Donald L. M iller, M D #{149} Richard Chang, M D #{149}W illiam T.

W ells, RN
#{149} Barbara A. Dowjat, RN #{149} Ruth M . M alinosky, RN #{149} John L. Doppm an, M D
Intravascular Contrast M edia:
Effect of Dose on Renal Function
607
Effect of contrast m aterial dose on
clinically evident change in renal
function was studied prospectively
in 200 exam inations requiring intra-
venous or intraarterial adm inistra-
tion of contrast m aterial. All pa-
tients were adequately hydrated.
Blood urea nitrogen and serum cre-
atinine were m easured before and
after the procedure. Ionic and non-
ionic contrast agents were used. To-
tal dose of contrast m aterial ranged
from 30 to 530 m L (m ean, 237 m L).
There was no tendency to give
sm aller doses to patients with pre-
existing renal im pairm ent and no
relationship between total dose and
patient age. No consistent clinical
effect on renal function was dem on-
strated with increasing dose, regard-
less of whether ionic or nonionic
agents were used. In adequately hy-
drated low-risk patients with pre-
dom inantly norm al initial renal
function and within the dose range
studied, there does not appear to be
any consistent clinical change in re-
nal function with increasing con-
trast m aterial dose.
Index term s: Angiography, complications.
96.448 . Angiography, contrast media #{149} Con-
trast media, comparative studies #{149} Contrast me-
dia, toxicity . Kidney. failure, 81.44
Radiology 1988; 167:607-611
I From the Diagnostic Radiology Depart-
ment, Clinical Center, National Institutes of
Health, Bldg 10, Rm 1C660, Bethesda, M D
20892(D.L.M ., R.C., W .T.W ., BAD., R.M .M .,
J.L.D.), and the Department of Radiology,
Georgetown University M edical Center, W ash-
ington, DC (D.L.M ., R.C., J.L.D.). From the 1987
RSNA annual meeting. Received October 12,
1987; revision requested November 10; final re-
vision received January 12, 1988; accepted Jan-
uarv 21. Address reprint requests to D.L.M .
RSNA, 1988
A common teaching in angiogra-
phy is that the total dose of con-
trast material administered in one sit-
ting should not exceed 2 mL/lb (300
mL for a 150-lb patient) or 4 mL/kg
(280 mL for a 70-kg patient) (1,2).
M anufacturers of contrast media of-
ten suggest similar maximum doses
of contrast material in their package
inserts. M any endocrine localization
studies require more contrast materi-
al than the suggested maximum dose
for an adequate examination, and we
have routinely exceeded these doses
for years, sometimes by almost a fac-
tor of two, without incident.
It has been our clinical impression
that, in most patients, contrast mate-
na! can be administered in doses con-
siderably higher than those usually
considered to be safe, without result-
ing in clinically important renal toxic
effects. This belief led us to conduct a
prospective study of the relationship
between contrast medium dose and
clinical renal dysfunction.
PATIENTS AND M ETHODS
W e prospectively studied 230 examina-
tions of inpatients that required intraarte-
na! or intravenous administration of con-
trast material in the special procedures
suite of the National Institutes of Health
(NIH) Clinical Center. W e did not in-
dude patients who underwent renal an-
giography or renal angioplasty as part of
a procedure or patients who had received
intravenous injections of contrast materi-
al elsewhere in the radiology department.
W e also excluded patients undergoing
surgery within 36 hours before or after
the radiologic procedure. W ith these ex-
ceptions, the series of examinations was
consecutive.
The contrast matera! used was either
ionic (diatrizoate meg!umine [Angiovist
282 and Angiovist 370; Ber!ex, W ayne
NJ]) or nonionic (iopamido! [Isovue 300;
Squibb Diagnostics, New Brunswick,
NJ]). Angiovist 282 is a 60 7 solution con-
taming 282 mg of iodine per mi!!i!iter,
Angiovist 370 is a 76 solution contain-
ing 370 mg iodine per mil!i!iter, and Iso-
vue 300 is a 61 i solution containing 300
mg iodine per milliliter. W hen ionic con-
trast material was used, Angiovist 282
was used for manual injections and An-
giovist 370 was used in the injector.
