Aust. N.Z. J. Surg. (1996) 66, 621-625
`
`ORIGINAL ARTICLE
`
`A SELF-EXPANDING ENDOLUMINAL GRAFT FOR TREATMENT OF
`ANEURYSMS: RESULTS THROUGH THE DEVELOPMENT PHASE
`
`M. K. Gorpbon,* M. M. D. LAWRENCE-BROWN,* D. HARTLEY,’ K. SIEUNARINE,* A. HOLDEN,*
`S. T. R. MACSWEENEY* AND M. J. HELLINGS*
`Departments of *Vascular Surgery, ‘Radiology, and *Anaesthesia, Royal Perth Hospital, Perth,
`Western Australia, Australia
`Background: The results of two and a half years’ experience of endoluminal treatment of aneurysmal disease (from March
`1993 to December 1995) are
`reported.
`are based on Dacron-coveredstainlesssteel
`were
`individually made at
`Royal Perth Hospital. They
`Methods: The endoluminalgrafts
`self-expanding ‘Z’ stents with Gianturco barbed stents (Cook Pty, Australia) for proximal anchoragefor grafts within the aorta.
`Results: Fourteen straight tube grafts (nine for aortic aneurysm, fourfor iliac aneurysm and one for subclavian aneurysm) and
`were
`deployed; all were in patients considered high-risk for conventional repair. Seventy-two per cent of
`24 bifurcate grafts
`the straight tube grafts successfully excluded the aneurysm. The bifurcate grafts, in use since July 1994, successfully excluded
`the aneurysm in 88%. There were two
`delayed deaths from rupture after the grafts failed to exclude the aneurysms; two
`patients
`required conversion to open repair and survived; three patients have persistent endoleaks; and three of the bifurcate grafts
`subsequently occluded a
`require further intervention. Ninety per cent of these complications occurred in
`graft limb but did not
`to
`the first half of the series (prior
`January 1995).
`curve was
`to those for
`clearly apparent. The results thereafter compare favourably
`Conclusions: A learning and development
`open repair in similar high-risk groups, suggesting that these techniques hold promise forall patients with aneurysms.
`
`Key words: aortic aneurysm, bifurcate, endoluminal, graft.
`
`INTRODUCTION
`This paper presents the results of two and a half years’ expe-
`of abdominal aortic aneurysms.
`rience in endoluminal grafting
`as a real alternative to open
`Endoluminal grafting is emerging
`to Parodi er al.’s initial report in 1991,
`surgery. Subsequent
`Mayet al., Dake et al. and Chuter et al. have confirmed the
`in selected cases.'-* However, long-term
`technique is possible
`results remain unknown.
`Conventional open repair of abdominal aortic aneurysms has
`become a well-established
`in the four decades since
`procedure
`wasfirst described by Dubostef al.5 Apart from
`this procedure
`modification in the 1960s where endo-aneurysmal
`repair
`replaced aneurysm resection, as
`popularized by DeBakey® and
`little change has occurred in the basic technique.
`Creech,’
`Results have been proven with time. Aneurysm exclusion is
`certain and durability is excellent; patency rates for the grafts
`are minimal.
`are
`high, and infective complications
`long-term
`However, morbidity and mortality remain significant. Published
`rates vary widely, with a rate of 4-9% for unselected
`mortality
`rates for octogenarians'® and lower rates
`population,*’ higher
`for groups where preliminary cardiac and cerebral vascular
`revascularization is aggressive.!'
`relates to
`Much of the morbidity of conventional repair
`trauma of access, and haemodynamic changes inducedbyaortic
`clamping. These effects are minimized by endoluminal
`cross
`to the
`must be compared
`repair. These potential advantages
`technique. We
`expense and uncertaintly of the endoluminal
`
`Mr M. M. D. Lawrence-Brown, Vascular Surgery Department, Royal
`Perth Hospital, GPO Box X2213, Perth 6001, Australia.
`Accepted for publication 15 May 1996.
