Several grading systems have been developed in the bone marrow transplantation setting in attempts to predict survival in patients with chronic graft-versus-host disease (cGVHD). In this study, we evaluated the prognostic value of the National Institutes of Health (NIH) scoring system and investigated for any additional prognostic factors in a series of 171 patients undergoing peripheral blood stem cell transplantation (PBSCT) from matched related donors. The cumulative incidence of cGVHD was 70%; cumulative incidences of mild, moderate, and severe cGVHD were 29%, 42% and 28%, respectively. Overall, 68% of patients were free from immunosuppression 5 years after transplantation. Absence of previous acute GVHD (aGVHD; hazard ratio [HR] = 2; P = .004) and mild cGVHD (HR = 4.2; P = .007) increased the probability of being off immunosuppressive treatment by the last follow-up. Overall survival (OS) at 5 years was 52%. Severe cGVHD, according to the NIH scoring system (HR = 13.27; P = .001) adversely influenced outcome, whereas de novo onset (HR = 0.094; P = .003) had a more favorable impact on survival. The combination of both variables allowed us to identify 4 different subgroups of patients with OS of 82%, 70%, 50%, and 25%. Our findings indicate that the NIH scoring system has some prognostic value in patients undergoing PBSCT and, together with the type of onset, must be considered to predict the possible outcome of patients who develop cGVHD.

The most common approach for the treatment of chronic graft-versus-host disease (cGVHD) has been the long-term use of systemic steroids. Beclomethasone dipropionate (BDP) is a topically active corticosteroid with relatively low absorption from the gastrointestinal mucosa. It has been successfully used to treat acute GVHD (aGVHD), but its use in the cGVHD setting is far more limited. In the current study, BDP was administered to 33 patients who underwent allogeneic transplantation and had biopsy-proven gastrointestinal cGVHD (GI cGVHD). Twenty-six patients with GI cGVHD received BDP as first-line and 7 as either second- or third-line treatment. All patients received BDP together with a calcineurin inhibitor, except for 1 patient who was also receiving mycophenolate mofetil (MMF). BDP was administered for a minimum of 16 weeks and was tapered during 4 additional weeks. Of those patients receiving BDP as the first line of treatment, 22 (84.6%) achieved complete remission (CR) of GI cGVHD, 2 (7.7%) achieved a partial response (PR) and 2 (7.7%) did not respond or progressed. Median time to response was 28 days. Nevertheless, only 7 (27%) patients had maintained the response at last follow-up, whereas 19 (73%) finally relapsed or progressed. Median time to relapse was 147 days after the end of BDP. In the case of the patients who received BDP as a second- or third-line treatment, 3 (42.9%) achieved CR and 2 (28.6%) PR. For the whole series of patients, 13 patients (39.4%) were not receiving immunosuppressive treatment at final follow-up. Only 4 patients developed cytomegalovirus (CMV) reactivation, which was successfully treated with antiviral drugs. No fungal infection was observed during the treatment period. In conclusion, the current study shows that BDP, in the absence of systemic steroids, is a highly effective initial therapeutic approach for GI cGVHD, which helps to avoid complications related to systemic steroids.

Background: Patient blood management (PBM) is a patient-centered, evidence-based approach for optimizing anemia management, minimizing blood loss, and ensuring appropriate transfusion. Artificial intelligence (AI) provides powerful tools for prediction, diagnosis, and decision support across PBM, but current evidence remains emerging and not yet consolidated. Objectives: This review synthesizes AI applications in PBM, summarizing predictive, diagnostic, and decision support models; highlighting methodological trends; and discussing challenges for clinical translation. Methods: PubMed, Scopus, and Web of Science were searched from inception to 31 March 2025. Eligible studies reported AI models addressing the three established PBM pillars. Studies on transfusion safety and blood bank operations relevant to PBM were also included. Extracted data covered study characteristics, predictors, models, validation strategies, and performance. The findings were narratively synthesized given study heterogeneity. Results: A total of 338 studies were included, spanning anemia detection, bleeding risk stratification, transfusion prediction, transfusion safety, and inventory management. Deep learning (DL) predominated in image-based anemia detection, while ensemble and gradient boosting methods frequently outperformed baselines in bleeding and transfusion risk prediction. Recurrent and hybrid architectures proved effective for blood supply forecasting. Across domains, machine learning and DL models generally surpassed logistic regression, clinical scores, and expert judgment. Despite strong internal performance, external validation and clinical deployment remain limited. Conclusions: AI is advancing PBM by enabling earlier anemia detection, more accurate bleeding and transfusion prediction, and smarter resource allocation. Translation into practice requires standardized reporting, robust external validation, explainability, and workflow integration. Future work should emphasize multimodal learning, prospective evaluation, and cost-effectiveness.

Pleural involvement in Non-Hodgkin Lymphoma (NHL) is well documented, but primary pleural lymphomas are extremely rare, occurring mostly in immunosuppressed patients or associated with chronic pleural inflammation. Nevertheless, the pathogenesis and therapeutic approaches to counteract primary pleural lymphomas are still matter of debate. The authors present the clinical case of an 81-year-old female with respiratory and constitutional symptoms. A valvular heart disease and bilateral pleural effusion were known. The study carried out showed a large right pleural effusion; the fluid analysis was compatible with Diffuse Large B-cell Lymphoma (DLBCL), and two lymphomatous masses with pleural origin were found at the ipsilateral hemithorax. Primary pleural lymphoma was considered and chemotherapy was initiated with a good response and evolution. The authors report this remarkable clinical case because of its rarity, its excellent clinical evolution and the absence of an immunodeficiency context.

Background/Objectives: Postoperative complications following adult cardiac surgery are biologically interrelated, yet most machine learning models predict single outcomes. We developed an explainable multi-output model integrating routinely collected clinical variables and patient blood management (PBM) biomarkers to predict multiple postoperative outcomes simultaneously, with complementary mono-output analyses for selected endpoints. Methods: This retrospective single-center cohort included 1414 adults undergoing cardiac surgery. In total, 513 complete cases were analyzed. Thirteen outcomes were modeled, including major binary complications and ICU/ward length of stay. An initial 80:20 train–test split was used only for algorithm screening across six candidate multi-output regressors and training-set-defined feature subsets. The selected regressor was then evaluated across five random states, and global permutation feature importance was used for multi-output explainability. Mono-output binary analyses using the selected regressor and the same training-set-only feature-selection workflow were evaluated along with accuracy, precision, recall/sensitivity, and F1-scores. Results: The Decision Tree Regressor was selected. Across five random states, global multi-output performance was R² = 0.83, MSE = 1.296, RMSE = 1.132, MAE = 0.298, and MAPE = 0.128. Based on global multi-output permutation importance, creatinine, ferritin, platelet count, estimated glomerular filtration rate, preoperative red blood cell units, and EuroSCORE II were ranked the highest. Atrial fibrillation had the lowest mono-output F1-score (0.719), whereas acute kidney injury, postoperative bleeding, infection, and 1-year hospital readmission yielded F1-scores of 0.928, 0.970, 0.963, and 0.975, respectively. Conclusions: This proof-of-concept study shows the feasibility of explainable multi-output modeling for postoperative outcomes after adult cardiac surgery using clinical and PBM variables. However, external validation is required prior to clinical use.

Click on the article title to read more.