Estimation Of Cumulative Incidence Function Research Articles

Radiofrequency ablation of hepatocellular carcinoma guided by real-time physics-based ablation simulation: a prospective study

Objectives To assess the safety and efficacy of radiofrequency ablation (RFA) guidance software that incorporated patient-specific physics-based simulation of each ablation volume. Materials and methods Patients referred for curative ablation of hepatocellular carcinoma (HCC) of 2–5 cm diameter were prospectively enrolled. RFA was performed under general anesthesia. Procedure planning and intraprocedural modifications were guided by computer simulation of each ablation. The segmented target (tumor with 5 mm margin) was registered to and superimposed on subsequent 3D multiplanar images. The applied RF energy was used to calculate a simulated ablation volume which was displayed relative to the electrode and segmented target, to depict any untreated target tissue. After each additional ablation, the software updated the accumulated simulated ablation volume in relation to the target. The primary endpoints were technical efficacy and rate of local tumor progression (LTP). Results Sixty-eight tumors were ablated during 57 procedures in 52 patients (68.3 ± 9.2 years old, 78.8% male); 15 (26.3%) had multiple lesions and 23 (39.1%) had prior HCC treatment. The mean tumor diameter was 2.73 (±0.64) cm. The intraprocedural simulation directed additional overlapping ablations in 75.9% of tumors. Technical success and efficacy were 100% at 3-month contrast enhanced CT or MRI follow-up after the single treatment session. Cumulative incidence function estimates for 1- and 2-year LTP were 3.9% and 20.2%, respectively. Conclusion This prospective study found computer-assisted guidance that simulated each ablation was both safe and efficacious. The low rate of LTP was similar to studies that employed stereotactic guidance and ablation confirmation, without requiring a second contrast enhanced study.

Robust Bayesian estimation of cumulative incidence function for competing risk data with missing causes

In competing risk problems, an identifiable quantity of interest is cumulative incidence function which is the probability of occurrence of a cause-specific event on or before a given point in time in presence of other competing events. For parametric Bayesian estimation of cumulative incidence function, a prior for the concerned parameter may be selected by seeking expert advice or can be elicited using information available in the form of past data. However, if there is a possibility of error in elicitation of an adequate prior for the parameter of concerned risk, one can obtain robust results by selecting a prior in a data-dependent fashion. In this paper, we discuss the elicitation of data-dependent prior in the presence of competing risk data with missing causes and obtain robust Bayes estimates of cumulative incidence function by employing Weibull cause-specific hazard. We provide procedure to construct Weibull cause-specific hazard plot for considered data. We present numerical illustrations using simulated as well as real data sets.

Outcomes of carotid artery stenting in patients with radiation arteritis compared with those with atherosclerotic disease

ObjectiveHead and neck malignancies are often treated with radiotherapy (RT). Nearly 80% of patients who have undergone RT will develop carotid radiation arteritis to some degree and 29% will develop stenosis >50%. Surgery in a radiated neck has higher rates of complications, and carotid artery stenting (CAS) has become the primary therapy. The outcomes for CAS in patients with radiation arteritis have not been rigorously evaluated. The objective of the present study was to evaluate the differences in perioperative outcomes, restenosis rates, the need for reintervention, and freedom from mortality between RT patients and patients with atherosclerotic disease who had undergone CAS. MethodsThe national Vascular Quality Initiative CAS dataset from 2016 to 2019 comprised the sample for analyses (n = 7343). The primary independent variable was previous head and/or neck RT. The primary endpoint was the interval to mortality. The secondary endpoints were the cumulative incidence of restenosis (>50% and >70% by duplex ultrasound) and reintervention. We also examined the following secondary perioperative endpoints: myocardial infarction, in-hospital mortality (death before discharge), neurologic events, ipsilateral stroke, and contralateral stroke. Kaplan-Meier and multivariable Cox proportional hazard models were used to assess for mortality, and cumulative incidence function estimates were used for the nonfatal endpoints. ResultsOf the 7218 patients, 1199 (17%) had undergone prior RT. We found a significant difference in the 3-year estimates of mortality for those with and without prior RT (9.4% and 7.5%, respectively; P = .03). Furthermore, on adjusted analysis, we observed a 58% increase in the risk of mortality for those with prior RT (adjusted hazard ratio, 1.58; 95% confidence interval, 1.13-2.21). We did not observe any differences in the risk of perioperative complications (myocardial infarction, in-hospital mortality, ipsilateral or contralateral stroke), restenosis (>50% or >70%), or reintervention for the prior RT group compared with those without RT. ConclusionsThe CAS patients with RT had significantly greater mortality at all time points compared with those without RT, even after adjusting for other covariates. No significant difference was found in the incidence of perioperative complications, reintervention, or restenosis between the two groups. The present study is unique because of the large sample size and length of follow-up. The results suggest that for this high-risk group, CAS provides the same patency as it does for atherosclerotic carotid stenosis and avoids potentially morbid cranial nerve injury and wound healing complications.

