Fibrosis 4 Index Predicts Haemocompatibility-Related Adverse Events in Patients Using Left Ventricular Assist Devices

Abstract

Background

This study aimed to explore the association between the liver fibrosis-4 (FIB4) index and clinical events, specifically haemocompatibility-related adverse events (HRAEs), in patients who underwent left ventricular assist device (LVAD) implantation.

Methods

The study involved patients who received LVADs for advanced heart failure (HF) treatment. FIB4 index was calculated both preoperatively and postoperatively, and its association with clinical events was examined, focusing on all-cause mortality and HRAEs, which included severe bleeding, cerebrovascular accidents, and intra-pump thrombosis.

Results

A total of 224 patients with LVAD were analyzed. The value of FIB4 index was significantly decreased after LVAD implantation, and the postoperative FIB4 index was independently associated with composite events, including all-cause death and HRAEs after adjustment of cofounders. Next, we clarified the association between postoperative FIB4 index and clinical events after LVAD implantation. Patients were divided into 2 groups based on a postoperative FIB4 index threshold of 1.45. The high FIB4 group was associated an elevated risk of HRAEs (P = 0.0038), with no significant difference in survival curves between the groups. Additionally, the high FIB4 group showed elevated von Willebrand factor activity, while no significant differences were found in prothrombin time international normalized ratio variability.

Conclusions

FIB4 index, a liver fibrosis marker, serves as a valuable prognostic indicator for HRAEs in patients receiving LVAD support.

The social impact of heart failure (HF) has significantly increased in recent years.

Graphical Abstract

INTRODUCTION

The social impact of heart failure (HF) has significantly increased in recent years. Despite the development of various effective interventions and treatments, there were several medically refractory HF cases.¹ Mechanical circulatory support, such as left ventricular assist devices (LVADs), and heart transplantation are available options for these refractory cases.

Recent studies have reported a growing connection between nonalcoholic fatty liver disease (NAFLD) and cardiovascular conditions.² However, there is limited research on the interaction between the heart and liver in the context of advanced HF.

The heart-liver interaction is complex in patients receiving LVAD treatment for advanced HF. LVAD support can alleviate the burden on the left hear system, but it can be challenging to support the right heart.³ According to the heart-liver interaction, LVAD support is expected to restore antegrade blood flow to the liver, whereas the increased load on the right heart may hinder blood flow from the liver to the right atrium, potentially exacerbating liver congestion.

Among the various methods for evaluating liver fibrosis, the fibrosis-4 (FIB4) index is a widely used marker of liver damage. FIB4 index is calculated using aspartate aminotransferase (AST), alanine aminotransferase (ALT), platelet count, and age. Initially developed to predict liver damage related to hepatitis viruses,⁴ it has also been shown to effectively predict outcomes in a range of conditions beyond liver disease.⁵

Liver damage linked to hepatic congestion might lead to pathological changes such as liver fibrosis in patients with advanced HF. Gelow et al⁶ performed liver tissue examinations in advanced HF patients requiring LVAD or heart transplantation, finding that 80% had liver fibrosis, with 47% exhibiting severe grade 3-4 liver fibrosis. However, there remains limited understanding of how liver damage impacts clinical outcomes in these patients.

In this study, we utilized FIB4 index, a marker of liver fibrosis, both preoperatively and postoperatively, to investigate the relationship between clinical events, particularly haemocompatibility-related adverse events (HRAEs), and liver damage in patients undergoing LVAD implantation.

METHODS

Study patients

The study included consecutive patients who underwent LVAD implantation for advanced HF at our hospital between October 2010 and December 2023. This group consisted of patients either awaiting heart transplantation or receiving destination therapy. The antithrombotic regimen in all patients consisted of low-dose aspirin or other thrombotic agents in addition to warfarin.

The LVADs utilized were HeartMate 3 (Abbott, Chicago, IL, USA), HeartMate II (Thoratec Corporation, Pleasanton, USA), EVAHEART (Sun Medical Company, Moriyama, Tokyo), Jarvik 2000 (Century Medical, Inc., Tokyo, Japan), DuraHeart (Terumo, Tokyo, Japan), and HVAD (Medtronic PLC, Dublin, Ireland). This study followed the principles of the Declaration of Helsinki and received approval from the Ethics Committee of the Faculty of Medicine at the University of Tokyo (approval number: [2650] November 20, 2009). The requirement for informed consent was waived due to the retrospective nature of the analysis. Data up to December 2024 were used.

