Same-Day Discharge Greatly Impacts Cost to PCI

Dr. Amin, let me first congratulate you and your colleagues on a very insightful publication. How would you generally characterize the impact of these data for any hospital or healthcare system? 

This paper is about percutaneous coronary intervention (PCI) care pathways. Despite tremendous improvements in PCI devices, technology and pharmacotherapy, we have not yet streamlined care pathways to mirror these technological advancements. Thus, the current care pathways are leading to unnecessary high costs of care. If we (the cath lab staff and interventional community) optimized the care pathway, it would lead to tremendous economic advantages while preserving or even improving PCI outcomes and patient satisfaction. In a large, nationally representative dataset of Medicare beneficiaries linked to the National Cardiovascular Data Registry (NCDR) CathPCI registry, we found that PCI care pathways of transradial intervention (TRI) and same-day discharge (SDD) were associated with much lower costs compared with transfemoral intervention (TFI) with no SDD (NSDD). The differences in cost were so large, exceeding $3,500 for TRI PCI with SDD, that it implies that even small shifts in the current practice of TFI NSDD to TRI SDD by 30% could save a hospital performing 1,000 elective PCIs each year $1 million and the country $300 million annually.

In this paper, we have provided a detailed analysis of hospital costs across various care pathways for PCI. We have thus identified a mechanism to reduce hospital losses while maintaining or even improving outcomes. Our findings show that TRI combined with SDD may offer a tremendous potential for hospitals across the country to reduce PCI costs. These potential savings are meaningful and offer insight to hospitals to help them modify staffing patterns or initiate SDD protocols.

Your data showed an approximate $3500 savings related to SDD. Can you break it down?

The largest savings come from reducing length of stay. This is the most powerful driver of costs. When we use technology or devices such as radial access that allow more efficient care and earlier discharge, then it leads to less resource consumption and a large magnitude of cost savings. The other big driver of costs is complications after PCI such as bleeding and vascular complications. Previous studies have quantified that these complications independently increase the costs of care. A bleeding complication increases costs of care by $5,000 to $15,000, depending on the definition of bleed used. Vascular complications are expensive as well and are the other driver of costs. In the recent study, we observed a 50-60% reduction in complications with transradial access, consistent with reports from prior randomized trials and observational studies. I must say that the evidence that transradial access reduces PCI complications is irrefutable; every single study ever performed consistently shows a reduction in PCI complications. Since transradial access reduces both complications and facilitates earlier discharge, it is no surprise that we see the magnitude of savings with TRI SDD is extremely large. If per-case cost differences are so large in magnitude, for a large-volume center, and by extension the whole country, it could quickly add up to millions of dollars.

SDD (same-day discharge) seems to be the pivot in achieving cost savings, regardless of the arterial approach. How do you explain the relatively low percentage of SDD use in your data and what is the key to increasing the use of SDD?

You are correct that SDD is the bigger driver of costs, whether by radial or femoral approach. Our data underscores the extremely large opportunity we have in terms of SDD. We saw that only 5% of our population was being discharged the same day after elective PCI, the denominator being all patients eligible for SDD. Thus, the room for improvement is remarkable! 

I can speak from personal experience. At Barnes Jewish Hospital, we realized that though SDD was safe, less costly, and preferred by patients, it was being performed only in low-risk patients, if at all. It was being used in less than 1% of our population. To increase the appropriate use of SDD and enable SDD to be utilized in higher risk patients, at Barnes Jewish Hospital (BJH), we implemented a novel “patient-centered” protocol based on risk for complications by using a health IT solution called ePRISM. The SDD program at BJH is based upon: (1) proactively identifying risks of bleeding, mortality, acute kidney injury (AKI), and using a personalized safe limit of contrast; (2) mitigating those risks by bleeding avoidance strategies (BAS) and limiting contrast; and, (3) if risks are successfully mitigated, performing SDD. Since implementation, we have seen radial access increase, as did SDD. It has allowed us to increase our SDD proportion from 1% to over 70%. We have seen no difference in adverse outcomes, and we see a tremendous improvement in patient satisfaction and cost savings. 

Having said that, every case needs to be approached individually, all factors and concerns pertaining to SDD should be carefully assessed, and SDD considered when felt to be safe by all stakeholders, the most important one being the patient.

Do you see any correlative value to these data (elective PCI), and the recently approved cardiac bundles for managing acute myocardial infarction through an index PCI admission? 