`
`documentthe results of 38 endoluminal procedures for arterial
`a
`aneurysm in 37 patients, using
`self-expanding graft system
`developed conjointly by the Royal Perth Hospital Departments
`of Vascular Surgery and Radiology.
`
`METHODS
`
`Patient selection
`Criteria have remained stringent due to lack of long-term data
`and have been based on these indications: (i) the risk of death
`to exceed the risk of death by
`by aneurysm rupture is thought
`for aortic aneurysm the diameter was
`concurrent disease;
`greater than 50 mm andfor iliac aneurysm the diameter was
`greater than 40 mm;
`(ii) life expectancy is less than normal for
`or
`age, due to
`(iii) advanced age (greater than
`co-morbidity;
`80); and (iv) quality of life is good.
`
`Assessment of the aneurysm
`This is carried out with a
`spiral computed tomography (CT)
`scan and aortobifemoral angiogram. Each patient is assessed
`which is tailored to
`individually for the stented graft,
`precise
`measurements of diameters and lengths taken from spiral CT
`images. Spatial appreciation is improved with 3D reconstruc-
`to allow assessment of:
`tion. Angiography is
`performed
`patency; number and
`of
`the renal arteries;
`the
`position
`contribution of the inferior mesenteric artery to the collateral
`at the
`circulation of the mid gut, particularly any hypertrophy
`marginal artery of Drummond; andthestate of the iliac arteries.
`
`Assessmentof the patient
`Operative fitness is assessed with anaesthetic consultation, as
`specialty consultation as
`for open surgery, with medical
`MEDTRONIC 1122
`
`MEDTRONIC 1122
`
`

`

`622
`
`GORDON FTAL.
`
`required. Although high-risk patients have been selected, some
`prospect of surviving salvage surgery (should it be necessary)
`is requisite. Ischaemic heart disease, as an indication for endo-
`was based on coronary angiography, cardiac
`luminal repair,
`catheter studies and the patient’s prospects of withstanding
`open aneurysm surgery, or cardiac and aneurysm surgery.
`
`Stented graft construction
`Each graft is individualized to each patient, using commercially
`available woven Dacron with the crimps ironed out. Stainless
`steel ‘Z’ stents line the graft, sutured in place with Prolene. For
`aneurysms of the aorta, an uncovered Gianturco barbed stain-
`assem-
`less steel stent is sewn to the proximal end of the graft
`bly. The stented graft is soaked in heparin (100 000 units/L)
`and compressed into a haemostatic sheath delivery system
`(ranging from 14 F to 22 F) using the narrowest sheath allowing
`to slide within. The system is held in place by the
`the assembly
`barbs of the uncovered Gianturco stent in the proximal
`aorta
`and by shear forces elsewhere.
`Construction and technical details of the Perth Bifurcated
`Graft for aortic aneurysms have been described elsewhere by
`our overall experience with
`this group’? and we now
`report
`endoluminal grafting. For the bifurcated graft, using the same
`delivery principles (proximal Gianturco stent and whole length
`a commercial bifurcated graft is employed with
`stent
`support),
`cut off 12 mm below the bifur-
`one limb of the bifurcate graft
`cation (Fig. 1). The remainderof this limb is introduced via the
`as an extension piece.
`contralateral femoral artery and deployed
`This docks into the expanded stump of the main graft already
`placed in the aorta.
`
`Anaesthesia
`In this series three anaesthetic techniques have been used to
`provide anaesthesia for the patients: (i) regional anaesthesia
`an
`epidural technique); (ii) general anaesthesia
`(provided by
`a muscle relaxant, intubation and controlled ventilation);
`(using
`and (iii) general anaesthesia (with spontaneous ventilation via
`a
`laryngeal mask airway).
`Fully informed consent was obtained from the patients and
`was
`their families, and hospital ethical committee approval
`given.
`
`RESULTS
`Thirty-eight endoluminal covered stents were
`deployed in 37
`patients since March 1993 (Fig. 2) (14 straight tubes and 24
`bifurcate grafts). All but one of the bifurcate systems have been
`deployed for abdominal aortic aneurysms, with either common
`iliac artery involvement or poor inferior necks. The other
`
`ereNe AOAF
`
`1. The Royal Perth Bifurcate Graft with detached limb
`Fig.