Choice of contrast material for a given
patient was not random. Nonionic con-
trast material was used for patients of any
age undergoing parathyroid angiography
and venous sampling, pulmonary angiog-
raphy, spinal angiography, and peripher-
a! angiography. Nonionic agents were
also used for cerebral angiography after
the 90th patient (previously ionic agents
were used for both hand and powered in-
jections in patients undergoing cerebral
angiography).
The dose of iodine (in grams) can be
calculated easily for those patients receiv-
ing iopamidol by multiplying the iopami-
do! dose (in milliliters) by 0.3. Patients
who received diatrizoate meglumine
were given the 60 solution for hand in-
jections and the 76T solution for powered
injections. M ost of these patients received
both concentrations during the procedure
but only the total volume of diatrizoate
meglumine was recorded. Therefore, the
correct conversion factor for determining
iodine dose in these patients varies be-
tween 0.282 and 0.37 but is not known ex-
actly for individual patients or for the
group receiving ionic agents as a whole.
No specific hydration orders were
used, except that the patient was asked to
consume only clear liquids from mid-
night before the procedure until after the
procedure was completed. Oral hydration
was always permitted and encouraged. In
genera!, the NIH nursing staff is well
aware of the importance of adequate hy-
dration. To avoid extra attention to hy-
dration during the study (with possible
resultant bias), the clinicians, ward nurs-
ing staff, and patients were not told of the
existence of the study. For the same rca-
son, input and output determinations and
urine specific gravity measurements were
not done. Curious clinicians were in-
formed that postprocedure blood urea ni-
trogen (BUN) and creatinine determina-
tions were being done as part of a quality
assurance program. This was possible be-
cause the only change from the regular
routine of the special procedures section
was the addition of BUN and serum creat-
mine determinations 24 hours after the
Types of Procedures and Doses of Contrast M gterial Used in 200 Exam inations
No. of Contrast M aterial
Type of Procedure Procedures Dose (m L)
Arterial procedures
Nonselective (n 23)
Aortography 23 22366
Selective arteriography (n 105)
Bronchial 1 400
Cerebral 13 17583
Parathyroid 24 303 124
Peripheral 5 16364
Spinal 17 256 141
Visceral 45 254 84
Venous procedures
Venous sampling (n = 50)
Adrenal 4 214 61
Parathyroid 8 33435
Petrosa!sinus 28 17486
Portal venous 9 397 128
Thymic 1 170
Venous imaging (n 22)
Upper-extremity venography 2 6030
Pulmonary angiograpl y 4 271 86
Inferior vena cavography
and intravenous digital
subtraction angiography 16 15355
*M ean standard deviation.
Table 3
Contrast M aterial Dose in 200 Special Procedures Examinations
Contrast M aterial Dose (m L)
M ean SD M inim um M axim um
Diatrizoate meglumine (n 88) 239 112 40 520
Iopamidol (n 112) 236 113 30 530
Overal!(n200) 237113 30 530
*SD standard deviation.
Table 1 Table 2
608 #{149} Radiology June 1988
Diagnoses In 153 Patients
Undergoing Special Procedures
Exam inations
Diagnosis No. (% )
Central nervous system dis-
ease , 16 (10.5)
Endocrine disorders 68 (44.4)
Other m alignant neoplasms
Vasculitis
40 (26.1)
17 (11.1)
Atherosclerosis 4 (2.6)
Other 8 (5.2)
procedure, and this did not require in-
formed consent. (In general, venipunc-
ture does not require informed consent.
In addition, information gained from the
venipuncture benefited the patient by
providing potential early warning of
acute rena! failure.) All patients gave in-
formed consent for the radiologic proce-
dure itself.
Height and weight measurements were
made by the ward nursing staff. W e used
the measurements obtained most closely
before the procedure. For preprocedure
determination of BUN and creatinine 1ev-
els, we used those values obtained most
closely before the procedure. Information
collected for each patient with respect to
the procedure itself included diagnosis,
type of procedure, contrast agent used, to-
tal contrast material dose (to the nearest 5
mL), and adverse reaction, if any. Post-
procedure BUN and creatinine levels
were m easured between 24 and 72 hours
after the procedure, most commonly at 24
hours. All data were recorded on a data
sheet and transferred to a computer for
analysis.
