Dengue infection in kidney transplant recipients during a major epidemic and the implementation of systematic donor screening: a cohort study

Summary

Background

During Brazil's unprecedented dengue epidemic in 2023–2025, we develop a rationale for organ donor screening to mitigate the risk of donor derived transmission using rapid, widely available, and low-cost tests. Here we aimed to describe the epidemiology, clinical presentation, and outcomes among kidney transplant recipients (KTRs) who developed different routes of dengue infection and establish a policy for mitigating the risk of donor transmission.

Methods

This study conducted at a large kidney transplant center in Brazil—Hospital do Rim (HRIM) examined dengue in KTRs through three components: initial cases of donor-derived infection that led to systematic donor screening; donor screening and adjudication for potential transmission from August 2023 to April 2025; and outcomes of vector-borne dengue among recipients followed at HRIM from December 2023 to July 2024. Donors were screened using immunochromatographic NS1 antigen and IgM rapid tests, with RT-PCR for those NS1-negative and IgM-positive.

Findings

Before August 2023, we recorded four donor-derived transmissions from two deceased donors, resulting in three deaths. After that, among 1252 deceased donors screened, eight were NS1-positive and four had clinical evidence of dengue; these 12 donors were excluded from donation. Among 58 NS1-negative/IgM-positive donors, 38 were considered for donation. From these, 52 recipients were transplanted, with no confirmed donor-derived transmission. Among 1542 recipients of kidneys from NS1- and IgM-negative donors, 17 developed dengue within 30 days after transplantation: one had fatal transfusion-transmitted dengue, two had confirmed donor-derived transmission, ten had potential donor-derived transmission, and four were considered to have vector-borne infection acquired before transplantation. Every recipient with confirmed or potential donor-derived transmission identified during the screened period recovered from the infection. Lastly, among 12,402 KTRs on follow-up, we recorded 373 confirmed vector-borne cases. They were 50·0 years old, 203 were male (54·4%), 170 were female (45·6%), and 232 were White (62·2%). The case-fatality rate was 1·88%, compared with 0·10% in the general population during the same period.

Interpretation

KTRs with vector-borne dengue experienced a high observed case-fatality rate. Organ or blood transmission led to life-threatening outcomes. The rapid test screening was associated with low transmission, and when it occurred, the risk of severe cases was mitigated. These findings support the incorporation of dengue screening strategies in endemic regions.

Funding

Fundação Oswaldo Ramos.

Research in context.

Evidence before this study

We searched PubMed for studies published between Jan 1, 2000, and May 31, 2026, without language restrictions, combining the terms “dengue”, “kidney transplantation”, “solid organ transplantation”, “donor-derived infection”, “organ donor”, “blood transfusion”, “screening”, “NS1”, and “transmission”. We also reviewed the reference lists of retrieved articles and relevant systematic reviews. We included observational studies, case series, case reports, and systematic reviews reporting dengue in solid-organ transplant recipients, donor-derived dengue, transfusion-transmitted dengue, or donor-screening strategies. Available evidence showed that dengue in kidney transplant recipients is associated with more severe clinical manifestations than in the general population. However, published data were limited to small observational studies and case reports. Reports of donor-derived dengue have been documented in both endemic and non-endemic regions, demonstrating the biological plausibility of transmission via organ transplantation. However, evidence regarding systematic donor-screening strategies, their effectiveness, and their impact on organ utilization remains scarce. No comprehensive study has evaluated the implementation of a pragmatic screening strategy during a major epidemic while simultaneously assessed outcomes of vector-borne and donor-derived dengue in kidney transplant recipients.

Added value of this study

This study provides integrated epidemiological, clinical, and translational evidence from one of the largest kidney transplant programs worldwide during the largest recorded dengue epidemic. We observed a substantially higher case-fatality rate among kidney transplant recipients with dengue than that reported in the general population during the same epidemic period. We also showed that donor-derived dengue, although uncommon, may cause fatal disease. Following these observations, we implemented a systematic donor-screening strategy based on low-cost and widely available rapid NS1 and IgM testing, supplemented by targeted RT-PCR. This strategy identified infected donors, was consistent with mitigation of transmission risk, preserved access to transplantation by allowing the use of carefully selected IgM-positive donors, and was associated with fewer severe donor-derived events.

Implications of all the available evidence

The available evidence indicates that dengue represents a relevant threat to kidney transplant recipients. Although donor-derived transmission is infrequent, its consequences can be disastrous, supporting the need for donor-screening strategy in endemic regions. Our findings suggest that a structured screening measure combining rapid antigen and serological testing may substantially reduce transmission risk while preserving organ availability. Given the rapid expansion into new regions as a consequence of climate change, urbanization, and increased mobility, these strategies are relevant worldwide. Transplant programs, based on their local dengue incidence, should consider integrating dengue surveillance and donor-screening strategies in order to effectively mitigate emerging risks.

