38
Rapid Communication
Assessment of Spleen Stiffness Using Acoustic Radiation Force Impulse Imaging (ARFI): Definition of Examination Standards and Impact of Breathing Maneuvers Messung der Milzsteifigkeit mit Acoustic Radiation Force Impulse Imaging (ARFI): Definition von Untersuchungsstandards und Einfluss von Atemmanövern Authors
T. Karlas1, 2*, F. Lindner2*, M. Tröltzsch2, 3, V. Keim2, 3
Affiliations
1
3
Leipzig University Medical Center, IFB Adiposity Diseases, Leipzig, Germany University Hospital Leipzig, Department of Medicine, Neurology and Dermatology, Division of Gastroenterology and Rheumatology, Germany University Hospital Leipzig, Interdisciplinary Ultrasound Unit, Leipzig, Germany
Key words
Abstract
Zusammenfassung
"
!
!
Purpose: Spleen elastography is a promising method for the characterization of portal hypertension in cirrhotic individuals. However, standardized examination procedures for spleen stiffness measurement have not been defined yet. We analyzed the distribution characteristics of spleen shear-wave velocity (ARFI) and assessed the influence of the respiratory position on spleen stiffness measured by ARFI. Materials and Methods: 25 healthy probands and 25 patients with Child A liver cirrhosis were prospectively characterized with conventional ultrasound, transient elastography, liver ARFI, and underwent spleen ARFI in two respiratory positions: breath hold after expiration (exp) and deep inspiration (insp). For each position 20 single measurements were performed. The distribution of spleen ARFI values was analyzed for normality and the appropriate number of measurements for spleen stiffness estimation was investigated. Results: Spleen ARFI results were normally distributed in > 95 % of cases. Performing 20 instead of 10 single measurements resulted in < 5 % deviation from the mean value after 20 measurements in the majority of cases. Cirrhotic patients had a higher spleen stiffness compared to healthy probands (exp: 3.25 ± 0.58 vs. 2.46 ± 0.35 m/s; p < 0.001). Deep inspiration caused an overall increase in spleen stiffness in both groups: probands 2.46 ± 0.35 m/s (exp) vs. 2.66 ± 0.36 m/s (insp), p = 0.01; cirrhotics 3.25 ± 0.58 m/s (exp) vs. 3.46 ± 0.38 m/s (insp), p = 0.03. However, cases with high spleen stiffness values (exp) show decreasing ARFI values in deep inspiration. Conclusion: ARFI values of the spleen are normally distributed and the mean of 10 valid measurements can be used as a representative value. Deep inspiration significantly modulates spleen stiffness. Therefore, the respiratory position needs careful standardization.
Ziel: Die Milzelastizität ist ein vielversprechender Parameter für die Diagnostik der portalen Hypertension bei Leberzirrhose. Bislang existieren jedoch noch keine einheitlichen Untersuchungsstandards für die Messung der Milzsteifigkeit. In dieser Studie wurden daher die Verteilungscharakteristika der Scherwellen-Geschwindigkeitsmessung (ARFI) der Milz und der Einfluss von Atemmanövern untersucht. Material und Methoden: Jeweils 25 gesunde Probanden und Patienten mit Leberzirrhose (Child A) wurden mit konventioneller Sonografie, transienter Elastografie und ARFI der Leber charakterisiert. Die Milzsteifigkeit wurde mittels ARFI in definierten Atempositionen untersucht: Atemruhe in Expiration (exp.) und tiefe Inspiration (insp.). Dabei wurden jeweils 20 ARFI-Messungen durchgeführt. Die Ergebnisse wurden auf Normalverteilung getestet und die Anzahl notwendiger Messungen wurde untersucht. Ergebnisse: ARFI-Messungen der Milz waren in > 95 % der Fälle normalverteilt. 20 statt 10 Einzelmessungen reduzierten die Abweichung vom Mittelwert nach 20 Messungen in der Mehrheit der Fälle um weniger als 5 %. Die tiefe Inspiration erhöhte die Milzsteifigkeit: Probanden 2,46 ± 0,35 m/s (exp.) vs. 2,66 ± 0,36 m/s (insp.), p = 0,01; Zirrhose-Patienten 3,25 ± 0,58 m/s (exp.) vs. 3,46 ± 0,38 m/s (insp.), p = 0,03. Fälle mit hoher Milzsteifigkeit (exp.) zeigten abnehmende Werte nach Inspiration. Schlussfolgerungen: Der Mittelwert aus 10 ARFIMessungen kann als repräsentativ für die Milzsteifigkeit betrachtet werden. Für künftige Anwendungen ist eine standardisierte Atemposition erforderlich.
● ARFI ● liver cirrhosis ● portal hypertension ● spleen stiffness ● ultrasound " " " "
received accepted
29.4.2013 8.11.2013
Bibliography DOI http://dx.doi.org/ 10.1055/s-0033-1356230 Published online: 2014 Ultraschall in Med 2013; 34: 38–43 © Georg Thieme Verlag KG Stuttgart · New York · ISSN 0172-4614
Correspondence Dr. Thomas Karlas University Hospital Leipzig, Department of Medicine, Neurology and Dermatology, Division of Gastroenterology and Rheumatology Liebigstraße 20 04103 Leipzig Germany Tel.: ++ 49/03 41/9 71 22 00 Fax: ++ 49/03 41/9 71 22 09 thomas.karlas@medizin. uni-leipzig.de
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
* These two authors contributed equally to this work.
Downloaded by: University of Queensland. Copyrighted material.
2
Rapid Communication
Introduction
Ultrasound evaluation and transient elastography
!
Abdominal ultrasound, TE and ARFI were performed after fasting for at least three hours by the same investigator. The liver morphology and spleen size were examined using a convex transducer (4C1, Siemens Healthcare). The spleen volume was estimated by the formula: volume [cm³] = π/6 × (length [cm] × width [cm] × depth [cm]) [16]. We further analyzed the portal venous flow velocity (liver hilus, intercostal view) and the resistive index (RI) of spleen arterial perfusion (arterial branch at 1 cm below the spleen capsule). The study participants underwent liver stiffness measurement with transient elastography (TE; Fibroscan®, Echosens, Paris): 10 valid measurements were taken (M probe) according to the manufacturer’s recommendation. Measurements with an interquartile range (IQR) > 30 % of the median value or a success rate below 60 % were considered as invalid results.
Patients and methods !
Patient cohorts Patients with Child A liver cirrhosis were consecutively recruited within a four-month period. The diagnosis of cirrhosis was based on clinical evaluation including typical ultrasound findings (e. g. irregular liver margins, heterogeneous echogenicity, nodularity), laboratory findings and liver histology in selected cases (n = 6). Only patients with cirrhosis due to chronic viral hepatitis, alcoholic and non-alcoholic fatty liver disease, or cryptogenic etiology were included, while patients with autoimmune liver disease and biliary cirrhosis were not considered. Healthy volunteers served as the control group. Liver disease was ruled out by medical history, normal ultrasound liver morphology, liver ARFI values ≤ 1.34 m/s and transient elastography results ≤ 7.9 kPa [4]. All participants provided written informed consent. The study was approved by the local ethics committee (registration number 352 – 08). Table 1
ARFI Acoustic Radiation Force Impulse Imaging (ARFI) (Acuson S2000; Siemens Medical Solutions, Mountain View, California, USA; Software Version 350.3.044.36; convex probe 4C1, Siemens Healthcare) of the liver and spleen was performed by two experienced examiners (TK and FL) with the patient in a supine position with both arms elevated above the head. The 10 × 5 mm region of interest (ROI) was adjusted in an area without visible bile ducts (liver) and vessels. The shear-wave velocity was indicated in m/s. The examiners aimed for a measuring angle close to 0° (region of interest position in the center of the transducer surface). For the assessment of liver stiffness, ARFI was applied in the right liver lobe through the intercostal space. The measurement depth from the liver capsule was between 20 and 55 mm at a region with liver parenchyma > 6 cm. 10 valid measurements were required and the median value was used for further analysis. Valid results required a success rate of > 60 % [5]. For spleen stiffness measurement, the region of interest (ROI) was placed between the central region and the lower pole in a position near the abdominal wall via an intercostal approach " Fig. 1). This region was chosen to reduce the influence of (● measuring depth variation and measuring angle [17]. The expiration position was defined as breath hold after relaxed expiration. For deep inspiration, the patient was asked to take a breath as deep as possible and hold the breath. The examiners aimed to place the ROI at the same spleen area in both respiratory posi-
Overview of examination modalities applied in recent studies assessing spleen stiffness.
