Band 52, Heft 5

Epigenetics refers to the study of mechanisms able to influence gene expression in a stable and potentially heritable manner without altering the DNA sequence. These mechanisms include posttranslational histone modifications, miRNA-mediated post-transcriptional regulation and DNA methylation. The accumulation of molecular errors over time resulting, at least partly, in the alteration of normal epigenetic patterns is being widely associated with aging. Epigenetic processes are also considered important mechanisms through which environmental and stochastic stressors promote numerous pathologies in humans. It is, therefore, reasonable to expect that several complex multi-factorial late-onset disorders, like osteoporosis and osteoarthritis, could have a strong epigenetic component. The focal point of all skeletal pathologies is the deregulation of bone remodeling, mediated by bone-forming osteoblasts and bone-resorbing osteoclasts. In order to keep both processes in balance, the activity, differentiation and apoptosis of both cell types have to be tightly regulated. In particular, the differentiation of osteoblasts and osteoclasts is accompanied by profound changes in gene expression. It has been shown that histone deacetylation and DNA methylation negatively regulate the expression of several genes associated with different stages of osteoblast differentiation; however, several miRNAs promote osteoblastogenesis. Furthermore, inactivating mutations in the miRNA coding regions could be associated with the pathogenesis of osteoporosis. The aim of this review is to highlight the role of epigenetic mechanisms in bone remodeling and bone homeostasis, so as to implicate their diagnostic and therapeutic potential in skeletal diseases.

Circulating fetal cell-free DNA (cffDNA) is emerging as the most reliable known target for prenatal molecular diagnostics. Different points of view are expressed in literature regarding the safe use of cffDNA for all types of molecular tests, above all those used to detect maternal DNA rather than fetal mutation. The aim of the present study was therefore to achieve consensus on guidelines conducive to standardizing current procedures, which differ between laboratories, and to design stringent technical protocols for the analysis of cffDNA.

Altered systemic iron metabolism is a key element of uremia, and functional iron deficiency mainly related to subclinical inflammation makes it difficult to maintain proper control of anemia in chronic hemodialysis patients (CHD). In the last decade, the hepatic hormone hepcidin has been progressively recognized as the master regulator of circulating iron levels through the modulation of cellular iron fluxes in response to iron stores, as well as to erythroid and inflammatory stimuli. Hepcidin is cleared by the kidney and progression of renal disease has been associated to increased serum hepcidin levels. This, in turn, reduces iron availability for erythropoiesis, suggesting anti-hepcidin strategies for improving anemia control. Moreover, hepcidin has been recently implicated in the pathogenesis of long-term complications of dialysis, like accelerated atherosclerosis. Initial studies almost invariably reported a sustained increase of serum hepcidin in chronic hemodialysis patients. Noteworthy, such studies included relatively few patients and controls that were poorly matched for major determinants of serum hepcidin at population level, i.e., age and gender. More recent data based on accurately matched larger series challenge the view that hepcidin is intrinsically increased in hemodialysis patients, showing a marked inter- and intra-individual variability of hormone levels. Here we take a critical look to the data published so far on hepcidin levels in CHD, analyze the reasons underlying the discrepancies in available studies and the hepcidin variability in CHD, and point out the need for further studies in large series of well-characterized CHD patients and controls.

D-dimer testing is currently considered a cornerstone in the diagnostic approach of patients with suspected venous thromboembolism (VTE) across different health-care settings, including the emergency department (ED). Nevertheless, inappropriate or incorrect activities developing throughout the total testing process may substantially impair the clinical usefulness of this test and delay or even challenge the fast rule out or diagnosis of VTE. The leading problem of D-dimer is represented by the poor specificity for diagnosing VTE, wherein a minority of patients with a positive D-dimer are finally diagnosed with VTE, and even more importantly, the specificity further decreases with ageing, thus contributing to increase the overcrowding in short stay units such as the ED. Due to the large heterogeneity that characterizes the use of D-dimer in the emergency room, three Italian societies of laboratory medicine (Italian Committee for Standardization of Hematology and Laboratory Methods, Italian Society of Clinical Biochemistry and Molecular Biology, and Italian Society of Laboratory Medicine), along with the Academy of Emergency Medicine and Care, have developed a consensus document about the use of D-dimer testing for diagnostics of patients with suspected VTE in this health-care setting. The evidence-based indications contained in this document will cover the leading preanalytical, analytical, and postanalytical issues that may impair the clinical efficacy of D-dimer testing in the ED.