Introduction

According to the World Health Organization, dengue is among the most common neglected tropical diseases, with more than 100 million cases reported worldwide each year.^1,2 The disease is an arbovirus transmitted mainly by Aedes mosquitoes and caused by four DENV serotypes.3, 4, 5 Between 2000 and 2019, the highest incidence rates were observed in South Asia, exceeding 2000 cases per 100,000 population, with additional endemic areas in Southeast Asia and South and Central Americas.^6,7 In 2024, among the 210 million Brazilian population, it was reported that more than 6·5 million persons were infected, reaching a historic peak, with more than 6000 deaths.⁸ In recent years, dengue has also emerged in non-tropical regions, including Europe and the United States, likely reflecting the combined effects of climate change and increased human mobility.9, 10, 11

Non-vector transmission of dengue, including blood transfusion¹² and through solid-organ transplantation (SOT), has been reported in endemic countries such as India, Singapore, and Colombia,13, 14, 15 as well as in non-endemic settings such as France.¹⁶ Reports of transmission through SOT have supported the implementation of systematic donor screening, particularly during periods of high incidence. However, those transmission routes present additional challenges: most infected individuals are asymptomatic,¹⁷ interpreting diagnostic tests in donors can be difficult,¹⁸ donation can occur when viremia is below the limit of detection,¹⁵ and transmission may still take place during the immunological window period, when viral markers are no longer detectable by antigen or nucleic acid testing and serological markers have not yet become positive.¹⁹ Furthermore, the clinical course of donor-derived dengue varies from completely asymptomatic to an unpredictable course, and the higher inoculum transmitted through an infected organ, as well as blood transfusion, could contribute to more severe clinical presentations.

Immunosuppression during dengue infection among kidney transplant recipients (KTRs) appears to pose additional challenges. A systematic review, including data from 168 KTRs across 11 studies and comparing them with more than 81,000 non-transplant individuals, showed that severe cases were more frequent among KTRs.²⁰ Furthermore, the real benefit of recently developed attenuated viral vaccines remains poorly established in at-risk populations and must be avoided in immunosuppressed patients.²¹

In this scenario, we hypothesized that dengue would be associated with substantial morbidity among KTRs and that systematic donor screening could mitigate the risk of donor-derived transmission without unnecessarily compromising organ utilization. Therefore, the present study aims to describe the epidemiology, clinical presentation, and outcomes of vector-borne, donor-derived, or other routes of infection in KTRs during the unprecedented Brazilian epidemic, and to establish a policy for identifying potentially infected organ donor and mitigating the risk of transmission through transplantation.

Methods

Study design and setting

The study was conducted at Hospital do Rim in São Paulo—HRIM, Brazil. This tertiary center performs more than 1000 kidney transplants annually and has completed more than 22,000 procedures in the past 27 years.

This is a retrospective, real-world observational study, comprising three components: (1) the first episodes of deceased donor-derived dengue in KTRs, occurring in March 2021 and July 2023, which prompted the pragmatic implementation of systematic dengue screening for all living and deceased donors evaluated for transplantation; (2) the screening of donors and adjudication of potential donor-derived transmission from August 2023 to April 2025; and (3) the frequency, clinical characteristics, and outcomes of vector-borne dengue among KTRs under clinical follow-up at HRIM from December 2023 to July 2024. During the latter period, DENV-1 and DENV-2 were the predominant circulating serotypes in the State of São Paulo, according to epidemiological surveillance data. The sanitary authorities recognized the formal epidemic in February 2024.

For the evaluation of donor-derived dengue, we retrospectively reviewed the outcomes of the first episodes of confirmed donor-derived transmission. Following the pragmatic implementation of systematic screening in August 2023, we adjudicated cases of dengue diagnosed within 30 days after transplantation.

For the evaluation of vector-borne dengue occurring >30 days after transplantation, the study population comprised 12,402 KTRs in active clinical follow-up at the HRIM outpatient clinic in July 2024. Dengue cases were identified through scheduled or unscheduled outpatient visits, emergency department visits, or telehealth contacts. Patients diagnosed or managed at other healthcare facilities remained under telehealth follow-up by the HRIM team, and external clinical information, including laboratory results, hospitalizations, and outcomes, was incorporated into the institutional electronic medical record. Further details on patient follow-up and case ascertainment are provided in the Supplementary Methods. We compared the case-fatality rate with that observed in the general population of the same geographical region.

The manuscript was prepared in accordance with the STROBE statement.

Participants

For the first component of the study, we included four adult KTRs with donor-derived dengue diagnosed between March 2021 and July 2023, comprising two recipients from each of two donors.

For the second component, the donors’ screening, we included 1252 consecutive deceased donors provided by state organ procurement organizations (OPOs) from August 2023 to April 2025. We also collected data for 1542 kidney transplants with deceased donors and 363 with living donors performed at HRIM during the same period to identify potential donor-derived transmission despite negative donor screening and review the outcomes of dengue cases diagnosed within 30 days after the procedure.

In the third and last component, we included adult (≥18 years) kidney transplant outpatients on follow-up who were more than 30 days post-transplant and with suspected or confirmed dengue between December 2023 and July 2024.

