Red blood distribution width and heart transplantation: any predictive role on patient outcome?

Red cell distribution width, eosinophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio can predict the development of infection and the number of antibiotics used in elderly patients undergoing revision total knee arthroplasty

Introduction: We aimed to evaluate the relationship between complete blood parameters reported to be associated with inflammation and development of complications, length of hospital stays and the number of antibiotics used in elderly patients undergoing revision total knee arthroplasty (rTKA). Material and Method: Our retrospective study was conducted in a single center and included 72 older patients who underwent rTKA operations. We recorded patients’ firs day preoperative, first day postoperative and 45th day postoperative whole blood parameters. Results: It was found that the development of postoperative infection and the number of antibiotics used were higher in patients with low preoperative Hb values and high platelet-to-lymphocyte ratio (PLR) rates. In patients with high RDW value and high eosinophil-lymphocyte ratio (ELR) one day after surgery, both the development of infection (P=0.002, P=0.002) and the number of antibiotics used during follow-up were found to be significantly higher ​(P

Red blood cell distribution width to predict mortality in heart transplant recipients: a systematic review

Red blood cell distribution width (RDW) has been shown to have prognostic value in a number of different clinical settings, such as cardiovascular disease, including heart failure. However, its prognostic value in heart transplant (HT) recipients remains unknown. The aim of this systematic review is to determine the prognostic value of pre-transplant RDW for mortality in HT recipients. There is a pre-published protocol of this review. The terms "Heart transplant", "Red cell distribution width" and their synonyms were used in the search strategy. PubMed/Medline, Embase, Scopus, Web of Science and LILACS were searched until May 17th, 2022, without date or language restrictions. Two authors independently carried out the selection, first by title and abstract, second by full-text revision. Discrepancies were discussed and resolved with three other authors. Quality of individual studies was assessed with Newcastle Ottawa Scale (NOS) for cohorts. After removing the duplicates, 3885 articles were identified. Four articles were included in the qualitative synthesis. Three studies were classified as “good quality”: whereas one as “poor quality” according to NOS scale. All the included articles evaluated long-term mortality and one study also evaluated short-term mortality. In this one, a correlation between higher RDW values and short-term mortality was reported. Meanwhile, in all the studies, a high pre-HT RDW was a marker of long-term mortality following cardiac transplantation. Our review shows that an elevated on-admission RDW is associated with long-term mortality in heart transplantation recipients.

Machine Learning Methods for Predicting Long-Term Mortality in Patients After Cardiac Surgery

Objective:

This study aims to construct and validate several machine learning (ML) algorithms to predict long-term mortality and identify risk factors in unselected patients post-cardiac surgery.

Methods

The Medical Information Mart for Intensive Care (MIMIC-III) database was used to perform a retrospective administrative database study. Candidate predictors consisted of the demographics, comorbidity, vital signs, laboratory test results, scoring systems, and treatment information on the first day of ICU admission. Four-year mortality was set as the study outcome. We used the ML methods of logistic regression (LR), artificial neural network (NNET), naïve bayes (NB), gradient boosting machine (GBM), adapting boosting (Ada), random forest (RF), bagged trees (BT), and eXtreme Gradient Boosting (XGB). The prognostic capacity and clinical utility of these ML models were compared using the area under the receiver operating characteristic curves (AUC), calibration curves, and decision curve analysis (DCA).

Results

Of 7,368 patients in MIMIC-III included in the final cohort, a total of 1,337 (18.15%) patients died during a 4-year follow-up. Among 65 variables extracted from the database, a total of 25 predictors were selected using recursive feature elimination and included in the subsequent analysis. The Ada model performed best among eight models in both discriminatory ability with the highest AUC of 0.801 and goodness of fit (visualized by calibration curve). Moreover, the DCA shows that the net benefit of the RF, Ada, and BT models surpassed that of other ML models for almost all threshold probability values. Additionally, through the Ada technique, we determined that red blood cell distribution width (RDW), blood urea nitrogen (BUN), SAPS II, anion gap (AG), age, urine output, chloride, creatinine, congestive heart failure, and SOFA were the Top 10 predictors in the feature importance rankings.

Conclusions

The Ada model performs best in predicting 4-year mortality after cardiac surgery among the eight ML models, which might have significant application in the development of early warning systems for patients following operations.

Donor-recipient age interaction and the impact on clinical results after heart transplantation

To evaluate the impact of donor-recipient age matching on clinical outcomes after heart transplantation, a total of 509 patients (January 1990-December 2018, mean follow-up 111 ± 80 months) were stratified into 4 groups (young-R/young-D, young-R/old-D, old-R/young-D, old-R/old-D) according to the recipient (young-R < 60, old-R ≥ 60 years) and the donor (young-D < 50, old-D ≥ 50 years) age. No difference was found among 30-day mortality (P = .11) and postoperative complications between groups. Both unadjusted and adjusted survival was significantly higher for group young-R/young-D than that of other groups, in which survival was similar [adjusted HR for mortality of 2.0(1.2-3.4), 2.1(1.4-3.8) and 2.5(1.6-4.1) for groups old-R/young-D, young-R/old-D, old-R/old-D, respectively]. Compared to other groups, the incidence of grade ≥ 2 CAV was significantly lower in old-R/young-D group [adjusted HR 0.4(0.2-0.7)]. Among young recipients, the rate of acute grade ≥ 2 rejection episodes was higher in those receiving an old donor graft (P = .04). Old recipient groups were more affected by neoplasms and severe renal failure than young recipient groups (P < .01). Employment of hearts from donors ≥50 years of age adversely affects survival in recipients <60 years of age but does not influence outcomes in older recipients. Also, donor and recipient ages seem to have opposite effects on incidence of rejections and CAV of high grade.