Background: Liquid biobanking is an important tool for laboratory diagnostics in routine settings and clinical studies. However, the current knowledge about adequate storage conditions for different classes of biomarkers is incomplete and, in part, contradictory. Here, we performed a comprehensive study on the effects of different storage conditions on the stability of various biomarkers in human serum and plasma. Methods: Serum and citrated plasma were aliquoted and stored at 4 °C, –20 °C, –80 °C, and <–130 °C for 0, 7, 30, and 90 days, respectively (5–10 pools/condition). Additionally, frozen aliquots were temporarily exposed to higher temperatures during storage to simulate removing individual samples. Stability was tested for 32 biomarkers from 10 different parameter classes (electrolytes, enzymes, metabolites, inert proteins, complement factors, ketone bodies, hormones, cytokines, coagulation factors, and sterols). Results: Biobanking at –80 °C and <–130 °C for up to 90 days did not lead to substantial changes (defined as >3 interassay coefficients of variation and p<0.01) of any biomarker concentration. In contrast, storage at 4 °C and –20 °C induced substantial changes in single biomarker concentrations in most classes. Such substantial changes were increases (<20%) in electrolytes, metabolites, and proteins, and decreases (<96%) in enzymes, ketone bodies, cytokines, and coagulation factors. Biomarker stability was minimally affected by occasional short-term thermal exposure. Conclusions: Based on these results, we provide recommendations for storage conditions of up to 90 days for several biomarkers. Generally, storage at ≤–80 °C for at least 90 days including occasional short-term thermal exposure is an excellent storage condition for most biomarkers.

Background: Non-invasive prenatal test of trisomy 21 (T21) is being researched using fetal specific epigenetic biomarkers present in maternal plasma. We applied a methyl-CpG binding domain-based protein (MBD) method based on epigenetic characteristics of fetal specific-methylated regions with a high CpG density in HLCS on chromosome 21 and RASSF1A on chromosome 3 for the non-invasive detection of fetal T21 and estimated the diagnostic accuracy of the method. Methods: A nested case-control study was conducted with maternal plasma collected from 50 pregnant women carrying 40 normal and 10 T21 fetuses. A MBD method was used for enrichment of methylated DNA regions in maternal plasma. The levels of methylated HLCS (M- HLCS ) and methylated RASSF1A (M- RASSF1A ) were simultaneously measured by multiplex qPCR. Results: Levels of M- HLCS and M- RASSF1A were obtained in all cases. Levels were not different according to fetal gender (p>0.05 in both). The level of M- HLCS was significantly increased in women with a T21 fetus compared with controls (p<0.001). The level of M- RASSF1A was not different between two groups (p>0.05). In non-invasive fetal T21 detection, the specificity of M- HLCS level and the epigenetic-epigenetic ratio (EER) using M- HLCS and M- RASSF1A levels were 82.5% and 92.5%, respectively, at 90.0% sensitivity. Conclusions: Our findings suggest that the EER may be useful as a potential biomarker for the non-invasive detection of fetal T21, regardless of fetal gender. The MBD method can be used as an effective tool in the detection of methylated fetal specific markers with a high CpG density in maternal plasma.

