Cross-sectional data were used from the ongoing and prospective TransplantLines Biobank and Cohort Study (ClinicalTrials.gov identifier: NCT03272841; METc 2014/077). This study collected data from all SOT recipients and donors (aged ≥18 years) with a written informed consent, starting from June 2015 at the University Medical Centre Groningen (UMCG, Netherlands). TransplantLines adheres with the UMCG Biobank Regulation, WMA Declaration of Helsinki and the Declaration of Istanbul [26, 27].

In the current study, we included adult (aged ≥18 years at inclusion) SOT recipients at 1-year after SOT (liver, lung, heart and kidney recipients). Assessments were performed during the standardized 1-year post-transplant visit, which occurred within a predefined window of approximately 9–16 months after transplantation. Data between June 2015 and December 2024 were used. If recipients had multiple transplant trajectories included in TransplantLines, data from the first transplant trajectory were used if the first transplant occurred at least 1 year before the second transplant, and the allograft did not fail. We excluded recipients who had a combined SOT. This study was performed in accordance with the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) statement (Supplementary Table S1).

Before and including 2021, clinical data and data on medication use (verified with the patient) were collected during TransplantLines study visits. SBP and DBP, were measured in seated position after 15 min of rest, with four consecutive seated measurements. Waist circumference was measured at least twice. Mean value was used if multiple measurements were available [28]. After 2021, study visits were discontinued, and data were extracted from the electronic patient dossier (EPD), with blood pressure values including both patient-reported and physician-recorded measurements, which ranged from single measurements or averages of repeated measurements. Laboratory values were obtained through standard laboratory measurements. For the calculation of eGFR, the 2009 CKD-EPI formula was used [29]. Metabolic syndrome was defined according to the criteria of the International Diabetes Federation [30]. Instead of fasting glucose, HbA1c was used with a threshold of 53 mmol/mol, as not all patients were able to fast at the time of blood sampling due to long travel times to our transplant center. Cardiovascular events (CVE) were defined as the occurrence of myocardial infarction, heart failure, peripheral vascular disease, cerebrovascular accident (CVA), or transient ischemic attack (TIA). These events were obtained from EPD using ICD-10 codes. A medical history of CVE was included as a composite marker to reflect overall cardiovascular risk burden. Pre-transplant hypertension was defined as documented hypertension before transplantation, based on ICD-10 codes. Diabetes was defined as documented diabetes, based on ICD-10 codes, or use of antidiabetic medication (metformin, sulfonylurea derivatives, SGLT2 inhibitors, DPP-4 inhibitors, GLP-1 receptor agonists, and insulin). Salt intake (g/day) was estimated from 24-h urinary sodium excretion [31]. Lifestyle parameters, including alcohol intake, and smoking status, were derived from self-reported questionnaires.

The primary outcome was the proportion of SOT recipients with suboptimal hypertension control. As there is no clear consensus about the optimal target values for SBP and DBP after SOT, a target value of ≤130/80 mmHg was applied. This decision was based on local, national and international guidelines, such as the Dutch national guidelines for cardiovascular risk management [32], the guidelines of the various SOT departments at the UMCG, the KDIGO guidelines [20], American guidelines for prevention, detection, and management of high blood pressure in adults [18], and two studies investigating different blood pressure target values [19, 21]. Hypertension was defined as a SBP >130 mmHg, a DBP >80 mmHg or current use of antihypertensive medication. Suboptimal hypertension control was defined as a SBP >130 mmHg or DBP >80 mmHg, regardless of antihypertensive treatment. The following medication were considered as antihypertensives: calcium channel antagonists, beta-blockers, angiotensin-converting enzyme-inhibitors (ACEi), angiotensin receptor blockers (ARB), alpha-blockers, loop diuretics, thiazide diuretics and potassium-sparing diuretics. The secondary outcome was the difference in suboptimal hypertension control between the transplantation programs (heart, liver, lung and kidney).

Statistical analyses were performed using SPSS software (version 28.0, IBM Corporation, Armonk, NY) and R version 4.4.1. Normality of distribution was visually assessed using histograms and quartile-quartile plots. Continuous data were presented as means with standard deviations (SD) or medians with interquartile ranges (IQR), depending on data distribution, or in numbers with valid percentages for categorical data. Differences between groups were assessed using independent t-tests or one-way ANOVA for normally distributed variables, with Tukey’s HSD or, when variances were unequal, the Games–Howell post hoc test applied when applicable. For non-normally distributed variables, the Mann–Whitney U test or Kruskal–Wallis test was used, with Dunn’s post hoc test (Holm adjustment) applied when applicable. Categorical variables were compared using chi-square or Fisher’s exact tests with Bonferroni-corrected pairwise comparisons when applicable. To assess associations between recipient characteristics and blood pressure status, univariable and multivariable multinomial logistic regression analyses were performed. Blood pressure status was categorized as no hypertension (reference group), optimal hypertension control, and suboptimal hypertension control. Multivariable analyses were adjusted for age, sex, and transplantation type. To assess factors associated with antihypertensive treatment among hypertensive SOT recipients, additional univariable and multivariable binary logistic regression analyses were performed comparing treated versus untreated recipients with suboptimal control.

