Preliminary Evaluation of Acoustic Radiation Force Impulse Shear Wave Imaging to Detect Hepatic Fibrosis in Morbidly Obese Patients Before Bariatric Surgery

Abstract

Background and Aims:

Acoustic Radiation Force Impulse (ARFI) shear wave imaging is a noninvasive method of assessment of the liver to detect fibrosis in patients with chronic hepatitis and nonalcoholic fatty liver disease (NAFLD). The aim of this retrospective study was to investigate whether noninvasive measurement of shear wave velocity (SWV) by ARFI shear wave imaging has a potential usefulness for detection of fibrosis secondary to NAFLD in patients with morbid obesity.

Methods:

Twenty-eight morbidly obese patients were included in this study. NAFLD and fibrosis were classified according to the nonalcoholic steatohepatitis (NASH) Clinical Research Network NAFLD activity score. SWV was quantified by ARFI imaging. Component steatosis, inflammation and ballooning scores, and fibrosis staging were correlated with SWV, and diagnostic accuracy of ARFI for fibrosis was assessed.

Results:

There was a decrease in mean SWV with increasing hepatic steatosis (P = .057). The SWV showed a significant negative correlation (r = −0.417, P = .011) with steatosis grade. The mean SWV was neither significantly different nor correlating with the obesity classes based on body mass index (BMI), steatosis grades, inflammation grades, ballooning grades, and fibrosis stages of NAFLD. Receiver operating characteristic analysis showed no significant area under curve for diagnosis of fibrosis using SWV. Valid SWV could be acquired in all subjects; however, only 21.42% fulfilled the interquartile range criterion.

Conclusion:

ARFI SWV values do not correlate with fibrosis on liver biopsy in morbidly obese patients and lack accuracy for diagnosis. Discordant values may be related to higher BMI and increasing hepatic steatosis.

Introduction

Nonalcoholic fatty liver disease (NAFLD) has a high incidence in morbidly obese patients ranging from 65% to 95% in different series.^1,2 The clinical course of most patients with NAFLD is benign, but patients who develop fibrosis have a high incidence of progression to cirrhosis and its complications; hence, this needs to be diagnosed and monitored for progression.^3,4 Currently, examination of the liver biopsy is considered the optimal method for diagnosis, staging, and monitoring changes in the liver histology over time. Liver biopsy can be safely combined with the bariatric surgery procedure for baseline diagnosis in morbidly obese patients undergoing surgical treatment.² Nevertheless, the liver biopsy is an invasive method when done under radiological guidance and has possible minor and major complications, with many patients reluctant to undergo this procedure.^4,5 Repeated liver biopsies are thus impractical during the course of disease in high-risk morbidly obese patients (defined by initial biopsy) as well as morbidly obese patients not subjected to bariatric surgery for diagnosis and monitoring. Noninvasive methods for the evaluation of liver fibrosis have become increasingly popular during the last few years replacing liver biopsy as a first option. Among these, the sonoelastography techniques are commonly used noninvasive methods and include transient elastography (TE) and Acoustic Radiation Force Impulse (ARFI) shear wave imaging.

The ARFI shear wave imaging acts by inducing transient tissue deformations of the liver tissue by an acoustic radiation force, which generates shear waves that can be used to estimate the tissue stiffness. ARFI has the advantage of combining real-time B-mode imaging with measurements of stiffness and can be used in patients with ascites. Numerous articles on the ARFI method have reported successful use in chronic hepatitis.^6–11 Studies on its use in NAFLD are still in progress, with few studies reporting usefulness.^12,13 Its use in morbidly obese patients is still not researched in detail, and data on applicability are lacking. A single study reported usefulness of ARFI in morbidly obese patients before bariatric surgery in identifying nonalcoholic steatohepatitis (NASH) and/or fibrosis. Two studies in abstract form failed to show usefulness for detection of NASH and/or fibrosis.^14–16 Studies have pointed out that severe obesity is a cause of unreliable and discordant values with ARFI as reported with TE.^9,12

In this study, we aimed to evaluate whether noninvasive measurement of shear wave velocity (SWV) by ARFI shear wave imaging has a potential usefulness for detection of fibrosis secondary to NAFLD in patients with morbid obesity.

Materials and Methods

Study population

Twenty-eight patients with morbid obesity scheduled for bariatric surgery were included in this study.

