British Journal of Radiology 75 (2002),573-577 © 2002 The British Institute of Radiology
Spleen size: how well do linear ultrasound measurements correlate with three-dimensional CT volume assessments?
P M Lamb, BSc, MRCP, FRCR,
A Lund, DCR, DMU,
R R Kanagasabay, BSc, MRCP, FRCS,
A Martin, DCR,
J A W Webb, MD, FRCP, FRCR and
R H Reznek, FRCP, FRCR
St Bartholomew's Hospital, West Smithfield, London EC1A 7BE, UK
Correspondence: Dr PM Lamb, 49 Stoneydeep, Twickenham Road, Teddington, Middlesex TW11 8BL, UK
 |
Abstract
|
|---|
Ultrasound measurement of splenic length is standard practice, but it is not known how well this represents the true size of the spleen. Previous studies, using a combination of measurements from in vivo and resected spleens, were subject to error because of changes in splenic size. The aim of this study was to correlate the dimensions of the spleen measured by ultrasound with the splenic volume measured by helical CT. Ultrasound examination was performed on 50 adult patients at the time of their attendance for abdominal CT. Linear dimensions of the spleen were measured with the patient first in the supine and then in the right lateral decubitus (RLD) position. The splenic volume was calculated from a three-dimensional reconstruction of the CT images. There was good correlation, using Spearman's rank correlation, between ultrasound measurements and CT volumes with correlation coefficients exceeding 0.7 for all parameters except one. The linear measurement that correlated most closely with CT volume was the spleen width measured on a longitudinal section with the patient in the RLD position (correlation coefficient (r)=0.89, p<0.001). There was also good correlation between splenic length measured in the RLD position and CT volume (r=0.86, p<0.001). We conclude that a good correlation exists between in vivo ultrasound assessment of splenic size and true splenic volume. The most accurate single measurement is spleen width measured on a longitudinal section with the patient in the RLD position. However, measurement of splenic length, which is the most commonly used in clinical practice, also correlates well with splenic volume, particularly when performed with the patient in the RLD position.
|
Introduction
|
|---|
The spleen is the largest organ in the reticuloendothelial system. It has been standard practice, for many years, to use splenic size as an indicator of disease activity in a variety of disorders of the reticuloendothelial system. However, it is not known how well the measurement of splenic length that is routinely obtained at ultrasound examination represents true splenic size. Previous studies of ultrasound measurement of the spleen have reported good correlation of in vivo measurements with the weight of the resected spleen [1] and with the volume of the resected spleen as measured by water displacement [2]. There also appears to be a good correlation between linear ultrasound measurements of cadaveric spleens in situ and ultrasound measurement of the volume of the excised spleen [3, 4]. However, these studies have some limitations. For example, there is evidence to show that spleen size decreases both during and after splenectomy, depending on the circulatory conditions at surgery, the timing and duration of vessel clamping and any uncontrolled blood loss [5]. It has been suggested that the spleen weighs 25–50% more in vivo because of the dynamic nature of its blood supply [6, 7].
The objective of this study was to correlate linear dimensions of the spleen measured by ultrasound in vivo with splenic size as assessed by helical CT with three-dimensional (3D) volume reconstruction.
|
Materials and methods
|
|---|
Adult patients attending for abdominal CT were recruited to this study. Ethics committee approval was obtained and all subjects gave informed consent to have an additional ultrasound scan of the spleen, performed at the time of their CT examination. All images were obtained by an experienced sonographer (AL), using an Acuson 128XP/10 machine (Acuson, Mountain View, CA) and a 3.5 MHz curvilinear probe.
Measurements were performed with the patient in the supine (SUP) position and were then repeated with the patient in the right lateral decubitus (RLD) position. Measurements were usually performed on deep inspiration because it was considered that these measurements would be more comparable with those obtained from the single breath-hold helical CT examination [8]. Occasionally, when the lung base obscured the spleen on deep inspiration, scans were obtained on shallow inspiration or at rest. In keeping with the authors' own clinical practice, the oblique intercostal approach was adopted for all patients as this view allows the optimal window for spleen imaging in the majority of patients.
All measurements were made on sections through the splenic hilum in order to create a constant reference point for repeating measurements (Figure 1
). Splenic length, defined as the maximum distance between the dome of the spleen and the spleen tip, was measured on a longitudinal section. Splenic width, defined as the maximum distance between the medial and lateral borders of the spleen, was measured in a plane perpendicular to the length. Splenic depth, defined as the maximum anteroposterior dimension, was measured on a transverse section. Splenic width was then measured in a plane perpendicular to the depth. Each measurement was recorded three times to the nearest millimetre. The median value was determined for the purposes of statistical analysis. In order to allow calculation of the mean intra-observer error, eight adult volunteers were scanned in an identical manner to the patients, and these measurements were repeated at a later date.
