HighlightsThe %pshreg SAS macro fits Fine-Gray models for competing risks.The macro first modifies a given data set and then uses PROC PHREG for analysis.Many useful features of PROC PHREG can now be applied to a Fine-Gray model.Time-dependent effects can be accommodated by time-by-covariate interactions.For small data sets, the Firth correction is available.

Statistical models are simple mathematical rules derived from empirical data describing the association between an outcome and several explanatory variables. In a typical modeling situation statistical analysis often involves a large number of potential explanatory variables and frequently only partial subject-matter knowledge is available. Therefore, selecting the most suitable variables for a model in an objective and practical manner is usually a non-trivial task. We briefly revisit the purposeful variable selection procedure suggested by Hosmer and Lemeshow which combines significance and change-in-estimate criteria for variable selection and critically discuss the change-in-estimate criterion. We show that using a significance-based threshold for the change-in-estimate criterion reduces to a simple significance-based selection of variables, as if the change-in-estimate criterion is not considered at all. Various extensions to the purposeful variable selection procedure are suggested. We propose to use backward elimination augmented with a standardized change-in-estimate criterion on the quantity of interest usually reported and interpreted in a model for variable selection. Augmented backward elimination has been implemented in a SAS macro for linear, logistic and Cox proportional hazards regression. The algorithm and its implementation were evaluated by means of a simulation study. Augmented backward elimination tends to select larger models than backward elimination and approximates the unselected model up to negligible differences in point estimates of the regression coefficients. On average, regression coefficients obtained after applying augmented backward elimination were less biased relative to the coefficients of correctly specified models than after backward elimination. In summary, we propose augmented backward elimination as a reproducible variable selection algorithm that gives the analyst more flexibility in adopting model selection to a specific statistical modeling situation.

Increased sclerostin serum levels in CKD patients are not due to decreased renal elimination. On the contrary, renal elimination increases with declining kidney function. Whether this has consequences on antisclerostin antibody dosing, efficacy, or safety in patients with CKD remains to be determined.

Background Recent studies show associations between inorganic phosphate and risk of heart failure in the general population as well as between fibroblast growth factor 23 (FGF-23) and outcome in coronary heart disease. This study was carried out to assess whether circulating levels of inorganic phosphate and FGF-23, a new central hormone in mineral bone metabolism, predict outcome in systolic heart failure.

AimsPrevious risk assessment scores for patients with coronary artery disease (CAD) have focused on primary prevention and patients with acute coronary syndrome. However, especially in stable CAD patients improved long-term risk prediction is crucial to efficiently apply measures of secondary prevention. We aimed to create a clinically applicable mortality prediction score for stable CAD patients based on routinely determined laboratory biomarkers and clinical determinants of secondary prevention. Methods and resultsWe prospectively included 547 patients with stable CAD and a median follow-up of 11.3 years. Independent risk factors were selected using bootstrapping based on Cox regression analysis. Age, left ventricular function, serum cholinesterase, creatinine, heart rate, and HbA1c were selected as significant mortality predictors for the final multivariable model. The Vienna and Ludwigshafen Coronary Artery Disease (VILCAD) risk score based on the aforementioned variables demonstrated an excellent discriminatory power for 10-year survival with a C-statistic of 0.77 (P , 0.001), which was significantly better than an established risk score based on conventional cardiovascular risk factors (C-statistic ¼ 0.61, P , 0.001). Net reclassification confirmed a significant improvement in individual risk prediction by 34.8% (95% confidence interval: 21.7-48.0%) compared with the conventional risk score (P , 0.001). External validation of the risk score in 1275 participants of the Ludwigshafen Risk and Cardiovascular Health study (median follow-up of 9.8 years) achieved similar results (C-statistic ¼ 0.73, P , 0.001). ConclusionThe VILCAD score based on a routinely available set of risk factors, measures of cardiac function, and comorbidities outperforms established risk prediction algorithms and might improve the identification of high-risk patients for a more intensive treatment.--