`extension piece.
`
`March 93
`
`B
`
`B BBB B
`
`
`aeA AAA AA
`Typeof
`Complications key
` orovedure
`Al Aorto-uni-iliac tube
`A Aortic straight tube
`Iliac straight tube
`|
`S Subclavian tube
`B Aortio-bi-iliac
`
`Death
`
`sae] Stroke
`{| Persisting endoleak
`NN Graft limb occlusion
`
`[z:] Conversion to open
`item failure
`onyee
`January Gar
`
`B
`
`B
`
`6B BBS B
`
`B BA B
`
`TOMEI
`
`Fig. 2. Timing and type of major complications of the 38 procedures
`performed.
`
`patient had large bilateral iliac aneurysms that developed sub-
`a
`sequent to an openrepair (of
`ruptured abdominal aortic aneu-
`rysm by straight tube graft) 3 years previously.
`Twenty-nine patients had endoluminalgrafts inserted as their
`treatment
`first
`for
`infrarenal abdominal aortic aneurysms.
`Twenty-two of these were bifurcated and two were aorto-uni-
`were inserted primarily for common
`iliac. Two bifurcated grafts
`iliac artery disease, where there was insufficient wall contact in
`a secure sealfor a
`commoniliac artery to
`the proximal
`provide
`were inserted.
`tube iliac graft. A total of 24 bifurcated grafts
`Of the 14 straight tube grafts, four were
`to treat
`deployed
`or iliac (2) anastomosis of
`false aneurysmsat the aortic (2)
`at open surgery. One straight
`tube
`grafts placed previously
`covered stent was used to exclude a subclavian artery
`aneu-
`rysm, and two were used for isolated commoniliac aneurysms.
`Five aortic aneurysms with adequate inferior necks allowed
`Two aorto-uni-iliac grafts
`deployment of straight tube grafts.
`were used. In both patients the contralateral iliac artery
`was
`was not used
`chronically occluded. A fem-fem cross-over
`graft
`was
`in either case as neither patient
`claudicating within their
`exercise distance.
`Figure 2 demonstrates the significant complications experi-
`enced and shows where these occurred within the series.
`Table |
`lists minor complications.
`Patients ranged in age from 60 years to 95 years (average
`74 years), and all were selected for an endoluminal procedure
`because of perceived high risk for conventional surgery and/or
`to an
`advanced age. The more
`frequent conditions contributing
`assessmentof high risk are listed in Table 2, with many patients
`having multiple risk factors. The aneurysm was
`successfully
`
`Table 1. Minor complications
`
`Complication
`
`Woundinfection
`Haematoma/seroma
`URTI
`UTI
`Small bowel obstruction
`
`n (%)
`
`1 (3)
`2 (5)
`2 (5)
`I (3)
`1 (3)
`
`URTI, upperrespiratory
`
`tract infection; UTI, urinary
`
`tract infection.
`
`

`

`SELF-EXPANDING ENDOLUMINAL GRAFT SYSTEM
`
`623
`
`Table 2. Indications for endoluminal repair
`
`n
` Indication
`(%)
`
`Ischaemic heart disease
`Respiratory insufficiency
`Hostile abdomen
`> 80 years
`Age
`Renalfailure
`Poorleft ventricular function
`Major cardiac valve disease
`Obesity
`Cerebrovascular disease
`
`Thrombocytopenia
`Jehovah’s Witness
`Colonic stoma
`Peritoneal dialysis
`
`21 (55)
`14 (37)
`9 (24)
`8 (21)
`7 (18)
`6 (16)
`5 (13)
`4 (10)
`3
`(8)
`2
`(6)
`1
`(3)
`1
`(3)
`1
`(3)
`
`excluded in 31 of 38 patients (81%) at the initial procedure, and
`in one further patient after deployment of an extension piece
`to
`his iliac straight tube graft (85% total). Two patients required
`conversion to open surgery when graft deploymentfailed. There
`were two
`peri-operative deaths (5.4%) where
`incomplete exclu-
`sion of the aneurysm occurred, and one further patient suffered
`debilitating stroke. Three patients sustained occlusion ofbifur-
`a
`cate
`graft limbs. The longest period of follow up is 33 months,
`with a range of 1-33 months and a median of 12 months.
