Late complications after femoral artery catheterization in children less than five years of age

Late complications after femoral artery catheterization in children less than five years of age L l o y d M. Taylor, Jr., M D , * Russell T r o u t m a n , M D , * Phillip Feliciano, M D , * Victor Menashe, M D , * * Cecille Sunderland, M D , * * and J o h n M. Porter, M D , *

Portland, Ore. Fifty-eight children who underwent diagnostic femoral artery catheterization before 5 years of age, from 5 to 14 years before the study, were randomly selected from approximately 300 surviving patients undergoing diagnostic femoral artery catheterization at our institution during the interval. Each patient underwent vascular laboratory segmental pressure and waveform examination and arterial duplex scanning, as well as lower extremity bone length radiographs, which were considered positive if the catheterized leg was ---1.5 cm shorter than the opposite leg. Thirteen children who had only venous catheterization served as controls. No arterial abnormalities were present in the control patients (mean ankle/brachial index, 1.01). Arterial occlusion was present in both limbs of five patients who had bilateral diagnostic femoral artery catheterization and in 14 limbs of 51 patients who had unilateral diagnostic femoral artery catheterization. Thus arterial occlusion was present in 33% of patients (19 of 58) and in 37% of limbs (24 of 65). The mean ankle/brachial index in the catheterized limbs was 0.79. Leg growxh retardation was present in four limbs (8%) of 51 children undergoing unilateral diagnostic femoral artery catheterization and in one (8%) control patient. The inverse relationship between ankle/brachial index and leg growth retardation was significant (R = 0.47,p < 0.0005). Only one patient had symptoms of arterial occlusion (clandication), and one patient had symptoms of leg growth retardation (gait dimtrbance). We conclude that arterial occlusion is common after diagnostic femoral artery catheterization in children less than 5 years of age, but that excellent collateral supply prevents leg growth retardation and/or symptomatic arterial insufficiency in most children. Children having diagnostic femoral artery catheterization should be monitored for arterial occlusion, and those detected should be monitored for leg growth retardation. Arterial repair in the absence of clinical symptoms appears unwarrented. (J VAsc SURG 1990;11:297-306.) Diagnostic femoral artery catheterization (DFAC) is a well-established technique with an extremely low complication rate in adults. ~ When occlusive complications o f DFAC occur, recognition is usually straightforward as most patients manifest both ischemic symptoms in the involved extremity

From the Division of Vascular SurgeryDepartment of Surgery,+ and the Section of Pediatric Cardiology Department of Pediatrics,~ Oregon Health SciencesUniversity. Supported by a grant From the Oregon Chapter of the American Heart Association, and in part by grant no. RR00334 from the General Clinical Research Center Branches, Division of Research Resources, National Institutes of Health. Presented at the Forty-third Annual Meeting of The Societyfor Vascular Surgery, New York, N.Y., June 20-21, 1989. Reprint requests: Lloyd M. Taylor,Jr., MD, Divisionof Vascular Surgery, Oregon Health SciencesUniversity, 3181 S.W. Sam Jackson Park Rd. OPll, Portland, OR 97201-3098. 24/6/17099

as well as absence o f peripheral pulses. Infants and young children must occasionally also undergo DFAC, primarily for evaluation o f congenital heart disease. Although the occurrence o f arterial occlusion after DFAC in children is well recognized; the reported incidence o f this complication varies widely from a low o f 3% 2 to as high as 40%. 2 This wide range reflects the rarity with which symptoms o f severe ischemia result from arterial occlusion in young children, in contrast to the situation in adults. Despite the rarity of'acute symptomatic limb ischemia resulting from arterial occlusion in children, concern regarding thi~ complication is appropriate as limb growth retardation may occur in extremities with arterial occlusion. This phenomenon was first described in humans by Harris ct al. 4 and systematically studied by Currarino and Engle. 5 As is true o f arterial ocdusion, the reported incidence o f leg growth retardation after DFAC varies widely, from 297

298

Journal of VASCULAR SURGERY

Taylor ¢t al.

