Abdominal Multidetector CT in Patients with Varying Body Fat

Document Sample
Abdominal Multidetector CT in Patients with Varying Body Fat Powered By Docstoc
					                                            Note: This copy is for your personal non-commercial use only. To order presentation-ready
                                            copies for distribution to your colleagues or clients, contact us at

                                                                                        Abdominal Multidetector CT in
                                                                                        Patients with Varying Body Fat
                                                                                        Percentages: Estimation of Optimal
                                                                                        Contrast Material Dose1
                           Hiroshi Kondo, MD

                                                                                             Purpose:    To determine if contrast material dose for abdominal mul-
                           Masayuki Kanematsu, MD
                                                                                                         tidetector computed tomography (CT), as determined by
                           Satoshi Goshima, MD
                                                                                                         using body weight (BW), overestimates the amount of
                           Yuhei Tomita, RT                                                              contrast material required in heavier patients.
                           Toshiharu Miyoshi, RT
                           Atsushi Hatcho, RT                                            Materials and   Institutional review committee approval and patients’ writ-
                           Noriyuki Moriyama, MD                                             Methods:    ten informed consent were obtained. CT images of the
                           Minoru Onozuka, PhD                                                           abdomen were obtained by using 2 mL per kilogram of BW
                           Yoshimune Shiratori, MD                                                       of intravenous contrast material (300 mg/mL iodine) in-
                           Kyongtae T. Bae, MD, PhD                                                      jected at 4 mL/sec in 161 consecutive patients (age range,
                                                                                                         28 –90 years; mean age, 63 years; 95 men, 66 women). CT
                                                                                                         scans were initiated 45 and 150 seconds after aortic en-
                                                                                                         hancement increased by 50 HU. The patients were divided
                                                                                                         into low (37–54 kg) and high (55–75 kg) BW groups. The
                                                                                                           HU/I, where HU is change in CT number and I is iodine
                                                                                                         dose in grams, and adjusted maximum hepatic enhance-
                                                                                                         ment ( HU/[I/kg]) were assessed for correlation with
                                                                                                         BW, body mass index (BMI), and body fat percentage
                                                                                                         (BFP) by using linear regression.

                                                                                              Results:     HU/I correlated (P     .001) inversely with BW in the
                                                                                                         aorta (r     0.78) and liver (r    0.80) and with BMI in
                                                                                                         the aorta (r      0.59) and liver (r      0.61) on portal
                                                                                                         venous phase images. Regression formula for the low BW
                                                                                                         group was HU/I       4.1    .044 BW (P      .001) and for
                                                                                                         the high BW group was HU/I         2.7    .017 BW (P
                                                                                                         .001), suggesting that the amount of contrast material
                                                                                                         required with increased BW is less in the high BW group.
                                                                                                         Adjusted maximum hepatic enhancement directly corre-
                                                                                                         lated with BFP (r 0.25, P .01).
                             From the Department of Radiology, Kizawa Memorial
                           Hospital, Gifu, Japan (H.K., Y.T.); Departments of Radiol-      Conclusion:   Excessive contrast material may inadvertently be given in
                           ogy (H.K., M.K., S.G.), Radiology Services (M.K., T.M.),                      heavier patients when the dose is determined by patient
                           and Medical Informatics (Y.S.), Gifu University Hospital,
                                                                                                         BW. Contrast material dose may need to be tailored in
                           1-1 Yanagido, Gifu 501-1194, Japan; Research Center of
                                                                                                         individual patients by using BFP.
                           Brain and Oral Science (M.K., M.O.) and Department of
                           Physiology and Neuroscience (M.O.), Kanagawa Dental
                           College, Yokosuka, Japan; Department of Radiology, Na-                         RSNA, 2008
                           gano Red Cross Hospital, Nagano, Japan (A.H.); Research
                           Center for Cancer Prevention and Screening, National
                           Cancer Center Hospital, Tsukiji, Japan (N.M.); and Depart-
                           ment of Radiology, University of Pittsburgh Medical Cen-
                           ter, Pittsburgh, Pa (K.T.B.). Received January 5, 2008;
                           revision requested February 26; revision received March
                           26; final version accepted June 23. Supported in part by
                           the Grant for Scientific Research Expenses for Health,
                           Labor, and Welfare Programs; Foundation for the Promo-
                           tion of Cancer Research; and Research on Cancer Pre-
                           vention and Health Services. Address correspondence
                           to H.K.

