Eur I Vasc Surg 4, 483489 (1990)
Long-term Results of Venous Thrombectomy Combined with a Temporary Arterio-venous Fistula* Gunnar Plate, Henrik Akesson, Eibert Einarsson 2, Per Ohlin 1 and Bo Ekl~f a Department of Surgery and 1Department of Clinical Physiology, Central Hospital, Helsingborg, Sweden, 2Department of Surgery, University of Lund, Sweden and 3Department of Surgery, Faculty of Medicine, University of Kuwait Forty-one patients with acute iliofemoral venous thrombosis were randomised to conventional anticoagulation or acute thrombectomy combined with a temporary arterio-venousflstula (A VF) and anti-coagulation. Follow-up after 5 years in 22 medical and 19 surgical patients revealed slightly more asymptomatic patients (37 vs. 18%) and lessfrequent severe post-thrombotic sequelae (16 vs. 27%) in the surgical group (N.s.). The iliacvein was morefrequently (P < 0.05) normalfollowing thrombectomy (71 vs. 30%) as demonstrated by radionuclide angiography, but occlusion plethysmography showed an outflow capacity (61 vs. 45 ml/min/ 100 ml) that was not signi~cantly better. There was no obvious difference in muscle pump function (EWrel) and reflux (Q/EVrel) assessed by foot volumetry. Still, the ambulatory venous pressure was significantly (P < 0.05) lower in the surgical group. There was a tendency towards better results following thrombectomy in patients with fresh thrombosis and a successful initial procedure. Although the numbers of observations in many cases were too small to provide statistical evidence of benefit with venous thrombectomy + A VF, this procedure seems to improve the long-term outcome following acute iliofemoral venous thrombosis. Since the difference in outcome is not very striking, anticoagulation treatment is still an acceptable alternative. Key Words: Iliofemoral venous thrombosis; Venous thrombectomy; Arterio-venous fistula; Post-thrombotic sequelae; Venous physiology.
Introduction Acute iliofemoral v e n o u s t h r o m b o s e s are followed b y acute complications as well as chronic sequelae. 1, 2 To o v e r c o m e these problems, most textbooks advocate systemic anticoagulation treatment, which usually controls p u l m o n a r y embolism and is frequently followed b y a rapid decrease of the leg swelling. 1' 3-s Still, the majority of patients with iliofemoral t h r o m boses will develop late post-thrombotic sequelae especially if the femoro-popliteal a n d crural veins are involved, w h i c h usually is the case. 6~9 These postthrombotic sequelae are due to chronic v e n o u s h y p e r tension caused by persistent d e e p v e n o u s obstruction a n d / o r reflux. 6,10-12 Surgical t h r o m b e c t o m y is most effective for removal of d e e p thrombi, especially in the iliofemoral * Preliminary results presented at the 18th World Congress of the International Society for Cardiovascular Surgery, Sydney, September 1987. Please address all correspondence to: Gunnar Plate, Department of Surgery, Central Hospital, S-25187 Helsingborg, Sweden 0950-821X/90/050483+07 $03.00/0 © 1990 Grune & Stratton Ltd.
veins with a t h r o m b u s duration of less than 5 - 7 days. 13-15 Still, rethrombosis has b e e n frequent 14' 16 resulting in chronic v e n o u s obstruction, valvular i n c o m p e t e n c e and late sequelae in 70-80% of the operated patients. 17'18 A l t h o u g h this has caused m a n y surgeons to a b a n d o n v e n o u s t h r o m b e c t o m y , further d e v e l o p m e n t of the surgical technique has p r o c e e d e d in some institutions especially in Europe. Initial vein clearance has been i m p r o v e d and the incidence of rethrombosis has b e e n decreased b y the use of the Fogarty catheter, 14 radiographic 19 or angioscopic is assurance of vein patency, a n d b y adding postoperative regional infusion of h e p a r i n 14"2o, 21 or thrombolytic agents. 22"23 We have d e m o n s t r a t e d that a flowincreasing arterio-venous fistula (AVF) constructed at the groin level is most effective in p r e v e n t i n g rethrombosis of the iliofemoral s e g m e n t providing c o n t i n u e d iliac vein p a t e n c y in 61-76% a n d valvular c o m p e t e n c e in 50% of operated cases as well as excellent clinical results at 6 m o n t h s postoperatively. < 24 Equally good results have b e e n achieved by others using similar surgical procedures. 18, 25-27
484
G. Plate et aL
Rutherford 28 has recently re-emphasised the definite role of thrombectomy in the management of acute iliofemoral venous thrombosis. His conclusions were to a large extent based on our early clinical and phlebographical results. 4 The present report describes the clinical, morphological and physiological results recorded at 5 years after the acute thrombotic event.