Table I. Ages of patients in study DFA C

Age at thne of procedure Age at time of study Follow-up in-

16 (0-55)

Controls

p < 0.05

3 (0-11)

110 (69-180)

118 (42-143)

95 (54-139)

115 (38-142)

terval All values are in months, and represent mean values (range).

a low of 2.8% reported by Rosenthal et al. 6 to a high of 86% as described by Bassett et al.7 Clinicians responsible for the care of infants and young children undergoing DFAC thus face a number of decisions: first, whether to rigorously pursue a diagnosis of arterial occlusion after DFAC; second, whether to recommend immediate and/or late repair of arterial occlusion when detected; and third, whether to monitor children with arterial occlusion for leg growth rctardation. The use of Dopplerderived ankle pressure measurements provides a sensitive and accurate method permitting detection of arterial occlusion after DFAC, as first described by Barnes et al. 8 in 1974. Radiologic comparison of lower extremity bone length ("orthoroentgenography" or "scanography") 9 is used to quantitate leg growth retardation. Precise information concerning the late incidence of both artcrial occlusion and leg growth retardation after DFAC is essential to assist clinical decision making in the care of infants and children after DFAC. Recently we examined 58 children chosen at random who underwent DFAC at less than 5 years of age 5 to 14 years previously to determine the incidence of late arterial occlusion and leg growth retardation. The results of this examination form the basis for this report. METHODS

Records of surviving children who underwent DFAC from 1974 to 1983 at the Oregon Health Sciences University Pediatric Cardiac Catheterization Laboratory were reviewed. Children less than 5 years of age at the time of DFAC were eligible for study. Subjects from this list were selected at random and contacted by telephone through parents or guardians and invited to participate in the study. Funds were provided for travel expense to protect against sample bias. Most patients were examined at the time of a regularly scheduled clinic follow-up visit. The study was fully approved by the Oregon Health Sciences University Human Studies Institutional Review Board.

The record of each child was reviewed to determine the following information: ( I ) D F A C site and technique, (2) use of anticoagulant (heparin) during DFAC, (3) whether arterial occlusion was detected, and (4) whether it was treated at the time of DFAC, and (5) the outcome of such treatment. Each subject then underwent noninvasive vascular laboratory lower extremity evaluation and lower extremity bone length radiographic examination. The vascular laboratory examination consisted of palpation of femoral, popliteal, and ankle pulses; recording of analog Doppler waveforms at these levels; and measurement of Doppler-derived pressure indexes at high-thigh, above-lmee, below-knee, and ankle levels with a 10 cm wide cuff. Anlde pressure responsc to trcadmill walking (1-1/2 mph at 0% grade for 5 minutes) was also recorded.I° The highest arm pressure was used as a reference value for pressure index calculations. Children with abnormal values underwent duplex scamling of involved areas to confirm the presence or absence of arterial occlusion. n Arterial occlusion was diagnosed on the basis of noninvasive vascular testing when the ankle/brachial pressure index (ABI) was ~0.90,12'1s and either palpable pulses were absent or analog Doppler waveforms were abnormal in the extremity, and duplex scanning documented occlusion or hemodynamically significant stenosis. Patients who had undergone bilateral interruption of the subclavian artery for use in shunt procedures and patients who had undergone coarctation repair in whom at lcast one lower extremity did not have normal hemodynamic values were excluded from the study. Lower extremity bone length radiographs were performed according to the technique of Green et al.9 and interpreted by a radiologist who was unaware of the results of the vascular laboratory examination. The comparative lengths of femur and tibia were recorded. The lower extremity bone length examination was considered abnormal if the total extremity bone length in the affected leg was 1.5 cm shorter than the contralateral leg. This value was chosen based on the mean age of the children (9 years) and growth charts TM as a value that would clearly result in a discrepancy of 2.0 cm at maturity. Patients who underwent bilateral DFAC were excluded from this part of the study. Identical vascular laboratory and leg radiographs were performed in a group of similary aged children with congenital heart disease who underwent diagnostic femoral venous catheterization during the same tinae period. These patients served as controls. All patients and/or their parents or guardians were

V o l u m e 11 Number 2

Complications after catheterization in children 299

February. 1 9 9 0

Uncatheterized Legs 1.3

1.2

Catheterized Legs

Control Legs

•2-------;

OOOO

0~OO

O~

8

O

O

O80

OaO

o8o

o 0

1.1

Ne Ankle Brachial Pressure Index

o O

O0 o

O

1.0

og

8

o

ogo

~o

0

8 O

0.9

&

Ip 0.8 ¸

l" ql

0.7

"1"

• Leg with arterial occlusion o Leg without arterial occlusion

0.6

Fig. 1. Distribution of ABI values for catheterized legs, control legs, and mlcatheterized legs.