                             RSNA, 2008

                           872                                                                                                      Radiology: Volume 249: Number 3—December 2008
GASTROINTESTINAL IMAGING: Optimal Contrast Material Dose for Abdominal CT                                                                      Kondo et al

       ontrast material– enhanced com-          [10], and injection rate of iodinated        overestimates the amount of contrast
       puted tomography (CT) is a com-          contrast material [11,12]; scan delay af-    material required in heavy patients who
       monly used imaging modality for          ter contrast material injection [2,13]),     are undergoing abdominal multidetec-
the detection and preoperative staging          as well as patient-related factors (eg,      tor CT.
of malignant liver tumors. Accurate             body weight [BW] [9,14], cardiac out-
evaluation of the degree of local tumor         put [15]). Ho et al (16) reported that
extension is crucial for estimating the         calculation of contrast material dose on      Materials and Methods
likelihood of benefit from surgical resec-       the basis of lean BW leads to increased
tion and the prognosis of patients with a       patient-to-patient uniformity of hepatic     Patients
malignant hepatic neoplasm (1). Be-             parenchymal and vascular enhance-            This study was approved by the Kizawa
cause the majority of metastases are hy-        ment. Heiken et al (14) advocated the        Memorial Hospital institutional review
povascular, the merits of routine acqui-        concept of adjusted maximum hepatic          committee, and patients gave written in-
sition of hepatic arterial– dominant            enhancement (MHE), which is the MHE          formed consent. During a 4-month period
phase images with contrast-enhanced             (in Hounsfield units) divided by the dose     (December 2003–March 2004), 217 con-
CT are still disputable. However, he-           (in grams of iodine) divided by BW (in       secutive patients who were suspected of
patic arterial– dominant phase images           kilograms), for determination of opti-       having abdominal disease because of pre-
may be obtained when primary or met-            mal total iodine dose required for a         vious ultrasonographic or laboratory eval-
astatic hypervascular tumors are sus-           given hepatic enhancement increase           uation underwent contrast-enhanced CT
pected or three-dimensional CT angiog-          (eg, 50 HU) in a patient of a given BW.      of the upper abdomen. We excluded 56
raphy is necessary. For detection of the        The adjusted MHE was reported to be          patients from our study for the following
majority of metastatic tumors, adequate         96 HU       19 (standard deviation) and      reasons: 12 had undergone partial hepa-
enhancement of hepatic parenchyma is            has been considered constant irrespec-       tectomy; 11 had numerous liver metasta-
of the utmost importance. It has been           tive of patient BW. For example, the         ses; nine had severe fatty liver (liver at-
reported that maximal enhancement of            iodine dose per kilogram of BW re-           tenuation 40 HU); five had undergone
the liver to ensure optimal tumor-to-           quired for a 50-HU increase in hepatic       total splenectomy; three experienced
liver contrast is obtained 45–55 seconds        enhancement should be 0.521 g/kg, ac-        technical failure related to contrast mate-
after the initiation of contrast material       cording to the equation.                     rial injection, breath holding, or machine
administration (2). To maximize the                  However, the fact that the blood        disorder during CT examination; and 16,
conspicuity of hypovascular hepatic tu-         vessel volume and liver weight are not       who weighed more than 75 kg, were ad-
