Sonographic Evaluation of Spleen Size in Tall Healthy Athletes

Transcription

1 Abdominal Imaging Spielmann et al. Sonography of Spleen Size Audrey L. Spielmann 1 David M. DeLong Mark A. Kliewer 3 Spielmann AL, DeLong DM, Kliewer MA Received March, ; accepted after revision June 1,. 1 Vancouver General Hospital, Department of Radiology, Vancouver Coastal Health, 99 W 1th Ave., Vancouver, BC V5Z 1M9, Canada. Address correspondence to A. L. Spielmann. Department of Radiology, Duke University Medical Center, Box 3, Durham, NC Department of Radiology, University of Wisconsin Hospital & Clinics, Highland Ave., Madison, WI AJR 5;1: X/5/11 5 American Roentgen Ray Society Sonographic Evaluation of Spleen Size in Tall Healthy Athletes OBJECTIVE. The purpose of this study was to establish the range of spleen sizes in tall healthy athletes. SUBJECTS AND METHODS. Sonographic measurements of spleen size and left renal length were performed on 19 college athletes ( men, 7 women). Length, width, and thickness of the spleen and left renal length were obtained. In addition, the height, weight, and age of each athlete were recorded. Pearson s product moment correlation coefficients were calculated, and linear regression analysis was used to create a model for calculating normative values. RESULTS. The mean body height for men was 7.3 (19 cm) ± (SD) 3.7 inches (9 cm) and for women was 9.3 (17 cm) ± 3.7 inches (9 cm). Spleen length was greater than 1 cm in 31.7% of the men (mean spleen length, 11. ± 1.7 cm) and in 1.% of the women (mean spleen length, 1.3 ± 1.3 cm). In women, height correlated with spleen length (r =.3, p =.5), width (r =., p =.1), and volume (r =.3, p =.) but not with thickness (r =., p =.). Spleen length did correlate with left renal length (r =.5, p =.5). In men, height correlated with spleen length (r =., p =.3), width (r =.5, p =.1), and volume (r =., p =.) and less with thickness (r =.3, p =.1). Spleen length and left renal length were poorly correlated (r =., p =.). Regression analysis showed that in women taller than 5 ft inches (1 cm), the mean splenic length of 1 cm increased by.1 cm for each 1-inch incremental increase in height. In men taller than ft (1 cm), the mean splenic length of 11 cm increased by. cm for each 1-inch incremental increase in height. CONCLUSION. Spleen size correlates with height in tall healthy athletes. Nomograms from this data can be used to gauge the risk of returning to play after episodes of acute splenomegaly, as with infectious mononucleosis. he accurate diagnosis of splenic enlargement is a matter of consider- T able importance for athletes of unusual size and height. Although splenic enlargement can be the result of a number of disorders (including infectious, infiltrative, immunologic, and malignant conditions), viral illnesses are by far the most common cause in the young population of college athletes. Splenic enlargement is particularly marked with infectious mononucleosis, a condition endemic on college campuses and one that places the patient at risk for splenic rupture. Although spontaneous rupture of the spleen is both uncommon and rarely fatal, most athletic team physicians will exclude athletes with mononucleosis from vigorous activity and contact sports until their spleen size returns to normal and the athletes feel well. Determining when it is safe for the young athletes to return to play is not straightforward. In the intensely competitive arena of modern collegiate sports, there is often pressure on physicians by coaches and the players themselves to sanction the resumption of play at the earliest possible time. The physical examination is unreliable to assess for splenic enlargement and therefore has been largely supplanted by imaging studies, particularly sonography. For the sonographic diagnosis and monitoring of splenomegaly, standards of normal spleen sizes have been developed for the general population, but not for those patients whose body habitus and size are at the extremes of the normal range. The extrapolation of data from the general population to those of unusual height is a dubious strategy. To our knowledge, this is the first study that has evaluated spleen size in the very tall population. The purpose of this study then was to determine normative spleen data for taller athletes to diagnose more accurately sple- AJR:1, January 5 5

