Predicting technique survival in peritoneal dialysis patients: comparing artificial neural networks and logistic regression

CliqueFluxNet: Unveiling EHR Insights with Stochastic Edge Fluxing and Maximal Clique Utilisation Using Graph Neural Networks

Electronic Health Records (EHRs) play a crucial role in shaping predictive are models, yet they encounter challenges such as significant data gaps and class imbalances. Traditional Graph Neural Network (GNN) approaches have limitations in fully leveraging neighbourhood data or demanding intensive computational requirements for regularisation. To address this challenge, we introduce CliqueFluxNet, a novel framework that innovatively constructs a patient similarity graph to maximise cliques, thereby highlighting strong inter-patient connections. At the heart of CliqueFluxNet lies its stochastic edge fluxing strategy - a dynamic process involving random edge addition and removal during training. This strategy aims to enhance the model's generalisability and mitigate overfitting. Our empirical analysis, conducted on MIMIC-III and eICU datasets, focuses on the tasks of mortality and readmission prediction. It demonstrates significant progress in representation learning, particularly in scenarios with limited data availability. Qualitative assessments further underscore CliqueFluxNet's effectiveness in extracting meaningful EHR representations, solidifying its potential for advancing GNN applications in healthcare analytics.

Artificial intelligence and machine learning in peritoneal dialysis: a systematic review of clinical outcomes and predictive modeling

BACKGROUND

The integration of artificial intelligence (AI) and machine learning (ML) in peritoneal dialysis (PD) presents transformative opportunities for optimizing treatment outcomes and informing clinical decision-making. This study aims to provide a comprehensive overview of the applications of AI/ML techniques in PD, focusing on their potential to predict clinical outcomes and enhance patient care.

MATERIALS AND METHODS

This systematic review was conducted according to PRISMA guidelines (2020), searching key databases for articles on AI and ML applications in PD. The inclusion criteria were stringent, ensuring the selection of high-quality studies. The search strategy comprised MeSH terms and keywords related to PD, AI, and ML. 793 articles were identified, with nine ultimately meeting the inclusion criteria. The review utilized a narrative synthesis approach to summarize findings due to anticipated study heterogeneity.

RESULTS

Nine studies met the inclusion criteria. The studies varied in sample size and employed diverse AI and ML techniques, reflecting the breadth of data considered. Mortality prediction emerged as a recurrent theme, demonstrating the significance of AI and ML in prognostic accuracy. Predictive modeling extended to technique failure, hospital stay prediction, and pathogen-specific immune responses, showcasing the versatility of AI and ML applications in PD.

CONCLUSIONS

This systematic review highlights the diverse applications of AI/ML in peritoneal dialysis, demonstrating their potential to enhance predictive accuracy, risk stratification, and decision support. However, limitations such as small sample sizes, single-center studies, and potential biases warrant further research and external validation. Future perspectives include integrating these AI/ML models into routine clinical practice and exploring additional use cases to improve patient outcomes and healthcare decision-making in PD.

Predicting transfer to haemodialysis using the peritoneal dialysis surprise question

BACKGROUND

People on peritoneal dialysis (PD) at risk of transfer to haemodialysis (HD) need support to remain on PD or ensure a safe transition to HD. Simple point-of-care risk stratification tools are needed to direct limited dialysis centre resources. In this study, we evaluated the utility of collecting clinicians' identification of patients at high risk of transfer to HD using a single point of care question.

METHODS

In this prospective observational study, we included 1275 patients undergoing PD in 35 home dialysis programmes. We modified the palliative care 'surprise question' (SQ) by asking the registered nurse and treating nephrologist: 'Would you be surprised if this patient transferred to HD in the next six months?' A 'yes' or 'no' answer indicated low and high risk, respectively. We subsequently followed patient outcomes for 6 months. Cox regression model estimated the hazard ratio (HR) of transfer to HD.

RESULTS

Patients' mean age was 59 ± 16 years, 41% were female and the median PD vintage was 20 months (interquartile range: 9-40). Responses were received from nurses for 1123 patients, indicating 169 (15%) as high risk and 954 (85%) as low risk. Over the next 6 months, transfer to HD occurred in 18 (11%) versus 29 (3%) of the high and low-risk groups, respectively (HR: 3.92, 95% confidence interval (CI): 2.17-7.05). Nephrologist responses were obtained for 692 patients, with 118 (17%) and 574 (83%) identified as high and low risk, respectively. Transfer to HD was observed in 14 (12%) of the high-risk group and 14 (2%) of the low-risk group (HR: 5.56, 95% CI: 2.65-11.67). Patients in the high-risk group experienced higher rates of death and hospitalisation than low-risk patients, with peritonitis events being similar between the two groups.

