Magnetic resonance imaging and MR angiography of endoluminally treated abdominal aortic aneurysms

OBJECTIVES

To evaluate magnetic resonance imaging (MRI) with gadolinium-based contrast medium-enhanced MR angiography (MRA) for the follow-up of endoluminally treated abdominal aortic aneurysms.

DESIGN

MRI/MRA, angiography and computed tomography (CT) were performed 1 month after endoluminal stent-graft placement. MRI/MRA was repeated at 6 and 12 months and angiography and CT were added to confirm unexpected findings.

MATERIALS

Fifteen male patients with endoluminally treated abdominal aortic aneurysms.

METHODS

MRI with MRA, spiral CT with transverse images and angiography were performed.

RESULTS

MRI/MRA demonstrated changes of stent-graft morphology, aortic neck- and aneurysmal diameter, stent-graft blood flow, stent-graft leakage, blood flow in lumbar arteries, intra-aneurysmal thrombus, periaortic inflammation and vertebral body infarction. For most of these features MRI/MRA provided more information than angiography and/or CT. MRI was the only method demonstrating thrombus reorganisation and vertebral body infarction.

CONCLUSIONS

MRI with MRA provides the relevant information needed for follow-up of endoluminally treated abdominal aortic aneurysms (AAA). This may be the method of choice because of its use of contrast media with very low nephrotoxicity, lack of ionising radiation and non-invasiveness.

Biological responses differ considerably between endovascular and conventional aortic aneurysm surgery

OBJECTIVES

To determine the inflammatory responses in endovascular abdominal aortic aneurysm (AAA) repair and their relation to clinical findings.

DESIGN

Prospective non-randomised study.

SETTING

University Hospital, Department of Surgery.

PATIENTS AND METHODS

Seven patients treated with an endoluminal procedure (AAA-E) and seven patients undergoing conventional surgery (AAA-C) were included. Inflammatory parameters were assessed by measurements of the cytokines interleukin (IL)-1 beta, IL-6, IL-8 and Tumour Necrosis Factor-alpha (TNF-alpha); analyses of complement proteins C1q, C4, C3, C5a and Terminal Complement Complexes (TCC); haematologic parameters and determination of C-reactive protein (CRP).

RESULTS

In six of seven patients in the AAA-E group blood pressure decreases were recorded during introduction of the device. IL-6 and CRP levels were found to be significantly higher in AAA-C patients compared to the AAA-E group. On the other hand, high TNF-alpha levels were recorded in the AAA-E group. Less consumption of the complement proteins C1q, C4 and C3 was observed in AAA-E compared to AAA-C patients. Increased C5a levels were recorded in the AAA-C group, whereas only slight fluctuations were noticed in the AAA-E group. TCC levels were unchanged in both groups.

CONCLUSION

Endovascular aortic aneurysm repair induced a significant inflammatory response, mainly involving TNF-alpha and differing from the findings during open AAA repair. These inflammatory responses were probably related to blood pressure decreases during the procedures. On the other hand, conventional repair induced responses related to the more extensive surgical trauma and reperfusion injury.

Surgical treatment versus thrombolysis in acute arterial occlusion: a randomised controlled study

Thrombolytic treatment has been tried in various forms for acute limb ischaemia with varying degrees of success but is also often accompanied by bleeding problems. The present investigation compares the effect of surgical thrombectomy (TE) and thrombolysis (TL) using recombinant tissue plasminogen activator (rt-PA). Twenty patients with a need for intervention owing to ischaemia lasting more than 24 h but less than 14 days were included. Patients randomised to TE were operated under epidural anaesthesia and patients in the TL group received 30 mg rt-PA during a 3 h period through a catheter placed into the thrombus and advanced as lysis was achieved. Thrombectomy resulted in an immediate restitution of blood flow in six out of nine cases, in three cases a bypass procedure was performed, and one of these failed with a resultant amputation. Thrombolysis gave a good primary result in six cases which lasted in four of them. Three had a subsequent percutaneous transluminal angioplasty. Partial lysis was seen in two cases and a further two failed. Five went to surgery with three bypass and two fogarty procedures being necessary. There was no hospital mortality and there were no bleeding complications due to the rt-PA treatment in this series. In 19 out of 20 patients the circulation was re-established. Appropriate handling of acute ischaemic conditions implies the use of both thrombolysis and appropriate surgical procedures, including distal bypass grafts.