Deceased and living donor screening

Every consecutive potential deceased donor was screened for dengue using an immunochromatographic rapid test (OnSite™ Duo Dengue Ag-IgG/IgM Rapid Test; CTK Biotech, Inc., San Diego, CA) from August 2023. Starting in March 2024, serum RT-PCR was added for donors with NS1-negative/IgM-positive results. RT-PCR was performed with the STANDARD™ M10 Arbovirus Panel (SB Biosensor, Inc., South Korea).

For deceased donors, we collected samples at the time of organ procurement and tested them before kidney allocation. Although the exact interval between sample collection and testing was not systematically recorded, based on the usual operational workflow, testing typically occurred approximately 4–12 h after organ removal. Donor acceptance or exclusion followed a prespecified protocol: NS1-positive donors were excluded for donation, and NS1-negative with IgM-positive donors were accepted if they had no clinical or laboratory findings suggestive of dengue and, after RT-PCR was incorporated into the algorithm, had negative RT-PCR results. In cases of retrospective adjudication of potential transmission, stored donor samples were additionally retested for RT-PCR in cases that this test had not been previously performed and, in three selected cases for IgM using a MAC-ELISA (IgM capture-ELISA) developed by the US CDC at a national reference laboratory. The dengue IgG results were not considered criteria for donor acceptance or exclusion decisions.

Living-donor candidates and corresponding recipients were screened on the day before transplantation using the same diagnostic strategy. In the case of any positive NS1 test, the transplant would be postponed.

Case definitions

Early post-transplant dengue was defined as occurring within 30 days after transplantation. Paired recipients identified the two kidney recipients sharing the same deceased donor. Donor-derived dengue was adjudicated based on timing, paired-recipient findings, donor and recipient virological results, and exclusion of alternative transmission pathways. Transmission was classified as confirmed when paired recipients developed dengue with pre-transplant negative test results and the donor had virological evidence of infection, and as potential when paired recipients developed dengue but the donor's infection was not confirmed. Virological evidence was defined as NS1 antigen positivity or RT-PCR positivity, or in epidemic scenario, IgM positivity with a clinically compatible syndrome, as recommended by the Brazilian Ministry of Health. Details on the classification of vector-borne infection and donor-derived transmission are provided in the Supplementary Material.

Predefined case definitions, systematic donor screening, structured adjudication of transmission routes, and institutional follow-up or telehealth were used to minimize misclassification and ascertainment bias. Institutional follow-up and telehealth use are detailed in the Supplementary Material. Residual selection, ascertainment, and diagnostic misclassification biases are acknowledged in the Discussion.

Outcomes for kidney transplant recipients

For recipients of screened donors, the primary outcome was the donor-derived dengue diagnosed in the first 30 days after transplantation.

For KTRs with vector-borne dengue, the main outcomes were hospitalization and death during the dengue episode. Follow-up was defined from dengue diagnosis until death, clinical recovery, or hospital discharge, whichever occurred first. Recovery was defined as being afebrile, with resolution of symptoms, and no need for dengue-related medical care within 14 days. The case-fatality rate among KTRs was compared with that observed in the general population of the State of São Paulo during the same period.

Study size

The study used a consecutive, all-eligible cohort within each component's time window, reflecting epidemic dynamics. Therefore, a prospective sample-size calculation was not applicable. All donors and recipients meeting the inclusion criteria were included.

Statistics analysis

Distribution of continuous variables was assessed with the Kolmogorov–Smirnov test. Because these variables were non-normally distributed, they are presented as median and interquartile range (IQR, 1st–3rd). Categorical variables are reported as counts and percentages. Dengue incidence rates for donors screened were expressed per 1000 (denominator: donors screened). Among KTRs with vector-borne infection, the differences in proportion of case-fatality rates in KTRs and in the general population were estimated using Chi-square test, and the risk ratio with 95% CI were calculated. Missing data were minimal; analyses were performed using available data without imputation, and the number of missing observations is reported where applicable.

Demographic characteristics, donor type, diabetes mellitus, hypertension, acetylsalicylic acid use, immunosuppressive regimen, time from transplantation to dengue diagnosis, dengue symptoms, and laboratory findings were considered as covariates in a multivariable logistic regression model for hospitalization. Variables in the final model were selected using backward stepwise selection, and results were expressed as odds ratios (ORs) with 95% confidence intervals (95% CIs).

Analyses were performed with IBM SPSS Statistics, version 29·0 (IBM Corp., Armonk, NY).

Ethical statement

The study was approved by the Ethics Committee of the Federal University of São Paulo (CAAE 81926924·0·0000·5505; approval number 7·994·210). Informed consent or a consent waiver was obtained in accordance with the committee's determinations and Brazilian regulations.

Role of funding source

This study was funded by Fundação Oswaldo Ramos. The funder had no role in study design, data collection, data analysis, interpretation, or writing of the report.

Results

First cases of confirmed donor-derived infection

We recorded two confirmed donor-derived dengue transmission episodes involving four KTRs that occurred before the screening implementation: two by DENV-1 subtype infections in March 2021 and two by DENV-2 subtype infections in July 2023, among 1990 deceased-donor transplants performed during that period.

