In accordance with the above-mentioned exclusion criteria, an overall of 281 out of 429 patients were included in the study. The median age of OLT recipients was 58 [51–63] years (Table 1). The majority of recipients were male (198; (70%)), and the most common indication for OLT was hepatocellular carcinoma (32%), closely followed by alcoholic liver cirrhosis (29%) (Table 1). The median pre-OLT laboratory Model of End-stage Liver Disease (labMELD) score [30] was 17 [10–27]. The median age of the donors was 58 [49–68] (Table 1). Some 145 donors were male and 136 were female (Table 1). The leading causes of death in the donor were cerebrovascular accidents (58%), anoxia (23%) followed by trauma (13%). Sixty-one percent of the donors met ECD criteria (Table 1). Further information on donor and recipient characteristics are shown in Table 1.

According to our predefined cutoffs, 87 (31%) of all patients showed relevant calcification of any of the regions of interest, from whom 81 (28.8%) had significant calcification of the visceral aortic segment (VAC) with a median Agatston score of 516 [211–805] mm3 and 19 (6.8%) of the celiac artery (CAC) with a median Agatston score of 115 [74–241] mm3. The AUROCs were 0.668 and 0.657, leading to the cutoff values of 100 mm3 and 50 mm3 (J = 0.331 and 0.278) for VAC and CAC, respectively.

Concerning demographics and clinical characteristics, patients with calcification of the visceral aortic segment or the celiac trunk were older (median age 62 vs. 55, p < 0.001; 64 vs. 57, p = 0.001, Table 1, respectively) than those without calcification. While labMELD did not differ significantly between the groups (20 [12–27] vs. 16 [9–27], p = 0.157; 21 [14–25] vs. 16 [10–27], p = 0.324, Table 1, respectively), patients with high VAC or CAC had lower ALT values prior to OLT (31 [22–52] vs. 44 [29–98], p = 0.001; 31 [23–40] vs. 42 [27–81], p = 0.03, Table 1, respectively) compared to patients without significant calcification. Concerning AST, lower values were found in VAC but not in CAC patients (45 [35–78] vs. 63 [45–134], p < 0.001; 54 [34–62] vs.42, 58 [42–122], p = 0.159, Table 1, respectively).

Chronic viral hepatitis was less frequent (2 vs. 9%, p = 0.039, Table 1, respectively) in patients with VAC, while patients with significant CAC showed a higher prevalence of hepatocellular carcinoma (53 vs. 31%, p = 0.048, Table 1, respectively). A higher number of patients with VAC had a history of pre-transplant abdominal surgery than those without VAC (42 vs 29%, p = 0.022, Table 1, respectively). Detailed patients’ characteristics are displayed in Table 1.

Patients with atherosclerotic involvement of either the visceral aorta or the celiac trunk showed a comparatively higher incidence of early postoperative renal function impairment, as determined by eGFR, compared to patients without calcifications in the first week following OLT (VAC: p = 0.0016; CAC: p = 0.0211; Fig. 2, respectively). In the analysis of kidney function, patients receiving pre-OLT dialysis were excluded (n = 42, 15%). Pre-OLT eGFR did not significantly differ between patients with significant calcification versus without (VAC: 60 [32–77] ml/min vs. 60 [44–91] ml/min p = 0.054; CAC: 60 [17–79] ml/min vs. 60 [42–88], p = 0.285; Fig. 3, respectively). Estimated GFR was prominently decreased with lowest values on the fourth postoperative day (32 [20–57] ml/min; Fig. 2, respectively) in patients suffering from VAC, while patients without aortic calcification showed worst renal function on POD 3 (44 [25–60] ml/min; Fig. 2, respectively). The most significant difference in renal function could be observed on POD 5 between the above-mentioned groups (46 [24–60] ml/min vs. 60 [34–78], p = 0.005; Fig. 3, respectively). Both, patients with and without CAC suffered the nadir of renal function impairment on POD 3 (28 [18–39] ml/min vs. 42 [24–60], p = 0.564; Fig. 2, respectively). Despite the continuous differences, observed graphically, the only statistically significant difference concerning eGFR levels was found on POD 2 (27 [16–42] ml/min vs. 56 [31–62], p = 0.029; Fig. 2, respectively).