Estimation of Cumulative Incidence Function in the Presence of Middle Censoring Using Improper Gompertz Distribution

In this paper we deal with the modelling of cumulative incidence function using improper Gompertz distribution based on middle censored competing risks survival data. Together with the unknown parameters, cumulative incidence function also estimated. In classical set up, we derive the point estimates using maximum likelihood estimator and midpoint approximation methods. The asymptotic confidence interval are obtained based on asymptotic normality properties of maximum likelihood estimator. We also derive the Bayes estimates with associated credible intervals based on informative and non-informative types of priors under two loss functions such as squared error and LINEX loss functions. A simulation study is conducted for comprehensive comparison between various estimators proposed in this paper. A real life data set is also used for illustration.

Competing Risks Methods Are Recommended for Estimating the Cumulative Incidence of Revision Arthroplasty for Health Care Planning Purposes.

Cumulative incidence of revision provides a measure of the failure rate of joint replacements and can be used to project demand for revisions. The most commonly applied survival analysis method (Kaplan-Meier [KM]) does not account for competing risks (eg, death). The authors compared the cumulative incidence function (CIF), a competing risks method, with the KM method through application to population-based cohorts. They measured time to revision, death, or censoring for unilateral total hip arthroplasty (THA; n=12,496) and total knee arthroplasty (TKA; n=19,172) cohorts in administrative databases in Alberta and TKAs (n=80,177) in the Swedish Knee Arthroplasty Register. The authors compared relative differences between the KM and CIF. They fitted Cox, Fine and Gray, and Royston and Parmar regression models and compared coefficients, standard errors, and P values. On sensitivity analysis, the authors included staged bilateral operations. Kaplan-Meier estimates exceeded the CIF at each time point. The magnitude of overestimation increased with follow-up time and was greatest for the Swedish cohort. At 5 years, relative differences between KM and CIF estimates for the Alberta THA and TKA and Swedish TKA cohorts were 1.8%, 2.3%, and 3.8%, respectively. These differences increased to 3.1%, 5.8%, and 8.2%, respectively, at 9 years, reaching 39.1% at 20 years (Swedish cohort). On sensitivity analysis (including staged bilateral operations), the Fine and Gray subdistribution hazard ratio differed from the Cox and Royston and Parmar hazard ratios. When the frequency of competing risks is high, competing risks methods are recommended to obtain accurate cumulative incidence estimates for informing health care planning and decision making. [Orthopedics. 2021;44(4):e549-e555.].

Implant survival of 662 dual-mobility cups and 727 constrained liners in primary THA: small femoral head size increases the cumulative incidence of revision