mors, CT images of the liver are com-           directly proportional to BW (17) sug-        ministered contrast material at a fixed
monly acquired at the portal venous             gests iodine dose might not be optimally     amount of 150 mL because that was the
phase with a single bolus of intravenous        determined by patient BW. In other           largest volume of iodinated contrast ma-
contrast material (3–8).                        words, since the liver accounts for a        terial commercially supplied in a syringe.
     Contrast enhancement of the liver          lower percentage of BW in a heavy pa-
at CT is affected by several radiologic         tient who has more fat tissue than does
                                                                                             Published online before print
factors (eg, dose [9], concentration            a lighter patient, an iodine dose tailored
                                                proportionately to BW might be more
                                                than required for MHE in heavier pa-         Radiology 2008; 249:872– 877
 Advances in Knowledge                          tients. We have wondered, in both our        Abbreviations:
    The change in aorta and liver CT            clinical and research practice, whether      BFP body fat percentage
    number per gram of iodine ad-               a double-dose of iodinated contrast ma-      BMI body mass index
    ministered correlates inversely             terial would ever be necessary to            BW body weight
    with body weight (BW) and body              achieve MHE in an obese patient who is        HU change in CT number
                                                                                             I dose of iodine (in grams)
    mass index.                                 twice the average weight. We con-
                                                                                             MHE maximum hepatic enhancement
    The regression slope of the high            ducted this study to determine if con-
    BW group was less steep than                trast material dose determined by BW         Author contributions:
    that of the low BW group, which                                                          Guarantor of integrity of entire study, H.K.; study con-
                                                                                             cepts/study design or data acquisition or data analysis/
    suggests that an excess of iodine
    was given to heavier patients.               Implication for Patient Care                interpretation, all authors; manuscript drafting or manu-
                                                                                             script revision for important intellectual content, all au-
    Adjusted maximum hepatic en-                    When designing an abdominal CT           thors; approval of final version of submitted manuscript,
    hancement (change in CT number                  protocol, optimal iodine doses           all authors; literature research, H.K., M.K., A.H., N.M.,
    divided by the dose of iodine in                required for visualizing aorta and       M.O., Y.S., K.T.B.; clinical studies, H.K., M.K., S.G., Y.T.,
    grams per kilogram of BW) was                   liver are more accurate when they        T.M., A.H., K.T.B.; statistical analysis, H.K., M.K., S.G.,
    virtually constant for BW, but                  are calculated on the basis of           Y.T., T.M., A.H., M.O., Y.S., K.T.B.; and manuscript edit-
                                                                                             ing, H.K., M.K., N.M., Y.S., K.T.B.
    correlated directly with body fat               body fat percentage rather than
    percentage.                                     total BW or body mass index.             Authors stated no financial relationship to disclose.