Intestinal hemodynamics during laparotomy: effects of thoracic epidural anesthesia and dopamine in humans

The effects of thoracic epidural anesthesia (TEA) and dopamine infusion (4 micrograms.kg-1.min-1) on superior mesenteric artery blood flow (SMABF), the mesenteric arteriovenous oxygen difference (AVDO2), and the mesenteric venous lactate concentration were studied in nine patients before abdominal aortic reconstruction. Thoracic epidural anesthesia reduced SMABF, as measured by electromagnetic flowmetry, to 77% +/- 8% (mean +/- SEM) of control (P less than 0.05), and mean arterial pressure to 46% +/- 4% of control (P less than 0.01). The mesenteric AVDO2 increased from 27 +/- 3 to 39 +/- 6 mL/L (P less than 0.05) and superior mesenteric venous lactate from 1.03 +/- 0.11 to 1.60 +/- 0.38 mmol/kg (P less than 0.05); systemic AVDO2 and lactate did not change. Dopamine had no significant effect on SMABF and mean arterial pressure before TEA. However, dopamine increased SMABF during TEA (from 77% +/- 8% to 137% +/- 21% of control; P less than 0.01), returned mesenteric AVDO2 to 27 +/- 3 mL/L (P less than 0.05), and elevated mean arterial pressure to 62% +/- 4% of control (P less than 0.05). It is concluded that the decrease in perfusion pressure during TEA reduces SMABF with resultant evidence of intestinal reductive metabolism. The intestinal blood flow during TEA was improved by dopamine.

Direct evidence of active sympathetic vasodilatation in the skin of the human foot

1. During operative aorto-femoral vascular reconstructions on sixteen patients, the sympathetic chain was stimulated electrically between the L2 and L4 ganglia while blood flow was monitored by laser doppler flowmeters from the skin on the sole of the foot and the ankle and by an electromagnetic flowmeter from the deep femoral artery. Epidural anaesthesia to at least the T6 level was used which excluded reflex effects. 2. Stimulation (10 Hz) at 1-12 mA current strengths for 30 s evoked both reductions and increases of blood flow in glabrous and hairy skin. Initial short-lasting flow increases (durations 9-19 s) followed by sustained decreases were common: sometimes there were sustained flow increases at low and decreases at high current strengths. 3. In the deep femoral artery (supplying predominantly muscle) only flow reductions were evoked. 4. The results provide evidence for sympathetically mediated vasodilatation in the skin of the human foot whereas leg muscles may be supplied by vasoconstrictor nerves only.

Secondary aortoenteric fistulae--changes from 1973 to 1993

AIM

To investigate a series of patients with secondary aortoenteric fistulas and compare it with a previous series (1985-93 vs. 1973-84).

DESIGN

Retrospective study of medical records.

SETTING

Sixteen vascular surgical centers in Sweden.

PATIENTS

Twenty-seven patients were identified making an overall incidence of 0.5% of all aortoiliac operations. Among aneurysm patients the incidence was significantly lower than in the previous series. One patient record could not be identified. Fourteen primary operations were for aortic aneurysm, 12 for occlusive disease and one was an aortorenal vein bypass.