At the NIH Clinical Center, the norm al
range for BUN level is 7-20 mg/dL (2.5-
7.1 mmo!/L of urea), and the normal
range for serum creatinine level is 0.7-1.3
mg/dL (60-110 mol/L). The day-to-day
variability in these measurements is ap-
proximately 2% for values up to 20 mg/dL
(7.1 mmol/L of urea) for BUN and 2.0
mg/dL (180 M mol/L) for creatinine, in-
creasing to approximately 5% for BUN
levels of 50 mg/dL (17.9 mmo!/L of urea)
and creatinine levels of 5.0 mg/dL (440
M mol/L).
Thirty of the 230 examinations were ex-
cluded from analysis because patient
height or weight was unavailable (n 3),
preprocedure or postprocedure renal
function data were unknown (ii 13),
preprocedure renal function data were
obtained more than 10 days before the
procedure (n 4), postprocedure renal
function was not measured between 24
and 72 hours after the procedure (n 7),
or a different contrast agent was used
(ti 3).
The remaining 200 examinations were
performed in 153 patients, with many un-
dergoing more than one. Patients ranged
in age from 10 to 79 years (mean, 43.7
years) and in weight from 31.4 to 176.6 kg
(mean, 77 kg). Patient diagnoses are given
in Table 1, and the types of procedures
and average doses of contrast material
used for each are given in Table 2.
Calculation of body surface area was
performed by means of the standard for-
mula of Dubois and Dubois (3). Data were
analyzed, and linear regression equations
and correlation coefficients were calculat-
ed with standard software (Symphony
[version 1 .2]; Lotus Development Corp.
Cambridge, M ass). Statistical analysis was
performed by standard methods and tech-
niques (4).
RESULTS
There were three contrast material
reactions during the 200 examina-
tions. Two episodes (one of nausea
and vomiting and one of pruritus)
occurred with diatrizoate meglu-
mine, and one (hypertension, confu-
sion, and transient blindness) oc-
curred with iopamidol.
The overall average contrast mate-
rial dose is given in Table 3. To eval-
uate the possible bias of reduction of
contrast material dose on the basis of
patient age or preprocedure renal
function, total dose was plotted
against patient age (Fig 1) and pre-
procedure serum creatinine level (Fig
2). W ithin the entire age range of pa-
tients studied, there was no clear
trend toward change in dose level for
any age group. In particular, there
was no evidence of a decrease in con-
trast material dose for older patients.
Of the 31 patients older than 59
years, 22 received iopamidol and
nine received diatrizoate meglumine.
This is not significantly different
from the distribution in patients
younger than 59 years, 90 of whom
received iopamidol and 79 of whom
received diatrizoate meglumine (x2.
P> .50).
Figure 2 demonstrates that the vast
majority of patients studied had nor-
mal renal function. There was no evi-
dence of a tendency to administer
less contrast material or a nonionic
agent to those few patients with
clearly decreased renal function be-
fore the procedure. Of the 15 patients
with a preprocedure creatinine level
greater than 1.5 mg/dL (130 tmol/L),
seven received iopamidol and eight
received diatrizoate meglumine.
Even in this small subset of patients,
there was no correlation between
+ OKTK2OUE , A
S GPAMIOOL
1 2 3
PRK-ARTERI000AG C00411NIPA( (,+q/dl(
05016120000
S 0*450001
A*
:+
+ 05010120000
S 0*461001
Volum e 167 Num ber 3 Radiology #{149} 609
S
1.
S00
500
400
300
200
too
0
20 40 00
AGO
000
500
I
400
2.
300
200
100
oA*
+ D01012001E
0*450001
Figures 1, 2. (1) Plot of total contrast m aterial dose as a function of age. Linear regression
line y = a + bx is shown and has the following parameters: a 178.5 111.3 (standard er-
ror), b = 1.338 0.529; r = .176, P > .10. There was no evidence of a tendency to use lower
doses of contrast material for older adults. (2) Plot of total contrast material dose as a func-
tion of initial creatinine level. M ost patients had an initial serum creatinine level of 1.5 mg/
dL (130 smol/L) or less. Linear regression line has the following parameters: a 233.9
113.1, b = 2.937 19.02; r .00, P > .50. There was no evidence of a tendency to give less
contrast material to patients with elevated initial serum creatinine levels.
