WhatsApp Us (24/7)
+91 926 888 0303 (24/7)
Abstract: Obesity has emerged as a global epidemic. Non-alcoholic fatty liver disease can progress to liver fibrosis, cirrhosis and hepatocellular carcinoma. Efforts to detect liver fibrosis early is crucial for effective intervention and preventing advanced stages like cirrhosis or liver failure. Liver Biopsy is gold standard for diagnosing liver fibrosis. The study assesses the efficacy of ultrasonography and non-invasive methods, including AST to platelet ratio index (APRI), fibrosis-4 (FIB-4) and NAFLD Score, in patients with morbid obesity scheduled for bariatric surgery.
A prospective observational study was conducted among 53 patients having morbid obesity and who were planned to undergo bariatric surgery. Pre-operative ultrasound (USG) and invasive scores, such as FIB-4 and APRI, were found to be inadequate screening tools for diagnosing advanced liver fibrosis in patients undergoing bariatric surgery. Conversely, the NAFLD score proved to be an excellent tool for this purpose.
Keywords: Morbid Obesity, Liver Fibrosis, Non-invasive Fibrosis score, Liver Biopsy, Bariatric Surgery
Introduction
Obesity has emerged as a global epidemic, impacting millions across developed and underdeveloped nations and giving rise to various health complications[1]. It is a complex, multifactorial condition associated with dyslipidaemia, metabolic syndrome, insulin resistance and non-alcoholic fatty liver disease (NAFLD)[2]. The consequences of obesity encompass hypertension, type 2 diabetes mellitus, coronary artery disease (CAD) and the development of liver fibrosis, posing significant threats to both liver function and overall well-being[3].
Non-alcoholic fatty liver disease, a prevalent condition linked to obesity, involves the accumulation of liver fat in the form of triglycerides, leading to hepatic steatosis and subsequent inflammation. This progression can result in non-alcoholic steatohepatitis (NASH) and ultimately, severe outcomes such as liver fibrosis, cirrhosis and hepatocellular carcinoma[4]. The management of obesity involves various strategies, including dietary modifications, physical activity, medications and surgical interventions like bariatric surgery[5]. Notably, the success of surgical procedures can be influenced by factors such as the status of liver fibrosis, adding a layer of complexity to patient outcomes.
Understanding the intricate relationship between obesity and liver fibrosis is imperative, considering the profound impact on morbidity and mortality during surgical procedures[6]. The study delves into evaluating the diagnostic methods employed for identifying and managing liver fibrosis.
Efforts to detect liver fibrosis early are crucial for effective intervention and preventing advanced stages like cirrhosis or liver failure. Various diagnostic methods, ranging from non-invasive imaging techniques to invasive procedures like liver biopsy, are employed. The study assesses the efficacy of ultrasonography and non-invasive methods, including AST to platelet ratio index (APRI), fibrosis-4 (FIB-4) and NAFLD score, in patients with morbid obesity scheduled for bariatric surgery. The evolving landscape of liver fibrosis diagnosis emphasises the need to compare newer modalities with established gold standards for more accurate assessments and improved patient care.
Methodology
Study Design
The present investigation adopts a Prospective Observational Study design.
Study Population
Conducted at a specialised tertiary care facility in the National Capital Region of Delhi, the study focuses on patients with morbid obesity planned for bariatric surgery
Study Duration
Spanning from March 2023 to January 2024, the study encompasses a ten-month duration at the specified centre.
Study Subjects
Encompassing individuals aged 18 years and above, the study involves patients with morbid obesity scheduled for bariatric surgery who consented to participate.
Inclusion criteria:
- Patients 18 years or older undergoing bariatric surgery.
- Clinically diagnosed with morbid obesity
Exclusion criteria:
- Exclusion of patients using hepatotoxic drugs or therapies causing hepatic steatosis.
- Exclusion of patients with HIV, HCV, or Hepatitis B infection.
- Exclusion of individuals with daily alcohol intake exceeding 20 g for women and 30 g for men.
Sample size:
Calculated based on the observational nature, the sample size follows the formula N = 4PQ/L², determining 53 participants considering a 10% non-response rate.