Radiology: Volume 249: Number 3—December 2008                                                                                                           873
GASTROINTESTINAL IMAGING: Optimal Contrast Material Dose for Abdominal CT                                                           Kondo et al

These 56 patients were excluded from the       scans. Real-time low-dose (120-kVp,            correlation with BW, BMI, and BFP by
study population because the hepatic vas-      50-mA) serial monitoring scans were            using a linear regression test. In addi-
cular alteration due to numerous tumors        initiated 5 seconds after starting con-        tion, linear regression analyses were
or previous surgery, abnormal hepatic          trast material injection. During the           performed separately for the low
density, or artifacts might adversely affect   5-second interval between contrast ma-         (BW      55 kg) and high (BW      55 kg)
the evaluation of contrast enhancement.        terial injection commencement and the          BW groups. The low BW group con-
The remaining 161 patients (age range,         start of the monitoring scan, patients         sisted of 75 patients (weight range,
28 –90 years; mean age, 63.3 years             were carefully monitored by a radiolo-         37–54 kg; mean weight, 47.3 kg 4.5;
13.6; BW range, 37–75 kg; mean BW,             gist for extravasation or acute adverse        BMI range, 15.8 –24.9 kg/m2; mean
55.8 kg 9.6) formed the study popula-          events due to the contrast material in-        BMI, 20.1 kg/m2 2.1), and there were
tion. There were 95 men (age range,            jection. The dual-phase scan was initi-        86 patients in the high BW group
28 – 85 years; mean, 63.7 years       13.2)    ated for the portal venous and equilib-        (weight range, 55–75 kg; mean weight,
and 66 women (age range, 34 –90 years;         rium phases at 45 and 150 seconds, re-         63.3 kg      5.8; BMI range, 19.3–28.9
mean, 62.7 years 14.2).                        spectively, after the bolus-tracking           kg/m2; mean BMI, 23.7 kg/m2         2.3).
                                               program detected the threshold change          Sex ratios were compared between the
Contrast Material Injection and Scan           (2,13).                                        low BW and high BW groups by using
Protocols                                                                                     the 2 test. Means of patient ages and of
Imaging was performed by using a multi-        Quantitative Image Analysis                    CT numbers of the liver on unenhanced
detector CT scanner (Light Speed QX/i;         Mean CT numbers of the aorta and the           images were compared between the low
GE Healthcare, Milwaukee, Wis) with            liver were measured (H.K., with 8 years        BW and high BW groups by using the
four sections at 2.5-mm detector collima-      posttraining experience interpreting body      unpaired Student t test. For determin-
tion, 15 mm per rotation table feed            CT images) in all patients on the CT con-      ing the required dose of iodine in corre-
speed, and 0.75 pitch. The images were         sole monitor by using a circular region-of-    lation with BFP to produce desirable he-
displayed as 40 5-mm-thick sections with       interest cursor ranging in size from           patic enhancement (40, 50, 60, or 70
no intersection gap for each phase set.        10 –30 mm in diameter on the unen-             HU), we composed a nomogram by
Gantry rotation time was 0.7 second, and       hanced and portal venous and equilib-          modifying the nomogram previously ad-