CONCLUSIONS

The PDSQ is a simple point of care tool that can help identify patients at high risk of transfer to HD and other poor clinical outcomes.

Improving Diagnostics with Deep Forest Applied to Electronic Health Records

An electronic health record (EHR) is a vital high-dimensional part of medical concepts. Discovering implicit correlations in the information of this data set and the research and informative aspects can improve the treatment and management process. The challenge of concern is the data sources' limitations in finding a stable model to relate medical concepts and use these existing connections. This paper presents Patient Forest, a novel end-to-end approach for learning patient representations from tree-structured data for readmission and mortality prediction tasks. By leveraging statistical features, the proposed model is able to provide an accurate and reliable classifier for predicting readmission and mortality. Experiments on MIMIC-III and eICU datasets demonstrate Patient Forest outperforms existing machine learning models, especially when the training data are limited. Additionally, a qualitative evaluation of Patient Forest is conducted by visualising the learnt representations in 2D space using the t-SNE, which further confirms the effectiveness of the proposed model in learning EHR representations.

Artificial intelligence in peritoneal dialysis: general overview

OBJECTIVE

This article is a general overview about artificial intelligence/machine learning (AI/ML) algorithms in the domain of peritoneal dialysis (PD).

METHODS

We searched studies that used AI/ML in PD, which were classified according to the type of algorithm and PD issue.

RESULTS

Studies were divided into (a) predialytic stratification, (b) peritoneal technique issues, (c) infections, and (d) complications prediction. Most of the studies were observational and majority of them were reported after 2010.

CONCLUSIONS

There is a number of studies proved that AI/ML algorithms can predict better than conventional statistical method and even nephrologists. However, the soundness of AI/ML algorithms in PD still requires large databases and interpretation by clinical experts. In the future, we hope that AI will facilitate the management of PD patients, thus increasing the quality of life and survival.

Longitudinal Studies 5: Development of Risk Prediction Models for Patients with Chronic Disease

Chronic diseases are now the major cause of ill health in both developed and developing countries. Chronic diseases evolve, over decades, from an early reversible phase, to a late stage of irreversible organ damage. Importantly, the trajectory of individual patients with a chronic disease is highly variable. This uncertainty causes substantial stress and difficulty for patients, care providers, and health systems. Clinical risk prediction models address this uncertainty by incorporating multiple variables to more precisely estimate the risk of adverse events for an individual patient. In the current chapter, we describe the general approach to developing a risk prediction model. We then illustrate how these methods are applied in the development and validation of the kidney failure risk equation (KFRE), which accurately predicts the risk of kidney failure in patients with chronic kidney disease stages 3-5.

Development of a scoring tool for predicting prolonged length of hospital stay in peritoneal dialysis patients through data mining

Background

The hospital admission rate is high in patients treated with peritoneal dialysis (PD), and the length of stay (LOS) in the hospital is a key indicator of medical resource allocation. This study aimed to develop a scoring tool for predicting prolonged LOS (pLOS) in PD patients by combining machine learning and traditional logistic regression (LR).

Methods

This study was based on patient data collected using the Hospital Quality Monitoring System (HQMS) in China. Three machine learning methods, classification and regression tree (CART), random forest (RF), and gradient boosting decision tree (GBDT), were used to develop models to predict pLOS, which is longer than the average LOS, in PD patients. The model with the best prediction performance was used to identify predictive factors contributing to the outcome. A multivariate LR model based on the identified predictors was then built to derive the score assigned to each predictor. Finally, a scoring tool was developed, and it was tested by stratifying PD patients into different pLOS risk groups.

Results

A total of 22,859 PD patients were included in our study, with 25.2% having pLOS. Among the three machine learning models, the RF model achieved the best prediction performance and thus was used to identify the 10 most predictive variables for building the scoring system. The multivariate LR model based on the identified predictors showed good discrimination power with an AUROC of 0.721 in the test dataset, and its coefficients were used as a basis for scoring tool development. On the basis of the developed scoring tool, PD patients were divided into three groups: low risk (≤5), median risk [5–10], and high risk (>10). The observed pLOS proportions in the low-risk, median-risk, and high-risk groups in the test dataset were 11.4%, 29.5%, and 54.7%, respectively.

