Assessing the value of fibro scan in the detection of liver fibrosis among patients with sickle cell disease– a single center study

Abstract

Background

Sickle cell hepatopathy (SCH) is a general term used to describe the acute and chronic manifestations of liver damage in sickle cell disease (SCD). One of the most effective methods used to examine the liver is transient elastography (Fibro Scan), a noninvasive, rapid, and reproducible method that can assess liver fibrosis by measuring liver stiffness. Objectives: This single-center study aims to investigate the role of transient elastography (Fibro Scan) in the detection of liver disease among patients with SCD in the Eastern Province of Saudi Arabia, where the rates of SCD are found to be among the highest. Design: This is a prospective, observational study.

Methods

The study was conducted among n=101 SCD patients of any phenotype between January and December 2024. Data on demographics, hospitalization rates, hydroxyurea use, transfusion, and complications were obtained from the health informatics system. All included patients were examined using non-invasive transient elastography (Fibro Scan) for the presence of fibrotic changes.

Results

Among n = 101 SCD patients, 13.9% had G6PD deficiency, 40.6% required one hospitalization per year, and 65.3% required blood transfusion. The transient elastography through Fibro Scan assessments revealed that 94.1% of patients had normal liver tissue (fibrosis score 2–7 kPa), 4.0% showed moderate scarring (7–11 kPa), and 2.0% had moderate to severe scarring (11–15 kPa). Radiographic evidence of fatty liver was present in 24.8% of cases, with a significant association with moderate or severe fibrosis (p = 0.049).

Conclusions

Patients with SCD without iron overload had a low prevalence of liver fibrosis. However, fatty liver in SCD likely leads to a risk of fibrosis progression, which is an area for further study.

Lay summary/Key points

• • Sickle cell disease (SCD) Patient without iron overload caries a low risk of having liver fibrosis

• • Fatty liver is a significant risk for liver fibrosis in SCD patients.

• • Fibro Scan can be used as a tool to investigate liver disease in SCD patients and in detection of liver fibrosis in individualized manner.

• • No available guidelines supporting early screening for liver fibrosis in SCD by using Fibro Scan.

• • No much data about liver condition and SCD with/without G6PD deficiency.

• • This study is considerable as a key for further research question around liver in hematopathies that will be interesting for the readers and researchers

Introduction

Sickle cell hepatopathy (SCH) is a general term used to describe the acute and chronic manifestations of liver damage in sickle cell disease (SCD), one of the most common monogenetic disorders.^1–3 The SCH occurs predominantly in patients with homozygous SCD and, to a lesser extent, in patients with sickle cell trait, hemoglobin SCD (HBG SC), and hemoglobin S-beta-thalassemia (HGB SB thalassemia). This health condition may be primarily caused by the sickle cell process; most commonly, it results from the multiple transfusions that SCD patients require throughout their lives.^1–3

Overall, SCH can be seen in approximately 10% of patients with SCD, with abnormal liver function tests in one-third, hepatomegaly noted in 91% of autopsies of patients with SCD, and cirrhosis noted in 16-29% of autopsies.^3–5 Acute sickle cell liver crisis occurs in approximately 10% of SCD patients and includes acute sickle cell liver crisis, hepatic sequestration, intrahepatic cholestasis, and overt liver failure. More chronic conditions of SCH include viral hepatitis, blood transfusion, cholelithiasis, and sickle cell cholangiopathy.^3–5 The risk of SCH is increased by the need for multiple blood transfusions and may exacerbate damage from other hepatotropic viruses, as well as predispose to cirrhosis and hepatocellular carcinoma. ⁵

In Saudi Arabia, SCD was first reported in the Eastern Province in 1960, leading to the initiation of screening studies to determine the clinical characteristics and gene frequencies of SCD in different regions of the country. According to the latest data, in the 21st century, SCD is still relatively common in Saudi Arabia, with a prevalence of about 2.3%.^6,7 The highest rates of SCD are found in the Eastern Province, followed by the southern, central, and northern provinces. The high prevalence of SCD and SCT in Saudi Arabia poses significant challenges to the healthcare system, as people experience complications such as pain crises, infections, and stroke, leading to increased hospitalizations and expensive treatments. As a result, Saudi Arabia bears a significant financial burden, spending US$395 million per year on SCD treatment.

The Saudi government has made significant efforts to address the problem of SCD through various measures, including the implementation of a mandatory premarital screening program, which has helped keep the prevalence of the disease stable in recent years.^6–8 In addition, healthcare providers and researchers focus on complications associated with SCD that may affect other organs. ⁹ However, little data is collected on liver complications associated with SCD. Globally, one of the most effective methods used to examine the liver is transient elastography (Fibro Scan), a noninvasive, rapid, and reproducible method that can assess liver fibrosis by measuring liver stiffness. ¹⁰

Taking into consideration, the high rates of SCD and related complications in Saudi Arabian patients, particularly in the Eastern Province, this is the first study to hypothesize that patients with SCD are at risk of developing liver disease, particularly liver fibrosis, even in the absence of laboratory evidence of iron overload. Therefore, this single-center study aims to investigate the role of transient elastography (Fibro Scan) in the detection of liver disease among patients with SCD in the Eastern Province of Saudi Arabia. Secondary objectives are to evaluate possible risk factors that contribute to the presence of initial stage of liver fibrosis in patients with SCD (other than iron overload) in correlation with transient elastography findings.

Methods

Study design and setting

This is a prospective, observational study conducted in a tertiary care hospital in the Eastern Province of Saudi Arabia. The study was conducted among n=101 patients with SCD of any phenotype seen in an adult hematology outpatient clinic between January and December 2024.

