JACC VoL 22, No . 3September 1993 : 1,67-7-1

11111h Pressure

Stenosas : Early

Objectives . The aim of this study was to evaluate the efficacyand safety of high pressure balloons (17 to 20 atria, Blue Max,Meditech) to dilate branch pulmonary artery stenosis .

Background. The low success rote (50% to 60%) for angio-plns-,fy of branch pulmonary artery stenosis using low pressure,balloons is due primarily to the inability to eliminate the balloonwaist . Hence, higher hdinflon pressures may improve results .

Mellow& flemodynande and anglographic data from 52 pa-tients (0.3 to 34 .8 years old) who underwent high pressure balloondilation of branch pulmonary artery stenosis between October1990 and February 1992 were reviewed retrospectively, as weredata from previous low pressure dilations in these patients .Common diagnoses included tetralogy of Fallot (a = 9), tetralogyof Fallot with pulmonary atresia (n = 23), single ventricle (n = 8)and isolateOd congenital pulmonary stenosis (n = 7) . The 52patients had 72 vessels dilated . Criteria for success were a X50%

Branch pulmonary artery stenosis remains a difficult thera-peutic problem . Stenoses distal to the hilum of the lung aredifficult to reach surgically, and incomplete relief of moreproximal lesions may contribute to operative morbidity andmortality (1-3) . The success rate of balloon angioplasty,using low pressure balloons with a burst pressure of 4 to10 atm, has remained a disappointing 50% to 60% despitetechnical advances and reduced morbidity (4) . In our expe-rience, failure is most often due to inability to eliminate theballoon waist during dilation . We therefore evaluated thesafety and efficacy of high pressure balloons with a burstpressure of 17 to 20 atm (Blue Pdax, Meditech) .

Methodsstudy patients . Fifty-two patients underwent high pres-

sure balloon dilation of 72 branch pulmonary arteries during53 catheterizations at Children's Hospital, Boston betweenOctober 1990 and February 1992 . Patient age ranged from

From the Department of Cardiology, Children's Hospital and the Depart-ment of Pediatrics, Harvard Medical School, Boston, Massachusetts .

Manuscript received July 15, 1992 ; revised manuscript received February18, 1993, accepted March 2, 1993 .

Address for correspondence : Stanton B . Perry, MD, Department ofCardiology, Children's Hospital, 300 Longwood Avenue, Boston, Massachu-setts 02115 .

(D1993 by the American College of Cardiology

nary Artery

THOMAS L. GENTLES, MB, CHB, JAMES E . LOCK, MD, FACC, STANTON B .Boston, Massachusetts

ERRY, MD

increase in vessel diameter or a >20% decrease in right ventricmular to aortic pressure ratio .

Results. Of 36 vessels with previously unsuccessful low pres-sure balloon dilation, 23 (63%) were successfully dilated with highpressure balloons . Of the 36 remaining vessels, 29 (81%) weresuccessfully dilated with high pressure balloons. Factors associ-aled Mitt success were stenosis at a surgical onaslomosLs anddisappearance of the balloon waist with dilation. Aneurysmsdeveloped m three vessels . Complications occurred in sevenpatients (13%) ; in two patients the distal pulmonary artery wasperforated by a stiff guide wire, causing death in one .

Conclusions . Dilation of stenotic peripheral pulmonary arterieswith high pressure balloons improves the success rate of angio .plasty both in patients who have had unsuccessful dilation with alow pressure balloon and in those without previous attempteddilation .

(J Apia Coll Cordial 1993,22 :867-72)

867

0.3 to 34 .8 years (mean 7.2 :, 6) . Diagnoses includedtetralogy of Fallot after repair in 9 (17%), tetralogy withpulmonary atresia after right ventricle to pulmonary arteryconduit placement with -ir without ventricular septal defectclosure in 23 (44%), single ventricle in 8 (15%), isolatedperipheral pulmonary stenosis in 7 (13%) and othk;, diag-noses in 5 (109 ) .

Angiograrns, hemodynamic data and lung scans werereviewed retrospectively . Data from prior low pressuredilations in these patients (either during the same procedureor from a previous procedure at this institution) were alsoreviewed . Angiographic measurements were corrected formagnification by comparison with the known catheter diam-

eter .Indications. Balloon pulmonary artery angioplasty was

performed when there was near systemic right ventricularpressure, a marked decrease in lung flow as assessed aloneor in combination by radionuclide lung scan, hypertension inunaffected portions of the vascular bed or symptoms (4,5) .Initially we used a high pressure balloon only after failure ofa low pressure balloon ; as our experience increased, webegan to use high pressure balloons without prior low

pressure dilation .Procedures. A single vessel was dilated in 38 patients

two vessels in I I and four vessels in 3 . In 34 vessels, stenosiswas at the site of surgical anastomosis (nonfunctioning

0735-1097/931$6,00

GENTLES ET AL.ANGIOPLASTY FOR PULMONARY ARTERY STENOSIS

systemic-pulmonary artery shunt in 20 and outflow tractpatch or conduit anastomosis in 14) . The remaining vesselshad congenital branch pulmonary artery stenoses .

We followed a technique previously described for lowpressure balloon dilation of branch pulmonary arteries (4,5)with several modifications that take into account the differ-ing characteristics of the Blue Max high pressL re balloons .These balloons, designed for angioplasty of peripheral sys-temic vessels, have a stiffer shaft and longer balloon length(2 to 4 cm) than those of low pressure balloons ofcomparable diameter. Advancing these catheters to straddleru stenotic pulmonary artery is facilitated by a stiff guide wire(Amplatz Super Stiff, Cook) . Because of its length, theballoon may injure normal parts of the pulmonary artery oroutflow tract as it straightens during inflation . Careful posi-tioning of the distal end of the balloon in the largest possibledistal branch is important to avoid aneurysm formation . Thecatheter shaft is 7F, making the use of these catheters lessdesirable in infants, and the maximal balloon size is 12 mm,necessitating the use of multiple balloons in vessels >4 to5 mm in diameter . Consequently, six vessels in this serieswere dilated with two balloons and one vessel was dilatedwith three .

Although the high pressure balloon catheters were usu-ally introduced through the femoral vein, manipulation maybe easier from the subclavian vein . On advancing the cath-eter and wire to the pulmonary artery from the femoral vein,force is transmitted through an unsupported bend in the rightatrium. If the wire is not sufficiently stiff or if the cathetermeets resistance, the wire and catheter will buckle at thispoint. Conversely, if the wire and catheter are advanced tothe pulmonary artery from the superior vena cava, force istransmitted by a single bend against the diaphragmaticsurface of the right ventricle (Fig . 1). The subelavian ap-proach was used in seven patients (in three because ofdifficulties in positioning the catheter from the femoral veinand in four because femoral venous access was not possi-ble). Early in this series, as with low pressure dilations, wechose a balloon with a diameter three to four times the lesiondiameter but not more than two times the diameter of then artery. Catheter and vessel constraints sometimesmandated the use of smaller balloons. Inflation pressureswere increased incrementally until the balloon waist disap-peared or to a maximum of 22 atm .

Criteria success. Criteria for success were similar tothose published previously, that is, a z-:~50% increase inp ' ation diameter in the angiogram immediately afterangioplasty or a > decrease in the ratio of right ventric-ular to aortic pressure (4) . As most patients had multiplevessels dilated by high and low pressure balloons during thesame Procedure, lung scans were not useful in assessing theCOM

five success of low versus high pressure balloons .. Results are expressed as mean value

t SD. Univariate comparisons between successful andunsuccessful dilations were made by two-tailed Student ttests. Multivariate analysis was performed using stepwise

3ACC Vol. 22, No. 3September 1993:867-72

Figure . A, Catheter course from the subclavian vein to the rightpulmonary artery . On advancing the catheter, force is transmitted tothe diaphragmatic right ventricular wall (

ws). D, When thecatheter course is from the femoral vein to the right pulmonaryartery, force is transmitted toward the superior vena cava (arrows) .If the guide wire is not sufficiently stiff or if the catheter meetsresistance, the wire and catheter will buckle .

logistic regression (SPSS/PC+ version 4 .0 software, SPSS).A p value < 0 .05 was considered significant .

Results

The overall success rate was 72% (52 of 72 vessels) . Meandiameter increased from 3 .8 t 1.5 mm to 6.3 ± 2.8 mm (p0.0001) in all vessels (mean increase in diameter 87 ± 50% ;range 33% to 340%, median 75%) in successful dilations and9 t 12% (range 0% to 42%, median 0%) in failures (p =0.0001). In the 44 patients with an intact ventricular

.'; eptum

(or restrictive ventricular septal defect), right ventricularpressure decreased by 14 t 13% in patients who had

JACC Vol . 22, No . 3September 1993 :867-72

8E

PreLPB

PostLPB

Pre

post1-113 13

HPB

Figure 2 . Lesion diameter before and after low (LPB) and high(HPB) pressure balloon dilation in the 36 vessels that had previouslyundergone low pressure dilation . Burs indicate ± I SD .

successful dilations, compared with 0 ± 10% in failures (p =0.008) . Two patients with a <50% inercase in vessel diame-ter (33% Rnd 42%, respectively) had a >20% decrease inright ventricular pressure .

Previous low pressure balloon dilations . In 36 vessels withunsuccessful low pressure balloon dilation performed 0 to 69months (median 3 months) before high pressure dilation,mean diameter increased from 3 .6 to 3 .9 mm with lowpressure 1,alloon dilation and increased from 4 .1 to 6.2 mmwith high pressure balloon dilation (Fig . 2) . In all, 23 of these36 vessels (63%) were successfully dilated with a highpressure balloon . When these 36 vessels were further clas-sified into congenital (n = 21) and surgery-related lesions(n = 15), the success rate was 48% (10 of 21) and 87% (13 of15), respectively. Of the 36 vessels that had not previouslyfailed low pressure balloon dilation, high pressure balloondilation was successful in 29 vessels (81%) (65% 111 of 17] inthe congenital group and 95% [18 of 19] in the surgicalgroup) . Successful dilation of a surgery-related stenosis isillustrated in Figure 3 .

Other determinants of success. Lesion diameter beforedilation, patient age at dilation, dilation of an old surgicalanastomosis site (rather than congenital pulmonary steno-sis), disappearance of the balloon waist and the presence ofan angiographic tear after dilation were significant predictorsof success by univariate analysis . Neither the ratio ofballoon diameter to the diameter of the lesion before dilationnor balloon inflation pressure were predictive (Table 1), andthere was no relation between the increase in vessel diame-ter with dilation and the ratio of balloon diameter to lesiondiameter (Fig . 4) . When factors found predictive of successby univariate analysis were entered into a multivariate modelas independent variables with outcome as the dependentvariable, only the type of lesion (surgical) and the disappear-ance of the balloon waist with dilation were significantpredictors of success (Table 1)

Dhgnosis. Dilation was successful in all 9 vessels in pa-tients with single ventricle, 5 (83%) of 6 vessels in the "others"

GENTLES ET AL.ANGIOPLASTY FOP PULMONARY ARTERY STENOStS

869

Figure 3. Before (A) and after (B) successful dilation of a surges y-related lesion at the site of a Blalock-Taussig shunt that hadpreviously been ligaled . An intimal tear (arrow) is seen afterdilation .

group, 25 (71%) of 35 vessels in patients with tetralogy of Fallotand pulmonary atresia, 7 (70%) of 10 vessels in patients withtetralogy of Fallot alone and 6 (50%) of 12 vessels in patientswith isolated peripheral pulmonary stenosis .

Table 1 . High Pressure Balloon Angioplasty of Branch PulmonaryArteries : Factors Associated With Success

*Significant (p < 0.05) by multivariate analysis using a logistic regressionmodel. Data are presented as number of patients in group (%) or mean valueSD. NS = not significant (p =- 0 .05) .

Success(n = 52)

Failure(n = 20) p value

Dilation at suirgic,,.l 3 .e 31 of 52 (60) 3 of 20 (15) 0.Disappearance of balloon waist 26 of 52 (57) 4 of 20 (20) 0.007*

Intimal tear 37 of 52 (71) 6 of M (M 0.002Predilation diameter (mm) 4.2

1 .6 3 .2 ± IA 0.01

Age (yr) 8.3

6.5 4.3 ± 3 .3 0.01Balloon/lesion diameter 3 .0

0.9 3.0 ± 1 .1 NSBalloon inflation pressure (aim) 13.8±3 .1 14.5±2 .1 NS

870

3503001,

•

250200150

•

10050

•

0-50

0

GENTLES ET AL .ANGIOPLASTY FOR PULMONARY ARTERY STENGPIS

a

2

3

4

5

BalloonNessel Ratio

0

__jj

Flprat 4. Percent increase in vessel diameter plotted against ratio ofballoon diameter to predilation vessel diameter The wide range ofballoon to lesion diameter ratio reflects the wide differences incompliance seen in stenosed pulmonary arteries .

Aneurysms . Two patents had aneurysmal dilation distalto the dilated lesion . Both had small distal pulmonaryarteries . A third patient developed a transmural tear with

JACC Vol. 22, No. 3September 1993 :867-72

aneurysm formation at the site of dilation (Fig . 5). One ofthese patients has been restudied ; angiograms showed reso-lution of the distal aneurysmal dilation .

Complications. Complications were seen in seven proce-dures (13% of patients) . Two patients had pulmonary arteryperforation by the stiff end of a guide wire. In one, withmultiple congenital branch pulmonary artery stenoses andsystemic right ventricular pressure, a distal pulmonary ar-tery was perforated by the stiff end of an Amplatz guide wireduring manipulation of the high pressure balloon catheter .Dilation of the upstream vessel was successful, exposing theperforation to high pressure . The patient died of pulmonaryartery hemorrhage 6 h later. The second patient had a briefepisode of hemoptysis immediately after dilation with twoballoons, a guide wire, had perforated the distal pulmonaryartery . Three other patients developed hypotension requir-ing transient inotropic support and, in one case, cardiopul-monary resuscitation . One patient (with tetralogy and pul-monary atresia) developed complete heart block duringcatheter manipulation . This persisted during the hospitalperiod but had resolved at follow-up several weeks later. A

Figure 5. Before (A), during (B) and after (C) dilation of a proximalright pulmonary artery stenosis in a 19-year old patient with tetralogyof Fallot repaired with a right ventricular patch . The predilationdiameter was 10 mm. The lesion was dilated with three 12-mmballoons with dissection and aneurysm formation at the lesion site(arrow) .

JACC Vol . 22, No. 3September 1993 :80-72

final patient with tetralogy of Fallot, pulmonary atresia,multiple congenital branch pulmonary artery stenoses andsmall distal pulmonary arteries had two vessels dilated witha high pressure balloon and two vessels dilated with otherballoons . Angiograms immediately after dilation showed anintimal tear at the site oT the high pressure :;illations but nochange in vessel diameter . Because of marked asymmetry offlow by lung scan, the patient was recatheterized I weeklater . There was obstruction by intimal flaps at the site ofhigh pressure dilation and at the site of one other dilation .Stents were successfully placed in these vessels .

Resource utilization . All but 6 of the 52 patients wereadmitted to the hospital the day before the procedure, and allwere observed in the hospital at least overnight after theprocedure Fifteen patients were transferred to the cardiacintensive care unit from the catheterization laboratory wheretheir mean stay was 1 .7 t 1 .2 days (range I to 4 .5) . For the51 surviving patients, the total length of hospital stay afterthe procedure was 2 .3 ± 2 .4 days (range I to 10) .

Follow-up study. In seven patients who had 13 vesselssuccessfully dilated by high pressure balloon, follow-upcardiac catheterization was undertaken 3 to 9 months (mean5.5 ± 2 .3) after dilation . Of these 13 vessels, 10 were studiedangiographically at follow-up. Mean vessel diameter was5.7 ± 1 .7 nun (range 3 .5 to 8.0) compared with 6 .0 ± 1 .1 mm(range 4 .4 to 7.5) immediately after dilation . Restenosis(narrowing to a diameter near the predilation value) occurredin one vessel .

DiwusAonIn this series successful dilation of branch pulmonary

artery stenoses using newly available high pressure angio-plasty balloons was achieved in 72% of vessels overall and in80% of vessels that had not previously been dilated . This isan improvement on the largest reported series of low pres-sure balloon dilations using similar dilation techniques andcriteria for success (4) . More important, a 63% success ratewas achieved in the 36 vessels that had previously hadunsuccessful low pressure balloon dilation, suggesting thathigh pressure dilation represents a genuine advance in thetreatment of these difficult lesions . These results were ob-tained in a patient group that was preselected to be unfavor-able, in that many of the patients had been referred fromother institutions after unsuccessful dilation .

Determinants of success. Dilation of surperl, related le-sions was nearly always successful . Although other investi-gators (6,7) have found these lesions difficult to dilate withlow pressure balloons, Rothman et al . (4) suggested that theywere more readily dilatable than were congenital lesions .High pressure dilation presumably opens the fixed fibroticstenosis at the site of surgical anastomosis . However, con-genital lesions differ from postoperative lesions with intimaland medial hypertrophy from increased muscle and elastin(8) . In this series, similar balloon to lesion ratio and inIationpressures were used for the congenital and surgical groups .

GENTLES ET AL .ANGIOPLASTY FOR PULMONARY ARTERY STENOSRS

871

la the congenital group, results were worse than in thesurgical group, but they were better wit : ; high than with lowpressure dilations . This observation, combined with the factthat congenital lesions were less likely to be torn and oftenhad a persistent waist, suggests that even higher inflationpressures may be needed . Although our results with congen-ital lesions were less spectacular than those with surgicallesions, half of these vessels were dilated successfully afterfailure of dilation with low pressure balloons, offering animprovement in therapy for patients who often have fewtherapeutic options .

Technique. We generally start with a balloon to lesiondiameter ratio three to four times the diameter of a narrowedperipheral pulmonary artery . However, in this study themean ratio (3 .0) was smaller than that previously reportedfor low pressure balloon dilation (mean 4,0) (4) . Although insome cases a large balloon to lesion diameter ratio was used,we tended to use smaller balloons than we would for lowpressure dilation for several reasons : 1) We were concernedthat high pressure dilation with a larger balloon mightdamage the distal artery, and 2) in some cases, using a largerhigh pressure balloon would have entailed double-balloondilation . The reason for the success of smaller, high pressureballoons may be that higher inflation pressures are necessaryto dilate the subtle balloon waist . Although high pressureballoons are rigid and track poorly, they may prove safer inthe long run by allowing a reduction in balloon size andthereby reducing of aneurysms in distal vessels . Al times weused an inflation pressure higher than that recommended bythe manufacturen because we were concerned that higherinflation pressure could increase the risk of balloon ruptureor vessel damage, or both, we used these higher inflationpressures only when other treatment option , had beenexhausted. Balloon rupture did not occur in any of thesecases .

As noted jxeviou~ly (4), intimal tears were associatedwith svccessful dilation . They were seen angiographically inonly 71% of successful di :otions, but they are probablypresent after almost all successtui dilations (8,9) . In unsuc-ce ,,sful dilations with an intimal tear, subsequent low pres-sure dilation with a larger balloon may be successful, as itwas in one patient in this series .

Complications . The Blue Max high pressure balloon shaftis stiffer than the low pressure balloon shaft and manipula-tion of the catheter within the heart and pulmonary arteriesis difficult . Several of the complications in this series prob-ably resulted from these features . The one death was causedby pulmonary artery hemorrhage after perforation of thedistal vessel by the stiff end of a guide wire and successfuldilation of a more proximal lesion . Although we prefer toposition the soft end of the guide wire in a distal vessel,difficulties manipulating the Blue Max catheter frequentlynecessitated positioning of the stiff end distally . In the otherpatient with pulmonary hemorrhage, dilation was performedwith two balloons . During manipulation of the catheters, aguide wire perforated a distal pulmonary vessel . It is likely

872

GENTLES ET AL .ANGIOPLASTY FOR PULMONARY ARTERY STENOSIS

that the increased technical demands of double-balloondilation contributed to this complication . It is also likely thatthe use of a stiff guide wire and difficulties passing thecatheter through the heart contributed to the development ofcomplete heart block in one patient . Modified extra stiffguide wires with short distal floppy lengths may reduce theincidence of such complications .

Other complications occurred at rates similar to thoseseen with low pressure angioplasty (4) . Recently, we used anew high pressure balloon (Ultrathin, Meditech) . The shaftis smaller (5F) and less stiff than that of the Blue Max and, inour preliminary experience, we found it easier to manipu-late. Because of these advantages, we now use it in prefer-ence to the Blue Max balloon for smaller lesions (themaximal balloon size is 10 mm), inflating first to low pressureand then to higher pressures, if needed .

R utilization. The mean length of hospital stayafter the procedure was 2 days, with only three patientshospitalized for > 1 week, indicating minimal morbidity inthe majority .

Follow-up. Although the rate of restenosis (10%) is sim-ilar to that described with low pressure balloon dilation (4),only 2 of successfully dilated vessels have been followedangiographically . Because many of these patients have com-plex cardiac defects, further follow-up angiography will beundertaken as part of ongoing management .

Conclusions . High pressure balloon angioplasty offers animprovement in treatment for patients with branch pulmo-nary stenosis, particularly for those with lesions at sites ofsurgical anastomosis. Even in the most difficult lesions,congenital stenoses that had been unsuccessfully treated

JACC Vol . 22 . No. 3September 1993 :867--72

with low pressure balloon dilation, the success rate wasalmost 50% . The Blue Max balloon catheter was designedfor peripheral systemic angioplasty . Because many compli-cations were related specifically to physical properties of thecatheter, one would expect the incidence of complications todecrease as high pressure balloons and extra stiff guide wiresdesigned for pulmonary artery dilation become available .

: fet°enees1 . Alfieri 0, Blackstone E13, Kirklln JW, Pacifico AD, Bargeron LM Jr .

Surgical treatment of tetralogy of Fallot with pulmonary atresia . J ThoracCardiovasc Surg 19711 ;76:321 .-35.

2. Kirklin JW, Blackstone EM, Kirklin JK, Pacifico AD, Aramendi J,Bargeron LM . Surgical results and protocols in the spectrum of tetralogy ofFallot . Ann Sing 1983;198:251--61 .

3. Mayer JE Jr, Helgason tt, Jonas RA, et al . Extending the limits formodified Fontan procedures. J Thorac Cardiovnsc Surg 198( ;92:1021-8.

4. Rothman A. Perry SB, Keane IF, Lock JE. Early results and follow-up ofballoon angioplasty for branch pulmonary artery stenoees . I Am CollCardiol 1990 ;15 :1109-17,

5 . Lock JE . Catheter intervention : balloon angioplasty . In : Lock JE . KeaneIF, Fellows KE, eds . Diagnostic and Interventional Catheterization inCongenital Inca Disease . Boston: Martins NijhoL 1987:104-7 .

6. Rocchini AP, Kveselis D, MacDonald D, Crowley D, Snider R,Rosenthal A . Use of balloon angioplasty to treat peripheral pulmonarystenosis. Am I Cardiol 1954 ;54 :1069-73 .

7 . Ring JC, Bass JL, Marvin W, et al. Management of congenital stenosis ofa branch pulmonary artery with balloon dilation angioplasty : report of 52procedures . J Thorac Cardiovasc Surg 1985 ;90:35-44 .

8 . Edwards BS. Lucas RV, Lock JE, Edwards JE . Morphological changes inthe pulmonary arteries after percutaneous balloon angioplasty for pulmo-nary arterial stenosis . Circulation 1985;71 :195-20I .

9 . Lock IF, Niemi T, Enzig S . Amplatz K, Burke B, Bass JL . Transvenousangioplasty of experimental branch pulmonary artery stenosis in newbornlambs. Circulation +981 ;64:886-93 .