Simple scintigraphic parameters with Tc-99m galactosyl human serum albumin for clinical staging of chronic hepatocellular dysfunction

Application of a machine learning approach to characterization of liver function using 99mTc-GSA SPECT/CT

PURPOSE

To assess the utility of a machine-learning approach for predicting liver function based on technetium-99 m-galactosyl serum albumin (^99mTc-GSA) single photon emission computed tomography (SPECT)/CT.

METHODS

One hundred twenty-eight patients underwent a ^99mTc-GSA SPECT/CT-based liver function evaluation. All were classified into the low liver-damage or high liver-damage group. Four clinical (age, sex, background liver disease and histological type) and 8 quantitative ^99mTc-GSA SPECT/CT features (receptor index [LHL15], clearance index [HH15], liver-SUVmax, liver-SUVmean, heart-SUVmax, metabolic volume of liver [MVL], total lesion GSA [TL-GSA, liver-SUVmean × MVL] and SUVmax ratio [liver-SUVmax/heart-SUVmax]) were obtained. To predict high liver damage, a machine learning classification with features selection based on Gini impurity and principal component analysis (PCA) were performed using a support vector machine and a random forest (RF) with a five-fold cross-validation scheme. To overcome imbalanced data, stratified sampling was used. The ability to predict high liver damage was evaluated using a receiver operating characteristic (ROC) curve analysis.

RESULTS

Four indices (LHL15, HH15, heart SUVmax and SUVmax ratio) yielded high areas under the ROC curves (AUCs) for predicting high liver damage (range: 0.89-0.93). In a machine learning classification, the RF with selected features (heart SUVmax, SUVmax ratio, LHL15, HH15, and background liver disease) and PCA model yielded the best performance for predicting high liver damage (AUC = 0.956, sensitivity = 96.3%, specificity = 90.0%, accuracy = 91.4%).

CONCLUSION

A machine-learning approach based on clinical and quantitative ^99mTc-GSA SPECT/CT parameters might be useful for predicting liver function.

Heterogeneous liver uptake of Tc-99m-GSA as quantified through SPECT/CT helps to evaluate the degree of liver fibrosis: A retrospective observational study

Tc-99m-galactosyl human serum albumin (GSA) scintigraphy is used to assess the hepatic functional reserve, and allows for visual assessment of the residual hepatocyte distribution on single-photon emission computed tomography/computed tomography (SPECT/CT) images. The association between heterogeneous liver uptake of Tc-99m-GSA and liver fibrosis remains to be studied in detail. We analyzed this association.Fifty-one patients with chronic hepatobiliary disease undergoing a Tc-99m-GSA scintigraphy were included in this study. The receptor (LHL15) and blood clearance (HH15) indexes (the uptake ratios of the liver and heart) were obtained from dynamic planar images. The liver uptake count maximum-to-mean ratio (LUC Max/Mean) was calculated from single-photon emission computed tomography/computed tomography (SPECT/CT) images as an indicator of the Tc-99m-GSA liver uptake heterogeneity. We assessed the relationship between these quantified values and liver fibrosis.There were 30 Child-Pugh classification grade A patients, 16 grade B patients, and 5 grade C patients. Among the 30 patients whose liver histopathology was evaluable, those with advanced liver fibrosis (F2-4) had a lower LHL15 than those with mild liver fibrosis (F0-1) (median, 0.90 vs. 0.92, P = .04), and a higher LUC Max/Mean (median, 1.80 vs. 1.70, P = .02). The multivariate analysis identified platelets (P = .04) and the LUC Max/Mean (P = .04) as contributing factors of advanced liver fibrosis.These findings suggest that Tc-99m-GSA SPECT/CT can be used not only to assess the hepatic functional reserve, but also to evaluate a degree of liver fibrosis.

[Novel Index (Hepatic Receptor: IHR) to Evaluate Hepatic Functional Reserve Using (99m)Tc-GSA Scintigraphy]

PURPOSE

This study aimed to evaluate the novel index of hepatic receptor (IHR) on the regression analysis derived from time activity curve of the liver for hepatic functional reserve.