Clinical parameters

Clinical information was collected from the medical records. von Willebrand factor (vWF) activity was assessed at the outpatient clinic after the patient was discharged from the hospital with the LVAD, using ristocetin cofactor activity as the measurement value. Data on prothrombin time-international normalized ratio (INR) values from 4 outpatient visits and the intervals between these visits were gathered after LVAD implantation discharge to calculate the mean INR, standard deviation, and INR variability. The data of blood test were obtained at designated times [the time of right heart catheterization (RHC) prior to LVAD operation, the time of RHC after LVAD implantation (about 1 month after operation), 3 months, 6 months, 1 year, 18 months, and 2 years after operation] for each case.

INR variability was determined using Fihn’s method to calculate the time-weighted INR variance.⁷ The difference between 2 INR measurements was squared and divided by the number of days, and the average of these values for each time interval was taken before calculating the square root:

$$\sigma =\sqrt{\frac{1}{n-1}\sum _{i=2}^n\frac{{\left(IN{R}_i-IN{R}_{i-1}\right)}^2}{\tau i}}, \tau ={\mathrm{t}}_1-{\mathrm{t}}_{\mathrm{i}-1}$$

The FIB4 index was calculated using the following formula⁸:

$$\mathbf{FIB}\mathbf{4} \mathbf{index}=\left[\mathbf{age} \left(\mathbf{years}\right)\times \mathbf{AST} \left(\mathbf{IU}/\mathbf{L}\right)\right]/\left[\mathbf{platelet} \mathbf{count} \left(\mathbf{1}{\mathbf{0}}^{\mathbf{9}}/\mathbf{L}\right)\times \mathbf{ALT} \left(\mathbf{IU}/\mathbf{L}\right)\right]$$

RHC was performed to evaluate HF status before LVAD implantation and to optimize LVAD rotation speed 1 month post-surgery. The data of cardiac ultrasound examinations were collected from tests conducted prior to LVAD implantation.

Clinical outcome

Data on all-cause mortality served as the primary event. Additionally, information on clinical events related to LVAD complications associated with haemocompatibility issues was gathered. The study follow-up was completed at December 2024. HRAEs included the following^9–11:

• Haemorrhage: Gastrointestinal or other nonsurgical bleeding episodes requiring hospitalization >30 days after LVAD surgery

• Cerebrovascular disorder: The emergence of new neurological symptoms, with confirmation of new cerebral haemorrhage or infarction lesions via computed tomography scan

• In-pump thrombosis: Replacement of the pump due to thrombosis, withdrawal of the LVAD because of thrombosis, or death resulting from in-pump thrombosis.

Statistical analysis

Data are presented as the median [interquartile range]. For continuous variables, Student’s t-test or Mann-Whitney U-test was employed, while Fisher’s exact test was used for categorical variables after patients were divided into 2 groups by the value of FIB4 index parameters. We performed pre-post comparison of FIB4 index using paired-sample test. We also conducted a linear mixed model analysis to examine the longitudinal changes in the FIB-4 index and platelet count over time. The Bonferroni method assessed the significance of multiple comparisons. Statistical analyses were conducted using 2-sided tests, with a significance level set at 5%, and P-values and confidence intervals (CIs) were calculated. To compare the predictive value of preoperative and postoperative FIB4 index, the association between the preoperative or postoperative FIB4 index and composite events including HRAEs and all-cause death was investigated by multivariable Cox proportional hazard models with its logarithmic transformation after adjustment of confounders including age, gender, device, estimated glomerular filtration rate (eGFR), and B-type natriuretic peptide (BNP). In the comparison, the postoperative FIB4 index was calculated by the data of the time of RHC after LVAD implantation (about 1 month after operation). After stratifying patient groups based on FIB4 index, the time-to-death curve was estimated using the Kaplan-Meier method with the day of LVAD implantation designated as “zero.” Since all-cause mortality serves as a competing risk to other clinical events, Gray’s test was utilized to estimate the event survival curves for HRAEs. Additionally, the Fine-Gray method was employed to calculate the subdistribution HR for the cumulative incidence rate of HRAEs. The subdistribution HR was adjusted by age, gender, and device. Spearman’s rank correlation analysis was conducted to calculate the correlation coefficient between postoperative FIB4 index and vWF activity. Statistical analyses were performed using JMP software (version14.2; SAS Institute, Cary, NC, USA), and EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan).¹²