study
method
aim/endpoint(s)
patients/ region of
depth below
total
controls
measurement
capsule
measurements tive value
Takuma et al. [7]
ARFI
esophageal varices 340/16
intercostal
1 cm
Colecchia et al. [3]
TE
portal hypertension and esophageal varices
141/0
intercostal, spleen parenchyma > 4 cm
Chen et al. [8]
ARFI
liver fibrosis
182/ 0
Ye et al. [9]
ARFI
liver fibrosis, 204/60 esophageal varices
Bota et al. [10]
ARFI
Rifai et al. [11]
ARFI
Grgurevic et al. [12] Stefanescu et al. [13]
representa-
respiratory position
5
median
not specified
not specified
10
median
not specified
intercostal and subcostal
not specified
5
mean
breath hold
intercostal, center of the spleen
1 – 2 cm
median
breath hold after expiration
esophageal varices 145/0
intercostal
1 – 2 cm
10
median
breath hold
portal hypertension
100/25
not specified
not specified
10
mean
not specified
ARFI
liver fibrosis
46/20
upper pole, center and lower pole
not specified
3
mean
not specified
TE
esophageal varices 174/17
intercostal, posterior axillary line
not specified
10
median
not specified
10
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
Downloaded by: University of Queensland. Copyrighted material.
The value of liver elastography for the detection of portal hypertension and associated complications is limited [1]. Spleen elastography is a promising method for the characterization of portal hypertension in cirrhotic individuals [2]. Recent data suggest that spleen shear-wave velocity can be used to identify patients with esophageal varices independent of the etiology of cirrhosis [2, 3]. However, standardized examination procedures for spleen elastography have not been defined yet, although it is mandatory to respect defined standards in case of liver elastography [1, 4 – 6]. Spleen measurement modalities (e. g. number of measurements) vary considerably between recently published studies and limit " Table 1) [2]. Breathing maneuvers are fretheir comparability (● quently performed during ultrasound examinations to improve spleen visibility. However, although the valsalva maneuver alters liver elastography and liver stiffness is directly related to central venous pressure, no data are available on the influence of respiratory positions on spleen stiffness [14, 15]. We therefore analyzed the distribution characteristics of spleen shear-wave velocity (ARFI) and assessed the impact of respiratory maneuvers on spleen ARFI examination.
39
Rapid Communication
tions and 20 valid single measurements were taken in each breathing position (three to five measurements per breath hold).
was used for the analysis of contingency tables. A p-value < 0.05 indicated a significant difference.
Statistical analysis Statistical analyses were conducted using MedCalc 12.4 (MedCalc Software, Belgium). Clinical and laboratory data were expressed as median/range and mean ± standard deviation (SD) where appropriate. The distribution of ARFI measurements was analyzed using the Kolmogorov-Smirnov test. The nonparametric Mann-Whitney U test was chosen to compare independent parameters which were not normally distributed. Normally distributed, independent samples and paired groups were compared with the appropriate t-tests. Fisher's exact test
Results !
The baseline characteristics and results of ultrasound assessment " Table 2. Healthy probands and patients with liver are shown in ● cirrhosis had significant differences in age and BMI (p < 0.001, respectively), but showed a comparable blood pressure and heart rate during examination. ARFI and TE values of the liver, spleen size and spleen RI were increased in cirrhotic patients whereas no significant difference in portal blood flow could be detected " Table 2). compared to the control group (●
Distribution characteristics of spleen shear-wave velocity
Fig. 1 Intercostal ARFI assessment of the spleen. The region of interest was placed between the central region and the lower spleen pole close to the abdominal wall avoiding high measuring depths. The examiners aimed for an equal distance to the spleen capsule at both sites. Abb. 1 Interkostale ARFI-Messung der Milz. Die Messregion wurde zwischen der Mitte des Organs und dem unteren Pol nahe der Bauchwand positioniert, um große Messtiefen zu vermeiden. Die Untersucher bemühten sich um einen identischen Abstand zur Milzkapsel auf beiden Seiten des Organs.
healthy probands
The data distribution of 20 single ARFI values for each subject and for both respiratory positions was analyzed: results were normally distributed in 97 of 100 measurement sites. Considering only the first ten valid values, the data were normally distributed in 99 of 100 measurement sites. The minimal number of ARFI measurements for reliable spleen stiffness estimation was investigated. Therefore, the mean values obtained after increasing the number of measurements (i. e., 2, 3, 4 … 19) was calculated in relation to the mean spleen " Fig. 2). The deviastiffness obtained after 20 measurements (● tion of stiffness measurements (in % of mean) was below 5 % after seven (healthy probands) and eight (compensated cirrhosis) measurements in both respiratory positions. Individual analysis revealed a deviation of less than 5 % in > 60 % of cases after eight and in > 70 % of individuals after ten measurements " Table 3). (●
Influence of respiratory position on spleen stiffness All study participants were capable of performing the breathing maneuvers. Deep inspiration significantly increased the mean " Fig. 3, spleen shear-wave velocity values in both groups (●
patients with liver
U test
cirrhosis (Child A) sex (female/male)
11/14
11/14
n. s. 2
age (years) 1
32.3 [22.1 – 63.0]
55.7 [37.2 – 67.8]
p < 0.001
body mass index (kg/m²) 1
23.0 [18.8 – 28.9]
26.0 [18.4 – 36.8]
p = 0.005
heart rate (bpm) 1
68 [60 – 100]
68 [60 – 88]
n. s.
middle arterial pressure (mmHg) 1
93.0 [76.7 – 112.7]
94.7 [76.7 – 120.0]
n. s.
transient elastography (kPa) 1 (no. of invalid results)
4.7 [2.6 – 7.8] (one invalid result)
27.2 [8.5 – 67.8] (three invalid results)
p < 0.001
ARFI of the liver (m/s) 1 (no. of invalid results)
1.10 [0.70 – 1.34] (all valid)
2.30 [1.80 – 3.84] (one invalid result)
p < 0.001 n. s.
portal venous flow (cm/s) 1
18.5 [11.7 – 38.2]
18.9 [11.3 – 35.4]
spleen length (mm) 1
109 [84 – 127]
143 [106 – 267]
p < 0.001
spleen volume (cm³) 1
161.2 [48.0 – 227.4]
402.5 [87.7 – 1937.6]
p < 0.001
arterial resistive index (spleen) 1
0.53 [0.42 – 0.60]
0.62 [0.51 – 0.74]
p < 0.001
skin-to-spleen-capsule distance at position of ARFI measurement (mm) 1
15.9 [10.0 – 26.7]
17.7 [11.4 – 28.9]
n. s.
spleen diameter at position of ARFI measurement (mm) 1
33.7 [21.0 – 49.0]
50.0 [24.2 – 73.4]
p < 0.001
ROI depth below spleen capsule (mm) 1
17.2 [11.0 – 35.0]
20.2 [9.7 – 40.0]
n. s.
