A translational pilot study was conducted, and a stepwise diagnostic accuracy strategy was applied to assess the predictive value of oxidative stress markers for DGF >7 days in adult patients with chronic end-stage kidney disease who underwent deceased-donor KTx. Following the Standards for Reporting of Diagnostic Accuracy Studies (STARD) Statement (https://www.equator-network.org/reporting-guidelines/stard/), we prospectively explored systemic levels of oxidative stress damage by thiobarbituric acid reactive substances (TBARS) and carbonyls; antioxidants by superoxide dismutase (SOD), catalase (CAT) activities and ferric reducing antioxidant power (FRAP); and pro-oxidants by hydrogen peroxide, nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, and nitrite, in both recipients and donors, as predictors of DGF >days. All patients underwent the operation at the Instituto de Medicina Integral Professor Fernando Figueira (IMIP) between June 2021 and June 2022. This study was reviewed by the Ethics Research Committee of the IMIP (CAAE 42858520.8.0000.5201, # 4.556.284) and complied with the current Brazilian ethical guidelines.

In terms of the scope, we limited our study to consecutive deceased-donor KTx performed in adult patients (≥18 years) with chronic end-stage kidney disease during the 1-year period of laboratory assessment that we defined at our convenience (1-year cohort study). Additional requirements for inclusion were transplantation being carried out at IMIP, with kidney graft retrieval performed at the Hospital da Restauração (RH), and voluntary informed consent provided by the recipients and donors’ families for participation in this study. The only exclusion criterion was missing data from patient charts. All kidney transplantations were performed by the same surgical team following standardized institutional protocols and routines.⁽¹³⁾ During the study period, the patients received the same institutional standards of immunosuppressive therapy, including induction with an anti-interleukin-2 receptor antibody, calcineurin inhibitor, mycophenolate, and steroids.

As part of our prospective study, we assessed and recorded clinical and paired laboratory data using electronic spreadsheets. Blood samples for oxidative stress marker analysis were collected at standardized time points to ensure procedural consistency. From the donors, peripheral blood was drawn immediately prior to the initiation of procurement surgery, and a sample of the preservation solution was obtained at the time of organ perfusion. In recipients, the first peripheral blood sample was collected upon hospital admission, before the administration of immunosuppressive therapy. A second sample was drawn from the renal vein immediately after graft reperfusion. All procedures were conducted under the direct supervision of the research team to ensure adherence to the protocol. Blood samples were processed using standard methods⁽¹⁴⁾ as detailed in the Table 1S, Supplementary Material. The endpoint of clinical interest was the severity of DGF, assessed as DGF >7 days, in patients requiring dialytic treatment beyond the first week after KTx.^(10–12)

The predictive value of the oxidative stress markers was assessed using a five-step hierarchical strategy. First, donor and recipient results were compared with those of healthy controls. Second, statistically significant parameters were correlated with the DGF duration using Spearman's rank test. Third, the overall accuracy was evaluated using the C-statistic, and optimal cut-offs were determined using Youden's J index. Fourth, significant continuous variables (i.e., area under the curve ≥0.7 and p<0.05) were dichotomized and reassessed using the χ² test. Finally, the diagnostic accuracy was determined by calculating the sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy.

Descriptive statistics are summarized as median (interquartile range) or frequency (percentage). Between-group comparisons were performed using the Mann-Whitney U test or χ² test, including Yates's correction or Fischer's exact test as appropriate. Analyses were performed using MedCalc v.19.4 (MedCalc Software, Ostend, Belgium), and correlation strength was classified according to ρ-values as very strong (0.91-1), strong (0.71-0.9), moderate (0.51-0.7), weak (0.31-0.5), very weak (0.01-0.3), and no correlation (zero), as previously reported.⁽¹⁵⁾ P-values (two-tailed) <0.05 were considered to indicate a statistically significant difference.