RESULTS

Study patients

A total of 230 patients who underwent LVAD implantation between December 2010 and December 2023 were analyzed, in which 214 underwent implantation as a bridge to transplant, while 16 were for destination therapy. Four cases were excluded due to difficulty in collecting most data, and 2 cases were excluded because they were believed to have experienced temporary severe liver damage prior to surgery. Consequently, 224 cases were included in the analysis (HeartMate 3; 40, HeartMate II; 66, EVAHEART; 44, Jarvik 2000; 32, DuraHeart; 18, and HVAD; 24). Among these, 102 patients received a transplant during the observation period, while 123 either continued with the LVAD or died. The follow-up period for patients who underwent transplantation was 1466 [979, 1836] days, whereas the follow-up period for the remaining 123 cases was 983 [449, 1474] days. During the observation period, 31 deaths occurred. Follow-up for transplanted patients concluded on the day of transplantation, and 15 patients had their LVAD removed during the follow-up due to recovery of cardiac function; follow-up for these patients was censored on the removal date.

Change in FIB4 index after LVAD implantation

Initially, we examined the distribution of FIB4 index before and after LVAD implantation within this cohort. Overall, there was a noted decrease in FIB4 index after LVAD implantation (P = 0.0001) (Figure 1A).

Figure 1.

(A) Histograms of FIB4 before and after LVAD implantation. (B) Changes in FIB4 over time before and after LVAD implantation. FIB4 (median + standard deviation) at each time point is plotted, with points showing significant differences (P <0.0083 by Bonferroni correction) compared to preoperative values marked with an *. Abbreviations: eGFR: estimated glomerular filtration rate; FIB4: liver fibrosis-4 index; LVAD: left ventricular assist device; Post: postoperative; Pre: preoperative; 3m: 3 months; 6m: 6 months; 1y: 1 year; 18m: 18 months; 2y: 2 years

Figure 1B shows the longitudinal changes in the FIB-4 index. Significant differences were observed 1 or 3 months after LVAD implantation, while the FIB-4 index was not significantly different with comparisons made to the presurgery values. FIB4 index showed a temporary improvement after surgery but returned to presurgery levels by 6 months.

Predictive value of composite events including death and HRAEs

We then evaluated the predictive value of preoperative and postoperative FIB4 index for composite events, including death and HRAEs (Table 1). The mean follow-up period was 1216 days (731-1693 days). Among the 224 patients with follow-up data, 94 experienced clinical events (death or HRAEs). Multivariable Cox proportional hazard analyses for composite events with its logarithmic transformation before adjustment, and after adjustment of age and sex (Model 1), or after adjustment of age, sex, and device (Model 2), or after adjustment of age, sex, device, eGFR, and BNP (Model 3) showed that the postoperative FIB4 index was an independent predictor for composite events. Therefore, postoperative FIB4 index, which was calculated by the data of the time of RHC after LVAD implantation, was used in further analyses.

Table 1.

Multivariable Cox Proportional Hazard Analyses for Events, Including Death and HRAEs

	Log preoperative FIB4 index	Log postoperative FIB4 index
Monovariable	2.00 [1.11-3.59] (P = 0.021)*	3.19 [1.38-7.25] (P = 0.0063)*
Model 1	1.80 [0.92-3.50] (P = 0.086)	3.65 [1.30-10.23] (P = 0.014)*
Model 2	1.79 [0.92-3.50] (P = 0.084)	3.64 [1.22-10.87] (P = 0.020)*
Model 3	1.87 [0.95-3.69] (P = 0.071)	3.88 [1.30-11.61] (P = 0.015)*

^* P < 0.05.

Model 1; age and gender.

Model 2; age, gender, and device.

Model 3; age, gender, device, eGFR, and BNP.

Abbreviations: BNP: B-type natriuretic peptide; eGFR: estimated glomerular filtration rate; FIB4: liver fibrosis-4 index; HAREs: haemocompatibility-related adverse events.

Postoperative FIB4 index and clinical course following LVAD implantation

Next, patients were categorized into 2 groups based on a postoperative FIB4 index threshold of 1.45, as defined in previous study.⁸

The groups differed significantly in age and platelet count, which is expected as both factors are components of the FIB4 index calculation. RHC data indicated that the high FIB4 group had notably higher preoperative right atrial pressure (RAP) (Table 2). Except for Jarvik 2000, there was no significant difference in LVAD models between the groups. Postoperatively, the high FIB4 group displayed significantly elevated pulmonary artery wedge pressure (PAWP).

Table 2.