ROI – ARFI region of interest 1 median value; range 2 Fisher’s exact test
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
Table 2 Characteristics of the study cohort.
Downloaded by: University of Queensland. Copyrighted material.
40
Rapid Communication
41
Table 3 Individual deviation of spleen stiffness measurement compared to the mean value after 20 measurements.
no. of measure-
n=5
n=8
n = 10
n = 15
expiration
13 (52 %)
15 (60 %)
21 (84 %)
22 (88 %)
inspiration
13 (52 %)
16 (64 %)
19 (76 %)
24 (96 %)
expiration
13 (52 %)
15 (60 %)
18 (72 %)
23 (92 %)
inspiration
13 (52 %)
17 (68 %)
19 (76 %)
23 (92 %)
ments healthy probands 1
cirrhotic patients 1
Cases with a deviation of less than 5 % of the mean value after 20 measurements
and ARFI), portal venous flow and spleen RI could be observed between these control subgroups. We further analyzed ARFI alterations of individual study participants: 13 healthy probands and 12 cirrhotic patients showed significant spleen ARFI differences between the two respiratory " Fig. 3) independent of positions (p < 0.05, pairwise t-test; ● age, sex, BMI, arterial pressure, and heart rate. However, significant respiratory ARFI variation was associated with a smaller spleen length (135.0 ± 42.4 vs. 159.0 ± 20.6 mm, p = 0.008) and volume (415.6 ± 490.0 vs. 558.4 ± 209.4 cm³, p = 0.007) in cirrhotic patients.
Discussion !
Fig. 2 Appropriate number of ARFI spleen measurements. The mean values for increasing numbers of measurements were analyzed. The deviation from the final mean was below 5 % after seven (healthy probands) and eight (Child A cirrhosis) measurements in both respiratory positions. Abb. 2 Benötigte Anzahl an ARFI-Messungen der Milz. Es wurden jeweils Mittelwerte für eine steigende Anzahl an Messungen analysiert. Die Abweichung vom endgültigen Mittelwert nach 20 Messungen fiel in beiden Atempositionen nach sieben (gesunde Probanden) bzw. acht (Patienten mit Zirrhose) Messungen unter 5 %.
● Table 4). The median ARFI alteration was 0.20 (range –0.45 – "
1.13) m/s in healthy individuals and 0.21 (range –0.55 – 1.12) m/ s in cirrhotic patients. The ARFI success rates and variation of single ARFI values did not differ significantly between both re" Table 4). spiratory positions (● The subgroup of cirrhotic patients with an inspiratory decrease of spleen stiffness (n = 8) had higher spleen stiffness values in a relaxed expiration position compared to cases with an inspiratory increase (3.80 ± 0.55 vs. 3.00 ± 0.40 m/s; p = 0.003). In addition, these patients had higher TE values (37.6 ± 7.6 vs. 25.3 ± 14.2 kPa; p = 0.01) as well as a tendency toward higher liver ARFI results (2.80 ± 0.37 vs. 2.39 ± 0.56 m/s; p = 0.11). Likewise, the spleen RI values (0.66 ± 0.06 vs. 0.62 ± 0.06, p = 0.15) and portal venous blood flow velocity (22.3 ± 6.0 vs. 17.6 ± 3.2 cm/s, p = 0.06) were slightly but not significantly higher in these cases. Healthy controls with an inspiratory spleen stiffness decrease (n = 8) showed a similar trend for higher spleen stiffness values in the relaxed expiration position (2.63 ± 0.38 vs. 2.38 ± 0.32 m/s; p = 0.08). However, no significant difference in liver stiffness (TE
Recent studies provide proof of concept of spleen elastography as a screening method for portal hypertension and esophageal varices [2]. However, in contrast to defined standards for liver elastography [1, 4 – 6], examination modalities for spleen stiffness measurement are still under discussion [18]. Up to now, the minimal number of measurements for reliable " Table 1). To this spleen stiffness assessment is not specified (● end, we analyzed the distribution characteristics of spleen ARFI. Data of 10 and 20 subsequent single measurements were normally distributed. Thus, the mean shear-wave velocity can be considered as a representative value of spleen stiffness. In addition, the accuracy benefit of performing 20 instead of 10 single measurements is less than 5 % for both healthy individuals and cirrhotic patients. The individual measurement deviation falls below 5 % in the majority of cases after 8 single measurements. Therefore, a minimum of 8 valid ARFI measurements can be considered sufficient for the accurate assessment of spleen stiffness, whereas ≤ 7 measurements may result in a less precise stiffness estimation [7, 8, 12]. However, 10 single valid ARFI values result in a higher proportion of patients within the 5 % deviation crite" Table 3) and are in line with the recommended number of rion (● measurements for liver ARFI [4 – 6] and TE [4]. Thus, performing 10 single ARFI values per examination is a suitable approach for spleen stiffness measurement in clinical practice. Furthermore, our data reveal a significant influence of the respiratory position on spleen stiffness in healthy individuals as well as in Child A cirrhotic patients. Subgroup analysis indicates that inspiratory variation depends on spleen stiffness in the relaxed expiration position. Moreover, inspiration reduces the range of spleen stiffness values in healthy controls (1.53 to 1.25 m/s) and cirrhotic patients (2.52 to 1.50 m/s). These observations strongly suggest a respiratory-associated bias on spleen elastography if the respiratory position is not specified [3, 7, 11 – 13]. In particular, the narrow Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
Downloaded by: University of Queensland. Copyrighted material.
1
Rapid Communication
Fig. 3 Influence of respiratory position on spleen stiffness in healthy probands a and patients with liver cirrhosis b. Data are shown as box plots (median, range, 25 – 75 % quartile). Line plots display the subgroups with significant spleen stiffness differences. Abb. 3 Einfluss der Atemposition auf die Milzsteifigkeit von gesunden Probanden a und Patienten mit Leberzirrhose b. Die Daten werden als Boxplots (Median, Spanne, 25 – 75 % Quartile) sowie für die Subgruppen mit signifikanter Atemvariabilität als Liniendiagramme dargestellt.
cohort
parameter
breath hold after
deep inspiration t-test for
relaxed expiration healthy probands mean individual ARFI (mean, SD, in m/s) 2.46 ± 0.35 (n = 25) median individual ARFI (mean, SD, in m/s) 2.47 ± 0.36
liver cirrhosis child A (n = 25)
paired samples 2.66 ± 0.36
0.01
2.65 ± 0.40
0.04
ARFI success rate (mean, SD, in %)
99.43 ± 1.58
99.26 ± 2.18
n. s.
ARFI IQR/median
0.18 ± 0.08
0.18 ± 0.07
n. s. n. s. 1
no. of cases with IQR/median > 0.30
2 (8 %)
1 (4 %)
mean individual ARFI (mean, SD, in m/s)
3.25 ± 0.58
3.46 ± 0.38
0.03
median individual ARFI (mean, SD, in m/s) 3.26 ± 0.63
3.50 ± 0.42
0.03
ARFI success rate (mean, SD, in %)
94.2 ± 6.08
96.05 ± 5.00
n. s.
ARFI IQR/median
0.21 ± 0.09
0.18 ± 0.07
n. s.
no. of cases with IQR/median > 0.30
4 (16 %)
1 (4 %)
n. s. 1
Table 4 ARFI assessment of the spleen in different respiratory positions.
IQR – interquartile range 1 Fisher’s exact test
range of spleen stiffness after deep inspiration in cirrhotic patients may affect ARFI accuracy as a screening method for esophageal varices and portal hypertension. Therefore, spleen stiffness should probably be assessed during a breath hold after expiration which has a higher feasibility than deep inspiration according to our clinical experience, especially for patients with comorbidities. However, we cannot exclude that spleen stiffness assessment in the in-
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
spiration position may improve diagnostic accuracy in particular situations (e. g. assessing esophageal varices). Future studies on the noninvasive assessment of portal hypertension should therefore consider the breathing position during spleen stiffness measurement. The underlying physiologic mechanisms of our findings remain to be defined. Potentially, alterations of intra-abdominal pressure
Downloaded by: University of Queensland. Copyrighted material.