RESULTS

Symptoms of the fistula occurred after a median interval of 90 months which is significantly later than the previous series (32 months; p<0.05). The commonest presentation was bleeding followed by septis. The median diagnostic delay was 10.5 days, which was significantly shorter than in the previous series. Most fistulas involved the duodenum (88%). One patient died before surgery. The postoperative mortality was 28%, significantly lower than in the previous series (58%) (p<0.05). At the end of follow up (median 43 months) significantly more patients were alive than in the previous series (42% vs 18%) (p<0.05).

CONCLUSION

Over a 21 year period there seems to have been a decrease in the frequency of secondary aortoenteric fistulas after aneurysm surgery, a longer interval before they occur, a shorter diagnostic delay, and a better survival.

Percutaneous Transluminal Angioplasty of Crural Arteries: Diabetes and other Factors Influencing Outcome

OBJECTIVE

to evaluate the efficacy of percutaneous transluminal angioplasty (PTA) of the crural arteries.

PATIENTS AND METHODS

a retrospective review of patients treated with PTA of at least one crural artery during an 8-year period (1990--1997).

RESULTS

one hundred and fifty-five legs in 140 consecutive patients (mean age 74 years, range 38--91 years) were treated. In 76% a more proximal lesion was also treated. After 1 year, results were significantly better in non-diabetics (improvement rate of 66% vs 32%p <0.05). The outcome for patients with a combination of diabetes, heart disease and renal disease was significantly worse compared to all other patients with an improvement rate of only 9% after 1 year. Patients alive and not amputated at 1 year were significantly more common (p <0.05) among non-diabetics (90%), compared to diabetics (66%). The 1-year mortality for the whole group was 15%, significantly higher for diabetic patients (p =0.04).

CONCLUSION

PTA of crural arteries produces reasonably good results in non-diabetic patients. Diabetic patients were doing worse than non-diabetics after a year, though 1-month results were not significantly different. Patients with diabetes, heart disease and renal disease make a high-risk group that has a significantly worse outcome.

Secondary aortoenteric fistulas--an analysis of 42 cases

Secondary aortoenteric fistulas are seen with an increasing frequency which parallels the expansion of reconstructive vascular surgery. During a 12-year period 42 cases have been collected from the hospitals, which perform most of the vascular surgery in Sweden (0.7% of vascular operations). Twenty-five were seen after operation for aortic aneurysm, 15 for aortoiliac occlusive disease, one after renal artery ligation (as a part of reconstruction for renovascular hypertension) and one after operation for an iliac pseudoaneurysm. The frequency of complications during and after the primary operation was high. The interval between operation and onset of fistula symptoms was significantly shorter if there had been infectious complications, the median interval was 32 months, the longest being 10 years. The most important symptom was gastrointestinal haemorrhage, consisting of several small bleeds often combined with septic complications. A large number of negative investigations usually preceded the final diagnosis which was made at exploratory laparotomy. After surgery for the fistula the frequency of complications and mortality, were very high. Mortality was 58%, the most common cause of death being a blow out of the aortic stump. Of those leaving hospital, several further operations were performed with a high mortality. Recurrence of the fistula occurred in 16 out of 34 patients who survived surgery. At follow-up (12-74 months after fistula closure) seven patients were still alive.

Immune response to collagen impregnated Dacron double velour grafts for aortic and aorto-femoral reconstructions

This study presents 20 patients, randomised to receive either a collagen-treated or an ordinary Dacron graft for aortic reconstructions, and the results of a skin-prick test, blood parameters and ELISA for anti-collagen antibodies as well as NMR pictures during a 6 week follow-up period. Forty per cent (4/11) of those receiving a collagen impregnated graft had a significantly increased titre of antibodies and NMR revealed in two out of 11 patients either a slightly increased amount of fluid or fibrosis around the graft, both collagen impregnated. No differences were found between the graft groups concerning body temperature and leucocyte or platelet counts. The skin-prick test for collagen was negative in all cases.

Intraoperative contamination of synthetic vascular grafts. Effect of glove change before graft implantation. A prospective randomised study

OBJECTIVES

to investigate the incidence of intraoperative graft contamination, bacterial species and the influence of change of surgeon's gloves on contamination.