2 Spielmann et al. nomegaly in this population and to better guide decisions regarding the safety of returning to contact sports. Materials and Methods Student athletes participating in basketball, football, volleyball, and soccer programs at a single large university were recruited for the study and prospectively evaluated during their preseason annual physical examinations. One hundred twentynine healthy students were assessed ( men and 7 women). Predetermined exclusion criteria for this study included a history of splenectomy, a recent history of infectious mononucleosis (within the past months), age younger than 1 years, a history of malignancy, hematologic disorders, prolonged febrile illness, or chronic fatigue syndrome. Our institutional review board approved the study, and informed consent was obtained in all subjects. All sonographic studies were performed on an HDI 5 or 35 sonographic unit (ATL-Philips) with a curvilinear -5 MHz transducer (C5- probe, ATL-Philips). The subjects were placed in a supine or right posterior oblique position and scanned during suspended respiration. The spleen size was measured in the sagittal plane in the standard oblique coronal orientation to record the maximal length (in centimeters) of the spleen. The transverse plane, perpendicular to the oblique coronal plane, was then measured to record the transverse width and anteroposterior thickness (in centimeters) and width or thickness (in centimeters) of the spleen (Fig. 1). The transverse dimension of the spleen was measured as the greatest dimension in the transverse plane, and the width was measured as the shortest distance from the hilum to the outer convexity of the spleen. The sonographic program calculated volume in cubic centimeters (using the formula for a prolated ellipse). The length of the left kidney was also documented. The same experienced sonographic technologist performed all of the scanning with a radiologist in attendance. In addition, the height and weight of the subjects were recorded. Pearson s product moment correlation coefficients were calculated for each of the variables, and linear regression analysis was used to create a predictive model for normative values of spleen size dimensions in this population. Results The average age of the participants was 19 ± 1 year. For the men, the mean height was 7.3 (19 cm) ± (SD) 3.7 inches (9 cm) (range, 5 3 inches [15 11 cm]) and the mean weight was (1 kg) ±.5 lb (1 kg) (range, lb [9 15 kg]). The mean splenic dimensions were 11. ± 1.7 cm sagittal (range,. 1.1 cm), 1. ± 1. cm transverse (range, 1 cm), 5. ±. cm in width, and 333. ± 11.1 cm in volume. The mean left kidney length was 11.1 ±.9 cm; 31.7% of the men had a spleen length of greater than 1 cm and 13.% had a spleen length of greater than 13 cm. In men, height correlated with spleen length (r =., p =.3) (Fig. ), transverse dimension (r =.5, p =.1) (Fig. 3), and volume (r =., p =.) but less with width (r =.3, p =.1). The left kidney length correlated with the transverse measurement of the spleen (r =., p =.1) and weakly with the splenic length (r =., p =.3). All splenic measurements correlated better with height than weight. Subject height correlated poorly with left kidney length (r =., p =.). Regression analysis showed that in men taller than 7 inches (13 cm), the mean splenic length of 11 cm increased by. cm for each inch of increase in height. The representative 95% confidence intervals were calculated at -inch intervals (Table 1). For the women, the mean body height was 9.3 (17 cm) ± 3.7 inches (9 cm) (range, 1 7 inches [ cm]) and the mean weight was 15 lb (9 kg) (range, lb [57 kg]). The mean splenic dimensions were 1.3 ± 1.3 cm sagittal (range, cm), 9.5 ± 1. cm transverse (range, cm),. ±.7 cm in width, and.3 ± 7. cm in volume. The mean kidney length was 1.3 ± 1.1 cm; 1.% of the women had a spleen length of greater than 1 cm. In women, body height correlated with spleen length (r =.3, p =.5) (Fig. ), width (r =., p =.1) (Fig. 5), and volume (r =.3, p =.), but not thickness (r =., p =.). The left kidney length correlated with spleen length (r =.5, p =.5), transverse dimension (r =., p =.), and volume (r =., p =.). The spleen size and renal length correlations were stronger in women Fig year-old man standing ft.5 inches (19 cm). A, Longitudinal sonogram shows craniocaudal length of spleen to be 1.5 cm. Electronically enhanced measurement cursors show measurement technique. B, Transverse sonogram of spleen shows measurement of anteroposterior thickness ( ) and transverse width (+) of spleen. A B AJR:1, January 5

