The development of esophageal varices as a consequence of portal hypertension is a serious complication of liver cirrhosis in children (1), with reported mortality rates for variceal bleeds ranging from 5% to 20% (2,3). In adults, primary prophylaxis of variceal hemorrhage using either β-blockers or prophylactic ligation of varices (esophageal varices ligation) is well established (4). In children there is presently no consensus among experts on whether primary prophylaxis to prevent variceal hemorrhage should be performed at all, and if so, which patients would be eligible (3). Research that addresses these open questions is impeded by several barriers that have been highlighted by Ling et al. These barriers include, among other issues, the fact that the epidemiology and the natural history of varices in children with chronic liver disease (LD) are not clear. To date, the only certain way to diagnose esophageal varices is still by endoscopy. Only few centers perform routine screening endoscopies, so the true prevalence of varices in children with advanced chronic LD is not known. As a consequence, the true incidence of bleeding in existing varices and the rate of progression of severity of varices are equally unknown.
A noninvasive tool to diagnose the presence and degree of esophageal varices would be highly useful to overcome the research barriers quoted above. Several noninvasive markers of portal hypertension have been investigated in adults for their potential to predict the presence of varices and the rate of complications. These include the presence and degree of splenomegaly (5,6), platelet count, platelet-to-spleen size z score ratio (5,7,8), serological markers of liver fibrosis such as the Fibrotest (9,10), and measurements of liver and spleen stiffness by acoustic radiation force impulse imaging (ARFI) or transient elastography (TE; FibroScan) (1,2,11,12). With the exception of a few studies describing TE of the liver for the prediction of esophageal varices (13–15), none of these studies have been performed in children. Three studies that investigated spleen stiffness measurement (SSM) using TE in adults showed that TE of the spleen is feasible and can predict the presence of esophageal varices (16–18). TE of the spleen appears to be attractive in children because TE is an easy-to-use, pain-free technique, which is not per se subject to the age-specific variations that impede the use of parameters such as spleen size or platelet-to-spleen size ratio. The aim of our pilot study, therefore, was to test the feasibility of SSM by FibroScan in children and to generate first data on its diagnostic use for the assessment of esophageal varices and their complications.
Between November 2011 and December 2012, 99 children (58 girls, 41 boys, 62 with chronic LD [ages 1.9–17.3, median 10.6 years], 6 children after liver transplantation [LTX; ages 6.3–15.4, median 11.1 years], 31 controls [ages 4.1–17, median 11.6 years]) underwent SSM using TE (FibroScan; Echosens, Paris, France). Demographic data of patients and controls are presented in Table 1. Within the group of patients with chronic LDs, 37 blood samples for Fibrotest were obtained simultaneously with SSM, 45 patients had an additional liver stiffness measurement (LSM), and 19 patients underwent an upper endoscopy for the assessment of esophageal varices simultaneously to the SSM. Clinical data of the patients with chronic LDs are summarized in Table 2. Patients with chronic LDs and transplanted patients were recruited either during a routine follow-up visit at our clinic or during an inpatient stay for endoscopy. Controls either were recruited in the department of pediatric surgery (minor surgical procedures, n = 20) or were inpatients with general pediatric issues (bronchial asthma, epileptic seizure, acute alcoholic intoxication) or with planned endoscopic evaluation of chronic inflammatory bowel disease or celiac disease (n = 11). Written informed consent was obtained from the children's parents and the children themselves if appropriate. The study protocol was approved by the ethics committee of Hanover Medical School.
Transient Elastography of the Spleen Via FibroScan
TE was originally developed for the examination of liver stiffness to quantify hepatic fibrosis (19–22). Liver stiffness is calculated from the propagation velocity of a pressure-induced shear wave. The results are given in kilopascals. The maximum value measured by the FibroScan is 75 kPa. For SSM the technique was adapted as follows.
The patients were in a supine position with the left arm in maximal abduction. After localizing the spleen using ultrasound, the measurement probe was positioned in an intercostal space, if possible above the middle part of the spleen with a tissue thickness under the probe of >4 cm. Two different FibroScan probes are available (S and M), offering 3 different measurement depths: S1, 15 to 40 mm; S2, 20 to 50 mm; and M, 25 to 65 mm. Probe choice for SSM was based on the spleen thickness visualized on B-mode ultrasound, and the largest possible probe was used. All measurements were performed by the same examiner (C.B.). The results are given as the median of 10 subsequent measurements and were accepted as valid if the ratio of the interquartile range (IQR) and median was <30%. For the examination of intraobserver variability, 15 subjects had 2 consecutive measurements directly one after the other and 11 subjects had SSM examinations by the same examiner on 2 consecutive days.
