Evaluation of a micro-spectrometer for the real-time assessment of liver graft with mild-to-moderate macrosteatosis: A proof of concept study

Evaluation of Minimum-to-Severe Global and Macrovesicular Steatosis in Human Liver Specimens: A Portable Ambient Light-Compatible Spectroscopic Probe

BACKGROUND AND AIMS

Hepatic steatosis (HS), particularly macrovesicular steatosis (MaS), influences transplant outcomes. Accurate assessment of MaS is crucial for graft selection. While traditional assessment methods have limitations, non-invasive spectroscopic techniques like Raman and reflectance spectroscopy offer promise. This study aimed to evaluate the efficacy of a portable ambient light-compatible spectroscopic system in assessing global HS and MaS in human liver specimens.

METHODS

A two-stage approach was employed on thawed snap-frozen human liver specimens under ambient room light: biochemical validation involving a comparison of fat content from Raman and reflectance intensities with triglyceride (TG) quantifications and histopathological validation, contrasting Raman-derived fat content with evaluations by an expert pathologist and a "Positive Pixel Count" algorithm. Raman and reflectance intensities were combined to discern significant (≥ 10%) discrepancies in global HS and MaS.

RESULTS

The initial set of 16 specimens showed a positive correlation between Raman and reflectance-derived fat content and TG quantifications. The Raman system effectively differentiated minimum-to-severe global and macrovesicular steatosis in the subsequent 66 specimens. A dual-variable prediction algorithm was developed, effectively classifying significant discrepancies (> 10%) between algorithm-estimated global HS and pathologist-estimated MaS.

CONCLUSION

Our study established the viability and reliability of a portable spectroscopic system for non-invasive HS and MaS assessment in human liver specimens. The compatibility with ambient light conditions and the ability to address limitations of previous methods marks a significant advancement in this field. By offering promising differentiation between global HS and MaS, our system introduces an innovative approach to real-time and quantitative donor HS assessments. The proposed method holds the promise of refining donor liver assessment during liver recovery and ultimately enhancing transplantation outcomes.

Potential of Raman-Reflectance Combination in Quantifying Liver Steatosis and Fat Droplet Size: Evidence From Monte Carlo Simulations and Phantom Studies

This study explores a combined strategy of Raman and reflectance spectroscopy for quantifying liver fat content and fat droplet size, crucial in assessing donor livers. By using Monte Carlo simulations and experimental setups with oil-in-water phantoms, our findings indicate that Raman scattering can solely differentiate between varying fat contents. At the same time, reflectance intensity is influenced by both fat content and oil droplet size, with a more pronounced sensitivity to fat droplet size. This study demonstrates the efficacy of combined Raman and reflectance spectroscopy in assessing liver steatosis and fat droplet size, potentially aiding in assessing donor livers for transplantation.

Enhanced Artificial Intelligence Methods for Liver Steatosis Assessment Using Machine Learning and Color Image Processing: Liver Color Project

BACKGROUND

The use of livers with significant steatosis is associated with worse transplantation outcomes. Brain death donor liver acceptance is mostly based on subjective surgeon assessment of liver appearance, since steatotic livers acquire a yellowish tone. The aim of this study was to develop a rapid, robust, accurate, and cost-effective method to assess liver steatosis.

METHODS

From June 1, 2018, to November 30, 2023, photographs and tru-cut needle biopsies were taken from adult brain death donor livers at a single university hospital for the study. All the liver photographs were taken by smartphones then color calibrated, segmented, and divided into patches. Color and texture features were then extracted and used as input, and the machine learning method was applied. This is a collaborative project between Vall d'Hebron University Hospital and Barcelona MedTech, Pompeu Fabra University, and is referred to as LiverColor.

RESULTS

A total of 192 livers (362 photographs and 7240 patches) were included. When setting a macrosteatosis threshold of 30%, the best results were obtained using the random forest classifier, achieving an AUROC = 0.74, with 85% accuracy.

CONCLUSION

Machine learning coupled with liver texture and color analysis of photographs taken with smartphones provides excellent accuracy for determining liver steatosis.