30
20
10
-10
a.
-20
-30
200 400
TOTAl. CONTRAST M ATORIAL DOSS ( .I)
800
1.2
I oe
, 0.6
0.2
0
-0.2
S -0.4
-0.6
-0.6
-1,2
b.
200 400
TOTK. CONTRAST 0011650. DOSE (45)
Figure 3. Plots illustrate effect of contrast m ateria! dose on renal function, m easured by
changes in BUN (a) and serum creatinine (b) levels. Preoperative level is subtracted from
postoperative level. Resultant value is positive for net increases in BUN and creatinine and
negative for net decreases in these tests. Regression line is plotted on each graph: for (a), a
-2.26 4.314, b 0.00 0.002; r .00 and for (b), a -0.03 0.195, b 0.00 0.00; r .00.
Note that there is no trend toward renal dysfunction with higher doses of contrast material.
Separate regression lines and correlation coefficients for diatrizoate meglumine and iopami-
do! are not shown, but they were similar to the values obtained for the overall series, which
indicates there were no demonstrable differences between the two contrast agents in this
setting.
contrast material dose and change in
creatinine level, regardless of wheth-
er dose was expressed in milliliters,
milliliters per kilogram, or milliliters
per square meter (r = .094, . 104, and
.031, respectively).
To evaluate the effect of contrast
material dose on renal function,
change in renal function was plotted
against contrast material dose (Fig 3).
Change in renal function was mea-
sured by subtracting the preproce-
dure value for BUN or creatinine
from the postprocedure value. A pos-
itive result represents an incremental
decrease in renal function, whereas a
negative result represents an incre-
mental improvement in renal func-
tion. No gross change in renal func-
tion was identified with increase in
dose. To correct for varying patient
weight and habitus, similar graphs
were constructed, but with contrast
material dose expressed as a function
of patient weight (Fig 4) or body sur-
face area (Fig 5). Again, no evidence
of deterioration of renal function
with increasing contrast material
dose was identified. No correlation
between the dose of contrast material
and change in renal function was
seen.
Although not shown, regression
lines and correlation coefficients
were also calculated for iopamidol
and diatrizoate meglumine separate-
ly for each graph. W ith respect to
change in creatinine level, the slope
of all regression lines was 0.00, as
was the correlation coefficient. W ith
respect to change in BUN level, the
slopes of regression lines were 0.00
to -0.37 for diatrizoate meglumine
and 0.005-0.247 for iopamidol. Corre-
lation coefficients were between .130
and .210 for diatrizoate meglumine
and .105 and .164 for iopamidol.
There was therefore no evidence of a
correlation between the dose of dia-
trizoate meglumine or that of iopa-
midol and gross change in renal
function.
Finally, we evaluated the effect of
repeated administration of contrast
material by evaluating renal function
in ten patients who underwent two
special procedures, each involving
the administration of at least 200 mL
of contrast material. In each patient,
the two procedures were separated
by a rest day. The combined contrast
material dose for the two procedures
averaged 639 mL (range, 430-925
mL) administered within 48 hours. In
these patients, there was no evidence
of a trend toward deterioration of re-
nal function (Fig 6).
DISCUSSION
There is no question that the intra-
arterial or intravenous administra-
tion of contrast material can impair
renal function. Various risk factors
have been im plicated as predisposing
to renal damage, including dehydra-
_600 tion, diabetes mellitus, multiple my-
eloma, preexisting renal disease, age,
and the type of study (5-9). Som e
studies have suggested that the vol-
ume of contrast material adminis-
tered is also a factor in determining
the likelihood of renal dysfunction
(8,9).