Statistical analysis:
Coded and entered MS Office Excel, data undergoes analysis in IBM SPSS Statistics 21.0. Descriptive statistics, chi-square tests, sensitivity analysis and kappa statistics are employed, with a significance level set at p < 0.05. Ethical clearance and informed consent are secured, ensuring confidentiality and participant rights.
Results
In this study, all participants scheduled for bariatric surgery underwent extensive laboratory investigations and ultrasound assessments for the evaluation of liver fibrosis. During the surgical procedures, liver biopsy samples were obtained and later subjected to histopathological analysis. The liver enzyme levels, specifically AST and ALT, were examined, revealing that 60.38% of participants had AST levels below or equal to 40 units/l and 75.47% had ALT levels below or equal to 56 units/l. The mean platelet count was found to be 275.22 ± 66.4, with only 5.66% of participants having platelet counts exceeding 4 lakh/cumm. Serum albumin levels were within the range of 3.4 to 5.4 g/dl for all participants.
The distribution of participants based on diabetes status showed that 75.47% did not have diabetes, while 24.52% were previously diagnosed with diabetes. Ultrasonography was employed to assess liver fibrosis, categorising participants into those with advance fibrosis (28.3%) and non-advance fibrosis (71.7%). Subsequent liver biopsy and histopathological analysis resulted in 83.02% of participants being identified as having non-advance fibrosis and 16.98% with advance fibrosis.
Non-invasive screening tools, including FIB-4 score, NAFLD score and APRI score, were utilised to determine the risk of advance fibrosis. FIB-4 score identified that 71.7% had a low risk, NAFLD score showed 52.8% with intermediate risk and APRI score indicated 83% with low risk. Sensitivity analysis revealed that ultrasound had a sensitivity of 33.33% and specificity of 72.7%, while FIB-4 Score demonstrated a sensitivity of 0% and specificity of 100%. NAFLD score exhibited a high sensitivity of 100% and low specificity of 29.5%, while APRI Score had a sensitivity of 0% and specificity of 100%.
Predictive values of these screening tools were compared, with NAFLD score showing the highest positive predictive value (PPV) of 75% and the best negative predictive value (NPV) of 100%. Kappa agreement analysis indicated that only ultrasound had slight agreement with the gold standard (liver biopsy), although the association was non-significant. The other screening modalities did not demonstrate significant agreement with the gold standard.
| Sl. No | Biopsy report | Frequency | Percentage (%) |
|---|---|---|---|
| 01. | Advance fibrosis | 09 | 16.98% |
| 02. | Non-advance fibrosis | 44 | 83.02% |
| 03. | Total | 53 | 100% |
Table 1: Distribution of study participants according to outcome of the biopsy and histopathological analysis (N=53)
| Ultrasound results | Fibrosis status
|
Total | Significance | ||
|---|---|---|---|---|---|
| Advance fibrosis |
|
15 | p=0.499$ | ||
| Non-advance fibrosis |
|
38 | |||
| Total |
|
53 |
Table 2: Sensitivity analysis with ultrasound as screening tool (N=53)
*Sensitivity and # Specificity for ultrasound as screening tool. $Fisher’s Exact Test
| FIB-4 scoring | Fibrosis status
|
Total | Significance | ||
|---|---|---|---|---|---|
| High |
|
00 | * | ||
| Low |
|
38 | |||
| Total |
|
38 |
Table 3: Sensitivity analysis with FIB-4 score as screening tool (n=38))
*Can’t be calculated as values in more than one column are zero
| NAFLD Score | Fibrosis Status
|
Total | Significance | ||
|---|---|---|---|---|---|
| High risk |
|
12 | p<<<0.05$ | ||
| Intermediate risk |
|
28 | |||
| Low risk |
|
13 | |||
| Total |
|
53 |
Table 4: Sensitivity analysis with NAFLD Score as screening tool (N=53)
*Sensitivity and # Specificity for NAFLD Score as screening tool. $Fisher’s exact test
| APRI score | Fibrosis Status
|
Total | Significance | ||
|---|---|---|---|---|---|
| High |
|
00 | * | ||
| Low |
|
44 | |||
| Total |
|
44 |
Table 5: Sensitivity analysis with APRI score as screening tool (n=44)
*Can’t be calculated as values in more than one column are zero
| Sl. No | Screening tool used | Predictive value | % age |
|---|---|---|---|
| 01. | Ultrasound | NPV | 84.2% |
| PPV | 20 | ||
| 02. | FIB-4 score | PPV | 00% |
| NPV | 92.1% | ||
| 03. | NAFLD score | PPV | 75% |
| NPV | 100% | ||
| 04. | APRI score | PPV | 00% |
| NPV | 86.4% |
Table 6: Predictive values of various screening tools
| Sl. No | Screening tool used | Kappa Agreement | Significance |
|---|---|---|---|
| 01. | Ultrasound | 0.048 (Slight) | P=1.0 |
| 02. | FIB-4 Score | 0.00 (Nil) | Can’t be Calculated |
| 03. | NAFLD Score | 0.048 (Slight) | P=1.0 |
| 04. | APRI Score | 0.0 (Nil) | Can’t be Calculated |
Table 7: Details of kappa agreement analysis for various scores used for risk analysis for liver fibrosis when assessed in comparison to liver biopsy (N=53)
Discussion
Discussion In the present study, it was observed that majority of them (60.38%) had their AST level at ≤40 units/l i.e. within normal limit. While for ALT enzyme, three-fourths of them (75.47%) had normal level ≤ 56 units/l. These estimates differed from the results reported by P Pathik et al13 and Jiang Wu et al[10], who reported that 45.45% and 12% of the study participants respectively had AST values in abnormal range and 58.18% and 11% of the study participants respectively having ALT values in abnormal range.
In the current study, patients were assessed for fatty liver using ultrasound. It was observed that close to three fourth of them (71.7%) had non-advance fibrosis (Grade 1 or 2 fatty liver), while 28.3% of the study participants had advance fibrosis (Grade 3 fatty liver). Petrick et al[21] and Jiang Wu et al[10] in their study reported completely different findings for ultrasound as a screening tool, with 65% and 70% of the study participants respectively being reported as having abnormal ultrasound and rest 35% and 30% participants respectively having normal ultrasound findings. The ethnicity and dietary behaviour may lead to such an observation in each set of population.
When histopathological analysis of liver biopsy was considered for the study participants in the present study, it was observed that many of them (83.02%) had non-advance (stage 0,1 and 2) liver fibrosis, while 09 (16.98%) of the study participants had advance fibrosis (stage 3 and 4). Similarly, Didoné Filho CN et al[7] reported in their study with only 3.6% of the patients having advance fibrosis and rest of them having non-advance fibrosis. But the results of the present study differed from the ones reported by Jiang Wu et al[10] in their study, who reported that 71% of the study participants were found positive on liver biopsy and 29% of the participants were found to be negative on liver biopsy.
Various non-invasive tests were employed to assess the risk of developing advance fibrosis among the study participants. These tests included FIB-4 score, NAFLD score and APRI score.
In case of FIB-4 score, it was observed that almost three fourth (71.7%) of the participants were found to be on low risk, while none of them could be categorised as having high risk for liver fibrosis. One third (28.3%) of participants had intermediate risk of developing advance fibrosis. A different result was reported by A Drolz et al[12] in their study where 45% of the patients were classified as having risk of advance fibrosis when score high on FIB-4 score. This difference of estimate may be attributed to the fact that there are multiple parameters which determine the results with respect to a particular test and given the fact that both studies were conducted in completely different set of populations, the results may vary in several estimates.
NAFLD score to ascertain risk of developing advance fibrosis was also calculated for study participants. It was found that majority (52.8%) of them had intermediate risk of developing advance fibrosis and close to one fourth of them had high (22.6%) and low (24.5%) risk for advance fibrosis. A comparative result was given in the studies conducted by A Drolz et al[12] and Didoné Filho CN et al[7], who could exclude 57% and 51.8% of patients from risk of advance fibrosis.
Didoné Filho CN et al[7] also reported to have excluded clinically significant fibrosis among 85.95 of the study participants using APRI score. Similarly in the current study also, when APRI score was used to ascertain this risk, it was observed that a large number (83%) of the participants had low risk. While none of them had high risk for the same in the present study, 17% of participants had intermediate risk of developing advance fibrosis.
All non-invasive tests along with ultrasound were assessed for sensitivity analysis to exclude advance stage of fibrosis as compared to results reported over liver biopsy. Ultrasound was found to have a sensitivity of 33.33% and a specificity of 72.7%. A sensitivity of 88.9% and specificity of 44.6% was reported for ultrasound by Leivas G et al[11] which differs from the findings of the current study but is similar to the review study by Karanjia et al14 who gave sensitivity of ultrasound in detecting liver fibrosis between 60-94% and a high specificity of 84% as well, which corresponds to the findings of the present study. Jiang Wu et al[10] reported a sensitivity of 86% and specificity of 68% for ultrasound in such case. These various results also emphasise the role of examiner while using ultrasound as a modality to exclude advanced liver fibrosis.
FIB-4 Score was also observed to be a very poor screening but an excellent diagnostic tool with a specificity of 79.55%. Similar results were reported by Sebastiani G et al[15] in their study measuring the sensitivity of FIB-4 at 56.3% and specificity at 91.2%. Huang et al8 also reported a similar specificity of 77.9% for Fib-4 but a different sensitivity of 67.7% for FIB-4. A different sensitivity of 97% and specificity of 34% was reported by A Drolz et al[12] in their study which questions the validity of the score as a tool.
Huang et al[8] and Sebastiani G et al[15] reported sensitivity of 55.9% and 50% respectively and specificity of 82.3% and 91.2% respectively for NAFLD Score in their study which is completely different from the results of this study where a sensitivity of 100% and a of 29.5% is being reported for NAFLD Score.
APRI Score was found to have a sensitivity of 0 % and an excellent specificity of 86.37%. Sebastiani G et al[15] also reported similar specificity for APRI (92.3%) with a sensitivity of 50%. Udelsman BV et al[9] reported a very low sensitivity of 24% for APRI in their study which is in comparison to the sensitivity reported in the present study.
Various predictive values (Positive Predictive Values and Negative predictive values) for different non-invasive tests including ultrasound were also compared. It was observed that though all tools except NAFLD score had low PPV of 75% and a negative predictive value (NPV) of 100% which was reported to be highest among all. Similar results were also reported by P Pathik et al13 and Udelsman BV et al[9], who estimated the PPV of NAFLD score to be 69% and 85% respectively and a NPV of 100% and 99% respectively. Though P Pathik et al[13] also reported another tool (APRI) to have a good NPV (84%).
Ultrasound was reported to have a PPV of 76.8% and a NPV of 66% as given by Leivas G et al11, which is different from the results of the current study where PPV for ultrasound is reported to be only 20% and NPV of ultrasound is reported to be 84.2%.
Given these values of sensitivity, specificity and predictive values of various non-invasive including ultrasound, it can be commented that this tools except ultrasound when used independently may not be able to exclude the correct proportion of patients having advance fibrosis of liver. Ultrasound on the other hand, though a moderate tool according to the findings of the current study can be used singly to exclude significant or advanced fibrosis among the patients.
All the non-invasive tests were also assessed in agreement with liver biopsy to identify the status of fibrosis among morbidly obese patients undergoing bariatric surgery. It was found that only ultrasound had slight but non-significant agreement with the gold standard. Similar accuracy for ultrasound was also reported by Jiang Wu et al[10] in their study with a diagnostic accuracy of 0.81. But Didoné Filho CN et al[7] also reported good agreement of all non-invasive tests except NAFLD score which contrasts with the results of the present study. This variation of agreement in present study can be seen because of the small sample size enrolled in the current study as compared to the other studies.
CONCLUSION:
Pre-op USG and invasive scores such as FIB 4, APRI scores are poor screening tools for diagnosing advance liver fibrosis, in patient undergoing bariatric surgery, but NAFLD Score is an excellent tool for the same.
References
- Tiwari A, Balasundaram P. Public Health Considerations Regarding Obesity. [Updated 2023 Jun 5]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2023 Jan-. Available from: https://www.ncbi.nlm.nih.gov/books/ NBK572122/ [last accessed on January 16, 2024]
- Brazil. Basic attention supplement 38. Strategies to health care — obesity.Health Attention Department; 2014, 212 p
- Seeberg, K.A., Hofsø, D., Borgeraas, H. et al. Association between hepatic steatosis and fibrosis with measures of insulin sensitivity in patients with severe obesity and type 2 diabetes - a cross-sectional study. BMC Gastroenterol 22, 448 (2022). https://doi.org/10.1186/s12876-022-02550-0
- Adams LA, Lymp JF, Sauver JS, et al. The natural history of non-alcoholic fatty liver disease: a population-based cohort study. Gastroenterology2005;129(1):113–21
- Brazil. Bariatric consent. Brazilian Society of Metabolic and Bariatric Surgery;2008, 212 p
- Khanam A, Saleeb PG, Kottilil S. Pathophysiology and Treatment Options for Hepatic Fibrosis: Can It Be Completely Cured? Cells. 2021; 10(5):1097. https://doi.org/10.3390/cells10051097
- Didoné Filho CN, Reginatto CJ, Ivantes CAP, Strobel R, Percicote AP, Petenusso M, Sobreira da Silva CJR, Benjamim CJR, Radominski RB. Comparison between non-invasive methods and liver histology to stratify liver fibrosis in obese patients submitted to bariatric surgery. Obes Res Clin Pract. 2021 Mar-Apr;15(2):152-156. doi: 10.1016/j.orcp.2021.02.004. Epub 2021 Feb 27. PMID: 33648885
- Huang Y, Dong S, Wang C, Dong Z and Chen W (2023) Significant fibrosis assessed by liver biopsy among Chinese bariatric surgery patients: A prospective crosssectional study. Front. Endocrinol. 14:1090598. doi: 10.3389/ fendo.2023.1090598
- Udelsman BV, Corey K, Hutter MM, Chang DC, Witkowski ER. Use of noninvasive scores for advanced liver fibrosis can guide the need for hepatic biopsy during bariatric procedures. Surg Obes Relat Dis. 2021 Feb;17(2):292-298. doi: 10.1016/j. soard.2020.09.037. Epub 2020 Oct 8. PMID: 33153965
- Wu J, You J, Yerian L, Shiba A, Schauer PR, Sessler DI. Prevalence of liver steatosis and fibrosis and the diagnostic accuracy of ultrasound in bariatric surgery patients. Obes Surg. 2012 Feb;22(2):240-7. doi: 10.1007/s11695-011-0509-8. PMID: 21901286
- Leivas G, Maraschin CK, Blume CA, Telo GH, Trindade MRM, Trindade EN, Diemen VV, Cerski CTS, Schaan BD. Accuracy of ultrasound diagnosis of nonalcoholic fatty liver disease in patients with classes II and III obesity: A pathological image study. Obes Res Clin Pract. 2021 Sep-Oct;15(5):461-465. doi:
- Drolz A, Wolter S, Wehmeyer MH, Piecha F, Horvatits T, Schulze Zur Wiesch J, Lohse AW, Mann O, Kluwe J. Performance of non-invasive fibrosis scores in non-alcoholic fatty liver disease with and without morbid obesity. Int J Obes (Lond). 2021 Oct;45(10):2197-2204. doi: 10.1038/s41366-021-00881-8. Epub 2021 Jun 24. PMID: 34168277; PMCID: PMC8455320
- Pathik P, Ravindra S, Ajay C, Prasad B, Jatin P, Prabha S. Fibroscan versus simple non-invasive screening tools in predicting fibrosis in high-risk non-alcoholic fatty liver disease patients from Western India. Ann Gastroenterol. 2015 AprJun;28(2):281-286. PMID: 25830783; PMCID: PMC4367221
- Karanjia RN, Crossey MM, Cox IJ, Fye HK, Njie R, Goldin RD, Taylor-Robinson SD. Hepatic steatosis and fibrosis: Non-invasive assessment. World J Gastroenterol. 2016 Dec 7;22(45):9880-9897. doi: 10.3748/wjg.v22.i45.9880. PMID: 28018096; PMCID: PMC5143756
- Sebastiani G, Alshaalan R, Wong P, Rubino M, Salman A, Metrakos P, Deschenes M, Ghali P. Prognostic Value of NonInvasive Fibrosis and Steatosis Tools, Hepatic Venous Pressure Gradient (HVPG) and Histology in Nonalcoholic Steatohepatitis. PLoS One. 2015 Jun 17;10(6):e0128774. doi: 10.1371/journal. pone.0128774. PMID: 26083565; PMCID: PMC4471234