Clinical value of increased serum creatinine concentration as predictor of short-term outcome in decompensated cirrhosis

Mid-infrared spectroscopy of serum, a promising non-invasive method to assess prognosis in patients with ascites and cirrhosis

Background & aims

Prognostic tests are critical in the management of patients with cirrhosis and ascites. Biological tests or scores perform poorly in that situation. Mid-infrared fibre evanescent wave spectroscopy (MIR-FEWS) which allows for global serum metabolic profiling may provide more relevant information by measuring a wider range of metabolic parameters in serum. Here we present the accuracy of a MIR-FEWS based predictive model for the prognosis of 6 months survival in patients with ascites and cirrhosis.

Methods

Patients with ascites were prospectively included and followed up for 6 months. MIR-FEWS spectra were measured in serum samples. The most informative spectral variables obtained by MIR-FEWS were selected by FADA algorithm and then used to build the MIR model. Accuracy of this model was assessed by ROC curves and 90%/10% Monte Carlo cross-validation. MIR model accuracy for 6 months survival was compared to that of the Child-Pugh and MELD scores.

Results

119 patients were included. The mean age was 57.36±13.70, the MELD score was 16.32±6.26, and the Child-Pugh score was 9.5±1.83. During follow-up, 23 patients died (20%). The MIR model had an AUROC for 6 months mortality of 0.90 (CI95: 0.88–0.91), the MELD 0.77 (CI95: 0.66–0.89) and Child-Pugh 0.76 (CI95: 0.66–0.88). MELD and Child-Pugh AUROCs were significantly lower than that of the MIR model (p = 0.02 and p = 0.02 respectively). Multivariate logistic regression analysis showed that MELD (p<0.05, OR:0.86;CI95:0.76–0.97), Beta blockers (p = 0.036;OR:0.20;CI95:0.04–0.90), and the MIR model (p<0.001; OR:0.50; CI95:0.37–0.66), were significantly associated with 6 months mortality.

Conclusions

In this pilot study MIR-FEWS more accurately assess the 6-month prognosis of patients with ascites and cirrhosis than the MELD or Child-Pugh scores. These promising results, if confirmed by a larger study, suggest that mid infrared spectroscopy could be helpful in the management of these patients.

Feasibility Study of NMR Based Serum Metabolomic Profiling to Animal Health Monitoring: A Case Study on Iron Storage Disease in Captive Sumatran Rhinoceros (Dicerorhinus sumatrensis)

A variety of wildlife species maintained in captivity are susceptible to iron storage disease (ISD), or hemochromatosis, a disease resulting from the deposition of excess iron into insoluble iron clusters in soft tissue. Sumatran rhinoceros (Dicerorhinus sumatrensis) is one of the rhinoceros species that has evolutionarily adapted to a low-iron diet and is susceptible to iron overload. Hemosiderosis is reported at necropsy in many African black and Sumatran rhinoceroses but only a small number of animals reportedly die from hemochromatosis. The underlying cause and reasons for differences in susceptibility to hemochromatosis within the taxon remains unclear. Although serum ferritin concentrations have been useful in monitoring the progression of ISD in many species, there is some question regarding their value in diagnosing hemochromatosis in the Sumatran rhino. To investigate the metabolic changes during the development of hemochromatosis and possibly increase our understanding of its progression and individual susceptibility differences, the serum metabolome from a Sumatran rhinoceros was investigated by nuclear magnetic resonance (NMR)-based metabolomics. The study involved samples from female rhinoceros at the Cincinnati Zoo (n = 3), including two animals that died from liver failure caused by ISD, and the Sungai Dusun Rhinoceros Conservation Centre in Peninsular Malaysia (n = 4). Principal component analysis was performed to visually and statistically compare the metabolic profiles of the healthy animals. The results indicated that significant differences were present between the animals at the zoo and the animals in the conservation center. A comparison of the 43 serum metabolomes of three zoo rhinoceros showed two distinct groupings, healthy (n = 30) and unhealthy (n = 13). A total of eighteen altered metabolites were identified in healthy versus unhealthy samples. Results strongly suggest that NMR-based metabolomics is a valuable tool for animal health monitoring and may provide insight into the progression of this and other insidious diseases.