3 Sonography of Spleen Size than in men. Splenic measurements did not correlate with subject weight. In women, subject height did correlate with left kidney length (r =.5, p =.). Regression analysis showed that in women taller than inches (1 cm), the mean splenic length of 1 cm increased by.1 cm for each inch increase in height. Representative 95% confidence intervals were calculated (Table ). 7 9 Fig.. Scatterplot graph shows correlation of body height with spleen length in male subjects. Pearson s product moment correlation coefficient was. (p =.3) Fig.. Scatterplot graph shows correlation of body height with spleen length in female subjects. Pearson s product moment correlation coefficient was.3 (p =.5). Spleen Width (cm) Discussion With infectious mononucleosis, splenomegaly may be clinically palpable as early as the second week of illness, but the clinical examination of splenic size is notoriously inaccurate. In one study evaluating spleen size in patients with sarcoidosis, splenomegaly was present in 57% of the patients (using sonographic criteria to evaluate size) but only clinically palpable in % [1]. The range in Fig. 3. Scatterplot graph shows correlation of body height with transverse spleen width in male subjects. Pearson s product moment correlation coefficient was.5 (p =.1). Spleen Width (cm) Fig. 5. Scatterplot graph shows correlation of body height with transverse spleen width in female subjects. Pearson s product moment correlation coefficient was. (p =.1). reported sensitivity of assessing splenomegaly by palpation of the abdomen varies from 5% at best to the more realistic estimate of 17% []. Therefore, imaging has become essential for the accurate assessment of spleen size, the serial monitoring of spleen size over course of the patient s illness, and development of guidelines for return to play. In these guidelines, it is generally recommended to withhold the athlete from vigorous AJR:1, January 5 7

4 Spielmann et al. TABLE 1 Height Men: Mean and Upper Limit of the Normal Range of Spleen Length for Increments of Body Height by Linear Regression Method Mean Normal Upper Limit a ft (13 cm) ft in (193 cm) ft in (3 cm) ft (13 cm) a Two SDs above the normal mean. TABLE Height Women: Mean and Upper Limit of the Normal Range of Spleen Length for Increments of Body Height by Linear Regression Method Mean Normal Upper Limit a 5 ft in (173 cm) ft (13 cm) ft in (1 cm) ft in (19 cm) a Two SDs above the normal mean. sports until the spleen has returned to normal size [3]. The problem is that current standards of spleen size may not be applicable to the very tall individual. In this study, we found that standard charts for spleen size may be misleading in the subpopulation of very tall individuals. Although the mean spleen size in this group falls within the normal range, the upper limit of the normal range needs to be revised upward (Tables 1 and ). Graphic representation of the data shows some variability of spleen size by height but also an unmistakable trend for the spleen length to increase in taller individuals (Figs. 5). By way of illustration, Figure 1 shows the sonographic images of a large spleen (1.5-cm length, cm transverse) in a healthy male subject who was ft.5 inches (19 cm) in height. Some investigators have sought to establish an internal reference standard against which spleen size can be calibrated. The most common such strategy is to compare the length of the spleen to the length of the left kidney. Loftus and Metreweli [] proposed a spleen kidney ratio of 1.5, as measured on sonography, as the upper limit of normal in a pediatric population. In our study, spleen length was correlated with kidney length in women, but not in men. We therefore found the spleen kidney ratio to be an unreliable index for spleen size. Rather we propose the use of the look-up Tables 1 and as references for daily use in a busy practice setting. Several prior studies have sought to develop standards for spleen size, using a variety of imaging techniques, such as CT, scintigraphy, MRI, and sonography One study used four measurements from two imaging planes in the volume formula for an ellipsoid to estimate splenic volume [5]. The spleen, however, has a variable 3D configuration, and its shape does not easily conform to the simple