scan acquisition time was 8.9 seconds.         rium phase images. CT numbers in he-           vocated by Heiken et al (14): A rela-
    All patients were administered non-        patic parenchyma were measured in              tional expression between BFP and ad-
ionic iodinated contrast material that con-    three areas (right anterior, right poste-      justed MHE obtained in our study was
tained 300 mg of iodine per milliliter         rior, and left lateral segments), and the      substituted into the Heiken et al rela-
(Omnipaque 300 [iohexol]; Daiichi              values were averaged. Focal hepatic le-        tional expression; I/kg          HU/(ad-
Sankyo, Tokyo, Japan) at body tempera-         sions, blood vessels, the bile duct, calcifi-   justed MHE), where I/kg is the dose of
ture by using a power injector at a rate of    cations, and artifacts were carefully ex-      iodine in grams needed per kilogram
4 mL/sec through a 21-gauge plastic intra-     cluded from the measurement areas.             of BW, and HU is the desired level of
venous catheter (32-mm long) placed in              The degree of contrast enhance-           hepatic enhancement.
an antecubital (158 of 161 patients, 98%)      ment was expressed as the change in CT
or radial (three patients, 2%) vein. In all    number ( HU), which was calculated by
161 patients, the volume of contrast ma-       subtracting CT numbers on unenhanced            Results
terial delivered was 2 mL per kilogram of      images from those on contrast-en-              There was no significant difference in
BW, thus resulting in a range of 74 –150       hanced images. Enhancement parame-             sex ratio, patient age, or CT number of
mL (mean, 112 mL).                             ters for the aorta and the liver calcu-        the liver on unenhanced images be-
    A bolus-tracking program (Smart-           lated in each patient were HU, HU              tween the low BW and high BW groups.
Prep; GE Healthcare) was used to de-           per iodine dose (in grams), and ad-            The correlation coefficients between the
termine the time for initiation of diag-       justed MHE ( HU adjusted for iodine              HU/I of the aorta and the liver and
nostic scanning following contrast mate-       dose and patient BW). The adjusted             BW, BMI, and BFP are summarized in
rial injection. This enabled real-time         MHE was calculated in accordance with          the Table. Linear regression showed
monitoring and the automatic calcula-          the method proposed by Heiken et al            that the HU/I of the aorta correlated
tion of CT numbers in a region of inter-       (14): adjusted MHE            HU/(I/BW),       inversely with BW on the portal venous
est, as well as manual initiation of diag-     where I is the amount of iodine adminis-       phase images (r        0.78, P     .001).
nostic scanning after the CT number in         tered in grams.                                Similarly, the HU/I of the liver corre-
the aorta reached the threshold of a                                                          lated inversely with BW on portal ve-
50-HU increase. The region-of-interest         Statistical Analysis                           nous phase images (r          0.80, P
cursor for bolus tracking was placed in        Statistical analyses were performed by         .001) (Table). The HU/I of the aorta
the aorta at a level just above the dia-       using statistical software (JMP, version       and liver showed inverse correlation
phragmatic dome; this level was also           5; SAS Institute, Cary, NC). The HU/I          with BMI in the portal venous phase
used as a starting position for diagnostic     and adjusted MHE were evaluated for            (r     0.59, P     .001; r     0.61, P