Table 1. Patient and thrombus characteristics in 41 patients with acute iliofemoral venous thrombosis available for follow-up at 5 years. Thrombus duration was based on the presence of leg swelling prior to institution of therapy. Extension of thrombus was defined as proximal if not involving femoro-popliteal or crural veins (impossible to evaluate in eight patients)
Medical group Surgicalgroup (n ~ 22) (n = 19) Patient
Material
From 1979 to 1982, 58 patients with unilateral acute iliofemoral venous thrombosis were included in a prospective randomised study comparing surgical thrombectomy supported by a temporary arteriovenous fistula with conventional anticoagulation treatment. The criteria for inclusion, exclusion, patient characteristics, and details of medical and surgical treatment as well as the early (6 months) clinical and phlebographic results have been described in previous reports. 4's Basically, patients with deep venous thrombosis with a presumed age of less than 7 days and a proximal extension above the inguinal ligament, but not into the vena cava, were included. Medical treatment consisted of continuous heparin infusion for 5-7 days and oral anticoagulation for at least 6 months. Venous thrombectomy was performed under general anaesthesia and fluoroscopic control using a venous Fogarty balloon catheter inserted via the common femoral vein. Distal clearance was aided by manual compression. The arterio-venous fistula was constructed by end-to-side anastomosis of the cranial portion of the divided long saphenous vein to the superficial femoral artery. The fistula was closed after 4-6 weeks. All surgical patients received anticoagulants in a similar w a y as the medical patients. Forty-one patients (22 medical and 19 surgical) were available for follow-up at 5 years after the initial thrombotic event. Sixteen patients had expired and one patient's leg was initially amputated due to progressive phlegmasia cerulea dolens with venous gangrene. When comparing the initial patient and thrombus characteristics in the remaining patients, no obvious difference between the two treatment groups was noted, although there were slightly more male patients and right-sided thromboses in the medical group (Table 1). Five medical and four surgical patients had a previous history of venous ulceration and/or DVT restricted to the calf. It should also be noted that the presence of thrombus in the iliac veins (criterion for inclusion) had been clearly verified during the operation in all surgical patients, whereas this was not clearly demonstrated in six (27%) medical Eur J VascSurg Vol4, October1990
Age (years)
59 (15-79)
55 (18-80)
Sex (males/females)
9/13
5/14
Side (right/left)
6/16
3/16
Extension (proximal/entire leg/ unknown)
5/15/2
2/11/6
Duration (days)
2 (1-7)
2 (0-5)
Thrombus
patients in w h o m puncture of the femoral vein had failed. These patients were included due to presumed presence of iliofemoral thrombosis, since no filling of the deep veins occurred with ascending phlebography from the foot. Surgical clearance of the iliac vein was mostly successful in all cases, but eight patients (42 %) had an underlying iliac vein obstruction probably caused by compression of the overriding common iliac artery, iliac compression syndrome, 6 which made catheter passage to the vena cava difficult or impossible.