METHODS

Sixty patients had undergone (99m)Tc-galactosyl serum albumin ((99m)Tc-GSA) scintigraphy in the retrospective clinical study. Time activity curves for liver were obtained by region of interest (ROI) on the whole liver. A novel hepatic functional predictor was calculated with multiple regression analysis of time activity curves. In the multiple regression function, the objective variables were the indocyanine green (ICG) retention rate at 15 min, and the explanatory variables were the liver counts in 3-min intervals until end from beginning. Then, this result was defined by IHR, and we analyzed the correlation between IHR and ICG, uptake ratio of the heart at 15 minutes to that at 3 minutes (HH15), uptake ratio of the liver to the liver plus heart at 15 minutes (LHL15), and index of convexity (IOC).

RESULTS

Regression function of IHR was derived as follows: IHR=0.025×L(6)-0.052×L(12)+0.027×L(27). The multiple regression analysis indicated that liver counts at 6 min, 12 min, and 27 min were significantly related to objective variables. The correlation coefficient between IHR and ICG was 0.774, and the correlation coefficient between ICG and conventional indices (HH15, LHL15, and IOC) were 0.837, 0.773, and 0.793, respectively. IHR had good correlation with HH15, LHL15, and IOC.

CONCLUSIONS

The finding results suggested that IHR would provide clinical benefit for hepatic functional assessment in the (99m)Tc-GSA scintigraphy.

Improved reproducibility of simple quantitative indices from ⁹⁹mTc-GSA liver functional imaging

Objective

We evaluated intra- and interoperator reproducibilities in calculating the conventional indices HH15 and LHL15 from ^99mTc-diethylenetriamine pentaacetic acid galactosyl human serum albumin (^99mTc-GSA) scintigraphy, and proposed new, simple methods for the calculation of quantitative indices.

Methods

The results of ^99mTc-GSA scintigraphy in 33 patients were retrospectively analyzed. Heart and liver ROIs were drawn manually to cover cardiac blood pool and entire liver, respectively, and HH15 and LHL15 were calculated. In addition, square regions of interest (ROIs) of fixed sizes were placed at the highest activity in blood pool and the liver. Using the square heart ROI, sHH15, an equivalent of HH15, was computed. Fractional liver uptake at 15 min (FLU15) was calculated using the square heart and liver ROIs. Intra- and interoperator reproducibilities, as well as correlation with Indocyanine green retention rate at 15 min (ICG R15), were assessed for these four indices by linear regression analysis.

Results

Substantial intra- and interoperator variabilities were found for HH15 and LHL15. The correlation coefficients for intra- and interoperator comparisons were 0.884 and 0.869 for HH15, respectively, and 0.919 and 0.917 for LHL15, respectively. The use of square ROIs instead of hand-drawn ROIs improved reproducibility. The correlation coefficients for intra- and interoperator comparisons were 0.988 and 0.973 for sHH15, respectively, and 0.989 and 0.975 for FLU15, respectively. Correlation with ICG R15 was better for sHH15 (r = 0.619) and FLU15 (r = −0.656) than for HH15 (r = 0.439) and LHL15 (r = −0.490).

Conclusions

HH15 and LHL15 showed substantial intra- and interoperator variabilities, and the use of square ROIs are indicated to provide better reproducibility.

Glycoconjugate probes and targets for molecular imaging using magnetic resonance

Recently, many research activities in medical diagnosis have been devoted to molecular imaging with MRI. A key issue is the evaluation of molecular targets that allow the early detection and characterization of diseases and the assessment of the effects of therapy. The majority of the current targeting vectors are peptides and proteins; reports on carbohydrate-based probes are relatively scarce. However, molecular recognitions involving carbohydrates are ubiquitous in both normal and pathological natural processes. Here, we critically review the literature on the development and validation of MRI probes using carbohydrates either as targets or targeting vectors. Exploitation of molecular recognition involving carbohydrates in MRI looks promising. Amplification techniques may be important for overcoming sensitivity problems.

