Patients
The study was carried out during the period over 24 months. 56 patients with NS were included in to the study. 26 children had SSNS and 30 had SRNS. In the SSNS group the mean of patients ages was 11.2 ± 0.6, 53.3% of them were males, all of them had generalized edema, and 12 of them (46.2%) of them were hypertensive.
In the SRNS group the mean of patients’ ages was 11.0 ± 0.6, 66.6% of them were males, all of them had generalized edema, and 70% of them were hypertensive (p values data are all summarized in Table 1.
The comparison of SSNS and SRNS patients’ clinical findings showed a statistically significant difference regarding the incidence of hypertension, which was evidently higher in the SRNS group (P value < 0.01). Major part of participants in SSNS and SRNS were boys (р<0,05 and р<0,05, respectively).
Basic laboratory biomarkers
Also, comparison of the laboratory findings showed statistically significant differences between SSNS and SRNS regarding WBC count – 6.87±0.47·109/L and 9.14±0.47·109/L, respectively (p<0.001). No difference in level of platelets count found in SSNS and SRNS - 258.5±11.5·109/L and 289.0±11.5·109/L, respectively (p>0.05) (Figure 1).
Biochemical markers i.e. blood urea, serum creatinine, serum cholesterol measured in all children. Serum creatinine and serum cholesterol found to be significantly different between the SSNS and SRNS groups. Serum creatinine concentrations documented at level 68.38±3.46 mcMol/L in children with SSNS and 80.23±2.72 mcMol/L in children with SRNS (p<0.01). Serum cholesterol levels were higher in SRSN 11.97±0.38 mMol/L children as compared to 9.89±0.38 mcMol/L in children with SRNS (p<0.001). Blood urea levels were 9.99±0.63 mMol/L and 11.08±0.5 mMol/L (p>0.05) (Figure 2).
Glomerular filtration rate (GFR) and 24-h urine protein excretion were selected as markers of kidney function impairment. GFR found to be significantly lower in SRNS children as compared to children with SSNS - 68.38±2.72 mL/min/1,73 m2 and 82.84±4.3 mL/min/1,73 m2 (p<0.01). No difference found in proteinuria level when comparing SSNS group and SRNS group - 9.1±0.71 g/24 h and 8.73±0.59 g/24 h (p>0.05) (Figure 3).
Apoptotic markers in children with nephrotic syndrome
The expression of proapoptotic and antiapoptotic factors were studied using immunoblotting. Bax, caspase-3 and caspase-8 were chosen as apoptotic markers, Bacl-xL as antiapoptotic.
Increased level of Bax expression in SRNS in comparison to SSNS documented - 90.15±2.0 mMol/L and 104.5±3.o a.u., respectively (p<0.01). Bax can initiate a cascade of events that eventually involves cleavage of caspase-3. Caspase-3 which is a final effector in mitochondrial level of apoptosis measured at level 131.1±1.45 a.u. in children with SSNS and 141.4±1.59 a.u. in children with SRNS (p<0.001).
Caspase-8 selected as a proapoptotic marker and effector of the extrinsic apoptotic pathway. No difference in caspase-8 level found between SSNS and SRNS - 137.3±1.5 a.u. 139.6±1.88 a.u., respectively (p>0.05).
Significantly lower value of antiapoptotic factor BcL-xL found in children with SRNS in comparison to SSNS - SSNS and SRNS - 110.5±1.84 a.u. 92.88±2.74 a.u., respectively (p<0.001). (Figure 4).
Beside apoptotic markers, studying of the NF-kB, HIF-1alfa, Mn-SOD done. NF-kB NF-κB is expressed in nearly all mammalian cells. Under physiologic (resting) conditions, NF-κB exists as an inactive form composed of a dimeric form of the transcription factor and its binding partner, IκB, which maintains NF-κB in an inactive form. NF-κB is activated in response to a various types of stress, including chronic inflamation [9,10]. We found that SRNS children have lower NF-kB level (107.3±1.98 a.u) as compared to SSNS children (97.42±3.34 a.u) (p<0.05).
In mammalian cells, manganese superoxide dismutase (MnSOD) is the essential mitochondrial antioxidant enzyme that detoxifies the free radical superoxide, the major by-product of mitochondrial respiration [11]. Mitocholdrial respiration changes have place under inflammatory conditions. Level of Mn-SOD was significantly lower in children with SRNS and compared to SSNS - 109.9±3.0 and 95.16±4.33, respectively (p<0.05).
Finally, levels of HIF-1alfa as a marker of intracellular hypoxia studied in nephrotic children [12]. Significantly higher HIF-1alfa in children with SRNS in comparison to group with SSNS found - 202.8±4.44 a.u and 212.9±2.14 a.u,, respectively (p<0.01).