interviewed by clinical nurse specialists in vascular disease. Symptoms of lower extremity vascular insufficiency including intermittent claudication, ischemic rest pain, and ischemic ulceration were specifically sought. The incidence of arterial occlusion, leg growth retardation, heparin use, and symptoms were compared by use of the Student t test. The relationships between ABI, age at catheterization, and leg length discrepancy were examined by univariate and multivariate analysis. The Kolmogorov-Smimov twosample test was used to compare the distribution of leg length discrepancy and ABI in study subjects and in controls. RESULTS Thirteen control subjects were studied. The mean ABI for these patients was 1.14 (0.95 to 1.20). Fiftyeight of approximately 300 surviving children who underwent DFAC from 1974 to 1983 were studied. The mean age of these children at the time of DFAC was 16 months (range, 0 to 55 months). The mean age at the time of the study was 110 months (range, 69 to 180 months), giving a mean follow-up interval

of 95 months (range, 54 to 139 months), The ages, follow-up intervals, and age at the time of study of the control children and the children who underwent DFAC are seen in Table I. The control patients were significantly younger at the time of DFAC (Student t test, p < 0.05). The technique of DFAC was uniform in all children and consisted of percutaneous insertion of catheters as described by Lurie, Armer, and Klatte is in 1963. Seven children tmderwent bilateral DFAC and 51 underwent unilateral DFAC. Arterial occlusion was present in both limbs of five children who underwent bilateral DFAC and in 14 limbs of the 51 children who underwent unilateral DFAC. Thus the incidence of arterial occlusion was 33% of patients (19 of 58) and 37% of limbs (24 of 65). No patient had arterial occlusion in a limb that had not been catheterized. The mean ABI in the legs with arterial occlusion was 0.79 (range, 0.60 to 0.90). The mean ABI in the uncatheterized legs of patients who underwent DFAC was 1.03 (range, 0.90 to 1.30). The mean ABI in the legs of the control patients who underwent venous catheterization only was 1.14 (range,

Journal of VASCULAR SURGERY

300 Taylor et al.

Ankle-Brachial

0 0

o

Index vs. Leg Length Difference

o

0

0

0

o

oo o

o

o

0

oO ~

o

_ o

o

0

o

o o

o o

^

[

O

- -

o

O

-

o

o

O

O

r = 0.447 -2 0.50

I 0.75

i 1.00

i 1.25

I

1.50

Ankle-Brachial Index Fig.. 2. Distribution of leg length difference in patients undergoing DFAC and in control pauents.

Table II. Age, follow-up, and heparin use in patients with and without arterial occlusion

Age at DFAC in months

(_ SE)

Follow-up interval in months (m SE) Heparin used (%)

Patients with arterial occlusion

Patients without arterial occlusion

17.64 (_+ 3,81)

15.44 (_+ 2.1)

104.23 (_+ 4.31)

88.69 (+ 3.7)

1 (5%)

9 (24%)

0.93 to 1.40). The ABI figures for those three groups are seen in graphic form in Fig. I. The distribution of ABI values for the legs that underwent DFAC is significantly different from the distributions of ABI values for uncatheterized legs and for control legs. (Kolmogorov-Smirnov two-sample test p < 0.01 and p < 0.01, respectively). The age at the time of DFAC and follow-up interval in patients with arterial occlusion did not differ significantly from that in patients without arterial occlusion. The use of heparin during DFAC was more common in patients without arterial occlusion (9 of 37, 24%) than in those with AO (1 of 2i, 5%) but this difference was

not significant (p < 0.10). These data are seen in Table II. Leg growth retardation was present in four (8%) of the 51 children who underwent unilateral DFAC (mean leg length difference = 1.6 cm, range, 1.5 to 1.7 cm), all of whom had arterial occlusion as defined above. No leg asymmetry was present in any of the seven patients who underwent bilateral DFAC. Leg growth retardation was also present in one (8 %) control patient who did not undergo DFAC and did not have arterial occlusion (leg length difference 1.6 cm). The difference in length of the two legs in catheterized patients is compared to the difference in length of the two legs in controls in Fig. 2. These two distributions were not significantly different (Kolmogorov-Smimov two-sample test, p > 0.10). The regression analysis of leg growth retardation as a function of ABI is seen in Fig. 3. (R = 0.47). The association between leg growth retardation and ABI was significant by univariate regression analysis (p < 0.005). Because of the difference in age at the time of DFAC between subjects and controls, the relationship between leg growth retardation and ABI was examined by multiple regression analysis with

Volume 11 Number 2 February 1990

Complications after catheterization in children 301

Difference in Leg Length Patients 2.0

Undergoing DFAC

Patients with no DFAC

Z~

1.5

z~ coo

Catheterized Leg Shorter

1.0 oo

0.5

(cm)

& eo oo oOo o

o oo

Oo

o

0.5

Catheterized Leg Longer 1.0

z~ Patient with Leg growth retardation

ooo

1.5 &

Fig. 3. Relationship between ABI in catheterized leg and leg length difference.