Clinical Application of Infrared Spectroscopy in Liver Transplantation for Rapid Assessment of Lipid Content in Liver Graft

Liver transplantation (LT) is a major treatment for patients with end-stage liver diseases. Steatosis is a significant risk factor for primary graft nonfunction and associated with poor long-term graft outcomes. Traditionally, the evaluation of steatosis is based on frozen section examination to estimate the percentage of hepatocytes containing lipid vesicles. However, this visual evaluation correlates poorly with the true lipid content. This study aimed to address the potential of infrared (IR) microspectroscopy for rapidly estimating lipid content in the context of LT and assessing its impact on survival. Clinical data were collected for >20 months from 58 patients who underwent transplantation. For each liver graft, macrovacuolar steatosis and microvesicular steatosis were evaluated through histologic examination of frozen tissue section. Triglycerides (TG) were further quantified using gas phase chromatography coupled with a flame ionization detector (GC-FID) and estimated by IR microspectroscopy. A linear relationship and significant correlation were observed between the TG measured by GC-FID and those estimated using IR microspectroscopy (R2 = 0.86). In some cases, microvesicular steatosis was related to high lipid content despite low levels of macrovacuolar steatosis. Seven patients experienced posttransplantation liver failure, including 5 deceased patients. All patients underwent transplantation with grafts containing significantly high TG levels. A concentration of 250 nmol/mg was identified as the threshold above which the risk of failure after LT significantly increased, affecting 35% of patients. Our study established a strong correlation between LT outcomes and lipid content. IR microspectroscopy proved to be a rapid and reliable approach for assessing the lipid content in clinical settings.

LiverColor: An Artificial Intelligence Platform for Liver Graft Assessment

Hepatic steatosis, characterized by excess fat in the liver, is the main reason for discarding livers intended for transplantation due to its association with increased postoperative complications. The current gold standard for evaluating hepatic steatosis is liver biopsy, which, despite its accuracy, is invasive, costly, slow, and not always feasible during liver procurement. Consequently, surgeons often rely on subjective visual assessments based on the liver's colour and texture, which are prone to errors and heavily depend on the surgeon's experience. The aim of this study was to develop and validate a simple, rapid, and accurate method for detecting steatosis in donor livers to improve the decision-making process during liver procurement. We developed LiverColor, a co-designed software platform that integrates image analysis and machine learning to classify a liver graft into valid or non-valid according to its steatosis level. We utilized an in-house dataset of 192 cases to develop and validate the classification models. Colour and texture features were extracted from liver photographs, and graft classification was performed using supervised machine learning techniques (random forests and support vector machine). The performance of the algorithm was compared against biopsy results and surgeons' classifications. Usability was also assessed in simulated and real clinical settings using the Mobile Health App Usability Questionnaire. The predictive models demonstrated an area under the receiver operating characteristic curve of 0.82, with an accuracy of 85%, significantly surpassing the accuracy of visual inspections by surgeons. Experienced surgeons rated the platform positively, appreciating not only the hepatic steatosis assessment but also the dashboarding functionalities for summarising and displaying procurement-related data. The results indicate that image analysis coupled with machine learning can effectively and safely identify valid livers during procurement. LiverColor has the potential to enhance the accuracy and efficiency of liver assessments, reducing the reliance on subjective visual inspections and improving transplantation outcomes.

Defatting of donor transplant livers during normothermic perfusion-a randomised clinical trial: study protocol for the DeFat study

BACKGROUND

Liver disease is the third leading cause of premature death in the UK. Transplantation is the only successful treatment for end-stage liver disease but is limited by a shortage of suitable donor organs. As a result, up to 20% of patients on liver transplant waiting lists die before receiving a transplant. A third of donated livers are not suitable for transplant, often due to steatosis. Hepatic steatosis, which affects 33% of the UK population, is strongly associated with obesity, an increasing problem in the potential donor pool. We have recently tested defatting interventions during normothermic machine perfusion (NMP) in discarded steatotic human livers that were not transplanted. A combination of therapies including forskolin (NKH477) and L-carnitine to defat liver cells and lipoprotein apheresis filtration were investigated. These interventions resulted in functional improvement during perfusion and reduced the intrahepatocellular triglyceride (IHTG) content. We hypothesise that defatting during NMP will allow more steatotic livers to be transplanted with improved outcomes.