Child-Pugh Versus MELD Score for the Assessment of Prognosis in Liver Cirrhosis: A Systematic Review and Meta-Analysis of Observational Studies

Child-Pugh and MELD scores have been widely used for the assessment of prognosis in liver cirrhosis. A systematic review and meta-analysis aimed to compare the discriminative ability of Child-Pugh versus MELD score to assess the prognosis of cirrhotic patients.PubMed and EMBASE databases were searched. The statistical results were summarized from every individual study. The summary areas under receiver operating characteristic curves, sensitivities, specificities, positive and negative likelihood ratios, and diagnostic odds ratios were also calculated.Of the 1095 papers initially identified, 119 were eligible for the systematic review. Study population was heterogeneous among studies. They included 269 comparisons, of which 44 favored MELD score, 16 favored Child-Pugh score, 99 did not find any significant difference between them, and 110 did not report the statistical significance. Forty-two papers were further included in the meta-analysis. In patients with acute-on-chronic liver failure, Child-Pugh score had a higher sensitivity and a lower specificity than MELD score. In patients admitted to ICU, MELD score had a smaller negative likelihood ratio and a higher sensitivity than Child-Pugh score. In patients undergoing surgery, Child-Pugh score had a higher specificity than MELD score. In other subgroup analyses, Child-Pugh and MELD scores had statistically similar discriminative abilities or could not be compared due to the presence of significant diagnostic threshold effects.Although Child-Pugh and MELD scores had similar prognostic values in most of cases, their benefits might be heterogeneous in some specific conditions. The indications for Child-Pugh and MELD scores should be further identified.

Assessment and modeling of routinely used biochemical laboratory data of healthy individuals and end-stage renal failure (ESRF) patients by three different chemometric methods

BACKGROUND

In recent years, the use of biochemical markers has received increasing attention for purposes of risk assessment and clinical management in renal failure patients. Chemometric methods are often used in medical studies and there are already indications for their specific role as a tool of the medical statistics.

METHODS

Three chemometric methods, discriminant analysis (DA), binary logistic regression analysis (BLRA), and cluster analysis (CA), were used for assessment and modeling of routinely used biochemical laboratory data of 18 parameters that were determined from 185 healthy individuals (HIs) and 173 end-stage renal failure (ESRF) patients.

RESULTS

The above-mentioned chemometric methods were performed using the data set of 14 parameters since the rest 4 parameters did not present significant difference between healthy and patients. DA created a model using only ALB (Albumin), K (Potassium), TG (Triglyceride), and ALP (Alkaline phosphatase); BLRA model also used the above four parameters; CA classified all the cases into two clusters using the same four parameters and one more parameter, AST (aspartate aminotransferase).

CONCLUSIONS

This study provides models for assessment and modeling of routinely used biochemical laboratory data, finding groups of similarity among clinical tests usually determined on HIs and ESRF patients, contributing in data mining and reducing costs.

Systematic review of noncancer presentations with a median survival of 6 months or less

PURPOSE

We report on clinical indicators of 6-month mortality in advanced noncancer illnesses and the effect of treatment on survival.

METHODS

The MEDLINE database was searched comprehensively to find studies evaluating survival for common advanced noncancer illnesses. We retrieved and evaluated studies that reported a median survival of ≤1 year and evaluated prognostic factors or effect of treatment on survival. We extracted data on presentations with median survivals of ≤6 months for heart failure, chronic obstructive pulmonary disease, dementia, geriatric failure to thrive, cirrhosis, and end-stage renal failure. Independent risk factors for survival were combined and included if their combination was associated with a 6-month mortality of ≥50%.