`
`DISCUSSION
`
`ofthis unit’s
`Theresults presented span the developmentperiod
`(Vascular Department, Royal Perth Hospital) endoluminal
`stented graft and delivery system. Despite disappointments, the
`remarkable benefits noted in the early convalescence phasepro-
`curve is well demon-
`vided a constant incentive. The learning
`strated (Fig. 2) by the decrease in numberof complications; all but
`occurredin thefirst halfof the series.
`one
`significant complication
`Two deaths (5.4%) occurred on the fifth and twelfth post-
`to stent-
`operative days; both from aneurysm rupture subsequent
`to
`rate is
`published results for
`ing. This mortality
`comparable
`open repair in unselected populations.’ In both cases a
`complete
`seal was unable to be obtained at the superior neck, butthe seal
`was
`complete. The aneurysm sac remained
`at the inferior neck(s)
`exposedto inflow atarterial pressure but without outflow. Minor
`leaks were detected at the top end aroundthe graft
`at
`completion
`angiography and wereeliminated by balloon expansion. Bedding
`of the graft with balloons is now routine prior
`to
`completion
`a
`angiography. If completion angiographyis satisfactory,
`duplex
`ultrasound scan is performed
`on the first postoperative day
`to
`on
`confirm aneurysm exclusion. Any unexplainedfinding
`duplex
`ultrasound is immediately examined with a
`spiral CT; otherwise
`follow-up CT are
`at 6 weeks, 6 months and then
`performed
`annually. Because a
`proximal leak without distal re-entry and
`outflow is associated with rupture of the aneurysm, correction
`with early intervention is performed.
`Twopatients required conversion to open operation; in both
`instances the delivery system failed. In one
`patient the graft
`was
`at the upper neck and at
`operation
`deployed successfully
`this attachment was found to be very secure. The proximal part
`was therefore left attached by the Gianturcostent,
`of this graft
`and the lower end was secured by conventional anastomosis at
`the aortic bifurcation.
`
`Persistent incomplete aneurysm exclusion (endoleak) contin-
`one at the superior neck and oneattheiliac
`ues in two
`patients;
`at the superior neck of one
`neck. A leak persists
`patient who
`sustained displacement of one iliac limb of his endoluminal
`to
`improve the seal at the superior
`prosthesis when angioplasty
`attempted. Attempted angioplasty via a brachial
`neck was
`was
`a
`severely debilitating stroke. His
`complicated by
`approach
`require intervention due to
`ischaemic leg did not
`immobility
`at rest. Where successful occlusion of
`and adequate perfusion
`the aneurysm was achieved and confirmed with a CT scan, no
`sofar.
`new endoleaks have been noted in the follow-up period
`was corrected with
`Oneleak aboutan iliac straight tube graft
`a covered stent extension piece. A subsequent correction pro-
`cedure was successful on one side but not on the contralateral
`side of the second patient with an iliac endoleak. A conservative
`was
`adopted dueto his
`failing health. The principal
`approach
`cause of leakage from the iliac limbs in bifurcate grafts
`was
`underestimation of the lengths of the commoniliac artery due
`on arterio-
`tortuosity and a foreshortened apparent length
`to
`gram. Curviplanar reconstructed images allowed for by the CT
`software have subsequently improved estimation of the correct
`length of the iliac limbs.
`A transient leak from the superior neck was noted in one
`su-
`patient where a
`was well lodged within an
`graft
`adequate
`perior neck, and this spontensously sealed within 1 week.
`We remain concerned about
`the safety and durability of
`‘endoleaks’ at the superior neck sealed by thrombosis. Endo-
`or
`leaks may occur where inadequate superior neck length,
`the neck, causes minimal contact
`at
`excessive angulation
`a
`between graft and true aortic wall. Although
`temporary seal
`may be obtained by apposition of the graft against the laminar
`thrombus, aneurysm exclusion is not achieved. In the above
`patient there wasat least 1.5 cm of contact
`length between the
`graft and the sound aortic neck, with no
`interposed thrombus.
