Role of Splenic Hepatic elastography ratio in differentiating Non Cirrhotic Portal Fibrosis and Chronic Liver Disease in children

doi:10.21203/rs.3.rs-4344311/v1

Download PDF

Research Article

Role of Splenic Hepatic elastography ratio in differentiating Non Cirrhotic Portal Fibrosis and Chronic Liver Disease in children

https://doi.org/10.21203/rs.3.rs-4344311/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 29 Jul, 2024

Read the published version in Hepatology International →

You are reading this latest preprint version

Background: Differentiation of Non Cirrhotic Portal Fibrosis(NCPF) from chronic liver disease(CLD) in children and adolescents with portal hypertension(PHT) is challenging especially in cases where liver stiffness measurement(LSM) and Hepatic venous pressure gradient are higher. This objective of the current study was to evaluate the diagnostic accuracy of the Splenic stiffness measurement (SSM)/LSM ratio in the diagnosis of NCPF.

Methods: From January 2019 to December 2023, consecutive children and adolescents of 6 months to 18 years of age with PHT(CLD and NCPF) were prospectively enrolled. Transient elastography(TE) for SSM and LSM, Upper gastrointestinal endoscopy(UGIE), liver biopsy/trans jugular liver biopsy, abdominal imaging and laboratory evaluation were done. The relationship of TE parameters for diagnosis of NCPF and CLD was evaluated. Receiver operating characteristic (ROC) statistics were applied using R Studio-4.2.2 statistical software

Results: One hundred and forty seven with CLD and 27 patients with NCPF were evaluated. Median age was 10.0(IQR 2.4-14.0) years, 68.4% were males. The AUROC of SSM/LSM ratio was better (0.992,95%CI 0.982-1.0001) than LSM (0.945,95%CI0.913-0.977) and SSM (0.626,95%CI0.258-0.489) for the diagnosis of NCPF. SSM/LSM ratio cut-off of 3.67 predicted NCPF with an excellent sensitivity(100%), specificity(95.9%) and diagnostic accuracy(95.91%). The AUROC of SSM/LSM ratio was excellent and outperformed other TE parameters in the subgroups i.e. LSM between 10-20 kPa (0.982,95%CI0.947-1.000), without clinically significant Varices(CSV) (1.000,95%CI1.000-1.000) and with CSV(0.993,95%CI0.983-1.000). Diagnostic performance of SSM/LSM Ratio was better than LSM for discriminating NCPF from CLD using McNemar test(p=0.01)

Conclusion: The SSM/LSM ratio is an excellent tool in differentiating NCPF from CLD.

Portal hypertension

Spleen Stiffness Measurement

Liver Stiffness Measurement

Cirrhosis

Non Cirrhotic Portal Hypertension

Transient elastography

Non Invasive

clinically significant Varices

Children

Adolescents

Non cirrhotic Portal Fibrosis (NCPF) unlike Chronic liver disease (CLD) is a syndrome with relatively good prognosis, hence its differentiation from latter is of utmost importance [1, 3–5]. NCPF is usually clinically characterised by massive splenomegaly with or without an episode of usually well tolerated variceal bleeding in children or young adults having near normal hepatic functions [6]. This diagnosis is made by the unique combination of findings of portal hypertension (PHT) in the absence of liver cirrhosis or any evidence of portal vein abnormality in abdominal imaging and having characteristic histological lesions like maintained lobular architecture, obliterative portal venopathy, abnormal approximation of portal-portal and portal-central vein etc in liver biopsy/ trans jugular liver biopsy (TJLB) [1]. Sometimes, due to technical challenges involved especially with TJLB in children, NCPF gets wrongly labelled as CLD. Hence a good non-invasive parameter is needed for screening and differentiating patients with NCPF from those with CLD [2–5]. Sometimes, Liver stiffness measurement (LSM) and hepatic venous pressure gradient (HVPG) comes higher, making diagnosis of NCPF difficult [1]. One study in adults has proposed that the cut-off values of < 10 kPa and > 20 kPa as significantly suggestive of porto-sinusoidal vascular disease (PSVD) and cirrhosis, respectively, however its sensitivity is poor (65%) [7]. A recent study in adult has shown that Splenic stiffness measurement (SSM)/LSM ratio has an excellent AUROC 0.940 (p < 0.001) in differentiating PSVD from cirrhosis in patients with PHT [8]. Studies have now shown that SSM by transient elastography is not only technically feasible in children and adolescents of 6 months to 18 years of age but also the best predicter of clinically significant varices (CSV) in patients with CLD as well as NCPF [9–14]. This study includes few patients, which have been the part of the earlier study on prediction of CSV by SSM from the same centre [11]. Diagnostic value of SSM/LSM for differentiating NCPF has never been studied in children yet, to the best of our knowledge. The present study was aimed to assess the diagnostic accuracy of SSM, LSM, and SSM/LSM ratio using transient elastography (TE), for discriminating NCPF from CLD in children and adolescents with PHT.