Study sample

The sample size determination was based on the formula for proportion in descriptive studies:

$$\text{Unlimited}\hspace{0.17em}\text{population}:n=\frac{Z^2\times \widehat{p}(1-\widehat{p})}{{\epsilon }^2}$$

$$\text{Finite}\hspace{0.17em}\text{population}:n^{\prime }=\frac{n}{1+\frac{Z^2\times \widehat{p}(1-\widehat{p})}{{\epsilon }^{2}N}}$$

Were (z) is the z score, (ε) is the margin of error, (N) is the population size, and (p̂) is the population proportion. ¹¹ This means that to achieve a 95% confidence level that the actual value is within ±5% of the measured/examined value, 95 or more measurements/examinations must be performed, the population proportion = 10%, and the population size of patients with SCD in an adult hematology clinic in one center = 300. Based on this formula, n=101 patients were included to achieve the study aim. ¹²

Participants were recruited using convenience sampling. The sample size was estimated based on an estimated prevalence of abnormal liver function among SCD patients of 50%, as reported in the regional literature, where estimates range from 40% to 60%.^7,8 Based on this prevalence, a 95% confidence level, and a precision (margin of error) of ±8%, the minimum required sample size was calculated to be approximately 120 patients after adjusting for the final population of 300 patients with SCD registered at the center at the time of data collection. As a result, a total of 101 participants were included in the study after applying eligibility criteria.

Eligibility criteria

Inclusion criteria

Any adult patient (aged ≥ 14 years as per hospital age policy) who is eligible for treatment at KFMMC-Dhahran and diagnosed with SCD of any phenotype (HGB SS, HGB SB, HGB SE, HGB SC or HGB SD). The patient should be in stable condition, both vitally and clinically, and not in an active crisis (absence of active pain, ongoing active hemolysis, organ sequestration, acute chest syndrome, or acute central nervous system crisis, such as active hemolysis associated with glucose-6-phosphate dehydrogenase deficiency). Patients included in the study should have no history of liver disease, HIV, or hepatitis of any type (viral or autoimmune), and no known cases of iron overload based on laboratory testing. During study enrollment, active serological testing excluded viral hepatitis and HIV. Autoimmune liver disease associated with SCD is typically not tested for unless suspected, and after ruling out hepatopathy, drug intoxication, and viral diseases.

After obtaining the expected result, additional immunological laboratory tests, such as serum immunoglobulin G (IGG), serum antinuclear antibodies (ANA), serum anti-smooth muscle antibodies (ASMA), serum anti-mitochondrial antibodies (AMA), and serum anti-liver-kidney microsome-1 (LKM) antibodies were also considered. To confirm the diagnosis, a liver biopsy was performed in the absence of contraindications. In addition, iron overload was detected in patients included in the study based on elevated ferritin levels and liver magnetic resonance imaging (MRI) in the T2-star mode.

Exclusion criteria

Patients under the age of 14-years-old, cases not eligible for treatment at KFMMC-Dhahran, cases of non-sickle cell disease or sickle cell trait. To evaluate the value of fibro scanning in SCD patients, it was decided to exclude cases of SCD in which the presence of an underlying disease damaging the liver parenchyma was known. This allowed for an objective assessment of the indications for fibro scanning in SCD patients. For example, without iron overload or hepatitis. Thus, patients with diagnosed liver disease, alcoholic liver disease (ALD), HIV, hepatitis, or iron overload were excluded. Pregnant patients with SCD and patients in active crisis were also excluded.

Procedures and data collection

All included patients were examined using Fibro Scan for the evidence of fibrotic changes. According to the patient preparation protocol, the patient must abstain from food for at least 4 hours before the examination. For greater accuracy and to minimize bias, the study was performed by one operator throughout the entire study period. The study was performed using Fibro Scan touch 502 (echo Sens), S/N: F60643. The analysis of the results was based on the machine reference standard (METAVIR scale) for liver fibrosis, which was assessed based on liver stiffness. Liver stiffness was measured in kilopascals (kPa) with refences as shown in Table 1.

Table 1.

References to measurements using transient elastography (Fibro Scan) to determine the presence of fibrotic changes in n = 101 patients.

Liver stiffness	Fibrosis score (kPa)	Interpretation
F0 to F1	2 - < 7 kPa	Normal liver tissue
F2	7- <11 kPa	Has moderate scarring
F2 to F3	11 - <15 kPa	Has moderate to severe scarring
F3	15 - <17 kPa	Has severe scarring
F4	≥ 17kPa	Has cirrhosis

Footnote: kPa = kilopascals, F=fibrosis, CAP = controlled attenuation parameter, dB/m= decibels per meter.

Sociodemographic data on age, gender, comorbidities, use of medications such as hydroxyurea, and transfusion requirements were documented. SCD severity was captured by extracting the frequency of hospitalizations per year, with patients requiring ≥3 hospitalizations per year considered to have more severe disease than those requiring <3 hospitalizations per year. In addition, hemoglobin electrophoresis, phenotypes, and SCD complications were recorded. In addition to hemoglobin level, white blood cell count (WBC), platelet count, reticulocytes, and hemoglobin electrophoresis (HGB S%, HGB F%, HGB A%, and HGB A2% + HGB C%, D%, or E% if available), results of biochemical studies, including screening (ALA, GGT, ALT, AST, LDH, and blood urea nitrogen, CR, ESR, CRP, and G6PD) were collected at the time of study entry. Patients were scheduled for a clinical examination and Fibro Scan no later than one to two weeks after the laboratory test. Clinical and laboratory data for these cases were obtained from the electronic information system with the ICD code for non-critical SCD. Furthermore, radiological study (as in abdominal ultrasound (US), computer tomography (CT), magnetic resonance cholangiopancreatography (MRCP), and magnetic resonance imaging (MRI)), that was done within less than one year before the study, were also assessed for better correlation with liver status. Typically, liver imaging for SCD is not performed on a set schedule for each patient; instead, it is determined by the clinical symptoms and risk factors. However, all patients with SCD are recommended to undergo a baseline ultrasound (US), a liver MRI every 1-2 years for those who frequently receive blood transfusions, and, if necessary, an ultrasound ±MRI, with a liver panel. In addition, US ± MRI should be performed every 6–12 months to monitor chronic liver diseases such as hepatitis in SCD. The incidence of liver fibrosis was reported after performing some correlation measurements between Fibro Scan results and disease severity (hospitalization rate), using hydroxyurea, HGB S and HGB F%, phenotyping and blood transfusion rate.

