Analysis of Survival Data with Competing Risks in ADRD (Alzheimer's Disease and Related Dementias) Research

Competing risk is an event that precludes the occurrence of the primary event of interest. For example, when studying risk factors associated with dementia, death before the onset of dementia serve as a competing event. A subject who dies is no longer at risk of dementia. This issue play more important role in ADRD research given the elderly population. Conventional methods for survival analysis assume independent censoring and ignore the competing events. However, there are some challenge issues using those conventional methods in the presence of competing risks. First, no one-to-one link between hazard function and cumulative incidence function (CIF), and Kaplan-Meier approach overestimates the cumulative incidence of the event of interest. Second, the effect of covariates on hazard rate cannot be directly linked to the effect of cumulative incidence (the risk). We will discuss two types of analyses in the presence of competing risk: Cause-specific hazard model and Fine-Gray subdistribution hazard model. Cause-specific hazard model directly quantify the cause-specific hazard among subjects who are at risk of developing the event of interest, while Fine-Gray subdistribution hazard model directly model the effects of covariates on the cumulative incidence function. The type of research questions (Association vs. Prediction) may guide the choice of different statistical approaches. We will illustrate those two competing risk analyses using the large national dataset from National Alzheimer’s Coordinating Center (NACC). We will analyze the association between baseline diabetes status and the incidence of dementia, in which death before the onset of dementia is a competing event.

Predicting Cancer-Specific Survival Among Patients With Prostate Cancer After Radical Prostatectomy Based on the Competing Risk Model: Population-Based Study

Introduction: We aimed to develop an easy-to-use individual survival prognostication tool based on competing risk analyses to predict the risk of 5-year cancer-specific death after radical prostatectomy for patients with prostate cancer (PCa).Methods: We obtained the data from the Surveillance, Epidemiology, and End Results (SEER) database (2004–2016). The main variables obtained included age at diagnosis, marital status, race, pathological extension, regional lymphonode status, prostate specific antigen level, pathological Gleason Score. In order to reveal the independent prognostic factors. The cumulative incidence function was used as the univariable competing risk analyses and The Fine and Gray's proportional subdistribution hazard approach was used as the multivariable competing risk analyses. With these factors, a nomogram and risk stratification based on the nomogram was established. Concordance index (C-index) and calibration curves were used for validation.Results: A total of 95,812 patients were included and divided into training cohort (n = 67,072) and validation cohort (n = 28,740). Seven independent prognostic factors including age, race, marital status, pathological extension, regional lymphonode status, PSA level, and pathological GS were used to construct the nomogram. In the training cohort, the C-index was 0.828 (%95CI, 0.812–0.844), and the C-index was 0.838 (%95CI, 0.813–0.863) in the validation cohort. The results of the cumulative incidence function showed that the discrimination of risk stratification based on nomogram is better than that of the risk stratification system based on D'Amico risk stratification.Conclusions: We successfully developed the first competing risk nomogram to predict the risk of cancer-specific death after surgery for patients with PCa. It has the potential to help clinicians improve post-operative management of patients.

Long-term durability of transcatheter aortic valve replacement: outcomes from a contemporary cohort from a tertiary reference center at 5-years and beyond

Background and aim Long-term data on the durability of transcatheter heart valves is scarce. This is of particular interest as indications expand to younger and lower surgical risk patients. We sought to assess the incidence of long-term structural valve dysfunction (SVD) and bioprosthetic valve failure (BVF) in a cohort of patients with TAVR who reached at least 5-year follow-up, as compared to surgical aortic valve replacement (SAVR), performed within the same time-frame at the same institution. Methods and results Consecutive patients with at least 5-year available follow-up, who underwent TAVR between November 2008 to December 2015 in a tertiary single center, were included. From a group of 246 patients undergoing TAVR, 126 had available follow-up data (age at implantation: 83.0 [77.8–87.0] years; EuroScore II: 4.54 [2.60–6.29]%; follow-up: 5.94 [5.06–7.67] years). First generation Corevalve® and Sapien® prosthesis were implanted in 56% and 38% patients, respectively. SVD and BVF were defined according to the new consensus statement from the EAPCI endorsed by the ESC and the EACTS. Mean transaortic pressure gradients decreased from 53.2±1.3 mmHg (pre-TAVR) to 10.4±0.4 mmHg (at discharge or up to one-year after TAVR, p<0.001), and there was a small non-significant increase at the fifth-year and the last available follow-up (11.2±0.6 mmHg; 14.7±1.8 mmHg, respectively). Moderate and severe SVD were reported in 12 and 4 patients, respectively (8-year cumulative incidence function to SVD: 2.67%; 95% CI, 2.12–3.89). Of these 8 had BVF, 7 of them with hospitalization for acute heart failure. A total of 4 patients died and none required reintervention (redo TAVR or SAVR). BVF for non-SVD were observed in 4 patients (2 subclinic thrombosis successfully treated with anticoagulation and 2 paravalvular regurgitation due to endocarditis). As comparator, from a cohort of 587 patients submitted to biological SAVR, 247 (age 75.0 [70.0–79.0] years; EuroScore II 1.43 [1.06–2.17]%) had available long-term follow-up (6.89 [6.08–8.19] years). Moderate and severe SVD were reported in 42 and 3 patients, respectively (8-year cumulative incidence function to SVD: 3.13%; 95% CI, 2.45–4.21). These events were clinically relevant (BVF) in 19 of them: 8 performed TAVR valve-in-valve procedures and 3 redo SAVR. At the fifth-year of follow-up the incidence of SVD was not statistically different between TAVR (8%) and SAVR (15%), with a p for comparison of 0.137. Conclusions In our population of patients with symptomatic severe aortic stenosis treated with first-generation percutaneous bioprostheses, TAVR was associated with a low incidence of BVF and SVD at the long-term follow-up. These outcomes seem indistinct from those occurring in patients submitted to conventional SAVR FUNDunding Acknowledgement Type of funding sources: None. KM curve reporting probability of SVD