DESIGN

a prospective randomised study.

MATERIALS AND METHODS

forty patients had implantation of synthetic vascular grafts. All patients received intraoperative cloxacillin (2.0 g) or clindamycin (0.6 g) intravenously. The procedures were randomised to two groups: Group 1 - surgeons changed the gloves before the first contact with the vascular prosthesis and Group 2 - operation without glove change. The growth of all bacterial species from graft segments and from the gloves was recorded. The susceptibility to antibiotics was tested.

RESULTS

the number of contaminated grafts was similar in the two groups. Growth of bacteria was recorded from 92.5% (37/40) of the graft segments and 33% (51/156) of glove imprints. Of the cultured species, 75% and 47%, respectively, were identified as coagulase-negative staphylococci (CNS). Twenty-eight per cent of CNS were resistant to cloxacillin, 15% to clindamycin, and 10% to cloxacillin and clindamycin. In all, 25% of the CNS strains were resistant to the prophylactic antibiotic used. In 50% of cases, the antibiogram of the CNS strain recovered from gloves agreed with that of the strain harvested from the graft.

CONCLUSIONS

a high incidence of graft contamination was found which was not reduced by changing gloves. However, changing gloves did seem to reduce the number of bacterial species.

Evidence of calcium phosphate depositions in stenotic arteriovenous fistulas

This study investigates vascular samples from patients with and without end-stage renal disease (ESRD) to determine the occurrence of calcium depositions. Findings in stenotic arteriovenous (AV)-fistula veins were compared with those of nonstenotic AV-fistula veins, non-AV-fistula veins, and atherosclerotic vessels. Calcium and phosphorus content was measured by means of scanning electron microscopy and its built-in method of energy-dispersive spectrometry (EDS) X-ray analysis. We found calcium and phosphorus in samples from AV fistulas with stenotic areas with a calcium/phosphorus molar ratio of 1. Based on EDS analysis and crystal shape comparison, we conclude that calcium phosphate precipitations in stenotic AV fistulas are brushites with the composition CaHPO(4)*2H(2)O. This specific calcium phosphate deposition was found solely in stenotic AV fistulas, not in nonstenotic AV-fistula veins or non-AV-fistula veins regardless of whether the patient had ESRD. Moreover, this calcium phosphate deposition was different from calcium compounds found in atherosclerotic samples. Whether the precipitation of brushite is primarily involved in the development of vascular-access stenosis or represents a secondary consequence cannot be determined from the present study.

Outcome of treatment of ruptured abdominal aortic aneurysms depending on the type of hospital

OBJECTIVE

To compare the outcome of patients operated on acutely for ruptured abdominal aortic aneurysms (AAA) or otherwise symptomatic aortic aneurysms in a university hospital and in two county hospitals by the same group of vascular surgeons.

DESIGN

Retrospective study.

SETTING

1 university and 2 county hospitals, Sweden.

SUBJECTS

108 patients operated on urgently for AAA, 81 at the university hospital, and 27 at the county hospitals between January 1992 and December 1998.

INTERVENTION

Repair of the AAA.

MAIN OUTCOME MEASURES

Morbidity and mortality.

RESULTS

21 of the 81 patients having urgent repair of an AAA at the university hospital (26%) had been transferred from the county hospitals. Thirteen patients were not operated on, 7 because of their poor general condition and great age (median 84 years), 3 who refused operation, and 3 in whom the diagnosis was incorrect. During the same time period a further 27 haemodynamically unstable patients were operated on by the same vascular surgeons at the county hospitals. The on-table mortality for patients with ruptured AAA and shock was 5/43 (12%) at the university hospital and 4/27 (15%) at the county hospitals. The corresponding in-hospital rates were 11/43 (26%) and 11/27 (41%). Mortality was significantly higher if the operation was delayed by more than 45 minutes. The incidence of postoperative complications was the same in both hospitals.