42
Rapid Communication
Acknowledgements !
This work was supported by the Federal Ministry of Education and Research (BMBF), Germany, FKZ: 01EO1001. We thank Hannelore Tenckhoff and Matthias Mende for their assistance in recruiting study participants. We further like to thank Katrin Moritz for her dedicated support in coordinating the study.
References 01 Cosgrove D, Piscaglia F, Bamber J et al. EFSUMB Guidelines and Recommendations on the Clinical Use of Ultrasound Elastography. Part 2: Clinical Applications. Ultraschall in Med 2013; DOI: 10.1055/s-00331335375 02 Chin JL, Chan G, McCormick PA. Spleen Stiffness: The New Kid on the Block for Diagnosing Portal Hypertension? Gastroenterology 2013; DOI: 10.1053/j.gastro.2013.02.047 03 Colecchia A, Montrone L, Scaioli E et al. Measurement of spleen stiffness to evaluate portal hypertension and the presence of esophageal varices in patients with HCV-related cirrhosis. Gastroenterology 2012; 143: 646 – 654 04 Friedrich-Rust M, Nierhoff J, Lupsor M et al. Performance of Acoustic Radiation Force Impulse imaging for the staging of liver fibrosis: a pooled meta-analysis. J Viral Hepat 2012; 19: e212 – e219 05 Karlas T, Pfrepper C, Wiegand J et al. Acoustic radiation force impulse imaging (ARFI) for non-invasive detection of liver fibrosis: examination standards and evaluation of interlobe differences in healthy subjects and chronic liver disease. Scand J Gastroenterol 2011; 46: 1458 – 1467 06 Goertz RS, Egger C, Neurath MF et al. Impact of Food Intake, Ultrasound Transducer, Breathing Maneuvers and Body Position on Acoustic Radiation Force Impulse (ARFI) Elastometry of the Liver. Ultraschall in Med 2012; 33: 380 – 385 07 Takuma Y, Nouso K, Morimoto Y et al. Measurement of spleen stiffness by acoustic radiation force impulse imaging identifies cirrhotic patients with esophageal varices. Gastroenterology 2013; 144: 92 – 101.e2 08 Chen S, Li Y, Lai H et al. Noninvasive assessment of liver fibrosis via spleen stiffness measurement using acoustic radiation force impulse sonoelastography in patients with chronic hepatitis B or C. J Viral Hepat 2012; 19: 654 – 663 09 Ye X, Ran H, Cheng J et al. Liver and spleen stiffness measured by acoustic radiation force impulse elastography for noninvasive assessment of liver fibrosis and esophageal varices in patients with chronic hepatitis B. J Ultrasound Med 2012; 31: 1245 – 1253 10 Bota S, Sporea I, Sirli R et al. Can ARFI elastography predict the presence of significant esophageal varices in newly diagnosed cirrhotic patients? Ann Hepatol 2012; 11: 519 – 525 11 Rifai K, Cornberg J, Bahr M et al. ARFI elastography of the spleen is inferior to liver elastography for the detection of portal hypertension. Ultraschall in Med 2011; DOI: 10.1055/s-0031-1281771 12 Grgurevic I, Cikara I, Horvat J et al. Noninvasive assessment of liver fibrosis with acoustic radiation force impulse imaging: increased liver and splenic stiffness in patients with liver fibrosis and cirrhosis. Ultraschall in Med 2011; 32: 160 – 1661 13 Stefanescu H, Grigorescu M, Lupsor M et al. Spleen stiffness measurement using Fibroscan for the noninvasive assessment of esophageal varices in liver cirrhosis patients. J Gastroenterol Hepatol 2011; 26: 164 – 170 14 Horster S, Mandel P, Zachoval R et al. Comparing acoustic radiation force impulse imaging to transient elastography to assess liver stiffness in healthy volunteers with and without valsalva manoeuvre. Clin Hemorheol Microcirc 2010; 46: 159 – 168 15 Millonig G, Friedrich S, Adolf S et al. Liver stiffness is directly influenced by central venous pressure. J Hepatol 2010; 52: 206 – 210 16 De Odorico I, Spaulding KA, Pretorius DH et al. Normal splenic volumes estimated using three-dimensional ultrasonography. J Ultrasound Med 1999; 18: 231 – 236 17 Chang S, Kim M, Kim J et al. Variability of Shear Wave Velocity using Different Frequencies in Acoustic Radiation Force Impulse (ARFI) Elastography: A Phantom and Normal Liver Study. Ultraschall in Med 2012; DOI: 10.1055/s-0032-1313008 18 Takuma Y, Morimoto Y, Yamamoto H. Reply. Gastroenterology 2013; DOI: 10.1053/j.gastro.2013.03.039. 19 Potthoff A, Attia D, Pischke S et al. Influence of different frequencies and insertion depths on the diagnostic accuracy of liver elastography by acoustic radiation force impulse imaging (ARFI). Eur J Radiol 2013; 82: 1207 – 1212 20 Abraldes JG, Reverter E, Berzigotti A. Spleen stiffness: toward a noninvasive portal sphygmomanometer? Hepatology 2013; 57: 1278 – 1280
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
Downloaded by: University of Queensland. Copyrighted material.
contribute to respiration associated spleen stiffness variation which may be less relevant in patients with splenomegaly. Furthermore, splenic arterial perfusion resistance and portal venous blood flow intensity may have contributed to the respiratory variability in cirrhotic patients although the observed differences failed to meet the level of significance. Our study has several limitations: (i) Healthy controls and cirrhotic patients show significant differences in age and body mass index which may impair comparability. However, elastography results reveal similar effects in both groups. (ii) Recent data on invasive screening for esophageal varices (upper endoscopy) and portal hypertension (measurement of hepatic venous pressure gradient) were not available for all cirrhotic patients. Hence, we were not able to assess the influence of breathing maneuvers on the association of portal hypertension and spleen stiffness. However, our patients with Child A cirrhosis represent a target cohort for screening strategies for portal hypertension [3, 7] and our observations underline the necessity of standardized spleen stiffness measure" Table 1) [2]. (iii) The relevance of the ment in clinical studies (● variation of single ARFI measurements and high interquartile ranges (IQR) has not been defined yet for spleen elastography. While some authors consider ARFI measurements with high IQR values as invalid results [7, 11], we decided not to exclude these patients because this criterion lacks validation. However, we could not detect any significant influence of respiratory position on ARFI " Table 4). (iv) Recent data indicate that ARFI accuquality criteria (● racy does not only depend on the distance between ROI and organ capsule but can also be influenced by the measurement depth from the body surface [19]. Although the examiner aimed for a comparable measurement position throughout the whole ARFI examination of the spleen, slight adjustments of measurement depth during the breathing maneuvers and variations in skin-to-spleen distances between single individuals may have contributed to the observed variability of spleen ARFI in our study. (v) We analyzed spleen stiffness based on shear-wave velocity only. Among other technique principles for spleen stiffness measurement, transient elastography can be used with promising results in selected patients [2, 3] but its clinical value is limited by low feasibility [20]. Based on our results, standardized respiratory positions should be evaluated for spleen TE as well. In conclusion, spleen ARFI assessment with a sufficient number of measurements in a defined respiratory position may improve ARFI accuracy for the detection of portal hypertension and esophageal varices. Further research on quality criteria for spleen elastography (e. g. definition of “invalid measurements”) is required.