Measurement of SWV

SWV was measured using an ACUSON S2000 Ultrasound System (SIEMENS Medical Solutions, Inc.) equipped with a 4-MHz frequency transducer (with ARFI technology implanted) via the Virtual Touch (VT) tissue quantification application (the probe and application used for SWV measurement by ARFI technology were the same for obese and nonobese subjects). This application allows measurement of SWV within a region of interest (ROI) chosen by the examiner, which is set as a rectangle. The ultrasonic transducer creates short-duration acoustic push pulses making the tissues displace downward within the ROI chosen by the examiner. Shear stress starts to restore the displaced tissue to its previous position, which generates shear waves, which travel perpendicular to the direction of the acoustic push pulse. The machine continuously monitors tissue positions via the same transducer in ROI and automatically calculates SWV in meters/second as well as the depth at which the measurement was performed. When VT tissue quantification was enabled, on pressing the update key, one measurement of SWV in meters/second in the ROI and the depth at which the measurement was performed were displayed. When no valid measurement could be acquired, the monitor would display the “X-X-X-X” symbol, and measurement was repeated. In each patient, 10 valid ARFI measurements were performed, and the median and mean values were calculated. The results were expressed in meters/second. The higher the tissue stiffness, the higher was the SWV. ARFI imaging was performed before scheduled bariatric surgery.

Diagnosis of NAFLD in morbidly obese patients

All patients included in the study provided specific written informed consent to undergo liver biopsy as part of their bariatric operation. They underwent a liver biopsy intraoperatively at the time of bariatric surgery. Using a 16-gauge Tru-Cut biopsy needle (BARD Max-Core), a liver biopsy of the left hepatic lobe, without targeting a specific area in the lobe, was performed under direct vision. There were no intraoperative or postoperative complications related to the liver biopsy in any of the patients. Biopsies were interpreted by a single experienced pathologist. Using standard laboratory methods, each liver biopsy was stained with hematoxylin–eosin, Masson trichrome, and reticulin stain for histopathological interpretation. A diagnosis of NAFLD and NASH was made after a thorough histologic evaluation of each patient's liver biopsy as well as carefully looking for evidence of other liver disorders, such as autoimmune hepatitis, hepatitis C, hepatitis B, iron overload, and primary biliary cirrhosis. The degree of NAFLD and stage of fibrosis in each biopsy were scored and staged using the NIH-sponsored NASH Clinical Research Network NAFLD activity score. Normal biopsy received a score of 0 and was staged as 0 for fibrosis. The maximum NAFLD score was 8 and fibrosis stage was 4.

Statistical analyses

The difference in means of SWV among component scores and fibrosis staging was tested using analysis of variance (ANOVA), and the relationship between SWV and component scores and fibrosis staging was assessed by calculating the Spearman's correlation coefficient. To judge the accuracy of ARFI for fibrosis, sensitivity, specificity, negative predictive value, positive predictive value, and area under a receiver operating characteristic curve (AUROC) were calculated for detection of fibrosis and significant fibrosis. All analyses were performed using SPSS software version 20. A two-sided P value less than .05 was considered statistically significant.

Results

Study population

Twenty-eight morbidly obese patients scheduled for bariatric surgery were included in the study. Mean age of the study group was 48.29 years (median = 46, range = 21–67). Mean body mass index (BMI) was 49.58 kg/m². Mean values of biochemical tests performed were 21.91 for alanine transferase, 23.41 for aspartate transferase, 81.88 for alkaline phosphatase, 109.71 for triglycerides, and 43 for high density lipoprotein (HDL). Three were morbid obesity Class I (BMI 30–35), 2 morbid obesity Class II (BMI 35–40), 13 morbid obesity Class III (BMI 40–50), 6 superobese (BMI 50–60), and 4 super superobese (BMI >60).

Liver histology according to the NASH Clinical Research Network NAFLD scoring system is shown in Table 1.

Table 1.

Liver Histology of the 28 Patients According to NASH CRN NAFLD Scoring System

	Number of patients
Steatosis score
<5% score 0	10
5%–33% score 1	8
33%–66% score 2	7
>66% score 3	3
Inflammation score
Nil score 0	12
<2 foci score 1	16
>2 foci score 2	0
Ballooning score
Nil score 0	19
Few score 1	9
Many score 2	0
Overall NAFLD score
0	8
1	5
3	9
4	6
Fibrosis stage
None	16
1a, mild zone three perisinusoidal fibrosis	4
3, bridging fibrosis	4
4, probable or definite fibrosis	4

NASH, Nonalcoholic steatohepatitis; CRN, Clinical Research Network; NAFLD, nonalcoholic fatty liver disease.

SWV by ARFI

The mean SWV was 2.73 m/s (median = 2.77). The mean SWV showed a decreasing trend with increasing hepatic steatosis (P = .057) (Table 2). The Spearman's rank correlation coefficient (r = −0.471) showed a significant negative correlation between SWV and hepatic steatosis grade, with increasing steatosis stage resulting in decreasing SWV (P = .011) (Table 3).