METHODS

In the proposed multi-centre clinical trial, we will randomly assign 60 livers from donors with a high-risk of hepatic steatosis to either NMP alone or NMP with defatting interventions. We aim to test the safety and feasibility of the defatting intervention and will explore efficacy by comparing ex-situ and post-reperfusion liver function between the groups. The primary endpoint will be the proportion of livers that achieve predefined functional criteria during perfusion which indicate potential suitability for transplantation. These criteria reflect hepatic metabolism and injury and include lactate clearance, perfusate pH, glucose metabolism, bile composition, vascular flows and transaminase levels. Clinical secondary endpoints will include proportion of livers transplanted in the two arms, graft function; cell-free DNA (cfDNA) at follow-up visits; patient and graft survival; hospital and ITU stay; evidence of ischemia-reperfusion injury (IRI); non-anastomotic biliary strictures and recurrence of steatosis (determined on MRI at 6 months).

DISCUSSION

This study explores ex-situ pharmacological optimisation of steatotic donor livers during NMP. If the intervention proves effective, it will allow the safe transplantation of livers that are currently very likely to be discarded, thereby reducing waiting list deaths.

TRIAL REGISTRATION

ISRCTN ISRCTN14957538. Registered in October 2022.

Point-of-care device for the noninvasive assessment of hepatic macrosteatosis in liver donors

BACKGROUND

Quantification of macrosteatosis (MS) in the liver is important given that it has shown to directly correlate with adverse post-liver transplant (LT) outcomes. With advances in medical technology and an implicit understanding of pathology, noninvasive methods of quantitatively assessing MS are in various stages of development. Each of these methods is based on the physical principles of differences between a fat-laden hepatocyte and a normal one.

METHODS

In this regard, after a proof-of-concept study on a prototype for a simple, real-time, handheld device using the principle of diffuse reflectance spectroscopy, this study presents an upgraded point-of-care (POC) device for the noninvasive assessment of hepatic MS in liver donors.

RESULTS

The device was validated on cohort of donor livers and showed a sensitivity (0.0021 V/% fat) and highly correlated (r = 0.9868, P < .0001) with gold-standard liver biopsy. Results showed that this upgraded POC device provides a reliable method for the noninvasive assessment of hepatic MS, which is crucial for selecting suitable donor livers for LT.

CONCLUSION

The device has the potential to be an invaluable apparatus at the hands of the organ-retrieving surgeon. It is noninvasive, portable (handheld), and economic; provides real-time readings of the percentage of MS; and can be efficaciously handled by any member of the organ-retrieving team.

Real-time assessment of liver fat content using a filter-based Raman system operating under ambient light through lock-in amplification

During liver procurement, surgeons mostly rely on their subjective visual inspection of the liver to assess the degree of fatty infiltration, for which misclassification is common. We developed a Raman system, which consists of a 1064 nm laser, a handheld probe, optical filters, photodiodes, and a lock-in amplifier for real-time assessment of liver fat contents. The system performs consistently in normal and strong ambient light, and the excitation incident light penetrates at least 1 mm into duck fat phantoms and duck liver samples. The signal intensity is linearly correlated with MRI-calibrated fat contents of the phantoms and the liver samples.

Radiomic analysis of liver grafts from brain-dead donors can predict early allograft dysfunction following transplantation: a proof-of-concept study

BACKGROUND

Selection of liver grafts suitable for transplantation (LT) mainly depends on a surgeon's subjective assessment. This study aimed to investigate the role of radiomic analysis of donor-liver CTs after brain death (DBD) to predict the occurrence of early posttransplant allograft dysfunction (EAD).

METHODS

We retrospectively extracted and analyzed the left lobe radiomic features from CT scans of DBD livers in training and validation cohorts. Multivariate analysis was performed to identify predictors of EAD.

RESULTS

From 126 LTs included in the study in the training cohort, 27 (21.4%) had an EAD. For each patient, 279 radiomic features were extracted of which 5 were associated with EAD (AUC = 0.81) (95% CI 0.72-0.89). Among donor and recipient clinical characteristics, cardiac arrest, steatosis on donor's CT, cold ischemic time and age of recipient were also identified as independent risk factors for EAD. Combined radiomic signature and clinical risk factors showed a strong predictive performance for EAD with a C-index of 0.90 (95% CI 0.84-0.96). A validation cohort of 23 patients confirmed these results.

