversed
Follow-up
W 10 x 68 mm Reoccluded at 8 weeks; successful thrombolysis/stent placement with a 10 X 20 mm Wallstent; patent on venography at 11months; symptom free at 18 months. W 10 X 68 mm Reoccluded at 11 weeks; successful thrombolysis; died at 5 months; no symptom recurrence. W 10 X 42 mm Died at 2 months; no symptom recurrence. W 10 x 42 mm, Died at 3 months; no P298 symptom recurrence. W 10 x 68 mm Patent on duplex at 9 months; symptom free at 16 months.
215
subclavian, Svc
Stent
=
superior vena cava, P
=
Palmaz stent,
with a guide wire even after thrombolysis (6-8). Such technical failures present a difficult management problem as both radiation and chemotherapy are limited by dose, and improvement following these treatments is not immediate in advanced cases. Symptomatic relief of the distressing sequelae of central venous occlusion is the goal in treating these patients, many of whom have a limited life expectancy. In addition, restoration of vascular patency is advantageous in those with limited venous access. As described by Campbell et a1 and Martin et al, upper limb venography often fails to define the proximal extent of obstruction as contrast material is diluted by flow in collaterals and washout from the internal jugular and opposite brachiocephalic veins (9,lO). We measured the length of occlusion more accurately by the simultaneous placement of transbrachial and transfemoral venous catheters. Our rationale for performing catheter-directed thrombolysis was
152
Sharp Recanalization of Central Venous Occlusions
February 1999 JVIR
a.
b.
C.
Figure 3. (a-c) Sharp recanalization technique with a transfernorally placed coaxial needle system, using a transbrachially
placed guide wire as a target. twofold. First, to recanalize the occluded vein and second to minimize the risk of pulmonary embolism in the event of successful recanalization. Campbell et a1 reported two deaths due to pulmonary embolism in a group of 22 patients treated for central venous stenoses while anticoagulated with heparin (9). None of our patients developed clinically significant pulmonary emboli after recanalization. Our standard practice was to continue UK infusion until either complete thrombolysis occurred or no further thrombolysis was noted on venography. We cannot predict if further thrombolysis would have made the occlusion easier to traverse with a guide wire instead of resorting to recanalization with the sharp puncture method. To increase the success of thrombolysis, we opted to infuse UK via a multisidehole catheter placed a s closely as possible to the occlusion rather than give the thrombolytic agent through a peripherally
placed intravenous catheter. Even catheter-directed thrombolysis is sometimes limited in its effectiveness, a s there is preferential flow via collaterals bypassing the venous occlusion. The dosage of UK (90,000 U/h) was dictated by our hospital protocol, a s patients receiving a higher dosage intravenously would have required admission to the intensive care unit for monitoring. We also opted to infuse UK for 15-18 hours (usually overnight) initially rather than giving it in a pulse-spray fashion. Sharp needle recanalization in our first two patients was performed with a modified RoschUchida transjugular liver set using a high basilic vein approach because the shortness of the needle prevented a femoral vein approach. We subsequently used a custommade coaxial system composed of a 21-gauge needle and tapered Teflon catheters, which were long enough
for use via a transfemoral venous access. The sheathed needle is flexible enough to follow the Teflon catheter around such angles a s the SVChrachiocephalic vein junction. Putting a curve on the distal needle prior to its introduction may facilitate its passage. Also withdrawing the trocar portion of the needle a centimeter so it lies inside the obturator tip helps advance i t within the Teflon catheter. Our only technical failure was in a patient with a long (5 cm) left brachiocephalic occlusion in whom we had successfully recanalized and placed a stent for a shorter right subclavian occlusion. The length and direction of the left-sided occlusion made accurate needle placement difficult and contributed to our reluctance to proceed with recanalization since this patient had already improved after thrombolysis. Two patients underwent computed tomography of the thorax, a s part of their work up for lung can-
Farrell et a1
153
Volume 10 Number 2
Figure 4. Venogram following thrombolysis, sharp puncture, balloon dilation, and stent placement showing a patent right subclavian and brachiocephalic veins.