Background: The use of dried blood spots on filter paper is well documented as an affordable and practical alternative to classical venous sampling for various clinical needs. This technique has indeed many advantages in terms of collection, biological safety, storage, and shipment. Amyloid β (Aβ) peptides are useful cerebrospinal fluid (CSF) biomarkers for Alzheimer disease diagnosis. However, Aβ determination is hindered by preanalytical difficulties in terms of sample collection and stability in tubes. Methods: We compared the quantification of Aβ peptides (1–40, 1–42, and 1–38) by simplex and multiplex ELISA, following either a standard operator method (liquid direct quantification) or after spotting CSF onto dried matrix paper card. Results: The use of dried matrix spot (DMS) overcame preanalytical problems and allowed the determination of Aβ concentrations that were highly commutable (Bland-Altman) with those obtained using CSF in classical tubes. Moreover, we found a positive and significant correlation (r 2 =0.83, Pearson coefficient p=0.0329) between the two approaches. Conclusions: This new DMS method for CSF represents an interesting alternative that increases the quality and efficiency in preanalytics. This should enable the better exploitation of Aβ analytes for Alzheimer’s diagnosis.

Background: The handling of cerebrospinal fluid (CSF) affects the biomarker quantification used to diagnose Alzheimer’s disease (AD). Only specialized centers can test for AD markers. The precise timing and freezing is required to correctly measure these biomarkers. Therefore, the effects of CSF storage temperature and repeated freeze/thaw cycles on CSF stability were investigated. Methods: Drop coating deposition Raman spectroscopy in combination with principal component analysis was used to analyze CSF and its dialyzed form (ELISA confirmed the removal of up to 80% of the AD markers). The advantage of this approach is that no prior knowledge of the biomarkers is necessary and that both the concentration and the protein structure of intact CSF are analyzed. Results: Dialyzed CSF was stable for up to 5 h after its collection, while native CSF started to denature nearly immediately. Most of the unstable proteins were denatured within 24 h. The dialyzed CSF was not affected by freeze/thaw cycles, but the native CSF exhibited significant progressive changes, even after the first freezing. The mechanism as well as the resulting structures of the freeze-denatured proteins differed from those of the temporally denatured proteins, although both protein sets began with the same initial proteins. Conclusions: CSF must be processed immediately, within 5 h of collection. Flash cooling is recommended for freezing CSF, but any freeze/thaw cycle will affect the protein component of CSF.

Background: The objective of this study was to evaluate the performance of the heavy/light chain and free light chain immunoassays in patients with heavy chain disease, and to assess the ability of the heavy/light chain assay to measure and confirm the abnormal, truncated heavy chain. Methods: Frozen serum samples from 15 γ-heavy chain disease patients were tested for IgGκ, IgGλ, total IgG, free light chains, and M-spike concentrations. Results: The (Gκ+Gλ)/IgG total ratio for these 15 patients ranged from 0.02 to 0.80. The 10 patients with IgG concentrations above 1 g/dL all had ratios below 0.3 indicating that a substantial portion of IgG was not quantitated by the Gκ and Gλ reagents. The average M-spike was 1.61 g/dL and the average calculated abnormal γ-chain concentration was 2.94 g/dL. Additionally, free light chain analysis revealed the presence of monoclonal free κ light chain in three of the 15 patients. Conclusions: This study demonstrates utility of a nephelometric assay to identify truncated immunoglobulin heavy chains in γ-HCD and that 20% of these patients also have monoclonal free light chain.

Background: Acute kidney injury (AKI) is a frequent complication of major surgery. The current study evaluated the power of two biomarkers [urinary neutrophil gelatinase-associated lipocalin (NGAL) and liver-type fatty acid binding protein (L-FABP)] to detect the occurrence of AKI and to predict the recovery from renal dysfunction in a major surgery cohort. Methods: In this prospective study, 199 patients undergoing major surgery were enrolled. Urinary samples collected from participants before surgery, and 0, 4, and 12 h and 1, 2, 7, and 14 days after surgery were tested for NGAL and L-FABP. Results: Thirty-seven (18.6%) subjects developed AKI. Urinary NGAL and L-FABP were significantly increased from the time surgery was completed (p<0.05). The peak levels of NGAL and L-FABP occurred 12 and 4 h postoperatively (16.4- and 172.0-fold compared to baseline) in AKI group, respectively. The area under the receiver operating characteristic (ROC) curve (AUC) in NGAL (at 12 h), L-FABP (at 4 h), the most predictive model (NGAL at 12 h+L-FABP at 4 h), and the best combination at the same time point (12 h) was 0.83 [95% confidence interval (CI), 0.74–0.91], 0.85 (95% CI 0.77–0.93), 0.94 (95% CI 0.89–0.98), and 0.91 (95% CI 0.85–0.97), respectively. However, the largest AUC of single and combined biomarkers for predicting non-recovery after AKI only reached 0.70. Conclusions: Urinary NGAL and L-FABP can be used to detect AKI and combining NGAL and L-FABP may improve the diagnostic performance; however, NGAL and L-FABP may be poor predictors for renal recovery after AKI.

