Skip to main content

Develop Measures for Primary Graft Dysfunction in Hearts

Skip secondary navigation

eye iconAt a glance

Current practice

Primary Graft Dysfunction (PGD) is the leading cause of death among heart transplant recipients within the first 30 days of transplant. The OPTN does not collect post-transplant information on recipients that could help identify those who are at risk for developing PGD.

Supporting media


View presentation slides

Requested feedback

  • Heart Transplantation Committee suggests adding new data elements to the Transplant Recipient Registration (TRR) form to collect information that could inform PGD risk.
  • Committee is asking for community feedback about:
    • Data elements
      • Is the Committee suggesting the right elements?
      • Are there other elements that should be included?
      • Should information be collected on donors?
      • Should the focus be on both moderate and severe PGD?
    • Timing
      • How many hours after transplant should data collection occur?
    • Logistics
      • What challenges would programs have in collecting this data?
      • Is the TRR the best way to collect this data?
      • How can the Committee ensure consistency in this data collection across programs?
      • Are there differences between adult and pediatric PGD that should be considered?

Anticipated impact

  • What it's expected to do
    • Inform a future Heart Transplantation Committee data collection proposal
  • What it won't do
    • Will not change policy at this time


  • Data elements
  • Timing of data collection
  • Transplant program considerations

Terms to know

  • Primary Graft Dysfunction (PGD): life-threatening complication when a heart transplant recipient develops left, right, or biventricular dysfunction within the first 24 hours of transplant and a secondary cause is unknown.
  • Transplant Recipient Registration (TRR): The form completed and submitted by the transplant center when a patient is transplanted. The form contains patient status, pre and post-transplant clinical measures, transplant procedure, graft status, treatment, and immunosuppression therapies.
  • Click here to search the OPTN glossary

eye iconComments

International Society for Heart and Lung Transplantation | 03/23/2021

• Data elements o Is the Committee suggesting the right elements? Additional data elements suggested include: 1) Granular hemodynamics: RAP, PAS/PAD/mPAP, PCWP, CO/CI, and derived RH unit function RAP/PCWP ratio, PAPi, CPO and RVSWI 24 hrs prior to transplant and at 8, 12, 24, 48 hrs. 2) Granular derived imaging: RV/LV dysfunction, pericardial effusion, LVOT VTI. 3) Return to the OR for mediastinal bleeding. 4) Granular inotrope/vasoactive measures via inotrope/vasoactive index measures 24 prior to transplant, and at 8, 12, 24, 48 hrs. 5) Granular blood product measures 24 prior to transplant, and at 8, 12, 24, 48 hrs. 6) Objective measurements of hypoperfusion: Lactate washout. 7) Granular adjuvant therapeutics including the use and dose of renal replacement therapy, methylene blue, type and dose of inhaled pulmonary vasodilators (i.e. inhaled nitric oxide, inhaled epoprosterenol) 8) Granular AMCS derived details: Counterpulsation (40, 50cc), Transvalvular pumps (Impella 5.5, CP, RP, 2.5, Protek Duo), LA-Ao bypass (Left Tandemheart) ECMO (VA, VV, VAV), Surgical axial flow (Centrimag LVAD, BiVAD), combination (i.e. EC-PELLA, VA-ECMO with IABP) and placement (axillary, femoral, central) prior and after transplantation. Are there other elements that should be included? • Should information be collected on donors? o Yes and suggest coinciding with other active registries for • Should the focus be on both moderate and severe PGD? o Yes we agree that the focus should be on moderate and severe PGD. • Timing • How many hours after transplant should data collection occur? o Baseline prior 24 hrs and post-transplant 8, 12, 24, 48 hrs. • Logistics • What challenges would programs have in collecting this data? o Is the TRR the best way to collect this data? o Requires constant review however performing a pilot test first among transplant centers will be of extreme value to identify gaps and opportunities prior to rollout. • There is an opportunity to enhance data collection and adjudication with innovation in critically ill post-transplant patient care, and this might be an opportunity to standardize such with a dedicated committee. o Suggest consultation with other active registries dedicated to invasive hemodynamic and AMCS profiling (i.e. CSWG, ELSO). o Several programs share similar EMR (Epic) and this is an opportunity to share methodologies for data extraction and adjudication. • How can the Committee ensure consistency in this data collection across programs? o Suggest having a specific committee that helps programs adjudicate data for example large systems used to upload data for granular registries in cardiogenic shock. • Are there differences between adult and pediatric PGD that should be considered? o Data should be uniformly collected for both and then using a short-term analysis to identify gaps and opportunities between this two populations.

