J. Paltiel,

Harriet

MD

Reta

C. Rupich,

#{149}

PhD

H. George

#{149}

Kiruluta,

MD

Enhanced Detection of Vesicoureteral Reflux in Infants and Children with of Cyclic Voiding Cystourethrography’

M

Cyclic voiding cystourethrography (VCUG) was prospectively evaluated to determine its ability to demonstrate

vesicoureteral

reflux

children initially

whose VCUG negative. The

assessed

the effect

(VUR)

results authors

of change

in

were also

in the

patient’s of VUR. younger negative

position on the detection Seventy-seven children than 3 years of age, with results from a VCUG study performed while they were supine, underwent a second cycle of bladder filling after they were placed prone (group 1). Sixty-five children who were also younger than 3 years of age and had negative results from an tial VCUG examination performed in the usual supine position underwent a second cycle of bladder filling, which was also performed with the patient supine (group 2). VUR occurred in three children (4%) in group 1 and in eight (12%) in group 2. Most children (68.8%) in the two groups combined had grade II reflux. Cyclic VCUG increased detection of VUR, which led to a change in cmical treatment. Prone positioning did not enhance detection of VUR to the same degree as did multiple studies performed with the patient supine. mi-

Index terms: Bladder, abnormalities, 83.848 Bladder, radiography, 83.123 #{149} Children, genitourinary system, 80.21 #{149}Infants, genitourinary system, 80.21 #{149}Ureter, reflux, 82.85 #{149}Urine,

reflux, Radiology

I

From

82.85, 83.85 1992;

the

184:753-755

Departments

of Radiology

(H.J.P.)

and Urology (H.G.K.), Montreal Children’s Hospital and McGill University, Montreal; and the Department of Pediatrics, University of Cmcinnati College of Medicine, Cincinnati (R.C.R.). From the 1991 RSNA scientific assembly. Received February 3, 1992; revision requested March 4; revision received April 8; accepted April 13. Address reprint requests to H.J.P., Department of Radiology, Children’s Hospital

Medical cinnati, C

Center,

Elland

OH 45229-2899. RSNA, 1992

and Bethesda

Ayes, Cm-

children who develop renal scarring after urinary tract infection (UT!) have vesicoureterab reflux (VUR), and higher grades of VUR are associated with an increase in parenchymal scarring (1,2). Detection of

reflux

OST

in infants

and

young

children

is particularly important, since they are more likely to develop renal scarring with UTI than are older children (3).

Retrograde voiding cystourethrography (VCUG) is widely used as a means of detecting VUR and has been used as the standard of comparison with newer methods of assessment for reflux such as radionuclide cystography (4) and ultrasonography (5). However, it is surprising that few reports on the reliability of VCUG to depict reflux have been published and that, to our knowledge, variations in technique that may affect VCUG results have not been studied systematically (6-12). In a recent retrospective study by Jequier and Jequier (12), the reliability of VCUG to demonstrate VUR was assessed in 177 patients who underwent two cycles of bladder fibbing and voiding at the same sitting and in 30 patients who underwent three cycles. In the subset of their study popubation with negative results from an initial VCUG examination, 3% of patients demonstrated reflux after two cycles; an additional 4% showed reflux after three cycles. The children in their study ranged in age from 4 days to 15 years, and all underwent multipbe bladder fillings because of a strong clinical suspicion of VUR. We performed a prospective evaluation of the ability of cyclic VCUG to enhance the detection of reflux in infants and young children in whom VCUG was initially negative for VUR, since the long-term implications of missed reflux in this patient popubation are potentially more serious than they are in older children and adolescents.

Use

We also explored the role of patient position in the detection of reflux, since observations in humans have suggested that postural changes may influence the detection of VUR (7), but no systematic investigation of this variable has, to the best of our knowledge, been clinically evaluated. The fact that VUR in the dog is prevented in the supine position, since the pressure of the bladder compresses the distal ureters, is well documented in the literature. ably demonstrated

placed (13,14).

However, when

in a lateral

decubitus

MATERIALS

AND

In the September

15 months

younger

than

ative

VUR dogs

is reliare

position

METHODS

between

July

1989

and

1990, 142 infants

results

and children of age who had neg-

3 years

from a VCUG

examination

when they were supine (first bladder filling) were studied a second time at the same sitting, either supine or prone. VCUG was performed after catheteriza-

tion with 15%

a 5- or 8-F infant

solution

feeding

of diatrizoate

tube.