Relationship between 99mTc-GSA scintigraphic indices of liver function reserve and portal circulation in patients with chronic liver disease

OBJECTIVES

In order to predict the prognosis or complications of portal hypertension in patients with chronic liver disease, it is important to evaluate both hepatic functional reserve and portal circulation. On (99m)Tc-galactosyl human serum albumin ((99m)Tc-GSA) scintigraphy, the index of blood clearance (HH15) and receptor index (LHL15) have been widely used to evaluate the hepatic functional reserve. However, the relationship between these indices and portal circulation is unknown. The purpose of this study was to examine the relationship between HH15 and LHL15 and portosystemic shunts evaluated with arteriographic portography or esophagogastroduodenoscopy.

METHODS

A total of 82 patients with chronic liver disease (mean age, 66.7 years) who underwent (99m)Tc-GSA scintigraphy, arteriographic portography, and esophagogastroduodenoscopy were enrolled. HH15 and LHL15 were obtained from dynamic (99m)Tc-GSA scintigraphy. The patients were divided into three groups according to the arteriographic portography findings: group 1, no portal collateral circulation; group 2, mild collateral development; and group 3, moderate to severe collateral development. They were also divided into three groups based on the esophagogastroduodenoscopic findings: group A, no varices; group B, small-caliber varices; and group C, enlarged varices. The Kruskal-Wallis test was used to compare each index among these groups. Receiver operating characteristic (ROC) analysis was used to determine whether each index was an indicator for the presence of portosystemic shunts.

RESULTS

Both HH15 and LHL15 differed significantly between groups 1 and 2 and between groups 1 and 3. However, only HH15 differed significantly between groups A and B and between groups A and C. On the basis of the ROC analysis, the HH15 threshold value of 0.62 yielded both excellent sensitivity (83.9%) and specificity (84.6%) for the presence of portosystemic shunts, as evaluated with arteriographic portography. The HH15 threshold value of 0.64 yielded both good sensitivity (66.1%) and specificity (69.2%) for the presence of portosystemic shunts, as evaluated with esophagogastroduodenoscopy, whereas no adequate threshold value of LHL15 was found for the presence of portosystemic shunts.

CONCLUSIONS

HH15 is a potent indicator of the presence of portosystemic shunts in chronic liver disease.

Targeting of lanthanide(III) chelates of DOTA-type glycoconjugates to the hepatic asyaloglycoprotein receptor: cell internalization and animal imaging studies

The characterization of a new class of hydrophilic liver-targeted agents for gamma-scintigraphy and MRI, consisting, respectively, of [(153)Sm](3+) or Gd(3+) complexes of DOTA monoamide or bisamide linked glycoconjugates (DOTA = 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid), is reported. In vitro studies show high uptake of radiolabeled [(153)Sm]-DOTAGal(2) by the human hepatocyte carcinoma cell line Hep G2 containing the asialoglycoprotein receptor (ASGP-R), which is decreased to less than 50% by the presence of its high-affinity ligand asialofetuin (ASF). In vivo biodistribution, gamma-imaging and pharmacokinetic studies on Wistar rats using the [(153)Sm](3+)-labeled glycoconjugates show a high uptake in the receptor-rich organ liver of the radiolabeled compounds containing terminal galactosyl groups, but very little uptake for those compounds with terminal glycosyl groups. Blocking the receptor in vivo reduced liver uptake by 90%, strongly suggesting that the liver uptake of these compounds is mediated by their binding to the asyaloglycoprotein receptor (ASGP-R). This study also demonstrated that the valency increase improves the targeting capability of the glycoconjugates, which is also affected by their topology. However despite the specific liver uptake of the radiolabeled galactose-bearing multivalent compounds, the animal MRI assessment of the corresponding Gd(3+) chelates shows liver-to-kidney contrast effects which are not significantly better than those shown by GdDTPA. This probably results from the quick wash-out from the liver of these highly hydrophilic complexes, before they can be sufficiently concentrated within the hepatocytes via receptor-mediated endocytosis.

