**9. Non-invasive assesment in NAFLD**

the calculation of the liver-to-spleen attenuation ratio, those correlate with steatosis degree. Liver density as measured by CT attenuation units has been shown to have an inverse correla‐ tion to the degree of fatty infiltration. Non-enhanced CT provides a high performance in quali‐ tative diagnosis of hepatic steatosis when fatty infiltration is over 30%, obtaining 82% of sensitivity and 100% specificity using histologic analysis of biopsies of liver donors as the refer‐ ence standard [45], however is not sensitive in detecting mild-to-moderate amounts of steato‐ sis between 5% and 30% [43]. New CT scanning techniques are developing, such as dualsource/dual energy scanners, but their evaluation needs further studies. A drawback of this technique is the liver iron overload because it increases the attenuation. This method is associ‐

Magnetic Resonance can detect steatosis by exploiting the difference of resonance frequen‐ cies between water and fat proton signals. The sensitivity and specificity of MRI in detecting as low as 5% of liver fat infiltration are 85% and 100%, respectively [46]. The detection of the fatty liver can be seen in "white/bright" when applying in-phase T1 images and "black" when applying out-of-phase images, compared to the signal intensity of the spleen and par‐ aespinal muscles. Another technic of MR imaging with fat saturation may quantify more ac‐

MR spectroscopy can reliably quantify even minimal steatosis, as low as 0,5% [47]. In has been based on the ubiquitous protons hydrogen and phosphorus [48], and more than 5% of fat content on MR spectroscopy indicates presence of steatosis [49]. Its routine application is

Operator dependent.

Radiation exposure.

confounding factors.

implantable devices.

infiltration.

steatosis.

images.

Qualitative assessment of steatosis. Only accurate when moderate-to-severe fat

Iron overload, copper and fibrous tissue could be

Not sensitive for mild-to-moderate amounts of

Limitation in patients with iron overload. Not suitable for patients with claustrophobia or

ated with radiation exposure which limits its use in children.

174 Liver Biopsy – Indications, Procedures, Results

curately liver fat infiltration, especially in patients who have fibrosis.

limited by cost and lack of availability, and it remains a research tool.

**Methods (S/s) Advantages Disadvantage**

Detects focal or diffuse infiltration.

Detects focal or diffuse infiltration.

Useful for screening

content.

exposure.

Noninvasive, widely aviable, low cost, repetition,

Noninvasive, semiquantitative assessment of fat

Noninvasive, semiquantitative, and no radiation

MR spectroscopy Noninvasive, reproducible, accurate quantification. High cost not widely available. Long time taking

**Table 9.** Pro's and con's of radiologic modalities for the study of NAFLD. S: sensitivity; s: specificity. Adapted from [41].

US, CT and MR are insensitive in differentiating hepatic steatosis from NASH, and they can‐ not be used to stage fibrosis [43, 48]. But in the near future, a novel method based on MRI

Ultrasonography. (60-95% / 84-100%)

Computer tomography, Contrast images. (50-86% / 75-87%)

Magnetic resonance imaging

(85% / 100%)

Liver biopsy remains a useful tool to confirm the diagnosis and exclude other disease or helps to discover concomitant chronic liver disease. It provides prognostic information by staging and grading this disease. At present non-invasive diagnostic markers could provide a new tool for differentiating fatty liver from NASH as well as for grading /staging NAFLD.

The investigation of these new diagnostic methods comes from the well known drawbacks of liver biopsy. These include sampling error, inadequate biopsy size, variability in patholo‐ gist interpretation, cost and associated morbidity (complications 0,3%, mortality rate 0,01%).

An ideal non-invasive test should be simple, reproducible, readily available, less expensive than liver biopsy, able to predict the full spectrum of liver fibrosis stages, and reflect changes occurring with therapy [48].