LSM was performed with the probe in an intercostal space in the anterior axillar line, after identifying the position of the liver via percussion. The results are also given as median of 10 measurements and were deemed to be successful if the success rate was >60% and the ratio of IQR and median was <30%.
The Fibrotest score is calculated from 5 different serological markers (total bilirubin, haptoglobin, γ-glutamyl transferase, apolipoprotein A1, α2-macroglobulin), sex, and age. Algorithms for the calculation were provided by BioPredictive (Paris, France). The score can reach values between 0 and 1. Corresponding histological degrees of hepatic fibrosis are ascribed based on validation data (23).
In most cases (14/19) upper endoscopy was performed in patients with known varices and/or a history of upper gastroesophageal bleeding to determine the present status of esophageal varices. Four of 19 patients underwent upper endoscopy for screening purposes as part of a diagnostic workup for hepatopathy of unknown etiology, and in 1 of 19 cases, the finding of varices was an incidental finding during an upper endoscopy in the context of inflammatory bowel disease (the patient was later found to have portal vein thrombosis). The varices were graded as follows: grade I, varices <5 mm, flattened by air insufflation; grade II, varices >5 mm, no luminal obstruction; grade III, great, winding veins, considerable luminal obstruction; grade IV, almost complete occlusion of the lumen (24). All endoscopies were performed under general anesthesia.
Spleen size was determined by ultrasound as the longest pole-to-pole diameter at the level of the spleen hilus. Care was taken only to include measurements performed in the correct plane. To quantify the degree of splenomegaly normalized for differences because of age, the ratio of measured spleen size and upper degree of normal for height was calculated (25).
SPSS version 20 (IBM SPSS Statistics, Armonk, NY) was used for statistical analysis. Continuous variables are presented as median and range. The Mann-Whitney U test was used to compare continuous variables between different patient subgroups. Correlations between continuous variables were expressed by using Pearson correlation coefficient. Paired Student t test for dependent samples was used to compare paired data. A P value ≤0.05 was considered statistically significant.
Feasibility of SSM
The first aim of our study was to test the feasibility of SSM using FibroScan TE in children. All of the patients with LD and all controls (N = 93) were included for determining success rates. SSM was technically possible in 89 patients (96%), meaning that the probe could be placed above the spleen and that the FibroScan was able to give stiffness values (Fig. 1A). Reasons for technical failure were high body mass index in n = 1, lack of cooperation in n = 1, and technical impossibility to reliably place the probe above the spleen in n = 2 (anatomical spleen position, breathing). Valid results for SSM with an IQR/median <30% could be obtained only in n = 71 subjects (76.3%). Success rates for valid results tended to be higher in patients with LD compared with controls (83.3% vs 72.4%, P = 0.27 n.s.) (Fig. 1B). Spleen size was found to be the major influencing factor in determining the success rate, with 90.5% valid measurements in patients with splenomegaly, compared with only 66.7% in patients with normal spleen size (P = 0.02) (Fig. 1C). Twenty-six subjects received 2 subsequent SSMs by the same observer, with a good intraobserver correlation of r = 0.946 (P < 0.001).
Diagnostic Value of SSM
The median value for SSM in patients with chronic LDs was 54.5 kPa (range 12.8–75), which was significantly higher compared with controls (median 16.3 kPa, range 6.8–38.2, P < 0.001). Within the group of patients with chronic LDs, spleen stiffness values were significantly higher in patients with splenomegaly compared with patients with normal spleen size (62.96 kPa, range 16.5–75 vs 18.4 kPa, range 12.8–42.2, P < 0.001). SSM results correlated strongly with both absolute spleen size (r = 0.692, P < 0.001) and with the height-adapted degree of splenomegaly (r = 0.799, P < 0.001, Fig. 2A–C).
Presence and Degree of Varices
Of the 19 patients who underwent an upper endoscopy, patients with varices presently present showed a significantly higher SSM than patients without varices (median 75 kPa, range 44–75 vs 24 kPa, 20.75–63.9, P < 0.01) (Fig. 3). Patients (n = 4) who received endoscopy, but had no history of varices whatsoever had a median SSM value of 22.85 kPa (range 20.75–41.5). SSM correlated strongly with the current grade of varices (r = 0.598, P < 0.01) and even stronger with the maximum grade of varices in the respective patient's history (r = 0.861, P < 0.001).