Background: Drug testing in hair is a unique analysis in pharmacotoxicology for establishing a past repeated history of consumption or passive exposure to psychotropic substances. A rather lengthy sample treatment is usually required before parent drugs and eventual metabolites are amenable to quali-quantitative analysis. Methods: We evaluated a high throughput screening and confirmation analysis of drugs of abuse in hair by immunoassay and a validated ultra-performance liquid chromatography tandem mass spectrometry (UPLC-MS/MS) after applying a rapid digestion of the keratin matrix with VMA-T reagent before screening assay and M3 reagent before confirmatory analysis. Results: Samples digestion with VMA-T reagent and immunometric screening analysis of hair calibrators, controls and clinical samples for a total of 150 samples was completed in 4 h. No false-positive and -negative results were found for the control material. UPLC-MS/MS analysis confirmed all of the 31 adult hair samples positive to the screening test using internationally established cut-offs, and identified and quantified drugs of abuse in 32 pediatric hair samples, applying lower limits of quantification from 0.01 to 0.1 ng analyte per mg hair. Analytical recovery was between 70.9% and 100.7%. Intra- and inter-assay imprecision and inaccuracy were always lower than 10%. Conclusions: Rapid extraction, identification and quantification of drugs of abuse in hair by immunoassay and UPLC-MS/MS was tested for its feasibility in clinical samples and provided excellent results for rapid and effective drug testing in hair in epidemiological studies.

Background: Hyperhomocysteinemia (Hhcy) is considered an independent risk factor for vascular diseases and, more recently, for dementia. The methionine loading test (MLT) is useful for diagnosing additional subjects with moderate Hhcy. However, it is a complex and time-consuming procedure. A noninvasive test for the diagnosis of moderate Hhcy is desirable. Methods: The study protocol consisted of three consecutive visits. During the first visit, we performed an MLT to characterize the Hhcy status of 75 healthy adult subjects. For the breath test protocol, we selected a subsample and assigned to the control group 17 subjects with fasting and post-loading homocysteine (Hcy) ≤12 and <42.3 μmol/L, respectively, and to the Hhcy group 16 subjects with fasting Hcy ≤12 and >42.3 μmol/L after loading. Selected subjects were requested to have a second visit to perform a breath test within 1–4 weeks following the MLT test and received an oral dose of 2.5 mg/kg of 1- 13 C-methionine dissolved in water. Breath samples were collected at basal, 20, 40, 60, 80, 100 and 120 min (test 1). The same procedure was repeated within 1 week (test 2). Results: MLT was useful for diagnosing almost twice the number of individuals with Hhcy (24%) in comparison with the fasting determination alone (13.3%). The 13 C-methionine breath test reported a sensitivity of 81.3% and a specificity of 64.7% against the MLT. The coefficient of variation between breath test 1 and breath test 2 was 9.0±5.4%. Conclusions: The 13 C-methionine breath test is a valid and reliable method for identifying subjects with moderate Hhcy.