Statistical analysis

The study included 101 patients with SCD who were followed up at the study site. They were included in the study based on the eligibility criteria and laboratory and radiographic data. The study then conducted fibro scanning as described previously. Data entry, cleaning and coding were performed using an Excel sheet. Data were analyzed using the Social Science Statistical Package (SPSS version 26, Armonk, NY: IBM Corp, USA). Quantitative data following normal distribution were presented by mean with standard deviation; skewed distribution was presented by median with range. Frequency (n) and percentage (%) of categorical variables were used for descriptive statistics. Qualitative data were expressed as frequency and percentage. Statical significance with demographic and clinical variables was assessed using t-tests, ANOVA and chi-square tests, while binary logistic regression was used for multivariant analysis. A p-value less than or equal to 0.05 with a 95% confidence interval was considered statistically significant. Liver stiffness was recorded in kPa and converted to mean values for analysis. CAP scores (S0–S2) were used to stratify groups. Continuous variables were summarized using count, mean, median, standard deviation, minimum, and maximum. Visualizations include tables, histograms, and boxplots.

Ethics

The study was conducted after ethical approval from the Eastern Province Armed Forces Hospital Institutional Review Board (AFHER-IRB-2023-029, IRB Review Date August 20, 2023). Written (signed) informed consent was obtained from patients after explanation of the purpose of the study. For patients included in the study under 18 years of age, consent was provided simultaneously by both the patient and legal guardian. No patient tissue was used. This study complies with the Declaration of Helsinki.

The reporting of this study complies with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement. ¹³

Results

The study was conducted among n = 101 patients with SCD, with a mean age of 27.28 years (SD = 10.98). Most of the participants were women 56.4%. Of the n = 101 patients, n = 100 (99%) had HGB SS and only one patient reported having HGB SB. Blood transfusions were administered to 65.3% of patients. Of these, 62.1% received 1 to 5 cumulative units of blood, and 37.9% received 6 to 10 cumulative units; no patient received more than 10 cumulative units of blood. The prevalence of comorbidities was generally low and included hypertension (5.9%), diabetes mellitus (3%), chronic kidney disease (2%), and rheumatological diseases (3%). Hypothyroidism and cardiac diseases were present in 2% and 3% of patients, respectively. Only one patient (1%) had bronchial asthma. Glucose-6-phosphate dehydrogenase (G6PD) deficiency was observed in 13.9% of patients, while the remaining 86.1% had normal G6PD levels (Table 2).

Table 2.

Demographic and clinical data of n = 101 SCD patients.

Variable		N (%)
Age	Mean (SD)	27.28 (10.98)
Gender	Male	44 (43.6)
	Female	57 (56.4)
Phenotype	Homozygous sickle cell disease (HGB SS)	100 (99.0)
	sickle cell+ Beta thalassemia (HGB SB)	1 (1.0)
Blood transfusion	No	35 (34.7)
	Yes	66 (65.3)
Accumulative of blood transfusion	1-5	41 (62.1)
	6-10	25 (37.9)
	>10	0 (0)
Diabetic mellitus (DM)	No	98 (97.0)
	Yes	3 (3.0)
Hypertension (HTN)	No	95 (94.1)
	Yes	6 (5.9)
Chronic kidney disease (CKD)	No	99 (98.0)
	Yes	2 (2.0)
Rheumatological disorder	No	98 (97.0)
	Yes	3 (3.0)
Hypothyroidism	No	99 (98.0)
	Yes	2 (2.0)
Cardiac disorders	No	98 (97.0)
	Yes	3 (3.0)
Bronchial asthma	No	100 (99.0)
	Yes	1 (1.0)
G6PD*	Normal	87 (86.1)
	Deficient	14 (13.9)

*Glucose-6-phosphate dehydrogenase (G6PD).

Laboratory findings revealed an average white blood cell (WBC) count of 8.76 × 10³/µL (SD = 3.38), a platelet count of 412.57 × 10³/µL (SD = 234.24), and reticulocyte count of 218.17 × 10³/µL (SD = 82.11). The mean hemoglobin (HGB) level was 9.31 g/dL (SD = 1.43). The results of hemoglobin electrophoresis showed the mean percentage values of HbS at 66.49% (SD = 25.62), HbA at 20.09% (SD = 18.38), HbA2 at 2.52% (SD = 1.59), and HbF at 43% (SD = 8.04). Liver enzymes were measured and ALK was found to be 100.24 IU/L (SD = 52.33), GGT was 33.78 IU/L (SD = 32.37), ALT was 25.51 IU/L (SD = 19.51) and AST was 34.30 IU/L (SD = 16.48). The mean LDH level was 373.43 IU/L (SD = 159.68) while CRP level showed wide variability with a mean of 29.18 mg/L (SD = 54.59) (Table 3).

Table 3.

Clinical and laboratory baseline data of n=101 patients with SCD.