`The transient endoleak was
`thought due to poor bedding of the
`self-expanding anchorage segmentof the graft against the aortic
`wall. We now
`routinely bed this segment with a soft latex
`balloon. The latex balloon is used rather than an
`angioplasty
`balloon becausethe relative rigidity of an
`angioplasty balloon,
`across an
`angulated neck,tends to cause traction on the
`applied
`and can cause
`displacement. Inten-
`inferior anchorage point(s)
`to monitor for recurrance of the endo-
`sive follow up is planned
`leak or continued aneurysm expansion.
`at
`limb occlusion has been discovered incidentally
`Graft
`patients. Limited exercise tolerance from co-
`follow up in two
`existent disease rendered both patients asymptomatic. However,
`this raises concerns about the compromisesituation of aorto-
`uni-iliac grafts. Here, the whole lower body is dependent upon
`a
`single outflow limb, and collaterals are
`to be suffi-
`unlikely
`cient to maintain viability of the legs in the event of occlusion.
`Both occlusions in our series occurred early in the development
`was modified to include self-
`of a bifurcate system. The system
`expanding ‘Z’ stents
`along the entire length of the iliac limbs,
`which should resist kinking and twisting.
`In the graft system described, the uncovered portion of the
`Gianturco stent
`anchoring the graft proximally in the aorta fre-
`quently extends above the renal artery origins.
`A fine wire may
`therefore cross one or both renal artery origins. Lawrenceet al.
`demonstrated no deterioration in renal function in the canine
`model with the use of the fine wire of this type of stent.'? Sig-
`was seenin three
`nificant deterioration of renal function (> 10%)
`patients in our series, each with significant pre-operative renal
`
`

`

`624
`
`GORDONETAL.
`
`failure. In two cases, a
`long superior neck allowed deployment
`of the Gianturco stent
`completely below the renal artery origin.
`In the third patient pre-operative stenting of a
`tight (> 60%)
`to preserve function ofthis
`renal artery stenosis was
`performed,
`the uncovered portion of the stent
`single kidney. Although
`extended abovethis renal artery origin, follow up on
`duplex and
`renal nuclear scans showedthat renal blood flow waswell pre-
`renal disease was
`served. Progression of the underlying
`thought
`to
`explain the deterioration in the patient’s renal biochemistry,
`an
`opinion sustained by the renal team
`responsible for his long-
`term care.
`A significant impact has not yet been seen on
`hospital stay
`(7.5 days average postoperative stay). Although disappointing,
`this is explained in part by the frail condition of the majority
`of patients selected. Almost all patients treated to date have
`exhibited a
`significant fever for 3-7 days. Blood cultures were
`universally negative. Patients were
`kept until afebrile, usually
`well past the date on which they considered themselves ready
`to result from thrombosis
`for discharge. This fever is thought
`within the aneurysm sac. The fever now has less influence on
`our decision for timing of discharge.
`were
`Minor complications
`infrequent (Table 1). Upper respi-
`ratory tract infection/basal atelectasis was
`radiologically evident
`patients, despite being frequently looked for as a
`two
`in only
`cause of the fever seen in the majority of patients post-
`potential
`woundinfection in an obesepatient healed
`procedure. One groin
`readily in the absenceof
`graft material externalto the arterial wall.
`to remain still during the acqui-
`It is necessary for the patient
`sition of digital subtraction angiography. Respiration therefore
`needs to be controlled. This can be achieved with a
`co-operative
`patient under epidural; otherwise general anaesthesia with
`relaxation, intubation and controlled ventilation is necessary.
`Spontaneousventilation via a
`laryngeal mask airway is unsuit-
`able when using digital subtraction angiography.