Study design

We performed a prospective study that included children and adolescents of 6 months to 18 years of age of PHT, who have been diagnosed as having CLD or NCPF. The study was conducted in the Pediatric Hepatology department of a tertiary centre in north India. Patients were enrolled between January 2019 and December 2023. Exclusion criteria were endoscopic treatment for varices prior to enrollment, treatment with beta-blockers, obesity, presence of grade 2–3 ascites, hepatocellular carcinoma or any other malignancy, extra hepatic portal vein obstruction (EHPVO), previous splenectomy, trans jugular intrahepatic Porto-systemic shunt (TIPS), status post liver transplant. CLD was defined as presence of either of the following histological evidence of advanced fibrosis > F2 on Metavir staging or imaging suggestive i.e. heterogenous echotexture, irregular nodular liver, caudate hypertrophy. Diagnosis of NCPF was based on presence of PHT, normal liver function tests, normal liver imaging and liver biopsy suggestive of NCPF i.e maintained lobular architecture, atretic portal tracts, approximation of portal-portal and portal-central areas, and aberrant peripheral portal channels [1, 2].

All patients underwent clinical exam, laboratory tests (hemogram, hepatic panel, renal biochemistry, and albumin), abdominal Imaging, LSM and SSM by TE, liver biopsy/TJLB, endoscopic oesophageal varices screening by Upper Gastrointestinal Endoscopy (UGIE). Informed consent forms were signed by the participants. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki (6th revision, 2008) as reflected in a prior approval by our institution’s human research committee. Approval for the study was obtained from Institutional Review Board (IEC/ IRB No: IEC 2019/67/NA01).

After overnight fasting (at least 6 hours), TE of liver and spleen was performed at baseline by using the Fibroscanner 502 FibroScan apparatus (Echosens, Paris, France), which consists of a 5-MHz ultrasound transducer probe mounted on the axis of a vibrator. The standard 7 mm “M” probe was used for children ≥ 6 years and 5 mm child “S1” probe used for those below 6 years of age. SSM was assessed from the 10th intercostal space in the left anterior axillary line or directly over spleen, where palpable. LSM was done from the 10th intercostal space in the right mid axillary line with arms in maximum abduction. Two separate sets of 10 measures each were performed for SSM and LSM and the average value was recorded. Quality criteria applied for SSM measurement were similar to LSM (≥ 80% success rate; interquartile range < 20% of median). We calculated the SSM/LSM index by dividing the median SSM value by the median LSM value in each patient. Esophageal varices were classified into grades 1 to 3 as per Westaby classification and modified as per Duche et al [15–17]. Red colour signs (RCS) were evaluated as previously defined [16]. Gastric varices were defined as per Sarin’s classification [18]. CSV were labelled as presence of grade 3 varices, grade 2 varices with RCS, and/or presence of gastric varix [10]. Portal hypertensive gastropathy was classified as mild or severe [19].

Sample size calculation

Analyses was done using R Studio-4.2.2 statistical software. Based on the study by Sutton et al, the sample size was calculated assuming sensitivity of splenic stiffness is 80%, specificity of 90%, alpha error of 5%, permissible error as 10% and prevalence of CSV in patients as 50% [10]. The sample needed was 120 children with portal hypertension for the present study. The final sample size was estimated at 126 patients in each group to account for a 5% dropout rate. As it was a pilot study, all children of PHT fulfilling the inclusion and exclusion criteria were prospectively enrolled in the CLD and NCPF groups, respectively.

The normality of variables was assessed using histograms and Shapiro-Wilk test. Univariable analysis were performed using independent samples T test or independent samples Mann-Whitney U test for continuous variables and Chi-square test for categorical variables. Cut off values for SSM, LSM, and SSM/LSM ratio was calculated as per receiver operating characteristic (ROC) curve and the area under the curve (AUROC) using the presence of NCPF as the binary system classifier. The ideal cut-off value for each ROC curve analysis was calculated using the Youden index. An additional ROC curve analysis of the SSM/LSM ratio was done in three subgroups separately (subgroup having LSM between 10 and 20 kPa, subgroup without CSV and subgroup with CSV) using the presence of NCPF as the binary system classifier. Sensitivity, specificity, positive and negative predictive values, positive and negative likelihood ratios, and diagnostic accuracy were calculated based on the ROC curve. Diagnostic accuracy of LSM and SSM/LSM ratio was then compared using the McNemar test.

A total of 254 children and adolescents with PHT were considered for inclusion, however only 174 could be finally included prospectively (Fig. 1). The median age (IQR) of the whole cohort was [10.0 years, IQR 2.4–14) with majority being males (n = 119, 68.4%). The median SSM and LSM by TE and SMM/LSM ratio were 29.5 (IQR 22-56.2) kPa and 26.5(IQR 11.7–42.0) kPa and 1.15 (IQR 0.8–2.23), respectively, in the whole cohort. The prevalence of CSV was 57.4%(n = 100) in the whole cohort, of which 18.9% (n = 33) were bleeders.