Background: In the Retic channel of DxH 800 (Beckman Coulter), the red blood cells (RBCs) resistant to hemoglobin clearing are counted as unghosted cells (UGCs). The aim of this study was to evaluate that the UGC is a surrogate marker for both the detection and counting of target cells. Methods: In total, 1181 samples including 22 from iron deficiency anemia (IDA) patients, 95 from jaundice, 2 from sickle cell anemia, 3 from thalassemia, 1 cord blood, and 269 from normal controls were analyzed. Slides were prepared from all samples except normal controls and target cells were counted for correlation analysis of target cell counts to UGCs. Results: The normal control samples showed 0.01% (0%–0.01%) UGCs, and the reference range was set at ≤0.02%. The IDA samples showed 0.015% (0.01%–0.03%) UGC count and 0.05% (0%–0.2%) target cell count. The jaundice samples showed 0.98% (0.1%–5.36%) UGC count, and 1.4% (0.1%–7.0%) target cell count. The two sickle cell anemia samples showed 0.41% and 3.74% UGC counts and 0.4% and 11.5% target cell counts. A cord blood sample showed 0.01% UGCs and 0% target cells. The three thalassemia samples showed 0.01%, 1.99%, and 7.82% UGC counts and 0%, 1.4%, and 15.5% target cell counts. The samples showing poikilocytosis other than target cells showed normal UGC count (≤0.02%). The positive predictive value of UGCs was 58.2% (124/213) and the negative predictive value was 96.8% (674/696). The UGC counts were well correlated to the manual target cell counts (r=0.944, p=0.000). Conclusions: This study demonstrates for the first time in the literature that a hematological parameter obtained automatically every time a reticulocyte counting is performed can be used to both screen for the presence of target cells and reliably quantify them.

Background: To minimise the volume of blood used for diagnostic procedures, especially in children, we investigated whether the size of sample tubes affected whole blood coagulation analyses. Methods: We included 20 healthy individuals for rotational thromboelastometry (RoTEM ® ) analyses and compared three blood sampling tubes of different size: 1.8, 2.7, and 3.6 mL. All tubes were made of plastic and contained 3.2% sodium-citrate as anticoagulant. Platelet aggregation was investigated in 12 healthy individuals employing the Multiplate ® Analyser comparing tubes of 3.6 mL and 1.8 mL. Platelet count was determined for each of the sampling tubes after 10, 60, and 120 min. Results: No significant differences were found in any of the RoTEM ® measurements between the three sizes of tubes (p=0.07–0.53). Platelet aggregation was significantly lower when using smaller tubes (p=0.0004). The platelet count remained stable using a 3.6 mL tube during the entire observation period of 120 min (p=0.74), but decreased significantly after 60 min when using tubes smaller than 3.6 mL (p<0.0001). Conclusions: RoTEM ® analyses were not affected by the size of blood sampling tubes. Therefore, 1.8 mL tubes should be preferred for RoTEM ® analyses in order to minimise the volume of blood drawn. With regard to platelet aggregation analysed by impedance aggregometry tubes of different size cannot be used interchangeably. If platelet count is determined later than 10 min after blood sampling using tubes containing citrate as anticoagulant, a 3.6 mL tube should be preferred to avoid false low values.

Background: Somatic mutation analysis is standard of practice for solid tumors in order to identify therapeutic sensitizing and resistance mutations. Our laboratory routinely performed standalone PCR-based methods for mutations in several genes. Rapid discovery and introduction of new therapeutics has demanded additional genomic information for adequate management of the cancer patient. We evaluated a next generation sequencing assay, the Ion Torrent AmpliSeq Cancer Hotspot Panelv2 (CHPv2), capable of identifying multiple somatic mutations in 50 genes in a single assay. Methods: Accuracy, precision, limit of detection, and specificity were evaluated using DNA from well-characterized cell lines, genetically engineered cell lines fixed and embedded in paraffin, and previously tested mutation positive or negative, formalin-fixed, paraffin-embedded (FFPE) tissues. Normal kidney, tonsil and colon FFPE tissues were used as controls. Results: Accuracy studies showed 100% concordance in each patient sample between previous PCR results and the corresponding variants identified using the Ion Torrent panel. Precision studies gave consistent results when libraries were prepared from the same original DNA and were run on multiple 316 chips. The limit of detection was determined to be 5% for single nucleotide variants (SNVs) and 20% for insertions and deletions (indels). Specificity studies using normal FFPE tissue previously tested by PCR methods were also 100%. Conclusions: We have evaluated the performance of the AmpliSeq Cancer Panel Hotspotv2 and show that it is suitable for clinical testing. This next generation sequencing panel has allowed the laboratory to consolidate a broader range of molecular oncology testing to a single platform and single assay.