In this study we examined gross
changes in renal function in a con-
secutive series of relatively low-risk
patients. Only 7.5% (15 of 200) had an
initial serum creatinine level greater
than 1.5 mg/dL (130 M mol/L). Severe
cardiac disease was unusual. Al-
though our patients are atypical with
respect to the relative frequency of
various categories of disease (Table
1), we believe they represent a rea-
sonable sample of otherwise relative-
ly low-risk individuals who undergo
angiography.
W e were unable to demonstrate ev-
idence of any gross change in renal
function resulting from increasing
contrast material dose (Figs 3-5). Fur-
thermore, there was no difference
between diatrizoate meglumine and
iopamidol in terms of their effect on
renal function. The absence of gross
renal deterioration extended to the
subset of 15 patients with initial cre-
atinine levels greater than 1.5 mg/dL
(130 mol/L), as well as the subset of
ten patients who received 430-925
mL of contrast material during a pen-
od of approximately 48 hours in two
30
20
10
r
-10
-20
-30
OI400IZOATE
S 0*401001
.++4.4+: # +5 (4
: 61 #{163}
7 2 4 6
00O+L CONTRAST M ATERIAL. DOSE (*1/N)
I #{176}#{176}
. 0.6
0.2
I
-0.2
a
-0.6
I
0500020000
4 0*461001
6 0 .
101*4. CONTRAST 61500041 DOSE ( /N)
a. b.
Figure 4. Plots illustrate effect of contrast m aterial dose on renal function corrected for dif-
ferences in patient weight. Dose of contrast material is expressed as a function of patient
weight (milliliters per kilogram) and plotted against changes in BUN (a) and creatinine (b)
levels, along with the regression line, as in Figure 1. Results are similar. For (a), a -2.25
4.314, b = -0.02 0.182; r .00. For (b), a -0.02 0.195, b 0.00 0.008; r .00.
0010120001
4 0*4611001
: 2 :
20
10
I
10
i -10
-20
-30
a.
+ DIAORIZOATE
S 0*450001
1.2
0.6
0.4
0.2
-0.2
-0.4
-06
-0.6
-1.2
-1.5
I 40 60 120 100 200 240 280
0004. CONTRAST 4*0(6141. 0050 (.4/,.)
I 40 80 120 60 200 240 280
00045. CONTRAST 80008541. DOSE (*l/oo)
b.
Figure 5. Plots illustrate effect of contrast m aterial dose on renal function corrected for
varying patient habitus. Dose of contrast material is expressed as a function of body surface
area (milliliters per square meter) and plotted against changes in BUN (a) and creatinine (b)
levels, as in Figure 1, along with the regression line, with similar results. For (a), a -1.92
4.310, b = 0.00 0.004; r .044. For (b), a -0.02 0.195, b 0.00 0.00; r .00.
::
!r
A+. .g.0+..C*+ M .o.45 A+... . 0+.. C++.. M.o.+
319.1 322,4
*10+60506 1 *10+644562
A00ig. Do.. Co.oout M .1 A+o . Do.. Cootoul M aleo45
31945 322m 1
a. b.
Figure 6. Graphs of serial m easurem ents of BUN (a) and serum creatinine (b) levels in ten
patients undergoing two special procedures 2 days apart. Procedures were performed on
days 1 and 3. Rena! function was measured on days 0, 2, and 4. Each procedure involved the
administration of at least 200 mL of contrast material; the average doses are shown. There
was no tendency toward renal dysfunction during the period shown.
610 . Radiology June 1988
separate procedures.
This study might be criticized on
the grounds that postprocedure BUN
and cneatinine levels were usually
measured 24 hours after the proce-
dune and not at 48 hours or longer, as
has been done in most other studies
(8-11). Dramatic increases in cneati-
nine level may not occur until 48
hours after the procedure, and we
may have missed cases of renal fail-
ure for this reason. However, Older
et al (12) showed that contrast medi-
um-induced renal failure resulted in
demonstrable increases in serum cne-
atinine level by 24 hours in seven of
their nine patients. None of our pa-
tients had similar increases in creati-
nine level. If we had used a more
conservative criterion (ie, consider-
ing renal failure to have occurred if
there was a 50% increase in creati-
nine level), only one of our patients
would have been judged as having
renal dysfunction (this patient ne-
ceived 105 mL of iopamidol and had
a serum creatinine level that in-
creased from 1.0 to 1.5 mg/dL [90 to
130 imol/L]). None of our patients
had clinically apparent acute renal
failure, while five of seven patients
in the series of Older et a! had oligu-
na. Furthermore, serial measure-
ments of BUN and creatinine were
available for the ten patients in our
series who underwent two high-dose
studies 48 hours apart, and these pa-
tients had no laboratory evidence of
renal deterioration (Fig 6). Finally, 47
of the examinations reported herein
were subsequent procedures per-
formed in patients already entered in
the study. W e reviewed the charts
and available laboratory data for
these patients at the time of their
subsequent study. None had clinical
or laboratory evidence of renal fail-
ure associated with their initial ex-
aminations.