Background and purpose — In total hip arthroplasty (THA), the risk for dislocation can be reduced using either dual-mobility cups (DMCs) or constrained liners (CLs). There are few studies comparing these concepts in primary THA. Therefore, we compared the cumulative incidence of revision in primary THA patients treated with DMC or CL with varying head sizes with conventional THA patients as reference group. Patients and methods — We performed a cohort study based on the Finnish arthroplasty register, comparing DMCs and CLs operated over the period 2000–2017. DMCs were divided into 2 groups based on the implant design: “DMC Trident” group (n = 399) and “DMC Others” group (n = 263). CLs were divided based on the femoral head size: “CL 36 mm” group (n = 425) and “CL < 36 mm” group (n = 302). All conventional primary THAs operated on in 2000–2017 with 28–36 mm femoral head were included as control group (“Conventional THA” group, n = 102,276). Implant survival was calculated by the corresponding cumulative incidence function with revision as the endpoint and death as competing event. Also, the prevalence of different reasons for revision was compared. Results — The 6-year cumulative incidence function estimates for the first revision were 6.9% (95% CI 4.0–9.7) for DMC Trident, 5.0% (CI 1.5–8.5) for DMC Others, 13% (CI 9.3–17) for CL < 36 mm, 6.3% (3.7–8.9) for CL 36 mm, and 4.7% (CI 4.5–4.8) for control group (conventional THA). The prevalence of dislocation revision was high (5.0%, CI 2.9–8.2) in the CL < 36 mm group compared with other groups. Interpretation — The DMC and CL 36 mm groups had promising mid-term survival rates, comparable to those of primary conventional THA group. The revision rate of CLs with < 36 mm head was high, mostly due to high prevalence of dislocation revisions. Therefore, CLs with 36 mm femoral head should be preferred over smaller ones.

Real-World Effectiveness and Safety of Blinatumomab in Adults with Relapsed or Refractory B-Cell Precursor Acute Lymphoblastic Leukaemia in Europe: 3-Year Results in Philadelphia Chromosome-Negative Patients and a Subset of Patients with Late First Relapse

Real-World Effectiveness and Safety of Blinatumomab in Adults with Relapsed or Refractory B-Cell Precursor Acute Lymphoblastic Leukaemia in Europe: 3-Year Results in Philadelphia Chromosome-Negative Patients and a Subset of Patients with Late First Relapse

P1507COMPETING RISKS MODELING IN PERITONEAL DIALYSIS: THE URGE FOR A CHANGE IN SURVIVAL ANALYSIS METHODOLOGY

Abstract Background and Aims Survival analysis is a cornerstone in medical research. For this purpose Kaplan-Meier is the most widely used statistical test, but the presence of competing risks violates the fundamental assumption that the censoring mechanism is independent of survival time. This leads to overestimation of the cumulative probability of cause-specific failure. Cumulative incidence estimate and competing risks analysis are preferred. The purpose of this study was to compare different survival analysis methods: Kaplan-Meier and cumulative incidence function estimates in a cohort of Peritoneal Dialysis (PD) patients. Method The survival of 115 incident patients on PD in a university hospital was evaluated after establishing 2 cohorts: patients starting renal replacement therapy with PD (PD first; n=85) and patients switching to PD on the first 6 months of dialysis (PD transfer; n=30). Kaplan-Meier, cumulative incidence function, cause-specific and subdistribution hazards were performed. The event of interest was death and the competing risk events were transfer to hemodialysis and renal transplantation. Results Besides higher residual renal function (RRF) and kt/V in the PD first group, there were no other significant differences between groups. There were 22 deaths. PD first group had a better survival with both Kaplan-Meier (log-rank test, p=0.013) and cumulative incidence function (p=0.021) approaches. The Cox regression model showed, as protecting variables, higher albumin (HR=0.174; CI95% 0.054-0.562), higher RRF (HR=0.785; CI95% 0.666-0.925) and PD first (HR=0.350; CI95% 0.132-0.927). Higher Charlson Index predicted worse outcome (HR=1.459; CI95% 1.159-1.835). PD as first dialysis therapy was associated with 65.0 % lower risk of death comparing with PD transfer. The subdistribution multivariable model found higher Charlson Index (HR=1.389; CI95% 1.118-1.725) and lower RRF (HR=0.798; CI95% 0.680-0.936) were statistically associated with death, but not PD transfer or albumin. This result differs from the obtained using the cause-specific hazard model. Analyzing the competing events, patients submitted to renal transplantation had a lower Charlson Index. Conclusion The probability of death was overestimated by the Kaplan-Meier method. The bias of Kaplan-Meier is especially great when the hazard of the competing risks is large. This study consisted on a statistical critical analysis of a real medical example, broader clinical conclusions related with “PD first initiative” should be cautious in this context. It is primordial to recognize the presence of competing risks in studies with multiple outcomes, as in Peritoneal Dialysis studies, to estimate cumulative incidence and yield more accurate results. This study shows how different conclusions are attained with different statistical methodology and its relevance in clinical context.