A

meglumine

and diatrizoate sodium was administered by drip infusion from a height of approximately 1 m above the table after the solution was warmed to body temperature.

When

the bladder

was full, voiding

oc-

curred around the catheter. A sustained void was necessary for the first cycle to be considered complete. The catheter was left

in place Spot

for the second radiographs

cycle.

(105 mm)

tamed during intermittent Routine views consisted oblique that

radiographs included

were

ob-

fluoroscopy. of right and left

of the

full bladder

the ureterovesical radiographs of the

junction,

one or two urethra during voiding, and radiographs of the bladder and each renal fossa after voiding. During the second cycle, only fluoroscopy

without less

reflux

radiography was

noted.

was performed If reflux

Abbreviations: UTI = urinary VCUG = voiding cystourethrography, vesicoureteral reflux.

was

unseen,

tract infection, VUR

=

753

additional

radiographs

document

its site and

occurred, classification

Reflux

were

obtained

extent.

to

When

reflux

it was graded according to the scheme of the International

Study

in Children

(grades

I-V)

The additional studies prolonged roscopy by an extra 10-30 seconds. col was approved by the Institutional view

Board

The

of our

hospital.

first 63 consecutively

tients

with

(15).

fluoProtoRe-

negative

referred

results

from

paa VCUG

examination performed with the patient supine underwent a second study, which was performed while the patient was prone. On review of the radiographs of this group, we decided that all subsequent studies

would

both

renal

because

of

reflux

with

The

next

tients VCUG

also the

after

of missing

with

the

negative

results

amination fants

from

study

(group

Thus,

studies

Patients

in group

in

by a prone

65 underwent

(group

days to 35.5 months 42 (54%) were boys.

77 in-

VCUG

followed

1), whereas

supine

ex-

bladder

underwent

position,

last

VCUG

position.

children

The with

a second

in the prone and

supine.

I ranged

in age

(mean, Patients

6.2 months); in group

7

from

of patient

tion

in groups

filling

cycle.

position

I and A

in reflux

2 during

x2 statistic

2

calcu-

bated to evaluate the significance of the difference in the proportion of patients with

reflux

in the

two

groups.

The effect of multiple fibbing cycles on the demonstration of reflux was analyzed separately for the prone and supine groups with a Fisher exact test. A K statistic was

used

ment,

to evaluate

beyond

between fibbing

the

that

due

the results cycles

degree

of agree-

to chance

alone,

group.

RESULTS In group 1 (patients were first supine and then prone), reflux occurred in three children (4%) into five ureters. Two of three children were boys, and two previously had a UTI. The mean age was 2.25 months. In group 2 (patients were supine both times), reflux occurred in eight children (12%)

on

11 ureters. boys,

and

the

second

Five seven

fibbing

of eight of eight

cycle

into

children were

were studied

(P

rate

at which

contrast

1 when

during

the study

on sequential examinations weeks or months apart or even from moment to mo-

performed (17,18)

supported

the

results

of

ment

(6,12,19).

that

Factors

determine

minute-to-minute variability are speculative but may include variations in urine production and flow rate,

first and second VCUG examinations was -0J2, or greater than that of chance for group 2; however, the result for group 1 was 0.04, which is very close to 0, or chance.

changes

Table 3 summarizes the number of ureters and the reflux grade identified with cyclic VCUG in groups 1 and 2. Most patients had grade II reflux; grade I reflux was identified in 25%.

Most scarring recent tively,

in intravesicab pressure, and muscle tone at the ureterovesicab junction (20). To our knowledge, the only pubbished study to date of the reproducibibity of VCUG findings from multiple-cycle examinations performed at the same sitting is that of Jequier and Jequier (12), in which VCUG was

evaluated

against

quence of two der fibbing and age of agreement

children who develop after a UTI have VUR. articles (1,2) document, a 75% and 80% sensitivity

renal Two respecof

and

second

negative

thought

otic

development

at greater

of renal

risk

scarring

are older children, so that larly important to identify younger patients (16).

VCUG has been used dard means of identifying

754

#{149}

years; however, in procedural

center to center the occurrence These

for

the

than

it is particureflux in

there technique

are variafrom

that may influence or demonstration variables

include

a se-

cycles

of blad-

voiding.

The

between

cycles

percentthe

first

for patients from

an

with

initial

VCUG

Our results of patients

in also

high first

rates of agreement and second cycles.