Table 2.

Analysis of Differences in Mean SWV According to Body Mass Index, Steatosis, Inflammation and Ballooning Grades, and Fibrosis Staging

	Mean SWV	Standard deviation	P
Obesity class according to body mass index
Morbid obesity Class I (30–35)	2.49	1.11	.883
Morbid obesity Class II (35–40)	3.04	1.90
Morbid obesity Class III (40–50)	2.58	0.90
Superobese (50–60)	2.95	0.99
Super superobese (>60)	2.94	0.67
Steatosis score
<5% score 0	3.23	0.72	.057
5%–33% score 1	2.85	0.99
33%–66% score 2	2.25	0.70
>66% score 3	1.92	1.19
<5% score 0	2.73	0.93
Lobular inflammation score			.372
Nil score 0	2.92	1.01
<2 foci score 1	2.60	0.87
Ballooning score
Nil score 0	2.75	0.92	.885
Few score 1	2.70	1.00
Fibrosis staging
None	2.70	1.02	.784
1a, mild zone three perisinusoidal fibrosis	3.15	0.81
3, bridging fibrosis	2.70	1.15
4, probable or definite fibrosis	2.47	0.51

SWV, shear wave velocity.

Table 3.

Analysis of the Relationship Between Mean SWV and Steatosis, Inflammation and Ballooning Grades, and Fibrosis Staging

	Correlation coefficient	P
Steatosis score	−0.471	.011
Lobular inflammation score	−0.174	.375
Ballooning score	−0.33	.867
Fibrosis staging	−0.23	.885

SWV, shear wave velocity.

The mean SWV was neither significantly different nor showed a significant correlation among the different morbidity classes, inflammation grades, ballooning grades, and fibrosis stages of NAFLD (Tables 2 and 3).

Eleven had the presence of fibrosis, and 17 had no fibrosis. Receiver operating characteristic analysis showed no significant area under curve (AUROC = 0.497, P = .114) for diagnosis of fibrosis using SWV. Valid SWV could be acquired in all subjects; however, only 21.42% (6/28) fulfilled the interquartile range criterion in the morbidly obese subjects.

Discussion

Liver biopsy is the gold standard for diagnosis of nonalcoholic liver disease complicated by secondary fibrosis or cirrhosis. However, because of poor patient acceptability and associated risk of liver biopsy, there has been an increasing interest in finding noninvasive methods for assessment of liver fibrosis. Detection of liver fibrosis in NAFLD has two aspects: first, its use as a diagnostic tool and second follow-up for disease progression. In obesity clinics, this holds potential for screening morbidly obese patients for liver fibrosis and follow-up for disease progression in bariatric surgery patients with high-risk histology detected at the time of surgery.

The usefulness of ARFI in chronic hepatitis–related fibrosis and cirrhosis has been proved in numerous studies.^6–11 Some recent articles have proved usefulness in NAFLD.^12,13 Its use in morbidly obese patients is still not researched in detail, and data on applicability are lacking. Several studies have identified that severe obesity is a cause of unreliable and discordant values as reported with TE.^9,12

In the present study, we analyzed a cohort of morbidly obese patients before bariatric surgery with the ARFI technique and studied the correlation of SWV with histology on liver biopsy taken at the time of bariatric surgery. Our data showed that SWV did not show significant differences or correlations among the inflammation stages or fibrosis stages. However, a significant negative correlation between hepatic steatosis and SWV was noted. Increasing hepatic steatosis resulted in decreasing SWV. This was reflected as a small difference in the means between the steatosis stages, although not statistically significant. Only 21.42% fulfilled the interquartile range criterion among the morbidly obese subjects.

A single study reported their preliminary results of the use of ARFI in morbidly obese patients before bariatric surgery. Based on liver histology, this study classified patients into three groups: simple steatosis, steatohepatitis, and fibrosis. They found increasing SWV with progression of disease, with 1.3 m/second being a cutoff for NASH or fibrosis. However, this group did not analyze the different fibrosis stages and looked at NASH and/or fibrosis as a single entity.¹³ A study published in abstract form, wherein ARFI was performed before and after very low–calorie diet before bariatric surgery, failed to show usefulness for detection or monitoring of NASH.¹⁴ Another study published in abstract form reported the ability to qualify the interquartile range criterion with ARFI in only 20% and inability to detect fibrosis or the cirrhosis in morbidly obese patients before bariatric surgery, similar to our findings.¹⁵ We speculate that our discordant findings may be because tissue displacement cannot be efficiently induced secondary to dampening of acoustic push pulse through the dense subcutaneous fat layers of morbidly obese patients. Hepatic steatosis also interferes with the SWV in these patients in contrast to those with a normal BMI as was noted in our study. Morbid obesity appears to be a possible limiting factor for implementation of ARFI in detecting fibrosis. Further prospective studies on larger group of patients are required to establish its role in this setting.