Female gender and mortality risk in decompensated heart failure

BackgroundStill there is conflicting evidence about gender-related differences in prognosis among patients with heart failure. This prognostic uncertainty may have implications for risk stratification and planning management strategy. The aim of the present study was to explore the association between gender and one-year mortality in patients admitted with acute decompensated heart failure (ADHF). MethodsWe studied 1513 patients. The Cumulative Incidence Function (CIF) method was used to estimate the absolute rate of mortality, heart transplantation (HT)/ventricular assist device (VAD) implantation, and survival free of HT/VAD implantation at 1year. An interaction analysis was performed to assess the association between covariates, gender, and mortality risk. Propensity score matching and Cox regression were used to compare mortality rates in the gender subgroups. ResultsThe CIF estimates of 1-year mortality, HT/VAD implantation, and survival free of HT/VAD implantation at 1year were 33.1%, 7.0%, and 59.9% for women and 30.2%, 10.2%, and 59.6% for men, respectively. Except for diabetes, there was no significant interaction between gender, covariates, and mortality risk. In the matched cohort, the hazard ratio of death for women was 1.19 (95% confidence intervals [CIs]: 0.90–1.59; p=.202). After adjusting for age and baseline risk, the hazard ratio of death for women was 1.18 (95% CIs: 0.95–1.43; p=.127). The use of gender-specific predictive models did not allow improving the accuracy of risk prediction. ConclusionsOur data strongly suggest that women and men have comparable outcome in the year following a hospitalization for ADHF.

Local Control and Toxicity of External Beam Reirradiation With a Pulsed Low-dose-rate Technique

To evaluate the efficacy and toxicity of external beam reirradiation using a pulsed low-dose-rate (PLDR) technique. We evaluated patients treated with PLDR reirradiation from 2009 to 2016 at a single institution. Toxicity was graded using the Common Terminology Criteria for Adverse Events, version 4.0, and local control was assessed using the Response Evaluation Criteria In Solid Tumors, version 1.1. On univariate analysis (UVA), the χ2 and Fisher exact tests were used to assess the toxicity outcomes. Competing risk analysis using cumulative incidence function estimates were used to assess local progression. A total of 39 patients were treated to 41 disease sites with PLDR reirradiation. These patients had a median follow-up time of 8.8months (range 0.5-64.7). The targets were the thorax, abdomen, and pelvis, including 36 symptomatic sites. The median interval from the first radiation course and reirradiation was 26.2months; the median dose of the first and second course of radiation was 50.4Gy and 50Gy, respectively. Five patients (13%) received concurrent systemic therapy. Of the 39 patients, 9 (23%) developed grade ≥2 acute toxicity, most commonly radiation dermatitis (5 of 9). None developed grade ≥4 acute or subacute toxicity. The only grade ≥2 late toxicity was late skin toxicity in 1 patient. On UVA, toxicity was not significantly associated with the dose of the first course of radiation or reirradiation, the interval to reirradiation, or the reirradiation site. Of the 41 disease sites treated with PLDR reirradiation, 32 had pre- and post-PLDR scans to evaluate for local control. The local progression rate was 16.5% at 6months and 23.8% at 12months and was not associated with the dose of reirradiation, the reirradiation site, or concurrent systemic therapy on UVA. Of the 36 symptomatic disease sites, 25 sites (69%) achieved a symptomatic response after PLDR, including 6 (17%) with complete symptomatic relief. Reirradiation with PLDR is effective and well-tolerated. The risk of late toxicity and the durability of local control were limited by the relatively short follow-up duration in the present cohort.