874                                                                                                Radiology: Volume 249: Number 3—December 2008
GASTROINTESTINAL IMAGING: Optimal Contrast Material Dose for Abdominal CT                                                                            Kondo et al

.001, respectively). The HU/I of the                           BW group was HU/I 5.2 0.032 BW                       expression, and a relational expression
aorta and the liver showed no correla-                         (r     0.39, P .001) (Fig 1). The regres-            (I/kg    HU/[83.12 0.62 BFP]) was
tion with BFP in the portal venous                             sion slope of the high BW group                      obtained to compose a modified nomo-
phase.                                                         ( 0.032) was less steep than that of the             gram (Fig 4).
    The regression formula for the                             low BW group ( 0.088). The regres-
  HU/I of the aorta in the portal venous                       sion formula for the HU/I of the liver in
phase versus BW for the low BW group                           the portal venous phase for the low BW                Discussion
was HU/I       8.4     0.088 BW (r                             group was HU/I        4.1     0.044 BW               Although several previous researchers
  0.50, P .001), and that for the high                         (r      0.54, P .001), and that for the              have evaluated optimal CT protocols for
                                                               high BW group was HU/I              2.7              the liver by adjusting contrast material
                                                               0.017 BW (r            0.38, P       .001)           dose, injection rate, and scan delays,
                                                               (Fig 2). The lower slope in the high BW              only a few have dealt with the effects of
   Correlation between HU/I and                                group in the aorta and liver suggests a              BW (9,14,16,18). Despite experimental
   Physical Parameters in Aorta and                            trend for slower decline of contrast en-             data that indicate an inverse relation-
   Liver on Portal Venous Phase CT                             hancement with increased BW and for                  ship between hepatic contrast enhance-
   Images                                                      less contrast material to be required per            ment at CT and patient BW (18), a uni-
   Parameter and Organ                              r Value    kilogram of BW to maintain contrast                  form dose of intravenous contrast mate-
                                                               enhancement.                                         rial, regardless of patient BW, is still
                                                                   The average adjusted MHE was 99.7                administered to patients undergoing ab-
      Aorta                                           0.78*
                                                               HU     15.0 in patients overall. The ad-             dominal CT at many centers. To achieve
      Liver                                           0.80*
                                                               justed MHE was virtually constant with               high-quality examinations that allow de-
                                                               the increase of BW (r 0.11, P .15)                   tection and characterization of abdomi-
      Aorta                                           0.59*
                                                               (Fig 3a). Adjusted MHE showed posi-                  nal diseases while minimizing cost, a
      Liver                                           0.61*
                                                               tive correlation with BMI in the portal              tailored dose of contrast material is de-
      Aorta                                           0.05     venous phase (r     0.16, P      .05) (Fig           sirable. However, this approach leads
      Liver                                           0.08     3b) and with BFP in the portal venous                to a dilemma since the iodine load in-
                                                               phase (r 0.25, P .01) (Fig 3c).                      creases as patient BW increases; iodine
   Note.—All correlations between HU/I and organ were              The relational expression between                may inadvertently be given in larger
   significant (P .001).
                                                               BFP and adjusted MHE (adjusted                       doses than are necessary in heavier pa-
   * Significantly greater than r value for BFP (P      .05).
                                                               MHE 83.12 0.62 BFP) was substi-                      tients. The advantages of reducing the
                                                               tuted into the Heiken et al relational               iodine load in heavy patients include a
                                                                                                                    potential decrease in nephrotoxicity,
                                                                                                                    particularly in patients with preexisting
 Figure 1                                                       Figure 2                                            renal insufficiency or other risk factors
                                                                                                                    associated with obesity (19,20), and
                                                                                                                    cost reduction. Thus, the thrust of our
                                                                                                                    study was to investigate whether iodine
                                                                                                                    load could be determined by other pa-
                                                                                                                    tient physical factors (eg, BMI, BFP) for
                                                                                                                    CT of the liver, to avoid excessive iodine
                                                                                                                    doses in heavy patients.
                                                                                                                         On contrast-enhanced CT images of
                                                                                                                    the liver, the dose of contrast material is
                                                                                                                    directly related to MHE, whereas pa-
                                                                                                                    tient BW is related inversely to it
                                                                                                                    (14,18). In our study, HU/I of the
                                                                                                                    aorta and of the liver correlated in-
                                                                                                                    versely with patient BW in the portal
 Figure 1: Graph of regression of HU/I of aorta                 Figure 2: Graph of regression of HU/I of liver      venous phase when a linear regression
 against BW on portal venous phase CT images.                   against BW on portal venous phase CT images.        test was used. Further, the lower slope
 For low BW group ( ), the regression formula is                For low BW group ( ), the regression formula is     in the high BW group indicated a trend
   HU/I 8.4 0.088 BW; P .001; for high                            HU/I 4.1 0.044 BW; P .001; for high               for a slower decline of contrast en-
 BW group (F), HU/I 5.2 0.032 BW; P                             BW group (F), HU/I 2.7 0.017 BW; P                  hancement with increased BW and for
 .001. Slope of regression line for high BW group               .001. Slope of regression line for high BW group
                                                                                                                    less contrast material to be required per
 ( 0.032) is less steep than that for low BW group              ( 0.017) is less steep than that for low BW group
                                                                                                                    kilogram of BW to maintain contrast
 ( 0.088).                                                      ( 0.044).
                                                                                                                    enhancement. This suggests that more