Long-term results of tricuspid annuloplasty with 3-dimensional-shaped rings: effective and durable!

OBJECTIVES 3-Dimensional (3D)-shaped rings are largely adopted for tricuspid annuloplasty, but evidence about their long-term results is scanty. The goal of this study was to analyse the long-term results of tricuspid annuloplasty with 3D-shaped rings. MATERIALS AND METHODS A retrospective review of our prospectively maintained database was carried out to identify all patients who underwent tricuspid valve repair with 3D-shaped rings between January 2011 and December 2014. Kaplan–Meier methods were used to analyse long-term survival. Cumulative incidence function using death as the competitive outcome was used to estimate cardiac death. RESULTS A total of 168 patients were identified. The median age was 66 years. Eighty-two patients (49%) were in advanced New York Heart Association functional class III–IV. Atrial fibrillation (AF) was present in 101 (60%); the median ejection fraction was 60%. In 82 (49%) patients, a Medtronic 3D Contour annuloplasty ring was employed; in the remaining 86 (51%) patients, an Edwards MC3 ring was used. Cumulative incidence function of cardiac death, with non-cardiac death as a competing risk, was 1.9 ± 1.1%, 95% confidence interval (CI) (0.51–4.95) at 7 years. The cumulative incidence function of recurrence of tricuspid regurgitation (TR) ≥2+ at 7 years was 14 ± 3.17%, 95% CI (8.49–20.82). Recurrence of TR ≥2+ at 7 years was not significantly different between the Medtronic 3D Contour and the Edwards MC3 rings (P = 0.3). AF was identified as the only independent predictor of recurrence of TR ≥2+. CONCLUSIONS 3D-shaped rings are effective and durable. TR recurrence was relatively low at 7 years and usually moderate (2+/4+) without a significant difference between the 2 types of rings. The role of AF as a predictor of TR recurrence was confirmed.

Hypertrophic cardiomyopathy with moderate septal thickness and mitral regurgitation: long-term surgical results

OBJECTIVES The aim of this study was to assess the long-term outcomes of different surgical strategies in patients with hypertrophic obstructive cardiomyopathy (HOCM) with septal thickness ≤18 mm and systolic anterior motion (SAM)-related moderate-to-severe mitral regurgitation (MR). METHODS Seventy-six HOCM patients with septal thickness 17 [16; 18] mm, resting left ventricle outflow tract gradient 60 [41; 85] mmHg and SAM-related MR ≥2+/4+, underwent septal myectomy alone (54%) or mitral valve (MV) surgery ± myectomy (46%). RESULTS No hospital death and no ventricular septal defect occurred. Patients undergoing MV surgery ± myectomy had longer cardiopulmonary bypass and X-clamp times (77 [60–106] vs 51 [44–62] min, P < 0.001 and 56 [45–77] vs 32 [28–41] min, P < 0.001) and higher incidence of low output syndrome (11% vs 0%, P = 0.04). Follow-up was 98.6% complete, median 8 years [3–11]. There were no statistically significant differences in overall survival (P = 0.069) with survival rates at 9 years of 96 ± 4% in the myectomy alone group and 81 ± 8% in the MV surgery ± myectomy one. At 9 years, cumulative incidence function of cardiac death was 12 ± 6% in the MV surgery ± myectomy group vs 0% in the myectomy one, P = 0.06. Multivariable analysis identified age and previous septal alcoholization as predictors of cardiac death (hazard ratio (HR) = 1.1, 95% confidence interval (CI) 1.0–1.1, P = 0.004 and HR = 2.9, 95% CI 1.0–8.3, P = 0.042). The 9-year cumulative incidence function of recurrence of MR ≥2+, with death as competing risk, was 3 ± 2.8% in the MV surgery ± myectomy group vs 25 ± 6.9% in the myectomy one, P = 0.005. CONCLUSIONS In HOCM patients with moderate septal thickness and SAM-related MR, as the degree of septal hypertrophy decreases, addressing the abnormalities of the MV apparatus may become necessary to provide a durable resolution of left ventricle outflow tract obstruction and SAM-related MR. However, performing myectomy alone, whenever possible, seems to be associated to a better postoperative course and a trend towards lower cardiac mortality at follow-up, despite a higher rate of residual moderate MR.

Cause of Death Among Patients With Peripheral Artery Disease

Background: Peripheral artery disease is common and associated with high mortality. There are limited data detailing causes of death among patients with peripheral artery disease. Methods: EUCLID (Examining Use of Ticagrelor in Peripheral Artery Disease) was a randomized clinical trial that assigned patients with peripheral artery disease to clopidogrel or ticagrelor. We describe the causes of death in EUCLID using mortality end points adjudicated through a clinical events classification process. The association between baseline factors and cardiovascular death was evaluated by Cox proportional hazards modeling. The competing risk of noncardiovascular death was assessed by the cumulative incidence function for cardiovascular death and the Fine and Gray method to ascertain the association between baseline characteristics and cardiovascular mortality. Results: A total of 1263 out of 13 885 (9.1%) patients died (median follow-up: 30 months). There were 706 patients (55.9%) with a cardiovascular cause of death and 522 (41.3%) with a noncardiovascular cause of death. The most common cause of cardiovascular death was sudden cardiac death (20.1%); while myocardial infarction (5.2%) and ischemic stroke (3.2%) were uncommon. The most common causes of noncardiovascular death were malignancies (17.9%) and infections (11.9%). The factor most associated with a higher risk of cardiovascular death was age per 5 year increase (HR, 1.26 [95% CI, 1.20–1.32]). Female sex was associated with a lower risk of cardiovascular death (HR, 0.68 [95% CI, 0.56–0.82]). To evaluate the effect of noncardiovascular death as a competing risk, we superimposed the cumulative incidence function curve with the Kaplan-Meier curve. These curves closely approximated each other. After accounting for the competing risk of noncardiovascular death, the magnitude and direction of the factors associated with cardiovascular death were minimally changed. Conclusions: Among patients with symptomatic peripheral artery disease, noncardiovascular causes of death reflected a high proportion (40%) of deaths. Accounting for noncardiovascular deaths as a competing risk, there was not a significant change in the risk estimation for cardiovascular death. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT01732822.

A Competing Risk Nomogram for Predicting Cancer-Specific Survival Among Patients with Prostate Cancer after Radical Prostatectomy

Background: We aimed to develop a detailed individual survival prognostication tool based on competing risk analyses to predict the risk of 5-year cancer-specific death after radical prostatectomy for patients with prostate cancer (PCa).Methods: We obtained the data from the Surveillance, Epidemiology, and End Results (SEER) database (2004-2016). The main variables obtained included age at diagnosis, marital status, race, pathological extension, regional lymphonode status, prostate specific antigen level, Gleason Score biopsy. In order to reveal the independent prognostic factors. The cumulative incidence function was used as the univariable competing risk analyses and The Fine and Gray’s proportional subdistribution hazard approach was used as the multivariable competing risk analyses. With these factors, a nomogram and risk stratification based on the nomogram was established. Concordance index (C-index) and calibration curves were used for validation.Results: A total of 95,812 patients were included and divided into training cohort (n = 67,072) and validation cohort (n = 28,740). Seven independent prognostic factors including age, race, marital status, pathological extension, regional lymphonode status, PSA level, and GS biopsy were used to construct the nomogram. In the training cohort, the C-index was 0.828 (%95CI, 0.812-0.844), and the C-index was 0.838 (%95CI, 0.813-0.863) in the validation cohort. The results of the cumulative incidence function showed that the discrimination of risk stratification based on nomogram is better than that of the risk stratification system based on D'Amico risk stratification.Conclusions: We successfully developed the first competing risk nomogram to predict the risk of cancer-specific death after surgery for patients with PCa. It has the potential to help clinicians improve postoperative management of patients