BackgroundIncreasing evidence is linking fluid intake, vasopressin suppression and osmotic control with chronic kidney disease progression. Interestingly, the association between urine volume, urine osmolarity and risk of dialysis initiation has not been studied in chronic kidney disease patients before.ObjectiveTo study the relationship between urine volume, urine osmolarity and the risk of initiating dialysis in chronic kidney disease.DesignIn a retrospective cohort analysis of 273 patients with chronic kidney disease stage 1–4 we assessed the association between urine volume, urine osmolarity and the risk of dialysis by a multivariate proportional sub-distribution hazards model for competing risk data according to Fine and Gray. Co-variables were selected via the purposeful selection algorithm.ResultsDialysis was reached in 105 patients over a median follow-up period of 92 months. After adjustment for age, baseline creatinine clearance, other risk factors and diuretics, a higher risk for initiation of dialysis was found in patients with higher urine osmolarity. The adjusted sub-distribution hazard ratio for initiation of dialysis was 2.04 (95% confidence interval, 1.06 to 3.92) for each doubling of urine osmolarity. After 72 months, the estimated adjusted cumulative incidence probabilities of dialysis were 15%, 24%, and 34% in patients with a baseline urine osmolarity of 315, 510, and 775 mosm/L, respectively.ConclusionsWe conclude that higher urine osmolarity is associated with a higher risk of initiating dialysis. As urine osmolarity is a potentially modifiable risk factor, it thus deserves further, prospective research as a potential target in chronic kidney disease progression.

BackgroundOvert chronic metabolic acidosis in patients with chronic kidney disease develops after a drop of glomerular filtration rate to less than approximately 25 mL/min/1.73 m2. The pathogenic mechanism seems to be a lack of tubular bicarbonate production, which in healthy individuals neutralizes the acid net production. As shown in several animal and human studies the acidotic milieu alters bone and vitamin D metabolism, induces muscle wasting, and impairs albumin synthesis, aside from a direct alteration of renal tissue by increasing angiotensin II, aldosteron and endothelin kidney levels. Subsequent studies testing various therapeutic approaches in very selected study populations showed that oral supplementation of the lacking bicarbonate halts progression of decline of renal function. However, due to methodological limitations of these studies further investigations are of urgent need to ensure the validity of this therapeutic concept.Methods/DesignThe SoBic-study is a single-center, randomized, controlled, open-label clinical phase IV study performed at the nephrological outpatient service of the Medical University of Vienna. Two-hundred patients classified to CKD stage 3 or 4 with two separate measurements of HCO3- of <21 mmol/L will be 1:1 randomized to either receive a high dose of oral sodium bicarbonate with a serum target HCO3- level of 24 ± 1 mmol/L or receive a rescue therapy of sodium bicarbonate with a serum target level of 20 ± 1 mmol/L. The follow up will be for two years. The primary outcome is the effect of sodium bicarbonate supplementation on renal function measured by means of estimated glomerular filtration rates (4-variable-MDRD-equation) after two years. Secondary outcomes are change in markers of bone metabolism between groups, death rates between groups, and the number of subjects proceeding to renal replacement therapy across groups. Adverse events, such as worsening of arterial hypertension due to the additional sodium consumption, will be accurately monitored.DiscussionWe hypothesize that sufficiently balanced acid–base homeostasis leads to a reduction of decline of renal function in patients with chronic kidney disease. The concept of an exogenous bicarbonate supplementation to substitute the lacking endogenous bicarbonate has existed for a long time, but has never been investigated sufficiently to state clear treatment guidelines.Trial registrationEUDRACT Number: 2012-001824-36

AimsTo compare the predictive value of estimated renal function calculated by the Chronic Kidney Disease Epidemiology Collaboration (eGFR CKD-EPI ), four-variable Modification of Diet in Renal Disease (eGFR , and CockcroftGault [estimated creatinine clearance (eCcr)] equation in terms of all-cause mortality in heart failure. Renal function is an important prognostic factor in heart failure. Established methods of estimating renal function are known to under-/overestimate true function in certain settings. Methods and resultsA total of 800 systolic heart failure outpatients (mean age 57 + 11.5 years, 82% male) were studied over a median follow-up of 121 (Q1 -Q3: 110 -130) months. The highest systematic difference was seen between eCcr and eGFR [+12.33 points (mean), 95% limits of agreement -22.35 to 47.01; generalized kappa ¼ 0.36]. eGFR and eGFR CKD-EPI were the most similar [-4.16 points (mean), 95% limits of agreement -11.56 to 3.25; generalized kappa ¼ 0.74]. Up to 35.4% of patients were reclassified into different estimated glomerular filtration rate (eGFR) categories when comparing eGFR CKD-EPI with eCcr and eGFR . eGFR CKD-EPI performed marginally better in terms of predicting all-cause mortality than eGFR , as univariate areas under the time-dependent receiver operating characteristic curves (AUC), marginal and partial proportions of explained variation (PEV), net reclassification improvement (NRI), and the integrated discrimination improvement (IDI) for 5 years of follow-up were significantly higher for eGFR CKD-EPI than for eGFR . ConclusionIn this cohort of heart failure patients, eGFR CKD-EPI was marginally better in predicting all-cause mortality than eGFR . Estimated function differed widely between equations and is likely to have an effect on therapy choice.--