Children and adolescents with NCPF (n = 27)

Of the 27 children and adolescents with NCPF, most patients were male (76.2%). Specific histologic finds of the NCPF cohort included maintained lobular architecture (n = 27, 100%), atretic portal tracts (n = 27, 100%), abnormal approximation of portal-portal and portal-central vein (n = 23, 85.1%), partial and complete absence of central vein profiles (n = 15, 55.5%), sinusoidal dilatation (n = 13, 48.1%), nodular regenerative hyperplasia (n = 9, 33.3%), phlebosclerosis (n = 3,11.1%) (Table 1).

Table 1

Specific histologic findings associated with the diagnosis of NCPF (n = 27)
Histologic findings	n (%)
Maintained lobular architecture	27(100%)
Atretic portal tracts	27(100%)
Abnormal approximation of portal-portal and portal-central vein	23(85.1%)
Partial and complete absence of central vein profiles	15(55.5%)
Sinusoidal dilatation	13(48.1%)
Nodular regenerative hyperplasia	9(33.3%)
Phlebosclerosis	3(11.1%)

Children and adolescents with cirrhosis

Of the 147 patients’ of CLD included (Fig. 1), again most of the patients were male 97(65.9%). Wilson disease was the most common (n = 41, 27.8%) etiology amongst all, followed by Biliary atresia (n = 36, 24.5%) and Auto immune liver disease (n = 32, 21.7%).

The baseline clinical and laboratory profiles of the children and adolescents with CLD and NCPF group are compared in Table 2. The median age of NCPF group (13.5 years, IQR 9.0–15.0) was significantly higher (p = 0.001) than the CLD group (9.0 years, IQR 1.7–14). Bleeders were significantly more (p < 0.001) in the NCPF group (n = 13, 48.1%) compared to CLD group (n = 20, 13.6%). Nutrition in terms of weight for height (WFH) z score (p = 0.022) and height for age (HFA) z score (p = 0.029) was significantly maintained in NCPF group relative to CLD group. Spleen size on ultrasound was significantly higher (p < 001) in the NCPF group (15.9 cms, IQR 12.8–17.6) compared to the CLD group (12.3 cms, IQR 9.6–15.2). Similarly, splenic z scores was also significantly higher (p < 001) in the NCPF group (7.38, IQR 4.55–8.58) compared to the CLD group (4.14, IQR 2.96–6.9). Amongst the laboratory investigations, patients with NCPF had lower levels of total leukocyte count (p < 0.001), platelet count (p = 0.001), while patients with CLD had higher median levels of bilirubin (p = 0.005), and prothrombin time (p = 0.021) compared to the CLD group at index presentation. CLD group had a significantly higher Pediatric End Stage Liver Disease (PELD) Score (p < 0.03) and Child Turcotte Pugh (CTP) score (p = 0.001) compared to the NCPF group (Table 2).

Table 2

Baseline clinical and laboratory parameters of the children with portal hypertension included in the study.
Parameter	Total n = 174 Median(IQR)	CLD n = 147 Median(IQR)	NCPH n = 27 Median(IQR)	Odd’s ratio or Mean difference (95% CI)	p value
Age(yrs)	10.0(2.4–14)	9.0(1.7–14)	13.5(9.0–15.0)	3.02 (1.04–5.2)	0.001
Male n(%)	119(68.4)	97(65.9)	22(81.5)	0.018 (0.014–0.19)	0.122
WFH (Z-score)	-0.99(-2.4- 0.15)	-0.70(-2.1- 0.19)	-1.9(-2.8- -0.46)	0.61 (0.23–1.23)	0.022
HFA (Z-score)	-0.94(-2.0- 0.30)	-0.75(-1.99- 0.50)	-1.3(-2.8- -0.86)	1.12 (1.55–3.78)	0.029
Bleeder n(%)	33(18.9)	20(13.6)	13(48.1)	0.24 (0.09–0.49)	< 0.001
Age index bleed (years)	11.0(3.4–13.8)	6.5(1.5–14.2)	11.7(8.0-13.8)	2.43 (1.94–6.9)	0.149
Spleen size USG (cm)	12.5(10.2–15.9)	12.3(9.6–15.2)	15.9(12.8–17.6)	3.2 (1.48–4.64)	< 0.001
Splenic Z-score (SZS)	4.47(2.4–7.1)	4.14(2.96–6.9)	7.38(4.55–8.58)	2.26 (0.95–3.23)	< 0.001
Laboratory Parameters
Bilirubin (mg/dL)	1.8(1.07-5.0)	2.1(1.1–6.8)	1.3(1.0-2.4)	4.38(1.3–7.4)	0.005
Albumin (g/dL)	3.2(2.4–3.7)	2.9(2.4–3.7)	3.7(2.9–3.8)	0.79(-0.64- 0.05)	0.097
INR	1.34(1.1–1.8)	1.37(1.1–1.8)	1.30(1.1–1.4)	0.28(0.11–0.45)	0.021
Hb (g/dL)	10.4(9.1–11.8)	10.6(9.1–11.8)	9.6(8.9–11.4)	0.72(-0.108- 1.55)	0.088
TLC (X10^3)/ cu.mm	6.5(4.2–9.04)	7.0(4.5–9.5)	4.09(2.2–4.5)	3.1(1.77–6.50)	< 0.001
Platelet (X10^3)/ cu.mm	109.5(65-169.2)	114(80–148)	102(52–148)	41.2(28.0.-94.3)	0.001
PELD	4.0(-3.7- 16.0)	5.0(-3.0- 17.0)	1.46(-5.0- 6.0)	5.9(0.53–11.4)	0.03
CTP	7.0(6.0–10.0)	8.0(6.0–10.0)	6.0(5.0–7.0)	1.6(0.65–2.6)	0.001