Within the group of patients who underwent an upper endoscopy, 9 patients had a history of variceal bleeding. Their median value for SSM was 75 kPa (range 63.9–75) versus 50.25 kPa (range 20.75–75) in 10 patients without a history of bleeding (P < 0.05). The frequency of 75 kPa results in patients with different severity of complications of portal hypertension is shown in Figure 3C, and was highest in patients with a history of variceal bleeding. The maximum stiffness result the FibroScan machine will deliver is 75 kPa. Variceal bleeding did not occur in patients with SSM results <60 kPa.
Comparison of SSM Results With LSM and Fibrotest
Compared with SSM, the assessment of liver stiffness by TE as well as serological assessment of hepatic fibrosis by Fibrotest did not discriminate between patients with and without varices or variceal hemorrhage. This may be the result of extremely small patient numbers in these respective groups. In contrast, both liver stiffness and Fibrotest scores were significantly higher in patients with splenomegaly versus patients with normal spleen size. The results are summarized in Table 3. Children with prehepatic block (portal vein thrombosis) were excluded from the analysis for Fibrotest and liver stiffness results.
Spleen Stiffness After LTX
Three patients with splenomegaly received serial SSM before and 1 to 12 months after LTX. Data are summarized in Table 4, and show a marked decrease in spleen stiffness after successful transplantation. An additional 6 patients underwent a single SSM after LTX. These 9 patients together had a median stiffness value of 27.50 kPa (range 10–40) after transplantation, which was significantly higher compared with the group of controls (P < 0.05) and significantly lower compared with patients with chronic LD and splenomegaly (P < 0.001). Five patients in this group had persistent splenomegaly after LTX.
The aim of our study was to test the applicability of SSM by FibroScan elastography as a new noninvasive marker for assessment of portal hypertension in children. Although in adults, SSM by FibroScan has been found feasible and of reasonable diagnostic value in 3 recent studies (16–18), data on the applicability in children have not been published so far. There is no consensus among pediatric experts about the criteria for elective endoscopies in children with portal hypertension. The need for noninvasive markers to predict the occurrence of portal hypertension-related complications, to focus the use of upper endoscopy on those patients with a high risk for variceal bleeding, has been emphasized repeatedly (3,11). Spleen elastography as a simple, noninvasive method appears to be an attractive candidate for such a marker.
We found SSM using the FibroScan to be technically feasible and reproducible in most children. Measurements were easier and more successful in children with splenomegaly. Although this gives some indication of anatomical limitations and renders the collection of normal spleen stiffness values more difficult, it does not represent an important technical limitation in the patient population of interest. Further studies that investigate possible influencing factors on spleen stiffness results, such as presence of obesity or ascites, are desirable to further refine knowledge on the technical applicability of SSMs.
Splenomegaly in portal hypertension appears to result not only from passive blood congestion but also from tissue fibrosis and tissue hyperplasia (6). It appears impossible to differentiate the 2 when performing spleen elastography. In LSMs, alterations of stiffness results secondary to increased post-prandial portal blood flow have been demonstrated in both adults and children (22,25a). The influence of blood congestion on spleen stiffness results, therefore, has to be borne in mind when interpreting results. Similarly, the presence of spleen fibrosis after long-standing portal hypertension may equally lead to falsely increased SSM values. The latter is supported by our results from patients with transplanted livers. Spleen stiffness decreased markedly after LTX in patients who underwent serial measurements, but remained elevated compared with controls. Five of 9 patients still had some degree of splenomegaly at the time of the investigation. This persistent increase in spleen stiffness may result from persistent spleen fibrosis in the absence of true portal hypertension. This suggests that the technique should not be used for assessment of portal hypertension in transplanted patients.
With regard to the diagnostic value of SSM by FibroScan, our data show that increased spleen stiffness reflected the presence of splenomegaly, the presence of varices, and the occurrence of variceal bleeding in the patient's history. Although the assessment of splenomegaly by ultrasound can be subject to some degree of error resulting from observer dependency, we believe it reflects the severity of LD adequately for the purpose of our study. Children with chronic LD but normal spleen size had a median value for SSM similar to controls. Variceal bleeding only occurred in children with SSM >60 kPa. These findings suggest a promising diagnostic value of SSMs. Two important limitations, however, need to be considered. The first limitation is the patient sample we used. Because screening endoscopies are not performed routinely in all of the children with chronic LD in our hospital, there is a strong bias in patient selection in that most patients with endoscopy results had known varices, and 50% had a history of variceal bleeding. We do not know the incidence of varices in those patients with LD who had signs of portal hypertension (ie, splenomegaly), but never underwent endoscopy. A prospective study using screening endoscopies would help to better delineate the correlation of spleen stiffness with lower degrees of varices and to truly define cutoff values that indicate increased risk of complications. Looking at our data, it is extremely likely that some of the patients with chronic LD who did not undergo endoscopy will have significant varices.