cer. One patient had a right suprahilar mass, which compressed the ipsilateral brachiocephalic vein. No mass lesion was seen in the second patient, and it is thought that the occlusion was mainly due to radiation. Although no significant extravasation occurred after sharp recanalization, such a complication could have a serious consequence with hemorrhage into the thoracic cavity leading to exsanguination. For this reason, we recommend passage of the recanalization needle in a cephalic rather than a caudal direction because of the lower venous pressure below the occlusion. If extravasation were to occur, prompt inflation of an angioplasty balloon or placement of a coated stent across the tract should prove effective in controlling the hemorrhage. Ferral et a1 describe a technique of accessing occluded veins with use of a transfemorally placed loop snare as a target at which a 21gauge needle is directed percutaneously (11).Once correct placement of the needle within the snare is confirmed, an 0.018inch guide wire is passed through the snare and
Figure 5. Coaxial covered needle system. A 21-gauge needle is within a 5-F tapered Teflon catheter. The 7-F Teflon catheter is not shown.
through-and-through access is obtained to facilitate the placement of central venous catheters. However, one of their patients developed moderate bleeding at the insertion site requiring vessel ligation due to the presence of extensive collaterals at the site of insertion. Although not essential, placement of a through-and-through guide wire facilitates subsequent balloon angioplasty and stent placement in the presence of a highgrade or long stenosis. A simultaneous cubito-femoral approach with a through-and-through guide wire has been described to deploy Wallstents in four patients with central venous occlusions and stenoses (12). As an alternative to the cubito-femoral route, Dondelinger et a1 recommend a simultaneous right and left axillary vein approach for lesions not involving the SVC (13). Thrombolysis followed by balloon dilation alone was associated with a
poor outcome in a series of patients with subclavian thrombosis reported by Begyui et al (14). We advocate balloon dilation to facilitate stent placement particularly for stents requiring a large introducing sheath. Several series report high technical success rates and good longterm patency following deployment of metallic stents including the Wallstent, the Gianturco Z stent, and the Palmaz stent for venous stenoses and occlusions (6,15-19). We placed Wallstents because of their flexibility and their relatively small delivery system (7-8 F). We considered the optimal Wallstent diameter to range from 6-10 mm in size for subclavian and brachiocephalic veins to 14-16 mm for the svc. In one patient, we also deployed a Palmaz 298 stent below a 10- x 68-mm Wallstent to treat a short SVC stenosis, which had prevented complete opening of the Wallstent despite attempted maximal dilation with an angioplasty balloon. Kee et a1 reported a similar experience following the placement of kissing Wallstents for stenotic brachiocephalic veins and SVC, as the lower ends of these stents were kept open after Palmaz stents were deployed (8).
Although recommended by most authors after placement of intravascular stents for venous stenoses and occlusions, anticoagulation with warfarin does not guarantee patency. Both patients in our study who developed stent occlusion were therapeutically anticoagulated with warfarin. One of these patients developed an interval stenosis adjacent to the existing stent, underlining the importance of surveillance preferably using noninvasive techniques such as Doppler ultrasonography. Both Vesely et a1 and Mickley et a1 describe good long-term patency rates following the use of Wallstents for venous stenoses in hemodialysis patients, and they emphasize the need for close clinical surveillance and multiple re-interventions to maintain patency (20,211. Patients with primary axillary-
154
Sharp Recanalization of Central Venous Occlusions
February 1999 JVIR
subclavian vein thrombosis (Pagetvon Schrotter disease) were not included in this study because these patients are generally younger and in good health, and the long-term results of stent placement in this population are unknown. Furthermore, there is evidence that these patients respond well to catheterdirected thrombolysis followed by anticoagulation and first rib resection (22,23). We realize that a limitation of our study was that patients were not randomized to receive sharp recanalization, as the technique was performed only if less invasive methods of recanalization had failed. We believe that this new technique is a minimally invasive option in the management of central venous occlusion and it results in prompt relief of symptoms and is associated with a low morbidity. References 1. Bhatia DS, Money SR, Ochsner JL, et al. Comparison of surgical bypass and percutaneous balloon dilatation with primary stent placement in the treatment of central venous obstruction in the dialysis patient: one-year follow-up. Ann Vasc Surg 1996; 10:452-455. 2. Doty DB, Doty JR, Jones KW. Bypass of superior vena cava. Fifteen years' experience with spiral vein graft for obstruction of superior vena cava caused by benign disease. J Thorac Cardiovasc Surg 1990; 99: 889-995; discussion 895-896. 3. Meier GH, Pollak JS, Rosenblatt M, Dickey KW, Gusberg RJ. Initial experience with venous stents in exertional axillary-subclavian vein thrombosis. J Vasc Surg 1996; 24: 974-981; discussion 981-983. 4. Sanders RJ, Cooper MA. Surgical management of subclavian vein obstruction, including six cases of subclavian vein bypass. Surgery 1995; 118:856-863.