In the first episode, both recipients developed dengue early after transplantation. One died on postoperative day (POD) 16 due to fulminant hepatic failure, whereas the paired recipient survived with recovering graft function and was discharged on POD 32.

In the second episode, one recipient developed fever on POD 6, and the other presented with hematemesis on POD 8. Both progressed to severe dengue; one died due to hemorrhagic shock on POD 9, and the other died from hemophagocytic lymphohistiocytosis on POD 17.

Stored pre-transplant serum samples from donors and recipients were retrospectively tested. All four recipients had negative NS1, IgM, and RT-PCR results before transplantation. Infection was retrospectively confirmed in both donors by positive RT-PCR and NS1, and the same viral serotypes were identified in recipients from each donor pair. Adjudication of these cases, together with rising community dengue transmission, prompted implementation of the systematic deceased-donor screening strategy described in the next section.

Systematic deceased-donor screening

From August 2023 to April 2025, 1252 deceased donors were screened. Donor demographic characteristics are detailed in Table S1, the geographic distribution of organ procurement is shown in Figure S1, and the monthly number of screened donors is presented in Figure S2.

Eight donors (0·6%) were NS1-positive and were excluded for donation. Fifty-eight were IgM-positive and considered for donation, but four were excluded due to recorded laboratory and clinical findings compatible with or resembling dengue. The timing of NS1- and IgM-positive results is presented in Figure S2. Other 16 were excluded due to non-dengue-related clinical limitations. Of the remaining 38 donors considered eligible for donation and found to be IgM-positive (30 with negative RT-PCR and eight evaluated before RT-PCR was incorporated into the screening), 76 kidneys were allocated by the state organization to recipients in the region (Figure S3).

From these donors, we followed outcomes in 52 recipients transplanted at HRIM. Only one recipient had symptoms related to dengue, which were confirmed by positive NS1 and RT-PCR on POD 5, with negative tests on the transplant day, including NS1, IgM, IgG, and RT-PCR (Fig. 1). NS1 and RT-PCR performed on the asymptomatic paired recipient from the same donor were negative. However, IgG and IgM were positive, possibly related to prior immunity acquired from a recent vector-borne infection. This event was considered the only potential donor-derived transmission from donors with isolated IgM-positive, leading to a non-severe disease that recovered within 2 weeks.

Fig. 1.

Outcomes of kidney transplant recipients of NS1 negative and IgM positive screening. Note: 58 donors were IgM-positive: four were excluded due to laboratory and clinical findings compatible with dengue, 16 were excluded due to non-dengue-related clinical limitations, and 38 were considered eligible for donation (see Supplementary Material). The state organization allocated 76 kidneys to recipients in the region, 52 being transplanted at HRIM.

Considering that eight NS1-positive donors were discarded, four IgM-positive donors were also discarded due to a clinical suspicion of dengue, and one of the IgM-positive donors was associated with a potential transmission event, a total of 13 donors were classified as having dengue (1·04%) at procurement.

Cases diagnosed early after transplantation in recipients from negative screened donors

During the same period, a total of 17 recipients from 1542 deceased-donor transplants performed at our center developed dengue within 30 days after kidney transplantation, despite testing negative at the transplant day and receiving kidneys from donors who also tested negative for NS1 antigen and IgM at the time of procurement. These cases were retrospectively rechecked for potential donor-derived transmission. The detailed findings are summarized in Table 1.

Table 1.

Panel with kidney transplant recipients with dengue <30 days after transplant from NS1 and IgM negative deceased donors.

Definition	Donor	Recipient 1	Recipient 2
Transfusion-transmitted dengue
	RT-PCR negative	NS1 and RT-PCR positives	No denguea
Confirmed donor-derived transmission
	RT-PCR positive	NS1, IgM, and RT-PCR Positives	NS1 and IgM Positives
Potential donor-derived transmission
	RT-PCR negative IgM (EILSA) positive	NS1 positive	NS1 positive
	RT-PCR negative	NS1 and IgM positives	NS1 positive
	RT-PCR negative IgM (ELISA) inconclusive	IgM (ELISA) positive	Denguea
	RT-PCR negative	NS1 and IgM positives	Denguea
	RT-PCR negative IgM (ELISA) positive	NS1 and PCR positives	IgM positive
	RT-PCR negative	RT-PCR negative IgM positive	IgM positive
Transplantation during the asymptomatic phase of vector-borne dengue
	RT-PCR negative	IgM positive (ELISA)	RT-PCR, NS1, and IgM Negative
	RT-PCR negative	NS1 and RT-PCR positives	RT-PCR, NS1, and IgM negatives
	RT-PCR negative	IgM positive	RT-PCR, NS1, and IgM negatives
	RT-PCR negative	NS1 and PCR positives	RT-PCR negative IgM positive

^aCases followed in other centers.