RESULTS

The search identified 1000 potentially relevant studies, of which 475 were retrieved and evaluated, and 74 were included. We report the common clinical presentations that are consistently associated with a 6-month median survival. Even though advanced noncancer syndromes differ clinically, a universal set of prognostic factors signals progression to terminal disease, including poor performance status, advanced age, malnutrition, comorbid illness, organ dysfunction, and hospitalization for acute decompensation. Generally, a 6-month median survival is associated with the presence of 2-4 of these factors. With few exceptions, these terminal presentations are quite refractory to treatment.

CONCLUSION

This systematic review summarizes prognostic factors common to advanced noncancer illness. There is little evidence at present that treatment prolongs survival at these terminal stages.

Models for prediction of mortality from cirrhosis with special reference to artificial neural network: a critical review

Prediction of mortality of patients with cirrhosis of liver, a common and potentially fatal disease, is important for timely listing of patients for liver transplantation. The Child-Pugh scoring system has been widely used for predicting the outcome of liver cirrhosis. The Model for End-Stage Liver Disease (MELD) score has recently become popular for prediction of short-term mortality for organ allocation. A few studies that evaluated artificial neural network (ANN)-based model for prediction of outcome of cirrhosis of liver in terms of mortality have consistently shown it to be superior to Child-Pugh scoring and logistic regression-based models; it is worth noting that MELD score is also derived using the logistic regression model. Due to the inherent ability of neural network-based systems in identifying complex nonlinear interactions, ANN-based models are expected to perform better than most linear models, such as regression-based models. More studies are needed on ANN-based models for prediction of mortality of patients with cirrhosis of liver and its value in prioritization of organ allocation for treatment of patients with cirrhosis of liver.

Renal function and cognitive impairment in patients with liver cirrhosis

OBJECTIVE

Cognitive impairment is a common problem in patients with liver cirrhosis. Its pathogenesis is multifactorial and ammonia is considered to play a central role. Renal function has been shown to be important for ammonia metabolism in cirrhosis. Although renal dysfunction is common in cirrhotic patients, its effect on cognitive function is largely unexplored.

MATERIAL AND METHODS

A total of 128 consecutive cirrhotic patients were prospectively evaluated for the presence of cognitive dysfunction according to the West-Haven criteria and by means of two psychometric tests. Serum creatinine, sodium and potassium as well as plasma ammonia concentrations were assessed. Glomerular filtration rate was also measured by (51)Cr- EDTA clearance in a subgroup of patients.

RESULTS

Forty-one patients (32%) were found to have cognitive dysfunction (clinical evaluation and/or psychometric tests). Sixteen patients (13%) found with serum creatinine levels above reference values had cognitive dysfunction more frequently than patients with creatinine within the normal range (69% versus 31%; p = 0.001), but did not differ in aetiology or severity of cirrhosis (p >0.1). Patients with loop diuretics versus without did not differ in creatinine values (p >0.1). Multivariate analysis showed that cognitive dysfunction was related to hospital admission at inclusion in the study, international normalized ratio and serum creatinine (p <0.05 for all), but not to potassium or sodium levels. Plasma ammonia concentration was related to serum creatinine (r = 0.26, p = 0.004) and the glomerular filtration rate (r = -0.44, p = 0.023).

CONCLUSIONS

Renal dysfunction seems to be related to cognitive impairment in patients with liver cirrhosis and might be implicated in the pathogenesis of hepatic encephalopathy.

A novel prognostic model based on serum levels of total bilirubin and creatinine early after liver transplantation

BACKGROUND/AIM

We aim to evaluate the impact of early renal dysfunction (ERD), early allograft dysfunction (EAD) on post-transplant mortality, and further explore a simple and accurate model to predict prognosis.

PATIENTS

A total of 161 adult patients who underwent liver transplantation for benign end-stage liver diseases were enrolled in the retrospective study. Another 38 patients were used for model validation.