Conclusions

This study developed a scoring tool to predict pLOS in PD patients. The scoring tool can effectively discriminate patients with different pLOS risks and be easily implemented in clinical practice. The pLOS scoring tool has a great potential to help physicians allocate medical resources optimally and achieve improved clinical outcomes.

An Equation Based on Fuzzy Mathematics to Assess the Timing of Haemodialysis Initiation

In order to develop an equation that integrates multiple clinical factors including signs and symptoms associated with uraemia to assess the initiation of dialysis, we conducted a retrospective cohort study including 25 haemodialysis centres in Mainland China. Patients with ESRD (n = 1281) who commenced haemodialysis from 2008 to 2011 were enrolled in the development cohort, whereas 504 patients who began haemodialysis between 2012 and 2013 were enrolled in the validation cohort comprised. An artificial neural network model was used to select variables, and a fuzzy neural network model was then constructed using factors affecting haemodialysis initiation as input variables and 3-year survival as the output variable. A logistic model was set up using the same variables. The equation’s performance was compared with that of the logistic model and conventional eGFR-based assessment. The area under the bootstrap-corrected receiver-operating characteristic curve of the equation was 0.70, and that of two conventional eGFR-based assessments were 0.57 and 0.54. In conclusion, the new equation based on Fuzzy mathematics, covering laboratory and clinical variables, is more suitable for assessing the timing of dialysis initiation in a Chinese ESRD population than eGFR, and may be a helpful tool to quantitatively evaluate the initiation of haemodialysis.

Artificial Neural Network: A Method for Prediction of Surgery-Related Pressure Injury in Cardiovascular Surgical Patients

PURPOSE

The aim of this study was to build an artificial neural network (ANN) model for predicting surgery-related pressure injury (SRPI) in cardiovascular surgical patients.

DESIGN

Prospective cohort study.

SUBJECTS AND SETTING

One hundred forty-nine patients who had cardiovascular surgery were included in the study. This study was conducted in a 1000-bed teaching hospital in Eastern China where 250 to 350 cardiac surgeries are performed each year.

METHODS

We performed a prospective cohort study among consecutive patients undergoing cardiovascular surgery between January and December 2015. The ANN model was built based on possible SRPI risk factors. The model performance was tested by a receiver operating characteristic curve and the C-index. A C-index from 0.5 to 0.7 is classified as having low accuracy, 0.7 to 0.9 as having moderate accuracy, and 0.9 to 1.0 as having high accuracy. We also compared the actual SRPI incidences based on the ANN stratification.

RESULTS

Thirty-seven of 147 patients developed SRPIs, yielding an incidence rate of 24.8% (95% CI, 18.1-32.6). The C-index was 0.815, which showed the ANN model had a moderate prediction value for SRPI. According to the ANN model, the SRPI predicting incidence ranged from 6.4% to 67.7%. Surgery-related pressure injury incidences were significantly different among 3 risk groups stratified by the ANN (P < .05).

CONCLUSION

We established an ANN model that provides moderate prediction of SRPI in patients undergoing cardiovascular surgical procedures. Identification and additional associated factors should be incorporated into the ANN model to increase its predictive ability.

The Peritoneal Dialysis Outcomes and Practice Patterns Study (PDOPPS): Unifying Efforts to Inform Practice and Improve Global Outcomes in Peritoneal Dialysis

♦

BACKGROUND

Extending technique survival on peritoneal dialysis (PD) remains a major challenge in optimizing outcomes for PD patients while increasing PD utilization. The primary objective of the Peritoneal Dialysis Outcomes and Practice Patterns Study (PDOPPS) is to identify modifiable practices associated with improvements in PD technique and patient survival. In collaboration with the International Society for Peritoneal Dialysis (ISPD), PDOPPS seeks to standardize PD-related data definitions and provide a forum for effective international collaborative clinical research in PD. ♦

