Journal of Vascular and Endovascular Therapy

Abstract

Background: Outpatient endovascular interventions have emerged as a preferred treatment option for patients with vascular disease in recent years. Procedures performed in this relatively new site of service are without adequately historical data to assess clinical best practices. Quality data collection with subsequent outcomes analysis, benchmarking, and direct feedback, is necessary to achieve optimal care. Outcomes-based registries can serve this purpose.

Methods and findings: The Outpatient Endovascular and Interventional Society (OEIS) established the OEIS National Registry in 2017 to collect data on safety, efficacy, and quality of care for outpatient endovascular intervention. Since then, it has grown to include a peripheral arterial disease module with plans to expand to include cardiac, venous, and dialysis access procedures in future modules. As a Centers for Medicare and Medicaid Services-approved Qualified Clinical Data Registry, it also supports new quality measure development under the Quality Payment Program. All physicians operating in an officebased lab or ambulatory surgery center can use the Registry to analyze de-identified data and benchmark performance against national averages. Since inception in 2017, 249 participating physicians from 63 centers located in 18 states have entered over 15,000 peripheral arterial interventions. Participants use the Registry analytics to view de-identified summary data and benchmark performance against national results.

Conclusions: This report describes the current structure and methodology of OEIS National Registry, an outcomes-based registry designed to collect quality performance data with subsequent outcome analysis, benchmarking, and direct feedback to aid clinicians in providing optimal care.

Keywords

Registry; Outpatient; Endovascular interventional office procedures

Abbreviations

CMS: The Centers for Medicare and Medicaid Services; EDS: Electronic Data System; EMR: Electronic Medical Health Record; MIPS: Merit-Based Incentive Payment System; NCDR: National Cardiovascular Data Registry; NSQIP: National Surgical Quality Improvement Program; OEIS: Outpatient Endovascular and Interventional Society; PAD: Peripheral Arterial Disease; QCDR: Qualified Clinical Data Registry; STS: Society of Thoracic Surgeons; VQI: Vascular Quality Initiativet.

Introduction

Outpatient endovascular interventions have grown rapidly into an increasingly prevalent treatment option for vascular disease [1-4]. Advances in pharmacology and technology for endovascular care and the changes in reimbursement by the Centers for Medicare and Medicaid Services (CMS) to shift the cost of care to outpatient sites of service have also led to the growth in outpatient endovascular procedures [1]. Although outpatient and office-based interventional suites have been operational for many years, there has been a marked proliferation of these sites nationwide during the past decade [5]. It is estimated, based on informal industry surveys and Medicare claims data, that nearly 750 office-based labs currently exist in the United States and that number continues to grow. Office-based labs, also referred to as office-based endovascular suites or outpatient interventional suites, offer distinct advantages for patient care being more efficient and cost-effective than hospital-based interventions [5]. In addition, patient satisfaction is consistently very high in these centers [6].

This increased utilization of outpatient and office-based sites of service has raised questions about potential overutilization, adverse patient outcomes, and the overall quality of care provided in this environment [7-9]. Though published reports from individual centers have shown excellent clinical results with very low morbidity and mortality [4,6,10], there is little state-wide or national data available for analysis and comparison. Similarly, there are no established guidelines nor metrics from which to assess best clinical practices. Broad-based patient and procedural data collection with associated outcome analysis, benchmarking, and clinician feedback is necessary to achieve optimal care. Outcomes-based registries can serve this purpose [11].

Over the past two decades, physician and professional societies’ voluntary clinical registries such as the Society of Thoracic Surgeons (STS) National Database, American College of Surgeons National Surgical Quality Improvement Program (NSQIP), Society for Vascular Surgery Vascular Quality Initiative (VQI), and the American College of Cardiology National Cardiovascular Data Registry (NCDR) have demonstrated the value of such initiatives in changing physician behavior and improving patient outcomes [12-32]. These established registries have focused primarily either on inpatient outcomes (NSQIP), specialty-specific procedures (STS), or a combination of both (VQI and NCDR).

However, none have been directed at a multi-specialty set of procedures performed by vascular surgeons, interventional cardiologists, and interventional radiologists in an office outpatient environment and, thus, they do not meet the unique needs of outpatient endovascular providers. Existing quality outcomes measures generally tend to be geared toward medical and inpatient hospital interventions rather than outpatient procedures [33]. Measured surgical outcomes focus primarily on open procedures rather than minimally invasive interventions. In addition, NSQIP requires follow-up data collected at 30 days [20]. Therefore, this does not allow for long-term outcome research or analysis to determine best treatment. Although the VQI requires follow-up at 1 year and allows further follow-up at data to be entered up to 5 years, this national registry is costly and requires a full-time on-site data entry manager [25]. Participation in existing registries often requires hospital financial support and are cost prohibitive to most private practice or office-based endovascular centers. Thus, there is a need to develop a patient-centered, physician-friendly and cost-effective registry specifically for outpatient endovascular procedures.

Outpatient Endovascular and Interventional Society and Registry

The Outpatient Endovascular and Interventional Society (OEIS) is a multi-specialty society composed primarily of vascular surgeons, interventional cardiologists, and interventional radiologists. It was established in 2013 with a mission to address the unique needs of patients and physicians working in the outpatient environment, to enhance the safety, quality and efficacy of outpatient endovascular procedures, and to improve healthcare quality by creating and adhering to professional quality standards. Such quality standards are summarized and referred to as SCOCAP: Safety, Credentialing, Outcomes Measures, Compliance, Appropriateness, and Peer Review. These guidelines promote excellence in outcomes through procedure selection, clinical appropriateness and safety.

To achieve these aims, OEIS in 2017 established a national registry of office-based and outpatient procedures to collect data on safety, efficacy, and quality of care for outpatient endovascular interventions for peripheral arterial disease (PAD). As a CMSapproved Qualified Clinical Data Registry (QCDR), it also supports new quality measures development under the Quality Payment Program [34]. The Registry consists of de-identified clinical and procedural data entered at the time of treatment and followup office visits from physicians operating in an office-based laboratory or ambulatory surgery center. This report describes the structure and methodology of the OEIS National Registry.

Method

Registry design

Site enrollment: The OEIS National Registry collects data related to endovascular interventions performed specifically in an outpatient setting. Centers operating as office-based labs or ambulatory surgery centers performing eligible procedures are able to enroll as participating sites. Each site contracts with OEIS for services and safeguarding protected health information in compliance with HIPAA while allowing the Registry to analyze and report on de-identified cumulative clinical data. Registry participants must agree to enter required data for all eligible cases, not just selected procedures, and comply with regular internal audit guidelines and requests for source documentation. Physicians of all specialties who perform endovascular procedures may participate. Figure 1 shows a summary of key features.

Figure 1: Registry features overview of key registry features.

Peripheral arterial disease module: The first module developed focuses on peripheral arterial diagnostic and interventional procedures including diagnostic angiography, balloon angioplasty, stent placement, and atherectomy in patients with lower extremity arterial disease. Data elements and definitions were developed utilizing established data standards [35]. The electronic data system (EDS) was structured to capture what are defined as data essential fields, which include all minimum data entry requirements, as well as a more expansive set of optional enhanced data fields. Data fields used to calculate supported Merit-Based Incentive Payment System (MIPS) and QCDR measures were included according to CMS standards.

QCDR-MIPS: The Registry has been a CMS-approved QCDR for the MIPS since 2017. QCDRs are one of several methods available for participation in MIPS which can collect and function by clinical data on behalf of participating physicians for submission to CMS for the purpose of improving the quality of patient care and establishing physician reimbursement adjustment levels. QCDRs enable collection of patient data regardless of patient insurance coverage, not just Medicare beneficiaries. Unlike other submission methods, a QCDR is uniquely able to develop custom quality measures that are subject to annual review and approval by CMS. This process allows QCDRs to create patient-relevant quality measures tailored to provide more meaningful feedback that participants can apply directly in their practices. The OEIS National Registry is the first and only QCDR to focus solely on detailed outcomes within outpatient interventional suites. For 2019, CMS approved three QCDR measures submitted by OEIS: QM OEIS2 - Emergent transfer from an outpatient, ambulatory surgical enter, or office setting; QM OEIS6 - Appropriate noninvasive arterial testing for patients with intermittent claudication who are undergoing a lower extremity peripheral vascular intervention; QM OEIS7- Structured walking program prior to intervention for claudication.