In conclusion, our preliminary findings suggest that SWV values do not correlate with fibrosis on liver biopsy in morbidly obese patients and lack accuracy for diagnosis. Discordant values are related to higher body mass index and increasing hepatic steatosis.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Beymer

, Kowdley

, Larson

, Edmonson

, Dellinger

, Flum

. Prevalence and predictors of asymptomatic liver disease in patients undergoing gastric bypass surgery. Arch Surg 1960, 2003; 138:1240–1244.

Praveenraj

, Gomes

, Kumar

, Karthikeyan

, Shankar

, Parthasarathi

, et al. Prevalence and predictors of non-alcoholic fatty liver disease in morbidly obese South Indian patients undergoing bariatric surgery. Obes Surg, 2015.

Teli

, James

, Burt

, Bennett

, Day

. The natural history of nonalcoholic fatty liver: A follow-up study. Hepatology, 1995; 22:1714–1719.

Praveen Raj

, Gomes

, Kumar

, Senthilnathan

, Karthikeyan

, Shankar

, et al. The effect of surgically induced weight loss on nonalcoholic fatty liver disease in morbidly obese Indians: “NASHOST” prospective observational trial. Surg Obes Relat Dis, 2015; 11:1315–1322.

Sumida

, Nakajima

, Itoh

. Limitations of liver biopsy and non-invasive diagnostic tests for the diagnosis of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis. World J Gastroenterol, 2014; 20:475–485.

Sporea

, Şirli

, Bota

, Fierbinţeanu-Braticevici

, Petrişor

, Badea

, et al. Is ARFI elastography reliable for predicting fibrosis severity in chronic HCV hepatitis?. World J Radiol, 2011; 3:188–193.

Sporea

, Bota

, Peck-Radosavljevic

, Sirli

, Tanaka

, Iijima

, et al. Acoustic radiation force impulse elastography for fibrosis evaluation in patients with chronic hepatitis C: An international multicenter study. Eur J Radiol, 2012; 81:4112–4118.

Rizzo

, Calvaruso

, Cacopardo

, Alessi

, Attanasio

, Petta

, et al. Comparison of transient elastography and acoustic radiation force impulse for non-invasive staging of liver fibrosis in patients with chronic hepatitis C. Am J Gastroenterol, 2011; 106:2112–2120.

Lupsor

, Badea

, Stefanescu

, Sparchez

, Branda

, Serban

, et al. Performance of a new elastographic method (ARFI technology) compared to unidimensional transient elastography in the noninvasive assessment of chronic hepatitis C. Preliminary results. J Gastrointest Liver Dis, 2009; 18:303–310.

10.

Lee

, Kim

. Non-invasive diagnosis of hepatitis B virus-related cirrhosis. World J Gastroenterol, 2014; 20:445–459.

11.

Bota

, Herkner

, Sporea

, Salzl

, Sirli

, Neghina

, et al. Meta-analysis: ARFI elastography versus transient elastography for the evaluation of liver fibrosis. Liver Int, 2013; 33:1138–1147.

12.

Palmeri

, Wang

, Rouze

, Abdelmalek

, Guy

, Moser

, et al. Noninvasive evaluation of hepatic fibrosis using acoustic radiation force-based shear stiffness in patients with nonalcoholic fatty liver disease. J Hepatol, 2011; 55:666–672.

13.

Osaki

, Kubota

, Suda

, Igarashi

, Nagasaki

, Tsuchiya

, et al. Shear wave velocity is a useful marker for managing nonalcoholic steatohepatitis. World J Gastroenterol, 2010; 16:2918–2925.

14.

Guzmán-Aroca

, Frutos-Bernal

, Bas

, Luján-Mompeán

, Reus

, Berná-Serna J de

, et al. Detection of non-alcoholic steatohepatitis in patients with morbid obesity before bariatric surgery: Preliminary evaluation with acoustic radiation force impulse imaging. Eur Radiol, 2012; 22:2525–2532.

15.

Chakravartty

, Jaffer

, Zen

, Dent

, Clarke

, Sidhu

, et al. PTH-088 non invasive monitoring of fatty livers in morbidly obese patients: Preliminary evaluation with acoustic radiation force impulse imaging. Gut, 2014; 63(Suppl 1):A248–A249.

16.

Poster Sessions.. Hepatology. 2013; 58(1_suppl):92A–207A.