METHODS

The PDOPPS is an international prospective cohort study in Australia, Canada, Japan, the United Kingdom (UK), and the United States (US). Each country is enrolling a random sample of incident and prevalent patients from national samples of 20 to 80 sites with at least 20 patients on PD. Enrolled patients will be followed over an initial 3-year study period. Demographic, comorbidity, and treatment-related variables, and patient-reported data, will be collected over the study course. The primary outcome will be all-cause PD technique failure or death; other outcomes will include cause-specific technique failure, hospitalizations, and patient-reported outcomes. ♦

RESULTS

A high proportion of the targeted number of study sites has been recruited to date in each country. Several ancillary studies have been funded with high momentum toward expansion to new countries and additional participation. ♦

CONCLUSION

The PDOPPS is the first large, international study to follow PD patients longitudinally to capture clinical practice. With data collected, the study will serve as an invaluable resource and research platform for the international PD community, and provide a means to understand variation in PD practices and outcomes, to identify optimal practices, and to ultimately improve outcomes for PD patients.

Simultaneous Catheter Replacement for Infectious and Mechanical Complications Without Interruption of Peritoneal Dialysis

♦

BACKGROUND

Conventional management for peritoneal dialysis (PD)-related infectious and mechanical complications that fails treatment includes catheter removal and hemodialysis (HD) via a central venous catheter with the end result that the majority of patients will not return to PD. Simultaneous catheter replacement (SCR) can retain patients on PD by avoiding the scenario of staged removal and reinsertion of catheters. The aim of this study was to evaluate a protocol for SCR without interruption of PD. ♦

METHODS

Clinical outcomes were analyzed for 55 consecutive SCRs performed from 2002 through 2012 and followed through 2013. ♦

RESULTS

Simultaneous catheter replacements were performed for 28 cases of relapsing peritonitis, 12 cases of tunnel infection, and 15 cases of mechanical catheter complications. All cases for peritonitis and tunnel infection and 80% for mechanical complications continued PD on the day of surgery using a low-volume, intermittent automated PD protocol. Systemic antibiotics were continued for 2 weeks postoperatively (up to 4 weeks for Pseudomonas). Simultaneous catheter replacement was performed as an outpatient procedure in 89.1% of cases. Only 1 of 55 procedures was complicated by peritonitis within 8 weeks. No catheter losses occurred during this postoperative timeframe. Long-term, SCR enabled a median technique survival of 5.1 years. ♦

CONCLUSIONS

In most instances, SCR can be safely performed without interruption of PD for selected cases of peritonitis and tunnel infection and for mechanical catheter complications. The procedure spares the patient from a central venous catheter, a shift to HD, the psychological ordeal of a change in dialysis modality, and a second surgery to insert a new catheter.

Longitudinal studies 5: Development of risk prediction models for patients with chronic disease

Chronic diseases are now the major cause of ill health in both developed and developing countries. Chronic diseases evolve, over decades, from an early reversible phase, to a late stage of irreversible organ damage. Importantly, the trajectory of individual patients with a chronic disease is highly variable. This uncertainty causes substantial stress and difficulty for patients, care providers and health systems. Clinical risk prediction models address this uncertainty by incorporating multiple variables to more precisely estimate the risk of adverse events for an individual patient. In the current chapter, we describe the general approach to developing a risk prediction model. We then illustrate how these methods were applied in the development and validation of the Kidney Failure Risk Equation (KFRE), which accurately predicts the risk of kidney failure in patients with Chronic Kidney Disease Stages 3-5.

Impact of hernias on peritoneal dialysis technique survival and residual renal function

OBJECTIVE

We evaluated the effect of hernias and their surgical or conservative management on peritoneal dialysis (PD) technique survival and residual renal function.

METHODS

This 10-year single-center retrospective case-control study (January 2001 - January 2011) compared patient survival, PD technique survival, and residual renal function in patients with a history of abdominal hernias and in a control cohort matched for age and PD vintage.

RESULTS

Of 73 hernias identified in 63 patients (mean age: 55 years; 63% men), umbilical hernias were the most frequent (40%), followed by inguinal (33%), incisional, and epigastric hernias. Some hernias were surgically repaired before (n = 10) or at the time of PD catheter insertion (n = 11), but most (71%) were diagnosed and managed after initiation of PD. Overall, 49 of 73 (67%) hernias were treated surgically. In 53% of subjects, early postoperative dialysis was not needed; only 7 patients required temporary hemodialysis. The occurrence of a hernia and its treatment did not significantly affect residual renal function. After a hernia diagnosis or repair, 86% of patients were able to continue with PD. ♢

CONCLUSIONS

The incidence of abdominal hernia and hernia management in patients on PD do not significantly influence residual renal function or PD technique survival. Timely management of hernias is advisable and does not preclude continuation with PD as a dialysis modality.