`removes many of the major
`stresses of
`Endovascular stenting
`an open procedure; the anaesthetist has only
`a
`to
`light
`provide
`level of anaesthesia. The cardiovascular stresses of cross-
`the aorta were not seen, although
`and
`clamping
`unclamping
`looked for during the brief periods of balloon inflation used to
`bed the graft in the aorta. Blood pressure manipulation is not
`required with the self-expanding stent, as
`to a balloon-
`opposed
`stent. The only major problem for the anaesthetist
`expanded
`was that the procedures
`were undertaken in an area
`containing
`sophisticated X-ray equipment. Reduced access and lighting,
`moving operating tables, and limited space combine to create
`as a ‘hostile environment’,
`what anaesthetists commonly regard
`emphasizing the need for the combination of high-quality
`imaging with the facilities of advanced operating theatres.
`were monitored in the Royal Perth Hospital
`All patients
`intensive care unit for the immediate postoperative period. All
`were returned to the vascular ward within 24h. As no
`patient
`we no
`intervention during this period,
`required significant
`longer plan intensive care observation as a routine.
`curve has been well demonstrated in this series.
`The learning
`All but one
`significant complication (Fig. 2) occurred in the
`first half of our series. Overall, 75% of patients had an uncom-
`plicated procedure and a further 10% had endoleaks corrected
`subsequently. In the second half of this series 95% had an
`were
`uncomplicated procedure. Modifications to the grafts
`effective in addressing technical failures associated with deliv-
`ery and short-term patency. The techniques and lessons learned
`on to new groupsin
`can be passed
`training programmes.
`
`The introduction of new
`minimally invasive techniques
`concerns
`recently, typified by laparoscopic surgery.
`prompted
`The debate in the UK has reached parliamentary level. Areas
`concern include:
`of particular
`(1) A procedure should not be modified to an inferior tech-
`to suit the method of access.
`nique
`(2) Unacceptable morbidity should not eventuate as each
`curve.
`surgeon goes through the learning
`(3) The procedure should remain cost-effective.
`(4) Indications for treatment should not be modified without
`justification.
`Studies evaluating the suitability of aneurysms for endolu-
`a
`> 5 cm in dia-
`minal techniques suggest that only
`minority
`meter have an inferior neck suitable for a
`straight tube graft."
`The prevalence of small aneurysms in screening studies exceeds
`that of aneurysms > 5 cm in diameter by
`ten times.'* Given that
`the incidence of rupture is rare in smal! aneurysms (< 5 cm)'*
`rates are slow in the vast
`majority,'’ those endol-
`and growth
`to smaller aneurysms cannotbe jus-
`uminal systems suited only
`tified on either medical or economic grounds. The system
`described allowed deployment of both straight tube and bifur-
`aneu-
`grafts, and was
`cate
`only used in clinically significant
`rysms from 5 to 9.5 cm in diameter (average 6 cm).
`Many important lessons were learnt during the development
`curve was due to
`period. Most of the
`of the learning
`morbidity
`ofthe graft, delivery system and imaging tech-
`the development
`niques, and should be largely avoidable for new surgeonslearn-
`ing under an
`‘apprenticeship’. Failure to seal at the superior
`neck appears to be a
`potentially lethal complication, perhaps
`increasing the risk of rupture.
`A tendency for the aorta to dilate with time in the perirenal
`some
`was noted by
`investigators,'* and dilatation
`segment
`> 10% wasseen in one of our
`It remains to be estab-
`patients.
`lished whether this will be a
`or
`self-limiting phenomenon,
`whetherit will progress to threaten the superior seal in an un-
`acceptable numberof patients. Also uncertain is whether the
`to continue to
`Gianturco self-expanding stent, with its ability
`to 40 mm in diameter at the anchorage barbs), will
`expand (up
`at the superior seal in this event than
`provide greater security
`stents such as the
`the fixed diameter of balloon expandable
`Palmaz (Johnson & Johnson, USA).
`use of the technique in patients
`at
`Results
`justify continuing
`higher than average risk for conventional surgery, but we con-
`to be the current
`sider open repair
`procedure of choice in those
`of average risk until] the durability and efficacy of the endolu-
`are proven in time.