age at the time of DFAC included. Anlde/brachial index remained a significant (p < 0.0005) predictor of leg growth retardation, whereas age was not (p = 0.41). An abnormality of arterial circulation was recognized immediately after DFAC in 10 patients. The abnormality recorded on the chart was absence of distal pulses in each patient. Nine of these patients had no treatment, and at the time of the study five had normal findings (ABI in the catheterized leg = (mean, 1.07, range 1.0 to 1.3). Four of these patients with no treatment had arterial occlusion at the time of the study (mean ABI = 0.73 range, 0.6 to 0.9). None of these children had leg growth retardation. The tenth patient with arterial occlusion recognized at the time of DFAC underwent femoral thrombectomy with restoration of pulses at the time. This patient had arterial occlusion at the time of study (ABI = 0.72) and also had leg growth retardation (leg length difference = 1.5 cm). Symptoms of arterial insufficiency were reported at the time of this study by only one patient. This child underwent DFAC at age 5 months. Arterial occlusion was recognized by absent pulses at the con-

clusion of the procedure and was not treated. At thc time of study, at age 8 years and 3 months, she complained of intermittent claudication, which was confirmed by findings of an ABI of 0.6 in the affected leg and an abnormal ankle pressure recovery time after treadmill walking in the noninvasive vascular laboratory. This child did not have leg growth retardation (leg length difference = 0.8 cm). One child had a gait abnormality attributed to leg growth retardation. This child had DFAC at age 4 months. No arterial occlusion was recognized at the time of DFAC. When studied at age 6 years 3 months she had ABI = 0.72 in the affected leg and a leg length discrepancy of 1.7 cm, the highest recorded in the study. This child did not have claudication. DISCUSSION Originally, arterial catheterization in infants was accomplished by direct arterial exposure, with passage of catheters through surgically created arteriotomies. After the catheterization, the arteriotomies were repaired by suture, or, especially in small infants, the arteries were ligated. The striking ability

302 Taylor et al.

of infants to compensatc for arterial occlusions resuiting from such procedures is exemplified by the study of Vlad et al. 16 in which 542 catheterizations were followed by absent peripheral pulses in one third of patients, only one of whom had symptoms of limb ischemia. Similarly, Vengsarkar and Swan 17 noted absent peripheral pulses in 31% of 125 procedures "most of which" returned to normal within 3 days. They recognized no patients with symptomatic ischemia. The study by Bassett et al. r documented an incidence of objectively determined (oscillometry) arterial occlusion of 86% (24 of 28) at 1 to 8 years of follow-up in a group of children most of whom had ligation of the superficial femoral artery after catheterization. No mention was made of symptoms of arterial insufficiency. The technique of percutaneous arterial catheterization introduced by Seldingcr in 1953,18 was modified for use in infants and children by Lurie et al. in 1963.15 These authors noted frequcnt absence of peripheral pulses after the procedure (39%) but observed that 97% of patients had normal pulses by 2 weeks after catheterization) Although pulse return after initial absence has been interpreted as return to normality, the few studies performed with objective criteria have confirmed the presence of persistent arterial occlusion after DFAC in a large percentage of infants. Jacobsson et al. 19 studied 33 children at a mean interval of 5 years after percutaneous DFAC with oscillometry and found arterial occlusion in 36%, most of whom had palpable pulses. Arterial occlusion with abundant collaterals was confirmed by arteriography in five of the children. Bloom et al. 2° and Hawker et al. 21 similarly noted the presence of palpable pulses and lack of ischemic symptoms in children with objectively documented arterial occlusion after percutaneous DFAC. The unique study of Mortensson 22 in which children who underwent DFAC were subsequently examined (mean 3.1 years later) angiographically demonstrated arterial occlusion or high-grade stenosis in nine of 44 arteries (20%), only one of which was associated with symptoms or physical findings of arterial ischemia. The present study objectively documents arterial occlusion in 19 of 58 patients undergoing DFAC. This incidence of arterial occlusion is quite comparable to that observed in previous studies described above, as is the finding that most children with arterial occlusion in the present study had palpable peripheral pulses in the affected limb, and only one had symptoms suggestive of limb ischemia. Available evidence thus confirms a remarkable ability of chil-

