Reproducibility in retrospective grading of acute graft-versus-host disease after allogeneic marrow transplantation

Abstract

Summary:the manifestations of acute GVHD. In the original reports, all patients had rash as the cardinal sign of GVHD, and We have undertaken a formal study to evaluate the virtually all patients with severe cutaneous GVHD also had reproducibility of retrospective assessments for grading visceral involvement, but it is now clear that severe exfolithe severity of acute GVHD. Using criteria previously ative cutaneous GVHD can occur in the absence of gut or established by the Seattle group, three reviewers indeliver disease and that active visceral GVHD can occur in pendently assigned GVHD severity grades for a set of the absence of cutaneous disease. 2,3 Severe diarrhea is less 100 marrow transplant patients. Significant differences common as a manifestation of GVHD than it used to be, were found in the distribution of GVHD grades assigned perhaps because current prophylactic regimens are more by one of the reviewers as compared to the other two effective than those used 20-25 years ago. 3 Nausea, anorreviewers. In only 40% of cases did all three reviewers exia and vomiting were not explicitly considered in the assign the same GVHD grade, and in only 68-71% of original criteria, but with the increasing use of endoscopic cases did all three reviewers assign the same grade biopsies for diagnosis, these symptoms are now recognized within 0-I vs II-IV or 0-II vs III-IV categories. Despite as manifestations of acute GVHD. 4 Furthermore, the origthe high rate of disagreement between any two inal criteria were developed before veno-occlusive disease, reviewers, at least two reviewers assigned the same viral enteritis and C. difficile enteritis were recognized as overall GVHD grade in 93% of cases. These results sugfrequent complications of marrow transplantation. 3 Intergest that current criteria for assessing the severity of actions between GVHD and other concomitant compli-GVHD by a single reviewer are not sufficiently reliable cations have not been well studied, 5 although guidelines for for rigorous clinical studies. As an alternative to the grading GVHD in such situations have been proposed. 6 original criteria, we have developed and tested simpliResults of an earlier study showed considerable diversity fied criteria that summarize the clinical course of among different reviewers in grading acute GHVD. 7 In that GVHD as reflected by the progression of disease and study, a brief clinical vignette was reviewed by 48 transthe amount of immunosuppressive treatment used to plant physicians worldwide, and the concordance rate for control the disease. Our results suggest that the revised grading GVHD was only 55%. Concordance rates were criteria might yield more reproducible retrospective similarly low with other vignettes designed to test the diaggrading than the original criteria. Although the original nosis of acute GVHD and decisions to treat acute GVHD. criteria and the revised criteria might produce different Although revisions of the original criteria have been proresults for individual patients, the overall distri...

Treatment of steroid-resistant acute graft-versus-host disease with anti-thymocyte globulin

“…Assessment of response of aGVHD to treatment remains complex, and considerable diversity exists among different reviewers in grading aGVHD. 24,25 A revised and simplified staging criteria was proposed by the Seattle group to facilitate retrospective grading of aGVHD. 25 When these criteria were applied to our patient population, the vast majority of patients (92%) would have been classified as grade IV aGVHD, mainly due to extended hospitalizations before post-transplant day 100 and the observations that aGVHDrelated complications represented the leading cause of death.…”

HLAs and risk of acute graft-vs.-host disease after marrow transplantation from an HLA-identical sibling

“…All grading before 1991 was done by a single individual and all grading in 1991 and thereafter was done by another individual [18]. In assigning GVHD grades, reviewers used research files containing copies of pre-and posttransplant inpatient, outpatient, and interim summaries written at Յ30-day intervals, biopsy reports, correspondence with referring physicians, and autopsy reports.…”

“…The analysis was stratified according to the reviewer who graded the GVHD because of significant differences in the distribution of grades between the two reviewers (Table 2) [18]. By permutation tests, we found statistically significant evidence of heterogeneity in risk of GVHD among patients categorized according to different HLA-DQ antigens (p ϭ 0.05) and suggestive evidence for heterogeneity in the risk of GVHD among patients categorized according to different HLA-B antigens ( p ϭ 0.09) ( Table 3).…”

Urinary Elafin and Kidney Injury in Hematopoietic Cell Transplant Recipients

“…The mean baseline urinary elafin (6SD) was 190263754 pg/ml (median, 1260 pg/ml; range, 30-48,273 pg/ml). The median number of post-HCT urinary elafin measurements per patient was 6 (range, [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]. The mean urinary elafin value among 1512 post-HCT measurements was 197661814 pg/ml (median, 1440 pg/ml; range, 3.6-20,179 pg/ml).…”