Clinically, however, ureter with reflux

detection of a of grade II or

higher

the

will

lead

therapy

up. Thus, tively ureters

medical as the stanVUR for

itself

results

demonstrate between

to be

during

or three

examination was high. a more selective group

reflux as a marker for renal scarring. In addition, higher grades of reflux are associated with an increase in parenchymal scarring (1). Children younger than 5 years of age are

of reflux.

vs

and

Fisher exact test analyses by indicating that the level of disagreement between the results of the

cycle

2 (prone

group

x2 and

ing

Radiology

prone

K

many tions

1 and

for

were .245). statistic

=

because they previously had a UT!. The mean age was 5 months (Table 2). When the effect of position on reflux was analyzed for the second fillin groups

volume,

severity

increased

DISCUSSION

of the first and second

in each

When

.059).

=

(a) catheterization technique (retrograde vs suprapubic puncture vs intravenous administration of contrast material); (b) type of catheter; (c) type of contrast material; (d) concentration,

sugsu-

cantly

the

detec-

then

the results rate with

group 2 when the second cycle was compared with the first (P = .006). The rate of reflux was not signifi-

A

the second

was

reflux (P

material is delivered to the bladder; (c) patient positioning; (f) imaging sequences; and (g) recording methods (7). Even when identical techniques are used, reflux may vary in

children

2).

ranged in age from 6 days to 32.5 months (mean, 10 months); 36 (55%) were boys. The reasons for referral for both groups are listed in Table 1. A 2 x 2 table was used for statistical analysis

positioning

x2 test,

groups 1 and 2 were analyzed separately for the effect of the second filling cycle (supine vs prone and supine vs supine), a Fisher exact test showed a significantly higher reflux rate in

a supine a second

patients

a supine

underwent

supine

two

patient

with a a higher

pine

pa-

from studied

referred

supine) gested

subtle

alone.

referred

results were

of

voiding

monitoring

65 consecutively

also

views

possibility

14 consecutively

fibbing

spot

the second

fluoroscopic

with negative examination

time,

the

include

fossae

to bong-term

and

the

search

small number is probably

treatment

The most

significance I reflux circumstances,

follow-

for even of refluxing justifiable,

a rebasince

of a patient

positive test result beads tially different follow-up pared with patients with findings at VCUG. grade

antibi-

prolonged

with

a

to substanwhen comnegative

of detection is debatable,

of

since,

it is not

in

treated

September

1992

3.

Ransley PG. Risdon lae, intrarenal reflux

phritis. In: Hodson Reflux nephropathy. 4.

5.

and

almost

always

resolves spontaneousby. However, in a child with recurrent UTI and demonstration of grade I VUR, many physicians would employ low-dose antibiotic prophylaxis. In our series of patients who demonstrated reflux during the second VCUG cycle, only 25% had grade I VUR. It would have been ideal to have been able to precisely control the volume of contrast material instilled in the bladder during each fiffing cycle. However, we were unable to tightly regulate this the children

variable, since most of in our study were not

toilet trained. Hence, voiding occurred spontaneously without vobuntary control. Our data reveal that prone positioning of the patient does not enhance reflux detection to the same extent as do cyclic studies performed with the patient supine. The reasons for this difference are purely speculative but might be related to the fact that contrast material is heavier than urine and will layer anteriorly in the bladder when the patient is prone. Thus,

it is conceivable

that

reflux

of

nonopaque urine into the ureters might not be detected with fluoroscopy. Second, the anatomy of the human bladder differs from that in the dog, which is a long, mobile organ (13).

The

bladder supine

compressive

on the distal dog is probably

effect

ureters not

of the

in the duplicated

in the human. Finally, it is possible that additional reflux might have occurred in the initial study group of 63 patients but was not recognized fluoroscopically or documented radiographically. We believe that further evaluation of cyclic VCUG is warranted, particularby in the very young child and in other patients at high risk for renal damage secondary to VUR, such as those with myebodyspbasia. Children with a history of recurrent UTIs and negative results from a VCUG examination, as well as those with a welldocumented episode of pyebonephritis, might also benefit from a cyclic study.

Cyclic

radionucide

cystogra-

phy tive

may prove to be a good to VCUG, as it provides

ous

monitoring

and

detection

of vesicoureteral

only

after

10.

second

11.

12.

13.

15.

16.

dictor with

of reflux nephropathy urinary tract infection.

Nielsen

JB.