CONCLUSION

If a patient with a ruptured AAA and shock is admitted to the county hospital and operated on by a specialist vascular surgeon the outcome is fully acceptable. The difference seems to be related to the postoperative period. To what extent the delay caused by the surgeon's journey to the county hospital has any influence on the outcome is not possible to evaluate.

Influence of some plasma proteins on in vitro bacterial adherence to PTFE and Dacron vascular prostheses

The in vitro adherence of Staphylococcus aureus, Staphylococcus epidermidis and Escherichia coli (one strain of each species) to commercially available, microporous polytetrafluoroethylene (PTFE) and woven Dacron vascular grafts before and after coating with human plasma was compared. Standard size segments of the materials were incubated with 35S-labeled bacteria for 0.5-18 h and, following washes, the radioactivity associated with the segment was measured. The binding of each of the tested species to native Dacron was higher than to PTFE. After coating with human plasma, however, the binding of all three species to PTFE was significantly enhanced, whereas the binding to Dacron was reduced. In addition, the influence of coating with serum albumin (HSA), immunoglobulin G (IgG), fibrinogen (Fg) or fibronectin (Fn) rather than whole plasma was tested. We found that coating with HSA reduced the binding of all three species to Dacron and of staphylococci to PTFE. IgG decreased the binding of S. epidermidis and E. coli to Dacron and of S. epidermidis to PTFE. In contrast, Fg enhanced the binding of S. aureus both to Dacron and PTFE, and that of E. coli to PTFE, but decreased the binding of S. epidermidis and E. coli to Dacron. Fn enhanced the binding of S. aureus to Dacron, and of E. coli to PTFE, but decreased the binding of S. aureus to PTFE and of S. epidermidis both to PTFE and Dacron. Thus, both whole plasma and some isolated plasma proteins were found to modulate bacterial adherence to two tested graft materials. From a clinical point of view, bacterial binding to plasma-coated rather than native materials may more adequately determine the likelihood of in vivo colonization of the various materials. Furthermore, precoating of materials with selected proteins may be of value in the prevention of graft colonization.

Bacterial adherence to synthetic vascular prostheses and influence of human plasma. An in vitro study

The in vitro adherence of Staphylococcus aureus, Staphylococcus epidermidis and Escherichia coli to five commercially available prosthetic vascular graft materials was compared. The influence of precoating the segments with human plasma for 2 h was also studied. S35-methionine was used to radiolabel bacteria. The segments were exposed to bacterial suspensions of approximately 10(7) CFU/ml at 37 degrees C for 0.5-18h. Following repeated washing in phosphate buffered saline (PBS), radioactivity associated with the segments was measured. The adherence of the three clinically relevant bacterial species was higher to untreated Dacron than to gelatin or collagen impregnated/coated Dacron or to PTFE. Furthermore, precoating of grafts with human plasma reduced bacterial adherence to woven Dacron, had a little effect on gelatin coated Dacron, but increased the adherence to collagen treated Dacron and, in particular, to PTFE.

Exposure of plasma proteins on Dacron and ePTFE vascular graft material in a perfusion model

OBJECTIVES

to compare the exposure of plasma proteins adsorbed onto three vascular graft materials (polytetrafluoroethylene ePTFE and two modifications of polyethyleneterephthalate Dacron).

METHODS

surface exposure of fibronectin, vitronectin, thrombospondin, antithrombin III, IgG, high molecular-weight kininogen, fibrinogen, albumin and plasminogen was studied by incubation with radiolabelled antibodies in a perfusion model. Perfusion times with human plasma were 1, 4, 24 and 48 hours.

RESULTS

all proteins could be detected at 1, 4, 24 and 48 hours after the start of perfusion. Overall, the least amount of proteins adsorbed onto ePTFE.