The endoscopic findings, transient elastography measurements liver (LSM), spleen (SSM), SSM/LSM index and portal hypertensive indices of the children and adolescents with CLD and NCPF group are compared in Table 3. Comparatively, SSM was significantly higher (p < 0.001) amongst the children with NCPH (32.0 kPa, IQR 26.0–75.0) than CLD (28.0 kPa, IQR 21.0–42.0). The LSM was significantly (p < 0.001) lower amongst the children with NCPH (8.0 kPa, IQR 5.0-10.2) than CLD (29.1 kPa, IQR 18.0–50.0). The prevalence of CSV was significantly (p = 0.01) more amongst the children with NCPH (n = 20,74.1%) than CLD (n = 80, 54.4%).

Table 3

Endoscopic features, transient elastography measurements (liver and spleen) and portal hypertensive indices of children with cirrhotic and non-cirrhotic portal hypertension.
Parameter	Total n = 174	CLD n = 147	NCPH n = 27	Odd’s ratio or Mean difference (95% CI)	p value
Endoscopic features
Esophageal varices n(%)	141(81.03%)	118(80.3%)	23(85.2%)	1.078 (0.480–2.315)	0.572
Grade 3 varices n(%)	33(18.9%)	24(16.3%)	9(33.3%)	1.842 (1.003–3.619)	0.042
Red colour signs	50(28.7%)	37(25.2%)	13(48.1%)	2.018 (1.081–3.788)	0.022
PHG	115(66.1%)	91(61.9%)	24(88.9%)	3.736 (1.185–11.880)	0.007
Gastric varices	17(9.7%)	9(6.1%)	8(29.6%)	2.881 (1.366–5.622)	0.004
CSV	100(57.4%)	80(54.4%)	20(74.1%)	0.192(0.008–0.41)	0.01
Transient elastography measurements (Liver and Spleen)
SSM (kPa)	29.5(22-56.2)	28.0(21.0–42.0)	32.0(26.0–75.0)	-11.3(-19.7- -2.82)	0.009
LSM (kPa)	26.5(11.7–42.0)	29.1(18.0–50.0)	8.0(5.0-10.2)	27.86(19.17–36.55)	< 0.001
SSM / LSM Index	1.15(0.8–2.23)	1.05(0.79–1.55)	8.3(6.7–11.6)	-8.24(-9.10- -7.39)	< 0.001
Portal hypertensive indices
EASS	17.9(14.9-21.97)	17.46(14.7-21.57)	22.2(18.07–24.08)	-3.46(-5.2- − 1.63)	< 0.001
KVaPS	54.53(34.5–79.1)	55.51(33.32–79.1)	50.8(34.9–79.3)	-1.91(-13.7- 9.91)	0.750
VPS	33.4(16.5–58.1)	35.8(18.6–62.4)	17.3(13.2–42.7)	20.1(4.4-35.78)	0.012
CPR	88.4(69.9-110.5)	88.6(69.3-110.5)	84.8(76.1-104.7)	1.34(-9.88–12.58)	0.81
APRI	2.2(1.1–4.6)	2.5(1.13–5.34)	2.09(0.87–3.31)	13.06(-35.17- 61.3)	0.59

Diagnostic performance of TE parameters in differentiating NCPF from CLD

The AUROC of LSM, SSM and SSM/LSM ratio index for the diagnosis of NCPF were 0.945 (95% CI 0.913–0.977), 0.374 (95% CI 0.258–0.489) and 0.992 (95% CI 0.982-1.000) respectively in the whole cohort (Fig. 2a). Even on restricting the analysis to the patient cohort which is most controversial i.e. having LSM values from 10 to 20 kPa (n = 36; 8 in NCPF group), SSM/LSM remained an excellent predictor for the diagnosis of NCPF with an AUROC of 0.982 (95% CI 0.949-1.000) unlike LSM whose AUROC of decreased to 0.848 (95% CI 0.695-1.000) in this subgroup (Fig. 2b). AUROC of SSM was good and better than LSM in this subgroup at 0.853 (95% CI 0.697-1.000) (Fig. 2b). To match the severity of PHT between NCPF and CLD groups, diagnostic accuracy of SSM, LSM and SSM/LSM ratio were also assessed in patient cohort sub groups without CSV and with CSV separately. In the cohort without CSV i.e. less severe PHT, AUROC of SSM/LSM was excellent (1.000, 95% CI 1.000–1.000) and much better than LSM (0.913, 95% CI 0.843-992) and SSM (0.864, 95% CI 0.752–0.976) (Fig. 2c). In cohort with CSV i.e. more severe PHT, AUROC of SSM/LSM ratio (0.993, 95% CI 0.983-1.000) was excellent and outperformed LSM (0.977, 95% CI 0.953-1.000 (Fig. 2d). The AUROC of SSM (0.472, 95%CI 0.286-658) was poor in patients with CSV (Fig. 2d).