43
Rapid Communication
Assessment of Spleen Stiffness Using Acoustic Radiation Force Impulse Imaging (ARFI): Definition of Examination Standards and Impact of Breathing Maneuvers Messung der Milzsteifigkeit mit Acoustic Radiation Force Impulse Imaging (ARFI): Definition von Untersuchungsstandards und Einfluss von Atemmanövern Authors
T. Karlas1, 2*, F. Lindner2*, M. Tröltzsch2, 3, V. Keim2, 3
Affiliations
1
3
Leipzig University Medical Center, IFB Adiposity Diseases, Leipzig, Germany University Hospital Leipzig, Department of Medicine, Neurology and Dermatology, Division of Gastroenterology and Rheumatology, Germany University Hospital Leipzig, Interdisciplinary Ultrasound Unit, Leipzig, Germany
Key words
Abstract
Zusammenfassung
"
!
!
Purpose: Spleen elastography is a promising method for the characterization of portal hypertension in cirrhotic individuals. However, standardized examination procedures for spleen stiffness measurement have not been defined yet. We analyzed the distribution characteristics of spleen shear-wave velocity (ARFI) and assessed the influence of the respiratory position on spleen stiffness measured by ARFI. Materials and Methods: 25 healthy probands and 25 patients with Child A liver cirrhosis were prospectively characterized with conventional ultrasound, transient elastography, liver ARFI, and underwent spleen ARFI in two respiratory positions: breath hold after expiration (exp) and deep inspiration (insp). For each position 20 single measurements were performed. The distribution of spleen ARFI values was analyzed for normality and the appropriate number of measurements for spleen stiffness estimation was investigated. Results: Spleen ARFI results were normally distributed in > 95 % of cases. Performing 20 instead of 10 single measurements resulted in < 5 % deviation from the mean value after 20 measurements in the majority of cases. Cirrhotic patients had a higher spleen stiffness compared to healthy probands (exp: 3.25 ± 0.58 vs. 2.46 ± 0.35 m/s; p < 0.001). Deep inspiration caused an overall increase in spleen stiffness in both groups: probands 2.46 ± 0.35 m/s (exp) vs. 2.66 ± 0.36 m/s (insp), p = 0.01; cirrhotics 3.25 ± 0.58 m/s (exp) vs. 3.46 ± 0.38 m/s (insp), p = 0.03. However, cases with high spleen stiffness values (exp) show decreasing ARFI values in deep inspiration. Conclusion: ARFI values of the spleen are normally distributed and the mean of 10 valid measurements can be used as a representative value. Deep inspiration significantly modulates spleen stiffness. Therefore, the respiratory position needs careful standardization.
Ziel: Die Milzelastizität ist ein vielversprechender Parameter für die Diagnostik der portalen Hypertension bei Leberzirrhose. Bislang existieren jedoch noch keine einheitlichen Untersuchungsstandards für die Messung der Milzsteifigkeit. In dieser Studie wurden daher die Verteilungscharakteristika der Scherwellen-Geschwindigkeitsmessung (ARFI) der Milz und der Einfluss von Atemmanövern untersucht. Material und Methoden: Jeweils 25 gesunde Probanden und Patienten mit Leberzirrhose (Child A) wurden mit konventioneller Sonografie, transienter Elastografie und ARFI der Leber charakterisiert. Die Milzsteifigkeit wurde mittels ARFI in definierten Atempositionen untersucht: Atemruhe in Expiration (exp.) und tiefe Inspiration (insp.). Dabei wurden jeweils 20 ARFI-Messungen durchgeführt. Die Ergebnisse wurden auf Normalverteilung getestet und die Anzahl notwendiger Messungen wurde untersucht. Ergebnisse: ARFI-Messungen der Milz waren in > 95 % der Fälle normalverteilt. 20 statt 10 Einzelmessungen reduzierten die Abweichung vom Mittelwert nach 20 Messungen in der Mehrheit der Fälle um weniger als 5 %. Die tiefe Inspiration erhöhte die Milzsteifigkeit: Probanden 2,46 ± 0,35 m/s (exp.) vs. 2,66 ± 0,36 m/s (insp.), p = 0,01; Zirrhose-Patienten 3,25 ± 0,58 m/s (exp.) vs. 3,46 ± 0,38 m/s (insp.), p = 0,03. Fälle mit hoher Milzsteifigkeit (exp.) zeigten abnehmende Werte nach Inspiration. Schlussfolgerungen: Der Mittelwert aus 10 ARFIMessungen kann als repräsentativ für die Milzsteifigkeit betrachtet werden. Für künftige Anwendungen ist eine standardisierte Atemposition erforderlich.
● ARFI ● liver cirrhosis ● portal hypertension ● spleen stiffness ● ultrasound " " " "
received accepted
29.4.2013 8.11.2013
Bibliography DOI http://dx.doi.org/ 10.1055/s-0033-1356230 Published online: 2014 Ultraschall in Med 2013; 34: 38–43 © Georg Thieme Verlag KG Stuttgart · New York · ISSN 0172-4614
Correspondence Dr. Thomas Karlas University Hospital Leipzig, Department of Medicine, Neurology and Dermatology, Division of Gastroenterology and Rheumatology Liebigstraße 20 04103 Leipzig Germany Tel.: ++ 49/03 41/9 71 22 00 Fax: ++ 49/03 41/9 71 22 09 thomas.karlas@medizin. uni-leipzig.de
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
* These two authors contributed equally to this work.
Downloaded by: University of Queensland. Copyrighted material.
2
Rapid Communication
Introduction
Ultrasound evaluation and transient elastography
!
Abdominal ultrasound, TE and ARFI were performed after fasting for at least three hours by the same investigator. The liver morphology and spleen size were examined using a convex transducer (4C1, Siemens Healthcare). The spleen volume was estimated by the formula: volume [cm³] = π/6 × (length [cm] × width [cm] × depth [cm]) [16]. We further analyzed the portal venous flow velocity (liver hilus, intercostal view) and the resistive index (RI) of spleen arterial perfusion (arterial branch at 1 cm below the spleen capsule). The study participants underwent liver stiffness measurement with transient elastography (TE; Fibroscan®, Echosens, Paris): 10 valid measurements were taken (M probe) according to the manufacturer’s recommendation. Measurements with an interquartile range (IQR) > 30 % of the median value or a success rate below 60 % were considered as invalid results.
Patients and methods !
Patient cohorts Patients with Child A liver cirrhosis were consecutively recruited within a four-month period. The diagnosis of cirrhosis was based on clinical evaluation including typical ultrasound findings (e. g. irregular liver margins, heterogeneous echogenicity, nodularity), laboratory findings and liver histology in selected cases (n = 6). Only patients with cirrhosis due to chronic viral hepatitis, alcoholic and non-alcoholic fatty liver disease, or cryptogenic etiology were included, while patients with autoimmune liver disease and biliary cirrhosis were not considered. Healthy volunteers served as the control group. Liver disease was ruled out by medical history, normal ultrasound liver morphology, liver ARFI values ≤ 1.34 m/s and transient elastography results ≤ 7.9 kPa [4]. All participants provided written informed consent. The study was approved by the local ethics committee (registration number 352 – 08). Table 1
ARFI Acoustic Radiation Force Impulse Imaging (ARFI) (Acuson S2000; Siemens Medical Solutions, Mountain View, California, USA; Software Version 350.3.044.36; convex probe 4C1, Siemens Healthcare) of the liver and spleen was performed by two experienced examiners (TK and FL) with the patient in a supine position with both arms elevated above the head. The 10 × 5 mm region of interest (ROI) was adjusted in an area without visible bile ducts (liver) and vessels. The shear-wave velocity was indicated in m/s. The examiners aimed for a measuring angle close to 0° (region of interest position in the center of the transducer surface). For the assessment of liver stiffness, ARFI was applied in the right liver lobe through the intercostal space. The measurement depth from the liver capsule was between 20 and 55 mm at a region with liver parenchyma > 6 cm. 10 valid measurements were required and the median value was used for further analysis. Valid results required a success rate of > 60 % [5]. For spleen stiffness measurement, the region of interest (ROI) was placed between the central region and the lower pole in a position near the abdominal wall via an intercostal approach " Fig. 1). This region was chosen to reduce the influence of (● measuring depth variation and measuring angle [17]. The expiration position was defined as breath hold after relaxed expiration. For deep inspiration, the patient was asked to take a breath as deep as possible and hold the breath. The examiners aimed to place the ROI at the same spleen area in both respiratory posi-
Overview of examination modalities applied in recent studies assessing spleen stiffness.