A second limitation lies in the fact that the FibroScan machine will not deliver stiffness values beyond 75 kPa. True spleen stiffness values are likely to be higher in a significant number of those patients whose result was 75 kPa. This finding suggests that a different elastography technique without upper limit, for example, ARFI, may be more useful for the diagnostic assessment of spleen stiffness. If the FibroScan is used, a result of 75 kPa certainly has to be regarded as indicating a high likelihood of advanced varices.
The problem of upper-limit results using the FibroScan seems not to have occurred in the 3 studies describing the use of FibroScan SSM in adults (16–18), wherein SSM values for patients with varices appear lower than in our sample. This may be a result of patient selection or sample size, or it may indicate a technical limitation that is specific to children.
Studies that investigated ARFI rather than FibroScan for spleen elastography in adults have shown mixed and partially contradictory results with regard to predicting the presence and complications of portal hypertension (26–29). One study already investigated the feasibility of ARFI spleen elastography in healthy children (30), but to the best of our knowledge, the use of ARFI elastography of the spleen in children to assess portal hypertension has not been described to date.
Studies in adults examining the diagnostic use of spleen stiffness assessment using FibroScan TE or ARFI have often compared and/or combined liver and spleen stiffness measurements, again with ambiguous results. Although Rifai et al (26) found liver stiffness assessment to be superior to spleen stiffness assessment, both Vermehren et al (27) and Takuma et al (29) using ARFI and Colecchia and Sharma using FibroScan (17,18), in contrast, describe superiority of spleen stiffness assessment. Stefanescu et al (16) suggest combining both liver and spleen TE for maximum diagnostic accuracy.
In children, a comparison between liver and spleen stiffness measurements to diagnose portal hypertension has not yet been published; however, some studies have demonstrated that LSM by FibroScan alone was predictive of the presence of varices in children (13–15). In contrast to some of the studies mentioned above, in our data LSM by FibroScan and serological assessment of hepatic fibrosis by Fibrotest both correlated with the degree of splenomegaly, but failed to distinguish between patients with or without varices and/or variceal bleeding. This lack of predictive effect of liver assessment in our cohort most likely results from the small sample size available for this particular assessment. Also owing to the sample size, we were not able to stratify children according to the etiology of their LD (with the exception of the exclusion of children with prehepatic block). Because cholestasis and hepatic inflammation affect both liver stiffness and Fibrotest results, the heterogeneity of a small pediatric sample may be another explanation for the observed difference between results from adults and our findings. Especially, Fibrotest has been predominantly investigated in adults, which makes comparison with pediatric data difficult.
From our sample, no definitive answer on the superiority of any one method (SSM, LSM, Fibrotest) can be given. The visible trend of higher LSM and Fibrotest values in children with a history of variceal bleed, however, suggests that a comparison of SSM, LSM, and Fibrotest is desirable and merits further investigation in a larger cohort.
Our pilot study delivers first data that show that spleen stiffness elastography using the FibroScan is feasible in children with chronic LD, and that its results reflect the degree and occurrence of complications of portal hypertension. A preliminary cutoff value is 60 kPa, above which there is a high risk of variceal bleeding. A prospective follow-up study incorporating larger patient numbers and including patients who undergo screening rather than indication endoscopies appears justified and desirable based on our pilot data. Given that the FibroScan limits its measurement range to 75 kPa, a comparison of FibroScan TE with another technique, for example, ARFI, is desirable. Spleen elastography could then become the much-needed tool to support further epidemiological and interventional research concerning the incidence and complications of portal hypertension in children.