1. Effect of empagliflozin on cardiovascular death in subgroups by age: Results from EMPA-REG outcome

Abstract type Clinical Research. Presentation type Poster Presentation. Introduction Empagliflozin is a potent and selective SGLT2-I used in the treatment of T2DM. Methodology Study design: Patients with T2DM and established CV disease, and estimated glomerular filtration rate (eGFR;) ⩾30 ml/min/1.73 m2 were randomized to receive empagliflozin 10 mg, 25 mg, or placebo once daily in addition to standard of care for T2DM and CV risk management. The trial continued until 691 patients experienced an adjudicated event included in the primary outcome: 3P-MACE Analyses. Outcomes were assessed in patients treated with ⩾1 dose of study drug CV death was analysed in the pooled empagliflozin group vs placebo in subgroups by baseline age Cumulative incidence function estimates were corrected for mortality as a competing risk. Results The reduction in risk of CV death with empagliflozin vs placebo was consistent across age categories. Conclusion Empagliflozin in addition to standard of care reduced the risk of CV death in patients with T2DM and high CV risk irrespective of age at baseline. Across age subgroups, reported adverse events were consistent with the known safety profile of empagliflozin. Greater proportions of patients had events consistent with genital infections with the empagliflozin than placebo in all age subgroups. Diabetic ketoacidosis adverse events were rare and balanced overall between the empagliflozin and placebo groups. Events consistent with volume depletion were comparable between empagliflozin and placebo groups, except for a slightly higher frequency with empagliflozin than placebo in patients aged ⩾75 years.

The Fine-Gray Model Under Interval Censored Competing Risks Data.

We consider semiparametric analysis of competing risks data subject to mixed case interval censoring. The Fine-Gray model (Fine & Gray, 1999) is used to model the cumulative incidence function and is coupled with sieve semiparametric maximum likelihood estimation based on univariate or multivariate likelihood. The univariate likelihood of cause-specific data enables separate estimation of cumulative incidence function for each competing risk, in contrast with the multivariate likelihood of full data which estimates cumulative incidence functions for multiple competing risks jointly. Under both likelihoods and certain regularity conditions, we show that the regression parameter estimator is asymptotically normal and semiparametrically efficient, although the spline-based sieve estimator of the baseline cumulative subdistribution hazard converges at a rate slower than root-n. The proposed method is evaluated by simulation studies regarding its finite sample performance and is illustrated by a competing risk analysis of data from an dementia cohort study.

Better survival of hybrid total knee arthroplasty compared to cemented arthroplasty

Background and purpose — There have been few comparative studies on total knee replacement (TKR) with cemented tibia and uncemented femur (hybrid TKR). Previous studies have not shown any difference in revision rate between cemented and hybrid fixation, but these studies had few hybrid prostheses. We have evaluated the outcome of hybrid TKR based on data from the Norwegian Arthroplasty Register (NAR).Patients and methods — We compared 4,585 hybrid TKRs to 20,095 cemented TKRs with risk of revision for any cause as the primary endpoint. We included primary TKRs without patella resurfacing that were reported to the NAR during the years 1999–2012. To minimize the possible confounding effect of prosthesis brands, only brands that were used both as hybrids and cemented in more than 200 cases were included. Kaplan-Meier survival analysis and Cox regression analysis were done with adjustment for age, sex, and preoperative diagnosis. To include death as a competing risk, cumulative incidence function estimates were calculated.Results — Estimated survival at 11 years was 94.3% (95% CI: 93.9–94.7) in the cemented TKR group and 96.3% (CI: 95.3–97.3) in the hybrid TKR group. The adjusted Cox regression analysis showed a lower risk of revision in the hybrid group (relative risk = 0.58, CI: 0.48–0.72, p < 0.001). The hybrid group included 3 brands of prostheses: LCS classic, LCS complete, and Profix. Profix hybrid TKR had lower risk of revision than cemented TKR, but the LCS classic and LCS complete did not. Kaplan-Meier estimated survival at 11 years was 96.8% (CI: 95.6–98.0) in the hybrid Profix group and 95.2% (CI: 94.6–95.8) in the cemented Profix group. Mean operating time was 17 min longer in the cemented group.Interpretation — Survivorship of the hybrid TKR at 11 years was better than that for cemented TKR, or the same, depending on the brand of prosthesis. Hybrid fixation appears to be a safe and time-efficient alternative to cemented fixation in total knee replacement surgery.