CONCLUSION

Radiomic signatures extracted from donor CT scan, independently or combined with clinical risk factors is an objective and accurate biomarker for prediction of EAD after LT.

Bluetooth-Connected Pocket Spectrometer and Chemometrics for Olive Oil Applications

Unsaturated fatty acids are renowned for their beneficial effects on the cardiovascular system. The high content of unsaturated fatty acids is a benefit of vegetable fats and an important nutraceutical indicator. The ability to quickly check fat composition of an edible oil could be advantageous for both consumers and retailers. A Bluetooth-connected pocket spectrometer operating in NIR band was used for analyzing olive oils of different qualities. Reference data for fatty acid composition were obtained from a certified analytical laboratory. Chemometrics was used for processing data, and predictive models were created for determining saturated and unsaturated fatty acid content. The NIR spectrum also demonstrated good capability in classifying extra virgin and non-extra virgin olive oils. The pocket spectrometer used in this study has a relatively low cost, which makes it affordable for a wide class of users. Therefore, it may open the opportunity for quick and non-destructive testing of edible oil, which can be of interest for consumer, retailers, and for small/medium-size producers, which lack easy access to conventional analytics.

Diffuse Reflectance Spectroscopy for The Assessment of Steatosis in Liver Phantom and Liver Donors - A Pilot Study

This paper reports the application of a low-cost diagnostic modality for fat analysis in a liver phantom as well as human liver donors. The device works on the principle of diffuse reflectance spectroscopy, which absorbs and/or scatters depending upon the molecules that compose a tissue. Here, we describe the development of liver phantom of varying fat concentration using saturated fat mimicking liver steatosis. Followed by a pilot study in the human liver donor setting. Later, handheld device based on Infrared-LED and Photodetector for real-time time assessment of live donor liver and fat assessment. Clinical Relevance- This device can be used in the development of an accurate and non-invasive for quantification of liver fat in the deceased donor selection process. It has an error margin of 10% in the quantification of fat which is comparable to a standard biopsy technique.

Proposal and validation of a liver graft discard score for liver transplantation from deceased donors: a multicenter Italian study

Several studies have explored the risk of graft dysfunction after liver transplantation (LT) in recent years. Conversely, risk factors for graft discard before or at procurement have poorly been investigated. The study aimed at identifying a score to predict the risk of liver-related graft discard before transplantation. Secondary aims were to test the score for prediction of biopsy-related negative features and post-LT early graft loss. A total of 4207 donors evaluated during the period January 2004–Decemeber 2018 were retrospectively analyzed. The group was split into a training set (n = 3,156; 75.0%) and a validation set (n = 1,051; 25.0%). The Donor Rejected Organ Pre-transplantation (DROP) Score was proposed: − 2.68 + (2.14 if Regional Share) + (0.03*age) + (0.04*weight)-(0.03*height) + (0.29 if diabetes) + (1.65 if anti-HCV-positive) + (0.27 if HBV core) − (0.69 if hypotension) + (0.09*creatinine) + (0.38*log10AST) + (0.34*log10ALT) + (0.06*total bilirubin). At validation, the DROP Score showed the best AUCs for the prediction of liver-related graft discard (0.82; p < 0.001) and macrovesicular steatosis ≥ 30% (0.71; p < 0.001). Patients exceeding the DROP 90th centile had the worse post-LT results (3-month graft loss: 82.8%; log-rank P = 0.024).The DROP score represents a valuable tool to predict the risk of liver function-related graft discard, steatosis, and early post-LT graft survival rates. Studies focused on the validation of this score in other geographical settings are required.

Non-invasive real-time assessment of hepatic macrovesicular steatosis in liver donors: Hypothesis, design and proof-of-concept study

Macrovesicular Steatosis (MS) is an independent risk factor for adverse post-liver transplant (LT) outcomes. The degree of MS is intimately related to the viability of the liver graft, which in turn is crucial to the success of the operation. An ideal liver graft should have no MS and most centres would find it unacceptable to use a donor liver with severe MS for LT. While a formal liver biopsy is the gold-standard diagnostic test for MS, given the logistical and time constraints it is not universally feasible. Other tests like a frozen section biopsy are plagued by issues of fallibility with reporting and sampling bias making them inferior to a liver biopsy. Hence, the development of an accurate, non-invasive, easy-to-use, handheld, real-time device for quantification of MS would fill this lacuna in the deceased donor selection process. We present the hypothesis, design and proof-of-concept of a study, which aims to standardise and determine the feasibility and accuracy of a novel handheld device applying the principle of diffuse reflectance spectroscopy for real-time quantification of MS.