Dengue was confirmed on POD 12 by positive NS1 antigen and RT-PCR in a single recipient who required transfusion of blood products six days before. Investigation of the transfused blood products identified DENV-1 RNA in two samples, confirming transfusion-transmitted dengue. The patient subsequently developed fulminant hepatitis and died two weeks after transplantation.

Confirmed donor-derived transmission was identified from a single donor, involving two recipients. Although the donor was negative for NS1 and IgM, retrospective RT-PCR testing of stored serum was positive. Both recipients developed dengue during the hospitalization for transplantation and recovered from the infection with preserved graft function.

Potential donor-derived transmission involved 12 recipients from six donors, including 10 recipients transplanted at our center and two paired kidney recipients transplanted at other centers. The latter were included only for adjudication of the transmission route and were not part of our center's cohort. In these cases, both recipients from the same donor developed dengue during the index hospitalization, while initial donor screening (NS1 and IgM) was negative, and retrospective RT-PCR testing of stored donor samples was also negative. Furthermore, retrospective donor testing with IgM ELISA was not supportive as it was positive in two donors, negative in three, and inconclusive in one case. Among the 10 recipients transplanted at our center, all recovered from the infection and maintained adequate graft function.

In the remaining four recipients, dengue occurred while the paired recipient from the same donor remained uninfected. Given the absence of virological evidence in donors and lack of concordance between paired recipients, these episodes were interpreted as transplantation occurring during the recipient asymptomatic phase of vector-borne infection. Affected recipients recovered from the infection with preserved graft function.

Systematic living-donor screening

After implementation of the screening strategy, 363 asymptomatic living-donor candidates were screened. None were NS1-positive, and nine were IgM-positive. Among these, seven also had positive IgG serology, suggesting previous exposure, and donation proceeded. Of the remaining two IgG-negative donors, one also had negative RT-PCR results and proceeded with donation, whereas donation was postponed in the other due to an unrelated clinical condition. No living donor-derived transmission occurred among these 363 transplants.

Vector-borne dengue during the epidemic

During the epidemic period, 536 suspected vector-borne cases were identified among KTRs under follow-up at HRIM. In 163 non-tested patients, the diagnosis was based on typical clinical findings without additional tests. Among the 373 confirmed cases, 295 (79·1%) were diagnosed using the NS1 rapid test, while the remainder were confirmed by IgM-positive serology. The median interval from symptom onset to diagnostic testing was 4 (2–8) days. Demographic characteristics and clinical features for confirmed and suspected cases are presented in Table 2. The population disposition is shown in Figure S4.

Table 2.

Demographic characteristics, symptoms, and laboratorial finds in kidney transplant recipients with vector-borne Dengue.

Variable	Confirmed cases N = 373	Suspected cases N = 163
Age, years	50 (42–59)	51 (39–58)
Male, n (%)	203 (54·4)	80 (49·1)
Ethnicity, n (%)
Withe	232 (62·2)	102 (62·6)
Afro-Brazilian	136 (36·5)	60 (36·9)
Asiatic	5 (1·3)	1 (0·5)
Deceased donor, n (%)	270 (72·4)	124 (76·1)
Diabetes mellitus, n (%)	114 (30·6)	52 (31·9)
Hypertension, n (%)	318 (85·3)	136 (83·4)
ASA, n (%)	95 (25·5)	46 (28·2)
Immunosuppression, n (%)
rATG as induction	318 (85·3)	135 (82·8)
Maintenance
CNI/MPA	190 (50·9)	80 (49·1)
CNI/AZA	120 (32·2)	48 (29·4)
CNI/mTORi	28 (7·5)	17 (10·4)
Other	35 (9·4)	18 (11·0)
Time after transplant, months	60 (16–132)	84 (36–148)
Dengue Symptoms, n (%)∗
Myalgia	275 (73·7)	126 (77·3)
Fever	219 (58·7)	84 (51·5)
Headache	138 (37·0)	76 (46·6)
Diarrhea or vomiting	146 (39·1)	51 (31·3)
Arthralgia	120 (32·2)	61 (37·4)
Cutaneous rash	11 (2·9)	6 (3·7)
Laboratory findings, n (%)∗∗
Leukopenia	139 (37·4)	40 (24·5)
Thrombocytopenia	221 (59·4)	68 (41·7)
High transaminases	53 (14·2)	12 (7·4)
Acute allograft dysfunction	163 (43·7)	39 (23·9)
Outcomes, N (%)
Hospitalization	106 (28·4)	31 (19·0)
ICU	26 (7·0)	10 (6·1)
RRT	13 (3·5)	6 (3·7)
Death	7 (1·88)	4 (2·45)

ASA, acetylsalicylic acid; AZA, azathioprine; CNI, calcineurin inhibitors; ICU, intensive care unit; MPA, mycophenolate acid; mTORi, mammalian target of rapamycin inhibitor; rATG, rabbit anti-thymocyte globulin; RRT, renal replacement therapy.

Note: missing: ∗n = 2, ∗∗n = 1.