Mortality predicted accuracy for hepatocellular carcinoma patients with hepatic resection using artificial neural network

The aim of this present study is firstly to compare significant predictors of mortality for hepatocellular carcinoma (HCC) patients undergoing resection between artificial neural network (ANN) and logistic regression (LR) models and secondly to evaluate the predictive accuracy of ANN and LR in different survival year estimation models. We constructed a prognostic model for 434 patients with 21 potential input variables by Cox regression model. Model performance was measured by numbers of significant predictors and predictive accuracy. The results indicated that ANN had double to triple numbers of significant predictors at 1-, 3-, and 5-year survival models as compared with LR models. Scores of accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) of 1-, 3-, and 5-year survival estimation models using ANN were superior to those of LR in all the training sets and most of the validation sets. The study demonstrated that ANN not only had a great number of predictors of mortality variables but also provided accurate prediction, as compared with conventional methods. It is suggested that physicians consider using data mining methods as supplemental tools for clinical decision-making and prognostic evaluation.

Risk prediction in chronic kidney disease: pitfalls and caveats

PURPOSE OF REVIEW

This review aims to describe the challenges and highlight recent advances in the field of risk prediction for patients with chronic kidney disease (CKD). We first focus on methods of model development and metrics of model performance in general, and then highlight important risk prediction tools for patients with CKD, for prediction of kidney failure and all-cause mortality.

RECENT FINDINGS

Investigators have used data from patients with CKD stages 1-5 and developed models for predicting the progression to kidney failure and all-cause mortality. Models for kidney failure have included estimated glomerular filtration rate, albuminuria, demographic and laboratory variables, and have achieved excellent discrimination. In contrast, model performance for prediction of all-cause mortality has been relatively modest. No validated models exist for predicting the risk of cardiovascular events in patients with CKD.

SUMMARY

Models for predicting kidney failure in patients with CKD are highly accurate and clinically usable. The kidney failure risk equation includes routinely collected laboratory data and can predict the progression of CKD to kidney failure with accuracy. Additional validation of the risk equation and development of new models for all-cause mortality and cardiovascular events in patients with CKD are needed.

Is transition between peritoneal dialysis and hemodialysis really a gradual process?

BACKGROUND

Transfer to hemodialysis (HD) is a frequent cause of peritoneal dialysis (PD) cessation. In the present study, we set out to describe the transition period between PD and HD.

METHODS

All patients in 4 centers of Basse-Normandie who had been treated with PD for more than 90 days and who were permanently transferred to HD between 1 January 2005 and 31 December 2009 were retrospectively reviewed. The rate of unplanned HD start was evaluated.

RESULTS

In the 60 patients (39 men, 21 women) included in the study, median score on the Charlson comorbidity index at PD initiation was 5 [interquartile range (IQR): 3 - 7], median age at HD initiation was 62 years (IQR: 54 - 76 years), and median duration on PD was 22 months (IQR: 12 - 36 months). Among the 60 patients, 37 had an unplanned HD initiation. Peritonitis was the most frequent cause of unplanned HD start (n = 20), and dialysis inadequacy (n = 11), the main cause of planned HD start. During the transition period, all patients were hospitalized. Median duration of hospitalization was 4.5 days (IQR: 0 - 25.5 days). Within 2 months after HD initiation, 9 patients died. Two months after starting HD, 6 of the remaining 51 patients were being treated in a self-care HD unit and only 23 patients had a mature fistula.

CONCLUSIONS

Unplanned HD start is a common problem in patients transferred from PD. Further studies are needed to improve the rate of planned HD start in PD patients transferred to HD.