~Iournal of VASCULAR SURGERY

dren to tolerate arterial occlusion and compensate through the mechanism of collateral circulation. Given these consistent findings, a conservative policy toward detection and or treatment of arterial occlusion after DFAC in infants seems initially justified and is recommended by some surgeons. 2a However, several experienced authorities continue to recommend aggressive efforts to diagnose and correct arterial ocdusion after DFAC in all children regardless of age. 24'z~ Others recommend an intermediate approach, suggesting surgery for those children in whom "simple thrombectomy" will suffice while reserving "complex procedures" for patients who are older and larger with a higher probability of success. 26 Still other authors recommend operation for children with occlusions proximal to the common femoral artery, and heparin therapy without operation for those with more distal occlusions.la The basis of such recommendation for treatment is the hope of preventing leg growth retardation, first recognized in the upper extremity after shunt operations that interrupted the subclavian artery. 4,s Studies intended to evaluate the incidence of leg growth retardation after DFAC in infants differ markedly from one another with reference to the definition of leg growth retardation, as well as methods used to determine its existence. Bassctt et al. 7 used x-ray orthoroentgenography 27 to assess leg length and found a shortened leg (which they defined as >3 mm difference) in 24 of 28 children. They based this definition of leg growth retardation on a simultaneously studied group of seven control children none of whom had differences greater than 3 mm. In contrast, Rosenthal et al. 6 exan~ined leg length by tape measure in 250 children who underwent DFAC and found leg growth retardation >8 mm (arbitrarily chosen) in only seven patients. Nonc of these legs had arterial occlusion as determined by oscillometry. Jacobsson et al. 19 found a decrease in circumference of the calf in 11 of 29 children, one of whom also had a length discrepancy of 1.5 cm. Hawker et al. 21 found no difference in leg length (defined as _+5%) in 42 patients. Bloom et al. 2° reported operations on three patients with external iliac occlusions and leg length discrepancies ranging from 1.5 to 2.5 cm. Surprisingly, little information is available regarding normal discrepancies between leg lengths in growing children. The only large scale study ever performed of leg lengths in children, that of Anderson et al., 28 does not address the subject of asymmetry. Several studies in adults, the largest of which

Volume 11 Number 2 February, 1990

are those of Rush and Steiner,29 who found some discrepancy in 71% of"normal" U.S. Army recruits, and that of Cox, 3° who documented leg length differences of up to 2.2 cm in healthy individuals, clearly indicate that some degree of leg asymmetry is the rule. It is interesting to note that no techniques currently used to assess leg length whether direct measurement or any of the various radiology methods has reproducible accuracy greater than _+1.0 cm. 31 In addition to a lack of clear definition of normal variation, the nature and severity of clinical symptoms resulting from mild to moderate degrees of leg length discrepancy are poorly defined. Abnormalities of gait obviously may occur as compensation for severe discrepancies and are well recognized.32 Widespread speculation exists that such gait abnormalities may result in arthritic and degenerative changes in the hips, knees, and spine, 33as but to date, no objective evidence has been collected to substantiate these hypotheses, al,a6 Given the significant vagaries in definition of normal values, in our ability to accurately measure discrepancy, most authorities reserve treatment of leg length discrepancy to those >2 cm in total length, at which point gait disturbance becomes common. 31 In children, recommendations for treatment may be made based on the calculation of the anticipated discrepancy at maturity by use of one of several methods) 4

In the present study leg length discrepancy was clearly related inversely to ABI in the catheterized leg, by both univariate and multivariate analysis confirming the tendency of arterial occlusion to produce leg growth retardation. Most differences recorded were slight however, as has been noted in previous studies. 6'7'19'21Four children in the present study had leg growth retardation as defined by leg length difference ---1.5 cm, a value chosen based on the mean age of the subjects (9 years), which is anticipated to produce a discrepancy >2.0 cm at maturity. One only child had symptoms questionably related to leg growth retardation at the time of study. Considering the above, it appears unlikely that any more than a small percentage of children with arterial occlusion after DFAC will ever have leg growth retardation of sufficient magnitude to require treatment. For those few children so affected, there is evidence that delayed arterial repair results in at least partial correction of the growth abnormality. Bloom et al.20 reported nearly complete resolution of leg growth retardation (mean 2.0 cm to mean 0.5 cm) in three children ages 8.5 years treated by arterial

Complications after catheterization in children

303

grafting. Our own experience includes a similar c a s e , a7 and similar cases have been reported by others.38 Thus it appears unlikely that children who suffer arterial occlusion after DFAC will ever become symptomatic either from arterial insufficiency or from leg length discrepancy. Cortsidering the less than perfect published results of arterial repairs in small children13'23-2swe recommend against immediate operative repair in infants and children, except in those rare instances when viability of the limb is threatened as indicated by complete absence of ankle Doppler signals and markedly delayed or absent distal foot capillary refill. Careful follow-up of infants with no symptoms after DFAC seems prudent. Vascular laboratory examinations will readily identify those with arterial occlusion who, in our opinion, should undergo radiographic determination of bone lengths at age 4 to 5 years. Those with discrepancies > 1.0 cm should have yearly follow-up, and arterial repair should be performed if it is calculated that leg growth retardation of >2.0 cm will be present at maturity. Optimism regarding both the long-term patency of such surgery as well as beneficial effect on leg growth retardation appears warranted. In contrast, we can find no evidence from this study or from those which have preceded it to justify prophylactic arterial repair in the absence of clinical symptoms and/or significant leg growth retardation. The authors acknowledge expert statistical support from Gary Sexton, PhD a~adDavid Wilson, MS.