Baseline Characteristics of High and Low FIB4 Index Groups

	FIB4 < 1.45	FIB4≧1.45	P-value
N	176	48
Age	40 [30, 50]	53 [41, 60]	<0.0001*
Gender (male/female)	133/43	29/19	0.052
Height (m)	1.69 [1.61, 1.74]	1.64 [1.58, 1.71]	0.0025*
Body weight (kg)	56.9 [50.0, 65.0]	55.4 [46.4, 62.2]	0.19
BMI (kg/m2)	20.0 [18.2, 22.1]	19.7 [18.2, 22.4]	0.88
mAP (mmHg)	68 [61, 74]	70 [65, 75]	0.16
HR (beats/min)	80 [70, 98]	80 [70, 91]	0.61
INTERMACS
Profile 2/3/4	78/71/27	22/20/6	0.25
LVDd (mm)	72 [63, 78]	69 [58, 81]	0.42
LVEF (%)	17 [12, 24]	18 [13, 26]	0.44
MR(moderate or severe)	88 (50.0)	20 (41.7)	0.30
TR (moderate or severe)	47 (26.7)	20 (41.7)	0.050
Aetiology (ischaemic/non-ischaemic)	11/165	10/38	0.0048*
Drugs
ACE/ARB	154 (88.0)	45 (93.8)	0.31
Beta blocker	133 (76.0)	34 (70.8)	0.46
MRA	143 (81.7)	44 (91.7)	0.12
Types of LVAD
Dura/EVA/HM2/	14/38/53	4/6/13
HM3/Jarvik/HVAD	35/16/19	5/16/4
Blood test
Alb (g/dl)	3.7 [3.2, 4.1]	3.4 [3.0, 3.8]	0.035*
AST (IU/l)	25 [19, 35]	26 [20, 41]	0.38
ALT (IU/l)	24 [15, 36]	18 [12, 32]	0.063
T.B (mg/dl)	1.1 [0.7, 1.8]	1.1 [0.7, 1.7]	0.75
eGFR (mL/min/1.73m2)	66 [51, 87]	53 [35, 80]	0.0098*
Hb (g/dl)	11.6 [10.3, 13.1]	11.0 [9.7, 12.0]	0.029*
Plt (×104/μl)	19.7 [15.7, 25.2]	16.4 [11.4, 21.5]	0.0018*
BNP (pg/mL)	588 [322, 952]	523 [298, 1080]	0.84
HbA1c (%)	5.9 [5.5, 6.4]	6.0 [5.7, 6.5]	0.29
Preoperative FIB4 index	1.04 [0.65, 1.67]	2.06 [1.40, 2.95]	<0.0001
Postoperative FIB4 index	0.74 [0.48, 1.00]	1.92 [1.68, 2.62]	<0.0001
Preoperative RHC
mRAP (mmHg)	8 [5, 12]	11 [7, 14]	0.010*
mPAP (mmHg)	29 [19, 36]	29 [21, 37]	0.80
mPAWP (mmHg)	22 [13, 28]	21 [12, 27]	0.91
Cardiac index (L/min/m2)	1.9 [1.6, 2.3]	2.0 [1.7, 2.4]	0.36
Postoperative RHC
mRAP (mmHg)	7 [5, 10]	8.5 [5.8, 12.2]	0.083
mPAP (mmHg)	16 [13, 20]	18 [14, 21]	0.18
mPAWP (mmHg)	8 [3, 11]	10 [7.5, 14]	0.025*
Cardiac index (L/min/m2)	2.5 [2.2, 2.8]	2.3 [2.0, 2.7]	0.14
mAP (mmHg)	78 [70, 82]	76 [72, 87]	0.95
HR (beats/min)	80 [72, 90]	85 [80, 90]	0.078

^* P < 0.05.

Abbreviations: ACE: angiotensin converting enzyme inhibitor; Alb: albumin; ALT: alanine aminotransferase; ARB: angiotensin II receptor blocker; AST: aspartate aminotransferase; BMI: body mass index; BNP: B-type natriuretic peptide; Dura: Duraheart; eGFR: estimated glomerular filtration rate; EVA: EVAHEART; Hb: haemoglobin; HM2: Heartmate II; HM3: Heartmate 3; HR: heart rate; INTERMACS: Interagency Registry for Mechanically Assisted Circulatory Support; LVAD: left ventricular assist device; LVDd: left ventricular diastolic diameter; LVEF: left ventricular ejection fraction; mAP: mean arterial pressure; mPAP: mean pulmonary artery pressure; mPAWP: mean pulmonary artery wedge pressure; MRA: mineralocorticoid receptor antagonist; mRAP: mean right atrial pressure; Plt: platelet; RHC: right heart catheterization; T.B: total bilirubin; TR: tricuspid regurgitation.

