Individualized dynamic prediction of survival with the presence of intermediate events

Often, in follow-up studies, patients experience intermediate events, such as reinterventions or adverse events, which directly affect the shapes of their longitudinal profiles. Our work is motivated by two studies in which such intermediate events have been recorded during follow-up. In both studies, we are interested in the change of the longitudinal evolutions after the occurrence of the intermediate event and in utilizing this information to improve the accuracy of dynamic prediction of their risk. To achieve so, we propose a flexible joint modeling framework for longitudinal and time-to-event data, which includes features of the intermediate event as time-varying covariates in both the longitudinal and survival submodels. We consider a set of joint models that postulate different effects of the intermediate event in the longitudinal profile and the risk of the clinical endpoint, with different formulations for the association structure while allowing its functional form to change after the occurrence of the intermediate event. Based on these models, we derive dynamic predictions of conditional survival probabilities which are adaptive to different scenarios with respect to the occurrence of the intermediate event. We evaluate the predictive accuracy of these predictions with a simulation study using the time-dependent area under the receiver operating characteristic curve and the expected prediction error adjusted to our setting. The results suggest that accounting for the changes in the longitudinal profiles and the instantaneous risk for the clinical endpoint is important, and improves the accuracy of the dynamic predictions.

Uncertainties and challenges in surgical and transcatheter tricuspid valve therapy: a state-of-the-art expert review

Tricuspid regurgitation (TR) is a frequent and complex problem, commonly combined with left-sided heart disease, such as mitral regurgitation. Significant TR is associated with increased mortality if left untreated or recurrent after therapy. Tricuspid regurgitation was historically often disregarded and remained undertreated. Surgery is currently the only Class I Guideline recommended therapy for TR, in the form of annuloplasty, leaflet repair, or valve replacement. As growing experience of transcatheter therapy in structural heart disease, many dedicated transcatheter tricuspid repair or replacement devices, which mimic well-established surgical techniques, are currently under development. Nevertheless, many aspects of TR are little understood, including the disease process, surgical or interventional risk stratification, and predictors of successful therapy. The optimal treatment timing and the choice of proper surgical or interventional technique for significant TR remain to be elucidated. In this context, we aim to highlight the current evidence, underline major controversial issues in this field and present a future roadmap for TR therapy.

Collapse

Statistical primer: checking model assumptions with regression diagnostics

Regression modelling is an important statistical tool frequently utilized by cardiothoracic surgeons. However, these models-including linear, logistic and Cox proportional hazards regression-rely on certain assumptions. If these assumptions are violated, then a very cautious interpretation of the fitted model should be taken. Here, we discuss several assumptions and report diagnostics that can be used to detect departures from these assumptions. Most of the diagnostics discussed are based on residuals: a measure of the difference between the observed and model fitted values. Reliable and generalizable results depend on correctly developed statistical models, and proper diagnostics should play an integral part in the model development.

Outcomes after surgery for functional tricuspid regurgitation: a systematic review and meta-analysis

Aims 

This study aims to provide a contemporary overview of outcomes after tricuspid valve (TV) surgery for functional tricuspid regurgitation (TR).

Methods and results 

The literature was systematically searched for papers published between January 2005 and December 2017 reporting on clinical/echocardiographic outcomes after TV surgery for functional TR. A random effects meta-analysis was conducted for outcome variables, and late outcomes are visualized by pooled Kaplan–Meier curves. Subgroup analyses were performed for studies with a within-study comparison of suture vs. ring repair and flexible vs. rigid ring repair. Eighty-seven publications were included, encompassing 13 184 patients (mean age: 62.1 ± 11.8 years, 55% females). A mitral valve procedure was performed in 92% of patients. Pooled mean follow-up was 4.0 ± 2.8 years. Pooled early mortality was 3.9% (95% CI: 3.2–4.6), and late mortality rate was 2.7%/year (95% CI: 2.0–3.5), of which approximately half was cardiac-related 1.2%/year (95% CI: 0.8–1.9). Pooled risk of early moderate-to-severe TR at discharge was 9.4% (95% CI: 7.0–12.1). Late moderate-to-severe TR rate after discharge was 1.9%/year (95% CI: 1.0–3.5). Late reintervention rate was 0.3%/year (95% CI: 0.2–0.4). Mortality and overall (early and late) TR rate were comparable between suture vs. ring annuloplasty (14 studies), whereas overall TR rate was higher after flexible ring vs. rigid ring annuloplasty (6 studies) (7.5%/year vs. 3.9%/year, P = 0.002).

