**Background:** Peritoneal dialysis (PD) is a home-based renal replacement therapy for chronic kidney disease (CKD). CKD patients exhibit accelerated aging and have 10- to 20-fold higher mortality risk than the general population, largely due to cardiovascular toxicity from uremia. The peritoneal membrane shows high person-to-person variability in its status before PD initiation, and deficiency of the anti-aging molecule α-Klotho has been linked to both cardiovascular damage and peritoneal membrane fibrosis. This study hypothesized that α-Klotho and other aging-related molecules could serve as markers of both membrane survival and patient survival in long-term PD.
**Methods:** This single-center prospective study at Santa Cruz Hospital, Portugal enrolled 58 incident PD patients with peritoneal biopsy at baseline. Patients were followed for 60 months. Peritoneal membrane histomorphology was scored as: Score 0 (no fibrosis, vasculopathy, or inflammation), Score 1 (no fibrosis but vasculopathy and/or inflammation), or Score 2 (fibrosis with/without vasculopathy/inflammation). Serum biomarkers including α-Klotho, galectin-3, FGF21, FGF23, Tweak, TNFα, and hr-CRP were measured by ELISA before PD initiation. Frailty was assessed using the Edmonton Frail Scale. Primary outcomes were PD technique failure and time to failure; secondary outcomes were MACE, time to MACE, and all-cause mortality. Statistical analyses included ROC curves, Kaplan-Meier survival curves, and Cox proportional hazards regression models.
**Key Results:** At baseline, 22 patients had peritoneal membrane fibrosis. The ROC-derived cutoff for α-Klotho to discriminate membrane fibrosis was 742 pg/mL (AUC = 0.860, p = 4 × 10⁻⁶, sensitivity 83%, specificity 71%). During follow-up (median 42 months), 41% of patients experienced PD technique failure (median time to failure 40 months), and 47% had a MACE (median time to MACE 17 months). Peritoneal membrane fibrosis was not associated with PD technique failure or all-cause mortality. However, membrane fibrosis was significantly associated with MACE occurrence and earlier time to MACE. Low α-Klotho (<742 pg/mL) independently predicted time to MACE in models adjusted for age, nutritional status, and peripheral arterial disease (PAD). This association persisted when α-Klotho cutoff replaced biopsy score in multivariate models, and when adjusted for age and antiplatelet use. Galectin-3 was associated with PD failure and time to PD failure. A cutoff of 8.88 ng/mL (sensitivity 92%, specificity 46%) discriminated PD failure and was independently associated with PD failure in a model adjusted for age, PAD, and calcium channel blockers. Six deaths occurred (5 cardiovascular, 1 malignancy), which were not related to biopsy score.
**Clinical Implications:** This study reveals that peritoneal membrane fibrosis reflects cardiovascular vulnerability rather than membrane survival risk in PD patients. Low α-Klotho (<742 pg/mL) identifies patients at higher risk for MACE and may serve as a non-invasive surrogate for peritoneal biopsy. Elevated galectin-3 (>8.88 ng/mL) identifies patients at risk for PD technique failure. These findings suggest that α-Klotho and galectin-3 could be clinically useful tools for risk stratification and personalized management in PD. The results also support that even older, frail, or high-cardiovascular-risk patients may benefit from home-based PD, as neither frailty nor membrane status predicted mortality or technique failure. Limitations include the small single-center sample (n=58), single-timepoint biomarker measurements, and lack of psychosocial/economic data. Further research is needed to validate these biomarkers as tools for tailored patient management.