Changing Trends in Liver Transplantation: Challenges and Solutions

Despite improvements in postliver transplant outcomes through refinements in perioperative management and surgical techniques, several changing trends in liver transplantation have presented challenges. Mortality on the waitlist remains high. In the United States, Europe, and the United Kingdom, there is an increasing need for liver transplantation, primarily as a result of increased incidence of nonalcoholic steatohepatitis-related cirrhosis and cancer indications. Meanwhile, donor suitability has decreased, as donors are often older and have more comorbidities. Despite a mismatch between organ need and availability, many organs are discarded. Notwithstanding this, many solutions have been developed to overcome these challenges. Innovative techniques in allograft preservation, viability assessment, and reconditioning have allowed the use of suboptimal organs with adequate results. Refinements in surgical procedures, including live donor liver transplantations, have increased the organ pool and are decreasing the time and mortality on the waitlist. Despite many challenges, a similar number of solutions and prospects are on the horizon. This review seeks to explore the changing trends and challenges in liver transplantation and highlight possible solutions and future directions.

A Model to Predict Significant Macrosteatosis in Hepatic Grafts

BACKGROUND AND AIMS

Assessing the risk of significant macrosteatosis in donors is crucial before considering hepatic graft procurement. We aimed to build a model to predict significant macrosteatosis based on noninvasive methods.

METHODS

From January 2012 to December 2018, liver attenuation indices and liver-to-spleen (L/S) ratio were measured in 639 brain-dead donors by local radiologists. Quantity and quality of steatosis were evaluated by an expert pathologist, blinded for attenuation indices measurement.

RESULTS

Macrosteatosis ≥ 30% was found in 33 donors (5.2%). Body weight, body mass index (BMI), abdominal perimeters, history of alcohol abuse, L/S ratio, and liver parenchyma attenuation were associated with macrosteatosis ≥ 30%. The L/S ratio, BMI, and a history of alcohol abuse remained independent predictors in multivariate analysis and were used to build a predictive model (C-index: 0.77). The optimal cutoff to predict macrosteatosis ≥ 60% was 0.85.

CONCLUSION

Our model, including L/S ratio, BMI, and history of alcohol, might be helpful to refine indication for liver biopsy before donation after brain death. External validation is required.

Deep learning quantification of percent steatosis in donor liver biopsy frozen sections

Background

Pathologist evaluation of donor liver biopsies provides information for accepting or discarding potential donor livers. Due to the urgent nature of the decision process, this is regularly performed using frozen sectioning at the time of biopsy. The percent steatosis in a donor liver biopsy correlates with transplant outcome, however there is significant inter- and intra-observer variability in quantifying steatosis, compounded by frozen section artifact. We hypothesized that a deep learning model could identify and quantify steatosis in donor liver biopsies.

Methods

We developed a deep learning convolutional neural network that generates a steatosis probability map from an input whole slide image (WSI) of a hematoxylin and eosin-stained frozen section, and subsequently calculates the percent steatosis. Ninety-six WSI of frozen donor liver sections from our transplant pathology service were annotated for steatosis and used to train (n = 30 WSI) and test (n = 66 WSI) the deep learning model.

Findings

The model had good correlation and agreement with the annotation in both the training set (r of 0.88, intraclass correlation coefficient [ICC] of 0.88) and novel input test sets (r = 0.85 and ICC=0.85). These measurements were superior to the estimates of the on-service pathologist at the time of initial evaluation (r = 0.52 and ICC=0.52 for the training set, and r = 0.74 and ICC=0.72 for the test set).

Interpretation

Use of this deep learning algorithm could be incorporated into routine pathology workflows for fast, accurate, and reproducible donor liver evaluation.

Funding

Mid-America Transplant Society