Conclusion 

This study shows that patients undergoing surgery for functional tricuspid regurgitation (FTR) have an acceptable early and late mortality. However, TR remains prevalent after surgery. The results of this study can be used to inform patients and clinicians about the expected outcome after surgery for FTR and can results serve as a benchmark for the performance of emerging transcatheter TV interventions.

Aortic Valve Surgery in Nonelderly Patients: Insights Gained From AVIATOR

Aortic valve surgery in non-elderly patients represents a very challenging patient population. The younger the patient is at the point of aortic valve intervention, the longer their anticipated life expectancy will be, with longer exposure to valve-related complications and risk for re-operation. Although the latest international guidelines recommend aortic valve repair in patients with aortic valve insufficiency, what we see in the real world is that the vast majority of these aortic valves are replaced. However, current prosthetic valves has now been shown to lead to significant loss of life expectancy for non-elderly patients up to 50% for patients in their 40s undergoing mechanical aortic valve replacement. Bioprostheses carry an even worse long-term survival, with higher rates of re-intervention. The promise of trans-catheter valve-in-valve technology is accentuating the trend of bioprosthetic implantation in younger patients, without yet the appropriate evidence. In contrast, aortic valve repair has shown excellent outcomes in terms of quality of life, freedom from re-operation and freedom from major adverse valve-related events with similar life expectancy to general population as it is also found for the Ross procedure, the only available living valve substitute. We are at a time when the paradigm of aortic valve surgery needs to change for the better. To better serve our patients, we must acquire high quality real-world evidence from multiple centers globally - this is the vision of the AVIATOR registry and our common responsibility.

Clinical and quality of life outcomes after aortic valve replacement and aortic root surgery in adult patients <65 years old

Selecting the optimal surgical treatment strategy in patients below the age of 65 years (i.e., non-elderly patients) with aortic valve or aortic root disease remains challenging. The objective of the current study is to summarize contemporary research on clinical and quality of life outcomes after aortic valve replacement (AVR) and aortic root surgery in non-elderly patients. Recent systematic reviews on clinical outcome after biological and mechanical AVR, the Ross procedure and aortic root surgery show that event occurrence is considerable after any type of AVR or aortic root surgery and-with the exception of the Ross procedure-survival is suboptimal. Although thromboembolism and bleeding events are more common after mechanical AVR and root surgery, these events are also considerably present after biological AVR, the Ross procedure and valve-sparing aortic root surgery (VSRR). Similarly, reoperation is more common after biological AVR, the Ross procedure and VSRR, but also occurs frequently after mechanical AVR and root replacement. Published evidence in AVR patients points to the direction of better health-related quality of life (HRQoL) outcomes with a biological solutions, while the HRQoL after aortic root surgery is limited and contradictory. This review illustrates that treatment for non-elderly aortic valve and aortic root disease patients needs to be tailored to the individual patient, considering both clinical and HRQoL outcomes as crucial factors to reach a treatment decision that best reflects the patient's values and goals in life.

Effectiveness of adherence to a preoperative antiplatelet and anticoagulation cessation protocol in cardiac surgery

OBJECTIVES

Reduction of blood loss after cardiac surgery remains challenging. The effectiveness of adherence to a protocol on cessation of anticoagulants and platelet-inhibiting medications was investigated together with the influence of protocol violations on blood loss after surgery, use of blood products, surgical re-explorations and 30-day mortality.