Shirazi F. Patterns Prediction of Chemotherapy Sensitivity in Cancer Cell lines Using FTIR Spectrum, Neural Network and Principal Components Analysis

Drug resistance enables cancer cells to break away from cytotoxic effect of anticancer drugs. Identification of resistant phenotype is very important because it can lead to effective treatment plan. There is an interest in developing classifying models of resistance phenotype based on the multivariate data. We have investigated a vibrational spectroscopic approach in order to characterize a sensitive human ovarian cell line, A2780, and its cisplatin-resistant derivative, A2780-cp. In this study FTIR method have been evaluated via the use of principal components analysis (PCA), ANN (artificial neuronal network) and LDA (linear discriminate analysis). FTIR spectroscopy on these cells in the range of 400-4000 cm(-1) showed alteration in the secondary structure of proteins and a CH stretching vibration. We have found that the ANN models correctly classified more than 95% of the cell lines, while the LDA models with the same data sets could classify 85% of cases. In the process of different ranges of spectra, the best classification of data set in the range of 1000-2000 cm(-1) was done using ANN model, while the data set between 2500-3000 cm(-1) was more correctly classified with the LDA model. PCA of the spectral data also provide a good separation for representing the variety of cell line spectra. Our work supports the promise of ANN analysis of FTIR spectrum as a supervised powerful approach and PCA as unsupervised modeling for the development of automated methods to determine the resistant phenotype of cancer classification.

Recognition Patterns Construction of Coronary Heart Disease Patients with Qi Deficiency Syndrome Based on Artificial Neural Network

Coronary heart disease (CHD), called “thoracic obstruction” in TCM, is one of the most important types of heart disease for its high incidence and mortality. The methods of syndrome studies in TCM can not be completely in accordance with these of modern medicine because of the complexity itself. In this paper, we investigated the ability of Artificial Neural Networks (ANNs) to predict CHD patients with or without qi deficiency syndrome. Predictions with Multilayer Perceptron Neural Network (MPLNN, one type of the ANNS), we obtained recognition patterns made up of eight biological parameters. The accuracy of this recognition pattern was 82.2%, and the accuracy of validation pattern was 80.0%.

Predicting mortality in incident dialysis patients: an analysis of the United Kingdom Renal Registry

BACKGROUND

The risk of death in dialysis patients is high, but varies significantly among patients. No prediction tool is used widely in current clinical practice. We aimed to predict long-term mortality in incident dialysis patients using easily obtainable variables.

STUDY DESIGN

Prospective nationwide multicenter cohort study in the United Kingdom (UK Renal Registry); models were developed using Cox proportional hazards.

SETTING & PARTICIPANTS

Patients initiating hemodialysis or peritoneal dialysis therapy in 2002-2004 who survived at least 3 months on dialysis treatment were followed up for 3 years. Analyses were restricted to participants for whom information for comorbid conditions and laboratory measurements were available (n = 5,447). The data set was divided into data sets for model development (n = 3,631; training) and validation (n = 1,816) using random selection.

PREDICTORS

Basic patient characteristics, comorbid conditions, and laboratory variables.

OUTCOMES

All-cause mortality censored for kidney transplant, recovery of kidney function, and loss to follow-up.

RESULTS

In the training data set, 1,078 patients (29.7%) died within the observation period. The final model for the training data set included patient characteristics (age, race, primary kidney disease, and treatment modality), comorbid conditions (diabetes, history of cardiovascular disease, and smoking), and laboratory variables (hemoglobin, serum albumin, creatinine, and calcium levels); reached a C statistic of 0.75 (95% CI, 0.73-0.77); and could discriminate accurately among patients with low (6%), intermediate (19%), high (33%), and very high (59%) mortality risk. The model was applied further to the validation data set and achieved a C statistic of 0.73 (95% CI, 0.71-0.76).

LIMITATIONS

Number of missing comorbidity data and lack of an external validation data set.

CONCLUSIONS

Basic patient characteristics, comorbid conditions, and laboratory variables can predict 3-year mortality in incident dialysis patients with sufficient accuracy. Identification of subgroups of patients according to mortality risk can guide future research and subsequently target treatment decisions in individual patients.