LSM with a best cut-off value of 11.4 kPa, predicted the diagnosis of NCPF with a sensitivity of 88.5%, specificity of 92.6%, positive predictive value of 98.5%, negative predictive value of 59.5%, positive likelihood ratio of 11.93 and diagnostic accuracy of only 81.1% (Table 4). While SSM with a best cut-off value of 28.0 kPa predicted the diagnosis of NCPF with a poor sensitivity of 54.5%, specificity of 51.8%, positive predictive value of 86.02%, negative predictive value of 17.30%, positive likelihood ratio of 1.13 and diagnostic accuracy of 6.27% (Table 4). However, SSM/LSM ratio with best cut-off value of 3.66 predicted the diagnosis of NCPF with an excellent sensitivity of 100%, specificity of 95.9%, positive predictive value of 81.8%, negative predictive value of 100%, positive likelihood ratio of 24.5 and diagnostic accuracy of 95.91% (Table 4). The best cut-off value of 3.66 of SSM/LSM ratio as obtained for the whole cohort was applied and evaluated for diagnostic accuracy in the subgroups of patients i.e. patient cohort having LSM values from 10 to 20 kPa, patient cohort sub groups without CSV and with CSV, separately (Table 4). Even in the patient cohort having LSM values from 10 to 20 kPa, which is considered as the most difficult to diagnose group, cut-off value of 3.66 of SSM/LSM ratio predicted the diagnosis of NCPF with a sensitivity of 100%, specificity of 66.7%, positive predictive value of 85.7%, negative predictive value of 100%, positive likelihood ratio of 3.0 and diagnostic accuracy of 88.88%. In the patient cohort sub group without CSV i.e. less severe PHT, it predicted the diagnosis of NCPF with a sensitivity of 100%, specificity of 70.0%, positive predictive value of 95.1%, negative predictive value of 100%, positive likelihood ratio of 3.3 and diagnostic accuracy of 95.58%. While in the patient cohort sub group with CSV i.e. more severe PHT, it predicted the diagnosis of NCPF with a sensitivity of 100%, specificity of 86.9%, positive predictive value of 95.9%, negative predictive value of 100%, positive likelihood ratio of 7.66 and diagnostic accuracy of 95.58% (Table 4).

Table 4

Diagnostic performance of TE measurements for discriminating NCPF from CLD (n = 174).
Elastographic Measurement	AUROC (CI)	Cut off	p value	Se %	Sp %	PPV%	NPV%	+LR	-LR	Diagnostic Accuracy
LSM (n = 174)	0.945 (0.913–0.977)	11.4 kPa	< 0.001	88.5	92.6	98.5	59.5	11.93	0.12	81.1%
SSM (n = 174)	0.374 (0.258–0.489)	28.0 kPa	0.015	54.5	51.8	86.02	17.30	1.13	0.87	6.27%
SSM/LSM (n = 174)	0.992 (0.982- 1.000)	3.67	< 0.001	100	95.9	81.8	100	24.5	0	95.91%
SSM / LSM (LSM 10-20kPA) (n = 36)	0.982 (0.949–1.016)	3.67	< 0.001	100	66.7	85.7	100	3.0	0	88.89%
SSM / LSM (CSV Absent) (n = 74)	1.0(1.0–1.0)	3.67	< 0.001	100	70	95.1	100	3.33	0	95.58%
SSM / LSM (CSV Present) (n = 100)	0.993 (0.983- 1.00)	3.67	< 0.001	100	86.9	95.9	100	7.66	0	96.77%

On comparing the diagnostic accuracy of LSM (90.8%) with SSM/LSM ratio (95.97%) for differentiating NCPF from CLD using McNemar test, this difference was found to be statistically significant(p = 0.01) in favour of SSM/LSM ratio(Table 5).