METHODS

Between 2009 and 2013, data were collected prospectively for all elective cardiac surgery procedures in adult patients (n = 1637). Two groups were distinguished: Group 1 adhered to the protocol for cessation or continuation of medication (n = 1287, 79%) and Group 2 violated the protocol (n = 350, 21%).

RESULTS

Median blood loss was 300 ml (interquartile range 175-500 ml). Eighty patients underwent re-exploration due to blood loss (5%). Thirty-day mortality was 2% (n = 27). Protocol violation was associated with increased blood loss [median 275 ml (175-475 ml) vs 350 ml (250-612); P ≤ 0.001] and with increased average use of fresh frozen plasma (226 ml vs 139 ml; P  << 0.00001), red blood cell transfusion (115 ml vs 87 ml; P = 0.081) and thrombocyte transfusions (52 ml vs 37 ml; P = 0.0082). The number of re-explorations (4% vs 6%; P = 0.39) and mortality risk (1% vs 2%; P = 0.72) did not differ.

CONCLUSIONS

Balancing the benefit of continuing platelet inhibitors or anticoagulants versus cessation before surgery remains challenging. Adherence to the protocol will lead to lower blood loss and in a lower consumption of blood products although the decision to go for re-exploration and 30-day mortality does not differ compared with the protocol violation. Stopping medication does not lead to thromboembolic events.

Beyond the clinical impact of aortic and pulmonary valve implantation: health-related quality of life, informal care and productivity†

OBJECTIVES

Our aim was to provide estimates of patient-reported health-related quality of life (HRQoL), use of informal care and productivity in patients after surgical aortic and pulmonary valve replacement and transcatheter aortic valve implantation.

METHODS

Consecutive cohorts of 1239 adult patients who had surgical aortic valve replacement or surgical pulmonary valve replacement and 433 patients who had transcatheter aortic valve implantation at 2 Dutch heart centres were cross-sectionally surveyed at a median time of 2.9 and 3.2 years after the intervention, respectively. The survey included questions on HRQoL (EQ-5D-5L and SF-12-v2), use of informal care and productivity in paid and unpaid work. All outcomes were compared with age and sex-matched individuals from the general population.

RESULTS

The response rate was 56% (n = 687) of patients who had surgical valve replacement and 59% (n = 257) of those who had transcatheter aortic valve implantation. Compared with the general population, patients reported poorer HRQoL on physical health domains, whereas their scores were comparable for mental health domains. After a heart valve implantation, patients reported using informal care more frequently than the general population, but labour participation was comparable. Patients with late complications [antibiotic treatment for endocarditis (n = 4), stroke (n = 11), transient ischaemic attack (n = 15)] reported lower HRQoL, greater use of informal care and greater productivity loss than patients without complications.

CONCLUSIONS

Patients who had aortic and pulmonary valve implantations experience relatively mild limitations in daily life compared to the general population. The consequences of a heart valve implantations beyond clinical outcomes should be considered to create realistic patient expectations of life after a heart valve implantation and unbiased resource allocation decisions at national levels.

Reconstructive surgery for Ebstein anomaly: three decades of experience

OBJECTIVES

Since 1988, our centre employs vertical plication repair with deattachment and reattachment of the tricuspid valve for Ebstein anomaly. This study describes the characteristics and long-term outcomes of our single-centre cohort.

METHODS

Data from all patients operated on between 1988 and 2016 were retrospectively collected. Kaplan-Meier analyses were done for survival data and mixed models were used to analyse longitudinally collected clinical and echocardiography data.