Background: We evaluated the analytical performance of a new one-step rapid quantitative sandwich immunoassay for total prostate-specific antigen (tPSA), the FREND™ PSA Plus (FREND PSA) (NanoEnTek Inc., Seoul, Korea). Methods: The imprecision, linearity, hook effect, detection limit (LoD), and interference were evaluated and trueness verification and matrix validation were performed. For method comparison, 79 patient specimens were analyzed with FREND PSA and two comparative tPSA assays (Architect ® total PSA and cobas ® total PSA assay). Results: Total CVs of the imprecision for low (0.208 ng/mL), medium (4.051 ng/mL), and high PSA levels (5.469 ng/mL) were 15.9%, 6.4%, and 9.1%, respectively. Linearity was observed from 1.01 to 19.15 ng/mL and the hook phenomenon was absent up to 171.48 ng/mL. The LoD was 0.094 ng/mL. The regression equations between FREND (y) and Architect or cobas were as follows: y=0.0133+1.054x (r=0.973), y=–0.2144+1.066x (r=0.977), respectively. Differences between FREND PSA and the comparative methods at a medical decision level of 4.0 ng/mL were less than the optimum specification bias (9.3%). The percentage biases from the trueness verification and interference test were less than the desirable specifications for bias (18.7%). The plasma tPSA level measured with lithium heparin or K2EDTA was comparable to that in the serum. Conclusions: The FREND PSA provided reliable analytical performance and test results in comparison to two widely used tPSA assays. It is a simple and rapid test for tPSA and can be applied in point-of-care testing.

Background: The aim of the article was to systematically review published data about the relationship between prostate-specific antigen (PSA) kinetics, including PSA doubling time (PSAdt) and PSA velocity (PSAvel), and detection rate (DR) of positron emission tomography/computed tomography (PET/CT) using radiolabelled choline in restaging prostate cancer (PCa). Methods: A comprehensive literature search of studies published through July 2013 regarding the relationship between PSA kinetics and DR of radiolabelled choline PET/CT was carried out. Furthermore, a meta-analysis was performed in order to establish the DR of radiolabelled choline PET/CT using different cut-off values of PSAdt (≤ or >6 months) and PSAvel [>1 or ≤1 ng/(mL year) and >2 or ≤2 ng/(mL year)]. Moreover, a pooled analysis to establish whether PSAdt and PSAvel (using the abovementioned cut-off values) may predict positive PET/CT results was carried out. Results: Fourteen articles were selected. The pooled DR of radiolabelled choline PET/CT in restaging PCa was 58% [95% confidence interval (CI) 55–60]. Most articles reported a relationship between PSA kinetics and DR of PET/CT. Pooled DR of radiolabelled choline PET/CT increased to 65% (95% CI 58–71) when PSAdt was ≤6 months and to 71% (95% CI 66–76) and 77% (95% CI 71–82) when PSAvel was >1 or >2 ng/(mL year), respectively. PSAdt ≤6 months and PSAvel >1 or >2 ng/(mL year) proved to be relevant factors in predicting the positive result of radiolabelled choline PET/CT. Conclusions: Due to the strong relationship between PSA kinetics and DR of radiolabelled choline PET/CT, beyond PSA values, PSAdt and PSAvel should be taken into account in the selection of PCa patients who should undergo radiolabelled choline PET/CT for restaging.

Background: Ovarian cancer is a lethal gynecological malignancy largely due to the lack of biomarkers for early detection and treatment options. Opioid binding protein/cell adhesion molecule-like gene ( OPCML ) has a tumor-suppressor function in ovarian cancer, and epigenetic inactivation of OPCML induces oncogenic transformation of human ovarian surface epithelial cells. Methods: This study investigated OPCML promoter methylation levels in ovarian cancer tissues. A total of 30 normal ovarian, 85 benign ovarian tumor, and 102 ovarian cancer tissues were subjected to quantitative methylation-specific PCR analysis of OPCML methylation. Four ovarian cancer cell lines were cultured and treated with the DNA demethylation agent 5-aza-2′-deoxycytidine (5-AZA) for restoring OPCML expression. Results: The data showed that 80 of 102 (78.4%) ovarian cancer tissues and 28 of 85 (32.9%) benign ovarian tumors had a methylated OPCML gene promoter. In contrast, there was no OPCML gene promoter methylation in any of the 30 normal ovarian samples. OPCML promoter methylation was significantly associated with an older age of the patients (p=0.022), an advanced pathological stage of ovarian cancer (p=0.023), and poor overall survival of ovarian cancer patients (p<0.001). Multivariate analysis data showed that pathological stage, age, and OPCML promoter methylation were all independent factors to predict overall survival of patients. Furthermore, 5-AZA was able to restore expression of OPCML mRNA and protein in ovarian cancer cell lines. Conclusions: These data indicate that detection of OPCML gene promoter methylation could be a useful biomarker for predicting the prognosis of ovarian cancer patients and disease progression.

Background: Previous studies identified an independent relationship between red blood cell distribution width (RDW) and prognosis in patients with pulmonary hypertension of mixed etiologies and idiopathic pulmonary arterial hypertension. This study aimed to investigate the significance of RDW for predicting survival in patients with Eisenmenger syndrome (ES). Methods: We retrospectively reviewed the clinical records and collected baseline data for patients newly diagnosed with ES in our hospital between January 2005 and October 2009. Follow-up data were collected periodically using a specifically designed network database until December 31, 2012. The end point was all-cause death. Results: A total of 109 patients with ES were included in the study. Twenty-one patients (19.3%) died during a median follow-up period of 4.2 years (interquartile range 3.7–5.0 years). Baseline RDW was significantly correlated with mixed venous oxygen saturation (r=−0.286, p=0.003), arterial oxygen saturation (r=−0.423, p<0.001), mean pulmonary arterial pressure (r=0.271, p=0.004) and total pulmonary resistance (r=0.465, p<0.001). The 1-, 3- and 5-year survival rates for all 109 patients were 94%, 87% and 78%, respectively. Kaplan-Meier analysis showed that patients with RDW ≥13.9% had a lower survival rate than patients with RDW <13.9% (p=0.001). Multivariate Cox regression analysis showed that RDW was an independent prognostic marker in ES, with a hazard ratio of 1.162 (95% CI 1.036–1.302; p=0.010). Conclusions: Baseline RDW correlates with hemodynamics and is an independent prognostic marker in ES.

Background: Diabetes increases the risk and severity of atherosclerosis. Adropin, a metabolic homeostasis-related protein, has been implicated in the maintenance of metabolic homeostasis. We examined the relationship between serum adropin level and angiographic severity of coronary atherosclerosis in diabetic and non-diabetic patients. Methods: A total of 392 patients with suspected coronary artery disease, who underwent coronary angiography, were assigned into the type 2 diabetic and non-diabetic groups and also classified into four groups according to the quartiles of adropin level. Venous serum samples were collected for adropin measurement by enzyme-linked immunosorbent assay and for biochemistry assay. The angiographic severity of coronary atherosclerosis was assessed by Gensini, Friesinger, and SYNTAX scores. Results: Compared with non-diabetic patients, diabetic patients had lower serum adropin level and higher Gensini, Friesinger and SYNTAX scores (all p<0.001). Serum adropin level was inversely correlated with the Gensini, Friesinger and SYNTAX scores (r s =–0.389, –0.390 and –0.386, respectively, all p<0.001) among all patients. Low adropin level was an independent predictor of clinically relevant coronary atherosclerosis (SYNTAX score >11), both in diabetic patients [odds ratio (OR) 0.66, 95% confidence interval (CI) 0.53–0.83; p<0.001] and in non-diabetic patients (OR 0.51, 95% CI 0.35–0.74; p<0.001). Conclusions: Serum adropin level was significantly lower in type 2 diabetic patients than in non-diabetic patients and was inversely and independently associated with angiographic severity of coronary atherosclerosis, suggesting that serum adropin serves as a novel predictor of coronary atherosclerosis.