1. Gugig R, Rosenthal P. Management of portal hypertension in children. World J Gastroenterol
2. Costaguta A, Alvarez F. Etiology and management of hemorrhagic complications of portal hypertension in children. Int J Hepatol
3. Ling SC, Walters T, McKiernan PJ, et al. Primary prophylaxis of variceal hemorrhage in children with portal hypertension: a framework for future research. J Pediatr Gastroenterol Nutr
4. Garcia-Tsao G, Sanyal AJ, Grace ND, et al. Practice Guidelines Committee of the American Association for the Study of Liver Diseases, Practice Parameters Committee of the American College of GastroenterologyPrevention and management of gastroesophageal varices and variceal hemorrhage in cirrhosis. Hepatology
5. Fagundes ED, Ferreira AR, Roquete ML, et al. Clinical and laboratory predictors of esophageal varices in children and adolescents with portal hypertension syndrome. J Pediatr Gastroenterol Nutr
6. Bolognesi M, Merkel C, Sacerdoti D, et al. Role of spleen enlargement in cirrhosis with portal hypertension. Dig Liver Dis
7. Gana JC, Turner D, Roberts EA, et al. Derivation of a clinical prediction rule for the noninvasive diagnosis of varices in children. J Pediatr Gastroenterol Nutr
8. Chang HK, Park YJ, Koh H, et al. Hepatic fibrosis scan for liver stiffness score measurement: a useful preendoscopic screening test for the detection of varices in postoperative patients with biliary atresia. J Pediatr Gastroenterol Nutr
9. Thabut D, Trabut JB, Massard J, et al. Non-invasive diagnosis of large oesophageal varices with FibroTest in patients with cirrhosis: a preliminary retrospective study. Liver Int
10. Thabut D, Imbert-Bismut F, Cazals-Hatem D, et al. Relationship between the fibrotest and portal hypertension in patients with liver disease. Aliment Pharmacol Ther
11. Shneider BL, Bosch J, de Franchis R, et al. Portal hypertension in children: expert pediatric opinion on the report of the Baveno v Consensus Workshop on Methodology of Diagnosis and Therapy in Portal Hypertension. Pediatr Transplant
12. Grimaldi C, de Ville de Goyet J, Nobili V. Portal hypertension in children. Clin Res Hepatol Gastroenterol
13. Malbrunot-Wagner AC, Bridoux L, Nousbaum JB, et al. Transient elastography
and portal hypertension in pediatric patients with cystic fibrosis transient elastography
and cystic fibrosis. J Cyst Fibros
14. Colecchia A, Di Biase AR, Scaioli E, et al. Non-invasive methods can predict oesophageal varices in patients with biliary atresia after a Kasai procedure. Dig Liver Dis
15. Chongsrisawat V, Vejapipat P, Siripon N, et al. Transient elastography
for predicting esophageal/gastric varices in children with biliary atresia. BMC Gastroenterol
2011; 11: 41-230X-11-41.
16. 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
17. 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
18. Sharma P, Kirnake V, Tyagi P, et al. Spleen stiffness in patients with cirrhosis in predicting esophageal varices. Am J Gastroenterol
19. Sirli R, Sporea I, Bota S, et al. Liver elastography for the diagnosis of portal hypertension in patients with liver cirrhosis. Med Ultrason
20. de Ledinghen V, Le Bail B, Rebouissoux L, et al. Liver stiffness measurement in children using FibroScan: feasibility study and comparison with fibrotest, aspartate transaminase to platelets ratio index, and liver biopsy. J Pediatr Gastroenterol Nutr
21. Engelmann G, Gebhardt C, Wenning D, et al. Feasibility study and control values of transient elastography
in healthy children. Eur J Pediatr
22. Goldschmidt I, Streckenbach C, Dingemann C, et al. Application and limitations of transient liver elastography in children. J Pediatr Gastroenterol Nutr
23. Poynard T, Ngo Y, Marcellin P, et al. Impact of adefovir dipivoxil on liver fibrosis and activity assessed with biochemical markers (FibroTest-ActiTest) in patients infected by hepatitis B virus. J Viral Hepat
24. Behrens R. Pfrimmer Nutricia, Endoskopie in der pädiatrischen Gastroenterologie: Lehrbuch und Atlas. 1. Auflage. Erlangen, Germany:2010.
25. Haber HP. Hofmann V, Deeg KH, Hoyer PF. Referenzwerte. Ultraschalldiagnostik in Pädiatrie und Kinderchirurgie: Lehrbuch und Atlas
3rd ed.Stuttgart, Germany:Georg Thieme Verlag KG; 2005. 699.
Mederacke I, Wursthorn K, Kirschner J, et al. Food intake increases liver stiffness in patients with chronic or resolved hepatitis C virus infection. Liver Int
26. 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 Med
2011; 32 (suppl 2):E24–30.
27. Vermehren J, Polta A, Zimmermann O, et al. Comparison of acoustic radiation force impulse imaging with transient elastography
for the detection of complications in patients with cirrhosis. Liver Int
28. Bota S, Sporea I, Sirli R, et al. Spleen assessment by acoustic radiation force impulse elastography (ARFI) for prediction of liver cirrhosis and portal hypertension. Med Ultrason
29. 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
30. Lee MJ, Kim MJ, Han KH, et al. Age-related changes in liver, kidney, and spleen stiffness in healthy children measured with acoustic radiation force impulse imaging. Eur J Radiol