The survival curves for both groups were then compared (Figure 2A), revealing no significant difference in survival rate, while there was an elevated risk of HRAEs in the FIB4 index ≥1.45 group (P = 0.0038) (Figure 2B).

Figure 2.

(A) Survival rates divided into 2 groups based on FIB4 = 1.45. Kaplan-Meier survival curve for the 2 groups (FIB4 ≥1.45 vs FIB4 <1.45), with time zero set at LVAD implantation. (B) Cumulative incidence curves of HRAEs calculated using the Fine-Gray method when divided into 2 groups based on FIB4 = 1.45. Abbreviations: FIB4: liver fibrosis-4 index; HRAEs: haemocompatibility-related adverse events; LVAD: left ventricular assist device

Sensitivity analysis

Subdistribution HR of the high FIB4 group for HRAEs was calculated using various thresholds for dividing the 2 groups by FIB4 index (Table S1), adjusting for age, gender, and device. With the FIB4 index quartile value at 1.30, thresholds were set by segmenting values from 1.3 to 1.6 into 5 groups. In each case, the high FIB4 group exhibited a significantly higher event rate for HRAEs, confirming FIB4’s utility in predicting HRAEs.

Association between FIB4 index and the effects of warfarin, platelet count, and vWF

In examining association between FIB4 index and HRAEs, we assessed potential effects on warfarin pharmacokinetics. Warfarin-related metrics did not significantly differ between high and low FIB4 groups (Table S2, Figure 3A) in the early postoperative phase.

Figure 3.

(A) Comparison of INR variability between high and low FIB4 groups. (B) Change in platelet count before and after LVAD implantation between high and low FIB4 groups. Platelet count (median + standard deviation) at each time point is plotted, with points showing significant differences (P <0.0083 by Bonferroni correction) compared to preoperative values marked with an *. (C) Comparison of vWF activity between high and low FIB4 groups. (D) Scatter plot of vWF activity against postoperative FIB4, FIB4 plotted on a logarithmic scale. Abbreviations: FIB4: liver fibrosis-4 index; INR: international normalized ratio; LVAD: left ventricular assist device; Post: postoperative; Pre: preoperative; vWF: von Willebrand factor; 3m: 3 months; 6m: 6 months; 1y: 1 year; 18m: 18 months; 2y: 2 years

Over time, platelet counts initially increased following LVAD surgery but showed a muted response in the high FIB4 group (Figure 3B). Furthermore, dividing groups at a threshold of FIB4 index =1.45 revealed significantly elevated vWF in the high FIB4 group (Figure 3C). Indeed, postoperative FIB4 index was finely correlated with vWF activity (Figure 3D).

DISCUSSION

This study examined the changes in FIB4 index, a marker of liver fibrosis, and its relationship to clinical outcomes in advanced HF patients with LVAD implants. Findings revealed that postoperative FIB4 index were closely related to the risk of HRAEs. FIB4 index showed no association with warfarin efficacy but was associated with vWF activity, suggesting that HRAEs after LVAD implantation are associated with liver damage, as indicated by liver fibrosis markers, likely in relation to vWF activity.

FIB4 index and HRAEs

Several studies have examined the association between FIB4 index and clinical outcomes in HF patients.^13–15 This study demonstrated a strong correlation between FIB4 index and HRAEs. Given the limited research on predicting HRAE complications, this paper contributes significantly by identifying a relationship between HRAEs and liver fibrosis markers.

In reviewing thrombotic disease literature, some papers indicated a relationship between liver fibrosis markers, such as FIB4 index, and the functional prognosis in patients with cerebral infarction.^16,17 Additionally, studies have highlighted a connection between FIB4 index and increased bleeding in cases of intracerebral haemorrhage.^18,19 A similar mechanism may connect FIB4 index with clinical events in LVAD- implanted patients.

FIB4 index, warfarin response, and coagulation status

In examining drug metabolism, we investigated the relationship between the effects of warfarin and FIB4 index. Several studies have previously linked INR variability with clinical events.²⁰ Additionally, studies indicate that INR variability calculated via Fihn’s method is associated with increased risks of both embolic cerebral infarction and bleeding.²¹ However, in this study, no clear association was found between warfarin variability and clinical events, nor was there a significant link between FIB4 index s and warfarin fluctuation.