As optimal blood pressure thresholds after SOT are still debated in the literature, sensitivity analyses were performed using target values of ≤135/85 mmHg and ≤140/90 mmHg. Furthermore, because blood pressure measurements differed, and included extractions from EPD and protocolized measurements, a sensitivity analysis restricted to protocolized blood pressure measurements was performed to evaluate the robustness of our findings. To assess potential selection bias due to missing blood pressure data in the liver transplant population, baseline characteristics of included and excluded liver transplant recipients were compared. As CVE was analyzed as a composite outcome, the individual cardiovascular disease components (myocardial infarction, heart failure, peripheral vascular disease, and CVA/TIA) were analyzed separately to assess potential heterogeneity between conditions.

Finally, concordance between prescribed antihypertensive agents and 24-h urinary metabolite detection was assessed in a random sample of the population for selected antihypertensives. Metabolites of ACEi, ARB, and beta-blockers were evaluated through reversed-phase liquid chromatography coupled to high-resolution quadrupole-time-of-flight mass spectrometry in positive electrospray ionization and with SWATH data-independent acquisition. Recipients were classified as concordant, partially concordant, or non-concordant based on urinary metabolite detection relative to prescribed medication.

Out of 1169 recipients with available data at 1-year post-transplantation included in the TransplantLines biobank and cohort study, 57 recipients (4.9%) were excluded because of missing data on SBP or DBP. This concerned one heart transplant recipient, 31 liver transplant recipients, 3 lung transplant recipients and 22 kidney transplant recipients. Consequently, 1112 SOT recipients were included in the current study, among which 37 heart, 119 liver, 162 lung and 794 kidney transplant recipients (Supplementary Figure S1). The mean age of this population was 57 ± 13 years, and 433 recipients (39%) were female. Mean SBP was 134 ± 16 mmHg and mean DBP was 78 ± 10 mmHg (Table 1; full version in Supplementary Table S2).

To assess potential selection bias in the liver transplant population, baseline characteristics of included and excluded liver transplant recipients were compared. Differences in baseline characteristics between groups were limited (Supplementary Table S3).

Of the SOT recipients, 997 (90%) had hypertension. Of the recipients with hypertension, 851 (85%) used antihypertensives. Within this hypertensive group, 721 (72%) had suboptimal hypertension control and 146 (20%) of them were untreated (Figure 1).

SOT recipients with hypertension were more often male, had a higher BMI, had more frequently metabolic syndrome, had more often cardiovascular events in their medical history, had more often pre-transplant hypertension, had higher hematocrit and hemoglobin levels, and had a lower eGFR compared to recipients without hypertension. Recipients with suboptimal control had more frequently diabetes compared to recipients without hypertension. Additionally, recipients with suboptimal control were older, had a higher pulse pressure and had a higher mean arterial pressure (Supplementary Table S4).

In multivariable multinomial logistic regression, using recipients without hypertension as the reference group, several characteristics were independently associated with hypertension status (optimal or suboptimal).

Compared with recipients without hypertension, optimal hypertension control was independently associated with male sex (OR = 1.74; 95%CI1.10–2.76), higher BMI (OR = 1.11; 95%CI1.05–1.18), presence of metabolic syndrome (OR = 8.28; 95%CI3.84–17.84), larger waist circumference (OR = 1.04; 95%CI1.01–1.08), prior cardiovascular events (OR = 2.38; 95%CI1.15–4.92), pre-transplant hypertension (OR = 2.82; 95%CI1.44-5.51) and lower HDL-cholesterol levels (OR = 0.52; 95%CI0.30–0.91) (Supplementary Table S5).

Compared with recipients without hypertension, suboptimal hypertension control was independently associated with older age (OR = 1.03; 95%CI1.01–1.04), male sex (OR = 1.98; 95%CI1.31–3.01), higher BMI (OR = 1.13; 95%CI1.07–1.19), presence of metabolic syndrome (OR = 9.05; 95%CI4.38–18.69), larger waist circumference (OR = 1.04; 95%CI1.01–1.07), prior cardiovascular events (OR = 2.08; 95%CI1.04–4.14), presence of diabetes (OR = 1.99; 95%CI1.18–3.36), pre-transplant hypertension (OR = 2.02; 95%CI1.07-3.84), higher levels of total cholesterol (OR = 1.23; 95%CI1.01–1.51), higher levels of triglycerides (OR = 1.33; 95%CI1.04–1.69), higher levels of creatinine (OR = 2.07; 95%CI1.05–4.08), and a lower eGFR (OR = 0.99; 95%CI0.97–1.00) (Supplementary Table S5).