View larger version (13K):
[in this window]
[in a new window]
|
Figure 1. Schematic diagrams to show how linear ultrasound measurements were obtained on longitudinal and transverse sections through the spleen.
|
|
Abdominal CT scans were performed using a GE HiSpeed Advantage Scanner (IGE Medical Systems, Milwaukee, WI). Images were obtained at 1 cm slice thickness using a single breath-hold helical CT technique with a pitch of 1 and following iv contrast material (100 mls Omnipaque 300; Nycomed, Amesham, UK). The CT spleen volume was determined using the Advantage windows software package (ADW 1.2) by summation of cross-sectional areas and 3D reconstruction (Figure 2). The outline of each of the 1 cm axial CT slices through the spleen was traced, from which the computer calculated a cross-sectional area for each slice. This was multiplied by the slice thickness to give the volume for each section of the spleen. Individual volumes were then added together to give the total spleen volume. This method has previously been validated for our scanning equipment using phantoms [9]. Two volume calculations were made for each patient and the mean value was determined for the purposes of statistical analysis. For 30 patients, a third measurement was made at a later date so that the mean intra-observer error could be calculated.
View larger version (54K):
[in this window]
[in a new window]
|
Figure 2. Axial CT section (left) and three-dimensional reconstructed image of the spleen (right) to demonstrate the summation of cross-sectional areas technique to calculate spleen volume.
|
|
Statistical methods
The mean intra-observer error was calculated for both linear ultrasound measurements and CT volume estimations. Ultrasound measurements were compared with CT volume estimations using Spearman's rank correlation.
|
Results
|
|---|
50 adult patients were recruited to the study: 28 men and 22 women. The mean intra-observer error of the linear ultrasound measurements was 9%, corresponding to a mean absolute error of 4.9 mm. The mean intra-observer error of the CT volume estimations was 3.5%, corresponding to a mean absolute error of 9.2 cm3.
The range of splenic sizes assessed by CT volume measurement is illustrated in Figure 3 and varied from 62 cm3 to 945 cm3. There was a good correlation between ultrasound measurements and CT volumes, with correlation coefficients exceeding 0.7 for all linear measurements except splenic width measured on a transverse section (Tables 1 and 2). Overall, measurements performed on a longitudinal section correlated more strongly than those performed on a transverse section. In addition, with the exception of splenic depth, measurements performed in the RLD position correlated more closely with CT volume measurements than their SUP equivalents.
View larger version (9K):
[in this window]
[in a new window]
|
Figure 3. Histogram to show the range of splenic sizes as assessed by helical CT three-dimensional volumetric analysis.
|
|
View this table:
[in this window]
[in a new window]
|
Table 1. Correlation between linear ultrasound measurements and CT splenic volume for patients in the supine position
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Correlation between linear ultrasound measurements and CT splenic volume for patients in the right lateral decubitus position
|
|
The linear ultrasound measurement found to correlate most closely with the CT volume was spleen width measured on a longitudinal section through the hilum, with the patient in the RLD position. This is illustrated by the scatter-plot in Figure 4, which shows the line of regression and 95% confidence limits for the 50 patients studied. Splenic length measured in the RLD position also showed a strong correlation with the CT volume (r=0.86, p<0.001).
View larger version (14K):
[in this window]
[in a new window]
|
Figure 4. Scatter-plot to demonstrate the correlation between ultrasound measurement of splenic width and CT measurement of splenic volume.
|
|
|
Discussion
|
|---|
The mean percentage error of volume calculations using the summation of cross-sectional areas CT technique is small, with errors in the range of 3–5% [10]. Analysis of the measurements of phantoms carried out on the CT scanning equipment used in this study revealed an excellent correlation between the CT volume estimations and the true volume as measured by water displacement [9]. There is, however, no consensus as to what constitutes a normal splenic volume. Henderson et al [11] reported a mean volume of 209±76 cm3 using a summation-of-areas technique. Other authors have attempted to derive a simplified measure of splenic volume by calculating a splenic index, defined as the product of length, width and thickness. The splenic index appears to correlate closely with the more cumbersome summation of cross-sectional areas technique [12] and a normal range of 120–480 cm3 has been proposed using this method [5, 13]. In a more recent study, the authors derived a formula based on the splenic index and suggested a more conservative range of normality, varying from 107–314 cm3 [6]. The majority of spleens in our study, approximately 70%, lay within the latter range, but both smaller and larger spleens were examined (Figure 3
).
There is marked variation in the ultrasound techniques used to evaluate the spleen [8]. Views described include transverse sections of the spleen with the patient supine or in the RLD position, longitudinal scans in the coronal plane and oblique intercostal scans that parallel the left rib cage. The choice of the oblique intercostal approach in this study was based on the authors' experience in their clinical practice that this view allows the optimal window for spleen imaging in the majority of patients. Measurements performed with the patient in the RLD position showed a consistently stronger correlation with CT volume measurements than those performed with the patient in the SUP position, indicating that the RLD position allows more accurate evaluation of splenic size. This is probably because gravity assists in moving the spleen away from the chest wall during inspiration, helping to optimize the ultrasonic window by reducing the possibility of interfering ribs and lung along the periphery of the spleen.
Measurements performed on a longitudinal section correlated more strongly with splenic volume than those performed on a transverse section. This is in agreement with the authors' clinical experience and with previous studies that report difficulty in obtaining reproducible transverse images through the spleen. In routine clinical practice, measurements are performed by the majority of radiologists and sonographers on longitudinal scans only [14, 15].