Table 5

Comparison of Diagnostic performance of SSM/LSM Ratio and LSM for discriminating NCPF from CLD using McNemar test (n = 174)
	LSM (Cut off 11.40)
	Correct Diagnosis	Incorrect Diagnosis	Total	P value
SSM/LSM Ratio (Cut Off 3.66)
Correct Diagnosis	128(73.6%)	27(15.5%)	155(89.1%)	0.01
Incorrect Diagnosis	4(2.3%)	15(8.6%)	19(10.9%)
Total	132(75.8%)	42(24.1%)	174

To the best of our knowledge, current study is the first study in children and the largest sample size ever to evaluate the role of the combination of SSM and LSM in the differentiation of NCPF and CLD in patients with PH. We demonstrated that, in patients with PHT, SSM/LSM ratio with a cut off of 3.6 predicted NCPF with greater accuracy when compared with LSM alone. Recently a study in adult patients with PHT, showed that SSM/LSM ratio of > 2.0 accurately predicted PSVD, compared to LSM alone [8].

The LSM has been used to discriminate non cirrhotic portal hypertension from CLD in adult as well as paediatric patients with PHT [7, 20, 21, 22]. A recent study in adults has shown that a cut off of LSM < 10 kPa has good diagnostic performance in predicting PSVD [7]. However, the sensitivity of this cut off was poor (65%) in differentiation from CLD, resulting in 25% of patients falling in the “grey zone” of 10–20 kPa. In children, Sood et al have found that the median value of LSM was 10.6(95% CI 7.6–13.5) kPa in children with NCPF, which is again higher than the proposed cut off for differentiation from CLD in a study done in adults [1]. Recent study done in adults, has shown that LSM alone at a best cut-off value of 12.5 kPa had a lower AUROC and diagnostic accuracy of 0.843 and 77.9% respectively, in differentiating CLD from PSVD, compared to SSM/LSM ratio, which at a best cut off of > 2.0 predicted PSVD with a better AUROC and diagnostic accuracy of 0.897 and 95.97% respectively [8]. Current study had shown similar results in children, wherein again SSM/LSM ratio at a best cut of 3.6 predicted NCPF with an excellent AUROC and diagnostic accuracy of 0.992 and 95.97% respectively compared to LSM alone, which at a best cut-off value of 11.4 kPa was found to have a lower AUROC and diagnostic accuracy of 0.945 and 81.1%, in differentiating NCPF from CLD. We propose that once EHPVO has been ruled out, higher SSM/LSM ratio (cut off 3.67) and lower LSM values strongly suggest NCPF in children and adolescents with PHT. However, liver biopsy should be done for confirmation.

In the current study, similar to adults, values of SSM were found to be significantly higher in patients with NCPF compared to patients with CLD [8]. Despite statistical significance,

difference was smaller in absolute terms compared with LSM values and thus explains the poorer diagnostic performance of SSM as has also been seen previously in an adult study [8].

Analogous to the findings in the adults with PHT, the creation of the SSM/LSM ratio in children and adolescent with PHT was also inspired by the finding of lower values of LSM and higher values of SSM in our study population with NCPF compared those having CLD. The combination of SSM and LSM in a single parameter (SSM/LSM ratio) allows to achieve an excellent diagnostic criteria, which not only is superior to LSM or SSM alone, but also preforms good in patients falling in the “grey zone i.e. those having LSM 10–20 kPa” (AUROC of 0.982, diagnostic accuracy of 88.89%), similar to adults (AUROC of 0.859, diagnostic accuracy of 76.9% ) [8]. Thus, current study suggests that SSM/LSM ratio is especially useful in this grey zone, and outperforms all other non-invasive diagnostic parameters.

NCPF group had severe PHT compared CLD group in current study as prevalence of CSV and history of previous variceal bleeding was significantly higher in the NCPF group. Earlier studies have correlated the severity of PHT with both SSM and LSM in CLD [10–11, 23–26]. Thus, the two groups were possibly not matched for the severity of PHT. Hence, to overcome this potential limitation cum speculation, diagnostic performance of the SSM/LSM index was evaluated in patients without CSV and with CSV separately too. In the current study, in the sub group without CSV subgroup, SSM/LSM retained its position as the best non-invasive predictor of NCPF with an excellent AUROC of 1.000, with a sensitivity of 100%, specificity of 70.1%, and diagnostic accuracy of 95.58% in differentiating NCPF from CLD. Similar findings have been reported in adults patient’s with PHT without CSV, wherein also SSM/LSM ratio outperformed other non-invasive TE parameters in differentiating PSVD from CLD with a AUROC of 0.99, with a sensitivity of 84.7%, specificity of 78.9%, and diagnostic accuracy of 81.2% [8]. Additionally, current study also evaluated the performance of SSM/LSM ratio in differentiating NCPF from CLD was in the subgroup with CSV i.e. more severe PHT for the first time ever in literature. In this subgroup too, SSM/LSM ratio outperformed other non-invasive TE parameters in differentiating PSVD from CLD with a AUROC of 0.993, with a sensitivity of 100%, specificity of 86.9%, and diagnostic accuracy of 96.77%.

However, the sample size is still small and can present a limitation. The direct measurement of the HVPG and portal pressures and their correlation with the SSM/LSM ratio would have added further to the value of the results.