Preoperative assessment of liver function

At the present time, the decision to resect and the choice of the extent ofa hepatic resection are largely based on surgical judgment. The CP score is the best assessment tool we can now employ. There is uniform agreement that even segmental resections are not possible in the vast majority of Child Class B patients, CP score 7 to 9. The CP score can be augmented by radiographic testing, ICG retention testing, and by assessing tumor extent and the severity of the patient's cirrhosis at surgery. Surgeons need a simple means to assist with liver function evaluation--a test to augment the CP score. Although determining ICG retention is simple, it is questionable whether it adds to one's ability to define the poor-risk patient with better accuracy than the CP score. Abundant data exist to dispute the accuracy and reproducibility of ICG retention. That surgeons use it says more about the fervent desire to find a test that supports clinical judgment in these difficult patients than the scientific validity of the test. Whether a series of tests would better define the Child-Pugh Class A patient who is also a relatively poor risk is not clear at present. Many investigations demonstrate the correlation of various assessment tools with each other, yet nothing distinguishes them in predicting risk beyond what is learned from the CP score. In a group of CP Class A patients, the extent of the disease, the nature of underlying cirrhosis, and the extent of resection provide the clinical backdrop against which a decision for resection must be made. It may well be that one test may not do it, or that one single assessment of the ICG or the 15-minute receptor volume of GSA may be inadequate to project the nuances of liver function. Thus, 99m-Tc GSA scintigraphy will provide volumetric receptor data, as well as kinetic distribution curves, and may prove a useful test in the future. Whether GSA is ultimately to be proven useful requires a correlation of the test with actual clinical outcomes, rather than correlation with other tests or with the CP score. Discovering which patients are the poor risk Child Class A patients is the desired goal. To have value, the GSA scan must augment, not mimic, the CP score. In view of the fact that experienced surgeons appear to be astute in their ability to select patients for hepatic resection, finding a more refined test will require large numbers of patients at several centers to correlate the test results and the outcomes against the spectrum of postoperative liver failure, including death. It appears that one lesson learned from portal vein embolization is that functional liver volume can be preserved. The compensatory hyperplasia that occurs in the contralateral hepatic lobe demonstrates two important features: (1) function of the opposite lobe has been transferred when evaluated by 99m-Tc-GSA, and (2) one considerable metabolic drain on the postoperative recovery from hepatic resection (ie, liver regeneration) can be attended to before the surgery. Cirrhotic livers do regenerate, but more slowly. Thus, pregrowing the remnant section of liver eliminates one stress on liver reserves following liver resection. For hepatocellular carcinoma or metastasis in cirrhotic patients, portal vein occlusion may be the best way to improve hepatic functional reserve. ICG retention may not corroborate return-to-baseline hepatic function within 2 weeks of portal vein occlusion,but may demonstrate a return to baseline when studied 6 to 8 weeks following the procedure. 99m-Tc-GSA is presently the best means to document compensatory hyperplasia and, possibly, a shift of functional reserve to the planned remnant of a more than four-segment hepatic resection. Whether this will predict the safe outcome of resection remains to be seen.

Clinical assessment of hepatic functional reserve using 99mTc DTPA galactosyl human serum albumin SPECT to prognosticate chronic hepatic diseases--validation of the use of SPECT and a new indicator

RATIONALE

It is generally known that scintigraphy of 99mTc diethylenetriamine pentaacetic acid-galactosyl human serum albumin (99mTc-GSA) is useful for assessing hepatic functional reserve. For hepatic functional indicators, the index of the calculated planar image has been used in previous studies. However, there have been few reports that suggest that the indicators calculated from static SPECT data would be useful for the assessment of hepatic function. The aims of this study were to establish a simple method for assessing hepatic functional reserve using the liver SPECT of 99mTc-GSA and to apply this method for rich stratification in patients with chronic hepatic diseases.