The monthly distribution of the 536 suspected or confirmed vector-borne dengue cases showed a progressive increase from December 2023 (n = 5) and January 2024 (n = 8), before the epidemic was formally recognized in February 2024 (n = 31), reaching a peak in April 2024 (n = 190), followed by a decline through July 2024 (Figure S5).

Among confirmed cases, the median age was 50·0 years, 54·4% (n = 203) were male, 45·6% were female (n = 170), and 62·2% (n = 232) were White. At the time of diagnosis, the most common maintenance immunosuppressive regimens were a calcineurin inhibitor and prednisone combined with either mycophenolate (50·9%, n = 190) or azathioprine (32·2%, n = 120). Hypertension was present in 85·3% of patients, diabetes in 30·6%, and 25·5% were receiving acetylsalicylic acid. The median time from transplantation to dengue diagnosis was 60 months.

The most frequent symptoms were intense myalgia (73·9%, n = 275), fever (58·7%, n = 219), headache (36·8%, n = 138), diarrhea or vomiting (39·2%, n = 146), and arthralgia (32·2%, n = 120). Cutaneous rash was observed in only 3·0% (n = 11) of cases. Leukopenia occurred in 37·4% (n = 139) of patients, thrombocytopenia in 59·2% (n = 221), elevated transaminase levels in 14·2% (n = 53), and acute allograft dysfunction in 43·8% (n = 163).

Hospitalization was required in 28·4% (n = 106), 7·0% (n = 26) were admitted to the intensive care unit, and 3·5% (n = 13) required renal replacement therapy. Bleeding occurred in 17 patients (4·6%). Seven patients died, corresponding to a case-fatality rate of 1·88%. During the same period, the case-fatality rate among confirmed dengue cases in the general population was 0·1%. The unadjusted comparison resulted in a RR of 18·2 (95% CI 8·8–37·2; p < 0·001).

Of note, dengue symptoms, laboratory findings, and outcomes were similar among patients in the confirmed group and those with suspected dengue, in line with the criteria recommended by Brazilian health authorities.

Among confirmed dengue cases (n = 373, 106 [28·4%] events of hospitalization), shorter time since transplantation, diabetes mellitus, and thrombocytopenia were associated with higher odds of hospitalization (Table 3). Each additional year since transplantation was associated with 4% lower odds of hospitalization (OR 0·96; 95% CI 0·92–0·99), whereas thrombocytopenia was associated with 2·43-fold higher odds (OR 2·43; 95% CI 1·46–4·04) and diabetes mellitus with 65% higher odds (OR 1·65; 95% CI 1·01–2·69).

Table 3.

Multivariable logistic regression for hospitalization in vector-borne dengue among kidney transplant recipients.

Variables	OR	95% CI	p
Fever (yes vs no)	0·65	0·41–1·04	0·07
Diabetes (yes vs no)	1·65	1·01–2·69	0·04
Time since transplant (each year)	0·96	0·92–0·99	0·04
Thrombocytopenia (yes vs no)	2·43	1·46–4·04	<0·001

CI, confident interval; OR, odds ratio.

Discussion

This study integrates epidemiological, clinical, and translational evidence to characterize dengue in KTRs during one of the largest epidemics ever recorded and to define a pragmatic framework for mitigating donor-derived transmission in endemic settings. Four key findings emerge. First, KTRs with dengue experienced substantial clinical morbidity and a higher observed case-fatality rate than the general population. Second, donor-derived transmission, although uncommon, can lead to severe and fatal disease. Third, a structured screening strategy for deceased donors reduces transmission risk while maintaining access to transplantation. Finally, although potential donor-derived transmission was observed from donors with negative screening results, infections in recipients were not associated with severe clinical course.

For vector-borne dengue in our cohort, as expected, fever and myalgia were the most frequent symptoms.²² However, the clinical course was often complicated by acute allograft dysfunction and the need for renal replacement therapy (Table 2). Volume depletion from frequent gastrointestinal manifestations may have contributed to allograft dysfunction, although other mechanisms of kidney injury are also likely. On the other hand, the frequent need for hospitalization and ICU admission further underscores the high severity associated with dengue in KTRs, beyond its impact on mortality. This pattern points to a particular vulnerability to systemic inflammatory and hemodynamic disturbances. Similar patterns have been reported in other non-opportunistic infections, both bacterial and viral, including COVID-19, where transplant recipients consistently experienced worse outcomes.²³ In a systematic review, KTRs with dengue showed higher rates of pleural effusion (17% vs 2%), ascites (35% vs 1%), and severe Dengue (16% vs 4%) than non-transplant individuals.²⁰ Taken together, these observations indicate that immunosuppression ultimately interacts with the burden of comorbidities to exacerbate disease severity. Our findings extend this concept by grounding it in data from a large cohort exposed to intense viral transmission.