geometry of an ellipsoid. Therefore, volumetric measurement is obtained most accurately on CT or MRI []. Nevertheless, routine CT for the diagnosis and serial follow-up of patients for suspected splenic enlargement is difficult to justify in view of the radiation exposure (especially in a pediatric or young adult population) and the expense. As an alternative, MRI is hampered by expense and limited availability in many areas of the world. One group estimated the normal splenic volume using 3D sonography in 5 normal volunteers [7]. The 3D technique was believed to provide a more accurate measurement of the spleen volume than conventional sonographic techniques but was cumbersome and time-consuming and therefore not practical for a busy sonographic practice. A recent study comparing sonography with CT for the estimation of splenic volume proposed the following formula:.5 width thickness (maximal length craniocaudal length) / as the most accurate estimation of splenic volume on sonography []. Regardless, many studies indicate that volumetric data are cumbersome, inherently flawed considering the complex 3D shape of the spleen, and usually not required to guide the clinical management of splenomegaly. As a practical clinical guideline, the distance that the spleen extends below the costal margin is often used to monitor spleen size, and of course a similar measurement could also be obtained on sonography. Conventional sonography is a well-established, widely used, and relatively inexpensive means of assessing the spleen without ionizing radiation. A study by Rosenberg et al. [9] found that a simple measurement of splenic length was accurate as a guide to spleen size. The authors proposed setting the upper normal limit of splenic length at 1 cm for girls 15 years or older and at 13 cm for boys 15 years or older. This study [9] and another by Dittrich et al. [5]. found that there was an approximately linear increase in spleen size, as measured on sonography, in the course of development in the pediatric population and that spleen size correlated best with the body height. A third study of the pediatric population by Konus et al. [1] with 37 subjects also found that height correlated best with spleen length (r =.), although their tallest individual was only inches (173 cm), which is of course much shorter than our study population. In a study of adults, Frank et al. [11] used conventional sonography to evaluate 793 healthy patients (17 years) and found that 95% of patients had a splenic length of less than 11 cm, a width (transverse dimension) of less than 7 cm, and a thickness of less than 5 cm. Niederau et al. [1] studied 915 healthy subjects using sonography and found the mean longitudinal and transverse diameters of the spleen to be 5. ± 1. cm and 5.5 ± 1. cm, respectively [1]. These dimensions are much smaller than those of other studies because the authors did not measure the maximal length of the spleen as authors in other studies have, but rather the true superior-to-inferior length. This study further found that spleen size did correlate weakly with height in the general population. Sonographic measurements of splenic length have also been correlated with actual spleen dimensions at autopsy [13]. A study by Loftus et al. [13] on 3 cadavers found a clear linear relationship between a sonographic measurement of splenic length and the actual length, volume, and weight as measured at autopsy. The Pearson s product moment correlation coefficient for maximal sonographic length and the actual length was.31 (p <.1). They suggested that a simple single sonographic measurement of length could be used for routine work, reserving the more complex volumetric measurement (splenic area or index) for problematic cases. This position is further bolstered by another study that showed good correlation between CT volume and splenic length measured on sonography (r =., p <.1) [1]. The AJR:1, January 5