A novel use of neural network model to determine the effects of multibiomarker on early health damage among Chinese steel workers

Coke-oven workers are exposed to many kinds of pollutants that can cause health damage even lead to carcinogenesis. Therefore, it is critical to identify biomarkers that predict early health damage in these exposed individuals in molecular epidemiological studies. We applied an artificial neural network (ANN) model to the identification of such predictors in a study of coke-oven workers. The study included 330 steel-factory workers who were exposed to different levels of polycyclic aromatic hydrocarbons (PAHs) in the workplace and their levels of early health damage were determined by cytokinesis-block micronuclei (CMN), heat shock protein 70 (Hsp70) expression, benzo(a)pyrene diolepoxide-albumin adduct (BPDE-AA), and olive tail moment (OTM). The ANN model was built to predict the early health damage index, and the receiver operating characteristic (ROC) curve was used to illustrate the judged criteria and the ANN model. Trend Chi-square modeling was also performed. We found that there were 55 subjects with early health damage among 330 workers based on the multibiomarker criteria using the 95 percentile of the control group as the cut-off value, while there were 22-35 positive subjects if screening by any single biomarker. The Cochran-Armitage trend test for these findings were statistically significant (Z = 3.21, P = 0.0013). Six variables were selected to simulate the ANN model. The area under ROC (AUROC) was 0.726 ± 0.037 (P < 0.001), and the predictors included workplace, cholesterol, waistline, and others. Therefore, collective using CMN frequency, Hsp70 level, BPDE-AA level, and OTM with equal weights to make an initial screening test for early health damage in coke-oven workers is feasible and superior to any single biomarker. The determinants of the effects of multibiomarker on early health damage screening can be identified by the ANN model and ROC curve method.

Preoperative scoring systems and prognostic factors for patients with spinal metastases from hepatocellular carcinoma

STUDY DESIGN

A retrospective study had been conducted to compare the existing preoperative scoring systems and to find useful prognostic factors for patients with spinal metastases from hepatocellular carcinoma (HCC).

OBJECTIVE

To evaluate different preoperative scoring systems and prognostic factors for patients with spinal metastases from HCC.

SUMMARY OF BACKGROUND DATA

Different scoring systems for metastatic spinal tumor have been designed for prognostic evaluation. However, these scoring systems were formulated from many different types of tumors, so that their efficacy for a certain type of cancer needs to be validated. Furthermore, some serologic test results may enhance the accuracy of the scoring system.

METHODS

We conducted a retrospective study to evaluate 4 prognostic scoring systems and factors in a series of 41 cases with spinal metastases from HCC in a single center. These scoring systems include Tokuhashi revised score, Tomita score, Bauer score, and a revised van der Linden score by the authors. Serologic test items including serum albumin, aspartate aminotransferase, alanine transaminase, and lactate dehydrogenase (LDH) were also evaluated.

RESULTS

The revised Tokuhashi scoring system provided statistically significant differences in survival time between different groups (P = 0.012), while the Tomita and Bauer systems did not show statistically significant differences (P = 0.918 and P = 0.754, respectively). Significantly improved survival was found in patients with good performance status and no visceral metastases (Group C, P = 0.008) in revised van der Linden scores. Univariate and multivariate analyses showed serum albumin and LDH were independent prognostic factors for survival time.

CONCLUSION

Revised Tokuhashi scoring system is practicable and highly predictive, while serum albumin and LDH also have prognostic value in patients with spinal metastases from HCC, especially those without visceral metastases. More accurate prognosis may be obtained if the scoring systems include clinical and laboratory data in future.

Encapsulating peritoneal sclerosis: incidence, predictors, and outcomes

Encapsulating peritoneal sclerosis is a complication of peritoneal dialysis characterized by persistent, intermittent, or recurrent adhesive bowel obstruction. Here we examined the incidence, predictors, and outcomes of encapsulating peritoneal sclerosis (peritoneal fibrosis) by multivariate logistic regression in incident peritoneal dialysis patients in Australia and New Zealand. Matched case-control analysis compared the survival of patients with controls equivalent for age, gender, diabetes, and time on peritoneal dialysis. Of 7618 patients measured over a 13-year period, encapsulating peritoneal sclerosis was diagnosed in 33, giving an incidence rate of 1.8/1000 patient-years. The respective cumulative incidences of peritoneal sclerosis at 3, 5, and 8 years were 0.3, 0.8, and 3.9%. This condition was independently predicted by younger age and the duration of peritoneal dialysis, but not the rate of peritonitis. Twenty-six patients were diagnosed while still on peritoneal dialysis. Median survival following diagnosis was 4 years and not statistically different from that of 132 matched controls. Of the 18 patients who died, only 7 were attributed directly to peritoneal sclerosis. Our study shows that encapsulating peritoneal sclerosis is a rare condition, predicted by younger age and the duration of peritoneal dialysis. The risk of death is relatively low and not appreciably different from that of competing risks for mortality in matched dialysis control patients.