In conclusion, in a patient with signs of PHT, LSM, SSM, and SSM/LSM ratio are useful in differentiating PSVD from CLD. Being non-invasive, SSM/LSM ratio with a superior diagnostic performance becomes an excellent parameter in the initial assessment of these patients, compared with LSM and SSM alone especially where LSM falls in the ‘grey zone’ i.e. between 10 and 20 kPa, and supportive clinical and laboratory evaluation is not revealing.

PHT: Portal Hypertension ; NCPF: Non Cirrhotic Portal Fibrosis ; CLD: Chronic Liver Disease ; EHPVO: Extra Hepatic Portal Vein Obstruction; IQR= Inter Quartile Range; HFA: Height for Age, WFH : Weight for height, TLC: Total leucocyte count, PELD: Pediatric End Stage Liver Disease, CTP : Child Turcotte Pugh score, Hb : Haemoglobin, T Bil : Total Bilirubin, USG: Ultrasound, cm : centimeter; UGIE: Upper Gastro Intestinal Endoscopy, SSM: splenic stiffness measurement, LSM: Liver stiffness measurement; CSV: clinically significant varices, EASS: Equivalent adult spleen size; KVaPs: Kings variceal prediction score, VPS: Variceal prediction score, CPR: Clinical prediction rule, APRI: Aspartate amino transferase(AST) to platelet ratio

AUROC: area under a receiver operating characteristic curve, Se: sensitivity, Sp: specificity, PPV: positive predictive value, NPV: negative predictive value, +LR: positive likelihood ratio

Acknowledgement

None

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Competing Interests

Piyush Upadhyay, Rajeev Khanna, Vikrant Sood, Bikrant Bihari Lal, Seema Alam have no have no relevant financial or non-financial interests to disclose.

Compliance with Ethical Requirements

The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki (6th revision, 2008) as reflected in a prior approval by our institution’s human research committee. Approval for the study was obtained from Institutional Review Board (IEC/ IRB No: IEC 2019/67/NA01). Informed consent was obtained from all patients for being included in the study.

Author contributions

Piyush Upadhyay, Seema Alam and Rajeev Khanna conceptualized the study, Piyush Upadhyay performed LSM and SSM, Piyush Upadhyay collected all data, Piyush Upadhyay and Seema Alam performed statistical analysis and derivation of results, all authors contributed to the conception and design of work, interpretation of data, critical revision of final manuscript for important intellectual content, approved the final version, and agreed to be accountable for all aspects of work. Seema Alam is the corresponding author of this manuscript and takes the responsibility of coordinating the work from its inception to publication and can be approached for access to raw data.