CONCLUSIONS

the low adsorption of proteins onto ePTFE may be one of the reasons for the lower incidence of infections reported with this material.

Infections and antibiotic prophylaxis in reconstructive vascular surgery

In 98 patients undergoing elective vascular surgery, specimens for bacterial cultures were obtained from urine, ischaemic ulcers, incisional wounds and the implanted grafts. Wound and graft infections were registered and compared with the results of these cultures and suspected risk factors in an attempt to find the source of infections. Antibiotic prophylaxis with cefuroxime was given for 24 h beginning at the start of surgery. Patients with ischaemic ulcers also received "spread prophylaxis", directed against isolated bacteria, for ten days. Three cases of graft infection and twelve cases of wound infection occurred. Positive postoperative cultures from wounds did not correlate with pre- or peroperative cultures. Peroperative cultures revealed small numbers of staphylococcus epidermidis in eleven patients, and none of them developed graft infection. Ischaemic ulcers, diabetes or re-do procedures were not accompanied by a significantly increased frequency of wound or graft infection, although each of three patients with graft infection had one of these risk factors. Bacteria, sensitive to cefuroxime, were found in one graft infection, six wound infections, and in two patients with urosepsis, whereas cefuroxime resistant organisms were isolated from one graft infection and three infected wounds. One of the three graft infections was probably caused by bacteria originating from the patient's ischaemic ulcer. In the other two patients the source of bacteria could not be determined. Cefuroxime seems to be an adequate alternative for prophylaxis of vascular graft infection, but in some patients with bacteriuria or indwelling catheters, a one day regimen may be too short.

Endothelialization of microporous polytetrafluoroethylene grafts in the infrarenal aorta and caval vein of the rat

To study healing and endothelialization of vascular grafts, microporous polytetrafluoroethylene (PTFE) prostheses 2 mm in inner diameter and 5 mm long were implanted into the infrarenal aorta or caval vein of the rat. Patency was assessed in six rats from each group at days 3, 7, 14, 28, and 56 after implantation. Four grafts were occluded, two in the aorta (56 days) and two in the caval vein (3 and 14 days). The prostheses were examined via scanning electron and light microscopy for evaluation of endothelialization. At 3 days, the inner surface of the aortic grafts was covered by a plasma proteinaceous layer and that of the caval vein grafts by a thin mural thrombus. Endothelial cells then migrated from aorta/caval vein edges over the graft. At 14 days, the caval vein grafts were completely re-endothelialized, and, at 28 days, the mural thrombus in these grafts was replaced by neointima. In contrast, endothelialization of the aortic grafts had advanced only 1 mm at about 56 days, never forming a complete endothelial layer. We conclude that endothelialization of microporous PTFE prostheses is more rapid and complete in the caval vein than in the aorta of the rat.

Experimental evaluation of polymerized Dacron grafts in the iliac position of pigs

Three differently polymerized Dacron grafts were compared to ordinary Dacron grafts and polytetrafluoroethylene grafts in the iliac artery of pigs. The grafts were inserted end-to-side and left for 40 days. Explantation was performed after glutaraldehyde pressure fixation. Gross morphology, light microscopy, and scanning electron microscopy were used to examine the specimens. We concluded that the polymerized grafts behaved similarly to polytetrafluoroethylene grafts and had a very limited foreign body reaction. Endothelial-like cells covered most of the graft surface.

Endothelialization of anastomoses in the aorta and inferior vena cava of the rat

Denuding endothelial damage at anastomoses may lead to thrombotic complications and failure of microvascular reconstructions. Confluent endothelial healing at anastomoses may reduce thrombotic and other complications. The progress of endothelial coverage of anastomotic structures in the aorta and inferior vena cava (IVC) of the rat was evaluated at 3, 5, 8, 11, and 14 days after completion of the anastomoses, by means of scanning electron microscopy. Arterial anastomotic patency was 100%; venous patency was 84%. It was found that endothelial healing progressed towards thin confluent coverage first, thereafter thickening. Coverage was first achieved at the needle defect, then the anastomotic junction, and lastly the suture. The same stage of endothelial coverage of anastomotic structures was reached 5 to 8 days sooner at the IVC anastomosis than in the aorta.