RESULTS

Thirty-six patients (mean age: 25.4 ± 15.9 years, 36% male) were operated on using the Carpentier-Chauvaud 21 (58%) or Cone repair 15 (42%). One patient (3%) died in hospital. Two late deaths were observed, yielding a survival of 97 ± 3% at 25 years. Reoperation was performed in 6 patients after a mean follow-up of 14.1 ± 10.3 years, resulting in a freedom of reoperation of 80 ± 8% at 25 years. During follow-up, predicted probability of being in New York Heart Association III/IV did not exceed 10%. Modelling longitudinal evolution of tricuspid regurgitation showed no major changes over time. Additionally, a rigid ring repair was associated with a higher probability of tricuspid regurgitation, especially after the first years after the operation. A full Cone repair was associated with less progression of tricuspid regurgitation over time.

CONCLUSIONS

Repair of Ebstein abnomaly is associated with low mortality and morbidity, acceptable reoperation rate and excellent valve function over time, especially in patients with completed Cone repair. Therefore, we conclude that in our centre, repair of Ebstein abnomaly is a durable technique to treat patients.

Should we involve patients more actively? Perspectives of the multidisciplinary team on shared decision-making for older patients with metastatic castration-resistant prostate cancer

OBJECTIVE

To evaluate perspectives of the multidisciplinary team concerning shared decision-making (SDM) in treatment decisions for older patients with metastatic castration-resistant prostate cancer (mCRPC).

MATERIALS AND METHODS

A survey among Dutch healthcare providers was conducted to assess healthcare providers' perspectives on patient involvement in decision-making and the value of a decision aid (DA) in the decision-making process. Treatment recommendations were assessed using hypothetical cases in which providers were asked to evaluate their likelihood of pursuing listed treatment options.

RESULTS

In total, 170 Dutch healthcare providers, including 82 urologists, 31 oncologists, and 57 oncology nurses completed the survey. Sixty-two percent of urologists, 65% of oncologists, and 51% of oncology nurses found that mCRPC patients take a passive role in decision-making and delegate treatment decisions to doctors due to advanced age (p = .45). Yet, 70% of urologists, 71% of oncologists, and 63% of oncology nurses agreed that mCRPC patients should be always involved in decision-making (p = .91). Fifty-two percent of urologists and 55% of oncologists stated that they are inadequately trained to apply SDM in clinical practice. Conversely, only 20% of oncology nurses believed that oncology nurses are inadequately trained. Fifty-four percent of all providers considered a DA suitable to support these patients and their healthcare providers in the decision-making process. All hypothetical cases showed variation in treatment recommendations among providers, with each of the five treatments ranging from extremely likely to extremely unlikely.

CONCLUSIONS

The wide variation of treatment recommendations observed among the multidisciplinary team suggests that mCRPC patients and their healthcare providers may benefit from implementation of informed SDM. Given the perceived passive role of older patients with mCRPC in decision-making, interventions to engage them are needed. With slightly more than half of respondents finding DAs useful to facilitate the decision-making process, development and implementation of a DA would be an interesting field of research.

Measuring what matters to the patient: health related quality of life after aortic valve and thoracic aortic surgery

With improved outcomes following cardiac surgery, health related quality of life (HRQoL) gains increasing importance for the better judgement of choosing the preferred treatment strategy in the individual patient. The physician perception of patient preferences can differ considerably from actual patient preferences, underlining the importance of gathering evidence of actual patient preferences before and quality of life after cardiac surgery. The objective of the current review is to provide an overview of current insights into the quality of life measurements after aortic valve and thoracic aortic surgery and to provide starting points for the application of HRQoL measurements toward the future. The amount and level of evidence on HRQoL outcomes after aortic valve and thoracic aortic surgery seems to be insufficient. Little has been investigated about the natural course of HRQoL after cardiac surgery, HRQoL outcomes between different surgical strategies, HRQoL outcomes between surgical patients and the general population, the different factors influencing HRQoL after cardiac surgery, and the effect of HRQoL on healthcare costs. More prospective studies should be performed, taking into account the knowledge gaps that need to be filled. Computerized adaptive testing methods through open source programs can be implemented to keep the burden to the patient as low as possible and catalyze the use of these tools. Our cardiovascular surgery community has the responsibility to deliberate how it can proceed to effectively fill in these knowledge gaps, and use this newfound knowledge to improve shared treatment decision making, patient outcomes, and ultimately optimize health care efficiency.