Laboratory data	Normal value	Mean	SD	Minimum	Maximum
White blood Cells (WBC)	4-11×10³/ul	8.76	3.38	3.00	16.00
Platelets (PLT)	140-450 ×10³/ul	412.57	234.24	56.00	1563.00
Retics	0.5–2.5 %	218.17	82.11	68.00	530.00
Hemoglobin (HGB)	12 -16 g/dL	9.31	1.43	6.00	13.00
Hemoglobin S	0 %	66.49	25.62	25.00	90.00
Hemoglobin A	95 - 98 %	20.09	18.38	.00	79.00
Hemoglobin A2	2.0–3.5 %	2.52	1.59	.00	8.00
Hemoglobin F	< 2 %	11.43	8.04	1.00	35.00
ALK	44–147 U/L	100.24	52.33	9.00	332.00
GGT	9–48 U/L	33.78	32.37	6.00	225.00
ALT	7-40 U/L	25.51	19.51	7.00	109.00
AST	5-34 U/L	34.30	16.48	11.00	107.00
LDH	140–280 U/L	373.43	159.68	33.00	887.00
BUN	2.5-6.7mmol/L	2.28	1.30	1.00	7.50
Cr	50-98 umol/L	56.36	14.07	33.00	139.00
ESR	0-20 mm/hr	17.48	21.97	2.00	107.00
CRP	Neg ≤ 5 mg/L	29.18	54.59	.00	290.00
Ferritin	150-400ng/mL	147.89	72.6	9	347

*Keynote: ALK=Alkaline Phosphatase, GGT= Gamma-glutamyl transferase, ALT= Alanine transaminase, AST= Aspartate transaminase, LDH=lactate dehydrogenase, BUN= Blood Urea Nitrogen, Cr= Creatinine, ESR= Erythrocyte Sedimentation Rate, CRP= C-Reactive Protein.

Among n = 101 patients who underwent liver imaging, the majority 89.0% underwent ultrasound, followed by 8.0% computed tomography (CT), 2.0% magnetic resonance cholangiopancreatography (MRCP), and 1.0% magnetic resonance imaging (MRI). Fatty liver changes were observed in 24.8% of patients, while 75.2% had no evidence of fatty changes. Cirrhotic presentation and extramedullary lesions were noted in 1% of patients. Most patients (90.1%) required less than three hospitalizations during the study period. Furthermore, of the 51 patients taking hydroxyurea, 48 (94%) reported drug-related complaints, and 3 (6%) were non-compliant with the hydroxyurea regimen. In fact, this cohort did not report using any other chelating agents (Table 4).

Table 4.

Liver imaging data in n=101 patients with SCD.

​		N (%)
Radiology	Ultrasound	89 (89.0)
	CT	8 (8.0)
	MRCP	2 (2.0)
	MRI	1 (1.0)
Fatty changes	No	76 (75.2)
	Yes	25 (24.8)
Cirrhotic picture	No	100 (99.0)
	Yes	1 (1.0)
Extramedullary lesion	No	100 (99.0)
	Yes	1 (1.0)
Admission 3 times or less	No	10 (9.9)
	Yes	91 (90.1)
Admission frequency past year	.00	27 (26.7)
	1.00	41 (40.6)
	2.00	17 (16.8)
	3.00	6 (5.9)
	4.00	6 (5.9)
	5.00	1 (1.0)
	6.00	1 (1.0)
	9.00	2 (2.0)
Hydroxyurea (HU)	No	50 (49.5)
	500 mg OD	11 (10.9)
	500 mg BID	39 (38.6)
	500 mg TID	1 (1.0)
	Compliance rate	48 (94%)
Iron chelating agent	No	101(100%)

Keynote: US= ultrasound, CT=computer tomography, MRCP=magnetic resonance cholangiopancreatography, MRI=magnetic resonance imaging.

The transient elastography through Fibro Scan assessments revealed that 94.1% of patients had normal liver tissue (fibrosis score 2–7 kPa), 4.0% showed moderate scarring (7–11 kPa), and 2.0% had moderate to severe scarring (11–15 kPa). According to the liver stiffness classification, 64.4% of patients were classified as F0, 28.7% as F1, 5.0% as F2, and 2.0% as F3. Controlled attenuation parameter (CAP) analysis showed that 62.4% had no steatosis (S0), 12.9% had mild steatosis (S1), and 24.8% had moderate steatosis (S2). Complications included acute chest syndrome in 14.9%, osteoporosis in 4.0%, cranial nervous system disorders in 4.0%, nephropathy (1.0%), and aseptic necrosis (AVN) in 1.0% (Table 5).

Table 5.

Fibro Scan findings in n = 101 SCD patients.

​		N (%)
Fibrosis score (kPa)	2- <7 kPa	95 (94.1)
	7- <11 kPa	4 (4.0)
	11- <15kPa	2 (2.0)
Liver stiffness	F0	65 (64.4)
	F1	29 (28.7)
	F2	5 (5.0)
	F3	2 (2.0)
Interpretation	Normal liver tissue	95 (94.1)
	Has moderate scarring	4 (4.0)
	Has moderate to severe scarring	2 (2.0)
CAP	S0	63 (62.4)
	S1	13 (12.9)
	S2	25 (24.8)
Complications	None	76 (75.2)
	acute chest syndrome	15 (14.9)
	Nephropathy	1 (1.0)
	Osteoporosis	4 (4.0)
	cerebrovascular accident, CVA	4 (4.0)
	AVN (Hip)	1 (1.0)

Keynote: AVN =A vascular necrosis, CVA = cerebrovascular accident.

Analysis of factors associated with Fibro Scan results revealed no statistically significant associations between fibrosis severity and variables such as gender, diabetes mellitus, hypertension, chronic kidney disease, rheumatological diseases, hypothyroidism, heart disease, bronchial asthma, phenotype, blood transfusions, cirrhosis status, or extramedullary lesions (p > 0.05 for all). However, in patients with fatty liver, there was a significant association with moderate or severe fibrosis (p = 0.049). No significant correlation was found between fibrosis severity and hospitalization rates, hydroxyurea dosage, or adherence to hydroxyurea. Moreover, the p-value was significant for changes in body fat at 0.049, adjusted for age, sex, BMI, and chronic diseases, with an odds ratio of 3.3 (0.61 - 17.1). However, no significant association was found for other variables when adjusted for age, gender, BMI, and chronic diseases (Table 6).