METHODS

A liver phantom (a 50% concentration of 99mTc solution) was used to compare the planar and SPECT methods. According to the definition of the new indicator, the liver SPECT of 99mTc-GSA was divided by a syringe SPECT of 99mTc-GSA and was called the liver uptake ratio (LUR). We correlated the LUR and the liver uptake ratio calculated according to the blood-sampling method. 99mTc-GSA SPECT was performed in 137 patients with hepatic diseases, including chronic hepatic diseases, and 20 healthy volunteers. The LUR was correlated between the formed subtypes for all subjects.

RESULTS

The acquired phantom-count ratio calculated by the SPECT method was more accurate than that acquired by the planar method. A good correlation was obtained between the LUR and the blood-sampling method (r = 0.971). The LUR was significantly lower in subjects with severe cirrhosis than in healthy subjects or those with chronic hepatitis and mild cirrhosis, and it was significantly lower in subjects with chronic hepatitis and mild cirrhosis than in healthy subjects. The LUR was significantly correlated with other hepatic function tests. Based on LUR, the chronic hepatic diseases were divided into two groups: Group A, with LURs 30% and higher, and Group B, with LURs below 30%. An LUR of 30% marked the 25th percentile of the mild-cirrhosis group. The cumulative survival rates were lower in Group B than in Group A.

CONCLUSION

The SPECT method was superior to the planar method for assessing LURs. LUR was a suitable indicator of 99mTc-GSA clearance from the blood pool and of binding to the asialo-glycoprotein receptor. LUR is a simple and clinically useful indicator for the assessment of hepatic functional reserve in chronic hepatic diseases.

A case of recurrent cholangitis after bile duct injury during laparoscopic cholecystectomy: value of scintigraphy with Tc-99m GSA and hepatobiliary scintigraphy for indication of lobectomy

A 39-year-old woman with acute cholecystitis and gallstones underwent laparoscopic cholecystectomy. She suffered from recurrent episodes of cholangitis due to injury of the major bile ducts during laparoscopic cholecystectomy. Hepatobiliary scintigraphy with Tc-99m Sn-N-pyridoxyl-5-methyltryptophan was performed. Although normal bile excretion was found from the left hepatic duct to the percutaneous transhepatic biliary drainage (PTBD) tube, excretion from the right hepatic lobe was prolonged. Scintigraphy with Tc-99m diethylenetriaminepentaacetic acid-galactosyl human serum albumin demonstrated atrophy of the right hepatic lobe and enlargement of the left hepatic lobe. Cholangiography via the PTBD tube revealed complete obstruction of the left hepatico-jejunal anastomosis and could not enhance the right intrahepatic bile duct. A right hepatic lobectomy was performed because of the atrophy, glissonitis and the absence of an appropriate bile duct for reconstruction. Postoperatively she was active and exhibited no evidence of recurrent cholangitis.

Course before and after percutaneous transhepatic portal vein embolization of a patient with cholangiocarcinoma monitored by scintigraphy with Tc-99m galactosyl human serum albumin

Percutaneous transhepatic portal vein embolization (PTPE) causes atrophy of the embolized lobe and compensatory hypertrophy of the nonembolized lobe, and improves the safety of hepatectomy. We report a patient with cholangiocarcinoma who underwent embolization of both anterior and posterior branches of the right portal vein before hepatectomy. Scintigraphy with Tc-99m galactosyl human serum albumin was performed before and 4 weeks after PTPE. After PTPE, the right lobe of the liver was atrophied and the left lobe of the liver was enlarged, compared with before PTPE. The receptor index of the entire liver was almost unchanged before and after PTPE, but the right lobe receptor index after PTPE was 23% less than the pre-PTPE value, whereas the left lobe receptor index had increased 37%. Scintigraphy with Tc-99m galactosyl human serum albumin is useful for evaluating segmental functional reserve before and after PTPE.