Methods
All 41 remaining patients were asked to return for clinical, morphological, and physiological evaluation at 5 years after the acute thrombotic event. Morphological evaluation consisted of radionuclide angiography, whereas venous physiology was assessed by venous plethysmography, foot volumetry and examination of foot vein pressures. All data are given as mean (range). Since a normal distribution was not proven, the two-tailed Mann-Whitney U test and Fisher exact probability tests were used for statistical analysis of quantitative and qualitative data, respectively. Presence of post-thrombic sequelae (leg swelling, ulceration, heaviness, venous claudication, varicose veins) was recorded in all cases. Venous claudication was defined as pain in the thigh or calf during exercise verified on tread mill testing and without signs of arterial occlusive disease. The clinical symptoms were graded according to the "Reporting standards in venous disease" by the A d Hoc Committee: Class 0 =
Long-term Results of Venous Thrombectomy
asymptomatic; 1 = mild chronic venous insufficiency (CVI); 2 = moderate CVI; 3 = severe CVI. Radionuclide angiography was performed as described by Akesson et al.3° to determine the patency of the iliac veins, which in each case were classified as normal, stenotic or occluded. Assessment of the 5year patency was supported by information obtained by radionuclide angiograms obtained at 6 months, 1 and 3 years and by contrast phlebography performed at 6 months. Occlusion plethysmography was performed to evaluate venous outflow capacity using strain-gauge technique as described by Hallb66k and G6thlin. 31 An abnormal outflow capacity was defined as having a maximum venous outflow (MVO) < 33 ml/min/100 ml or a k-value 20s), mild (15-20s), moderate (5-15s), or severe
( 33ml/min/ 100ml and k > 0.21s-i), no reflux assessed by foot volumetry (Q/EVrel < 2 sD) and/or foot vein pressure (90% refill time > 20 s), normal muscle p u m p (EVrel > --2 SD) and (if tested) no ambulatory venous hypertension (AVP -< 45 mmHg).
Results
The presence of post-thrombotic sequelae in all 41 (22 medical and 19 surgical) patients are presented in Fig. 1. There was no significant difference in the incidence of separate late post-thrombotic sequelae. There was no significant difference in the clinical results (Fig. 2), but severe chronic venous insufficiency (Class 3) tended to be more common (27 v s . 11%) and asymptomatic patients (Class 0) tended to be less common (18 v s . 37%) in the medical than in the surgical group. In subgroups of medical and surgical patients, respectively, the proportion of asymptomatic individuals in patients aged 2 SD. Foot vein pressures are demonstrated in Table 2. CurJ VascSurgVol4, October1990
The ambulatory venous pressure was significantly (P < 0.05) lower in the surgical group of patients. Nine (50%) of the medical and two (20%) of the surgical patients had an AVP above 60mmHg (N.S.). Two (11%) of the medical and five (50%) of the surgical patients had an AVP ~ 4 5 m m H g (0.05 < P < 0.1). Three (17%) of the medical and five (50%) of the surgical patients had a pressure reduction exceeding 50% during ambulation (N.S.). Venous reflux (90% refill time) was normal in one (6%) medical and four (40%) surgical patients, mild or moderate in 11 (61%) medical and five (50 %) surgical patients and severe in six (33 % ) medical and one (10%) surgical patient (N.S.). The overall physiological results in the 19 medical and 14 surgical patients with sufficient tests performed are demonstrated in Fig. 4. All tests were normal in two (11%) medical and in five (36%) surgical patients (N.S.). Comparing subgroups of medical and
Long-term Results of Venous Thrombectomy
487
Table 2. Mean values (ranges) for physiological tests performed 5 years after an acute iliofemoral venous thrombosis. Since one patient in each group had an EVrel = 0, the Q/EVr~I could only be calculated for 18 and 13 patients, respectively. The M a n n - W h i t n e y U test was used for the statistical analyses
Treatment Medical
Surgical
P-value Mann-Whitney U test
Occlusion plethysmography MVO (ml/min/100 ml)
n = 20 45 (12-87)
n = 17 61 (24-161)
N.S.
Foot volumetry EVre~(ml/100ml/min) Q/EVrel (min- 1)
n = 19 0.9 (0.0-1.8) 4.7 (0.5-18.3)
n = 14 1.0 (0.0-1.9) 3.3 (1.1-7.4)
N.s. N.s.
Foot vein pressure Pressure reduction (%) Ambulatory venous pressure (mmHg) 90% refill time (s)
n = 18 36 (-6-78) 60 (21-95) 13 (0-71)
n = 10 52 (26-81) 43 (18-70) 14 (4-38)
P < 0.05
MEDICAL GROUP
N.S.
N.S.