Table 6.

Association between patient demographic and clinical data and Fibro Scan results in n = 101 SCD patients.

Variables		Normal liver tissue n (%)	Moderate scarring n (%)	Moderate to severe scarring n (%)	Un adjusted p-value	Adjusted p-value for age, gender, BMI, chronic condition)	Adjusted OR (for abnormal outcomes)	P-value
Gender	Male	41 (93.2)	3 (6.8)	0 (0.0)	0.206	0.165	Reference
	Female	54 (94.7)	1 (1.8)	2 (3.5)			0.75 (0.14-3.9)	0.165
DM	No	92 (93.9)	4 (4.1)	2 (2.0)	0.907	0.651	Reference
	Yes	3 (100.0)	0 (0.0)	0 (0.0)			2.03 (0.09-43.7)	0.651
HTN	No	90 (94.7)	3 (3.2)	2 (2.1)	0.246	0.281	Reference
	Yes	5 (83.3)	1 (16.7)	0 (0.0)			3.6 (0.35-36.9)	0.281
CKD	No	93 (93.9)	4 (4.0)	2 (2.0)	0.938	0.509	Reference
	Yes	2 (100.0)	0 (0.0)	0 (0.0)			2.87 (0.12-66.4)	0.509
Rheumatological issues	No	92 (93.9)	4 (4.1)	2 (2.0)	0.907	0.650	Reference
	Yes	3 (100.0)	0 (0.0)	0 (0.0)			2.03 (0.09-43.7)	0.650
Hypothyroidism	No	93 (93.9)	4 (4.0)	2 (2.0)	0.938	0.509	Reference
	Yes	2 (100.0)	0 (0.0)	0 (0.0)			2.87 (0.12-66.4)	0.509
Cardiac issues	No	92 (93.9)	4 (4.1)	2 (2.0)	0.907	0.650	Reference
	Yes	3 (100.0)	0 (0.0)	0 (0.0)			2.03 (0.09-43.7)	0.650
Bronchial asthma	No	94 (94.0)	4 (4.0)	2 (2.0)	0.969	0.348	Reference
	Yes	1 (100.0)	0 (0.0)	0 (0.0)			4.85 (0.18-131.1)	0.348
Phenotype	HGB SS	94 (94.0)	4 (4.0)	2 (2.0)	0.969	0.348	Reference
	HGB SB	1 (100.0)	0 (0.0)	0 (0.0)			4.85 (0.18-131.1)	0.348
Blood transfusion	No	31 (88.6)	3 (8.6)	1 (2.9)	0.196	0.112	Reference
	Yes	64 (97.0)	1 (1.5)	1 (1.5)			0.24 (0.04-1.4)	0.112
Fatty changes	No	73 (96.01)	3 (4.0)	0 (0.0)	0.049*	0.042*	Reference
	Yes	22 (88.0)	1 (4.0)	2 (8.0)			3.3 (0.61-17.1)	0.042*
Cirrhotic picture	No	92 (93.9)	4 (4.1)	2 (2.0)	0.968	0.355	Reference
	Yes	1 (100.0)	0 (0.0)	0 (0.0)			4.7 (0.17-128.34)	0.355
Extramedullary lesion	No	92 (93.9)	4 (4.1)	2 (2.0)	0.968	0.355	Reference
	Yes	1 (100.0)	0 (0.0)	0 (0.0)			4.7 (0.17-128.34)	0.355
Admissions 3 times or less	No	10 (100.0)	0 (0.0)	0 (0.0)	0.733	0.813	Reference
	Yes	85 (93.4)	4 (4.4)	2 (2.2)			1.43 (0.07-27.3)	0.813
Admission frequency past year	.00	24 (88.9)	2 (7.4)	1 (3.7)	0.991	0.997	Reference
	1.00	38 (92.7)	2 (4.9)	1 (2.4)			0.63 (0.11-3.38)	0.592
	2.00	17 (100.0)	0 (0.0)	0 (0.0)			0.20 (0.01-4.1)	0.297
	3.00	6 (100.0)	0 (0.0)	0 (0.0)			0.54 (0.02-11.8)	0.694
	4.00	6 (100.0)	0 (0.0)	0 (0.0)			0.54 (0.02-11.8)	0.694
	5.00	1 (100.0)	0 (0.0)	0 (0.0)			2.3 (0.07-69.3)	0.624
	6.00	1 (100.0)	0 (0.0)	0 (0.0)			2.3 (0.07-69.3)	0.624
	9.00	2 (100.0)	0 (0.0)	0 (0.0)			1.40 (0.05-35.6)	0.838
Hydroxyurea (HU)	No	46 (92.0)	2 (4.0)	2 (4.0)	0.843	0.832	Reference
	500 mg OD	11 (100.0)	0 (0.0)	0 (0.0)			0.44 (0.02-8.95)	0.600
	500 mg BID	37 (94.9)	2 (5.1)	0 (0.0)			0.62 (0.11-3.6)	0.595
	500 mg TID	1 (100.0)	0 (0.0)	0 (0.0)			3.44 (0.12-97.6)	0.468
	Non compliance	3 (6)	0 (0.0)	0 (0.0)			-	-

Keynote: DM= Diabetic mellitus, HTN=Hypertension, CKD= chronic kidney disease.

*Odds ratios were added and calculated for all variables with 95% confidence intervals.

In terms of liver stiffness, the primary analysis included cases with non-missing CAP and kPa values for all 101 patients. Sensitivity analysis excluded cases with CRP > 10 mg/L or AST/ALT > 80 IU/L (n = 61) (Table 7). Liver stiffness values clustered around 4.5 kPa (corresponding to F0–F1), with a small number of elevated values. Figure 1 shows histogram of fibrosis score(kPa), and Figure 2 Boxplot of kPa by steatosis/imaging status.

Table 7.

