2015 ISCD Position Development Conference
Executive Summary of the 2015 ISCD Position Development Conference on Advanced Measures From DXA and QCT: Fracture Prediction Beyond BMD

https://doi.org/10.1016/j.jocd.2015.06.013Get rights and content

Abstract

There have been many scientific advances in fracture risk prediction beyond bone density. The International Society for Clinical Densitometry (ISCD) convened a Position Development Conference (PDC) on the use of dual-energy X-ray absorptiometry beyond measurement of bone mineral density for fracture risk assessment, including trabecular bone score and hip geometry measures. Previously, no guidelines for nonbone mineral density DXA measures existed. Furthermore, there have been advances in the analysis of quantitative computed tomography (QCT) including finite element analysis, QCT of the hip, DXA-equivalent hip measurements, and opportunistic screening that were not included in the previous ISCD positions. The topics and questions for consideration were developed by the ISCD Board of Directors and the Scientific Advisory Committee and were designed to address the needs of clinical practitioners. Three task forces were created and asked to conduct comprehensive literature reviews to address specific questions. The task forces included participants from many countries and a variety of interests including academic institutions and private health care delivery organizations. Representatives from industry participated as consultants to the task forces. Task force reports with proposed position statements were then presented to an international panel of experts with backgrounds in bone densitometry. The PDC was held in Chicago, Illinois, USA, contemporaneously with the Annual Meeting of the ISCD, February 26 through February 28, 2015. This Executive Summary describes the methodology of the 2015 PDC on advanced measures from DXA and QCT and summarizes the approved official positions. Six separate articles in this issue will detail the rationale, discussion, and additional research topics for each question the task forces addressed.

Introduction

The International Society for Clinical Densitometry (ISCD) is a nonprofit professional organization dedicated to the advancement of assessment of musculoskeletal health, particularly (but not limited to) bone densitometry. A major focus of the society is the development of guidelines and establishment of standards for bone densitometry, assessment of fracture risk, and other aspects of musculoskeletal measurement. In recognition of the many scientific advances in the use of dual-energy X-ray absorptiometry (DXA) scans for measures other than bone density, and further advances in fracture risk assessment using computed tomography (CT), the ISCD convened a Position Development Conference (PDC) on the use of advanced measures from DXA and quantitative computed tomography (QCT) to assess fracture risk.

The society conducts PDCs every 2–3 years to develop guidelines and standards (expressed as position statements) for new technologies used to assess musculoskeletal health and fracture risk, and to update older guidelines and standards as new data become available. The ISCD official positions are widely used by clinicians and densitometry technologists as a reference regarding the indications for, acquisition of, and interpretation and reporting of measures of musculoskeletal health, as well as incorporation of those measures into fracture risk assessment. The curricula of the densitometry educational courses provided by ISCD are heavily influenced by these positions.

The ISCD PDC process is designed to summarize and use the best scientific evidence available to develop and update position statements regarding musculoskeletal assessment. Because musculoskeletal assessment technologies are evolving, clinically important issues are sometimes addressed in the absence of robust evidence and thus are largely based on expert opinion. However, the PDC process grades and highlights the limitations of the available evidence pertinent to each statement and indicates where additional research is needed to improve the scientific evidence on which positions are based as well as to resolve areas of ambiguity and controversy. Position statements from prior PDC's held in 2001, 2003, 2005, 2007, 2010, and 2013 have been published 1, 2, 3, 4, 5, 6, 7, 8, 9. The most recent PDC was held in Chicago, IL, USA, February 26 through February 28, 2015. This article describes the methodology of this PDC, and the results from the topics regarding advanced measures from DXA and QCT for fracture prediction beyond BMD.

Potential topics for the 2015 ISCD PDC were solicited from the ISCD Board of Directors, executive committee, and members of the scientific advisory committee. Candidate topics were then ranked in order of importance. Trabecular bone score (TBS) was identified as an important new application of DXA requiring new guidelines and standards. However, TBS is only one in a class of non-BMD measures from DXA scans that had not been previously reviewed by the ISCD: hip geometry measures including hip structural analysis (HSA), hip axis length (HAL), and neck-shaft angle (NSA). In addition, there have been recent advances in the use of CT to assess fracture risk using either QCT of the hip, advanced analysis algorithms such as finite element analysis (FEA), or the opportunistic evaluation of CT scans acquired for nonbone-related clinical reasons.