SURGICAL G R O U P
[] [] •
Normal Obstruction Reflux Obstruction + reflux
Fig. 4. Physiological results in 19 medical and 14 surgical patients at 5-year follow-up after an acute iliofemoral venous thrombosis.
surgical patients, respectively, the p r o p o r t i o n w i t h n o r m a l p h y s i o l o g y in patients a g e d
Long-term Results of Venous Thrombectomy Combined with a Temporary Arterio-venous Fistula* Gunnar Plate, Henrik Akesson, Eibert Einarsson 2, Per Ohlin 1 and Bo Ekl~f a Department of Surgery and 1Department of Clinical Physiology, Central Hospital, Helsingborg, Sweden, 2Department of Surgery, University of Lund, Sweden and 3Department of Surgery, Faculty of Medicine, University of Kuwait Forty-one patients with acute iliofemoral venous thrombosis were randomised to conventional anticoagulation or acute thrombectomy combined with a temporary arterio-venousflstula (A VF) and anti-coagulation. Follow-up after 5 years in 22 medical and 19 surgical patients revealed slightly more asymptomatic patients (37 vs. 18%) and lessfrequent severe post-thrombotic sequelae (16 vs. 27%) in the surgical group (N.s.). The iliacvein was morefrequently (P < 0.05) normalfollowing thrombectomy (71 vs. 30%) as demonstrated by radionuclide angiography, but occlusion plethysmography showed an outflow capacity (61 vs. 45 ml/min/ 100 ml) that was not signi~cantly better. There was no obvious difference in muscle pump function (EWrel) and reflux (Q/EVrel) assessed by foot volumetry. Still, the ambulatory venous pressure was significantly (P < 0.05) lower in the surgical group. There was a tendency towards better results following thrombectomy in patients with fresh thrombosis and a successful initial procedure. Although the numbers of observations in many cases were too small to provide statistical evidence of benefit with venous thrombectomy + A VF, this procedure seems to improve the long-term outcome following acute iliofemoral venous thrombosis. Since the difference in outcome is not very striking, anticoagulation treatment is still an acceptable alternative. Key Words: Iliofemoral venous thrombosis; Venous thrombectomy; Arterio-venous fistula; Post-thrombotic sequelae; Venous physiology.
Introduction Acute iliofemoral v e n o u s t h r o m b o s e s are followed b y acute complications as well as chronic sequelae. 1, 2 To o v e r c o m e these problems, most textbooks advocate systemic anticoagulation treatment, which usually controls p u l m o n a r y embolism and is frequently followed b y a rapid decrease of the leg swelling. 1' 3-s Still, the majority of patients with iliofemoral t h r o m boses will develop late post-thrombotic sequelae especially if the femoro-popliteal a n d crural veins are involved, w h i c h usually is the case. 6~9 These postthrombotic sequelae are due to chronic v e n o u s h y p e r tension caused by persistent d e e p v e n o u s obstruction a n d / o r reflux. 6,10-12 Surgical t h r o m b e c t o m y is most effective for removal of d e e p thrombi, especially in the iliofemoral * Preliminary results presented at the 18th World Congress of the International Society for Cardiovascular Surgery, Sydney, September 1987. Please address all correspondence to: Gunnar Plate, Department of Surgery, Central Hospital, S-25187 Helsingborg, Sweden 0950-821X/90/050483+07 $03.00/0 © 1990 Grune & Stratton Ltd.
veins with a t h r o m b u s duration of less than 5 - 7 days. 13-15 Still, rethrombosis has b e e n frequent 14' 16 resulting in chronic v e n o u s obstruction, valvular i n c o m p e t e n c e and late sequelae in 70-80% of the operated patients. 17'18 A l t h o u g h this has caused m a n y surgeons to a b a n d o n v e n o u s t h r o m b e c t o m y , further d e v e l o p m e n t of the surgical technique has p r o c e e d e d in some institutions especially in Europe. Initial vein clearance has been i m p r o v e d and the incidence of rethrombosis has b e e n decreased b y the use of the Fogarty catheter, 14 radiographic 19 or angioscopic is assurance of vein patency, a n d b y adding postoperative regional infusion of h e p a r i n 14"2o, 21 or thrombolytic agents. 22"23 We have d e m o n s t r a t e d that a flowincreasing arterio-venous fistula (AVF) constructed at the groin level is most effective in p r e v e n t i n g rethrombosis of the iliofemoral s e g m e n t providing c o n t i n u e d iliac vein p a t e n c y in 61-76% a n d valvular c o m p e t e n c e in 50% of operated cases as well as excellent clinical results at 6 m o n t h s postoperatively. < 24 Equally good results have b e e n achieved by others using similar surgical procedures. 18, 25-27
484
G. Plate et aL
Rutherford 28 has recently re-emphasised the definite role of thrombectomy in the management of acute iliofemoral venous thrombosis. His conclusions were to a large extent based on our early clinical and phlebographical results. 4 The present report describes the clinical, morphological and physiological results recorded at 5 years after the acute thrombotic event.
Table 1. Patient and thrombus characteristics in 41 patients with acute iliofemoral venous thrombosis available for follow-up at 5 years. Thrombus duration was based on the presence of leg swelling prior to institution of therapy. Extension of thrombus was defined as proximal if not involving femoro-popliteal or crural veins (impossible to evaluate in eight patients)
Medical group Surgicalgroup (n ~ 22) (n = 19) Patient
Material
From 1979 to 1982, 58 patients with unilateral acute iliofemoral venous thrombosis were included in a prospective randomised study comparing surgical thrombectomy supported by a temporary arteriovenous fistula with conventional anticoagulation treatment. The criteria for inclusion, exclusion, patient characteristics, and details of medical and surgical treatment as well as the early (6 months) clinical and phlebographic results have been described in previous reports. 4's Basically, patients with deep venous thrombosis with a presumed age of less than 7 days and a proximal extension above the inguinal ligament, but not into the vena cava, were included. Medical treatment consisted of continuous heparin infusion for 5-7 days and oral anticoagulation for at least 6 months. Venous thrombectomy was performed under general anaesthesia and fluoroscopic control using a venous Fogarty balloon catheter inserted via the common femoral vein. Distal clearance was aided by manual compression. The arterio-venous fistula was constructed by end-to-side anastomosis of the cranial portion of the divided long saphenous vein to the superficial femoral artery. The fistula was closed after 4-6 weeks. All surgical patients received anticoagulants in a similar w a y as the medical patients. Forty-one patients (22 medical and 19 surgical) were available for follow-up at 5 years after the initial thrombotic event. Sixteen patients had expired and one patient's leg was initially amputated due to progressive phlegmasia cerulea dolens with venous gangrene. When comparing the initial patient and thrombus characteristics in the remaining patients, no obvious difference between the two treatment groups was noted, although there were slightly more male patients and right-sided thromboses in the medical group (Table 1). Five medical and four surgical patients had a previous history of venous ulceration and/or DVT restricted to the calf. It should also be noted that the presence of thrombus in the iliac veins (criterion for inclusion) had been clearly verified during the operation in all surgical patients, whereas this was not clearly demonstrated in six (27%) medical Eur J VascSurg Vol4, October1990
Age (years)
59 (15-79)
55 (18-80)
Sex (males/females)
9/13
5/14
Side (right/left)
6/16
3/16
Extension (proximal/entire leg/ unknown)
5/15/2
2/11/6
Duration (days)
2 (1-7)
2 (0-5)
Thrombus
patients in w h o m puncture of the femoral vein had failed. These patients were included due to presumed presence of iliofemoral thrombosis, since no filling of the deep veins occurred with ascending phlebography from the foot. Surgical clearance of the iliac vein was mostly successful in all cases, but eight patients (42 %) had an underlying iliac vein obstruction probably caused by compression of the overriding common iliac artery, iliac compression syndrome, 6 which made catheter passage to the vena cava difficult or impossible.