study
method
aim/endpoint(s)
patients/ region of
depth below
total
controls
measurement
capsule
measurements tive value
Takuma et al. [7]
ARFI
esophageal varices 340/16
intercostal
1 cm
Colecchia et al. [3]
TE
portal hypertension and esophageal varices
141/0
intercostal, spleen parenchyma > 4 cm
Chen et al. [8]
ARFI
liver fibrosis
182/ 0
Ye et al. [9]
ARFI
liver fibrosis, 204/60 esophageal varices
Bota et al. [10]
ARFI
Rifai et al. [11]
ARFI
Grgurevic et al. [12] Stefanescu et al. [13]
representa-
respiratory position
5
median
not specified
not specified
10
median
not specified
intercostal and subcostal
not specified
5
mean
breath hold
intercostal, center of the spleen
1 – 2 cm
median
breath hold after expiration
esophageal varices 145/0
intercostal
1 – 2 cm
10
median
breath hold
portal hypertension
100/25
not specified
not specified
10
mean
not specified
ARFI
liver fibrosis
46/20
upper pole, center and lower pole
not specified
3
mean
not specified
TE
esophageal varices 174/17
intercostal, posterior axillary line
not specified
10
median
not specified
10
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
Downloaded by: University of Queensland. Copyrighted material.
The value of liver elastography for the detection of portal hypertension and associated complications is limited [1]. Spleen elastography is a promising method for the characterization of portal hypertension in cirrhotic individuals [2]. Recent data suggest that spleen shear-wave velocity can be used to identify patients with esophageal varices independent of the etiology of cirrhosis [2, 3]. However, standardized examination procedures for spleen elastography have not been defined yet, although it is mandatory to respect defined standards in case of liver elastography [1, 4 – 6]. Spleen measurement modalities (e. g. number of measurements) vary considerably between recently published studies and limit " Table 1) [2]. Breathing maneuvers are fretheir comparability (● quently performed during ultrasound examinations to improve spleen visibility. However, although the valsalva maneuver alters liver elastography and liver stiffness is directly related to central venous pressure, no data are available on the influence of respiratory positions on spleen stiffness [14, 15]. We therefore analyzed the distribution characteristics of spleen shear-wave velocity (ARFI) and assessed the impact of respiratory maneuvers on spleen ARFI examination.
39
Rapid Communication
tions and 20 valid single measurements were taken in each breathing position (three to five measurements per breath hold).
was used for the analysis of contingency tables. A p-value < 0.05 indicated a significant difference.
Statistical analysis Statistical analyses were conducted using MedCalc 12.4 (MedCalc Software, Belgium). Clinical and laboratory data were expressed as median/range and mean ± standard deviation (SD) where appropriate. The distribution of ARFI measurements was analyzed using the Kolmogorov-Smirnov test. The nonparametric Mann-Whitney U test was chosen to compare independent parameters which were not normally distributed. Normally distributed, independent samples and paired groups were compared with the appropriate t-tests. Fisher's exact test
Results !
The baseline characteristics and results of ultrasound assessment " Table 2. Healthy probands and patients with liver are shown in ● cirrhosis had significant differences in age and BMI (p < 0.001, respectively), but showed a comparable blood pressure and heart rate during examination. ARFI and TE values of the liver, spleen size and spleen RI were increased in cirrhotic patients whereas no significant difference in portal blood flow could be detected " Table 2). compared to the control group (●
Distribution characteristics of spleen shear-wave velocity
Fig. 1 Intercostal ARFI assessment of the spleen. The region of interest was placed between the central region and the lower spleen pole close to the abdominal wall avoiding high measuring depths. The examiners aimed for an equal distance to the spleen capsule at both sites. Abb. 1 Interkostale ARFI-Messung der Milz. Die Messregion wurde zwischen der Mitte des Organs und dem unteren Pol nahe der Bauchwand positioniert, um große Messtiefen zu vermeiden. Die Untersucher bemühten sich um einen identischen Abstand zur Milzkapsel auf beiden Seiten des Organs.
healthy probands
The data distribution of 20 single ARFI values for each subject and for both respiratory positions was analyzed: results were normally distributed in 97 of 100 measurement sites. Considering only the first ten valid values, the data were normally distributed in 99 of 100 measurement sites. The minimal number of ARFI measurements for reliable spleen stiffness estimation was investigated. Therefore, the mean values obtained after increasing the number of measurements (i. e., 2, 3, 4 … 19) was calculated in relation to the mean spleen " Fig. 2). The deviastiffness obtained after 20 measurements (● tion of stiffness measurements (in % of mean) was below 5 % after seven (healthy probands) and eight (compensated cirrhosis) measurements in both respiratory positions. Individual analysis revealed a deviation of less than 5 % in > 60 % of cases after eight and in > 70 % of individuals after ten measurements " Table 3). (●
Influence of respiratory position on spleen stiffness All study participants were capable of performing the breathing maneuvers. Deep inspiration significantly increased the mean " Fig. 3, spleen shear-wave velocity values in both groups (●
patients with liver
U test
cirrhosis (Child A) sex (female/male)
11/14
11/14
n. s. 2
age (years) 1
32.3 [22.1 – 63.0]
55.7 [37.2 – 67.8]
p < 0.001
body mass index (kg/m²) 1
23.0 [18.8 – 28.9]
26.0 [18.4 – 36.8]
p = 0.005
heart rate (bpm) 1
68 [60 – 100]
68 [60 – 88]
n. s.
middle arterial pressure (mmHg) 1
93.0 [76.7 – 112.7]
94.7 [76.7 – 120.0]
n. s.
transient elastography (kPa) 1 (no. of invalid results)
4.7 [2.6 – 7.8] (one invalid result)
27.2 [8.5 – 67.8] (three invalid results)
p < 0.001
ARFI of the liver (m/s) 1 (no. of invalid results)
1.10 [0.70 – 1.34] (all valid)
2.30 [1.80 – 3.84] (one invalid result)
p < 0.001 n. s.
portal venous flow (cm/s) 1
18.5 [11.7 – 38.2]
18.9 [11.3 – 35.4]
spleen length (mm) 1
109 [84 – 127]
143 [106 – 267]
p < 0.001
spleen volume (cm³) 1
161.2 [48.0 – 227.4]
402.5 [87.7 – 1937.6]
p < 0.001
arterial resistive index (spleen) 1
0.53 [0.42 – 0.60]
0.62 [0.51 – 0.74]
p < 0.001
skin-to-spleen-capsule distance at position of ARFI measurement (mm) 1
15.9 [10.0 – 26.7]
17.7 [11.4 – 28.9]
n. s.
spleen diameter at position of ARFI measurement (mm) 1
33.7 [21.0 – 49.0]
50.0 [24.2 – 73.4]
p < 0.001
ROI depth below spleen capsule (mm) 1
17.2 [11.0 – 35.0]
20.2 [9.7 – 40.0]
n. s.
ROI – ARFI region of interest 1 median value; range 2 Fisher’s exact test
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
Table 2 Characteristics of the study cohort.
Downloaded by: University of Queensland. Copyrighted material.
40
Rapid Communication
41
Table 3 Individual deviation of spleen stiffness measurement compared to the mean value after 20 measurements.
no. of measure-
n=5
n=8
n = 10
n = 15
expiration
13 (52 %)
15 (60 %)
21 (84 %)
22 (88 %)
inspiration
13 (52 %)
16 (64 %)
19 (76 %)
24 (96 %)
expiration
13 (52 %)
15 (60 %)
18 (72 %)
23 (92 %)
inspiration
13 (52 %)
17 (68 %)
19 (76 %)
23 (92 %)
ments healthy probands 1
cirrhotic patients 1
Cases with a deviation of less than 5 % of the mean value after 20 measurements
and ARFI), portal venous flow and spleen RI could be observed between these control subgroups. We further analyzed ARFI alterations of individual study participants: 13 healthy probands and 12 cirrhotic patients showed significant spleen ARFI differences between the two respiratory " Fig. 3) independent of positions (p < 0.05, pairwise t-test; ● age, sex, BMI, arterial pressure, and heart rate. However, significant respiratory ARFI variation was associated with a smaller spleen length (135.0 ± 42.4 vs. 159.0 ± 20.6 mm, p = 0.008) and volume (415.6 ± 490.0 vs. 558.4 ± 209.4 cm³, p = 0.007) in cirrhotic patients.
Discussion !
Fig. 2 Appropriate number of ARFI spleen measurements. The mean values for increasing numbers of measurements were analyzed. The deviation from the final mean was below 5 % after seven (healthy probands) and eight (Child A cirrhosis) measurements in both respiratory positions. Abb. 2 Benötigte Anzahl an ARFI-Messungen der Milz. Es wurden jeweils Mittelwerte für eine steigende Anzahl an Messungen analysiert. Die Abweichung vom endgültigen Mittelwert nach 20 Messungen fiel in beiden Atempositionen nach sieben (gesunde Probanden) bzw. acht (Patienten mit Zirrhose) Messungen unter 5 %.
● Table 4). The median ARFI alteration was 0.20 (range –0.45 – "
1.13) m/s in healthy individuals and 0.21 (range –0.55 – 1.12) m/ s in cirrhotic patients. The ARFI success rates and variation of single ARFI values did not differ significantly between both re" Table 4). spiratory positions (● The subgroup of cirrhotic patients with an inspiratory decrease of spleen stiffness (n = 8) had higher spleen stiffness values in a relaxed expiration position compared to cases with an inspiratory increase (3.80 ± 0.55 vs. 3.00 ± 0.40 m/s; p = 0.003). In addition, these patients had higher TE values (37.6 ± 7.6 vs. 25.3 ± 14.2 kPa; p = 0.01) as well as a tendency toward higher liver ARFI results (2.80 ± 0.37 vs. 2.39 ± 0.56 m/s; p = 0.11). Likewise, the spleen RI values (0.66 ± 0.06 vs. 0.62 ± 0.06, p = 0.15) and portal venous blood flow velocity (22.3 ± 6.0 vs. 17.6 ± 3.2 cm/s, p = 0.06) were slightly but not significantly higher in these cases. Healthy controls with an inspiratory spleen stiffness decrease (n = 8) showed a similar trend for higher spleen stiffness values in the relaxed expiration position (2.63 ± 0.38 vs. 2.38 ± 0.32 m/s; p = 0.08). However, no significant difference in liver stiffness (TE
Recent studies provide proof of concept of spleen elastography as a screening method for portal hypertension and esophageal varices [2]. However, in contrast to defined standards for liver elastography [1, 4 – 6], examination modalities for spleen stiffness measurement are still under discussion [18]. Up to now, the minimal number of measurements for reliable " Table 1). To this spleen stiffness assessment is not specified (● end, we analyzed the distribution characteristics of spleen ARFI. Data of 10 and 20 subsequent single measurements were normally distributed. Thus, the mean shear-wave velocity can be considered as a representative value of spleen stiffness. In addition, the accuracy benefit of performing 20 instead of 10 single measurements is less than 5 % for both healthy individuals and cirrhotic patients. The individual measurement deviation falls below 5 % in the majority of cases after 8 single measurements. Therefore, a minimum of 8 valid ARFI measurements can be considered sufficient for the accurate assessment of spleen stiffness, whereas ≤ 7 measurements may result in a less precise stiffness estimation [7, 8, 12]. However, 10 single valid ARFI values result in a higher proportion of patients within the 5 % deviation crite" Table 3) and are in line with the recommended number of rion (● measurements for liver ARFI [4 – 6] and TE [4]. Thus, performing 10 single ARFI values per examination is a suitable approach for spleen stiffness measurement in clinical practice. Furthermore, our data reveal a significant influence of the respiratory position on spleen stiffness in healthy individuals as well as in Child A cirrhotic patients. Subgroup analysis indicates that inspiratory variation depends on spleen stiffness in the relaxed expiration position. Moreover, inspiration reduces the range of spleen stiffness values in healthy controls (1.53 to 1.25 m/s) and cirrhotic patients (2.52 to 1.50 m/s). These observations strongly suggest a respiratory-associated bias on spleen elastography if the respiratory position is not specified [3, 7, 11 – 13]. In particular, the narrow Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
Downloaded by: University of Queensland. Copyrighted material.
1
Rapid Communication
Fig. 3 Influence of respiratory position on spleen stiffness in healthy probands a and patients with liver cirrhosis b. Data are shown as box plots (median, range, 25 – 75 % quartile). Line plots display the subgroups with significant spleen stiffness differences. Abb. 3 Einfluss der Atemposition auf die Milzsteifigkeit von gesunden Probanden a und Patienten mit Leberzirrhose b. Die Daten werden als Boxplots (Median, Spanne, 25 – 75 % Quartile) sowie für die Subgruppen mit signifikanter Atemvariabilität als Liniendiagramme dargestellt.
cohort
parameter
breath hold after
deep inspiration t-test for
relaxed expiration healthy probands mean individual ARFI (mean, SD, in m/s) 2.46 ± 0.35 (n = 25) median individual ARFI (mean, SD, in m/s) 2.47 ± 0.36
liver cirrhosis child A (n = 25)
paired samples 2.66 ± 0.36
0.01
2.65 ± 0.40
0.04
ARFI success rate (mean, SD, in %)
99.43 ± 1.58
99.26 ± 2.18
n. s.
ARFI IQR/median
0.18 ± 0.08
0.18 ± 0.07
n. s. n. s. 1
no. of cases with IQR/median > 0.30
2 (8 %)
1 (4 %)
mean individual ARFI (mean, SD, in m/s)
3.25 ± 0.58
3.46 ± 0.38
0.03
median individual ARFI (mean, SD, in m/s) 3.26 ± 0.63
3.50 ± 0.42
0.03
ARFI success rate (mean, SD, in %)
94.2 ± 6.08
96.05 ± 5.00
n. s.
ARFI IQR/median
0.21 ± 0.09
0.18 ± 0.07
n. s.
no. of cases with IQR/median > 0.30
4 (16 %)
1 (4 %)
n. s. 1
Table 4 ARFI assessment of the spleen in different respiratory positions.
IQR – interquartile range 1 Fisher’s exact test
range of spleen stiffness after deep inspiration in cirrhotic patients may affect ARFI accuracy as a screening method for esophageal varices and portal hypertension. Therefore, spleen stiffness should probably be assessed during a breath hold after expiration which has a higher feasibility than deep inspiration according to our clinical experience, especially for patients with comorbidities. However, we cannot exclude that spleen stiffness assessment in the in-
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
spiration position may improve diagnostic accuracy in particular situations (e. g. assessing esophageal varices). Future studies on the noninvasive assessment of portal hypertension should therefore consider the breathing position during spleen stiffness measurement. The underlying physiologic mechanisms of our findings remain to be defined. Potentially, alterations of intra-abdominal pressure
Downloaded by: University of Queensland. Copyrighted material.