RESULTS

Poor patient survival was associated with ERD or EAD. A post-transplant model for predicting mortality (PMPM) based on serum levels of total bilirubin and creatinine at 24-h post-transplantation was then established according to multivariate logistic regression. At 3 months, 6 months and 1 year, the area under receiver operating characteristic curves (AUC) of PMPM score at 24-h post-transplantation (0.876, 0.878 and 0.849, respectively) were significantly higher than those of pre-transplant model for end-stage liver diseases (MELD) score (0.673, 0.674 and 0.618, respectively) or the post-transplant MELD score at 24-h post-transplantation (0.787, 0.787 and 0.781, respectively) (P<0.05). Patients with PMPM score <or=-1.4 (low-risk group, n=114) achieved better survival than those with PMPM score >-1.4 (high-risk group, n=47) (P<0.001). The patients in the high-risk group showed a relatively good outcome if their PMPM scores decreased to <or=-1.4 at post-transplant day 7. The subsequent validation study showed that PMPM functioned with a predictive accuracy of 100%.

CONCLUSION

The PMPM score could effectively predict short- and medium-term mortality in liver transplant recipients.

Serum creatinine in patients with advanced liver disease is of limited value for identification of moderate renal dysfunction: are the equations for estimating renal function better?

BACKGROUND

The Cockcroft-Gault formula (CGF) is used to estimate the glomerular filtration rate (GFR) based on serum creatinine (Cr) levels, age and sex. A new formula developed by the Modification of Diet in Renal Disease (MDRD) Study Group, based on the patient's Cr levels, age, sex, race and serum urea nitrogen and serum albumin levels, has shown to be more accurate. However, the best formula to identify patients with advanced liver disease (ALD) and moderate renal dysfunction (GFR 60 mL/min/1.73 m2 or less) is not known. The aim of the present study was to compare calculations of GFR, using published formulas (excluding those requiring urine collections) with standard radionuclide measurement of GFR in patients with ALD.

METHODS

Fifty-seven consecutive subjects (40% women) with a mean age of 50 years (range 16 to 67 years) underwent 99m-technetium-diethylenetriamine pentaacetic acid (99mTc-DTPA) (single injection) radionuclide measurement of GFR. To calculate GFR, three formulas were used: the reciprocal of Cr multiplied by 100 (100/Cr), the CGF and the MDRD formulas. Pearson's correlation coefficient (r) and Bland-Altman analyses of agreement were used to analyze the association between 99mTc-DTPA clearance and the three equations for GFR.

RESULTS

The mean 99mTc-DTPA clearance was 83 mL/min/1.73 m2 (range 28 mL/min/1.73 m2 to 173 mL/min/1.73 m2). Mean calculated GFRs by 100/Cr, the CGF and the MDRD formula were 106 mL/min/1.73 m2, 98 mL/min/1.73 m2 and 86 mL/min/1.73 m2, respectively. Regression analysis showed good correlation between radionuclide GFR and calculated GFR with r(100/Cr)=0.74, r(CGF)=0.80, r(MDRD)=0.87, all at P > or = 0.0001. The MDRD formula provided the least bias. The Bland-Altman plot showed best agreement between GFR calculated by the MDRD formula and 99mTc-DTPA clearance, with only 3 mL/min/1.73 m2 overestimation. There was higher variability between radionuclide GFR and calculated GFR by the CGF and by 100/Cr. Although there was no difference in precision, GFR calculated by the MDRD formula had the best overall accuracy. The sensitivity and specificity for detection of moderate renal dysfunction by the MDRD formulas were 73% and 87%, respectively.

CONCLUSIONS

Among the Cr-based GFR formulas, the MDRD formula showed a larger proportion of agreement with radionuclide GFR in patients with ALD. In clinical practice, the MDRD is the best formula for detection of moderate renal dysfunction among those with ALD.