Methods
All 41 remaining patients were asked to return for clinical, morphological, and physiological evaluation at 5 years after the acute thrombotic event. Morphological evaluation consisted of radionuclide angiography, whereas venous physiology was assessed by venous plethysmography, foot volumetry and examination of foot vein pressures. All data are given as mean (range). Since a normal distribution was not proven, the two-tailed Mann-Whitney U test and Fisher exact probability tests were used for statistical analysis of quantitative and qualitative data, respectively. Presence of post-thrombic sequelae (leg swelling, ulceration, heaviness, venous claudication, varicose veins) was recorded in all cases. Venous claudication was defined as pain in the thigh or calf during exercise verified on tread mill testing and without signs of arterial occlusive disease. The clinical symptoms were graded according to the "Reporting standards in venous disease" by the A d Hoc Committee: Class 0 =
Long-term Results of Venous Thrombectomy
asymptomatic; 1 = mild chronic venous insufficiency (CVI); 2 = moderate CVI; 3 = severe CVI. Radionuclide angiography was performed as described by Akesson et al.3° to determine the patency of the iliac veins, which in each case were classified as normal, stenotic or occluded. Assessment of the 5year patency was supported by information obtained by radionuclide angiograms obtained at 6 months, 1 and 3 years and by contrast phlebography performed at 6 months. Occlusion plethysmography was performed to evaluate venous outflow capacity using strain-gauge technique as described by Hallb66k and G6thlin. 31 An abnormal outflow capacity was defined as having a maximum venous outflow (MVO) < 33 ml/min/100 ml or a k-value 20s), mild (15-20s), moderate (5-15s), or severe
( 33ml/min/ 100ml and k > 0.21s-i), no reflux assessed by foot volumetry (Q/EVrel < 2 sD) and/or foot vein pressure (90% refill time > 20 s), normal muscle p u m p (EVrel > --2 SD) and (if tested) no ambulatory venous hypertension (AVP -< 45 mmHg).
Results
The presence of post-thrombotic sequelae in all 41 (22 medical and 19 surgical) patients are presented in Fig. 1. There was no significant difference in the incidence of separate late post-thrombotic sequelae. There was no significant difference in the clinical results (Fig. 2), but severe chronic venous insufficiency (Class 3) tended to be more common (27 v s . 11%) and asymptomatic patients (Class 0) tended to be less common (18 v s . 37%) in the medical than in the surgical group. In subgroups of medical and surgical patients, respectively, the proportion of asymptomatic individuals in patients aged 2 SD. Foot vein pressures are demonstrated in Table 2. CurJ VascSurgVol4, October1990
The ambulatory venous pressure was significantly (P < 0.05) lower in the surgical group of patients. Nine (50%) of the medical and two (20%) of the surgical patients had an AVP above 60mmHg (N.S.). Two (11%) of the medical and five (50%) of the surgical patients had an AVP ~ 4 5 m m H g (0.05 < P < 0.1). Three (17%) of the medical and five (50%) of the surgical patients had a pressure reduction exceeding 50% during ambulation (N.S.). Venous reflux (90% refill time) was normal in one (6%) medical and four (40%) surgical patients, mild or moderate in 11 (61%) medical and five (50 %) surgical patients and severe in six (33 % ) medical and one (10%) surgical patient (N.S.). The overall physiological results in the 19 medical and 14 surgical patients with sufficient tests performed are demonstrated in Fig. 4. All tests were normal in two (11%) medical and in five (36%) surgical patients (N.S.). Comparing subgroups of medical and
Long-term Results of Venous Thrombectomy
487
Table 2. Mean values (ranges) for physiological tests performed 5 years after an acute iliofemoral venous thrombosis. Since one patient in each group had an EVrel = 0, the Q/EVr~I could only be calculated for 18 and 13 patients, respectively. The M a n n - W h i t n e y U test was used for the statistical analyses
Treatment Medical
Surgical
P-value Mann-Whitney U test
Occlusion plethysmography MVO (ml/min/100 ml)
n = 20 45 (12-87)
n = 17 61 (24-161)
N.S.
Foot volumetry EVre~(ml/100ml/min) Q/EVrel (min- 1)
n = 19 0.9 (0.0-1.8) 4.7 (0.5-18.3)
n = 14 1.0 (0.0-1.9) 3.3 (1.1-7.4)
N.s. N.s.
Foot vein pressure Pressure reduction (%) Ambulatory venous pressure (mmHg) 90% refill time (s)
n = 18 36 (-6-78) 60 (21-95) 13 (0-71)
n = 10 52 (26-81) 43 (18-70) 14 (4-38)
P < 0.05
MEDICAL GROUP
N.S.
N.S.
SURGICAL G R O U P
[] [] •
Normal Obstruction Reflux Obstruction + reflux
Fig. 4. Physiological results in 19 medical and 14 surgical patients at 5-year follow-up after an acute iliofemoral venous thrombosis.
surgical patients, respectively, the p r o p o r t i o n w i t h n o r m a l p h y s i o l o g y in patients a g e d