This is a terrific question! I do see great value of these data in the context of cath labs preparing for the recently announced cardiac bundles for acute myocardial infarction (AMI). Elective PCI is approximately 50% of the volume at most cath labs; the other 50% being acute coronary syndromes and AMI. Our data could be used to prepare for cardiac bundles in the following two ways: First, if we optimize the SDD program, we know it leads to large cost savings. A successful SDD program that reduces costs and frees up resources would then allow those same resources to be used for AMI care while maintaining overall cath lab profitability. Those resources could be rechanneled for reducing costs of AMI care such as rehab and home health, that could then pay higher dividends in preventing readmissions, which in turn, would also drive down the costs of AMI care. Second, lessons learnt in the elective PCI population on optimizing care pathways could then be applied to the ACS population as well. If ACS patients can undergo transradial access, they will likely experience lower complications and can be considered for lower acuity beds, earlier discharge, and shorter length of stay. Optimizing care pathways for the ST-elevation MI (STEMI) and non-STEMI patients would perhaps also lead to cost savings similar to the findings seen in our paper in the elective PCI population. There may be a tradeoff between shorter length of stay and readmisssion prevention, and we cannot lose sight of preventing readmission when trying to achieve a shorter length of stay. Appropriate follow-up within a week after PCI and focus on medication compliance is critically important as well. In the era of cardiac bundles, this multifaceted approach is going to be the key to success.

You identified one of your study limitations as the costs associated with a greater societal impact as it relates to use of TRI. Can you elaborate?

I believe that we as physicians owe society lower costs of care we deliver. I have long thought that “the system”is responsible for  these exorbitant costs, but over the last few years, I have realized that ultimately I am the one delivering care, so shouldn’t I take responsibility for costs, just as I do for outcomes? And I think that by becoming cost conscious and fiscally responsible, eventually everyone wins! In health economic theory, there are “variable” and “fixed” costs. While we may have little control over fixed costs, the variable costs are within our control. But please note that recommendations based on cost alone are not appropriate. Better patient care at a lower cost should be our goal. Transradial access with same-day discharge is one way of doing that. In health economic theory, such therapies are considered “economically dominant”.

You are correct in pointing out that the costing methodology in our paper was from a hospital perspective and not the broader societal perspective. Some downstream effects of TRI SDD not captured in our costing methodology are reduced readmissions from lower complications, lower indirect costs such as the expenses incurred from the cessation or reduction of work, productivity losses as a result of the morbidity and mortality associated with longer length of stay, and even family members’ transportation costs and work productivity losses. These indirect costs from reduced complications, readmission prevention, work loss, and reduced productivity are typically valued from either societal, individual, or employer perspectives, and have not been captured in our study.

Can you elaborate on your findings regarding the use of bivalirudin and vascular closure devices when used in support of femoral access?

This is an interesting question. If bleeding avoidance therapies can be consistently and effectively used with femoral access, then SDD after femoral access should be feasible, too. There are some single-center studies that describe SDD in patients undergoing transfemoral PCI. However, the question remains as to whether bleeding and transfusion can be consistently prevented via transfemoral access, especially as we try to ramp up SDD in a broader, all-comer population. In our study, we observed a lower rate of bleeding and transfusion in the TRI group, despite frequent use of other bleeding avoidance strategies such as 67% bivalirudin use and 58% closure device use in the TFI group. 

While overall SDD was low in our population, it was lower still with TFI than with TRI (4.5% vs 13.5%). These data highlight the limitation that bleeding avoidance therapies (BATs) cannot always be consistently used via a femoral approach and perhaps the interventionalists’ comfort with SDD is lower via the TFI approach. Transradial access may be the solution to this problem. Personally, when I send patients home the same day of PCI after transradial access, I do not have to worry about bleeding complications. It is one less thing to worry about. 

A counterargument could be whether transradial access is feasible for the vast majority of PCI patients. Although the feasibility of TRI in older patients with more complex coronary disease has been challenged, several studies have established the success of TRI in older patients and those with complex coronary anatomy, including bifurcations, bypass grafts, left main disease, long lesions, and calcified vessels. Certainly the studies on the transradial learning curve indicate these challenges may be overcome and that transradial access is feasible in the vast majority of PCI patients. Our findings further extend and confirm lower complication rates in a large national population and establish that better outcomes may be achieved at a lower cost.

What are the barriers to adoption of TRI?

There is a perception that transradial PCI is more technically challenging than transfemoral PCI. The reality is that with the technological and engineering advances that have occurred over the last decade in guide catheters, guidewires, balloons, stents, and PCI device technology, these challenges may be easily overcome. There are studies that characterize the transradial learning curve, and suggest that it is not very steep and that most cardiologists can gain proficiency in weeks to months.

Higher radiation exposure to cath lab staff is another barrier I have seen being discussed. There is some truth to this: prior studies have shown slightly longer procedure and fluoroscopy time, and radiation dose, but this mostly occurs during the initial phase of the learning curve. As operators become more proficient, an equivalence of radiation dose between transradial and transfemoral is achieved. A survey from the VA cath labs in the U.S. showed that interventional cardiologists practicing in cath labs with high rates of transradial PCI rated radial and femoral access as equivalent on procedure time and technical results. This has been my own personal experience as well. 