Data collection and transmission: Detailed data is collected at the time of initial treatment and during subsequent follow-up office visits. Variables were selected to provide robust information regarding patient demographics, clinical history, diagnostic testing, procedure indications, procedural details, complications, and clinical follow-up (Figure 2). Data is submitted using a cloudbased EDS system (Syncrony; Syntactx Technologies, New York). OEIS contracted to develop and maintain the registry data entry system and database using this customizable and fully secure system [36].

Figure 2: Data collection f Figure 2 orm, peripheral arterial disease data collection form containing the required registry procedural data.

Registry users can log into the EDS through a secure web page using any device or computer with web access. Once the procedural data is submitted, the Registry receives deidentified data for analysis. Data is regularly aggregated and reports with summary statistics, benchmarking, and quality measure feedback are provided to users via web portal.

Database structure and analytics: The registry utilizes an enterprise-grade cloud-based NoSQL database architecture (MongoDB Atlas/Enterprise; MongoDB, New York) as the database engine and back-end. Datapoints collected through the EDS are formatted, stored, and analyzed from the database platform. The current data architecture is optimized to add additional future modules.

Currently analytics are performed, and the results are published to dashboards contained within the EDS as described above. These reports are accessible to registry participating physicians and sites for their own data. Additional graphical and statistical reports are separately made available to the Registry Committee and OEIS Board of Directors. Individual sites have access to their own data as well as de-identified aggregate measures from the Registry as a whole (Figure 3). The reports include a multitude of clinically relevant metrics and information, basic and advanced demographics, and the individual site/physician performance on supported Quality Measures as well as summary comparative benchmarking data compared to national averages.

Figure 3: Sample registry report, sample generated report demonstrating a clinical summary view that individual registry participants have available to them.

Regulatory oversight

Organizational oversight: The Registry is organizationally administered and directed in a layered and complementary fashion. The structure consists of three supervisory layers: A Medical Director, a Technical Director, and the Registry Committee. The OEIS Board of Directors appoints and approves both the Medical and Technical Directors and all members of the Registry Committee.

Roles of medical director and technical director: The Medical Director and Technical Director provide direction and clinical and administrative oversight and are responsible for Registry operations. Both positions are currently filled by OEIS member physicians and function in an overlapping but complementary fashion. The Medical Director is charged with providing overall direction, strategic vision and goals, global medical oversight, and advocacy and/or negotiative functions. These duties include interfacing with existent and potential registry sites/ participants, interested physician(s), physician groups, health care organizations, other health industry entities, CMS and regulatory bodies, and other professional societies and related registries.

The Technical Director is charged with the day-to-day operation of the Registry, maintenance, upgrading, and development of the underlying IT interfaces, including the web interfaces and back-end data structure. Primary responsibilities also include direct interaction and technical/medical direction to the Registry Manager, interaction, direction, and negotiations with technical vendors, quality measures, and QCDR functionality. The Medical and Technical Directors receive input and feedback from the Registry Committee with additional direct reporting to the OEIS Board of Directors.

The Registry Committee is comprised of physician members of OEIS appointed by the President. The role of the committee is to oversee and give input regarding the strategic direction of the registry, provide medical expertise in the development and evolution of both existing and future modules, QCDR quality measure development, internal quality measures and adjudication, and feedback to the Medical and Technical Directors. The chairperson reports on committee function and registry activities to the OEIS Board of Directors.

Data standards and validation: Data management is conducted internally and in conjunction with the Syntactx Ltd contract research organization. The web-based EDS database is used to record and manage data and also provides an audit trail. The EDS data can be exported to various file formats for statistical analysis.

Data integrity is achieved through several avenues, including data entry staff training, clinical feedback, and data entry system edit checks and help functions. The Registry EDS is designed to prevent error during entry using indicators of required field completion, detailed help text and definitions, and accepted value limits on data fields. All Registry users are required to complete webbased training on Registry protocol, data collection methods, and data entry standards to ensure accuracy and consistency. Existing users receive regular updates and access to individual assistance and additional training when needed. Participating sites also receive real-time online feedback reports containing a summary of commonly observed errors, such as missing or out of range dates resulting from human error, that they can then use to review their data independently and make corrections.

Internal audits are conducted at regular intervals to ensure data completeness and accuracy. During this process, randomly selected records entered into the Registry database are reviewed to ensure that data accurately reflects the contents of the subject’s medical record or other source documentation maintained by the submitting practice. Error detection triggers a subsequent additional action to determine the scope and source of the error. Once identified, centers receive improvement feedback, additional training, and/or corrective action, as appropriate.

Participating site reporting: Participants can obtain immediately usable data in real time through structured reports provided via the Registry web portal. The Registry reports offer detailed clinical and quality performance feedback reflecting both individual site-level data and aggregated national averages for comparison. Reports and can be customized and filtered by date range, individual physician, and site name (available for large groups with multiple practices). Dynamic interactive functionality allows for cross filtering to explore trends in the data. The tabular reports consist of page views showing a clinical summary overview plus detailed views of patient population, lesion, intervention, and complication data. The MIPS quality measure performance dashboard offers a summary view with functionality allowing participants to drill down and granularly review their data. The Registry national benchmarking reports closely reflect the content and structure of the individual center reports and facilitate comparison between physicians within one center, centers within a group, and anonymously between physicians/centers versus national averages.

Cost: The cost to participate in the Registry as of 2019 is $295 per month charged to each participating site, defined as operating under a unique tax ID number, regardless of procedure volume or number of enrolled physicians. Sites where all enrolled physicians are current members of OEIS, however, are eligible for a discounted monthly rate of $175 per month for Registry participation. This fee covers access to the registry database, web-based services, and reporting.

Enrolled providers and groups may also elect for OEIS National Registry QCDR MIPS data submission services once annually for $399 per each physician NPI number. These annual costs are substantially lower than those for other large U.S. registries, especially considering there is no initial set-up cost associated with new site enrollment and centers are only responsible for the nominal monthly fee for registry participation.

Results

Current status: In just over two years, the Registry has recruited 249 participating physicians from 63 centers located in 18 states (Figure 4a). The most common physician specialties contributing data include interventional cardiology, vascular surgery, and interventional radiology (Figure 4b). The current database includes 15,288 peripheral endovascular interventions performed in 10,629 patients with PAD between January 2017 and August 2019 (Figure 5).

Figure 4: Registry participation summary, (a) geographic distribution of centers currently enrolled and participating in oeis national registry. distribution increase represented by color shift from light (< 1) to dark (>21). (b) physician specialty breakdown of registry participants.

Figure 5: Case volume, case volume entered into the registry. *2019 value includes cases entered through July 31, 2019

Discussion

Future directions

Additional module development: The current PAD module was designed and developed to act as a framework on which future additional modules can be constructed. Near future plans include expansion to add modules supporting cardiac (coronary interventions and rhythm management procedures including device implantation), venous (deep and superficial venous system procedures including inferior vena filter management, deep vein thrombosis management, and superficial ablative procedures), and dialysis interventions (including arteriovenous fistula/ arteriovenous graft formations and interventions and catheter management). The cardiac module is next in development and will focus on same day interventions in the outpatient setting.

QCDR measures: As the registry database grows, new opportunities arise for QCDR quality measure development based on historical data demonstrating gaps in care. CMS re-evaluates measures for inclusion in MIPS on an annual basis. Measures with average performance rates that are too high to demonstrate a gap in care are typically rejected and are not eligible for the topped-out measure timeline as it applies to MIPS measures, so there is a more urgent interest in developing new concepts as high performing measures are phased out. New measure development is an ongoing process and is an opportunity for the OEIS to help lead the way in advancing value-based care.

Electronic medical record integration: The current process for manual data entry does require some time and resource allocation. Most Registry participating sites also utilize electronic medical health record (EMR) systems to collect clinical data.

Developing avenues for electronic integration to reduce the need for redundant manual entry would ease the workflow strain associated with clinical registry participation. Direct integration with existing EMR systems, however, can be difficult to implement due to the variation in EMR vendor platforms, the individual centers’ custom configurations, and lack of data uniformity. For example, data stored in unstructured clinical notes is not readily parsed by direct automated extraction without developing advanced natural language processing algorithms. This approach, however, does not work consistently due to differences in terminology and/or definitions between participating physician and sub-specialties. Therefore, the Registry is working with select vendors to develop custom EMR forms based on registry specifications. This would allow centers using these forms to capture all required data into their existing EMR simultaneously with entry into the Registry thus significantly lowering participation cost and reducing error associated with manual data abstraction and entry.