Estimated glomerular filtration rate pre-OLT and over the course of the first two weeks after liver transplantation. Patients receiving pre-OLT dialysis were excluded. Parameter stratified by critical calcification of the aorta (a) or the celiac artery (b). Data are shown as median and interquartile ranges (IQR). OLT orthotopic liver transplantation, POD postoperative day, eGRF estimated glomerular filtration rate, CAC celiac artery calcification, VAC visceral aortic calcification

Results comparing VAC and VAC+ . a Correlation analysis between VAC and extended visceral aortic calcification values or VAC+ . Spearman correlation plot including ± 95% confidence interval. b, c Patients receiving pre-OLT dialysis were excluded. Estimated glomerular filtration rate stratified by VAC+ (a) and comparing VAC+ with VAC. Data are shown as median and interquartile ranges (IQR). CAC celiac artery calcification, VAC visceral aortic calcification, VAC+  extended visceral aortic calcifications including the renal artery ostia;, HU: Hounsfield units, eGRF estimated glomerular filtration rate

Overall, severe postoperative complications (CD ≥ 3b) occurred in 56% of cases and the 90-day mortality after OLT was 9%. Neither of these parameters was significantly associated with visceral calcification (Table 2). Patients with significant VAC showed higher rates of EAD (38 vs. 26%, p = 0.031; Table 2, respectively), while patients with CAC did not (37 vs. 29%, p = 0.316; Table 2, respectively). While no statistically significant difference in estimated procedural costs was found between VAC and non-VAC patients (58 [40–92] TEuro vs 53 [39–80] TEuro, p = 0.284; Table 2, respectively), relevant CAC was significantly associated with higher procedural costs (70 [48–101] TEuro vs 53 [39–79] TEuro, p = 0.049; Table 2, respectively). Concerning length of ICU stay and length of overall hospitalization, no difference was found between the groups. The analysis of aortic and celiac trunk calcification levels and their association with perioperative outcome are shown in Table 2.

Finally, in the uni- and multivariable logistic regression analyses (Table 3), recipient age (0.634 OR, 0.239–1.089 CI, p = 0.007; Table 3, respectively), cold ischemic time (1.931 OR, 1.115–3.346 CI, p = 0.019; Table 3, respectively), and VAC (2.387 OR, 1.290–4.418 CI, p = 0.006; Table 3, respectively) were identified as independent predictors of EAD.

As VAC has been shown to have a significant impact on renal outcomes, as a subsequent step VAC+ has been introduced, involving a secondary measurement with the inclusion of both renal artery ostia (Fig. 1). VAC+ has demonstrated a similar AUROC and Youden index (0.623, J = 0.298) like VAC, yielding a cutoff value of 200 mm3. A total of 85 patients (30%) suffered from a significant level of calcification based on VAC+. 

VAC and VAC+ have shown a very strong association (r = 0.922; p < 0.0001, Fig. 3). Renal outcomes showed a largely similar pattern like with VAC and eGFR values of patients with critical calcification based on VAC versus VAC+  were nearly identical at each time point (Fig. 3). Further perioperative data including complications, hospital stay, transfusions were almost identical between VAC+ and VAC patients (Supplementary Table 1).

Therefore, the extended measurement of VAC+ has failed to provide a significantly superior value in assessing perioperative outcomes compared to VAC.

Patients who died within the first 90 days following OLT (n = 24) were not excluded from this analysis as no significant effects of VAC or CAC on short-term mortality were detected (Table 2, respectively). The median length of follow-up for the patient cohort was 70 months. Calcification of neither the visceral aorta nor the celiac trunk had significant effects on long-term graft and patient survival. Even though not statistically significant, high levels of VAC showed slightly inferior graft (1 year: 76 vs. 81%, 3 years: 71 vs. 73%, 5 years: 64 vs. 69%, p = 0.483; Fig. 4, respectively) and patient survival rates (1 year: 76 vs. 85%, 3 years: 71 vs. 78%, 5 years: 65 vs. 73%, p = 0.217; Fig. 4, respectively). In line with these findings, graft (1 year: 71 vs. 81%, 3 years: 65 vs. 73%, 5 years: 52 vs. 68%, p = 0.216; Fig. 4, respectively) and patient survival (1 year: 71 vs. 83%, 3 years: 65 vs. 80%, 5 years: 52 vs. 72%, p = 0.105; Fig. 4, respectively) were slightly inferior in patients with CAC, even though this difference did not reach the level of statistical significance (Fig. 4).

Graft and patient survival stratified by significant calcification of the visceral aorta and celiac artery. a 5-year patient and b graft survival by critical calcification of the visceral aorta (64%) vs. no significant calcification (69%). c 5-year patient and d graft survival by critical calcification of the celiac artery (52%) vs. no significant calcification (68%). CAC celiac artery calcification, VAC visceral aortic calcification