Sensitivity analysis of confounding of stiffness excluding patients with CRP above 10 and AST/ALT above 2.

​	Analysis excluding patients with CRP above 10		Analysis excluding patients with AST/ALT above 2
	Standardized coefficients beta (B)	P-value	Standardized coefficients beta (B)	P-value
Gender	0.008	0.957	0.139	0.306
DM	0.025	0.856	-0.191	0.253
HTN	-0.160	0.283	0.390	0.047*
CKD	0.027	0.844	-0.042	0.766
Rheumatological	-0.109	0.440	-0.069	0.589
Hypothyroidism	-0.020	0.887	-0.014	0.914
Cardiac	0.054	0.693	-0.091	0.510
Blood transfusion	-0.633	0.530	0.039	0.782
Fatty changes	0.353	0.015*	0.229	0.088
Extramedullary lesion	-0.127	0.360	-0.111	0.391
Admission less than 3 X	-0.044	0.821	0.102	0.619
Admission frequency past year	-0.210	0.306	-0.042	0.853
HU Hydroxyurea	-0.130	0.416	-0.152	0.275

*Significant p-value less than or equal to 0.05.

Figure 1.

Histogram of fibrosis score(kPa).

Figure 2.

Boxplot of kPa by steatosis/imaging status.

The sensitivity analysis excluding patients with CRP > 10 and those with AST/ALT > 2 showed that most clinical variables did not demonstrate a significant association with stiffness, indicating that potential inflammation did not substantially confound these relationships. In the analysis excluding patients with elevated CRP, only fatty changes remained significantly associated with stiffness (β = 0.353, p = 0.015), while all other variables, including gender, diabetes, hypertension, chronic kidney disease, rheumatological disease, hypothyroidism, cardiac disease, history of blood transfusion, extramedullary lesions, admission frequency, and hydroxyurea use, were not significant. When excluding patients with AST/ALT ratio above 2, hypertension emerged as a significant predictor (β = 0.390, p = 0.047), with fatty changes showing a trend toward significance (p = 0.088), while all other variables again remained non-significant (Table 7).

Discussion

It is well known that the spectrum of liver disease in SCD is broad and multifactorial. Frequent hemolysis with intrahepatic red blood cell deposition, frequent blood transfusion with the side effect of iron overload, hepatotoxic or hemolytic effects of drugs, inflammatory process, gallstones with or without obstructive sequelae, and viral hepatitis are the most common causes of liver disease in SCD.^14,15 Additional data suggest a significantly higher incidence of liver fibrosis in patients with SCD due to iron overload, which acts as a contributing factor, both directly and indirectly, to both morbidity and mortality.^15,16 Patients with SCD without iron overload are also at risk of developing liver fibrosis, which was an essential factor in initiating this first-of-its-kind study in Saudi Arabia to investigate the role of Fibro Scan in the early detection of liver disease as well as liver fibrosis in adult patients with SCD.

The overall data reflect a small percentage of liver fibrosis in this cohort without laboratory evidence of iron overload. Thus, a fibrosis score of less than 7 kPa was found in 94.1% of cases (normal). In 6% of cases, the score was greater than 7 kPa, reflecting moderate to severe liver scarring (fibrosis). CAP was found to be S0 in 62.4%, S1 in 12.9%, and S2 in 24.8%. Liver stiffness was found to be F0 in 64.4%, F1 in 28.7%, F2 in 5.0%, and F3 in 2.0%.

Regarding radiographic examination, to better understand the correlation between liver disease manifestations and Fibro Scan results, fatty liver was detected in 24.8% of SLT cases with a significant association with fibrotic changes (p = 0.049). Although there is no precise national data on the prevalence of fatty liver disease in Saudi Arabia, an earlier study by Alswat et al. (2018) that estimated the prevalence of non-alcoholic fatty liver disease in the general Saudi population found that fatty liver disease was approximately 30%.^17,18 These data suggest the possibility of differences in the prevalence of non-alcoholic fatty liver disease between SCD and non-SCD cases. Albeit, to the authors’ knowledge, no studies have been found worldwide assessing the incidence of non-alcoholic fatty liver disease in SCD populations. ¹⁹

In this study, the majority of SCD patients had normal liver stiffness values, typically ranging from 4 to 6 kPa. Over 80% of patients had values between 2 and 7 kPa, which corresponds to minimal or no fibrosis (F0–F1), indicating functional changes rather than profound liver damage. Differences between CAP S0–S2 were minimal when comparing steatosis grades, suggesting that fat accumulation may be due to metabolic changes in adipose tissue. After excluding patients with elevated inflammatory markers (CRP >10 mg/L or AST/ALT >80 IU/L), mean liver stiffness values decreased slightly and became more uniform. This can confirm that inflammation is the cause. This test has several advantages, including the use of proven non-invasive methods (VCTE and CAP) and the inclusion of inflammatory factors, which increases the reliability of the results.

The mean age of patients with SCD in this study was 27.28 years (SD 10.98), which was younger than reported in previous studies. However, research on the epidemiology of liver fibrosis and age analysis is limited. To date, the available knowledge from three reports conducted in European countries showed that most cases of liver fibrosis with a cutoff value ≥ 8 kPa occurred in people over 45 years of age.^20–22 In Saudi Arabia, a study conducted in Jizan showed that liver fibrosis in the general population was higher in males and the mean age was 40-50 years. ¹⁷ However, occult liver fibrosis is one of the barriers that affect the accurate reporting of the prevalence of such disease. ²³

Similarly, the female-to-male ratio was 1.3:1, contrary to worldwide data, with a prevalence of males.^18,19 Moreover, European data have shown that the male gender is an independent risk factor for high liver stiffness and liver fibrosis in individuals without SCD. ²³ Whether gender is a risk factor for liver fibrosis in patients with SCD or not is unknown, as this topic has not been well studied previously. This raises the question of age and gender differences as a risk for liver fibrosis in patients with SCD.In this cohort, regression adjusted for age, sex, BMI, and chronic diseases yielded slightly significant findings. Moreover, the current study found slightly significant fatty liver disease in patients with SCD, with no significant association with other variables after adjusting for age, sex, BMI, and chronic diseases. Therefore, further studies involving a larger number of patients with SCD are needed to confirm the results obtained in the current cohort.