Research opportunities

The Registry’s large sample sizes, routine review for data completeness and accuracy, and detailed dataset, all create excellent opportunities for directed clinical research with the potential for improved patient care protocols. Topics will include short- and long-term safety and efficacy, device usage, and appropriateness. These results will soon be analyzed and reported in future publications.

Conclusions

The OEIS National Registry is now the standard for quality and outcomes analysis related to outpatient endovascular interventions. Registry data can provide new insights into the safety and efficacy of outpatient endovascular interventions. Opportunities abound for investigators to utilize Registry resources to develop quality standards for best practice in an outpatient setting and provide operator and site-level feedback to drive quality improvement and positive patient outcomes.

Funding

This study was funded by the Outpatient Endovascular and Interventional Society.

Competing and Conflicting Interests

Dr. Samuel Ahn is the Founder and CEO of Vascular Management Associates. Lauren Jones is an employee of the Outpatient Endovascular and Interventional Society. Dr. Jeffrey Carr, Dr. Robert Tahara, and Dr. John Blebea have no disclosures.

References

1. Mukherjee D, Hashemi H, Contos B (2017) The disproportionate growth of office-based atherectomy. J Vasc Surg 65: 495-500.
2. Jain KM, Munn J, Rummel M, Vaddineni S, Longton C (2010) Future of vascular surgery is in the office. J Vasc Surg 51: 509-514.
3. Mesbah Oskui P, Kloner RA, Burstein S, Zhiroff K, Kartub BR, et al. (2015) The safety and efficacy of peripheral vascular procedures performed in the outpatient setting. J Invasive Cardiol 27: 243-249.
4. Lin PH, Yang KH, Kollmeyer KR, Uceda PV, Ferrara CA, et al. (2017) Treatment outcomes and lessons learned from 5134 cases of outpatient office-based endovascular procedures in a vascular surgical practice. Vascular 25: 115-122.
5. Schudy JB (2018) How Office Based Procedural Suites Are Impacting the Health Care Industry: Central Michigan University.
6. Jain K, Munn J, Rummel MC, Johnston D, Longton C (2014) Office-based endovascular suite is safe for most procedures. J Vasc Surg 59: 186-191.
7. Lawrence P (2016) "Better" (sometimes) in vascular disease management. J Vasc Surg 63: 260-269.
8. Creswell J, Abelson R (2015) Medicare payments surge for stents to unblocked blood vessels in limbs.
9. Lapetina EM, Armstrong EM (2002) Preventing errors in the outpatient setting: a tale of three states. Health Aff 21: 26-39.
10. Criado FJ, Abdul-Khoudoud O, Twena M, Clark NS, Patten P (1998) Outpatient endovascular intervention: is it safe? J Endovasc Ther 5: 236-239.
11. Kerezoudis P, Devin CJ, Goncalves S, Alvi MA, Asher AL, et al. (2018) The role of clinical registries in health care. In quality and safety in neurosurgery. Academic Press: 53-67.
12. https://www.sts.org/registries-research-center/sts-national-database.
13. Badhwar V, Rankin JS, Thourani VH, D’Agostino RS, Habib RH, et al. (2018) The society of thoracic surgeons adult cardiac surgery database: 2018 update on research: outcomes analysis, quality improvement, and patient safety. Ann Thorac Surg 106: 8-13.
14. Onaitis MW, Furnary AP, Kosinski AS, Kim S, Boffa D, et al. (2018) Prediction of long-term survival after lung cancer surgery for elderly patients in the society of thoracic surgeons general thoracic surgery database. Ann Thorac Surg 105: 309-316.
15. Gammie JS, Chikwe J, Badhwar V, Thibault DP, Vemulapalli S, et al. (2018) Isolated mitral valve surgery: the society of thoracic surgeons adult cardiac surgery database analysis. Ann Thorac Surg 106: 716-727.
16. Cooper DS, Riggs KW, Zafar F, Jacobs JP, Hill KD, et al. (2019) Cardiac surgery in patients with trisomy 13 and 18: an analysis of the society of thoracic surgeons congenital heart surgery database. J Am Heart Assoc 8: e012349.