Sood V, Lal BB, Khanna R, Rawat D, Bihari C, Alam S. Noncirrhotic Portal Fibrosis in Pediatric Population. J Pediatr Gastroenterol Nutr. 2017;64(5):748–53.
Khanna R, Sarin SK. Non-cirrhotic portal hypertension - diagnosis and management. J Hepatol. 2014;60(2):421–41. 10.1016/j.jhep.2013.08.013. Epub 2013 Aug 23.
Hillaire S, Cazals-Hatem D, Bruno O, et al. Liver transplantation in adult cys- tic fibrosis: clinical, imaging, and pathological evidence of obliterative portal venopathy. Liver Transpl. 2017;23(10):1342–7. (In eng).
Meijer B, Simsek M, Blokzijl H, et al. Nodular regenerative hyperplasia rarely leads to liver transplantation: a 20-year cohort study in all Dutch liver trans- plant units. United Eur Gastroenterol J. 2017;5(5):658–67. (In eng).
Krasinskas AM, Goldsmith JD, Burke A, Furth EE. Abnormal intrahepatic por- tal vasculature in native and allograft liver biopsies: a comparative analysis. Am J Surg Pathol. 2005;29(10):1382–8. (In eng).
Sarin SK, Kapoor D. Non-cirrhotic portal fibrosis: current concepts and management. J Gastroenterol Hepatol. 2002;17(5):526–34.
Elkrief L, Lazareth M, Chevret S et al. Liver stiffness by transient elastogra- phy to detect porto-sinusoidal vascular liver disease with portal hypertension.
Ferreira-Silva J, Gaspar R, Liberal R, Cardoso H, Macedo G. Splenic-hepatic elastography index is useful in differentiating between porto-sinusoidal vascular disease and cirrhosis in patients with portal hypertension. Dig Liver Dis. 2023;55(1):75–80.
Goldschmidt I, Brauch C, Poynard T, Baumann U. Spleen stiffness measurement by transient elastography to diagnose portal hypertension in children. J Pediatr Gastroenterol Nutr. 2014;59(2):197–203.
Sutton H, Fitzpatrick E, Davenport M, Burford C, Alexander E, Dhawan A, Grammatikopoulos T. Transient Elastography Measurements of Spleen Stiffness as a Predictor of Clinically Significant Varices in Children. J Pediatr Gastroenterol Nutr. 2018;67(4):446–51.
Upadhyay PMD, Khanna DM, Sood RMD, Vikrant MD, Lal DM, Bikrant Bihari MD, Patidar DM. Yashwant MD*; Alam, Seema MD. Splenic stiffness is the best predictor of Clinically significant varices in children with Portal hypertension. J Pediatr Gastroenterol Nutr ():10.1097/MPG.0000000000003674, December 01, 2022.
Ferreira-Silva J, Gaspar R, Liberal R, Cardoso H, Macedo G. Transient splenic elastography predicts high-risk esophageal varices.
Stefanescu H, Grigorescu M, Lupsor M, Procopet B, Maniu A, Badea R. Spleen stiffness measurement using Fibroscan for the noninvasive assess- ment of esophageal varices in liver cirrhosis patients. J Gastroenterol Hepatol. 2011;26(1):164–70.
Colecchia A, Montrone L, Scaioli E, et al. Measurement of spleen stiffness to evaluate portal hypertension and the presence of esophageal varices in pa- tients with HCV-related cirrhosis. Gastroenterology. 2012;143(3):646–54.
Westaby D, Macdougall BR, Melia W, Theodossi A, Williams R. A prospective randomized study of two sclerotherapy techniques for esophageal varices. Hepatology. 1983 Sep-Oct;3(5):681–4.
Duché M, Ducot B, Ackermann O, Guérin F, Jacquemin E, Bernard O. Portal hypertension in children: High-risk varices, primary prophylaxis and consequences of bleeding. J Hepatol. 2017;66(2):320–7.
Idezuki Y. General rules for recording endoscopic findings of esophagogastric varices (1991). Japanese Society for Portal Hypertension. World J Surg. 1995 May-Jun;19(3):420–2. discussion 423.
Sarin SK, Lahoti D, Saxena S, et al. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients. Hepatology. 1992;16:1343–9.
Sarin SK, Shahi HM, Jain M, Jain AK, Issar SK, Murthy NS. The natural history of portal hypertensive gastropathy: influence of variceal eradication. Am J Gastroenterol. 2000;95(10):2888–93.
Chougule A, Rastogi A, Maiwall R, Bihari C, Sood V, Sarin SK. Spectrum of histopathological changes in patients with non-cirrhotic portal fibrosis. Hep- atol Int. 2018;12(2):158–66. 10.1007/s12072-018-9857-y. (In eng).
Sharma P, Agarwal R, Dhawan S, et al. Transient elastography (fibroscan) in patients with non-cirrhotic portal fibrosis. J Clin Exp Hepatol. 2017;7(3):230–4. (In eng).
Laharie D, Vergniol J, Bioulac-Sage P, et al. Usefulness of noninvasive tests in nodular regenerative hyperplasia of the liver. Eur J Gastroenterol Hepatol. 2010;22(4):487–93. (In eng).
Jansen C, Bogs C, Verlinden W, Thiele M, Möller P, Görtzen J, Lehmann J, Vanwolleghem T, Vonghia L, Praktiknjo M, Chang J, Krag A, Strassburg CP, Francque S, Trebicka J. Shear-wave elastography of the liver and spleen identifies clinically significant portal hypertension: A prospective multicentre study. Liver Int. 2017;37(3):396–405.
Vizzutti F, Arena U, Romanelli RG, Rega L, Foschi M, Colagrande S, Petrarca A, Moscarella S, Belli G, Zignego AL, Marra F, Laffi G, Pinzani M. Liver stiffness measurement predicts severe portal hypertension in patients with HCV-related cirrhosis. Hepatology. 2007;45(5):1290–7.
Hu X, Huang X, Hou J, Ding L, Su C, Meng F. Diagnostic accuracy of spleen stiffness to evaluate portal hypertension and esophageal varices in chronic liver disease: a systematic review and meta-analysis. Eur Radiol. 2021;31(4):2392–2404. doi: 10.1007/s00330-020-07223-8. Epub 2020 Sep 24. Erratum in: Eur Radiol. 2020;: PMID: 32974686; PMCID: PMC7979650.
Takuma Y, Nouso K, Morimoto Y, Tomokuni J, Sahara A, Takabatake H, Matsueda K, Yamamoto H. Portal Hypertension in Patients with Liver Cirrhosis: Diagnostic Accuracy of Spleen Stiffness. Radiology. 2016;279(2):609–19.

Graphicalabstract.png

Download PDF

Journal Publication

published 29 Jul, 2024

Read the published version in Hepatology International →

Editorial decision: Major Revisions Needed
31 May, 2024
Reviewers agreed at journal
06 May, 2024
Reviewers invited by journal
05 May, 2024
Editor assigned by journal
30 Apr, 2024
First submitted to journal
29 Apr, 2024

You are reading this latest preprint version

Role of Splenic Hepatic elastography ratio in differentiating Non Cirrhotic Portal Fibrosis and Chronic Liver Disease in children

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Material and methods

Results

Discussion

Abbreviations

Declarations

References

Supplementary Files

Status:

Journal Publication

Version 1