42
Rapid Communication
Acknowledgements !
This work was supported by the Federal Ministry of Education and Research (BMBF), Germany, FKZ: 01EO1001. We thank Hannelore Tenckhoff and Matthias Mende for their assistance in recruiting study participants. We further like to thank Katrin Moritz for her dedicated support in coordinating the study.
References 01 Cosgrove D, Piscaglia F, Bamber J et al. EFSUMB Guidelines and Recommendations on the Clinical Use of Ultrasound Elastography. Part 2: Clinical Applications. Ultraschall in Med 2013; DOI: 10.1055/s-00331335375 02 Chin JL, Chan G, McCormick PA. Spleen Stiffness: The New Kid on the Block for Diagnosing Portal Hypertension? Gastroenterology 2013; DOI: 10.1053/j.gastro.2013.02.047 03 Colecchia A, Montrone L, Scaioli E et al. Measurement of spleen stiffness to evaluate portal hypertension and the presence of esophageal varices in patients with HCV-related cirrhosis. Gastroenterology 2012; 143: 646 – 654 04 Friedrich-Rust M, Nierhoff J, Lupsor M et al. Performance of Acoustic Radiation Force Impulse imaging for the staging of liver fibrosis: a pooled meta-analysis. J Viral Hepat 2012; 19: e212 – e219 05 Karlas T, Pfrepper C, Wiegand J et al. Acoustic radiation force impulse imaging (ARFI) for non-invasive detection of liver fibrosis: examination standards and evaluation of interlobe differences in healthy subjects and chronic liver disease. Scand J Gastroenterol 2011; 46: 1458 – 1467 06 Goertz RS, Egger C, Neurath MF et al. Impact of Food Intake, Ultrasound Transducer, Breathing Maneuvers and Body Position on Acoustic Radiation Force Impulse (ARFI) Elastometry of the Liver. Ultraschall in Med 2012; 33: 380 – 385 07 Takuma Y, Nouso K, Morimoto Y et al. Measurement of spleen stiffness by acoustic radiation force impulse imaging identifies cirrhotic patients with esophageal varices. Gastroenterology 2013; 144: 92 – 101.e2 08 Chen S, Li Y, Lai H et al. Noninvasive assessment of liver fibrosis via spleen stiffness measurement using acoustic radiation force impulse sonoelastography in patients with chronic hepatitis B or C. J Viral Hepat 2012; 19: 654 – 663 09 Ye X, Ran H, Cheng J et al. Liver and spleen stiffness measured by acoustic radiation force impulse elastography for noninvasive assessment of liver fibrosis and esophageal varices in patients with chronic hepatitis B. J Ultrasound Med 2012; 31: 1245 – 1253 10 Bota S, Sporea I, Sirli R et al. Can ARFI elastography predict the presence of significant esophageal varices in newly diagnosed cirrhotic patients? Ann Hepatol 2012; 11: 519 – 525 11 Rifai K, Cornberg J, Bahr M et al. ARFI elastography of the spleen is inferior to liver elastography for the detection of portal hypertension. Ultraschall in Med 2011; DOI: 10.1055/s-0031-1281771 12 Grgurevic I, Cikara I, Horvat J et al. Noninvasive assessment of liver fibrosis with acoustic radiation force impulse imaging: increased liver and splenic stiffness in patients with liver fibrosis and cirrhosis. Ultraschall in Med 2011; 32: 160 – 1661 13 Stefanescu H, Grigorescu M, Lupsor M et al. Spleen stiffness measurement using Fibroscan for the noninvasive assessment of esophageal varices in liver cirrhosis patients. J Gastroenterol Hepatol 2011; 26: 164 – 170 14 Horster S, Mandel P, Zachoval R et al. Comparing acoustic radiation force impulse imaging to transient elastography to assess liver stiffness in healthy volunteers with and without valsalva manoeuvre. Clin Hemorheol Microcirc 2010; 46: 159 – 168 15 Millonig G, Friedrich S, Adolf S et al. Liver stiffness is directly influenced by central venous pressure. J Hepatol 2010; 52: 206 – 210 16 De Odorico I, Spaulding KA, Pretorius DH et al. Normal splenic volumes estimated using three-dimensional ultrasonography. J Ultrasound Med 1999; 18: 231 – 236 17 Chang S, Kim M, Kim J et al. Variability of Shear Wave Velocity using Different Frequencies in Acoustic Radiation Force Impulse (ARFI) Elastography: A Phantom and Normal Liver Study. Ultraschall in Med 2012; DOI: 10.1055/s-0032-1313008 18 Takuma Y, Morimoto Y, Yamamoto H. Reply. Gastroenterology 2013; DOI: 10.1053/j.gastro.2013.03.039. 19 Potthoff A, Attia D, Pischke S et al. Influence of different frequencies and insertion depths on the diagnostic accuracy of liver elastography by acoustic radiation force impulse imaging (ARFI). Eur J Radiol 2013; 82: 1207 – 1212 20 Abraldes JG, Reverter E, Berzigotti A. Spleen stiffness: toward a noninvasive portal sphygmomanometer? Hepatology 2013; 57: 1278 – 1280
Karlas T et al. Assessment of Spleen … Ultraschall in Med 2013; 34: 38–43
Downloaded by: University of Queensland. Copyrighted material.
contribute to respiration associated spleen stiffness variation which may be less relevant in patients with splenomegaly. Furthermore, splenic arterial perfusion resistance and portal venous blood flow intensity may have contributed to the respiratory variability in cirrhotic patients although the observed differences failed to meet the level of significance. Our study has several limitations: (i) Healthy controls and cirrhotic patients show significant differences in age and body mass index which may impair comparability. However, elastography results reveal similar effects in both groups. (ii) Recent data on invasive screening for esophageal varices (upper endoscopy) and portal hypertension (measurement of hepatic venous pressure gradient) were not available for all cirrhotic patients. Hence, we were not able to assess the influence of breathing maneuvers on the association of portal hypertension and spleen stiffness. However, our patients with Child A cirrhosis represent a target cohort for screening strategies for portal hypertension [3, 7] and our observations underline the necessity of standardized spleen stiffness measure" Table 1) [2]. (iii) The relevance of the ment in clinical studies (● variation of single ARFI measurements and high interquartile ranges (IQR) has not been defined yet for spleen elastography. While some authors consider ARFI measurements with high IQR values as invalid results [7, 11], we decided not to exclude these patients because this criterion lacks validation. However, we could not detect any significant influence of respiratory position on ARFI " Table 4). (iv) Recent data indicate that ARFI accuquality criteria (● racy does not only depend on the distance between ROI and organ capsule but can also be influenced by the measurement depth from the body surface [19]. Although the examiner aimed for a comparable measurement position throughout the whole ARFI examination of the spleen, slight adjustments of measurement depth during the breathing maneuvers and variations in skin-to-spleen distances between single individuals may have contributed to the observed variability of spleen ARFI in our study. (v) We analyzed spleen stiffness based on shear-wave velocity only. Among other technique principles for spleen stiffness measurement, transient elastography can be used with promising results in selected patients [2, 3] but its clinical value is limited by low feasibility [20]. Based on our results, standardized respiratory positions should be evaluated for spleen TE as well. In conclusion, spleen ARFI assessment with a sufficient number of measurements in a defined respiratory position may improve ARFI accuracy for the detection of portal hypertension and esophageal varices. Further research on quality criteria for spleen elastography (e. g. definition of “invalid measurements”) is required.
43