Reduced susceptibility of polytetrafluoroethylene vascular prostheses to colonization by Staphylococcus aureus following in situ endothelialization

The impact of endothelialization of polytetrafluoroethylene (PTFE) grafts on susceptibility to experimental colonization by Staphylococcus aureus was studied in a rat model. One hundred and eight grafts (2 mm inner diameter, 5 mm length) were implanted into the infrarenal aorta (54 rats) or the infrarenal caval vein (54 rats). The progress of endothelialization following graft implantation was evaluated by SEM at 1, 3, 7, and 14 days on 6 grafts from each group. We found that the endothelialization was more rapid in caval vein than in aorta: The caval vein grafts were completely endothelialized 2 weeks following implantation compared with endothelialization of approximately 0.5 mm of aorta grafts, measured from each anastomosis. During this time, the resistance to standardized intravenous challenge with 10(8) colony-forming units of S. aureus increased gradually in both groups, and all caval vein grafts tested at 2 weeks were sterile. However, all aorta grafts challenged at 2 weeks were colonized, although to a somewhat lower degree than at earlier challenge. Resistance of PTFE grafts to colonization with S. aureus thus correlated to the degree of endothelialization.

Intra-Operative Digital Angiography as a Control of the in Situ Saphenous Vein By-Pass Grafts

Forty-two patients undergoing in situ saphenous vein by-pass grafting procedures, in two patients bilaterally, were examined intra-operatively with digital subtraction angiography. In 19 (43%) of the examinations the graft and the anastomoses appeared adequate. In 8 cases (18%) significant abnormalities were found, including stenoses (11 %), deficient anastomoses (5%) and graft kinking (2%). Remaining arteriovenous fistulas were found in 17 patients (39%). In most cases immediate correction was possible avoiding later re-operation. At follow up 11 of the 44 grafts were occluded, 10 of these during the first five months and of these five during the first week.

Intra-operative digital angiography as a control of the in situ saphenous vein by-pass grafts

Forty-two patients undergoing in situ saphenous vein by-pass grafting procedures, in two patients bilaterally, were examined intra-operatively with digital subtraction angiography. In 19 (43%) of the examinations the graft and the anastomoses appeared adequate. In 8 cases (18%) significant abnormalities were found, including stenoses (11%), deficient anastomoses (5%) and graft kinking (2%). Remaining arteriovenous fistulas were found in 17 patients (39%). In most cases immediate correction was possible avoiding later re-operation. At follow up 11 of the 44 grafts were occluded, 10 of these during the first five months and of these five during the first week.

Early development of suture holding capacity in vascular anastomoses in the rat

Early development of suture holding capacity (SHC) of end-to-end anastomoses in the infrarenal aorta and inferior vena cava (IVC) of the rat was studied by measuring the resistance to tensile forces at 0, 3, 5, 8, 11, or 14 days postoperatively. Comparable anastomoses were histologically evaluated. The SHC of the aortic wall decreased following anastomoses and did not regain the strength of nonanastomosed vessels during the first two weeks. The SHC of the IVC also decreased following anastomoses, but regained the strength of nonanastomosed vessels by the 5th day. The relative ultimate strain of the arterial and venous perianastomotic walls decreased slightly, while the relative compliance increased in the aorta and decreased in the IVC. Perianastomotic media necrosis was found exclusively in the arteries. The reduced perianastomotic SHC in the aorta indicates impaired healing, compared to the IVC. Changes in biomechanical properties of the aortic wall following anastomosis may increase its susceptibility to anastomotic complications such as pseudoaneurysms.