Can our experience be generalized
to other patient populations? The pa-
tient population in this series was
representative of NIH experience but
was otherwise atypical. M ost (92.59k)
of the patients in this study had nor-
mal renal function, none had multi-
pie myeloma, and no more than five
had diabetes. The prevalence of heart
disease and atherosclerosis was also
extremely low in this population (Ta-
ble 1) compared with that in the typi-
cal population seen in most special
procedures laboratories.
The prevalence of relatively un-
usual disorders, especially endocrine
disorders, was much higher in this
series than would be expected in a
standard angiography practice. This
has a direct effect on the results for
two reasons. First, these patients tend
to be relatively healthy except for
their endocrine problem. it has been
shown that diabetes (5,6,9), heart dis-
ease (8), and underlying renal insuf-
ficiency (5-9) can predispose to the
development of renal failure after
contrast material administration, al-
though they do not invariably do so
(11). Caution is appropriate in the ex-
tension of our results to patients with
these disorders. Second, patients
with endocrine disorders constituted
the majority of those who received
the highest doses of contrast materi-
al. Of the 13 examinations in this Se-
nies in which 450 mL or more of con-
trast material was administered, all
were parathyroid arteriograms, spi-
nal arteniograms, or portal venous
sampling. All ten patients who un-
derwent two examinations 48 hours
apart (Fig 6) had hyperparathyroid-
ism, Zollinger-Ellison syndrome, or a
Volum e 167 Num ber 3 Radiology #{149} 611
spinal arteriovenous malformation.
Those patients who received the
highest doses of contrast material
tended to have undergone relatively
lengthy, time-consuming procedures,
so that the total dose of contrast ma-
terial was administered during a pe-
nod of 1-2.5 hours. Administration
of contrast material doses this high
during a period of a few minutes
might not be tolerated as well. How-
ever, Hayman et al (13) showed that
high doses of contrast material (80 g
of iodine, equivalent to 265 mL of a
60% contrast material such as Isovue
300) can be given over 10 minutes
with no increase in the rate of renal
dysfunction compared with the rate
resulting from a smaller dose of 40 g
of iodine given in the same fashion.
Furthermore, Rubin et al (14) showed
in a rabbit model that even with in-
termittent (every 10 minutes) injec-
tion of aliquots of diatrizoate meglu-
mine/sodium, serum iodine concen-
tration and osmolality increased with
each successive injection. This sug-
gests that the period during which
contrast material is administered may
not be of major significance.
A final factor is the emphasis on
hydration in our patients. Although
we cannot prove that all patients
were well hydrated, clinicians at the
NIH Clinical Center are well aware
of the doses of contrast material we
commonly use. In consequence, some
patients, particularly those with hy-
perparathyroidism and Zollinger-El-
lison syndrome, received intrave-
nous hydration in addition to oral
hydration. The decreases in BUN 1ev-
el noted at high contrast material
doses (Figs 3-5) may be accounted for
by the overhydration and diuresis
caused by this practice. It must be
noted, however, that hydration is not
a panacea, and contrast material-in-
duced renal failure can occur despite
vigorous hydration (9).
There is both theoretical (15) and
experimental (16) evidence to sug-
gest that iopamidol should be less
nephrotoxic than is diatrizoate me-
glumine. However, our study
showed neither agent to be particu-
larly toxic clinically, and we identi-
fied no real difference between
them. This is in accordance with the
study of Gale et al (17) in humans in
which a sensitive enzyme marker of
renal injury was used and no differ-
ence in nephrotoxicity was found
among iopamidol, iothalamate, and
diatrizoate.