Radiology: Volume 249: Number 3—December 2008                                                                                                              875
GASTROINTESTINAL IMAGING: Optimal Contrast Material Dose for Abdominal CT                                                                                Kondo et al

 Figure 3

 Figure 3: Scatterplots show relations between adjusted MHE and physical parameters in portal venous phase. (a) There was no significant correlation between ad-
 justed MHE and BW (r 0.11, P .15). Adjusted MHE showed positive correlation with (b) BMI (r 0.16, P .05) and (c) BFP (r 0.25, P .01).

iodine was given when the dose was                     level of hepatic enhancement in a pa-
determined solely by BW than would be                  tient of known BW and BFP: This may                        Figure 4
required for optimal enhancement in                    obviate the use of excessive iodine in
heavier patients.                                      overweight or obese patients. With the
     Heiken et al (14) also advocated the              nomogram we proposed, for example,
use of a formula that incorporated MHE                 the dose of iodine required for desired
adjusted for iodine dose (adjusted                     enhancement levels of 50 HU in a pa-
MHE) and BW. For a patient of known                    tient with BFP of 22% was about 0.52 g
BW, one can calculate the dose of io-                  of iodine per kilogram, whereas that in
dine needed to achieve a desired level of              a patient with BFP of 40% was about
hepatic enhancement by using this for-                 0.46 g. By calculating iodine dose on the
mula: adjusted MHE            HU/(I/BW),               basis of BFP, we may be able to more
where HU is the desired level of he-                   efficiently optimize the iodine dose for
patic enhancement, I is the dose of io-                imaging the liver in these patients.
dine in grams, and BW is given in kilo-                    Ho et al (16) reported that calcula-
grams. Heiken at al reported that ad-                  tions of contrast material dose on the
justed MHE was virtually a constant                    basis of measured lean BW (total                           Figure 4: Nomogram for determining dose of
value at 96 HU. Hence, for example, the                BW [1       {BFP/100}]) marginally in-                     iodine needed to achieve desired level of hepatic
dose of iodine required for desired en-                creased patient-to-patient uniformity                      enhancement in a patient with known BW and BFP.
hancement level of 50 HU is approxi-                   with respect to hepatic parenchyma and                     Iodine (gI/kg) necessary for a given hepatic en-
mately 0.521 g/kg.                                     vascular enhancement. Our results are                      hancement can be expressed as follows: I/kg
     Because we found the highest corre-               consistent with their observations that                      HU/(83.12 0.62 BFP).
lation between adjusted MHE and BFP                    an iodine dose determination that con-
among the three physical parameters in                 siders the weight of fat in the human
our study, we propose a modified ad-                    body is a more optimal way to keep the                   We did not take this issue into account
justed MHE formula as follows: adjusted                intensity of hepatic and vascular en-                    in our study. Second, we applied the
MHE 83.12 0.62 BFP. Hence, the                         hancement constant and to reduce the                     regression formula, adjusted MHE
iodine weight in grams per kilogram of                 intersubject variability in enhancement                  83.12     0.62 BFP, for the construc-
BW necessary for a given level of he-                  intensity.                                               tion of the nomogram, despite the fact
patic enhancement can be expressed as                      There were some limitations to our                   that the correlation coefficient between
follows: I/kg    HU/(83.12 0.62 BFP).                  study. First, as patient BW increases,                   adjusted MHE and BFP was as small as
The nomogram we derived from this                      the image noise also increases, which                    0.25. The nomogram needs to be vali-
equation may serve as a reference from                 suggests that an increased dose of io-                   dated in future clinical studies. Third,
which one can easily determine the io-                 dine may be necessary to compensate                      our study findings may not apply to pa-
dine dose needed to produce a desired                  for the image degradation due to noise.                  tients whose BW is greater than 75 kg,

876                                                                                                                    Radiology: Volume 249: Number 3—December 2008
GASTROINTESTINAL IMAGING: Optimal Contrast Material Dose for Abdominal CT                                                                         Kondo et al