The authors recognize that this study is limited by the use of a single observer to perform ultrasound measurements for each patient. The use of two observers and the measurement of an inter-observer error would be ideal. Nevertheless, the mean intra-observer error was less than 10% for ultrasound measurements.
Finally, in this study the linear ultrasound measurement that correlated most strongly with CT splenic volume measurement was spleen width, estimated on a longitudinal section with the patient in the RLD position (r=0.89). However, splenic length also correlated extremely well (r=0.86 for the RLD position and r=0.83 for the SUP position). We conclude that the measurement of splenic length in routine clinical practice is a very good indicator of actual splenic size.
Received for publication March 30, 2001.
Revision received January 18, 2002.
Accepted for publication March 7, 2002.
|
References
|
|---|
- Ishibashi H, Higuchi N, Shimamura R, Hirata Y, Kudo J, Niho Y. Sonographic assessment and grading of spleen size. J Clin Ultrasound 1991;19:21–5.[Medline]
- Koga T. Correlation between sectional area of the spleen by ultrasonic tomography and actual volume of the removed spleen. J Clin Ultrasound 1979;7:119–20.[Medline]
- Loftus WK, Chow LT, Metreweli C. Sonographic measurement of splenic length: correlation with measurement at autopsy. J Clin Ultrasound 1999;27:71–4.[Medline]
- Rodrigues A, Rodrigues C, Germano M, Rasera I, Cerri G. Sonographic assessment of normal spleen volume. Clin Anat 1995;8:252–5.[Medline]
- Strijk S, Wagener D, Bogman M, Pauw B, Wobbes T. The spleen in Hodgkin disease: diagnostic value of CT. Radiology 1985;154:753–7.[Abstract/Free Full Text]
- Prassopoulos P, Daskalogiannaki M, Raissaki M, Hatjidakis A, Gourtsoyiannis N. Determination of normal splenic volume on computed tomography in relation to age, gender and body habitus. Eur Radiol 1997;7:246–8.[Medline]
- Griffith R, Janney C. Hematopoietic system: bone marrow and blood, spleen and lymph nodes. In: J K, editor. Anderson's pathology. St Louis, MO: Mosby, 1990:1373–492.
- Perlmutter GS. Ultrasound measurements of the spleen. In: Goldberg BB, Kurtz AB, editors. Atlas of ultrasound measurements (1st edn). Chicago, IL: Yearbook Medical Publishers Inc, 1990:126–38.
- Turner B. Volume determinations of phantoms using computed tomography. Eur Radiol 1999;9(Suppl.):S216.
- Breiman R, Beck J, Korobkin Mea. Volume determinations using computed tomography. AJR 1982;138:329–33.[Abstract/Free Full Text]
- Henderson J, Heymsfield S, Horowitz J, Kutner M. Measurement of liver and spleen volume by computed tomography. Radiology 1981;141:525–7.[Abstract/Free Full Text]
- Cools L, Osteaux M, Divano L, Jeanmart L. Prediction of splenic volume by a simple CT measurement: a statistical study. J Comput Assist Tomog 1983;7:426–30.[Medline]
- Lackner K, Brecht G, Janson R, Scherholz K, Luetzeler K, Thurn P. Wertigkeit der computertomographie bei die stadieneinteilung primaerer lymphknotenneoplasier. Fortschr Geb Rontgenstr Nuklearmed Erganzungsbd 1980;132:21–30. (In German.)
- Rosenberg HK, Markowitz RI, Kolberg H, Park C, Hubbard A, Bellah RD. Normal splenic size in infants and children: sonographic measurements. AJR 1991;157:119–21.[Abstract/Free Full Text]
- Prassopoulos P, Cavouras D. CT Assessment of normal splenic size in children. Acta Radiologica 1994;35:152–4.[Medline]
This article has been cited by other articles:
|
|
|
|
 
M. C. B. Godoy, P. M. Vos, P. L. Cooperberg, C. P. Lydell, P. Phillips, and N. L. Muller
Chest Radiographic and CT Manifestations of Chronic Granulomatous Disease in Adults
Am. J. Roentgenol.,
November 1, 2008;
191(5):
1570 - 1575.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R G Hosey, C G Mattacola, V Kriss, T Armsey, J D Quarles, J Jagger, D Vigil, and C M Lebrun
Ultrasound assessment of spleen size in collegiate athletes * Commentary * Commentary.
Br. J. Sports Med.,
March 1, 2006;
40(3):
251 - 254.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. S. Bezerra, G. D'Ippolito, S. Faintuch, J. Szejnfeld, and M. Ahmed
Determination of Splenomegaly by CT: Is There a Place for a Single Measurement?
Am. J. Roentgenol.,
May 1, 2005;
184(5):
1510 - 1513.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. L. Spielmann, D. M. DeLong, and M. A. Kliewer
Sonographic Evaluation of Spleen Size in Tall Healthy Athletes
Am. J. Roentgenol.,
January 1, 2005;
184(1):
45 - 49.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