Patient and physician view on patient information and decision-making in congenital aortic and pulmonary valve surgery

Background

To assess the current state of patient information and decision-making in congenital aortic and pulmonary valve disease, we conducted a survey among patients, parents and physicians.

Methods

A questionnaire was sent by ground mail to 157 adults and 32 parents of children who previously underwent surgery for congenital aortic or pulmonary valve disease at 0-40 years of age between January 2005 and February 2014 at the Erasmus University Medical Center and to all paediatric and adult congenital cardiologists and congenital cardiac surgeons in the Netherlands (n=88).

Results

73 patients/parents (39% response rate, 62 adult patients, 11 parents of paediatric patients) and 35 physicians (40% response rate) responded. Median patient age at the time of surgery was 25.7 years. Basic disease-specific knowledge was adequate in 42% of patients/parents and numeracy was sufficient in 47%. Patients/parents reported that they rely heavily on their physicians for information and often experience difficulty in finding reliable information elsewhere. They lack information on psychosocial aspects of disease (29% of respondents) and risks and benefits of treatment options (26%). They feel less involved in decision-making than they would prefer to be (p=0.014). Decisional conflict at the time of surgery was experienced by 31% of patients/parents. If they had to do it again, 72% of patients/parents would want the same treatment. Quality of life is often impaired due to various valve-related anxieties and lifestyle changes. Physicians reported that they are unable to fully inform and sufficiently involve patients, due to limited patient/parent knowledge and understanding (56%) and limited time during consultations (32%). Patients/parents (98%) and physicians (97%) agree that they should have shared roles in decision-making.

Conclusion

The substantial shortcomings in our current practice of patient information and decision-making underline the need for innovative solutions, such as careful implementation of patient information tools and shared decision-making in the care path.

Bioprosthetic aortic valve replacement in elderly patients: Meta-analysis and microsimulation

OBJECTIVE

To support decision-making in aortic valve replacement (AVR) in elderly patients, we provide a comprehensive overview of outcome after AVR with bioprostheses.

METHODS

A systematic review was conducted of studies reporting clinical outcome after AVR with bioprostheses in elderly patients (mean age ≥70 years; minimum age ≥65 years) published between January 1, 2000, to September 1, 2016. Reported event rates and time-to-event data were pooled and entered into a microsimulation model to calculate life expectancy and lifetime event risks.

RESULTS

Forty-two studies reporting on 34 patient cohorts were included, encompassing a total of 12,842 patients with 55,437 patient-years of follow-up (pooled mean follow-up 5.0 ± 3.3 years). Pooled mean age was 76.5 ± 5.5 years. Pooled early mortality risk was 5.42% (95% confidence interval [CI], 4.49-6.55), thromboembolism rate was 1.83%/year (95% CI, 1.28-3.61), and bleeding rate was 0.75%/year (95% CI, 0.50-1.11). Structural valve deterioration (SVD) was based on pooled time to SVD data (Gompertz; shape: 0.124, rate: 0.003). For a 75-year-old patient, this translated to an estimated life expectancy of 9.8 years (general population: 10.2 years) and lifetime risks of bleeding of 7%, thromboembolism of 17%, and reintervention of 9%.

CONCLUSIONS

The low risks of SVD and reintervention support the use of bioprostheses in elderly patients in need of AVR. The estimated life expectancy after AVR was comparable with the general population. The results of this study inform patients and clinicians about the expected outcomes after bioprosthetic AVR and thereby support treatment decision-making. Furthermore, our results can be used as a benchmark for long-term outcomes after transcatheter aortic valve implantation in patients who were eligible for surgery and other (future) alternative treatments (eg, tissue-engineered heart valves).

Male-female differences in aortic valve and combined aortic valve/coronary surgery: a national cohort study in the Netherlands

Objective

The outcome of female patients after adult cardiac surgery has been reported to be less favourable compared with the outcome of male patients. This study compares men with women with respect to patient and procedural characteristics and early mortality in a contemporary national cohort of patients who underwent aortic valve (AV) and combined aortic valve/coronary (CABG/AV) surgery.

Methods

All patients who underwent AV (n=8717, 56% male) or a combined CABG/AV surgery (n=5867, 67% male) in the Netherlands between January 2007 and December 2011 were included.

Results

In both groups, women were generally older than men (p<0.001) and presented with higher logistic EuroSCORES. In isolated AV surgery, men and women had comparable in-hospital mortality (OR 1.20, 95% CI 0.90 to 1.61; p=0.220). In concomitant CABG/AV surgery, in-hospital mortality was higher in women compared with men (OR 2.00, 95% CI 1.44 to 2.79; p<0.001). The area under the curve for logistic EuroSCORE 1 was systematically higher for men versus women in isolated AV surgery 0.82 (95% CI 0.78 to 0.86) vs 0.75 (95% CI 0.69 to 0.80) and in concomitant CABG/AV surgery 0.78 (95% CI 0.73 to 0.82) vs 0.69 (95% CI 0.63 to 0.74). Finally, (the weight of) risk factors associated with in-hospital mortality differed between men and women.

Conclusions

There are substantial male-female differences in patient presentation and procedural aspects in isolated AV and concomitant CABG/AV surgery in the Netherlands. Further studies are necessary to explore the mechanisms underlying the observed differences. In addition, the observation that standard risk scores perform worse in women warrants exploration of male-female specific risk models for patients undergoing cardiac surgery.Brief title:

Downsized cryopreserved and standard-sized allografts for right ventricular outflow tract reconstruction in children: long-term single-institutional experience

OBJECTIVES

The objective of this study was to determine long-term results with bicuspidalized allografts compared to non-bicuspidalized allografts in children under 2 years undergoing primary correction of the right ventricular outflow tract.

METHODS

Thirty-five consecutive bicuspidalized allografts were compared to 45 consecutive non-bicuspidalized allografts implanted during the same period. Valve-related events were analysed with Kaplan-Meier and Cox-regression techniques. Mixed-effects modelling was used to analyse serial echocardiographic measurements of pulmonary gradient. In addition, a systematic review with meta-analysis of the published literature concerning implantation of bicuspidalized allografts was performed.

RESULTS

Perioperative characteristics and in-hospital mortality [bicuspidalized 5 (14.3%), non-bicuspidalized 6 (13.3%)] were comparable (P = 0.902). Bicuspidalized allografts were smaller (14.7 vs 16.5 mm, P = 0.023) and always (100%) of pulmonary origin compared to 26 (57.8%) of the standard-sized allografts. There were no differences in late mortality between the bicuspidalized and non-bicuspidalized group (6.7% vs 7.7%, P = 0.798) or freedom from allograft replacement at 10 years (82 ± 10% and 71 ± 8%, for bicuspidalized and non-bicuspidalized allografts, respectively). Evolution of peak pulmonary gradient (P = 0.273) was comparable between bicuspidalized and non-bicuspidalized allografts. Meta-analysis showed a pooled early and late mortality for bicuspidalized allograft patients of 10.72% [95% confidence interval (CI) 6.13-18.75] and 1.6% per year (95% CI 0.99-2.79), respectively. Pooled estimated late reintervention and replacement rates were 5.94% per year (95% CI 3.42-10.30) and 3.78% per year (95% CI 2.69-5.32), respectively.

CONCLUSIONS

Bicuspidalization seems to be a viable alternative to combat limited supply of small-sized allografts with acceptable survival and reintervention rates comparable to non-bicuspidalized allografts.