Non-alcoholic fatty liver disease (NAFLD) is the most common leading cause of liver fibrosis and, ultimately, liver failure. ²³ In Saudi Arabia, among the non-SCD population, obesity and diabetes were found to be common comorbidities associated with NAFLD. ²⁴ The results of the current study showed that hypertension was the most common comorbidity affecting SCD patients, followed by type 2 diabetes, heart disease, and chronic kidney disease. However, none of the comorbidities were significantly correlated with Fibro Scan results and liver scarring severity. Additional data in the literature have shown that metabolic syndrome and atherosclerotic risk factors are significant as the risk of liver fibrosis, elevated ALT and AST levels. ²³ This study showed that although the mean values of ALT (25.51) and AST (34.30) were within normal limits, the maximum values were 2-3 times higher than the normal level (109 and 107, respectively). This increases the risk of further liver injury and a possible increase in the incidence of liver fibrosis shortly if the causal predictor is not controlled. For example, Kamani et al. (2020) reported two brothers with acute viral illness and concomitant G6PD deficiency; both cases ended in fulminant liver failure. ²⁵ In addition, a Chinese study in reproductive-age women with low G6PD levels found an association with HBV liver disease. ²⁶ Given the previous evidence and the pattern of hemolysis that occurs with G6PD deficiency, the results of the current study suggest that the risk of liver fibrosis may be higher in SCD patients with G6PD deficiency than in SCD patients without G6PD deficiency. However, this is a hypothesis that requires further investigation.

Iron overload in patients with SCD, which is a result of multiple blood transfusions, is usually detected by high ferritin levels in addition to liver MRI and/or liver biopsy. As an acute-phase reactant, ferritin levels can increase due to liver disease or inflammation. Therefore, patients with SCD should be free of active crises, infections, and inflammatory processes before iron overload is assessed. Iron overload is defined as a serum ferritin level greater than 1000 ng/mL. However, levels greater than 2500 ng/mL are associated with an increased risk of organ damage, particularly the heart and liver. Additionally, a liver T2-MR value >6.3 ms indicates normal or absent iron deposition, a T2 value between 2.0 and 6.3 ms indicates mild iron overload, and a T2 value <2.0 ms indicates moderate or severe iron overload. Values <1.4 ms indicate significant iron deposition in the liver. An alternative measurement method, Ferriscan (R2-MRI), measures liver iron concentration (LIC) in milligrams of iron per gram of dry weight.

An LIC value of less than 1.8 mg/g is considered normal, a value of 1.8 to 7 mg/g indicates mild iron overload, and a value greater than 7 mg/g indicates moderate or severe iron overload. Iron loading in patients with SCD is more accurately reflected by combining serum ferritin levels with quantitative liver iron determination using MRI. Based on laboratory and radiological findings in this study, cases of iron overload were excluded early in the recruitment process before further data were collected, as they were beyond the scope of the current study. ²⁷ In a study of iron overload in children with SCD, Adamkiewicz et al. (2009) found that the increase in serum ferritin (SF) was nonlinear compared with liver iron concentration (LIC) or pre-chelation predicted transfusion iron loading (TIL). ²⁸ On the other hand, Badawy et al. (2016) conducted a study to evaluate the correlation between LIC determined by liver biopsy and MRI and long-term average serum ferritin and spot ferritin levels. On the other hand, Badawy et al. (2016) conducted a study to evaluate the correlation between LIC determined by liver biopsy and MRI and long-term average serum ferritin and spot ferritin levels. The authors found a strong correlation between the LIC measured by P2-MRI and long-term mean serum and spot ferritin values. Furthermore, a clear positive association was found between the LIC measured by liver biopsy and the LIC measured by P2-MRI. ²⁹

However, there is currently no need for multiple liver biopsies after Fibro Scan implantation with high diagnostic ability for liver disease. The results of the current study showed that 65.3% had a history of blood transfusion. However, this did not correlate significantly with Fibro Scan findings of prior tissue scarring, which is contrary to existing evidence.^30,31 A later study by Manganas et al. (2023) reported predictors of liver fibrosis in SCD and concluded that the frequency of crises, ongoing hemolysis, and frequent blood transfusion was significantly associated with liver fibrosis. ³⁰ On the other hand, minimizing blood transfusion using exchange transfusion instead of supplemental transfusion in addition to the use of chelators has a great impact on iron overload and liver health.^30,32 The reversibility of liver fibrosis due to iron deposition after treatment of iron overload is another controversial issue. Njoku et al. (2023) in their study highlighted the difficulty of answering this question in the presence of an ongoing crisis and the need for blood transfusions due to the nature of the disease, even if the patient continued measurements and medications to lower iron levels. ³³ However, based on the study by Delicou et al. (2018), in which Fibro Scan results were correlated with serum ferritin and LIC in patients with sickle cell disease, they were found to be an effective method for assessing the effectiveness of liver iron chelating agents. ³⁴ However, according to a study by Castéra et al. (2010), fibro scanning may be an unreliable examination in one in five patients due to obesity, increased waist circumference, metabolic syndrome, and lack of operator experience. ³⁵

This study showed that 49.5% of cases did not use hydroxyurea, either due to previous allergy, gastrointestinal disorder, patient’s refusal of medication, or uneventful course of the disease, when hydroxyurea is not indicated. On the other hand, 50.5% of patients took hydroxyurea in different doses depending on the patient’s weight, hematological markers (CBC), and patient’s tolerance. However, there was no significant correlation with liver fibrosis in either group (p=0.843). Although hydroxyurea has been shown to be effective against liver fibrosis by inhibiting stellate cell proliferation and liver fibrogenesis without relieving hepatocyte necrosis or blocking liver regeneration, there are limited data on the actual effects of hydroxyurea on the liver, as well as on some other organs such as the retina.^35,36