REFERENCES

1. Ross RS. Arterial complication. In: Brannwald E, Swan H, eds. The cooperative study on cardiac catheterization. Circulation 1968;37[supp111I]:3%41. 2. Kirkpatrick SE, Takahashi M, Petry EL, Stanton RE, Lurie PR. Percutaneous heart catheterization in infants and children II. Prospective study of results and complications in 127 consecutive cases. Circulation 1970;42:1049-56. 3. Freed MD, Keane JP, Roscnthal A. The use ofheparinization to prevent arterial thrombosis after percutaneous cardiac catheterization in children. Circulation 1974;50:565-9. 4. Harris AM, Segel N, Bishop JM. Blalock-Taussig anastomosis for tetralogy of Fallot. A ten to fifteen year follow-up. Brit Heart I 1964;26:266-74. 5. Currarino G, Engle ME. The effects of ligation of the subclavian artery on the bones and soft tissues of the arms. l Pediatr 1965;67:808-11. 6. Rosenthal A, Anderson M, Thompson S)', Pappas AM, Tyler DC. Superficial femoral artery catheterization; effect on extremity length. Am J Dis Child 1972;124:240-2. 7. Bassett FH III, Lincoln CR, KingTD, Canent RV. Inequality

304

8.

9.

10. 11.

12.

13. 14. 15.

16.

17.

18.

19.

20.

21.

22.

Taylor et aL

in the size of the lower extremity following cardiac catheterization. South Med J 1968;61:I013-7. Barnes RW, Petersen JL, Krugmire RB, Strandness DE Jr. Complications of percutaneous femoral arterial catheterization; prospective evaluation with the Doppler ultrasonic velocity detector. Am J Cardiol 1974;33:259-63. Green WT, Wyatt GM, Anderson M. Orthoroentgenography as a method of measuring the bones of the lower extremities. J Bone Joint Surg [Am] 1946;28:60-5. Baur GM, Zupan TL, Gates KH, Porter JiM. Blood flow in the common femoral artery. Am J Surg 1983;145:585-8. Nicholls SC, Kohler TR, Martin RL, Neff R, Phillips DJ, Strandness DE Jr. Diastolic flow as a predictor of arterial stenosis. J VASC SURG i986;3:498-501. Yao JST, Hobbs JT, Irvine WT. Ankle systolic pressure measurements in arterial disease affecting the lower extremities. Br J Surg 1969;56:676-9. Hanigan DP, Keifer TJ, Schuler IL Ryan TI, Castronuovo JJ. Experience with iatrogenic pediatric vascular injuries. Ann Surg i983;198:430-42. Moseley C. A straight line graph for leg length discrepancies. Clin Orthop 1978;i36:33-40. Lurie PR, Armer RM, ICIatte EC. Percutaneous guide wire catheterization: diagnosis and therapy. Am J Dis Child 1963; 106:189-95. Vlad P, Hohn A, Lambert EC. Retrograde arterial catheterization of the left heart. Experience with 500 infants and children. Circulation 1964;29:787-93. Vengsarkar AS, Swan HJC. Arteriotomy for cardiac catheterization and angiocardiography in infants and children. Mayo Clin Proc 1962;37:619-26. Seldinger SI. Catheter replacement of the needle in percutaneous arteriography: a new technique. Acta Radiol 1953;39:368-71. Jacobsson B, Curlgren LE, Hedvall G, Sivertsson R. A review of children after arterial catheterization of the leg. Pediatr Radiol 1973;1:96-9. Bloom JD, Mozersky DJ, Buckley CJ, Hagood CO. Defective limb growth as a complication of catheterization of the femoral artery. Surg Gynecol Obstet 1974;138:524-6. Hawker RE, Palmer J, Bury RG, Bowdler JD, Celermajer JM. Late results of percutaneous retrograde femoral arterial catheterization in infants. Br Heart I I973;35:447-9. Mortensson W. Angiography of the femoral artery following percutaneous catheterization in infants and children. Acta Radiol Diag i976;17:581-93.