Suboptimal control was observed in 721 (72%) of recipients with hypertension, of whom 146 (20%) were not receiving antihypertensive treatment. These participants were younger, more often female, and had a lower BMI. They also had a higher pulse rate, lower pulse pressure, higher mean arterial pressure, lower salt intake, a lower prevalence of cardiovascular events, and a lower prevalence of pre-transplant hypertension compared with those who received treatment. Additionally, they used less often prednis(ol)on. In terms of metabolic and renal parameters, they showed lower hematocrit levels, lower hemoglobin levels, higher LDL-cholesterol values and higher eGFR values (Supplementary Table S6).

In multivariable logistic regression analyses several factors were independently associated with treatment: a higher age (OR = 1.03; 95%CI1.01–1.05), male sex (OR = 1.55; 95%CI1.03–2.34), higher salt intake (OR = 1.07; 95%CI1.00–1.14), medical history of cardiovascular events (OR = 2.06; 95%CI1.14–3.95), higher levels of glucose (OR = 1.16; 95%CI1.04–1.31), and lower eGFR values (OR = 0.99; 95%CI0.97–1.00) (Supplementary Table S7).

Comparing the different SOT transplantation programs, hypertension prevalence was different between the groups (p < 0.001), with hypertension present in 92% of heart, 69% of liver, 80% of lung and 95% of kidney transplant recipients. Suboptimal hypertension control was observed in 74% heart, 84% liver, 73% lung and 71% kidney transplant recipients, with no significant difference between groups (p = 0.09). However, the proportion of untreated recipients within the suboptimal hypertension control group varied significantly between organ groups (p < 0.001): 24% of heart, 54% of liver, 47% of lung, and 11% of kidney transplant recipients, with suboptimal hypertension control, remained untreated. Additionally, 32% of heart, 13% of liver, 12% of lung and 59% of kidney transplant recipients with suboptimal treatment, received multiple antihypertensive agents (Figure 2).

Kidney transplant recipients were more likely to have both optimally and suboptimally controlled hypertension compared with liver and lung transplant recipients (Supplementary Table S5). No significant differences were observed between kidney and heart transplant recipients.

Kidney transplant recipients without treatment were primarily younger, more frequently women, had less often diabetes and had fewer cardiovascular events in their medical history (Supplementary Table S8). Liver transplant recipients without treatment were characterized by a lower frequency of metabolic syndrome, lower frequency of pre-transplant hypertension and lower salt intake (Supplementary Table S9). Untreated lung transplant recipients were younger compared to those treated (Supplementary Table S10). No significant differences were found in the heart transplant subgroup (Supplementary Table S11).

Kidney transplant recipients were more likely to receive antihypertensive treatment compared to liver and lung transplant recipients (Supplementary Table S7). Heart transplant recipients did not differ significantly from kidney transplant recipients.

Different blood pressure cut-off values (≤135/85 mmHg and ≤140/90 mmHg) were applied in sensitivity analyses. The proportion of recipients reaching these target values increased compared to the primary threshold of ≤130/80 mmHg. Using a target cut-off of ≤135/85 mmHg, hypertension occurred in 957 (86%) of SOT recipients and suboptimal control was present in 524 (55%) SOT recipients with hypertension (Supplementary Figure S2). Across transplant groups, the proportion of SOT recipients with suboptimal hypertension control ranged from 52% to 76% (Supplementary Figure S3).

Using a target cut-off of ≤140/90 mmHg, hypertension occurred in 921 (83%) of SOT recipients and suboptimal control was present in 370 (40%) SOT recipients with hypertension (Figure 3). Across transplant groups, the proportion of SOT recipients with suboptimal hypertension control ranged from 44% to 62% (Supplementary Figure S4).

To address potential heterogeneity in blood pressure measurements, an additional sensitivity analysis was performed restricted to recipients with protocolized study visit-based blood pressure measurements (Supplementary Table S12). In the primary analysis, suboptimal hypertension control was present in 721 of 997 hypertensive SOT recipients (72%). Findings remained consistent when restricting the analysis to recipients assessed with protocolized study visit-based measurements (298/386, 77%).

Because CVE was analyzed as a composite outcome, separate sensitivity analyses were performed for the individual cardiovascular disease components (myocardial infarction, heart failure, peripheral vascular disease, and CVA/TIA) to assess potential heterogeneity. These analyses showed broadly comparable directions of association to those observed for the composite CVE outcome. However, confidence intervals were wider and statistical significance was less consistent across individual cardiovascular conditions (Supplementary Tables S13, 14).

In a randomly selected subgroup of 156 recipients, metabolites of selected antihypertensive agents (ACEi, ARB, and beta-blockers) were analyzed in 24-h urine samples. Of these, 60 recipients used no antihypertensive medication according to medication lists, consistent with urinary metabolite measurements. Of the 96 recipients prescribed one or more ACEi, ARB, or beta-blockers, 88 (92%) showed concordance between prescribed medication and urinary metabolite detection, whereas 4 (4%) showed partial concordance and 4 (4%) showed no concordance (Supplementary Table S15).