Bloodless kidney transplants proved feasible in matched cohort of Jehovah’s Witnesses

Surgeons at a single Spanish centre have shown that kidney transplantation can be carried out in Jehovah’s Witnesses without any blood transfusion at all, by pairing rigorous preoperative haemoglobin optimisation with a structured perioperative blood conservation protocol. In a retrospective propensity score-matched cohort study published in Frontiers in Nephrology, all 27 Jehovah’s Witness recipients who underwent kidney transplantation at Hospital del Mar in Barcelona between January 2016 and December 2025 completed the procedure without peri-transplant red blood cell transfusion, while nearly a third of matched non-Witness controls required transfusion support. The findings, drawn from one of the few truly transfusion-free transplant cohorts reported to date, offer a detailed look at how a religiously motivated surgical constraint can be met through systematic anaemia management rather than blood products.

Jehovah’s Witnesses decline allogeneic blood transfusions on religious grounds, a stance that creates a distinctive challenge in kidney transplantation, an operation in which perioperative and postoperative bleeding are recognised complications. In routine clinical practice, perioperative red blood cell transfusion remains common in kidney transplantation, with reported rates around 30 percent. Transfusion, although often necessary, carries recognised risks including HLA sensitisation, infection, immunologic effects and other transfusion-related adverse outcomes. Approximately 127,000 Jehovah’s Witnesses live in Spain, and while successful solid organ transplantation in Witness recipients has been reported for lungs, liver, heart, simultaneous kidney and pancreas, and kidneys, the literature remains dominated by case reports and small single-centre series, with only a limited number of comparative studies against standard transplant populations.

Since 2016, the Barcelona team has operated a transfusion-free protocol for Jehovah’s Witness patients undergoing kidney transplantation. The protocol centres on pre-transplant haemoglobin optimisation to at least 13 grams per decilitre, achieved with erythropoiesis-stimulating agents, often combined with intravenous iron, before surgery. It also includes review of antiplatelet and anticoagulant therapy, multidisciplinary planning and documented consent, meticulous intraoperative haemostasis with antifibrinolytics such as tranexamic acid and cell salvage when acceptable and clinically indicated, minimisation of phlebotomy using paediatric tubes for blood extractions, early postoperative monitoring of haemoglobin and iron status with escalation of ESAs and intravenous iron according to predefined triggers, and a bleeding contingency plan involving rapid imaging and surgical or radiological intervention for clinically significant bleeding or haematoma. Acceptability of blood products, fractions and procedures such as intraoperative cell salvage was documented individually for each patient, reflecting the fact that beliefs about blood components are not uniform among Witnesses.

The study included all Jehovah’s Witness recipients who underwent kidney transplantation under the protocol during the ten-year period, matched one-to-three with non-Witness controls from the institutional transplant database using propensity score matching on recipient age, sex, donor type and transplant date. Two planned transplantations were deferred on the day of surgery because predefined protocol criteria were not met: one patient had a pre-transplant haemoglobin of 12.5 grams per decilitre, below the 13 gram per decilitre target, and another was receiving clopidogrel. Both subsequently underwent transplantation once the relevant criteria were satisfied and were included in the cohort. Patients were censored at last follow-up on 31 March 2026, and the study was approved by the local ethics committee with written informed consent waived due to its retrospective, non-interventional design.

The primary outcome was the feasibility of transfusion avoidance, assessed by the occurrence of peri-transplant red blood cell transfusion, which the authors treated as a process indicator of protocol feasibility rather than a comparative clinical outcome. Peri-transplant transfusion was completely avoided in the 27 Jehovah’s Witness recipients, whereas 24 of 74 controls, or 32.4 percent, received peri-transplant transfusion, corresponding to an absolute risk difference of minus 32.4 percentage points with a 95 percent confidence interval of minus 43.1 to minus 21.8. The authors emphasise that this finding reflects successful implementation of the protocol in clinical practice and should not be read as a head-to-head comparison of surgical safety.

Haemoglobin data illuminate how the avoidance was achieved. Witness recipients presented with higher haemoglobin at admission than controls, 14.2 versus 11.5 grams per decilitre, a mean difference of 2.69 with a confidence interval of 2.15 to 3.22, because they had undergone systematic preoperative optimisation. This advantage persisted during hospitalisation, with higher nadir haemoglobin levels of 10.0 versus 8.5 grams per decilitre, a mean difference of 1.50. However, the absolute haemoglobin decrease from admission to nadir was actually greater in the Witness group, 4.14 versus 3.01 grams per decilitre, a mean difference of 1.13, indicating that the higher nadir occurred despite a larger perioperative haemoglobin decline and was largely a consequence of the higher starting level. No relevant differences were observed at discharge or at three months. Anaemia management was correspondingly more intensive: ESA therapy was used in 74.1 percent of Witness recipients versus 56.8 percent of controls, with more doses and a higher cumulative epoetin alfa dose, and intravenous ferric carboxymaltose was used in 59.3 percent versus 31.1 percent, although cumulative iron doses did not differ. Iron status at admission, including ferritin and transferrin saturation, was comparable between groups.

