MO411ACUTE KIDNEY INJURY AFTER PARATHYROIDECTOMY FOR PRIMARY HYPERPARATHYROIDISM: PREVALENCE, PREDICTORS AND RISK ASSESSEMENT

Abstract

Abstract Background and Aims Surgery is the most effective and the only definitive treatment of primary hyperparathyroidism (PHPT). We aimed to assess the prevalence of acute kidney injury (AKI) among patients underwent surgery for PHPT, to determine the possible risk factors. Method A retrospective cohort study included 290 patients who underwent successful selective parathyroidectomy (PTx) for PHPT. We did not include patients who underwent re-operative surgeries. AKI was defined according to KDIGO-2012 criteria. Results In our cohort, 106 patients (36,6%) met AKI criteria after PTx. Most of the patients developed AKI stage 1. In univariate analysis preoperative serum PTH level (р=0,0004) aa well as degree of its decrease before/after PTx (р<0,0001), preoperative serum total calcium level (р=0,0158), size of the parathyroid adenoma (р=0,0184), presence of proteinuria (RR=1,9 [95%CI: 1,19; 3,54], р=0,0061), hypertension (р=0,019) and anemia (р=0,0313), older age (RR=1,32 [95%CI: 1,03; 1,72], р=0,0265) were significant risk factors of AKI development. In multivariate analysis age (OR 1,05 [95%CI: 1,02; 1,08] per a year, р=0,002), body-mass index (OR 1,07 [95%CI: 1,02; 1,13] per each kg/m2, р=0,005), anemia (yes/no OR 3,38 [95%CI: 1,38; 8,2], р=0,008), preoperative PTH (OR 1,03 [95%CI: 1,01; 1,05] per each pmol/l, р=0,002), proteinuria (yes/no OR 3,45 [95%CI: 1,34; 8,93], р=0,011), use of ACE inhibitors/ARBs (yes/no OR 2,84 [95%CI: 1,58; 5,12], р=0,001) were discovered as independent predictors of AKI. Considering the most significant risk factors we developed two regression models for AKI risk assessment: the model 1 for patients with preserved kidney function (estimated glomerular filtration rate (eGFR ≥60 ml/min/1,73 m2) and the model 2 for those with decreased kidney function (eGFR less than 60 ml/min/1,73 m2) – tab. 1 and 2. Both models were statistically significant: χ2=25,39, df=5, р<0,001, RN2=0,341 for the model 1, χ2=19,355, df=3, р<0,001, RN2=0,428 for the model 2. The proposed models had good discrimination to predict AKI with area under the receiver operating characteristic curves (AUC-ROC) of 0,792 [95%CI: 0,691; 0,894], р<0,001 for the model 1 (normal kidney function) and 0,84 [95%CI: 0,73; 0,951], р<0,001 for the model 2 (decreased kidney function). Optimal cut-off values for predicted probability of AKI to define high-risk individuals were > 0,57 (Youden’s index 0,525) for the model 1 and > 0,439 (Youden’s index 0,589) for the model 2. Conclusion We observed high prevalence of AKI in patients after PTx for primary HPT. Developed risk models predict AKI with adequate accuracy. Risk factors of AKI should be considered when planning PTx, special attention should be paid to modifiable ones.