DISCUSSION Dr. R i c h a r d D e a n (Winston-Salem, N.C.). As quoted in this paper and given in the presentation, femoral artery catheterization in children is a frequent source of thrombosis, and the literature holds a wide variety of viewpoints concerning the frequency o f sequelae and necessity of their correction. I think it is important to point out one bit of philosophy in the context that we as vascular surgeons abhor the presence o f an occluded vessel and feel compelled to be aggressive about opening it. As in many other areas of

Journal of VASCULAR SURGERY

23. Smith C, Green RaM. Pediatric vascular injuries. Surgery, 1981;90:20-31. 24. Shaker !J, White II, Signer Rd, Golladay ES, Hailer JA Jr. Special problems of vascular injuries in children. I Trauma I976;16:863-67. 25. O'Neill JA Jr. Traumatic vascular lesions in infants and children. !n: Dean RH, O'Neill JA Jr, eds. Vascular disorders of childhood. Philadelphia: Lea and Febiger, I983:181-93. 26. Perry MO. Iatrogenic injuries of arteries in infants. Surg Gynecol Obstet 1983;157:415-8. 27. Bell IS, Thompson WAL~ Modified spot scanography. AIR I950;63:915-8. 28. Anderson M, Messner M, Green W. Distribution of lengths of the normal femur and tibia in children from one to eighteen years of age. J Bone Joint Surg [Am] 1964;46:1197-202. 29. Rush W, Steiner H. A study of lower extremity length inequality. AJR I946;56:616-27. 30. Cox WC. On the want of symmetry in the length of opposite sides of persons who have never been the subjects of disease or injury to their lower extremities. Am J Med Sci 1875;69: 438-9. 31. Moseley CF. Leg length discrepancy. In: Lovell Winter, ed. Pediatric orthopedics. 3rd ed. New York: CF Moseley, 1989: [In press]. 32. Mahar R, Kirby R, MacLeod D. Simulated leg-length discrepancy: its effect on mean center-of-pressure position and postural sway. Arch Phys Med Rehabil 1985;66:822-4. 33. Gofton J, Trueman G. Studies in osteoarthritis of the hip II. Osteoarthritis of the hip and leg length disparity'. Can Med Assoc J i971;104:791-9. 34. Kujata U, Fiberg O, Aalto T, et al. Lower limb asymmetry and patellofemoral joint incongrivence in the etiology of knee exertion injuries in athletes. Int J Sports Md 1987;8:214-20. 35. Gofton J, Persistent low back pain and leg length disparity. J Rheumatol I985;12:747-50. 36. Grundy P, Roberts C. Does unequal leg length causes back pain? A case-control study. Lancet 1984;2:'256-8. 37. Rubenstein RA, Edwards JM, Taylor LM Jr, Porter JM, Beals RIC Limb growth recovery following correction of femoral artery, occlusion. J Bone Joint Surg [Am]. (In press.) 38. IClein MD, Coran AG, Whitehouse WM Jr, et al. Management of iatrogenic arterial injuries in infants and children. J Ped Surg 1982;17:933-9.

vascular surgery and other disciplines of surgery, this common sense is frequently untested and when tested is found to be invalid. Certainly that is the case with this premise. In the discussion of their manuscript, which I also commend, I was somewhat surprised to find that one of the outlandish recommendations to be somewhat empirically aggressive about attempts at restoring patency in small infants was one of mine. In a book that we wrote on vascular disease in childhood, we did recommend an aggressive approach toward reconstructive procedures in small children.

Volume 11 Number 2 February 1990

I would only hasten to add that we are not so naive, however, and an aggressive approach in a neonate is frequently composed of frequent visits to the neonatal intensive care unit for observation. Those o f you who have tried to operate on little children at 8, 6, 8, 10, 15 pounds will quickly recognize how small the vessel becomes once it is manipulated. I am not surprised at all that the vessel thrombosis rate is above 30% and even would have suspected that it would be higher. They have made a number of observations that are worth reiterating again. One that was not pointed out in the presentation but is in the article was the high correlation between the frequency of thrombosis at the time of catheterization when no heparin was used at the time of angiography. There was a correlation, as they stated, between the ABI and leg length discrepancies as well as the presence of occlusion, not surprisingly. The wide variation in limb length regardless of the patency of the vessel or the presence or absence of previous catheterization, I think, is an important point and points out the lack of standardization of the data in the literature and the wide variation of normal anatomy. Limb growth retardation is exceedingly uncommon, as documented in this presentation, and in contrast to those of us who have suggested a relatively aggressive approach in the past relative to restoring patency to prevent this late sequelae. I think this paper is important in placing in front of us some informed common sense when one recognizes the low likelihood of success of restoration of flow in small infants. I concur with the authors' premise that in the absence of significant ischemia, empiric operative intervention is unjustified. I would like to ask two questions: At what age do they believe that their observations become invalid? Obviously, this is basically in the first year of life. The m~an age of their group is about 14 to 16 months. When one meets a 20-year-old or S-year-old, does one equally continue to observe this? My biases would be that when one gets beyond the first year of life, that the likelihood of success in restoration of flow is high enough to justify a much more liberal use of simple thrombectomy. My only other question is, I do not know if their experience would provide us an answer, but do they have any special techniques that they employ when managing these very, very young children who indeed because of the severity of ischemia require operative intervention? Our success has not been any greater than the rest of the world as relates to managing s,mall children, and I wonder if they have any special techniques to improve the patency rate. Dr. Lloyd Taylor#We thank Dr. Dean for his kind comments about our paper. Dr. Dean asked at what age we recommend empiric operation of children with arterial occlusion. Our experience parallels his. By the time children reach 5 or 6 years of age, their arteries are of sufficient size that they can be operated on with confidence, and the patency results of these procedures are g0bd. Our experience with operations in tiny infants is extremely limited and we do not have any special techniques to improve patency in these children.