Some reviews provide an overview of the role of non-invasive test in NAFLD [48, 50, 51]. We will try to present many of these scores through a table (number 10) to summarize their characteristics. AUROC is a numerical data that assess the performance of a scoring system. AUROC value greater than 0,8 indicate good diagnostic performance. The closer the value to 1, the better performing the scoring system.



**SCORE [Reference] Variables Cutoff AUROC Sens. (%) Spec. (%) PPV (%) NPV (%)**

<0,611 F0-1 vs F2-4:

16 kPa

1,90 m/sec

Means: - 2,51kPa. - 3,24kPa. - 4,16kPa.

ARFI [69] 4,24 kPa F0-2 vsF3-4: 0,90 90 90

Fibrometer [67] Glucose, AST, platelets, ferritin, ALT, weight, age.

Fibroscan [68] Transient elastography 9,9 kPa

ARFI [68] 1,77 m/sec

simple steatosis (SS)

both techniques can be performed at the same time.

likelihood of having bridging fibrosis and/or cirrhosis [1].

**Table 10.** Non-invasive assessment of NAFLD.

MAGNETIC RESONANCE IMAGING BASED TECHNIQUES MR elastography[70] Discriminating NASH from

ULTRASOUND BASED TECHNIQUES IN FIBROSIS DETECTION IN NAFLD

F0-2 VS F3-4: 0,81-0,92

0,936-0,952

F ≥3: 0,99 F4: 0,998

F ≥3: 0,973 F4: 0,976

Simple steatosis. NASH no fibrosis. With fibrosis

Abbreviations: AUROC: area under receiver operator curve. Sens.: Sensitivity; Spec.: Specificity; PPV: positive predictive value; NPV: negative predictive value; BMI: body mass index; GGT: Gamma-glutamyl-transpeptidase; HA: hyaluronic acid; AST: aspartate transaminase; AAR: AST/alanine aminotransferase, TIMP-1: tissue inhibitor of matrix metallopro‐ teinase 1; P3NP: aminoterminal peptide of pro-collagen III; ARFI: acoustic radiation force impulse; \* Sample: bariatric surgery patients. COMMENTS: BARD score can reliably exclude advance fibrosis, particularly among non-diabetics. FIB-4, as happened with BARD score, is useful in excluding advance fibrosis due its high NPV. Transient elastography and ARFI are based on the variation of the speed wave through liver tissue (generated by vibrator/short-duration acoustic pulses, respectively), this can be measured and converted to a numerical value (in kPa and m/sec, respectively, but ARFI could also be expressed as kPa) which is the liver stiffness and it is proportional to liver fibrosis. An important difference between both systems is that ARFI consists in a probe which can be plugged to a common US machine so

APRI and FIB-4 have been evaluated in obese children and they might be useful in this special population [71]. Pediatric NAFLD scores is a noninvasive model evaluated in obese children,

In the future, new serologic markers, such as CD36, will help to differentiate more accurate‐ ly between NAFLD stages, we would be able to distinguish simple steatosis from NASH.

Although clinical and laboratory models may be useful in identifying a group of patients at a low risk of advance fibrosis and liver biopsy might be avoided, they are not enough for staging and prognostic purposes if patients are at risk of advance fibrosis [48]. NAFLD Prac‐ tice Guideline of 2012 recommends NAFLD fibrosis score to identify patients with higher

and it may help clinicians to predict liver fibrosis but external validation is needed [60].