Herein, we also provide insight into non-vector transmission pathways. Although donor-derived dengue was uncommon before the screening implementation, it was associated with higher fatal outcomes compared with vector-borne transmission. This finding supports the biological plausibility that the magnitude of viral inoculum, whether delivered through an organ graft or blood products, may influence disease severity. In addition, in the early period after transplantation, more than overall intensity of immunosuppression, its qualitative features may also play a role. In this setting, lymphocyte depletion induced by agents such as rATG could further impair the host response and contribute to a more severe clinical course.²⁴

A key contribution of this study is the implementation of a structured screening strategy for deceased donors during an epidemic, using simple, widely accessible tools.¹⁸ In addition to providing rapid results in less than 30 min, this strategy has a low direct testing cost of less than five American dollars per donor in our country, compared with 81 for RT-PCR. Combining NS1 antigen testing with serology and utilizing targeted RT-PCR in only a few cases enabled the identification and exclusion of donors with active infection. Concurrently, this approach allowed the continued use of organs from donors with isolated IgM positivity and no clinical evidence of dengue.

Donor-derived dengue has been previously reported after solid organ transplantation in India, Singapore, Colombia, and France, including severe and fatal cases. These reports, however, have largely been restricted to isolated cases or small donor-recipient clusters (Table S2). In contrast, our study describes the programmatic response to donor-derived transmission through systematic screening of 1252 consecutive deceased donors and structured assessment of early post-transplant dengue.13, 14, 15, 16 Although a small number of transmission events occurred despite screening, this pattern suggests that while the protocol is effective, it may not entirely eliminate the risk of transmission. Crucially, none of these transmissions resulted in death. This outcome suggests that screening reduced recipient exposure to a higher viral inoculum, thereby shifting disease expression towards a milder phenotype resembling community-acquired infection outcomes.¹⁷ In contrast, the single transfusion-transmitted infection in our cohort had a fatal outcome. Although transfusion-transmitted dengue appears to be rare, with few cases reported worldwide, its potential severity deserves attention. During the same Brazilian epidemic, six suspected cases among pediatric cardiac patients receiving multiple transfusions were reported, including four deaths.²⁵

The occurrence of early post-transplant dengue despite negative donor screening poses challenges for attributing transmission routes, as community or nosocomial vector exposure and transfusion represent alternatives to donor-derived transmission. In addition, inherent limitations of testing strategies, particularly the eclipse phase and the diagnostic window period, may preclude identification of donor infection.^19,26 Even when antigen and serological assays are combined, very early infections may remain undetected at the time of procurement. This challenge is amplified by the high frequency of asymptomatic infection, almost 50% in the population setting,¹⁷ and the potential for transmission during the immunological window, when antigen tests are negative and serological markers are not yet detectable. Although universal nucleic-acid testing could mitigate this risk,^27,28 transmission may still occur during the eclipse phase, when viremia falls below the threshold of detection, as described for HIV and hepatitis B and C.¹⁹

A structural limitation of this strategy is that NS1-negative and IgM-negative donors did not undergo RT-PCR, leaving a diagnostic window during early infection in which both markers may be negative. During the same epidemic period, a large study based on asymptomatic Brazilian blood donations demonstrated detectable arboviral RNA in donors without clinical suspicion.²⁹ Although asymptomatic NAT positivity does not necessarily imply NS1 negativity, these findings reinforce the potential complementary value of molecular testing. Universal NAT, however, must be balanced against cost and turnaround-time constraints inherent to deceased-donor allocation, particularly in low- and middle-income endemic settings.

Fig. 2 outlines a conceptual model of the temporal dynamics of infection, extrapolated from HIV and B and C hepatitis models, and the performance of available diagnostic assays in organ transplantation. Persistent detection of DENV RNA in urine after clearance from blood has been reported, suggesting that urine RT-PCR could extend the diagnostic window and potentially provide an additional signal when blood-based tests are negative.^15,30,31 Although the feasibility and incremental value of systematic urine testing during time-sensitive organ procurement remain uncertain, this approach may represent a useful adjunct to donor screening and warrants further investigation. By contrast, infections occurring during the eclipse phase would not be detected by currently available assays,¹⁹ underscoring the need to integrate epidemiological context into donor assessment. Screening strategies should therefore be regarded as tools for risk reduction rather than absolute safeguards and should be complemented by clinical vigilance and post-transplant monitoring.

Fig. 2.

Conceptual model of the temporal dynamics of infection and the performance of available dengue diagnostic assays in organ transplantation. Note: The eclipse phase is defined as the period after the mosquito bite during which viremia is absent or falls below the detection threshold. The interval between a mosquito bite and the active replication phase (incubation) is not well defined for dengue. However, this model is extrapolated to HIV, hepatitis B, and C donor-derived infection. The immunological window is the period during which viremia by quantitative nucleic acid testing and antigen tests are negative and serological markers are not yet detectable.

These findings illustrate the trade-off inherent to a pragmatic screening strategy. NS1 and IgM testing is rapid, inexpensive, and scalable, but may miss early infections before antigen or antibody detection, as well as low-level infections later in the diagnostic window. On the other hand, universal NAT could improve the detection of early viremia but would entail greater cost, laboratory complexity, and longer turnaround times.