Heart Failure Society of America | 03/23/2021

In review of the OPTN proposal, HFSA supports the following: 1. Capturing donor and recipient data in the early perioperative window following transplant; 2. Tracking DCD donors and key DCD donor elements that may impact PGD; and 3. The differential data capture of PGD in pediatrics vs. adults, as we feel there are likely some different contributors to the PGD pathophysiology. More detailed recommendations and considerations can be found in our submitted comment letter.


Region 10 | 03/23/2021

Members of the region are in support of the project’s goal to develop measures for Primary Graft Dysfunction. One member noted that it seems like a good measure to develop if it has value and will help define criteria for re-listing, quality measures, or program activity outcomes. Another member suggested that heart/lung or other multi-organ combinations should also be included in the data collection. It was also suggested that venting method, cardioplegia use in the recipient’s transplant surgery, and repercussion time be considered as possible data elements. Another member expressed support the proposed data collection elements, but also feels that further procurement and donor data should be collected. Possible procurement data to collect includes: solution, warm ischemia time, and time from initiation of procurement of all organs until beginning of warm ischemia time. Heart that struggle through long procurement and acidosis from multi-visceral organ procurement could be affected. Lastly, possible donor data that could be collected: detailed cause of death information to include presence of smoke inhalation and unknown downtime or CPR.

Region 2 | 03/23/2021

One member commented that in terms of combining the vasoactive agents into a scoring system, it’s still important to distinguish vasopressors from pure inotropes. The reason for this distinction is because right heart failure can be confounded with someone that has vaso-dilatory collapse. Leaving off that equivalence would not be wise. In terms of timing, the member suggested documentation in the first 24 hours and then the first 72 hours up to one week. Another member commented that the committee consider collecting both moderate and severe PGD. To reduce confounding, especially in the early post-transplant time period consider a later time point such as 1 week and/or 2 weeks. Hemodynamics will be challenging to acquire routinely and reliably across centers. There was some concern over the amount of data being collected and it was suggested that organizing pertinent data points will be important. Another member commented that the amount of data being requested is large and fluctuates over time so capturing the data to define primary graft dysfunction is not an appropriate use of time and resources.

Region 9 | 03/23/2021

During the meeting there were comments voiced during the discussion, submitted online and submitted as comments from the OPTN representatives. Overall the region supports the Heart Committee working on this project. One attendee commented that for pediatric patients, some of these data elements won’t be able to be captured in the first 24 hours and agreed that perhaps requesting these values at 24 and 48 hours would be a good idea. There were several comments that only collecting data at 24 hours post transplant is too short of a time. One attendee agreed that in the first 24 hours, it can be hard to figure out what’s going on, so they suggested considering 72 hours. A comment was made about concern with using LV and RV function as qualitative values, but the commenter liked the idea of using the ejection fraction, although it’s sometimes hard to capture. There was some discussion that collecting inotropic support by the highest dose in 0-24 hours and then 24-48 hours would be a good idea.

International Consortium on Primary Graft Dysfunction | 03/23/2021

We write today to comment on the Heart Transplantation Committee’s proposal to add additional data elements to the TRR, to capture post-transplant data that could identify recipients who develop primary graft dysfunction – a significant contributor to early morbidity and mortality following heart transplantation. This comment was authored by members of the International Consortium on Primary Graft Dysfunction, a multicenter collaboration which has retrospectively collected data on over 2500 heart transplant recipients from 8 centers in order to identify ISHLT defined PGD. Given our experience in collecting this data, we feel we are well suited to answer this request for public commentary. With regards to the proposed data, we suggest the following fields which should be captured within the first 24 hours following heart transplantation: · Changing PGD to be a calculated field, not one that is ‘user entered’. This will prevent different centers from using different algorithms to define PGD, which would hinder analysis and interpretation of the data. A simple yes/no binary response will likely not be sufficient. The possible field values should be Moderate BIV PGD, Severe BIV PGD, or isolated RV PGD, which will be calculated based upon entry of the other specified data fields. · Quantify LV function based upon the presence or absence of LVEF <= 40%. · Right ventricular dysfunction should be quantified as mild, moderate, or severe. · Was there vasoplegia requiring treatment? · The highest doses of inotropes/vasopressors recorded in the first 24 hours post-transplant, including dopamine, dobutamine, milrinone, epinephrine, and norepinephrine. · The worst hemodynamics from the first 24 hours: highest right atrial pressure, highest pulmonary capillary wedge pressure, and lowest cardiac index. · We would recommend four individual data elements instead of a binary variable for mechanical circulatory support device. In so doing there is no need to identify right, left or biventricular support. o Post-Transplant IABP (Y/N) – specify that this is a NEW IABP placed post-transplant o Post-Transplant VA-ECMO (Y/N) o Post-Transplant RVAD (Y/N) o Post-Transplant LVAD (Y/N) · Duration of mechanical circulatory support Finally, we feel that the following pre-transplant data elements to be added to the TCR would be important for risk stratification based upon published risk factors for PGD: · Inclusion of right atrial pressure in pre-transplant hemodynamics · Inclusion of pre-transplant amiodarone use In terms of new procurement factors, we feel it will be important to include/update the following data elements related to preservation and procurement: · Include binary variable to determine whether the organ was procured using normothermic ex-vivo perfusion · Include total ischemic time as well as warm ischemic time and cold ischemic time if normothermic ex-vivo perfusion was used. Lauren K. Truby, MD Fellow, Cardiovascular Disease Duke University Medical Center Maryjane Farr, MD, MS Associate Professor of Medicine at CUIMC Medical Director, Heart Transplant Program Region 9 Associate Councilor, Member MPSC Columbia University Irving Medical Center Yasbanoo Moayedi, MD, FRCPC Department of Medicine, University of Toronto Kiran K. Khush, MD, MAS Professor of Medicine Director, Heart Transplant Research Director, Advanced Heart Failure Transplant Cardiology Fellowship Program Division of Cardiovascular Medicine Stanford University School of Medicine

American Society of Transplant Surgeons | 03/23/2021

The American Society of Transplant Surgeons (ASTS) is neutral on this policy proposal and submits the following comments. While the ASTS agrees overall with the intent of collecting data regarding Primary Graft Dysfunction (PGD), we have concerns with the data elements as the committee proposes. For example, Primary Graft Dysfunction/LV Dysfunction/RV Dysfunction are all Yes/No questions. What is absolutely needed is a crisp and robust definition of when a program makes those responses. The definition must be clear with low inference for data entry and abstraction. In Appendix A, the various definitions for PGD LV and PGD RV are stated. Presently, this definition is not specific enough and is open to interpretation by practitioners and transplant programs. We have concerns about data collection and they include the following. As to capturing data on support device(s) and what device/methods of use and duration of use, we agree but see the need for well-delineated definitions and specifics. In addition, we have several questions about potential quantitative elements that are proposed. The proposal states that there are multiple time points at which the data will need to be captured. Are all considered required? One should recognize that this may prove to be a significant data burden on transplant centers. As examples, the committee requests an LVEF to be documented. Do they want to specify modality (e.g. only with TTE/TEE)? Is post-op cardiac cath acceptable? Some official echocardiogram reports will give a range of EF (e.g. 30-40%). If so, should the data abstractor pick the lower number, the higher number, or the middle of the range? Alternatively, instead of an integer #, the committee could ask for categories (e.g. >60%, 50-60%, 40-50%, 30-40%, <30%). The committee should consider how the data field would be constructed such that data abstraction and data entry will be simplified and made as accurate as possible. Regarding RA and PA pressure measurements, are only direct measurements to be included via a PA catheter/right heart catheterization? Will an estimated PA pressure by echocardiogram be considered an acceptable method and suffice? If so, will the methodology used be a recorded field? As far as cardiac output is concerned, will the committee specify methodology? Is echocardiographic estimation adequate? What about thermodilution and Fick? Will the committee allow different methods, and will the method be recorded? With regard to Inotropes, this in particular seems like a high intensity variable that may not produce meaningful data. The committee should consider requiring the minimal data needed to document adequate use of pressors to support the diagnosis of PGD, but not more granular data beyond that. In closing, ASTS offers the following feedback to the OPTN Heart Transplantation Committee’s questions. 1) We believe the Committee has suggested reasonable elements and does not recommend adding others at this time. 2) There should be a focus on both moderate and severe PGD. 3) We agree information should be collected on donors to include: EF and (if available) right atrial pressure, pulmonary artery pressures, and cardiac output which may not have to be a required data field. 4) How many hours after transplant should data collection occur? One suggestion is at 0, 12, 24, 48, and 72 hours and including discrete data elements as on page 6. We may suggest that the committee be consistent with the ISHLT definition and remain consistent with, or at least be cognizant of other societal guidelines (e.g. lung PGG which is now defined using 72-hour data). 5) As to the challenges programs face in collecting data, the committee should consider how transplant programs would manage the extra data abstraction and collection requirements, and be as parsimonious as possible to achieve the stated goals. 6) We agree the TRR is likely the best way to collect the data. 7) ASTS recommends using standard definitions with discrete data elements to ensure consistency in data collection across programs.

American Society of Transplantation | 03/22/2021

The American Society of Transplantation is supportive of this proposal in concept, and offers the following feedback for consideration: Given the mortality associated with PGD, collection of both donor, recipient and intra-operative data is imperative to permit a better understanding of this condition. With regard to PGD related data elements for assessing transplant related mortality, the suggested potential data elements for addition to the TRR form are reasonable. However, a more expansive set of data elements will be needed. The timing of collection of hemodynamic data needs to be carefully considered. In patients with defined PGD, the worst hemodynamic parameters and highest doses of inotropes at 24 hours (defined as 1st 24 hours out of OR) would ideally be collected as PGD consensus definition has a 24-hour timeframe. Repeat parameters at 72 hours (+/- 4 hours) should also be collected. Additionally, consideration should be given to capture hemodynamic data measured at the time of a first biopsy (if performed in the first 7-10 days) to determine the persistence or resolution of PGD. Finally, PGD grade should be included (i.e. whether moderate or severe). There is significant literature to suggest that Pre-tx recipient factors contribute to the risk of PGD. These data should be collected if not already available, including mechanical circulatory support (temporary or durable and type, duration of temporary support), inotropes including doses, medications at transplant including amiodarone, ACEI, ARB, ARNI. Given developments with ex vivo perfusion and evolving technologies (e.g., new preservation solutions and techniques), it is important to collect donor data. These should include organ preservation technique, ex vivo Y/N, preservation solution, warm ischemic time. Elements from the UNOS donor management registry could possibly be incorporated for this purpose. However, collecting more detailed data to include amount of perfusion solution and bag pressure may be cumbersome and onerous. For DCD (direct procurement method), time to asystole after withdrawal of life support, warm ischemic time, back table time and total ischemic time, time from admission to declaration of death should be collected. Where ex vivo platforms are used, both warm and cold ischemic times should be collected, time on device and maximum lactate values. Transplant perioperative factors such as transfusions (number and type of units) may be important. We concur with the removal of airways dehiscence as it is not applicable to heart transplantation. Primary Graft Non-Function should be replaced by Primary Graft Dysfunction as it now has a definition by consensus. Acute rejection should stipulate type and severity by ISHLT Classification, whether mixed or biopsy negative. Chronic rejection should also be removed and replaced with cardiac allograft vasculopathy but may not be relevant if data is collected early after transplant (see below). The collection of additional data may provide challenges for programs, but this could possibly be counterbalanced by removal of less relevant elements and automation of collection of certain data elements from EMR systems. Transplant programs should have easier access to data elements due to EMR systems in place. The TRR is a reasonable tool to use in the absence of a relational database. The data collection should be part of collection of EARLY post-transplant data. Issues relating to ABO incompatible transplants, use of different temporary and durable mechanical support devices and congenital heart disease may be unique to collecting PGD data for pediatric transplants. Simplification of data elements (e.g., use of PGD grading scale) and avoidance of free-text data elements would ensure consistency in data collection. OPOs should be able to provide important information about DCD donors, including etiology, admission time, donor support, hemodynamics. From the pediatric perspective: The data elements seem reasonable. We do not recommend including mild but should include moderate to severe PGD with a clear definition of each. We do feel that some donor elements should be collected. This work has previously been done by ISHLT and published as a consensus document (Report from a consensus conference on primary graft dysfunction after cardiac transplantation). We recommend using the data recommended in this document as a lot of time, effort and thought was put into its development. Donor risk factors used in this document included age, cause of death, trauma, cardiac dysfunction, inotropic support, comorbidities (DM, HTN), downtime of cardiac arrest, drug use (alcohol, cocaine, amphetamines), Left ventricular hypertrophy, valvular disease, hormone treatment, CAD/wall motion abnormalities on TTE, sepsis, alternate list/marginal donor allocation-not increased risk, troponin trend and hypernatremia. We suggest that it is important to avoid the temptation to include or address items that DO NOT factor into PGD (e.g. rejection). From Table 1: PGD, LV dysfunction, RV dysfunction, we recommend that the definitions be VERY clear and include when the events occur. From a consistency perspective, we recommend that “Y/N” not be used for PGD as that is too vague without a clear definition. We would NOT collect hemodynamics as this likely will not be available for many patients and is very labor intensive. Likewise, we would not include drug doses as this will be very labor intensive as well particularly in pediatrics where dosing is not a single standard dose. We also believe that the challenges of collecting WIT and organ preservation techniques will likely outweigh the benefit. The only exception would be to indicate DCD and whether ex vivo perfusion is used Timing: There are differences between adults and pediatrics, so the committee should ensure pediatric input. We strongly suggest sticking to the 24-hour definition. 24 hours is an accepted definition and meets with the current practices and studies internationally. If changed to the 72 hours, then the U.S. would differ from the consensus guidelines and the studies published since 2013, and blur the field from a PGD perspective. We recommend reporting the highest value in that 24-hour period (or the maximal support).

Paragonix Technologies, Inc. | 03/22/2021

As the Co-Founder, President, and CEO of Paragonix Technologies, Inc. I respectfully submit this comment on behalf of the company in response to the proposed policy, “Develop Measures for Primary Graft Dysfunction in Hearts.” Paragonix Technologies, Inc. is a medical device company headquartered in Cambridge, MA, that designs, produces, and markets organ transportation devices that safeguard human organs intended for transplant during the journey between the donor procurement facility and the transplant recipient center. Paragonix supports the addition of data collection elements to the Transplant Recipient Registration (TRR) form as proposed by the committee to better understand what can be done to minimize primary graft dysfunction (PGD), the main reason for 30-day mortality in heart transplant recipients. We offer the following in response to the committee’s specific questions. Paragonix supports inclusion of data for patients with “moderate” PGD as the classification is defined by the International Society for Heart and Lung Transplantation (ISHLT) and recommends including data on both pediatric and adult patients. Regarding additional predictive and operational data points, Paragonix supports the inclusion of data that identifies the method and quality of preservation of the donor heart prior to the transplant. For example, as has been recognized by the 2020 Consensus Statement of the ISHLT, certain preservation data points, including the transport temperature, have been linked to donor heart injury as well as PGD incidence (discussed in the 2014 ISHLT Consensus Statement on PGD). The wording of the data to be collected should be broadly stated to capture heart preservation data to fully research and evaluate possible correlations between quality of preservation, certain preservation parameters and the incidence of PGD. As a leading manufacturer of heart preservation devices, Paragonix applauds the OPTN/UNOS Heart Committee for bringing awareness to what is needed to better understand how to reduce the incidence of PGD and we look forward to reviewing our heart transplant registry data with the findings of the OPTN/UNOS in this effort.

NATCO | 03/22/2021

NATCO thanks the Heart Committee for their work on this proposal. Overall, NATCO supports the proposal. One unidentified consequence may involve challenges that transplant programs may face with data reporting and collection. Data collected may impact patient education in regard to informing patients of post-operative recovery risks that may include increased LOS, return to OR for bleeding/MCS or increased risk of infections, need for rehab, etc. The proposed data collection in Table 1 would be a good start as this coincides with the ISHLT Consensus statement on PGD and the Severity Scale for PGD. We recommend not including airway dehiscence in heart collection data. It is important to focus on moderate PGD in addition to severe PGD. It would be helpful to collect this data when the patient arrives to the ICU for 24 hours as it is defined as “ventricular dysfunction occurring within 24 hours post-transplant.” We recommend adding clear definitions for what is to be documented and when data is to be collected to this proposal. The amount of data to collect outlined in Table 1 seems sufficient. For VAD related data, collecting information on date of implantation, duration of MCS prior to removal should be collected. It may also be helpful to know if they were using MCS prior to transplant to determine if any correlation exists with PGD post-transplant. Collecting this additional data would increase the amount of time to gather the information, possibly leading to the need to hire an additional staff member to perform this duty. Centers do have the necessary information to collect all proposed additional data. The TRR form is the correct tool to use for data collection.

Evan Kransdorf | 03/22/2021

I would like to offer suggestions on this proposal based on my review of the literature and research performed at our transplant center. First, based on research performed in the literature, I would suggest that recipient and donor predicted heart mass (PHM) be collected at the time of transplant. Recent work by Gong et al. (Gong et al., Journal of Heart and Lung Transplant 2018) has shown that PHM mismatch is a better predictor of PGD than weight. Second, we have performed a comprehensive review of factors associated with the development of mild, moderate and severe PGD at our institution's heart transplant program. Similar to others, we have found that mild or moderate PGD were not associated with an increase in mortality at one year post-transplant, whereas severe PGD was associated with a significantly increased risk of mortality at one year post-transplant. Factors strongly associated with severe PGD included: treatment with amiodarone plus beta-blocker, treatment with ACE or ARB or ARNI plus MRA, previous cardiac surgery, as well as transfusion of PRBC and platelets. Thus, based on our data, collection of both the severity of PGD (graded as no PGD, mild/moderate PGD and severe PGD) and the above variables would be helpful in continued efforts at better identification of PGD. Our study is under review for publication at the Journal of Heart and Lung Transplantation.

Transplant Coordinators Committee | 03/19/2021

The Transplant Coordinators Committee thanks the Heart Transplantation Committee for presenting its request for feedback “Develop Measures for Primary Graft Dysfunction in Hearts” and offers the following feedback: A member shared feedback that the device seemed to be the most accurate measure of primary graft failure as opposed to inotropes due to their varied use by facilities or the existing definition as 48 hours was too short but 72 was too long to determine dysfunction. A member also suggested adding cross match information for highly sensitized patients. A member emphasized the importance of using the device as what appears to be PGD shortly after the transplant can resolve itself, but this waiting period can lead to inaccuracies between the patient’s chart and DonorNet. A member inquired about the specification between ‘procurement team’ and ‘local team,’ asking that those two terms be clearly defined.

Region 11 | 03/18/2021

During the meeting there were comments voiced during the discussion and submitted online. Overall, there was general agreement and support for this proposal. Two attendees commented that one advantage of looking at severe PGD is that the data is much easier to track and understand. One attendee added that severe PGD needs to be clearly defined based on literature. Another attendee commented that hemodynamic data collection might be complicated and onerous because of doses at different times. Expanding use of DCD hearts was supported by several attendees, with one attendee recommended a safety net when transplanting DCD or marginal donor organs, allowing for the ability for clinician judgement to transplant if necessary and increase utilization. They went on to add that if using these organs is outside the guidelines or is punitive, it creates reluctance in using organs. Another attendee commented that PGF measures need to be developed. One attendee commented that perhaps the data collection should include biopsy results or positive/negative cross match results.

Maya Barghash | 03/17/2021

I think capturing data points up to 48 hours would be sufficient to capture the hemodynamic variables of interest for PGD. For EF quantification, given that many of the quantifications will be done with TEE which is not the most accurate assessment of EF, perhaps we can include a qualitative in addition to a quantitative drop down menu for selection (ie Mild dysfunction (LVEF 40-49%), Moderate dysfunction (LVEF 30-39%), Severe dysfunction (LVEF less than 30%)) and I think it should be specified at what time point this EF was taken – may be interesting to see EF POD 0, POD1, POD 2. For hemodynamic variables, given the continuous monitoring with Swan Ganz catheters, it may be helpful to specify whether to include the worst values in the 24 hour period for CVP, PAS, PAD, PAM, PCWP, CI, CO, and may also be helpful to include PA saturation since there may be variation in which CI/CO is being reported. Other hemodynamic variables could be underlying rhythm as well as lowest mean arterial pressure 0-24 hours, and 24-48 hours. I am not sure if these variables are already captured, but I think peak lactate, Cr, AST, Bilirubin, pH would of interest as well as whether renal replacement therapy was needed, duration of mechanical ventilation, ICU length of stay As for the inotrope data collection, would also include inhaled pulmonary vasodilator use. I think that the inotrope type, and max dose in 0-24 hours as well as 24-48 hours should be included. It would be interesting to collect data on vasopressor use (number, type, and dose) to understand the interaction between PGD and vasoplegia. For support device, in addition to collecting whether it was used and which device was used, I think it would be helpful to identify when it was instituted (time from OR) and the duration of use because I wonder if the time to institution of a support device plays into outcomes in addition to the PGD severity I assume pre-operative and perioperative risk factors are already being collected but if not, would be important to record such as donor risk factors, recipient risk factors, and surgical procedural risk factors noted in the ISHLT PGD consensus document table 2.


Region 7 | 03/12/2021

During the meeting there were comments voiced during the discussion, submitted online and submitted as comments from the OPTN representatives. An attendee suggested the ISHLT consensus statement should be used, as it is more user-friendly, and uses data that can translate across all transplant programs. One attendee also recommended that a clean definition of PGD come from the Heart Committee. Overall, there was general discussion around using the PGD trends and understanding the amount of severe verses moderate outcomes.

Region 6 | 03/09/2021

Overall the comments were in support of the Heart Committee pursuing this project. A member commented that the Committee should align this data collection with the International Society for Heart and Lung Transplantation (ISHLT) guidelines as much as possible. A couple members commented on the timeframe for data collection and thought that the suggested 24 hours was too short. One member said that 48 hours or even longer should be considered, and another suggested considering 24 hours and then a second timeframe. A comment was made that an important data element might be whether the patient had been relisted for transplant. An attendee wondered if this information would eventually be used to inform patient care or if primary graft dysfunction rates might be reported as patient-facing data. One member asked if it might be possible to add an “other” field where a program could enter something like the presence of non-HLA antibodies that might be associated with PGD. Another attendee commented that a standardized approach to data collection is important, and this project is a great start.

Region 8 | 03/09/2021

During the meeting there were comments voiced during the discussion, submitted online and submitted as comments from the OPTN representatives. Several members discussed the opportunity and challenges of interfaces with information systems to collect data, one member stated that electronic does not necessarily make gathering data easier. A member questioned whether it was necessary to collect this data from all transplant hospitals versus a subset. Another member voiced support and recommended collecting the data for partial dysfunction to help evolve the concept, diagnosis, and treatment. One member suggested adding preservation device information. A member suggested adding ECHO, BNP or proNT BNP results for pediatrics who are not always taken for a heart catheter to get hemodynamics if they are unstable.

OPTN Data Advisory Committee | 03/09/2021

The Data Advisory Committee thanks the Heart Transplantation Committee for its effort to further identify PGD through their request for feedback document. The Committee supports data collection on all heart recipients, rather than a self-identified subset. The Committee supports collecting data on the use of support devices post-transplant but is concerned about the challenges associated with collecting inotrope data.

OPTN Transplant Administrators Committee | 03/05/2021

The Transplant Administrators Committee appreciates the opportunity to comment on the Develop Measures for Primary Graft Dysfunction in Hearts proposal. Overall, the members support this proposal. One member stated their heart team shared that that recipient data elements are already collected and would not be difficult to report. The team raised concern with the perfusion and donation associated data elements because these elements are harder to collect and asked if organ procurement organizations (OPOs) would be required to provide the data. A member commented that with an increase in broader sharing, relationships with additional OPOs will need to be established in order to collect this data.

Region 5 | 02/19/2021

During the meeting there were comments voiced during the discussion, submitted online and submitted as comments from the OPTN representatives. Overall, there was general agreement and support for this proposal. One member agreed that collecting data from organ recovery is critical. In addition to the data elements mentioned, we should consider including time to standstill. Another member stated they are seeing this in addition to pre-donation warm ischemic time stating that it does make a difference and depends on the operator. The member did not know if this is something that can be measured stating that it is important to know what kind of pressure you have.

Region 3 | 02/18/2021

During the meeting there were comments voiced during the discussion, submitted online and submitted as comments from the OPTN representatives. One member commented that data collection on DCD hearts could expand the donor pool. Another member commented that this is great and necessary work and believe the committee should look at moderate and severe PGD and stated that it is different in pediatrics and adults. They recommended the committee look at different ways to preserve it if supported by technology as well as looking at DCD vs traditional donors. They also suggest looking at data related to the donor including hemodynamic support, mode of death, CPR and duration, and time to retrieval. Several members commented that the data burden and cost to transplant centers needs to be considered. One of the members stated that it needs to be considered whether this data is for fair policy development or to answer a potential research question.

Region 4 | 02/04/2021

During the meeting there were comments voiced during the discussion, submitted online and submitted as comments from the OPTN representatives. Overall, there was general agreement and support for this proposal. One member commented that a review at their center found worse mortality and prolonged hospitalizations for those identified with PGD. This member recommended using inotropic score to define PGD- this is validated in the literature, as it may be more useful and easier to pull from EHR. She added that hemodynamic data collection was not used as much in their study because it was too onerous and inconsistently collected. The attendee also commented that it would be beneficial to capture first biopsy or some other marker for hyper acute rejection in order to learn from this data. Another member commented that data collection will be challenging for transplant programs, but will help differentiate PGD Left ventricle (PGD-LV) and PGD-Right Ventricle (PGD-RV) RVPGD. He added that cross match data needs to be factored in since programs make decisions early in the transplant process about how to address highly sensitized candidates’ needs. The attendee also commented that such pre-transplant decisions may eventually play into who has PGD and should probably be considered as well as post-transplant metrics. Another attendee commented that defining PGD at 24 will be more critical when looking at moderate to severe cases as clinical teams are comfortable dealing with mild PGD. The attendee added that use of inotropic score is problematic as use of inotropes is variable, even within centers. The attendee also commented that reporting the use of devices will be a more definite method in defining PGD. The member also supported collecting data on predictive heart mass between recipient and donor for matching. Several members were concerned with data burden, with one reminding meeting attendees that the information the community would like to collect needs to be balanced against the requirements/challenges programs would face with performing the collection and reporting. One member suggested pulling data directly from the EMR to reduce data burden. One attendee commented that they were concerned about how the diagnosis is made and potentially used by centers to get organ offers. Finally, an attendee commented that the committee will need to ensure we have clear definitions that are documented.