I think these are the two greatest barriers to increasing transradial and SDD adoption, which may be overcome if we educate and emphasize that procedure times and technical success correlate with achieving proficiency.  

There has been a great deal of discussion in regard to SDD protocols and which patients can be sent home safely. Can you describe your practice?  

Discharging a patient home within hours of the PCI is a very personal and complex decision that is based on many factors, including patient, physician, and administrative/hospital factors. Protocols for such a decision are difficult and this is the reason why most SDD protocols can’t be applied universally across hospitals. Additionally, these protocols also tend to be too conservative and do not match up to the complexity of patients seen in real-world practice. In our own practice at BJH, less than 10% of the population met criteria for SDD according to existing SDD expert consensus guidelines. This is also reflected in the NCDR CathPCI registry data in our study, where less than 10% of the population was sent home the same day.

Within our own practice, we have seen an increase in complexity and comorbidity burden among our elective PCI population. Patients seem to be surviving longer — perhaps a result of improvements in therapy and better survival in patients with heart failure, cancer, diabetes etc. — allowing them to accumulate an increasing burden of comorbidities. To overcome these barriers, at Barnes Jewish Hospital, we use a customized or individualized “patient-centered” approach, where all potential risks of complications are proactively and explicitly identified before the PCI and mitigated successfully with use of multiple bleeding avoidance therapies or lowest possible dose of contrast. In addition, our mobile pharmacy delivers a 30-day supply of post-PCI medications to the patient at the bedside prior to discharge, and we ensure early follow-up within less than 7 days and follow-up via phone calls for the first 1-4 days to ensure the patient is doing well after discharge. Having a successful SDD program also requires that cath lab processes and workflow issues are streamlined. According to Samantha Miller, the cath lab manager at BJH, “Elective cases need to start on time, workflow between cases needs to be smooth, and cases need to finish in a timely manner so that an appropriate observation time is available prior to SDD.” For a successful SDD program, this is vital. As you are aware, the first few hours after PCI are the most important and any complications that have the potential to occur, will occur in the first few hours. According to Patricia Crimmins-Reda, the BJH Heart & Vascular service line director, “Aligning cath lab staff and streamlining workflow plays an almost equally important role in SDD as does the clinical decision making.”

Lastly, and perhaps most importantly, involving patients in this decision making from the get-go is critically important. Says Samantha Miller, the cath lab manager at BJH, “We discuss SDD vs overnight observation, and elicit patient preference early on in the cath lab holding area where patients are prepped. I have yet to see any patient request overnight observation, though!” Brandon Rahn, the manager of performance, states, “We have seen our patient satisfaction scores increase rapidly, too. I believe this stems  not only from earlier discharge, and less pain and discomfort from transradial access, but also from all the processes we have aligned that reassure patients they are well cared for!”

Aligning all these factors, takes time and effort, but by doing so, we have been successful in effectively discharging almost 3 out of 4 elective PCI patients home the same day, despite some of the challenges that we face in terms of patient complexity and comorbidity burden, as well as social and geographic factors.  

These data and the cost savings represented should garner significant attention from hospital administrators. Can they really be achieved?  

These data are very real. We have demonstrated in our paper the differences in costs associated with various care pathways. But while designing the study, we needed to ensure that patients were truly eligible for SDD and we had to perform numerous exclusions. On the other hand, in real-world practice at BJH, we have expanded our approach to a broader, more complex, elective PCI population as described above. We are now in the process of closely examining our patient satisfaction and cost data. Our patients prefer SDD, as reflected in prior studies of SDD, and at our hospital, there appears to be a favorable impact on patient satisfaction scores. Our preliminary cost results are also very consistent with the model projection available in our paper and serve as a good external validation of our published data.  

I must add that to streamline clinical care, and improve the efficiency of the care pathway, there needs to be coordination and cooperation between the physicians and cath lab, strong leadership, and nursing and physician champions who can implement and align these care pathways. The role of leadership in supporting this is critical. At Barnes Jewish Hospital, we have been very privileged to have strong leadership and support from Dr. Jasvindar Singh, the medical director of the cath lab, Dr. Douglas Mann, the divisional chairman, and Patricia Crimmins-Reda, the Heart & Vascular service line director.

Final Thoughts
Gary Clifton, Vice President, 
Terumo Business Edge

As you can see from Dr. Amin’s interview, SDD represents a significant opportunity for cost-saving efficiencies, improving patient experience, streamlining staffing resources, and other associated direct and indirect benefits. SDD may even be a necessity for cath labs interested in cost containment and meeting the demands of an ever-growing, consumer-driven patient population. Terumo Business Edge and MedAxiom Consulting are here to help your lab make the transition.