as our study excluded patients weighing                 3. Small WC, Nelson RC, Bernardino ME,              12. Tublin ME, Tessler FN, Cheng SL, Peters TL,
greater than 75 kg because we could not                    Brummer LT. Contrast-enhanced spiral CT              McGovern PC. Effect of injection rate of con-
                                                           of the liver: effect of different amounts and        trast medium on pancreatic and hepatic heli-
obtain doses of iodine greater than 45 g.
                                                           injection rates of contrast material on early        cal CT. Radiology 1999;210:97–101.
Many patients in North America and                         contrast enhancement. AJR Am J Roentge-
Europe exceed 75 kg, and iodine doses                      nol 1994;163:87–92.                              13. Kondo H, Kanematsu M, Satoshi G, et al.
of greater than 45 g are not infrequently                                                                       MDCT of the pancreas: optimizing scanning
                                                        4. Freeny PC, Gardner JC, von Ingersleben G,            delay with a bolus-tracking technique for
given to such patients. We believe that                    Heyano S, Nghiem HV, Winter TC. Hepatic              pancreatic, peripancreatic vascular, and he-
the iodine load per kilogram of BW                         helical CT: effect of reduction of iodine dose       patic contrast enhancement. AJR Am J
needed to achieve a given hepatic con-                     of intravenous contrast material on hepatic          Roentgenol 2007;188:751–756.
trast enhancement can be reduced for                       contrast enhancement. Radiology 1995;197:
                                                           89 –93.                                          14. Heiken JP, Brink JA, McClenan BL, et al.
these heavy patients as well, but this
                                                                                                                Dynamic incremental CT: effect of volume
assumption needs to be verified by clin-                 5. Valls C, Andia E, Sanchez A, et al. Hepatic          and concentration of contrast material and
ical studies in western countries whose                    metastases from colorectal cancer: preoper-          patient weight. Radiology 1995;195:353–
populations include patients who weigh                     ative detection and assessment of resectabil-        357.
                                                           ity with helical CT. Radiology 2001;218:
more than 75 kg.
                                                           55– 60.                                          15. Bae KT, Heiken JP, Brink JA. Aortic and
    In conclusion, contrast material                                                                            hepatic contrast medium enhancement at
dose for the enhancement of the aorta                   6. Abdelmoumene A, Chevallier P, Chalaron               CT. II. Effect of reduced cardiac output in a
                                                           M, et al. Detection of liver metastases under
and the liver, as determined by patient                                                                         porcine model. Radiology 1998;207:657–
                                                           2 cm: comparison of different acquisition            662.
BW, may be excessive in heavier pa-                        protocols in four row multidetector-CT
tients because the regression slope of                     (MDCT). Eur Radiol 2005;15:1881–1887.            16. Ho LM, Nelson RC, Delong DM. Determin-
the high BW group was less steep than                                                                           ing contrast medium dose and rate on basis
                                                        7. Valette PJ, Pilleul F, Crombe-Ternamian A.
that for the low BW group, and the ad-                     MDCT of benign liver tumors and metasta-
                                                                                                                of lean body weight: does this strategy im-
justed MHE increased as BFP in-                                                                                 prove patient-to-patient uniformity of he-
                                                           ses. Eur Radiol 2003;13(suppl 5):M31–M41.
                                                                                                                patic enhancement during multi– detector
creased. The iodine dose needed to
                                                        8. Marchiano A. MDCT of primary liver malig-            row CT? Radiology 2007;243:431– 437.
achieve a desired level of hepatic en-                     nancies. Eur Radiol 2003;13(suppl 5):M26 –
hancement at contrast-enhanced CT                          M30.                                             17. Bae KT, Heiken JP, Brink JA. Aortic and
may be tailored to individual patients by                                                                       hepatic contrast medium enhancement at
                                                        9. Yamashita Y, Komohara Y, Takahashi M,                CT. I. Prediction with a computer model.
using their BFP.
                                                           et al. Abdominal helical CT: evaluation of           Radiology 1998;207:647– 655.
                                                           optimal doses of intravenous contrast mate-
                                                           rial—a prospective randomized study. Radi-       18. Kormano M, Partanen K, Soimakallio S,
                                                           ology 2000;216:718 –723.                             Kivimaeki T. Dynamic contrast enhance-
 1. Haider MA, Amitai MM, Rappaport             DC,                                                             ment of the upper abdomen: effect of con-
    et al. Multi– detector row helical CT in    pre-   10. Awai K, Takada K, Onishi H, et al. Aortic            trast medium and body weight. Invest Radiol
    operative assessment of small ( 1.5         cm)        and hepatic enhancement and tumor-to-liver           1983;18:364 –367.
    liver metastases: is thinner collimation    bet-       contrast: analysis of the effect of different
    ter? Radiology 2002;225:137–142.                       concentrations of contrast material at multi–    19. Lasser EC, Lyon SG, Berry CC. Reports on
                                                           detector row helical CT. Radiology 2002;             contrast media reactions: analysis of data
 2. Goshima S, Kanematsu M, Kondo H, et al.                                                                     from reports to the U.S. Food and Drug Ad-
    MDCT of the liver and hypervascular hepa-                                                                   ministration. Radiology 1997;203:605– 610.
    tocellular carcinomas: optimizing scan de-         11. Foley WD, Hoffmann RG, Quiroz FA, Kahn
    lays for bolus-tracking techniques of hepatic          CE Jr, Perret RS. Hepatic helical CT: con-       20. Tublin ME, Murphy ME, Tessler FN. Current
    arterial and portal venous phases. AJR Am J            trast material injection protocol. Radiology         concepts in contrast media–induced nephropa-
    Roentgenol 2006;187:W25–W32.                           1994;192:367–371.                                    thy. AJR Am J Roentgenol 1998;171:933–939.

Radiology: Volume 249: Number 3—December 2008                                                                                                            877

Shared By: