Disseminated Intravascular Coagulation (DIC)
6 August 2005
16:00 to 19:30
Tumbalong Meeting Room
Sydney Convention and Exhibition Centre
Chairman: K. Hoots, USA
Co-chairs: J. D. Nielsen, Denmark; C-H. Toh, UK; H. Wada, Japan
Part A
The DIC Subcommittee meeting of ISTH met for its annual meeting in the Tumbalong Room of the Sydney Convention Center on Saturday, August 6, 2005. Approximately 90-100 members participated for the session which summarized Subcommittee progress (See Part B, below), discussed areas in need of continued investigation, and reached consensus on important proposed modifications on the DIC algorithm for future consideration. Specifically, the following recommendations were made that were a direct result of discussion of participant members (and were in addition to other proposed alterations discussed in Part B which were likewise affirmed without discussion):
Part B
The DIC Subcommittee of the International Society of Thrombosis and Haemostasis met in Houston, TX, USA on April 29- May 1, 2005 to plan the agenda for the Subcommittee meeting at the ISTH meeting in Sydney in August. In addition, we discussed the accomplishments arising from subcommittee endeavors and assessed future challenges in advancing the diagnosis and therapy of DIC-associated syndromes. The following summarize these latter discussions. Attached to this document are the agenda and a roster of subcommittee chairs and other subcommittee members who participated actively in this planning process.
Most of the recent effort of the Subcommittee has focused on validating and refining the DIC Algorithm [1] that was developed by us to be used in the diagnosis of Overt and Non-overt DIC. Further, an important secondary purpose for developing the Overt and Non-overt Algorithms was to develop prospective stratification strategies for therapeutic trials of new (or established) drugs for DIC. We have made significantly greater progress in the former. Specifically, since our publication of the Taylor, et al. article in 2001, no fewer than 6 peer-reviewed manuscripts whose original design was predicated (at least in part) on validating the overt DIC algorithm have been published by members of this Subcommittee.
Specifically, Wada, et al has compared the sensitivity and specificity of our ISTH algorithm with a refined version of the original algorithm developed out of the Japanese Ministry of Health and Welfare. [2] Levi, et al (past chair of this Subcommittee) has published 2 validation studies of the overt algorithm using sequentially larger cohorts (>50 to >240) that indicate that the positive and negative predictive value among consecutive patients admitted to the ICU of a single institution (Univ. of Amsterdam) is extraordinarily high when compared to blinded “expert” clinical assessments of DIC. [3, 4] Further, the presence of DIC by the algorithm was highly predictive of mortality in these cohorts.
Similarly, Dempfle, et al has published similar ICU cohort data that validates the utility of the Overt algorithm in cohorts in Mannheim, Germany, assessing, in particular, the essential role that measurement of fibrin markers (indicative of both fibrin cross-linking and fibrinolysis) play in the Overt DIC Algorithm. [17]
Toh and his colleagues in Liverpool have demonstrated that among subjects consecutively admitted to their ICU, both the Overt and Non-Overt Algorithm predict patients with high mortality, although not typically with the same pathogenetic course. [6] Data from this paper suggest that early abnormalities of coagulant markers parallel those of inflammatory harbingers of bad outcome. This observation is consistent with a basic pathogenetic premise described in the original Taylor paper: that is that quantification of dysfunction of the “microvascular-endothelial” organ provides important predictability of clinical outcome when biologic markers of inflammation and coagulation are markedly abnormal early in a disease process predisposing to DIC.
Single reports, as important as they have been as markers of Subcommittee progress, however, are not sufficient. It is for this reason that we propose two new submissions as Subcommittee Reports for publication in JTH. This first to be developed this year, will summarize the progress (cited above) to date. The second will suggest refinements to the algorithm based on the studies to date and those that are presently on-going.
As these published reports have indicated, the ultimate clinical utility of these DIC algorithms requires their being married to validated scoring of injury and disease in the critical care setting. This proves to be a significant challenge for the Subcommittee as most of the potential use of the algorithms (and their application to describe the pathogenesis of vascular injury in severely ill patients) lies with physicians whose primary training and affiliation are in Critical Care Medicine rather than Hematology. Our strategies to address this formidable challenge will be described below.
However, before we can export our coagulopathic/inflammatory view of severe disease, the Subcommittee has been working to incorporate the language of vascular injury (which is, again, the theoretical basis to our algorithm development) into the lexicon of severe disease. To achieve this ambitious aim, Subcommittee Chairs, Co-Chairs and members have been striving to incorporate this vascular “world-view” more completely into the concept of DIC for the breadth of the clinical community. One early, but promising strategy has been for us to “re-write” standardized textbooks and reviews of DIC from this vascular injury perspective. Over the last 3-5 years our members have incorporated this biologic view into a neurosurgical journal and textbook, [7, 8] updated classical textbooks of hematology [9-11], in a fundamental discussion in the microvascular research arena [12] and to the broader community of clinicians via a comprehensive review. [13] Further, the concepts on which we have predicated the work of our Subcommittees have been cited in State of the Art discussions of Sepsis [14, 15] and Trauma. [15] These slow, but steady inoculations of new DIC concepts (e.g. Overt versus Non-Overt DIC as clinical spectra of disease) appear to be gradually gaining acceptance in diverse medical subspecialties. However, as a subsequent paragraph will discuss, some areas of medicine (e.g. Obstetrics and Gynecology) have to date largely been ignored in this endeavor. Efforts to fill these gaps constitute significant future work by this Subcommittee. One approach will be to initiate a dialogue with the subcommittee of Hemostatic/Thrombotic Issues of Women chaired by Dr. Marilyn Manco-Johnson. We will focus this initial effort by having Dr. Benjamin Brenner of the University of Illinois-Chicago discuss obstetrical DIC at our Subcommittee Meeting in Sydney. In addition, we have identified several other clinician scientists with both hematologic and obstetrical expertise to work with us in this initiative.
One clinical discipline where we are progressing is pediatrics. Drs. Prasad Mathew (Co-Chair, Pediatrics) and Marilyn Manco-Johnson (Chair, Women’s Health in Thrombosis & Haemostasis Subcommittee) are initiating a retrospective analysis in the Mountain States Region of the U.S. of whether the algorithms (in the present iteration, or perhaps in a modification specifically for very young patients) are as reflective of the biology and outcome as has been demonstrated in adults. Dr. Deborah Brown, a Subcommittee member, will initiate a single institution retrospective and prospective investigation which will contribute to this effort. Both will examine pre-term and term neonates as well as older children in the Pediatric ICU setting. These efforts constitute a major emphasis of the Subcommittee for 2005-2006.
In addition to efforts by Subcommittee chairs (past/present), Co-chairs and members to extend the clinical capacity to diagnose and treat DIC, we have been working with collaborators whose laboratories are attempting to define more precisely what the vascular response to injury is at the cellular and subcellular level. Okajima has presented at our meeting in Houston and has submitted for publication exciting new work on the role that sensory neurons appear to play in the upregulation of inflammation and importantly, how tight thrombin regulation may be essential to alleviating this effort. Recent work on the relationship between PAR receptor activation and endothelial protein C receptor [16] have intensified efforts to define the cellular and molecular response modifiers of vascular injury-all of which offer the prospect of defining more precisely how therapeutic molecules (e.g. activated protein C, Antithrombin, Tissue Factor Pathway Inhibitor, etc.) may or may not be relevant for clinical subsets of at-risk ill patients prone to develop DIC. In this spirit, members of this subcommittee (e.g. Nigel Key) are initiating a dialogue on our behalf with the Vascular Biology Subcommittee (particularly as it relates to Tissue Factor and platelet microparticles) to explore future opportunities for collaboration.
As discussed, we will prepare two Subcommittee Reports for consideration by the SSC for publication in JTH. The first will be a straight-forward report of or progress in validating the Overt and Non-Overt DIC algorithms. The second will make refinements to the algorithms based on new clinical data that we may be able to improve the positive and negative predictive values of the scores by defining parameters more precisely and/or by making categorical distinctions based on the cause of the DIC.
A goal which has preoccupied us for the last 18 months and which has engendered much investigation by Dr. Karl-Ehrich Dempfle (Co-Chair, Fibrinolysis; Member, DIC) has been to establish reliable cut-offs for the fibrin markers. Since most clinical labs employ D-dimers as their measurement of choice, Dr. Dempfle has extensively studied standardizing reagents for the D-dimers test as well as trying to establish quantitative “cut-points” of a “standardized” D-dimer measurement to score in both the Overt and Non-Overt algorithm. [17] Significant progress has been made and, based on this work, new parameters for scoring fibrin marker (D-dimer) in the algorithms should be ready for the next subcommittee report next year. Additionally, we propose to re-examine the contribution of fibrinogen measuremen to the algorithms.
Another challenging dilemma has arisen from the work of Wada and colleagues. They have demonstrated that the utility of the overt algorithm may be better for sepsis-induced DIC compared to cancer-associated DIC. [18] This led our Japanese colleagues to modify the JMHW DIC algorithm which appears to make it somewhat more sensitive to cancer-associated DIC. Our approach may be different: one possibility is to utilize a different cut-off in the overt Algorithm (the standard >5) for sepsis and other non-cancer associated DIC states and lower score (e.g. 4) for cancer-related DIC. This approach will require significant further investigation and Dr. Wada is obtaining some preliminary data about such an adaptation.
Although microvascular thrombosis leading to organ failure (MODS) clearly predominates as the morbid and/or mortal outcome of DIC initiated by sepsis and most other causes, there are subsets of DIC patients in whom a hemorrhagic diathesis from procoagulant consumption contributes significantly to clinical morbidity (certain oncologic-associated DIC syndromes are examples). The work by Wada and others will hopefully better define these specific entities. Patients in these unique sub-groups of DIC my benefit from newer hemostatic options (e.g. recombinant Factor VIIa). [19] Precise sub-categorization of The DIC syndrome by the Subcommittee may help identify patient subsets in whom these theoretically risky interventions may confer a positive benefit risk.
Assessing the utility of other biological markers (e.g. Antithrombin, TAT or Protein C levels) for the Non-Overt Algorithm in particular needs further data. Like the D-dimer, these tests suffer from standardization problems. Further, it is still unclear whether measuring them in real time (in laboratories where this is even feasible) contributes significantly to enhanced sensitivity and specificity of the algorithm.
Equally important is the question of whether serial determinations of such biological markers will assist treating physicians in monitoring the impact of therapeutic maneuvers. Assessment of the latter is likely a much longer-term goal and will likely require the implementation of our algorithms in prospective stratification of large clinical trials to answer (see below). Another area of interest which is being pursued in local settings by several of the members (e.g. Dempfle, Escobar) is the use of “point of care” coagulation testing and monitoring in the ICU setting. It is still quite unclear whether we can adapt such measurements for incorporation into the algorithms. Nonetheless, it is a valid area of investigation that the subcommittee will follow with great interest.
Another important pathogenetic clarification that we need to refine in light of recent data [6] is how to discriminate the Overt-Non-Overt spectrum of DIC from a previously employed concept of “Compensated” versus “Uncompensated” DIC. It is apparent that these terms are not entirely interchangeable since MODS and death can occur in some individuals who by scoring have “Non-Overt” DIC, never progress before death to Over DIC but die from organ failure nonetheless. Removing all confusion with regards to these labels will likely require better understanding of individual phenotype. [20]
As was alluded previously, it has always been a primary goal of this Subcommittee to have its algorithms employed prospectively to stratify and then to monitor longitudinally (in the case of Non-overt algorithm) patients randomized in clinical trials of therapeutic interventions. [5] This will require a more proactive approach by the Subcommittee Members than we have previously assumed since most of these trials originate through collaborations between Pharma and Critical Care Physician Investigators. This, however, is consistent with our purposeful effort to insinuate ourselves as hematologists into a more aligned position with these investigators. Our plans for accomplishing this are discussed below. However, even if we are successful, we will need to bring to the table an organized proposal which incorporates expertise in epidemiology, trial design and statistics (skills that have not been traditional emphases for us up to now). However, M. Levy has recent experience in DIC trials and Dr. Hoots has directed a AT-placebo DIC Trauma trial in the past. [8, 21] In addition, we hope to attract individuals from the academic and Pharma Communities to extend our competencies in these areas. Cogent arguments based on accepted pathogenetic models of DIC definable with our algorithms will likely improve the likelihood that sponsors would be receptive to their use in trial design of new therapies. Possibilities for exploring this approach may include proposed trials by Genzyme, ESINAI, APEX, and Chiron.
For the next year, one of our highest priorities is to “sell” our biologic concepts of DIC and our algorithms to the critical care medicine community internationally. We want to work together with them to harmonize disease scoring (e.g. APACHE II and MOD Score) with the DIC Score using published studies and recent data. Dr. Gary Kinesawitz, an academic Critical Care Physician and colleague of Fletcher Taylor at the University of Oklahoma as well as a member of our Subcommittee for the last several years, will lead this effort. He is working with Subcommittee members to create a pre-planned symposium on DIC pathogenesis and prognostic scoring of very ill patients to be offered to organizers of key Critical Care international meetings (e.g. American Thoracic Society, European Critical Care). We hope this will open a dialogue about such harmonization. It may also create opportunities to explore the development of clinical guidelines. This represents an initial, but hopefully significant step towards bridging the gap between the hematologic and critical care view of DIC.
In summary, the DIC Subcommittee continues to push a very aggressive agenda. Summarized are our accomplishments to date and our short-and intermediate-term aims and strategies for their respective pursuits. For the long term, we hope to incorporate new knowledge about individual risk factors as implied by genetic polymorphisms that influence response to vascular injury and new means of measuring biologic processes (proteomics) into an evolving strategy to define DIC more precisely.
References: