Adult Clinical Perfusion Practice Survey: 2020 results

Background: Cardiovascular perfusion is a dynamic healthcare profession where new practices are frequently introduced. Despite the emergence of evidence-based clinical practice guidelines, little is known about their dissemination at the institutional level. Clinical practice surveys have been used to identify current trends in perfusion practice in the areas of equipment, techniques, and staffing. This survey aims to describe clinical perfusion practices across adult cardiac surgical programs located in a large, single, geographical region of the United States. Methods: Following Institutional Review Board (IRB) approval, an 81-question survey was distributed to 167 adult perfusion programs across the Zone IV region of the American Society of Extracorporeal Technology (AmSECT), a non-profit professional society representing the extracorporeal technology community. Surveys were distributed to chief perfusionists through the Research Electronic Data Capture (REDCap) web-based survey response system. Results: Responses were received from 58 of 167 centers across (34.7% response rate). Centrifugal pumps were used at 81% (n = 47) of centers and 96.6% (n = 56) use an open venous system or hard-shell venous reservoir. Del Nido was the most frequently used cardioplegia strategy with 62.1% (n = 36) of centers reporting its use. The use of electronic medical records was reported in 43% (n = 25) of centers, while 84.5% (n = 49) reported using Cardiopulmonary Bypass (CPB) protocols (>75% of all CPB activities). Extracorporeal Membrane Oxygenation (ECMO) support was reported in 93.1% (n = 54) of programs, with 59.2% of programs (n = 34) employing a perfusionist as ECMO Coordinator. The n + 1 staffing model was reported by 50% (n = 29), with 24% supporting the n + 1 staffing for after-hours and on-call procedures. Conclusion: Clinical practice surveys can be effective tools to inform clinicians about contemporary perfusion practice and identify deviations from professional standards and guidelines. Subsequent surveys may describe trends over time, assess standardization of practice, measure adherence to evidence-based guidelines, and foster improved patient care and outcomes.


Introduction
Cardiac surgery is one of the most performed inpatient operations in the United States [1]. Despite large-scale improvements in outcomes over the nearly seven decades since the first cardiopulmonary bypass (CPB) procedure [2], appreciable interhospital variability persists in both the conduct of and outcomes associated with CPB [3]. Several efforts have emerged to advance the conduct and safety of CPB, including clinical registries, evidence-based guidelines [4][5][6] as well as professionally based standards and guidelines [7][8][9].
Despite the promulgation of these resources for practicing clinicians, little is known about contemporary CPB practices for adult cardiac surgery. While several large-scale clinical registries have emerged, publications to date often focus on a discrete area of practice (e.g., nadir hematocrit during CPB) rather than broadly covering the conduct and practice of CPB. With several topic-specific (e.g., blood management) guidelines have been published, evaluations of the implementation into practice have been infrequent [10][11][12]. Finally, while The American Society of ExtraCorporeal Technology (AmSECT) has developed professional standards and guidelines for adult CPB, few studies have evaluated real-world practice patterns. This survey aim to describe clinical perfusion practices across adult cardiac surgical programs located in a large, single, geographical region of the United States.

Materials and methods
Following exempt status approval from the Institutional Review Boards (IRB) of the University of Michigan (HUM00194742) and the Medical University of South Carolina (Pro00107747), an 81-question, the closed-question survey was distributed through the Research Electronic Data Capture (REDCap; Nashville, TN). REDCap is a web-based and secure application for data capture and clinical research database development [15]. The survey topics included program staffing and demographics, equipment, techniques and monitoring, and clinical protocols (Appendix). The unit of analysis were centerlevel adult cardiovascular perfusion programs located in the eastern and mid-Atlantic regions of the United States. A database repository consisting of chief perfusionists (or designee) contact information from programs located in the Zone IV region of AmSECT was previously collected and maintained at the University of Michigan (Cardiovascular Perfusion Data Repository: Submission ID: REP00000060). AmSECT Zone IV comprises 15 states and the District of Columbia across the eastern region of the United States, ranging from Maine to South Carolina. A total of 234 cardiac surgical programs were identified in the Zone IV region (Appendix).
The survey questionnaire requested program and procedural data that described clinical practice for the 2020 calendar year. Several questions were also included to identify practice trends over the previous three-year period. Subjects were recruited directly via email invitation on June 8th, 2021. The survey invitation remained open for five weeks, with a closure date of July 16th, 2021. To maximize response rates, non-respondent subjects received up to three notices for participation prior to survey closure. The initial survey was sent to 234 cardiac institutions; however, removing pediatric programs, duplicate or erroneous entries, and missing data resulted in 167 confirmed centers.
Descriptive statistics were used to analyze survey responses. Responses collected from the University of Michigan REDCap repository were de-identified by a database analyst and provided to study investigators. De-identified data were imported into SPSS (IBM SPSS Statistics for Macintosh, Version 27.0, Armonk, NY: IBM Corp.) for analysis and reporting of completed surveys.

Results
Responses were received from 58 of 167 centers for an overall response rate of 34.7%. Survey responses were received from 13 states with the number of responses between states ranging from 1 to 16 (Table 1). Pennsylvania, New Jersey, New York, and Massachusetts represented 67.3% (n = 39) of the responses received.

Staffing and demographics
Thirty-two (55.2%) centers reported performing 600 CPB procedures or less in 2020, with 16 (28%) reporting less than 300 ( Figure 1). Centers performing 150 CPB standby or less were reported in 35 (60.3%) of responses ( Figure 2). The reported distribution of surgical case type is reported in Figure 3.
The mean number of full-time perfusionists was 6.43, with teams ranging from 1 to 33 perfusionists ( Figure 4). Eleven programs (19%) had part-time perfusionists on staff, with 32 (55.2%) reporting the use of per diem or locum tenens coverage over the previous three years. Most respondents (n = 43, 72.4%) indicated that the annual number of clinical hours provided by the perfusion team had increased from 2017 to 2020.
The n + 1 staffing model, as defined by Guideline 15.1 in the AmSECT Standards and Guidelines for Perfusion Practice, establishes the minimum safe number of perfusion staff required to support operating rooms at any one given time per surgical center [7]. This guideline calls for the presence of one staff member above the number of procedures underway [7]. Respondents reported implementing N + 1 during the elective schedule in 50% (n = 29) of centers. A total of 4 (6.9%) respondents indicated using more than the n + 1, and nine programs (15.5%) reported staffing the minimum 1 perfusionist per procedure. The use of the n + 1 model for both elective scheduled and off-hour procedures was reported by 14 (24%) of respondents.
The sites for temperature monitoring on patients are listed in Table 4. The most frequently reported core temperature used was a bladder catheter (n = 54, 93%). The least reported temperature sources were peripheral temperatures, both skin and tympanic, with an incidence of 1.7% (n = 1) and 6.9% (n = 4), respectively.

ECMO utilization and support
Of the 58 centers, 54 (93.1%) offer ECMO at their institution and 59.2% (32/54) had a perfusionist as the ECMO Coordinator. Of those, 39.7% performed 10 ECMO procedures or less ( Figure 6). Most centers (87.9%) indicated that the perfusion team was responsible for ECMO initiation and discontinuation, with 91.4% assigned to troubleshooting (Table 6).
Plasmalyte-A was the most used priming solution ( Table 8). The most common priming additives were heparin, mannitol, and sodium bicarbonate (Table 8). Other additives reported include dexamethasone, solumedrol, MgSO 4 , or no additives at all (Table 8).

Protocol adoption and adherence
All centers reported using a checklist for some part of their practice (Table 12). Most centers used a checklist for assembly/ priming, and initiation of CPB. 65.5% (n = 38) of centers keep a dry assembled circuit for routine on-call coverage, and 34.5% (n = 20) keep a primed assembled circuit. Once a pump is primed, 4 programs (6.9%) keep it for up to 48 hours, and 10 (17.2%) will use it for up to 72 h. Six centers (10.3%) will keep a primed circuit for up to 1 week. Most centers reported 76-100% of their CPB clinical practice is guided by written departmental protocols (Table 13).

Discussion
This survey provides an analysis of adult perfusion practice in the Northeast and Mid-Atlantic regions of the United States. We identified substantial variation in multiple areas of perfusion clinical techniques, equipment, staffing, and adoption of clinical practice guidelines. Due to the paucity of comprehensive surveys conducted across US adult clinical perfusion programs, few opportunities have existed to identify and report trends in operative equipment and techniques. The most recent surveys Abbreviations: CPB = cardiopulmonary bypass. Abbreviations: TXA = Tranexamic acid. conducted outside of the country may not reflect contemporary US perfusion practice [13,14]. North American-centric comprehensive surveys date back nearly 27 years [15]. Regional, multi-institutional clinical registries have previously described similar variations in CPB equipment and clinical management strategies; however, the generalizability of these findings is restricted by the number of participating programs [16,17]. Validated survey tools may broaden the scope of program recruitment and improve our understanding of current perfusion practice. Further, a longitudinal survey design can not only assess clinical trends and guideline adoption, but also identify where the gaps exist for continued consensus development.
The results highlight areas of contrast between published evidence-based guidelines and real-world clinical practice. In 2007, a collaboration between the Society of Thoracic Surgeons (STS) and the Society of Cardiovascular Anesthesiologists (SCA) resulted in a seminal publication of 57 perioperative clinical practice guidelines in cardiac surgical blood conservation and management [18]. Since then, updates to the guidelines were published in 2011 and 2021 [6,19]. Although several perfusion and intraoperative interventions were assigned high-level recommendations (ACC/AHA Class I and IIa), our findings suggest the application of these techniques have not yet achieved widespread adoption. Examples include the use of autologous circuit priming and perioperative viscoelastic testing. Only 34.5% of respondents reported using perioperative viscoelastic testing, and 69% of centers indicated autologous   priming in at least half of all CPB procedures. Both interventions are class I recommendations [16]. Mitigating unwanted variation in clinical practice has been associated with a higher quality of care and lower hospital costs [20,21]. Specifically, several perfusion-related initiatives have highlighted the importance of evidence-based guidelines adherence, outcome reporting, and the reduction in practice variability [22][23][24]. Most respondents indicated that the large majority of CPB care plans are supported by institutional protocols. One of the primary responsibilities of a professional society is to develop standards and guidelines of practice to guide the community in safe and effective patient care. The AmSECT Standards and Guidelines, first formed in 1993, aim to define the minimum requirements for safe cardiopulmonary bypass [25]. These guidelines serve as a framework for developing institutionspecific CPB protocols [7]. Clinical practice surveys can assist in reporting guideline dissemination and inform key stakeholders of opportunities to support their adoption. For example, Standard 12.1 recommends the discontinuation of CPB cardiotomy suction at the onset of protamine administration to avoid circuit thrombus formation [7]. However, most respondents reported continued suction use after protamine initiation, despite the inability to predict ACT responsiveness. Jansa et al. reported a 40% decrease in the ACT value following a partial test dose of protamine, resulting in a value lower than the institutional standard for safe CPB support [26]. While the decision to continue suction use may not ultimately be at the perfusionist's discretion, reducing these discrepancies in care may require further collaboration and endorsement between surgical and perfusion societies.
The survey results also identify several other areas of noncompliance with professional standards and guidelines. Among them are backup CPB battery availability (64%), backup gas supply (81%), medical gas scavenging of the oxygenator output port (83%), and arterial line bubble detection (86%). Each of these elements is recommended by both AmSECT and the EACTS/EACTA/EBCP guidelines as minimum standards for the safe conduct of CPB. These findings may highlight the importance of understanding the barriers that prevent their adoption. Such barriers may include a lack of awareness of the standards, economic constraints, or perceived benefit of their usage. Professional societies may offer opportunities in facilitating the implementation of these practices at the local level.
Practice surveys may also inform the community about techniques that lack guideline support or clear consensus. Substantial variation is observed across various aspects of cardioplegia use. A recent international survey of cardioplegia practices by Ali and colleagues reported significant variation in myocardial protection strategies [27]. Although blood substrates were the most frequently reported formulas, the dilution ratios and cardioprotective additives were highly variable. Similarly, our survey found both center-level and procedural differences in formulas. In particular, del Nido cardioplegia was the most frequently reported formula in both CABG and non-CABG procedures. The transition to del Nido solution in adults is a recent phenomenon, with the first reported case in 2014 [28]. This was preceded by nearly 25 years of experience in pediatric congenital surgery [29]. The pediatric perfusion survey, first conducted by Groom and colleagues in 1990, has described both domestic and international pediatric perfusion for over 30 years [30][31][32][33][34][35]. The 2016 survey reported a 74% use of del Nido in North America, a 2.3-fold increase from the 2011 survey results [34,35]. Similar opportunities for adult surveys performed serially over time would provide valuable insight into the diffusion of new and emerging techniques. Other reported practices lacking clear consensus in the survey included pulsatile perfusion (7%), heparin concentration monitoring during CPB (20.7%), and albumin as a circuit prime additive (47%).
Practice surveys can help describe current and future requirements regarding workforce and staffing. The expansion of adult ECMO services, MCS device implants, and transplant procurement services may necessitate additional perfusion clinical support resources. Respondents reported an increase in clinical workload over the last three years, and most centers indicated the use of part-time and per diem perfusion coverage for relief. Trends in workforce demographics are essential for perfusion supervisors, hospital administrators, and perfusion education programs. Approximately 38% of the certified clinical perfusionists that responded to the 2015-2016 American Board of Cardiovascular Perfusion (ABCP) Perfusion Profile Survey anticipated working 10 more years before retiring [36]. A 2019 survey of perfusion vacancy and turnover estimated rates of 12.3% and 14.7%, respectively [37]. Both rates exceeded those reported in nursing. Considering the timing of these findings and the unknown long-term consequences of the COVID-19 pandemic, workforce survey data is paramount in supporting practitioners and optimizing the quality of care.
There are several limitations to our survey. There are recognized sources of bias in conducting survey research such as  sampling bias, compound questioning, recall bias, and respondent misinterpretation of questions. Survey participation was limited to programs located in the Zone IV region of AmSECT, with a response rate of 34.7%. A majority of center responses were received from Northeastern programs. Data collected from these centers may not be generalizable to centers outside of this region. Survey recruitment across additional zones would appreciably increase our understanding of adult perfusion practice. Additionally, the timing of the survey may have been influenced by the Covid-19 pandemic, as centers may have experienced changes in both cardiac surgical and ECMO procedural volume. Recruitment was also restricted to a database containing names and emails of chief perfusionists and managers. The lapse in time between database data collection and survey recruitment may have contributed to lower survey response rates. Many of the non-respondents were attributed to invalid contact information, as several perfusion managers may have left their place of employment prior to the recruitment period. Lastly, this survey did not collect data on all aspects of clinical perfusion service. Examples include ultrafiltration, departmental quality improvement initiatives, or perfusion position vacancy rates.

Conclusion
Clinical practice surveys can be effective tools in identifying current perfusion staffing, techniques, and equipment utilization. Professional organizations may offer opportunities to promote the recruitment of member center participation. Additional research is warranted to help explain the gaps that may exist between clinical practice guidelines and current perfusion practice. This survey identified several elements of perfusion practice that did not reach thresholds consistent with high-level clinical practice guidelines and professional standards. Longitudinal surveys may describe clinical trends over time and measure adherence to evidence-based and professional standards and guidelines. Benchmarking and trending analysis is necessary to identify areas of improvement, help predict changes in clinical resource management, and foster improved patient outcomes.