`minal grafts
`
`ACKNOWLEDGEMENT
`
`We would like to thank William A. Cook Australia for the
`manufacture and supply of devices to our
`
`specification.
`
`REFERENCES
`1. Parodi JC, Palmaz JC, Barone HD. Transfemoralintraluminal
`graft implantation for abdominal aortic aneurysm. Ann. Vasc.
`Surg. 1991; 5: 491-9.
`2. May J, White G, Waugh R ef al. Treatment of complex
`a combination of intraluminal
`abdominal aortic aneurysms by
`and extraluminal aorto-femoral grafts. J. Vasc. Surg. 1994, 9:
`924-33.
`
`

`

`SELF-EXPANDING ENDOLUMINAL GRAFT SYSTEM
`
`625
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`10.
`
`11.
`
`Dake MD, Miller DC, Semba GP er al. Transluminal place-
`ment of endovascularstent-grafts for the treatment of descend-
`ing thoracic aortic aneurysms. N. Engl. J. Med. 1994; 331:
`1729-34,
`Chuter TA, Green RM, Ouriel K. Transfemoral endovascular
`aortic graft placement. J. Vasc. Surg. 1993; 18: 185-97.
`Dubost C, Allan M, OeconomosN. Resection of an aneurysm
`of the abdominal aorta. Re-establishment of continuity using
`a
`preserved humanarterial graft; with results after 5 months.
`Arch, Surg. 1952; 64: 405-81.
`Rutherford RB (ed.). Vascular Surgery, 3rd edn. Sydney:
`W. B. Saunders Co., 1989; 11: 907.
`Creech O Jr. Endo-aneurysmorraphy and treatment of aortic
`aneurysms. Ann. Surg. 1966; 164: 935-46.
`Johnston KW, Scobit TK. Multicenter prospective study of
`non-ruptured abdominal aortic aneurysms. Population and
`operative management. J. Vasc. Surg. 1988; 1: 69-79.
`Henderson A, Effeney D. Morbidity and mortality after
`abdominal aortic surgery in a
`population of patients with high
`cardiovascularrisk. Aust. N.Z. J. Surg. 1995; 65: 417-20.
`Cormier F, Veroux P, Fichelle J et al. Surgery of aneurysms
`over 80 years of
`of the sub-renal abdominal aorta in patients
`age. Presse. Med. 1993; 27: 1861-4.
`Golden MA, Whittemore AD, Donaldson MC, Mannick JA.
`
`Selective evaluation and management of coronary artery
`disease in patients undergoing repair of abdominal aortic aneu-
`rysms: A 16 year experience. Ann. Surg. 1990; 212: 415-20.
`Lawrence-Brown MMD, Hartly D, MacSweeney STRet al.
`The Perth endoluminal bifurcated graft system: Development
`and early experience. J. Cardiovasc. Surg. (in press).
`Lawrence DD, Charnsanggarij C, Wright KC et al. Percuta-
`neous endovascular graft: Experimental evaluation. Radiology
`1987; 163: 357-60.
`Andrews SM, Cumig R, MacSweeney STR er a/. Assessment
`of feasibility for endovascular prosthetic tube correction of
`aortic aneurysms. Br. J. Surg. 1995; 82: 917-19,
`Scott RAP, Wilson NM, Ashton HA et ail. Is surgery necessary
`for abdominal aortic aneurysms less than 6 cm in diameter?
`Lancet 1993; 342: 1395-6.
`Brown MP, Ballard DJ, Hallet JV Jr. Prognosis of abdominal
`aortic aneurysms: A population based study. N. Engl. J. Med.
`1989; 321: 1009-13.
`MacSweeney STR, O’Meara M, Alexander C er al. High
`prevalence of unsuspected abdominal aortic aneurysms in
`or cerebro-
`patients with confirmed symptomatic peripheral
`vascular disease. Br. J. Surg. 1993; 80: 582-4.
`Parodi JC. Endovascular repair of abdominal aortic aneurysms
`and otherarterial lesions. J. Vasc. Surg. 1995; 21: 549-55.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`.
`
`