In this pilot study, we explored the value of eight oxidative stress markers in both recipient and donor blood samples to predict clinically relevant endpoints in patients who underwent deceased-donor kidney transplantation. Accordingly, we identified the systemic level of donor's hydrogen peroxide as a potential predictor of DGF >7 days by applying a comprehensive stepwise diagnostic approach. Herein, the donor systemic levels of this oxidative stress marker were found to be higher than those in healthy controls and correlated with the DGF duration after transplantation. Using c-statistics, we also determined the optimal cutoff for dichotomizing this marker as a positive/negative test. The cutoff was then statistically associated with DGF >7 days, achieving 85.7% sensitivity and a 91.7% negative predictive value.

Clinical factors such as the donor and recipient age, ischemia time, and revision surgery in association with human leukocyte antigen mismatching have been the most explored predictors for the occurrence of DGF in patients undergoing KTx.^(16–19) On the other hand, markers of oxidative stress have also been demonstrated a potential value in a few numbers of studies. For example, by exploring the role of malondialdehyde as an early predictive marker of graft dysfunction, Fonseca et al.⁽²⁾ found that the levels of this product of lipid peroxidation were significantly higher shortly after KTx and increased further in the subsequent days in patients developing DGF. In these settings, the authors reported that malondialdehyde levels accurately predicted DGF and demonstrated that high levels at one week were independently associated with poorer 1-year allograft function.⁽²⁾ Similarly, assessing the oxidative status by the OxyScore and AntioxyScore indexes, Rodriguez-Sanchez et al.⁽³⁾ recently demonstrated that donation after circulatory death induced greater short-term oxidative, whereas the early levels of oxidative damage were predictive of the graft function at 1 year among recipients of donation after brain death. Among the eight oxidative stress markers explored in recipient and donor blood samples, we added the donor's hydrogen peroxide level at the time of KTx as a potential predictor of DGF >7 days.

Because of a long cold ischemia period and other socioeconomic-related factors, the rates of DGF have remained high in Brazil, ranging from 29.9% to 87.7% among different transplantation centers.⁽⁹⁾ Accordingly, we observed a DFG rate of 77.8% in the current study despite a lower median DGF duration (4 days) in comparison with the American (10 days)⁽¹²⁾ and European (5 days) series.⁽¹¹⁾ Of note, the length of the DGF, rather than the DGF itself, is associated with long-term kidney graft function according to several studies.^(10–12,20) As confirmation of the duration-dependent effect of DGF on graft survival, the need for dialytic treatment beyond the first week after transplantation occurred in 33.3% of our cases and severe DGF lasting more than 14 days occurred in 25.9% of our cases, similar to 22.5%⁽¹²⁾ and 26%⁽¹¹⁾ reported in large previous studies. Since most of our cases (92.6%) achieved resolution of DGF within 4 weeks after KTx at a rate similar to those previously reported,^{(11, 12)} we chose DGF >7 days as a more clinically relevant endpoint in our study.

The main strengths of this study include its comprehensive accuracy analysis using five stepwise approaches and the use of paired prospectively collected data to assess oxidative stress markers in both donor and recipient blood samples. To our knowledge, this study is the first to investigate oxidative stress markers as predictors of DGF in deceased-donor KTx. Notably, the study was conducted during the COVID-19 pandemic, which required additional logistical efforts to perform complex surgical procedures such as kidney transplantation and may have contributed to longer cold ischemia times and higher rates of DGF in this study. Although the number of procedures may seem limited, this is inherent to the pilot study design, and reflects the rigorous methodological control applied to sample collection at each stage of the donation and transplantation processes.

Oxidative stress, one of the most important components of the ischemia-reperfusion process, has been implicated in relevant complications after KT, including DGF and allograft rejection.^{(2, 3)} However, effective treatment of oxidative stress in transplant patients requires further investigation.⁽²¹⁾ Additionally, identifying patients at high risk of allograft malfunction is important to improve KTx outcomes and graft allocation policies. Overall, our pilot study provides pivotal information that will aid researchers in exploring oxidative stress markers in future large clinical studies. These studies should focus on identifying patients who require early intervention and intensive postoperative surveillance to mitigate the effects of ischemic injury and DGF on long-term graft function.