A steering committee was formed consisting of the President-Elect of ISCD (John A. Shepherd, who served as chair), the current ISCD president (William D. Leslie), and past president (John T. Schousboe), all with extensive experience leading and moderating prior PDCs. Scientific questions within each topical area were chosen by the PDC steering committee. The questions asked of the task forces regarding the clinical utility of the non-BMD measures from DXA and QCT were as follows.

  • Can the measure be used to diagnose osteoporosis?

    • Are there adequate reference data to use the measure clinically?

    • Are reference data for the measure affected by age, sex, and ethnicity?

  • Can the measure be used to assess fracture risk?

    • How should the measure be used relative to other fracture risk assessment tools?

  • Can the measure be used for monitoring?

    • Monitoring of individuals not receiving treatment?

    • Monitoring of individuals receiving treatment?

    • What is the short-term and long-term precision of the measure in routine practice?

    • What are the quality assurance and quality control (QA/QC) criteria for the measure?

  • Are there special conditions where the measure has greater or lesser clinical utility?

    • Clinical role in Glucocorticoid-induced osteoporosis (GIOP) or other secondary osteoporosis?

    • What are the limitations and contraindications to using the measure?

  • What are the required elements in a clinical report of the measure?

Formation of and function of task forces: 3 task forces were formed to address the questions in 3 primary areas: TBS, hip geometry (including HSA), and QCT. Task force chairs were selected because of their high level of expertise in these areas. Along with the steering committee, they selected additional experts in the field to serve as task force members.

Each task force was asked to perform a literature search using the methodology of previous PDCs, focusing on PubMed, MEDLINE, and Embase databases. Informed by the published literature, the task forces had the option to refine the questions initially proposed, in consultation with the steering committee. Each task force then drafted proposed position statements to address the questions assigned. The task force chairs, in consultation with task force members, then wrote detailed reports describing the proposed position statements, supporting rationale and literature.

Formation of and function of the expert panel: starting with the adult and pediatric PDCs of 2007, the ISCD PDCs have followed the RAND-UCLA method of rating the appropriateness of candidate position statements RAND/UCLA Appropriateness Method (RAM) that is described in more detail in the next section (10). This method requires that an expert panel, completely separate from the task forces, be formed to review and rate the proposed position statements after considering the supporting documents. Expert panel members were chosen on the basis of 4 criteria; (1) acknowledged expertise in at least one of the topic areas; (2) representative of different geographic regions; (3) representative of other professional societies with a commitment to the field of musculoskeletal measurement; and (4) a keen ability to weigh scientific evidence. Accordingly, 12 international experts were invited and agreed to serve on the expert panel, were able to attend all presentations from each task force as well as the deliberations of the expert panel, and rated the proposed position statements.

The 2015 PDC procedures, including the literature reviews performed by the task forces, formulation of position statements to address the questions posed to the task forces, and the rating of those statements, all followed a modified RAM 2, 4, 10. The RAM not only includes ratings as to whether or not a statement is appropriate but also explicitly grades the quality of the evidence on which the statement is based. This process recognizes that although based on the currently available evidence, a statement may be considered to be appropriate although the evidence may be weak. Grading of the quality of evidence places the statement in the appropriate context.

The task forces presented the proposed statements and the supporting scientific evidence for each statement in 2 steps. The task forces finished their literature reviews and draft documents by January, 2015. These were sent to all expert panelists, who conducted their initial round of rating from January 15 through January 29, 2015 without consulting each other.

Grading of the official positions: all statements were rated by each expert panelist in 4 areas.

  • (1)

    Appropriateness. The appropriateness of each statement was voted on a scale of 1–9 with 1 representing “highly inappropriate” and 9 representing “highly appropriate”. Statements that had a median score of 1–3 were rated as “inappropriate”, those with a median score of 4–6 as “uncertain”, and those with a median score of 7–9 as “appropriate”. Statements were considered to be “appropriate without disagreement” if, in addition to a median score of 7–9, no more than 3 expert panelist ratings fell outside this range.

  • (2)

    Quality of evidence. The quality of the evidence supporting each statement was rated as being good, fair, or poor. “Good” evidence was from 2 or more well-designed prospective studies (randomized controlled trials or high quality observational studies). “Fair” evidence was judged to be sufficient to determine effects on outcomes but limited by the number, quality, or consistency of the available studies. “Poor” evidence was judged to be insufficient to determine effects or consequences of implementing the statement on outcomes, due to the number of available studies, flaws in their design or conduct, major gaps in the chain of evidence, or conflicting evidence.

  • (3)

    Strength of recommendation. The strength of the recommendation to implement the statement was rated as; (A) strong recommendation supported by the available evidence; (B) recommendation supported by the available evidence; or (C) recommendation supported primarily by expert opinion. It should be noted that a statement could be rated as being supported by a good quality of evidence and yet have a level C strength of recommendation, if the beneficial consequences of implementation of the statement were judged to be slight or offset by negative consequences.

  • (4)

    Applicability. Statements were rated as applicable throughout the world (worldwide) or applicable according to local requirements (local).

Second round expert panel voting at the PDC in Chicago, IL: based on the initial rating by the expert panel, the task forces revised their initial documents and presented these revised documents in detail to the expert panel in sessions open to the public on February 28, 2015. Statements that were rated as inappropriate in the first round of rating were not considered further or presented. Statements that were rated as appropriate without disagreement in the first round of expert panel voting were presented briefly and, with a few exceptions, accepted and rated again as appropriate in their original wording and content. A few of these statements, however, were altered at the behest of the expert panel before being rated as appropriate during the second round.

Most of the time in the open sessions was spent debating statements that were rated as uncertain. Some of these statements were rated as inappropriate on the second round votes, some were again voted as uncertain, and some that were voted as uncertain on the first round were voted appropriate without disagreement on the second round.

Final selection of and approval of the 2015 ISCD official positions: the expert panel, task force chairs, and the 2015 PDC steering committee met again in closed session on March 1, 2015, to refine the final wording of the statements that would be submitted to the ISCD Board of Directors for review and approval. The final statements rated as appropriate without disagreement, along with their ratings regarding quality of evidence, strength of recommendation, and applicability, and the supporting literature reviews, were reviewed by the ISCD Board of Directors. The ISCD Board of Directors voted to approve all these statements on May 14, 2015. The rationale supporting these positions is presented in 6 separate articles also published in this issue of JCD 11, 12, 13, 14, 15, 16.

The list of individuals who comprised the 2015 PDC steering committee, served on each of the task forces, or served on the expert panel is shown in Appendix A. All PDC participants were required to fully disclose any real or perceived conflicts of interest using the ISCD disclosure form. Any potential conflict of interest was thoroughly vetted with the Chair of the respective task force, steering committee liaison to the task force, and the PDC steering committee before appointment. Industry scientists and representatives were excluded from participating as task force or expert panel members but were allowed to act as technical consultants to the task force to ensure that position statements were not incompatible with the technology itself.

The following industry partners gave financial support that made the PDC possible; Amgen, Inc. (Thousand Oaks, California, USA) and Merck, Inc. (Kenilworth, New Jersey, USA) Industry partners providing financial support played no role in the PDC or its recommendations.

A summary of all the ISCD positions for this and all past PDCs that remain current is shown in Appendix B.

Trabecular Bone Score

  1. (1)

    TBS is associated with vertebral, hip, and major osteoporotic fracture risk in postmenopausal women.

    • Grade: good-B-W

  2. (2)

    TBS is associated with hip fracture risk in men >50 yr.

    • Grade: fair-B-W

  3. (3)

    TBS is associated with major osteoporotic fracture risk in men >50 yr.

    • Grade: fair-C-W

  4. (4)

    TBS should not be used alone to determine treatment recommendations in clinical practice.

    • Grade: good-A-W

  5. (5)

    TBS can be used in association with FRAX and BMD to adjust FRAX-probability of fracture in postmenopausal women and older men.

    • Grade: good-B-W

  6. (6)

    TBS is not useful for monitoring bisphosphonate treatment in postmenopausal women with osteoporosis.

    • Grade: good-A-W

  7. (7)

    TBS is associated with major osteoporotic fracture risk in postmenopausal women with type II diabetes.

    • Grade: fair-B-L

Hip Geometry
  1. (8)

    HAL derived from DXA is associated with hip fracture risk in postmenopausal women.

    • Grade: fair-B-L

  2. (9)

    The following hip geometry parameters derived from DXA (cross-sectional area [CSA], outer diameter [OD], section modulus [SM], buckling ratio [BR], cross-sectional moment of inertia [CSMI], NSA) should not be used to assess hip fracture risk.

    • Grade: fair-B-W

  3. (10)

    Hip geometry parameters derived from DXA (CSA, OD, SM, BR, CSMI, HAL, NSA) should not be used to initiate treatment.

    • Grade: fair-B-W

  4. (11)

    Hip geometry parameters derived from DXA (CSA, OD, SM, BR, CSMI, HAL, NSA) should not be used for monitoring.

    • Grade: good-A-W

Quantitative CT
  1. (12)

    QCT acquisition of the proximal femur should extend from the femoral head to the proximal shaft.

    • Grade: good-A-W

  2. (13)

    For density-based QCT measurements the in-scan calibration phantom can be replaced by asynchronous calibration if scanner stability is maintained.

    • Grade: fair-B-W

  3. (14)

    Opportunistic CT to screen for patients with low BMD or low bone strength of the spine or proximal femur is possible only if validated machine-specific cutoff values and scanner stability have been established.

Grade: Fair-C-W
  1. (15)

    Femoral neck and total hip T-scores calculated from 2D projections of QCT data are equivalent to the corresponding DXA T-scores for diagnosis of osteoporosis in accordance with the WHO criteria.

    • Grade: fair-B-W

  2. (16)

    Total femur trabecular BMD measured by QCT predicts hip fractures and hip BMD measured by DXA in postmenopausal women and older men.

    • Grade: fair-B-W

  3. (17)

    Where QCT and DXA are both available and provide comparable information, DXA is preferred to limit radiation exposure.

    • Grade: fair-C-W

  4. (18)

    Integral and trabecular BMD of the proximal femur measured by QCT can be used to monitor age- and treatment-related BMD changes.

    • Grade: fair-B-W

FEA by QCT
  1. (19)

    Vertebral strength as estimated by QCT-based FEA predicts vertebral fracture in postmenopausal women.

    • Grade: fair-B-W

  2. (20)

    Vertebral strength as estimated by QCT-based FEA is comparable to spine DXA for prediction of vertebral fractures in older men.

    • Grade: fair-B-W

  3. (21)

    Femoral strength as estimated by QCT-based FEA is comparable to hip DXA for prediction of hip fractures in postmenopausal women and older men.

    • Grade: fair-B-W

  4. (22)

    FEA cannot be used to diagnose osteoporosis using the current WHO T-score definition.

    • Grade: good-A-W

  5. (23)

    Vertebral or femoral strength as estimated by QCT-based FEA can be used to initiate pharmacologic treatment using validated thresholds and in conjunction with clinical risk factors.

    • Grade: fair-B-W

  6. (24)

    Vertebral or femoral strength as estimated by QCT-based FEA can be used to monitor age- and treatment-related changes.

    • Grade: fair-B-W

Section snippets

Acknowledgments

As was true of prior Position Development Conference's (PDC's), the 2015 International Society for Clinical Densitometry (ISCD) PDC was made possible only through the extensive voluntary efforts of large numbers of individuals, notably the clinical and scientific experts who donated countless hours over the past 18 mo as task force chairs, task force members, or expert panel members. The ISCD wishes to acknowledge and thank these individuals for their extraordinary service. Additionally, the

Glossary

ALMI–appendicular lean mass index
BMC–bone mineral content
BMD–bone mineral density (equivalent to areal BMD, aBMD)
BMI–body mass index
BR–buckling ratio
CSA–cross-sectional area
CSMI–cross-sectional moment of inertia
DXA–dual-energy X-ray absorptiometry
FEA–finite element analysis
FMI–fat mass index
FRAX–World Health Organization Fracture Risk Assessment Tool
HAL–hip axis length
ISCD–International Society for Clinical Densitometry
LMI–lean mass index
LSC–least significant change
NHANES III–National Health

References (16)

There are more references available in the full text version of this article.

Cited by (216)

View all citing articles on Scopus
View full text