Nonparametric estimation of cumulative incidence functions for competing risks data with missing cause of failure

In this paper, we develop a fully nonparametric approach for the estimation of the cumulative incidence function with Missing At Random right-censored competing risks data. We obtain results on the pointwise asymptotic normality as well as the uniform convergence rate of the proposed nonparametric estimator. A simulation study that serves two purposes is provided. First, it illustrates in detail how to implement our proposed nonparametric estimator. Second, it facilitates a comparison of the nonparametric estimator to a parametric counterpart based on the estimator of Lu and Liang (2008). The simulation results are generally very encouraging.

Nonparametric Estimation of Cumulative Incidence Functions for Competing Risks Data with Missing Cause of Failure

In this paper, we develop a fully nonparametric approach for the estimation of the cumulative incidence function with Missing At Random right-censored competing risks data. We obtain results on the pointwise asymptotic normality as well as the uniform convergence rate of the proposed nonparametric estimator. A simulation study that serves two purposes is provided. First, it illustrates in details how to implement our proposed nonparametric estimator. Secondly, it facilitates a comparison of the nonparametric estimator to a parametric counterpart based on the estimator of Lu and Liang (2008). The simulation results are generally very encouraging.

Estimation of cumulative incidence functions when the lifetime distributions are uniformly stochastically ordered

In the competing risks problem an important role is played by the cumulative incidence function (CIF), whose value at time t is the probability of failure by time t from a particular type of risk in the presence of other risks. Assume that the lifetime distributions of two populations are uniformly stochastically ordered. Since this ordering may not hold for the empiricals due to sampling variability, it is natural to estimate these distributions under this constraint. This will in turn affect the estimation of the CIFs. This article considers this estimation problem. We do not assume that the risk sets in the two populations are related, give consistent estimators of all the CIFs and study the weak convergence of the resulting processes. We also report the results of a simulation study that show that our restricted estimators outperform the unrestricted ones in terms of mean square error. A real life example is used to illustrate our theoretical results.

Tutorial in biostatistics: competing risks and multi‐state models

Standard survival data measure the time span from some time origin until the occurrence of one type of event. If several types of events occur, a model describing progression to each of these competing risks is needed. Multi-state models generalize competing risks models by also describing transitions to intermediate events. Methods to analyze such models have been developed over the last two decades. Fortunately, most of the analyzes can be performed within the standard statistical packages, but may require some extra effort with respect to data preparation and programming. This tutorial aims to review statistical methods for the analysis of competing risks and multi-state models. Although some conceptual issues are covered, the emphasis is on practical issues like data preparation, estimation of the effect of covariates, and estimation of cumulative incidence functions and state and transition probabilities. Examples of analysis with standard software are shown.

Estimation of cumulative incidence functions in competing risks studies under an order restriction

In the competing risks problem an important role is played by the cumulative incidence function (CIF), whose value at time t is the probability of failure by time t for a particular type of failure in the presence of other risks. Its estimation and asymptotic distribution theory have been studied by many. In some cases there are reasons to believe that the CIFs due to two types of failure are order restricted. Several procedures have appeared in the literature for testing for such orders. In this paper we initiate the study of estimation of two CIFs subject to a type of stochastic ordering, both when there are just two causes of failure and when there are more than two causes of failure, treating those other than the two of interest as a censoring mechanism. We do not assume independence of the two types of failure of interest; however, these are assumed to be independent of the other causes in the censored case. Weak convergence results for the estimators have been derived. It is shown that when the order restriction is strict, the asymptotic distributions are the same as those for the empirical estimators without the order restriction. Thus we get the restricted estimators “free of charge”, at least in the asymptotic sense. When the two CIFs are equal, the asymptotic MSE is reduced by using the order restriction. For finite sample sizes simulations seem to indicate that the restricted estimators have uniformly smaller MSEs than the unrestricted ones in all cases.