0,70 98 73

100 100

100 100

2,74 kPa SS vs NASH: 0,93 94 73 85 89

93 93

Nonalcoholic Fatty Liver Disease: A Pathological View

http://dx.doi.org/10.5772/52622

177

91 96 77 86

71 75 100 100

100 100


Abbreviations: AUROC: area under receiver operator curve. Sens.: Sensitivity; Spec.: Specificity; PPV: positive predictive value; NPV: negative predictive value; BMI: body mass index; GGT: Gamma-glutamyl-transpeptidase; HA: hyaluronic acid; AST: aspartate transaminase; AAR: AST/alanine aminotransferase, TIMP-1: tissue inhibitor of matrix metallopro‐ teinase 1; P3NP: aminoterminal peptide of pro-collagen III; ARFI: acoustic radiation force impulse; \* Sample: bariatric surgery patients. COMMENTS: BARD score can reliably exclude advance fibrosis, particularly among non-diabetics. FIB-4, as happened with BARD score, is useful in excluding advance fibrosis due its high NPV. Transient elastography and ARFI are based on the variation of the speed wave through liver tissue (generated by vibrator/short-duration acoustic pulses, respectively), this can be measured and converted to a numerical value (in kPa and m/sec, respectively, but ARFI could also be expressed as kPa) which is the liver stiffness and it is proportional to liver fibrosis. An important difference between both systems is that ARFI consists in a probe which can be plugged to a common US machine so both techniques can be performed at the same time.

**Table 10.** Non-invasive assessment of NAFLD.

**SCORE [Reference] Variables Cutoff AUROC Sens. (%) Spec. (%) PPV (%) NPV (%)**

0,74-0,83 63-81 84-97

0,3825 0,73 71,4 72,7 83,3 57,1

0,88 56

0,79 29 98 91 71

0,85 98,5 44,5

0,81 96

0,80 43

ELF: -0,2070. No fibrosis 0,76 61 80 81 79

0,84 100 47 100

94 74 94 74

90

69%

80

75 86 54 93

80 90 71 94

70 80 70 80

70 90

93 85

90 83

75,4\*\* 75\*\* \*\*Pre-test probability:

oxNASH [55] Detection of lipid peroxidation

176 Liver Biopsy – Indications, Procedures, Results

SHIMADA [56] Serum adiponectin, type 4

NashTest [58] Age, gender, BMI, triglycerides, cholesterol,

NONINVASIVE MANAGEMENT OF FIBROSIS IN NAFLD

NASH Diagnostics

NAFLD fibrosis score

Pediatric NAFLD fibrosis index [60]

[57]\*

[59]

products by chromatographymass spectrometric

collagen 7s level, HOMA-IR

alfa-2macroglobulin, GGT, haptoglobin, apolipoprotein-

Age, BMI, IFG/diabetes, AAR,

Age, waist circumference and

≤1,455 ≥0,676

≥9: rule in liver fibrosis.

<3:rule out fibrosis

≥2: advanced fibrosis

0,85

0,90.

fibrosis 0,82

075-085.


0-1, for septal fibrosis

>2,67

APRI [63] AST, platelet 0,98 Advanced fibrosis:

ELF [64] HA, TIMP1, P3NP. 0,3576 For severe fibrosis

A1, total bilirubin

platelet, albumin

triglycerides.

FIB-4 [62] Age, AST, platelet, ALT <1,3

BAAT [65] Age ≥50years, BMI ≥28kg/m2

ALT ≥2N

=1, triglycerides ≥7mmol/L=1,

Fibrotest [66] 0,30 F0-1 VS F2-4:

BARD score [61] BMI ≥28 =1, AAR ≥0,8 =2, diabetes=1,

CK-18, cleaved CK-18, adiponectin, resistin

> APRI and FIB-4 have been evaluated in obese children and they might be useful in this special population [71]. Pediatric NAFLD scores is a noninvasive model evaluated in obese children, and it may help clinicians to predict liver fibrosis but external validation is needed [60].

> In the future, new serologic markers, such as CD36, will help to differentiate more accurate‐ ly between NAFLD stages, we would be able to distinguish simple steatosis from NASH.

> Although clinical and laboratory models may be useful in identifying a group of patients at a low risk of advance fibrosis and liver biopsy might be avoided, they are not enough for staging and prognostic purposes if patients are at risk of advance fibrosis [48]. NAFLD Prac‐ tice Guideline of 2012 recommends NAFLD fibrosis score to identify patients with higher likelihood of having bridging fibrosis and/or cirrhosis [1].