Bleeding-related surgical complications occurred in four of 27 Witness recipients, 14.8 percent, and eight of 74 controls, 10.8 percent, giving a relative risk of 1.65 with a confidence interval of 0.44 to 6.21, an estimate the authors describe as imprecise and compatible with both lower and higher risk. All four affected Witness recipients developed severe postoperative anaemia, with haemoglobin below 7 grams per decilitre and nadir values of 6.3, 5.6, 4.1 and 6.9 grams per decilitre within 24 to 48 hours after transplantation. Two required surgical reintervention, one for an abdominal wall haematoma and one with mesh placement for evisceration associated with a haematoma, while the other two developed retroperitoneal haematomas managed conservatively after imaging showed no active bleeding. All four received ESA and intravenous iron therapy, and their haemoglobin at discharge ranged from 6.3 to 10.1 grams per decilitre. These four patients had a mean hospital stay of 30.3 days, compared with 10.9 days among the remaining 23 Witness recipients, suggesting that severe postoperative anaemia and its associated complications contributed substantially to the longer mean hospitalisation observed in the Witness group overall, which was 13.7 versus 8.6 days, a mean difference of 5.2 days with a confidence interval spanning from minus 0.10 to 10.31.

Other early outcomes were broadly reassuring but statistically imprecise. Delayed graft function, defined as the need for dialysis within the first seven days, occurred in 11 of 27 Witness recipients, 40.7 percent, and 23 of 74 controls, 31.2 percent, a relative risk of 1.30 with a confidence interval of 0.63 to 2.69. Non-bleeding surgical complications were seen only in the control group, at 5.4 percent, with the relative risk not estimable due to zero events among Witness recipients. Renal function at three months was comparable, with serum creatinine of 1.68 versus 1.92 milligrams per decilitre. At three months, patient survival was 100 percent in both groups, a descriptive finding the authors explicitly state should not be interpreted as evidence of equivalent safety given the small sample and short follow-up. One graft loss occurred in the Witness group due to primary non-function, whereas no graft losses were observed among controls.

The authors are careful about the limits of their evidence. The retrospective single-centre design restricts causal inference and generalisability, and residual confounding cannot be excluded despite matching, particularly because clinically relevant imbalance persisted for dialysis vintage greater than 36 months, with a standardised mean difference of 0.73, and for pre-emptive transplantation. The propensity score model included only recipient age, sex, donor type and transplant date, and pre-transplant haemoglobin was deliberately excluded as a potential mediator of the transfusion-free strategy rather than a baseline confounder. Given the small sample size, low number of events and observational design, all analyses were considered exploratory and descriptive, and non-significant findings were not interpreted as evidence of equivalence, absence of clinically relevant differences, or safety. The study is underpowered to detect moderate differences in clinical outcomes, and confidence intervals remain wide for several key endpoints. The largest previous cohort, a Brazilian study of 143 Witness kidney transplant recipients and 142 matched controls, similarly demonstrated feasibility, although some of those patients received transfusions during surgical reinterventions, whereas all patients in the Barcelona series completed the transplant process without blood.

Beyond the clinical findings, the study engages with the ethics of organ allocation when patients decline blood products. The authors note that such refusal is widely accepted as an expression of patient autonomy, but that in transplantation, where organs are scarce and waiting lists substantial, it may introduce tension between autonomy and justice in allocation. Some centres have required transfusion rescue agreement before listing, an approach that assumes transfusion refusal may compromise outcomes, yet available evidence, including this study, has not demonstrated a clear short-term disadvantage, although clinically meaningful differences cannot be excluded. The authors argue that excluding such patients in the absence of evidence of inferior outcomes may raise concerns regarding equity, and that refusal of transfusion should not be equated with non-adherence to medical care, since unlike immunosuppressive non-adherence it does not inherently imply poor long-term graft outcomes. From a broader patient blood management perspective, they suggest that systematic haemoglobin optimisation, proactive anaemia management and careful perioperative planning may provide a useful framework for transfusion minimisation beyond the Witness population, while cautioning that the haemoglobin targets used to avoid transfusion may exceed standard chronic kidney disease recommendations and should be individualised in view of the potential risks of higher levels. Further multicentre studies with larger cohorts and longer follow-up are needed to define long-term outcomes, safety and optimal haemoglobin targets.

Subject of Research: Feasibility and early outcomes of transfusion-free kidney transplantation in Jehovah's Witnesses using a propensity score-matched cohort design

Article Title: Transfusion-free kidney transplantation in Jehovah’s witnesses: a propensity-score matched cohort study

Article References: Chamoun, B., Jiménez De Parga, A., Juez, A., Burballa, C., Buxeda, A., Cidraque, I., Redondo Pachón, D., Frances, A., Cecchini, L., Carazo, J., Pascual, J., Crespo, M., & Pérez Sáez, M. J. (2026). Transfusion-free kidney transplantation in Jehovah’s witnesses: a propensity-score matched cohort study. Frontiers in Nephrology, 6, Article 1938525. https://doi.org/10.3389/fneph.2026.1938525

Image Credits: AI Generated

DOI: 10.3389/fneph.2026.1938525

Keywords: kidney transplantation, Jehovah's Witnesses, bloodless surgery, transfusion avoidance, haemoglobin optimisation, erythropoiesis-stimulating agents, intravenous iron, propensity score matching, perioperative blood management, delayed graft function, bleeding complications, patient blood management