Another issue that needs to be addressed in the discussion is the relevance of SCD phenotyping and the percentage of HGB S (HGB S%) in SCD with liver disease. The results of the current study showed that, despite a predominance of HGB SS patients over HGB SB patients, there was no statistically significant difference in the frequency of liver scarring between the two groups (p=0.969). Unfortunately, no other phenotypes were identified that could strengthen this association with Fibro Scan results. In their study, Banerjee et al.(2025) provided a detailed review of the impact of sickle cell hepatopathy, hepatic sequestration, and iron excess on the liver. ³⁷ The authors state that in patients with HGB SS, blood iron levels typically exceed 50%, leading to sickle cell anemia in the hepatic sinusoids and the development of liver problems. ³⁸ Iron overload and liver crisis are among the acute and chronic symptoms of liver disease that were examined by Suddle et al. (2019), who emphasized that liver disease in SCD is complex and does not depend solely on the HGB S%. ³⁹ Furthermore, Kaminski et al. (2023) investigated how poor hemoglobin clearance in the liver sinusoidal endothelium leads to vaso-occlusion and liver damage, thus indirectly linking HGB S load to liver damage. ⁴⁰ Finally, an observational study by Bortolotti et al. (2020) examining liver damage in different SCD genotypes (HGB SS, HGB SC, and HGB S/B-Thal) showed that genotype, and therefore HGB S%, alters the pattern of liver damage, particularly cholestatic damage. ⁴¹ However, further larger-scale research may shed light on the true association between HGB S% and liver health.

Alas, studies have shown a high mortality rate in SCDs due to hepatobiliary changes, with no clear guidelines or methods for monitoring patients with SCD for liver or hepatobiliary alterations. ⁴² Although there were no reports of liver cancer in this study, patients with a positive F4 Fibro Scan indicating cirrhosis would be considered at risk for additional liver changes, including cancer. To address this issue, a follow-up plan was developed with laboratory (alpha-fetoprotein) and radiological (liver ultrasound) monitoring every six months, and annual repeat Fibro Scans for follow-up. However, there are currently no specific recommendations for SCD screening for any type of cancer. In terms of cost and effectiveness, a standard liver ultrasound costs between $100 and $250 per patient, depending on the region. By comparison, a Fibro Scan in private clinics costs between $200 and $350 per test.^43,44 Although ultrasound cannot assess the degree of fibrosis, it remains the method of choice for general liver imaging. To reduce unnecessary biopsies and referrals, fibro scanning is more cost-effective in certain patient groups (e.g., those with elevated liver enzymes).

In summary, although this study did not find an increased incidence of liver fibrosis among patients with SCD who do not have iron overload, the role of fibro scanning and its value in detecting fibrosis should be balanced against clinical and radiographic indications and not be used routinely without good justification. In addition, there is no data on the prevalence of liver disease in SCD, with broad knowledge of NAFLD in SCD, the reversibility of liver injury after treatment of iron overload, how G6PD-deficient SCD behaves differently than non-G6PD-deficient SCD in liver disease, and what is the exact effect of hydroxyurea on liver disease. Further national studies are needed to have a potential answer to the proposed hypotheses.

Limitations

In addition to its unique contribution to SCD research in Saudi Arabia and global health, this study has some limitations that should be considered in future studies. First, this was a single-center study conducted in a single military hospital with limited eligibility criteria. Future studies are expected to be multicenter to provide a more holistic picture. Secondly, Fibro Scan is procedure dependent, and the result may be altered if it is performed or interpreted by different personnel. A third limitation is that cases known to have iron overload were excluded, which impacts the true incidence of liver fibrosis in the Saudi SCD population. Another point is that, due to data collection limitations, no further association was established between Fibro Scan results and the type of blood transfusion (exchange or simple). Finally, this study uses the METAVIR system cutoff values to determine the overall liver stiffness stage. However, in patients with SCD, cutoff values may be higher than in patients without SCD due to cholestasis, liver congestion, hemolysis, and inflammation, even in the absence of histological fibrosis. To fully understand the risk of liver fibrosis in SCD patients, future studies are recommended to include patients with and without iron overload.

Conclusion

The results of the current study suggest that patients with SCD without iron overload had a small risk of liver fibrosis, despite the frequency of hospitalizations, medication use, and/or presence of other complications of SCD. Fatty liver in SCD is likely at risk for fibrosis progression, which is an area for further study and a larger national study. The transient elastography (Fibro Scan) as part of a liver complication investigation in patients with SCD is a case-specific rather than a general standard of care.

Although the results of the current study suggest that SCD patients without iron overload have a low prevalence of liver fibrosis, this observation may reflect a selection effect. Specifically, individuals without iron overload are less likely to develop iron-mediated liver disease to begin with, which may confound this association. Thus, caution should be exercised in interpreting this as a causal relationship, and further studies are needed to disentangle the contribution of iron load and other fibrosis risk factors in SCD. Finally, these data do not support the need for routine universal screening in stable outpatients without known iron overload; targeted use in combination with clinical and biochemical assessment merits study in larger multicenter cohorts.

ORCID iD

Rehab Y. Al-Ansari https://orcid.org/0000-0003-4331-5769

Ethical considerations

Approval for the research was obtained from the Armed Force Hospitals Eastern Region Institutional Review Board (IRB), protocol No. AFHER-IRB-2023-029. IRB Review Date August 20, 2023.

Consent for publication

Informed consent was acquired for Fibro Scan study and publication of this article from the patients.

Data Availability Statement

All data already included in the manuscript.*