Contrast material toxicity is not
limited to renal dysfunction, of
course. Toxicity to the central ner-
vous system, heart, and lungs must
also be considered when high doses
of these agents are used. Although
we encountered only three instances
of contrast material reaction or toxic-
ity, none of which appeared to be
dose related, these factors must also
be considered before proceeding
with the administration of large
doses of contrast material.
W e conclude that in patients with
relatively normal renal and cardiac
function and without diabetes melli-
tus or multiple myeloma, very large
doses of contrast material appear to
be tolerated as well as smaller doses,
if adequate patient hydration is used.
In our patients there was no identifi-
able trend toward deterioration of re-
nal function with increasing contrast
material dose over the dose range
studied. No difference was evident
between diatrizoate meglumine and
iopamidol in this study. These results
may not apply to high-risk patients,
especially those with preexisting re-
nal impairment. U
References
1. Reuter SR. Redman HC. Gastrointestinal
angiography. 2d ed. Philadelphia: Saun-
ders, 1977; 17.
2. Abrams HL. The opaque media: physio-
logic effects and systemic reactions. In:
Abrams HL, ed. Abrams angiography: vas-
cular and interventional radiology. 3d ed.
Boston: Little, Brown, 1983; 20.
3. Dubois D, Dubois EF. A formula to esti-
mate the approximate surface area if
height and weight be known. Arch Intern
M ed 1916; 17:863-871.
4. Swinscow TDV. Statistics at square one.
London: British M edical Association,
1980.
5. W einrauch LA, Healy RW , Leland OS Jr,
et a!. Coronary angiography and acute
renal failure in diabetic azotemic ne-
phropathy. Ann Intern M ed 1977; 86:56-
59.
6. Alexander RD. Berkes SL, Abuelo JG.
Contrast media-induced oliguric renal
failure. Arch Intern M ed 1978; 138:381-
384.
7. Cochran ST. W ong W S, Roe DJ. Predict-
ing angiography-induced acute renal
function impairment: clinical risk model.
AJR 1983; 141:1027-1033.
8. M artin-Paredero V. Dixon SM , Baker JD,
et al. Risk of renal failure after major an-
giography. Arch Surg 1983; 118:1417-
1420.
9. Comes AS, Baker JD. M artin-Paredero V.
et al. Acute renal dysfunction after major
arteriography. AJR 1985; 145:1249-1253.
10. Cramer BC, Parfrey PS, Hutchinson TA, et
a!. Rena! function following infusion of
radiologic contrast material: a prospective
controlled study. Arch Intern M ed 1985;
145:87-89.
1 1 . Crux C, Hricak H, Samhouri F, Smith RF,
Eyler W R, Lavin NW . Contrast media for
angiography: effect on renal function. Ra-
dio!ogy 1986; 158:109-112.
12. Older RA, Korobkin M , Cleeve DM ,
Schaaf R, Thompson W . Contrast-in-
duced renal failure: persistent nephro-
gram as clue to early detection. AJR 1980;
134:339-342.
13. Hayman LA, Evans RA, Fahr LM , Hinck
VC. Renal consequences of rapid high
dose contrast CT. AJR 1980; 134:553-555.
14. Rubin DI, Kido DK, M orris TW , Fischer
HW . Effect of multiple intravenous injec-
tions of diatrizoate on iodine concentra-
tions and osmola!ity: an experimental
study in rabbits. Invest Radio! 1987; 22:
223-226.
15. Dawson P. Contrast agent nephrotoxi-
city: an appraisal. Br J Radiol 1985; 58:121-
124.
16. Lund C, Rysavy J, Salomonowitz E, et al.
Nephrotoxicity of contrast media assessed
by occlusion arteriography. Radiology
1984; 152:615-619.
17. Gale M E, Robbins AH, Hamburger RJ, W i-
drich W C. Renal toxicity of contrast
agents: iopamidol, iothalamate, and diatri-
zoate. AJR 1984; 142:333-335.