Skip to main content

Report Primary Graft Dysfunction in Heart Transplant Recipients

eye iconAt a glance

Current policy

Primary Graft Dysfunction (PGD) is the leading cause of patient death in the first 30 days after a heart transplant. Currently, the Organ Procurement and Transplantation Network (OPTN) does not collect some of the information that could help identify patients who are more likely to develop PGD.

Supporting media


View presentation

Proposed changes

  • Collect post transplant graft data
  • Monitor outcomes for patients who receive a heart transplant
  • Allow program-level comparisons of PGD incidences

Anticipated impact

  • What it's expected to do
    • Collect post-transplant data to help identify PGD
    • Monitor how often PGD occurs in patients who receive a heart transplant
    • Help the Committee monitor outcomes for patients with PGD
    • Provide data to guide future policy development
  • What it won't do
    • The proposal does not change heart policy or allocation


  • Data elements
  • Timing of data collection
  • Inotrope and vasotrope reporting

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

Data Advisory Committee | 10/01/2021

The Data Advisory Committee appreciates the opportunity to provide feedback on the Heart Committee’s proposal Report Primary Graft Dysfunction in Heart Transplant Recipients. The Committee overall supports the proposed data collection as it will provide the community with a better understanding of PGD but acknowledges that collecting and reporting this data will create a significant burden. The Committee recommends providing adequate education to ensure the staff responsible for entering the data has access to the PGD, left ventricular dysfunction, and right ventricular dysfunction diagnosis information as these data elements may not be readily discoverable in EMRs or medical charts.

Pediatric Heart Transplant Society | 09/30/2021

As one of the largest organizations representing the interests of pediatric heart transplantation, the Pediatric Heart Transplant Society, thanks you very much for your efforts to better understand and evaluate primary graft dysfunction. We would agree that this is an issue of importance and appreciate that pediatric patients are considered. With that said, there are a few concepts that we feel require further definition and clarification, especially as they relate to pediatric heart transplantation. Firstly, the definition of PGD has been a long-standing issue in the field and is likely one of the reasons why this topic has been difficult to study. The concept of “secondary to other causes” we feel is of utmost importance, particularly in pediatrics as recipient factors can often be significantly complicated by congenital heart disease and differing physiology of heart failure as compared to adults. We often encounter patients with elevated PVRi pre-transplant or elevated PVRi that is unmasked in the early post-transplant period that results in RV dysfunction and at the moment it is unclear from the current data collection if RV dysfunction secondary to elevated pulmonary vascular resistance is considered PGD. . This dysfunction often responds to the management of the pulmonary vascular bed and most of our community would consider this to be secondary RV dysfunction. In addition, the assessments of pulmonary pressures proposed are not routinely collected at most sites in pediatrics as we do not typically do not use Swan- Ganz catheters due to size constraints. We wondered if other measurements were considered for collection such as echo parameters of RV pressure/function. Additionally, we also feel that considerations should be given to how cardiac dysfunction is defined, how it is measured (as there can be differences between measurements ie: m-mode vs. Biplane), and how the RV is assessed. The lack of definitions will likely lead to much variability in the data and may make interpretations of the findings difficult and lack external validity. In addition, some of the data points may not be available at the 24 or 72-hour time points. Is there flexibility with entering these time-driven points? How will changes in function over time be accounted for (ie: RV dysfunction at 24 hours than LV at 72hrs)? Furthermore, for the proposed inotropic data collection, while it is mostly accommodating of pediatric practice, we do order vasopressin and phenylephrine as weight-based drugs (ie: per Kg), and this option for reporting is not available in this proposal. Lastly, while we strive to understand more about our patients it is important that data collection is designed to answer specific questions so that we can ensure the extra work required by the transplant teams does result in meaningful outcomes.

Operations & Safety Committee | 09/30/2021

The Operations & Safety Committee thanks the OPTN Heart Transplantation Committee for their efforts on this proposal. The Committee supports this proposal and acknowledges that the community is long over due in collecting primary graft dysfunction data. The Committee agreed that utilizing the data to help identify a definition of primary graft dysfunction would be beneficial for the community. The Committee agreed that the suggested timeframes for data collection are reasonable.

Region 11 | 09/30/2021

Region 11 sentiment: 3 strongly support, 8 support, 6 neutral/abstain, 0 oppose, 0 strongly oppose Comments: Region 11 supported the proposal with no comments.

NATCO | 09/29/2021

NATCO supports the proposed data elements and believe the requirements at 24 and 72 hours (plus or minus 4 hours) after patient arrival to ICU is reasonable because the care team would likely conduct assess those elements at those intervals anyway. In addition, we feel the ease of drop down choices would ease the burden the forms would place on large transplant centers. One comment is about the data collection on LVEF, we agree that LVEF should be left for collection on all transplants and we would propose noting on the 24 hours form, that LVEF from any method can be reported since transplant (for example: intra op or postop TEE or TTE); however, a repeat echocardiogram may not be completed at each center at the 72 hour data point based on center specific protocols and to make this a requirement adds undo cost and staffing burden for each center. For the Flolan data, this should be labeled specify to include "epoprostenol", not just brand Flolan.

Thomas Jefferson University Hospital | 09/29/2021

PATJ supports collection of data to monitor primary graft dysfunction. While we would support data collection, there are concerns and clarification needed in how the current proposal is written. An additional 30-60 minutes per form adds significant work for the clinical team. We would also like clarification over the timing of data entry. While pulling data from 24 and 72 hours post-transplant is reasonable it was not clear if the data would have to be entered at the 24 hour and 72 hour mark? If so, this could create undue burden as it could potentially be overnight/early morning hours.

American Society of Transplant Surgeons | 09/29/2021

The American Society of Transplant Surgeons (ASTS) is neutral on this policy proposal. In heart transplantation PGD, though a very important metric of periop and long-term outcomes, has been somewhat ill defined. The proposed data elements begin to address the development of a concrete definition. The hemodynamic assessment, the medication use, and dosing additions coupled with the outcome data already collected would facilitate the development of a more specific PGD definition. ASTS supports the concept to better define PGD but recommends the OPTN facilitate a pilot project to vet and establish the validity of the proposed data elements before finalizing the policy. ASTS provides the following feedback on questions posed by the OPTN Heart Transplantation Committee. 1. Data elements: Are there additional data elements that should be considered for inclusion? Exclusion? a. Do any of the proposed data elements create unreasonable burden to collect and report? The data elements are reasonable though the primary metric being assessed, PGD, LVD, and RVD are qualitative in nature and do not provide good definition for PGD. The hemodynamic data are good to include. The use of inotropes, nitric oxide and epoprostenol are good to include though the duration is of importance in gauging degree of heart dysfunction, not only their initiation. Flolan (Page 9) is a trade name and the use of epoprostenol would be more appropriate. b. Would any modification reduce the level of effort required? A concern for “ease” of collection and use of EMR is that automatically transferred and recorded data, lack some fidelity. The time and conditions on where the hemodynamics and the dose of medications collected could cause added data collection and need for validation. c. Could any modifications better align with patient data currently reported in a program’s electronic medical records (EMR)? The time of the data elements collection at T0, T24, T48, and T72 would be important. Timing of data collection: Does offering a window of +or- 4 hours at 24 and 72 hours from arrival at ICU reduce the need to modify existing workflows? Will this window significantly impact the ability to compare patient outcomes? Is arrival at ICU an appropriate starting point? Should additional time points be considered in addition to 24 and 72 hours (+or- 4 hours)? A time of 48 hours would be reasonable as is having a window of time. Overall, it would be reasonable to collect at T0, T24, T48, and T72 to minimize impact on workflow and allow for comparison. 2. Inotrope and Vasopressor Reporting: a. Are the proposed ranges of inotrope and vasopressor dosing applicable to pediatric patients? Yes, the dosing would be appropriate as they are listed at weight based doses. b. Are the proposed ranges appropriately stratified to indicate high, medium, and low dosages? The ranges would be appropriate stratification as well. c. Is collecting vasopressor dosing in mcg/kg/mins with the option of also reporting in mcg/mins reasonable or is there another preferred unit of measure that would allow easier reporting or alignment with what is reported in a program’s EMR? Whether a center reports as mcg/kg/mins or in mcg/mins is not necessarily relevant as long as the weight based dosing and stratification is what is reported out from OPTN and for broad comparisons of outcomes and management (not applicable for vasopressin though would be for phenylephrine and norepinephrine). 3. Other: What challenges would this request present for transplant programs responsible for collecting the additional data? Is the data requested readily accessible? Should the data collection be part of the “Clinical Information: POST TRANSPLANT” section of the TRR, or is there a more appropriate section? Are there differences and/or similarities between adult and pediatric PGD the Heart Committee should consider as part of its future reviews? How can the Committee ensure the data collection is reported consistently by all transplant programs? The challenge would be in identifying the data point to report at the time window. The data are/would be readily accessible to the transplant center though a manual extraction or a recording which would be the most likely real-world event. Including the data in “Clinical Information: Post Transplant” is reasonable and appropriate. Ensuring data entry ease via buttons or drop downs and minimizing free text would facilitate collection.


American Society of Transplantation | 09/29/2021

The American Society of Transplantation strongly supports the proposal to collect data on Primary Graft Dysfunction (PGD). While additional time points for data collection would be desirable, we believe that there needs to be a fair balance between the amount of data collected and the additional administrative burden it imposes on transplant centers. Therefore, the collection of data at two time points, 24h and 72h is reasonable. Long-term data already collected (morbidity, hospitalization and mortality), should be linked to the new data points. There was not unanimous agreement on this point, recognizing that hemodynamics are important in defining primary degrees of graft function and that this cannot be checked at another time interval. While the desire to streamline or minimize data collection to reduce the data burden on programs is admirable, it may come at the expense of greater granularity in understanding the outcomes of those patients who do not experience primary graft dysfunction. The proposed data elements provide a reasonable framework. However, data acquisition should be streamlined such that, for example, if PGD is not present, then most of the other data element boxes should be automatically filled in as negative (eg device used, right or left PGD, EF). Since the intent is to collect hemodynamics on all patients, these would remain open for completion. This would prevent misclassification in cases where a temporary support device (I.e. IABP) may be left in position after transplant in the absence of PGD and would minimize the clicks for the person entering the data. If PGD is present, the next stem should define PGD-LV or PGD-RV in line with the ISHLT Consensus definition which is currently the only working definition. While most data may be reasonably obtained in the proposed timeframe defined after arrival in the ICU, initial EF may only be available from intra-operative TEE and second assessment may occur within a broader timeframe than 72h +/- 4h. We therefore propose that EF values are not strictly restricted to the 24h and 72h timepoints. We believe that the starting timepoint for PGD should be in the OR as in rare cases severe PGD may lead to mortality prior to arrival in ICU or significant delay in arrival into ICU. The occurrence of PGD should be within 24h of transplant not within 24h of arrival in the ICU. All vasoactive drugs should be named primarily by their generic nomenclature. Units for vasopressin should be unit/min. The ranges appear appropriate for adult and pediatric patients. Collecting vasopressor dosing in mcg/kg/mins with the option of also reporting in mcg/mins is reasonable. The collection and entry of these data elements, while readily available in most EHR systems, will require additional resources, particularly for larger programs. Accordingly, we suggest clarification on the data collection to determine if there will be a potential increased burden of data entry. With regard to consistency between programs, there may be a need to better define time points for entering vasoactive drug dosages – should they be captured at a single time point or should the highest dose within the +/- 4h time window be entered? This is particularly important as dosing and number of vasoactive drugs may change rapidly in the first 72h after transplant. While this proposal does not suggest the PGD rates of centers will be made public, if PGD rates for centers are to be published, there should also be comparative analyses of factors which may contribute to it. Examples include, donor age, ischemic time, procurement distance, preservation method. This would allow a balanced assessment of programs willing to comply with broader sharing at risk of potentially increased PGD rates. We acknowledge that the use of the data is outside the scope of the proposal to collect the data but recognize that future utilization of the information should be carefully vetted. We also ask that pediatrics be taken into consideration in this reporting as the indications for pediatric heart transplant are not the same as in adults and outcomes will be different. This will require clear definitions?(including pediatrics)?and timeframes to ensure consistent, meaningful data is collected.

Region 10 | 09/28/2021

Region 10 sentiment: 1 Strongly Support; 13 Support; 5 Neutral/Abstain; 0 Oppose; 0 Strongly Oppose Comments: There was support for the 24 and 72 hour post-transplant timeframe, but suggested that the committee add a 7 day time point as some PDG cases take more time to improve. Another member noted that there needs to be clarity on when the documentation has to occur for the 24 and 72 hour timeframes. The data should be readily available in the patient’s medical record and could be captured in a retrospective review. Additionally, there is a need for clarification on data collection to determine if there will be a potential increased burden of data entry. The committee should also consider pediatrics in this reporting as the indications for pediatric heart transplant are not the same as in adults and outcomes will be different. This will require clear definitions (including pediatrics) and timeframes to ensure consistent, meaningful data collection.

Region 1 | 09/24/2021

Region 1 sentiment: 2 Strongly Support, 2 Support, 4 Neutral/Abstain, 0 Oppose, 0 Strongly Oppose. Comments: Overall, the region supports the proposal. There were no additional comments made.

Region 6 | 09/23/2021

Region 6 sentiment: 2 Strongly Support; 6 Support; 2 Neutral/Abstain; 0 Oppose; 0 Strongly Oppose. Comments: An attended recommended the Heart Committee ask the Pediatric Committee for additional input.

Region 8 | 09/22/2021

Region 8 sentiment: 4 strongly support, 15 support, 3 neutral/abstain, 0 oppose, 0 strongly oppose. Comments: Region 8 supports this proposal. A member supported the proposal and suggested to also look at donor data, ischemic time, and distance travelled. A member stated that the need for this information is understandable but the impact on programs to collect this data should not be underestimated. He stated that there will be a significant increase in the time required to complete this form.

Region 7 | 09/15/2021

Region 7 sentiment: 5 strongly support, 6 support, 5 neutral/abstain, 0 oppose, 0 strongly oppose. Comments: During the meeting, there was discussion regarding pediatric candidates, DCD related data, and the definition and utilization of surgical complications data. An attendee suggested replacing, "Flolan following transplant" with the generic term "inhaled prostacyclin following transplant.” One attendee suggested making recommendations for the dose of pressors/inotropes along with hemodynamics carefully to define primary graft dysfunction however would not clearly make it’s a requirement, in the end, it should be the treating physician/surgeon to define graft failure. The attendee also expressed concern that vasoplegic patients who may get ECMO despite normal heart function could get mixed into "primary graft dysfunction" definition. Another attendee suggested that the Committee provide specific examples of “surgical complications” within the definition for PGD to provide direction and more standardization for these exceptions.

Region 9 | 09/14/2021

Region 9 sentiment:? 1 Strongly Support; 5 Support; 5 Neutral/Abstain; 0 Oppose; 0 Strongly Oppose. Comments:?Overall Region 9 is supportive of this proposal. There were comments regarding the ranges of epinephrine, including that the three ranges were not granular enough and that the upper range was too high.

OPTN Transplant Coordinators Committee (TCC) | 09/14/2021

The Transplant Coordinators Committee (TCC) appreciates the work of the Heart Transplantation Committee in developing this proposal and for the opportunity to comment on it. Members voiced support for the proposed data elements and believed the requirements at 24 and 72 hours after patient arrival were reasonable because the care team would likely conduct those tests anyway. There is concern about the burden the forms would place on large transplant centers if taking 30 – 60 minutes to complete and suggested sending an automated email that reminds coordinators to check for those fields in real time throughout the day. This could cause undue burden when coupled with the increased workload from COVID-19, but with a yearlong implementation timeline this burden will likely decrease by the time of implementation. There was also a suggestion to provide additional details to explain why this data is not available for pediatric candidates. Overall, the TCC is supportive of the Heart Committee’s proposal and is appreciative of the opportunity to provide the transplant coordinator perspective.

Maryl Johnson | 09/12/2021

I am strongly supportive of increasing the collection of data elements to better define the incidence, predictors, and outcomes of PGD. However, I would suggest that information regarding whether or not the donor was DCD, and, if so, whether the method was ex vivo perfusion or normothemic regional perfusion, should be collected. It is true that the ex vivo study has been completed, however, as stated by Dr. Jason Smith during the Region 7 meeting, FDA approval for the use of the ex vivo device is in process and many centers are proceeding with DCD heart donors using normothermic regional perfusion (recent case series from the Vanderbilt group has been published on line in JHLT). As adding data elements to collect always takes some time it would seem appropriate to start collecting DCD information now so we don't miss the opportunity to have information available in this rapidly developing field.

Region 3 | 09/10/2021

Region 3 sentiment: 3 strongly support, 9 support, 2 neutral/abstain, 0 oppose, 0 strongly oppose. Region 3 supported this proposal. One attendee believes the dosing scheme is appropriate; and the proposal will not overburden coordinators. Another attendee expressed concern the dosing for vasopressin does not include per Kg, which may be problematic for pediatrics. A third attendee commented dosing per IU vs mg of vasopressin needs to be considered.

Region 2 | 09/10/2021

• Region 2 sentiment: 10 Strongly Support, 10 Support, 6 Neutral/Abstain, 1 Oppose, 0 Strongly Oppose • Comments: Overall, the proposal was supported by members in the region; however, members in the region raised concern over the added data burden being proposed. The proposal states that each form will require an extra 30-60 minutes worth of data collection and entry, which is a significant amount of work for transplant coordinators. Another member noted that it would be very helpful for transplant program staff if there was a way to track Primary Graft Dysfunction using fewer data points.

Region 5 | 08/30/2021

Region 5 sentiment: 8 strongly support, 12 support, 4 neutral/abstain, 2 oppose, 0 strongly oppose. Region 5 supports the proposal for Reporting Primary Graft Dysfunction but points out the need for clarification on data collection and potential burden of the data collection. There were two oppositions for this proposal. A member suggested that the committee figure out what are the acceptable modalities and what values you are looking for. Regarding cardiac output, the member suggested to look at the desired methods. A member asked whether the 24-72 hours data was needed Several members cautioned on the committee on the large amount of data collection the proposal requires. Another member thought the data elements and type were reasonable. Another member stated they support this concept but oppose it in its current state due to the data burden and lack of specificity regarding the data points. Another member wants to see more clarity on the data collection end points and asked the committee to streamline the data collection measures to lessen the burden on transplant programs A member wants clarification of variable definition prior to implementation.

Region 4 | 08/27/2021

3 strongly support, 10 support, 5 neutral/abstain, 0 oppose, 0 strongly oppose. Region 4 supported this proposal and had the following comments. One attendee commented that primary graft dysfunction remains highly important for every center. They went on to comment that reporting inotropes that are used is variable between centers, and even within centers from cardiologist to cardiologist and surgeon to surgeon. They were also concerned that data could affect individual practice or center practices. The attendee recommended collecting data on whether the patient has primary graft dysfunction: yes or no. They went on to add that collecting hemodynamic data may add to the paperwork and data entry burden of each center, but supported acknowledging the presence of primary graft dysfunction and which device has been used to support. Another attendee commented that the lung TRR forms ask for specific information at 72 hours, adding that as long as the descriptions are as clear as possible, it will not be a big burden for the transplant centers. One attendee commented that this will require clear definitions and timeframes to ensure consistent, meaningful data. Finally, one attendee commented that more data entry is always a problem and was in favor or eliminating data submission when possible.

Jondavid Menteer | 08/18/2021

I strongly support the proposal. There are a couple of details of the data collection that I think could be improved before this is fully approved. 1) primary graft dysfunction Yes/no should be changed to "Is primary graft dysfunction present?" yes/no. This would help in cases where PGD was present but resolved by day 3. This way, if resolved at day 3, the form would say yes at day 1 and no at day 3, reflecting this milder form of PGD. 2) After the data element primary graft dysfunction is selected, I would propose that a "yes" response causes the appearance of two more questions; LV dysfunction and RV dysfunction. If the response is PGD:no, I think these questions should not appear. 3) For clarification, the LV dysfunction and RV dysfunction should be changed to "PGD: left ventricular dysfunction" and "PGD: right ventricular dysfunction" 4) I agree that LV EF should be left for collection on all transplants. I would propose noting on the 24 hours form, that EF from any modality can be reported since transplant (e.g from the postop TEE); and that at 72 hours, EF from any modality performed AFTER 24 hours can be reported (e.g. not from the same study as on the first day, but anything since then). I feel strongly that the EF does not need to be from within the +/- 4 hour window, as this will severely diminish the number of data points we receive in the database. 5) For the elements nitric oxide following transplant. The element on the 24 hour form should read "inhaled nitric oxide after transplant (any time 1st 24 hours)" and on the 72 hour form it should read "inhaled nitric oxide after transplant, still in use". This will help clarify the data and also provide greater utility by specifying whether this was a brief therapy or needed for longer. 6) Same for Flolan, but importantly it should be labeled "epoprostenol", not brand Flolan. The element on the 24 hour form should read "epoprostenol after transplant (any time 1st 24 hours)" and on the 72 hour form it should read "epoprostenol after transplant, still in use". 7) for inotrope dosing, epinephrine dose should read "0.1" not "1.0" in the moderate and high doses. 8) Vasopressin should be in Units per min 9) For vasopressors consider allowing equivalent completion of the doses by allowing mcg/hour and per minute. It seems this should be simple programming. Thank you for doing this critical work to collect this data!

Anonymous | 08/17/2021

As a heart transplant recipient, I believe this change will help care be more consistent with recipients experiencing PGD and possibility the information learned from the additional testing help predict this complication's occurring. I believe an informed patient - with both the good and the bad risks is important in having confidence in the team and decisions made in the care.

Anonymous | 08/13/2021

The fact that there is not a standard definition for primary graft dysfunction is concerning - it seems that this is a wide swath attempt to have heart transplant centers collect data for someone to do a research study. I have concerns that if the initial answer is no to primary graft dysfunction, the center is required to complete the rest of the data. Of particular concern from a cost standpoint is the EF required at 24 and 72 hours. The addition of 2 echo measurements, of which the patient might not need, is concerning. It places a burden on the staff assessing EF, and for those patients with a global payment rate who don't need the test, will cause the center to have to absorb those costs. This is a significant amount of data that will not take a small amount of time to collect and enter. I don't disagree with the concept, but I think the reality needs more work.

Bryan Barrus | 08/12/2021

This data collection has been missing for years and will be very helpful especially as we move toward increasing DCD donors in the country. The time frames for reporting at 24 and 72 hours seem appropriate. The epinephrine range limits seem a little high at >1 mcg/kg/min. If the goal is to detect at what doses of inotropes/vasopressors there is a risk of PGD, lowering the limits to something like 0.5 mcg/kg/min might improve the sensitivity of the analysis. Please make more clear if the committee looking for the max dose of drugs in the first 24 hours or just the dose at the 24 hour time mark post-transplant.

Robert Goodman | 08/04/2021

I agree with the recommendations of the committee as presented in the report.