We then compared the platelet counts between the 2 groups, revealing a consistently low platelet count in the high FIB4 group over the long-term following LVAD implantation. Several studies have reported an association between platelet count and prognosis in LVAD-implanted patients.^22,23 These studies indicated that a high platelet count is linked to worse outcomes, while Luo et al²⁴ recently identified a U-shaped relationship between platelet count and prognosis in this population. The lower platelet count observed in the high FIB4 group may contribute to the poor prognosis in this cohort, although the exact mechanism has not yet been clarified.

We also examined vWF activity to investigate its relationship with HRAEs. vWF deficiency can result from mechanical damage caused by the LVAD.²⁵ However, in cases of acquired vWF syndrome due to mechanical circulatory support, quantification of vWF or a decrease in its activity is rarely detected; analyzing high-molecular-weight multimers is deemed necessary. Indeed, the average vWF activity value was 84.1%, with 70% of cases falling within the normal range of 50-150% in this cohort. Novel point of this study was a correlation between FIB4 index and vWF. Several studies have linked elevated vWF activity to a higher risk of thrombotic diseases, such as cerebral infarction.^26,27 Regarding LVAD, Jahangiri et al²⁸ recently reported that patients with low vWF activity after HeartMate 3 implantation had a higher incidence of haemorrhagic events, while those with high vWF activity experienced more cerebral infarctions. In our study, vWF activity was significantly higher in patients with cerebrovascular events [vWF activity, with cerebrovascular events, 100 (59 127) vs without cerebrovascular events, 75 (46 103) (P = 0.018)]. Therefore, the association between elevated vWF and increased risk of HRAEs in the high FIB4 group may be plausible. Several studies have showed an association between elevated vWF activity and both liver disease and its severity.²⁹ The association between FIB4 index and HRAEs may be linked to increased vWF activity resulting from liver injury.

Limitations

This study was a retrospective analysis conducted at a single institution, and the sample size was relatively small. It should be conducted in the future while collecting data from multiple facilities. Additionally, the study included various types of LVADs, each with distinct characteristics; therefore, the presentation of HRAEs was different in each device (Table S3). Future studies should focus on a single device for consistency. Notably, differences in postoperative FIB4 levels were found between the Jarvik 2000 and other devices [postoperative FIB4 index, Jarvik 2000 vs others, 1.39 (0.89-1.84) vs 0.88 (0.53-1.14) (P = 0.0003)]. However, the subdistribution HR for HRAEs with Jarvik 2000 was 1.13 (CI 0.67-1.91) (P = 0.65), indicating that this device did not significantly affect clinical events. Although patients with evident liver disease were excluded from the study, some may not have been adequately filtered out. Furthermore, liver fibrosis was evaluated using only FIB4 index, so incorporating other evaluation methods would be beneficial. In this study, we used the FIB4 index as a marker for liver damage, but there are other markers for liver damage, such as model for end-stage liver disease (MELD), and further study is required to determine which marker is most suitable as a marker for liver damage in patients under LVAD support. In terms of metabolic dysfunction-associated steatotic liver disease (MASLD), the pathologic process includes the fibrogenesis and stiffening of the liver in advanced disease. FIB4 index is the recommended first-line screening test for the pathological process.³⁰ Recently Fontan-associated liver disease (FALD), one striking example of liver damage caused by right HF, had been reported to have some common features with MASLD.³¹ From these results, the FIB4 index is expected to be useful in the setting of post-LVAD surgery, a condition of right HF somewhat similar to FALD. While extracardiac organ damage such as renal dysfunction and pulmonary vascular resistance after LVAD has traditionally received attention, liver damage has not received much attention. By creating a new marker that indicates changes in condition after LVAD, it will be possible to more accurately understand the complications under LVAD support, which might result that the treatment targets that have not been known could be identified. There were some differences in the value of vWF between each device. However, the relationship between vWF and postoperative FIB4 index was independent from the kinds of devices because regression analysis showed that FIB4 index associated with vWF [t value; 3.10 (P = 0.0023)] after the adjustment of device. The role of age as a cofactor of FIB4 index is a source of bias because age is included in the calculation of FIB4 index. However, age is closely related to the risk of events; therefore, we also included age as a cofactor in multivariate analysis in accordance with previous reports.^30,32

CONCLUSION

FIB4 index, as a liver fibrosis marker, shows a strong prognostic value for HRAEs in patients receiving LVAD support.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

FUNDING

None declared.

CONFLICTS OF INTEREST

The other authors have no conflicts of interest to disclose. There are no patents, products in development, or marketed products to declare.

DATA AVAILABILITY

Data available on request.