Identification of clinical and metabolic factors defining and predicting steroid resistance in nephrotic children
Multivariate linear regression analysis used to determine the clinical predictors of steroid resistance in nephrotic children among categorical factors. Demographic parameter i.e. age, gender and presence of arterial hypertension were included into analysis. Factor predicting SRNS these were identified arterial hypertension presence (β - 0,31, SE 0.15, 95% CI -0.6 to -0.007, p<0.05). None of other studied factors found as a predictors of SRNS – age (β - 0,02, SE 0.03, 95% CI -0.07 to 0.03, p>0.05), gender - (β 0.11, SE 0.13, 95% CI -0.16 to 0.39, p>0.05) (Table 2).
Logistic regression models for the odds of steroid resistance among numerical factors in nephrotic children made using a step-wise selection procedure. Candidate predictors entering the model at p<0.05. Three models created: (I) basic laboratory factors – RBC, WBC, PTC; (II) biochemical and factors: Serum creatinine, Serum urea, Serum cholesterol; Functional factors: GFR, proteinuria; (III) apoptosis controlling factors: Bax, BcL-xL, caspase-3, caspase-8; (IV) – transcriptional factors: NF-kB, HIF-1alfa, Mn-SOD. Receiver operating characteristics curve and area under the curve (AUC) analyses used to assess each model’s ability to define clinical steroid resistance.
Logistic regression analysis (Table3) of basic laboratory factors selected WBC level (OR: 1.4; OR 95% CI: 1.12–1.85) and RBC level (OR: 1.14; OR 95% CI: 0.38–3.48) but not PLT count (OR: 1.0; OR 95% CI: 1.0-1.02) as a markers able to define the steroid resistance in nephrotic children (AUC: 0.75; SE: 0.06; 95% CI: 0.63-0.88; p<0.001) (Figure 6A).
Among biochemical factors and parameters showing kidney function impairment (Table3) serum creatinine (OR: 1.04; OR 95% CI: 0.92–1.14), serum urea (OR: 1.14; OR 95% CI: 0.89–1.49), serum cholesterol (OR: 1.97; OR 95% CI: 1.37–3.18) were found to be a factors defining SRNS. GFR level (OR: 0.95; OR 95% CI: 0.87–1.04) and proteinuria level (OR: 0.85; OR 95% CI: 0.67–1.06) did not fit the model (AUC: 0.88; SE: 0.05; 95% CI: 0.78-0.97; p<0.001) (Figure 6B).
Apoptosis regulating factors Bax, BcL-xL, caspase-3, caspase-8 (Table3) were analyzed in terms of factors predicting SRNS. BcL-xL (OR: 1.4; OR 95% CI: 1.17–1.91), Bax (OR: 1.15; OR 95% CI: 1.04–1.37), caspase-3 (OR: 1.66; OR 95% CI: 1.27–2.68) but not caspase-8 (OR: 0.95; OR 95% CI: 0.84–1.05) (AUC: 0.93; SE: 0.03; 95% CI: 0.87-0.99; p<0.001) (Figure 6C).
Finally, group of transcriptional factors NF-kB, HIF-1alfa, Mn-SOD included into the logistic analysis (Table3). We found that HIF-1alfa, which is a marker of intracellular hypoxia selected as a factor defining steroid resistance phenomenon (OR: 1.02; OR 95% CI: 0.98–1.07). Nuclear transcriptional factor NF-kappaB (OR: 0.95; OR 95% CI: 0.9–1.01) and antioxidative factor Mn-SOD did not fit the model (OR: 0.95; OR 95% CI: 0.92–0.98) - (AUC: 0.78; SE: 0.06; 95% CI: 0.67-0.91; p<0.001) (Figure 6D).
Pearson correlation was run to correlate principal factors found to be those defining steroid resistance in nephrotic children – Bax, BcL-xL, caspase-3, HIF-1alfa in both SSNS group (Figure 7-A) and SRNS group (Figure 7B).
For SSNS group significant negative correlation was observed between antiapoptotic factor BcL-xL and proapoptotic factor Bax (r = - 0.45, P<0.05), significant negative correlation was observed between antiapoptotic factor BcL-xL and proapoptotic factor caspase-3 (r = - 0.79, P<0.05), significant positive correlation was observed between marker of cellular hypoxia HIF-1alfa and proapoptotic factor caspase-3 (r = 0.56, P<0.01) and proapoptotic factor Bax (r = 0.43, P<0.05).
For SRNS group significant negative correlation was observed between antiapoptotic factor BcL-xL and proapoptotic factor Bax l (r = - 0.46, P<0.05), significant negative correlation was observed between antiapoptotic factor BcL-xL and proapoptotic factor caspase-3 (r = - 0.78, P<0.01), significant positive correlation was observed between marker of cellular hypoxia HIF-1alfa and proapoptotic factor caspase-3 (r = 0.92, P<0.05) and proapoptotic factor Bax (r = 0.53, P<0.05).