17. Kormos RL, Cowger J, Pagani FD, Teuteberg JJ, Goldstein DJ, et al. (2019) The society of thoracic surgeons intermacs database annual report: evolving indications, outcomes, and scientific partnerships. J Heart Lung Transplant 38:114-126.
18. D’Agostino RS, Jacobs JP, Badhwar V, Fernandez FG, Paone G, et al. The society of thoracic surgeons adult cardiac surgery database: 2018 update on outcomes and quality. Ann Thorac Surg 105: 15-23.
19. Jacobs JP, Mayer Jr JE, Pasquali SK, Hill KD, Overman DM, et al. (2018) The society of thoracic surgeons’ congenital heart surgery database: 2018 update on outcomes and quality. Ann Thorac Surg 105: 680-689.
20. Khuri SF, Daley J, Henderson W, Hur K, Demakis J, et al. (1998) The department of veterans affair’s NSQIP: the first national, validated, outcome-based, risk-adjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care. National VA surgical quality improvement program. Ann Surg 228: 491.
21. Liu Y, Cohen ME, Hall BL, Ko CY, Bilimoria KY (2016) Evaluation and enhancement of calibration in the american college of surgeons NSQIP surgical risk calculator. J Am Coll Surg 223: 231-239.
22. Neumayer L, Mastin M, Vanderhoof L, Hinson D (2000) Using the veterans administration national surgical quality improvement program to improve patient outcomes. J Surg Res 88: 58-61.
23. Cohen ME, Liu Y, Ko CY, Hall BL (2016) Improved surgical outcomes for ACS NSQIP hospitals over time. Ann Surg 263: 267-273.
24. Desai SS, Kaji AH, Upchurch G (2018) Practical guide to surgical data sets: society for vascular surgery vascular quality initiative (SVS VQI). JAMA Surgery 153: 957-958.
25. Cronenwett JL, Kraiss LW, Cambria RP (2012) The society for vascular surgery vascular quality initiative. J Vasc Surg 55: 1529-1537.
26. Wang GJ, Cambria RP, Lombardi JV, Azizzadeh A, White RA, et al. (2019) Thirty-day outcomes from the society for vascular surgery vascular quality initiative thoracic endovascular aortic repair for type B dissection project. J Vasc Surg 69: 680-691.
27. Columbo JA, Kang R, Hoel AW, Kang J, Leinweber KA, et al. (2019) A comparison of reintervention rates after endovascular aneurysm repair between the Vascular Quality Initiative registry, Medicare claims, and chart review. J Vasc Surg 69: 74-79.
28. Dhillon AS, Li S, Lewinger JP, Shavelle DM, Matthews RV, et al. (2018) Comparison of devices used in carotid artery stenting: A vascular quality initiative analysis of commonly used carotid stents and embolic protection devices. Catheter Cardiovasc Interv 92: 743-749.
29. Available from: https://cvquality.acc.org/NCDR-Home/Registries.
30. Bhardwaj B, Spertus JA, Kennedy KF, Jones WS, Safley D, et al. (2019) Bleeding complications in lower-extremity peripheral vascular interventions: insights from the NCDR PVI registry. JACC: Cardiovasc Interv 12: 1140-1149.
31. Fan W, Song Y, Inzucchi SE, Sperling L, Cannon CP, et al. (2019) Composite cardiovascular risk factor target achievement and its predictors in US adults with diabetes: The diabetes collaborative registry. Diabetes Obes and Metab 21: 1121-1127.
32. Thourani VH, O'Brien SM, Kelly JJ, Cohen DJ, Peterson ED, et al. (2019) Development and application of a risk prediction model for in-hospital stroke after transcatheter aortic valve replacement: a report from the society of thoracic surgeons/american college of cardiology transcatheter valve therapy registry. Ann Thorac Surg 107: 1097-1103.
33. Aiello FA, Shue B, Kini N, Rosen A, Messina L, et al. (2014) Outcomes reported by the vascular quality initiative and the national surgical quality improvement program are not comparable. J Vasc Surg 60: 152-159.
34. https://qpp.cms.gov
35. Jones WS, Krucoff MW, Morales P, Wilgus RW, Heath AH, et al. (2018) Registry assessment of peripheral interventional devices (RAPID): Registry assessment of peripheral interventional devices core data elements. J Vasc Surg 67: 637-644.
36. Evans D, Shure D, Clark L, Criner GJ, Dres M, et al. (2019) Temporary transvenous diaphragm pacing vs. standard of care for weaning from mechanical ventilation: study protocol for a randomized trial. Trials 20: 60.