5. Wisselink W, Money SR, Becker MO, et al. Comparison of operative reconstruction and percutaneous balloon dilatation for central venous obstruction. Am J Surg 1993; 166: 200-204; discussion 204-205. 6. Stock KW, Jacob AL, Proske M, Bolliger CT, Rochlitz C. Steinbrich treatment of malignant obstruction of the superior vena cava with the self-expanding Wallstent. Thorax 1995; 50:1151-1156. 7. Crowe MT, Davies CH, Gaines PA. Percutaneous management of superior vena cava occlusions. Cardiovasc I n t e n Radiol 1995; 18:367372. 8. Kee ST, Kinoshita L, Razavi MK, Nyman UR, Semba CP, Dake MD. Superior vena cava syndrome: treatment with catheter-directed thrombolysis and endovascular stent placement. Radiology 1998; 206:187-193. 9. Campbell CB, Chandler JG, Tegtmeyer CJ, Bernstein EF. Axillary, subclavian, and brachiocephalic vein obstruction. Surgery 1977; 82:816826. 10. Martin EC, Koser M, Gordon DH. Venography in axillary-subclavian vein thrombosis. Cardiovasc Radiol 1979; 2:261-266. 11. Ferral H, Bjarnason H, Wholey M, Lopera J , Maynar M, CastanedaZuniga W. Recanalization of occluded veins to provide access for central catheter placement. JVIR 1996; 7~681-685. 12. Link J, Brossmann J, Muller-Hulsbeck S, Heller M. Venous stent application with a simultaneous cubitofemoral approach. [German] ROFO 1995; 163:81-83. 13. Dondelinger RF, Gofette P, Kurdziel JC, Roche A. Expandable metal stents for stenoses of the vena cava and large veins. Semin Interv Radiol 1991; 8:252-263. 14. Beygui RE, Olcott C, Dalman RL. Subclavian vein thrombosis: outcome analysis based on etiology and modality of treatment. Ann Vasc Surg 1997; 11:247-255.
15. Dyet JF, Nicholson AA, Cook AM. The use of the Wallstent endovascular prosthesis in the treatment of malignant obstruction of the superior vena cava. Clin Radiol 1993; 48:381-385. 16. Hennequin LM, Fade 0 , Fays JG, et al. Superior vena cava stent placement: results with the Wallstent endoprosthesis. Radiology 1995; 196:353-361. 17. Rosch J , Bedell JE, Putnam J, Antonovic R, Uchida B. Gianturco expandable wire stents in the treatment of superior vena cava syndrome recurring after maximumtolerance radiation. Cancer 1987; 60:1243-1246. 18. Gaines PA, Belli AM, Anderson PB, McBride K, Hemingway AP. Superior vena caval obstruction managed by the Gianturco Z stent. Clin Radiol 1994; 49:202-206; discussion 207-208. 19. Elson JD, Becker GJ, Wholey MH, Ehrman KO. Vena caval and central venous stenoses: management with Palmaz balloon-expandable intraluminal stents. JVIR 1991; 2:215-223. 20. Vesely TM, Hovsepian DM, Pilgram TK, Coyne DW, Shenoy S. Upper extremity central venous obstruction in hemodialysis patients: treatment with Wallstents. Radiology 1997; 204:343-348. 21. Mickley V, Gorich J , Rilinger N, Storck M, Abendroth D. Stenting of central venous stenoses in hemodialysis patients: long-term results. Kidney Int 1997; 51:277-280. 22. Machleder HI. Thrombolytic therapy and surgery for primary axillosubclavian vein thrombosis: current approach. Semin Vasc Surg 1996; 9:46-49. 23. Sheeran SR, Hallisey MJ, Murphy TP, Faberman RS, Sherman S. Local thrombolytic therapy as part of a multidisciplinary approach to acute axillosubclavian vein thrombosis (Paget-von Schroetter syndrome). JVIR 1997; 8:253-260.