5 Sonography of Spleen Size preponderance of evidence, then, indicates that a simple measurement of spleen length is a practical and reasonably accurate estimate of spleen size. What is the true risk of splenic injury and bleeding for a young athlete recovering from infectious mononucleosis? With splenomegaly from mononucleosis, splenic rupture is reported to occur in one to two patients per 1, affected individuals [15] and usually takes place between days and 1 after the onset of illness, when the splenic capsule is heavily infiltrated with lymphocytes [3, 1, 17]. In a -year Mayo clinic review of,11 patients with infectious mononucleosis, only nine had splenic rupture. Although the most vulnerable period is during weeks of illness, rupture can occur as early as 3 days or as late as months [3]. These statistics are not drawn from a population of athletes engaged in vigorous contact sports, but the risk of splenic rupture is relatively low even when the spleen is large. This risk is probably much less still with normal spleen sizes. There are a few limitations to this study. Our study population included only tall healthy athletes. It seems likely that the data from this study would also apply to tall individuals who are not athletes, but we cannot be certain. A larger study that includes nonathletes might improve the precision of our estimates and also the generalizability of the data. In conclusion, the writing of guidelines for return to play must take into account revised standards of spleen size. Our data support adjusting the upper limit of the normal range for spleen length to avoid the false-positive diagnosis of splenomegaly in very tall patients. Certainly, the health and safety of these young athletes must remain the paramount concern, but also it would be unfortunate to bench a healthy athlete because inappropriate standards are applied to the interpretation of sonographic measurements, especially considering that the risk of splenic rupture is likely quite small after recovery from the acute period of splenomegaly with infectious mononucleosis. References 1. Kataoka M, Nakata Y, Maeda T, et al. Ultrasonographic analysis of splenomegaly in patients with sarcoidosis [in Japanese]. Nihon Kyobu Shikkan Gakkai Zasshi 199;: Dommerby H, Stangerup SE, Stangerup M, Hancke S. Hepatosplenomegaly in infectious mononucleosis: assessed by ultrasonic scanning. J Laryngol Otol 19;1: Eichner R. Infectious mononucleosis: recognizing the condition reactivating the patient. Phys Sportsmed 199;:9 5. Loftus WK, Metreweli C. Ultrasound assessment of mild splenomegaly: spleen/kidney ratio. Pediatr Radiol 199;: Dittrich M, Milde S, Dinkel E, Baumann W, Weitzel D. Sonographic biometry of liver and spleen size in childhood. Pediatr Radiol 193;13: 11. Breiman RS, Beck JW, Korobkin M, et al. Volume determinations using computed tomography. AJR 19;13: De Odorico I, Spaulding KA, Pretorius DH, Lev- Toaff AS, Bailey TB, Nelson TR. Normal splenic Notice to Authors volumes estimated using three-dimensional ultrasonography. J Ultrasound Med 1999;1:31 3. Yetter E, Acosta K, Olson M, Blundell K. Estimating splenic volume: sonographic measurements correlated with helical CT determination. AJR 3;11: Rosenberg HK, Markowitz RI, Kolbeg H, Park C, Hubbard A, Bellah RD. Normal splenic size in infants and children: sonographic measurements. AJR 1991;157: Konus OL, Ozdemir A, Akkaya A, Erbas G, Celik H, Isik S. Normal liver, spleen, and kidney dimensions in neonates, infants, and children: evaluation with sonography. AJR 199;171: Frank K, Linhart P, Kortsik C, et al. Sonographic determination of spleen size: normal dimensions in adults with a healthy spleen [in German]. Ultraschall Med 19;7: Niederau C, Sonnenberg A, Muller JE, Erckenbrecht JF, Scholten T, Fritsch WP. Sonographic measurements of the normal liver, spleen, pancreas, and portal vein. Radiology 193;19: Loftus WK, Chow LT, Metreweli C. Sonographic measurement of splenic length: correlation with measurement at autopsy. J Clin Ultrasound 1999;7: Lamb PM, Lund A, Kanagasabay RR, Martin A, Webb JA, Reznek RH. Spleen size: how well do linear ultrasound measurements correlate with three-dimensional CT volume assessments? Br J Radiol ;75: Farley DR, Zietlow SP, Bannon MP, et al. Spontaneous rupture of the spleen due to infectious mononucleosis. Mayo Clin Proc 199;7: Haines JD Jr. When to resume sports after infectious mononucleosis: how soon is safe? Postgrad Med 197;1: Maki DG, Reich RM. Infectious mononucleosis in the athlete: diagnosis, complications, and management. Am J Sports Med 19;1:1 173 Authors of accepted manuscripts now receive page proofs via . The author s computer system must have version 3. (or later) of Adobe Acrobat Reader. The ARRS Web site ( has a link to the Adobe Web site, where authors can download free copies of the most recent version of the reader. AJR:1, January 5 9