A direct comparison of donor-derived transmission before and after screening should be interpreted cautiously. The apparent severity of pre-screening donor-derived dengue may also reflect selection bias, as milder or asymptomatic transmissions may have remained unrecognized. Before implementation, two donor-derived events involving four recipients occurred among 1990 deceased-donor transplants, whereas only potential transmission events were identified among 1594 transplants after implementation. However, these periods are not directly comparable: the pre-screening events occurred during non-epidemic periods, whereas screening encompassed the largest dengue epidemic recorded in Brazil, and case ascertainment differed between eras. Moreover, screening itself altered the population proceeding to transplantation by excluding 12 donors with evidence of dengue who would otherwise have entered the usual allocation pathway. Thus, the observed post-screening transmission rate does not account potentially infectious donors that the strategy intercepted and should not be interpreted as a direct estimate of screening effectiveness.

Our data do not allow us to define a validated incidence threshold at which donor screening can be safely discontinued. However, because donor-derived transmission occurred even during non-epidemic periods, systematic screening should be maintained in endemic settings with sustained dengue transmission. In settings with a well-defined seasonal pattern, screening could be adapted to local epidemiology, with universal screening during periods of high transmission and a reduced strategy during predictably low-transmission periods. During epidemic periods, including when local dengue incidence exceeds 300 cases per 100,000 population, systematic donor screening must be strongly considered.

From a global health perspective, these findings have implications that extend beyond Brazil. Dengue is rapidly expanding into new geographic regions, driven by climate change, urbanization, and increased population mobility.9, 10, 11 Transplant programs in endemic and emerging areas face similar challenges in balancing the risks of infection transmission against the consequences of organ shortage.13, 14, 15, 16 Importantly, donor-derived transmission occurred even during a non-epidemic period, supporting systematic screening in endemic settings, particularly in low- and middle-income countries where the low cost and rapid turnaround of our approach may facilitate implementation.

Some limitations should be acknowledged. The observational design does not permit definitive causal inference about the effectiveness of screening strategies, and comparisons with unscreened populations were not conducted. In addition, classification of donor-derived transmission relied on a combination of clinical and virological criteria, and some cases were categorized as potential rather than confirmed, reflecting inherent diagnostic uncertainty. Yet, the small number of donor-derived transmission events limits statistical power to detect differences in these rare events and precludes robust estimates of screening effectiveness. Not all vector-borne cases were laboratory-confirmed, which may have introduced some degree of misclassification, particularly in the context of intense community transmission, and the nonspecific clinical presentation of dengue may also have resulted in misclassification of some suspected cases. However, as shown in Table 2, both groups had similar clinical presentations, and this classification followed the Brazilian health authorities' guidelines. The comparison of case-fatality rates between KTRs and the general population was unadjusted. Differences in age, comorbidity burden, diagnostic intensity, follow-up, and attribution of dengue-related deaths may have contributed to the observed difference; therefore, this comparison should be interpreted as a descriptive epidemiological comparison rather than an estimate of the independent effect of transplantation or immunosuppression on mortality. In addition, because systematic virological screening of all KTRs was not performed, asymptomatic or mild infections that did not result in contact with the healthcare system may not have been captured. Finally, this was a single-center study conducted in a high-volume transplant program, and the findings may not be fully generalizable to centers operating under different epidemiological conditions or resource constraints.

In conclusion, dengue in kidney transplant recipients was associated with substantial morbidity and a higher observed case-fatality rate than that reported in the general population. Although uncommon, donor-derived transmission represents a clinically significant risk. A structured organ donor screening strategy during epidemic periods can minimize transmission at allocation while preserving access to transplantation. Given the disastrous consequences of transfusion-transmitted dengue, a similar screening approach should be considered for blood donors, despite the overall low risk of transmission via this route. As arboviral infections continue to expand globally, and because live-attenuated vaccines are not used in solid organ recipients, integrating infectious disease surveillance into safety protocols is instrumental in ensuring both recipient safety and the optimal use of scarce donor organs.

Contributors

JMP and LRM conceived, designed, and supervised the study. DFR, CLA, NFO, CMS, HSG, ELF, RDF were involved in data collation. JMP, DRF, TR, ELF, and LRM were involved in data curation and analysis. JMP, TR, CFHG, and LRM were involved in data interpretation. JMP and LRM wrote the original draft of the manuscript. JMP, CFHG, and LRM reviewed and edited the original draft of the manuscript. All authors had accessed and verified the underlying data. All authors approved the final version of the manuscript. All authors had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Data sharing statement

De-identified individual participant data, including the data dictionary, will be available beginning upon publication and ending 5 years after publication to researchers who provide a methodologically sound proposal. Data will be made available after approval of the proposal and completion of a data-sharing agreement by contacting the corresponding author.

Editor note

The Lancet Group takes a neutral position with respect to territorial claims in published maps and institutional affiliations.

Declaration of interests

JMP and LRM are members of the Brazilian Ministry of Health Advisory Board for Kidney Transplantation. LRM serves as a nephrology consultant to the Department of Specialized and Thematic Care (DAET) of the Brazilian Ministry of Health. All other authors have no conflicts of interests to declare.