Grading to Guidelines: Using the GRADE Methodology to Guide Plasma Transfusion
By Ashley Smith

Physicians often are faced with questions regarding the best way to treat a patient, and many refer to guidelines developed by professional associations or groups with specific expertise. In the session “Evidence-Based Guidelines for Plasma Transfusion” (9238-TC), attendees learned how the AABB Clinical Transfusion Medicine Committee (CTMC) has approached the development of guidelines on plasma transfusions, including a systematic review of the literature and the implementation of the GRADE methodology for analyzing data and developing the guidelines.

Why Develop Guidelines?

The question of why guidelines are needed and wanted by the medical community is a common one, and as John D. Roback, MD, PhD, of Emory University and chair of CTMC, explained, “Guidelines identify best clinical practices and are most valuable when they are developed by pulling in lots of expertise from different areas to answer a complex clinical question.” They also are valuable tools because guidelines can provide useful and needed information to those in a variety of specialties, can promote more acceptance of these practices, and stimulate research initiatives into areas where evidence showing efficacy is lacking.

Recognizing the need for clinical guidelines on the appropriate use of blood components, the AABB Board of Directors included guideline development in the association’s strategic plan and asked CTMC to begin by focusing on plasma usage guidelines.

Traditionally, there are varying approaches to developing guidelines, including literature reviews, consensus conferences and systematic reviews, but each has drawbacks. “Consensus conferences often are dominated by one or two strong personalities, limiting full discussion of the issues, and they are not explicit in how the information was developed,” Roback said. When evaluating what process to follow for its guidelines, CTMC wanted to ensure there were explicit, transparent systematized approaches for deriving the guidelines from study evidence. The committee decided to follow a new system gaining popularity, the GRADE methodology — Grading of Recommendations, Assessment, Development and Evaluation.

What is GRADE?

The GRADE system, which has been endorsed by numerous U.S. and worldwide professional societies and other health care organizations, relies on four main steps: 1) assemble the guidelines review group; 2) formulate the clinical question(s); 3) perform a thorough search of the relevant literature followed by a systematic review and statistical analysis; and 4) prepare evidence-based guidelines following an explicit step-by-step methodology.

To develop the review group, CTMC decided to expand its own expertise by including outside stakeholders and consultants familiar with the GRADE methodology. This group then was tasked with brainstorming the clinical questions to be evaluated. “We asked the group to determine the most important areas of plasma transfusion that should be considered, and we wanted to generate the questions from scratch before anyone reviewed the data,” Roback explained. The group divided plasma transfusions into two categories — medical treatments with no procedures and surgical treatments with minor or major procedures — and then these groups where subdivided until the areas that most often are associated with questions regarding transfusions were identified.

Step Three: Systematic Review

AABB contacted the Knowledge and Encounter Research (KER) Unit at the Mayo Clinic to perform the systematic review of literature based on four questions drafted by the committee. “The overarching question we focused on was what are the benefits and harms of plasma transfusion in different clinical scenarios,” said M. Hassan Murad, MD, MPH, of the KER Unit. As background for the review, the group noted that the number of fresh frozen plasma (FFP) transfusions continues to steadily increase, FFP use in the United States exceeds use in other countries, and the use often is outside guidelines and consensus panel statements.

For the literature review, the KER Unit used numerous electronic databases to identify 713 potential references. After an evaluation process that specified eligibility criteria — randomized control trials and observational studies that enrolled a group of patients who received FFP and a control group that received no FFP, a lower dose of FFP or a lower FFP:RBC ratio  — 29 studies were selected for a detailed statistical analysis, including a meta-analysis.

Blinded pairs of reviewers extracted the data included in these 29 studies and assessed the quality of the references based on predetermined criteria. The group then evaluated the data and focused on specific clinical outcomes, including death, nonfatal myocardial infarction, stroke, acute lung injury, multi-organ failure, blood loss and transfusion requirements.

According to Murad, the quality of the data in the studies was low, but it showed that transfusion of FFP led to significant reduction in the risk of death in patients undergoing massive transfusion. This effect, however, was not demonstrated in other populations. In addition, the transfusion of FFP led to a significant reduction in the risk of multi-organ failure but increased the risk of acute lung injury. FFP transfusion also did not reduce blood loss or transfusion requirements, and there were “sparse data and poor reporting regarding the risk of stroke.”

Evidence Summaries for Guidelines Development

Once the systematic review was completed, CTMC also worked with Mayo to produce evidence summaries that will assist in the development of the guidelines. “Decisions should be based on systematic summaries of the pertinent research, and the better the research, the more confident the decision,” said Victor M. Montori, MD, MSc, also from the KER Unit. He reminded the session attendees, however, that evidence alone is never sufficient to make a clinical decision.

The systematic and explicit approaches of the GRADE system aid in the development of evidence summaries and guidelines because they help protect against errors, highlight sources of disagreement, facilitate critical appraisal and offer clear guidance.

Recognizing these capabilities, the KER group has encouraged CTMC to review the evidence summaries in order to make what Montori dubs “good recommendations.” To be a good recommendation, it must have specific actions (such as “do” or “don’t do”), be pertinent and tailored to specifics, and be offered with confidence that following the recommendations will yield more good than harm and that the assessment is accurate.

When developing the evidence summaries, the KER unit focused on two components of the grading system — the quality of the evidence and the strength of recommendation. Montori defined quality of evidence as the extent to which our confidence in an estimate of effect is adequate to support a particular recommendation, and explained that it can be divided into four levels:

·          High, which means further research is very unlikely to change our confidence in the estimate of effect.

·          Moderate, which indicates that additional research is likely to have an important impact on the our confidence and may change the estimate.

·          Low, indicating that more research is very likely to have an important impact on our confidence and is likely to change the estimate.

·          Very low, which means any estimate of effect is very uncertain.

The strength of recommendation is the degree of confidence that patients will be better off following the recommendation than following an alternative course of action. “You must look at the balance of desirable effects — health benefits, less burden and savings — versus undesirable effects — harm, more burden and costs to the community,” Montroi said. “In a strong recommendation, almost all patients would choose the recommendation, and a weak recommendation indicates that informed patients would vary between choosing the recommendation or another treatment option.”

The Next Steps for Plasma Guidelines

The KER Unit at the Mayo Clinic has completed the systematic review and produced the evidence summaries. CTMC recently signed off on the information gathered, and Roback noted that “This was a longer and more challenging process than we expected because we wanted to understand why certain studies were or were not included.” The committee now will use this information to develop the evidence-based guidelines following the explicit step-by-step GRADE methodology.

The original four questions for the guidelines have morphed to six, which Roback said is not uncommon. As the questions are assessed and the guidelines are drafted, CTMC will examine the quality of the evidence for each outcome, determine the relative importance of each outcome, and work to balance the benefits and harms to create net benefits. “The net benefits will then be balanced with potential costs to the community, both financial and operational, before we determine the strength of the recommendation,” Roback said. “And, the final step will be for the panel to endorse the results.”

Once the guidelines are finalized, they will be made available for member comment. Once revisions are completed, the final guidelines will be published in a peer-reviewed journal, and Roback emphasized that they will include an emphasis on the areas in which there is a lack of evidence supporting appropriate clinical practice. He also noted that CTMC already is thinking ahead to the next guidelines. “We plan to focus on red blood cell transfusions in the next set of guidelines,” he concluded.