The

of examination

clinical

producing

(edito-

intrinsic

cys-

significance

of the

bladder

pressure

and bladder volume in reflux and reflux nephropathy. Scand J Urol Nephrol Suppl 1989; 125:9-13. Jequier 5, JequierJC. Reliability of voiding cystourethrography to detect reflux. AJR 1989; 153:807-810. Cass AS, Lenaghan D. The influence of

Newman L, Bucy dence of naturally

JG, McAlister WH. mcioccurring vesicoureteral

reflux in mongrel 8:354-356.

dogs.

Lebowitz Smellie

mnvest Radiol

RL, Olbing H, Parkkulainen JM, Tamminen-M#{246}bius TE.

Cardiff-Oxford

Sequelae 17.

18.

Bacteriuria

of covert

Ky, Inter-

Pediatr

Study

bacteriuria

1973;

Group.

in school

girls. Lancet 1978; 1:889-893. Gelfand MJ, Strife JL, Hertzberg VS. Lowgrade vesicoureteral reflux: variability in grade on sequential radiographic and nuclear cystograms. Clin Nucl Med 1991; 16: 243-246. Young DL, Treves ST. Potter CS. Natural

history of low grade vesicoureteral reflux: evaluation with radionuclide cystography. Presented at the 30th annual meeting of the Society for Pediatric Radiology, To19.

ronto, Cremin

May 30 to June 4, 1987. BJ. Observations on vesico-ureteric reflux and intrarenal reflux: a review and survey of material. Clin Radiol 1979; 30:607-621. Theobald RJJr. Changes in ureteral peri-

staltic activity induced by various Neurourol Urodyn 1986; 5:493-504.

References

2.

method

national reflux study in children. Radiol 1985; 15:105-109.

20.

Bisset GS III, Strife JL, DunbarJS. Urography and voiding cystourethrography: findings in girls with urinary tract infection. AJR 1987; 148:479-482. HellstrOm M,Jacobsson B, M#{226}rildS,Jodal U. Voiding cystourethrography as a pre-

Vol

posture on the occurrence of vesicoureteral reflux. Invest Radiol 1965; 2:523-529.

der

1.

radiology.

rial). IntJ Pediatr Nephrol 1980; 1:2-3. Hannerz L, Wikstad I, Broberger 0, Aperia A. Influence of diuresis on the degree of vesicoureteral reflux. Acta Radiol Diagn 1983; 24:395-399. Kogan SJ, Sigler L, Levitt SB, Reda EF, Weiss R, Greifer I. Elusive vesicoureteral

reflux

blad-

filling. In conclusion, cyclic VCUG increased detection of VUR in our study population, which in turn bed to a change in medical treatment. Prone positioning of the patient did not enhance demonstration of VUR to the same degree as did multiple bladder fillings performed with the patient in the supine position. #{149}

in pediatric

reflux in children with normal contrast tograms. J Urol 1986; 136:325-328.

14.

substantially

the

9.

reflux

3. Basel, Switzerland: Karger, 1970; 41-42. Friedland GW. The voiding cystourethrogram: an unreliable examination. In: HodsonJ, Kincaid-Smith P. eds. Reflux nephropathy. New York: Masson, 1979; 91101. Hodson CJ. Micturating cystography: an

unassessed

alternacontinu-

decreased gonadal radiation (4). A recent study by Pozderac et al (21) comparing the findings from singleand double-filling cyclic radionuclide cystography performed for evaluation of VUR in 25 children has yielded promising results. VUR to the level of the pelvicaliceal system occurred in 13 patients; in two of them, reflux was identified

8.

J, Kincaid-Smith P. eds. New York: Masson,

in children. AJR 1982; 138:1033-1036. Nogrady MB, DunbarJS. The technique of roentgen investigation of the urinary tract in infants and children. In: Kaufmann

HJ, ed. Progress 7.

The renal papilchronic pyelone-

1979; 126, 133. Conway JJ, King LR, Belman AB, Thorson T Jr. Detection of vesicoureteral reflux with radionucide cystography. AJR 1972; 115: 720-727. Kessler RM, Altman DH. Real-time sono-

graphic 6.

RA. and

21.

Pozderac RV, Becker CJ, Reitelman Kuhns LR. Comparison of single

two-stage radionucide for evaluation of reflux 1990;

stimuli. C, and

cystography (RNC) (abstr). J Nucl Med

31 :893.

in children AJR 1989; 152:

801-804.

Volume

184

#{149} Number

3

Radiology

#{149} 755