Complications after catheterization in children

305

Dr. D. Preston Flanigan (Orange, Calif.). The authors are to be commended on their retrospective attempt to define the late results in young children with arterial occlusion. Most authors would agree that these injuries are seldom limb threatening, but many reports have shown that limb growth retardation can be significant. In our prospective study, which we reported to the American Surgical Association in 1983 on 76 such children, the incidence of limb length discrepancy was 23% and ranged from 0.5 to 3 cm in unoperated children having a mean age of only 31 months. The ABI in our children after injury was much lower than the ABI reported by the authors in late follow-up. For these data to be compatible, we would have to assume that collateralization stabilizes or reverses limb growth retardation. Do the authors have any evidence that this may be occurring? It is interesting that despite a mean ABI of 0.79 in this study, all children except one with arterial occlusion were asymptomatic. Although treadmill testing was described in the methods, the results of such testing were not reported in the manuscript. What were the results of these tests? Such an index in active adults likely would be associated with intermittent claudication and certainly would be limiting in athletics. The authors have recommended against surgery in the absence of limb-threatening ischemia, indicating that careful observation followed by delayed revascularization should be adequate to treat those few children in whom significant growth retardation occurs. Our experience would dictate a slightly different approach. A case can be made for the routine monitoring of vascular status at the time of catheterization in light of the high incidence of arterial injury. We believe the high incidence, and sometimes severe nature, of limb growth retardation dictates the need for an operative approach in chiildren with arterial occlusion except in those with high operative risk. Further rationale for this approach is based on the safety, simplicity, and high success rate of femoral artery thrombectomy and repair, and the fact that delayed revascularization is not always possible and is often unsuccessful in reversing growth retardation. It is important to realize that femoral artery catheter injuries are probably the most benign type of arterial injuries in children and that other types of arterial occlusion as a result of problems such as umbilical artery catheters, complications of surgery, trauma, or spontaneous thromboembolism often behave differently. Extrapolation of the author's recommendation,; to the management of these severe problems may not be appropriate. I congratulate the authors on a carefully performed study and agree that their suggested method of management is certainly reasonable. However, the significant risk of growth retardation and/or ischemia symptoms in untreated children contrasted with the high success rate and low risk of operative therapy support a surgical approach in children who are good risks. Dr. Taylor. Dr. Flanigan asked if we had any information regarding stabilization of limb growth abnormality

306

]ottm± of VASCULAR SURGERY

Taylor et al.

with the passage of time and the development of collateral circulation. Obviously we do not. We studied these children at one point after their cardiac catheterization and will attempt to learn more about those few children who have limb growth retardation as time passes. There is a relative lack of information regarding normal values for leg length discrepancy in children. There has been only one large study in the United States regarding normal limb growth progression in children, and this study did not address the issue of asymmetry. Several large studies of Army recruits indicated that differences in length of 2 to 2.2 cm between the two legs may be entirely normal.

So we have to be a little bit careful with our definition of limb growth retardation. Dr. Flanigan also asked about the results of treadmill testing, All these children underwent treadmill testing, and interestingly, there was no change in ankle pressure with treadmill walking in these children despite duplex scan documented external iliac and common femoral occlusion. We agree totally with Dr. Flanigan that these results should not be generalized to the treatment of other more severe arterial injuries, which may result in significant ischemia and obviously mandate repair.

B O U N D VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the Jou~'~AL OF VASCULARSURGERY for 1990 are available to subscribers only. They may bc purchased from thc publisher at a cost of $54.00 ($68.00 international) for Vol. 11 (January to June) and Vol. 12 (July to December). Price includes shipping charges. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volumc. The binding is durablc buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Circulation Fulfillment, The C.V. Mosby Company, 11830 Wesflinc Industrial Drive, St. Louis, MO 63146-3318, USA. In the United States call toll free: (800)325-4177, ext. 351. In Missouri call collect: (314)872-8370, ext. 351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular JOUI~X:ALsubscription.