Around 3% of individuals diagnosed with type 2 diabetes mellitus have evident nephropathy at the moment of diagnosis, whereas an additional 20%-30% develop nephropathy at a later stage in their lives. Diagnosing diabetic individuals for renal damage as soon as possible is critical. In addition to decreasing the necessity for renal replacement therapy, it also mitigates the significant morbidity and death rates that are commonly linked to impaired renal function [13]. However, increasing evidence suggests that many patients with diabetes who are followed as DNP have features typical of NDNP or diabetic nephropathy in overlap histology with other nephropathies [7]. Diabetic individuals who present with active urine sediment, quick onset nephrotic syndrome, or rapid decline in renal function are deemed to warrant a biopsy. [13]. According to the classical premise, KB is indicated when a differential diagnosis that could affect clinical care is required. It is widely acknowledged that NDNP has a better prognosis than DNP, as renal lesions in DNP are considered difficult to reverse. In contrast, NDNP is often treatable and may even be reversible [14]. Detection of these pathologies by KB in diabetic patients will affect renal survival. Clinicians may utilize clinical and biochemical indicators that forecast NDNP to aid in the determination of the necessity of a biopsy [15]. In this study, we analyzed the frequency of non-diabetic nephropathy in patients with diabetes mellitus, as well as the clinical and laboratory findings that may help predict its occurrence.
The prevalence of NDNP in patients with DM varies significantly across different regions [16, 17]. In our study, non-diabetic nephropathy was found in half of diabetic patients. The substantial heterogeneity observed in the prevalence of NDNP among studies could potentially be attributed to the divergent judgments rendered by hospitals and physicians regarding the necessity of renal biopsy. Additionally, regional differences in the prevalence of DM and its accompanying complications may have an effect on the incidence of NDNP. The high frequency of NDRD detected in our study suggests that nephrologists appropriately select only diabetic patients for biopsy who are considered atypical for DNP or have clinical features suspicious for NDNP.
While the literature lists numerous histopathological diagnoses for NDNP, IgA nephritis, FSGS, and membranous nephropathy are the most prevalent [18, 19]. In the study of Erdogmus et al., NDNP was identified at a frequency of 50%, with membranous nephropathy being the most prevalent disease in this cohort [20]. In the study conducted by Heybeli et al. FSGS was most frequently detected in the NDNP group [21]. The increased occurrence of FSGS symptoms in our cohort of patients may be ascribed to the greater incidence of comorbidities, including obesity, diabetes mellitus, metabolic syndrome, and hypertension. These diseases may result in FSGS due to hyperfiltration-induced glomerular injury.
DNP symptoms often take a long time to become apparent and are therefore often not diagnosed in the early stages. [16]. As a result, when conventional inducers of renal biopsy in DM patients are absent, patients are often observed until advanced stages, with the majority of patients opting not to have KB. Thus, it is not surprising that patients with DNP, as in our study, have more chronic changes in the pathological findings compared to those with NDNP.
Clinical characteristics that have been previously demonstrated to be predictive of renal involvement in diabetic patients by NDNP include hematuria, acute kidney failure, abrupt onset of proteinuria, and proteinuria in the absence of retinopathy or neuropathy [14, 16]. In our study, female gender, absence of DR, non-HbA1c variability, positive serology were found to be risk factors in predicting non-diabetic nephropathy.
In the clinical setting, serological tests (ANA, ANCA, ENA, protein electrophoresis) are often performed on diabetic patients with acute kidney failure and proteinuria with a sudden onset or increase. Serological tests have been investigated for their utility in differentiating between DNP and NDNP. The literature indicates that serological tests may have some diagnostic value in NDNP, but their efficacy varies. When pre-biopsy serological tests were matched with pathological diagnoses in the Swedish population, the presence of antiphospholipid antibodies and low complement levels were found to reduce the likelihood of DNP the most. [22]. Positive serological testing, as anticipated, aids the clinician in the differential diagnosis of NDNP in DM patients and in the determination of whether to do a biopsy. Positive serological marker results were identified as the risk factor with the greatest specificity in our study.
Both male and female genders have been demonstrated to be predictors of NDNP in the literature [23, 24]. In our study, more male dominance was found in the DNP group and female gender was found to be a risk factor for NDNP. There may be several reasons for this. Firstly, in studies conducted in diabetic patients, the prevalence of clinically diagnosed DNP was found to be lower in women, and it was stated that this may be related to the protective effects of estrogen [25]. In a study conducted in Turkey investigating the frequency of DNP by blood-urine analyses, it was found to be significantly more frequent in males (female 22.9% vs. male 29.2%, p = 0.007). [26]. Secondly, autoimmune diseases are known to occur more frequently in women [27]. Finally, it is known that FSGS, the most frequently detected renal pathology in the NDNP group, is mostly due to secondary/adaptive causes and obesity is one of these causes. The prevalence of obesity is higher in women in Turkey [28]. From this point of view, it may be realistic to assume that female gender is a risk factor for NDNP.
Diabetic retinopathy is a microvascular complication of DM that may share pathogenetic pathways with DNP. When retinopathy coexists with nephropathy, it is considered a potential indicator of renal complications [7, 16]. The KDOQI 2007 guidelines state that if DR with albuminuria is present, the cause of CKD may be attributed to DNP [29]. Therefore, retinopathy can be considered as an indicator of DNP [30]. In our study, DR was detected more frequently in patients with DNP than in the NDNP group and the absence of DR was determined to be a risk factor for NDNP. In the ROC analysis, it was found to be the most successful parameter in differentiating NDNP from DNP.
The HbA1c value is widely regarded as the most reliable method for assessing glycemic control. Long-term glycemic control is the primary risk factor for the development of micro and macrovascular complications in diabetic patients, as evidenced by cumulative research [31]. In the literature, it has been repeatedly reported that elevated HbA1c is a risk factor in predicting DNP. [32–34]. In this study, the parameters indicating uncontrolled diabetes (such as intensive diabetes treatment and HbA1c levels) were significantly more common in the DNP group. These results suggest that higher HbA1c levels in DM patients with kidney disease are associated with a higher likelihood of DNP, while lower HbA1c levels are more likely to be associated with NDRD. In contrast, the ACCORD trial found that intensive treatment aimed at achieving normal HbA1c levels for 3.5 years did not significantly reduce major cardiovascular events compared with standard treatment. Moreover, it increased mortality [35]. In order to test their hypothesis that HbA1c variability in intensive therapy would be associated with all-cause mortality, Sheng et al. performed a post hoc analysis of the ACCORD trial. HbA1c variability is a significant predictor of all-cause death, as shown by the research [36]. Recent studies have found that HbA1c variability, expressed as intrapersonal SD or coefficient of variance, is also an independent risk factor for diabetic complications in type 2 patients [37]. It may even be a more reliable measure of glycemic control than the mean HbA1C in type 2 DM [12, 38]. In the short term and in a limited number of measurements, the HbA1c value may be influenced by various factors, such as anemia treatment, uremia, and laboratory errors. For this reason, analysis of HbA1c variability calculated separately for each patient over at least two years and more than two measurements may be more valuable in detecting diabetes complications such as DNP. In our study, HbA1c variability was found to be a risk factor in DNP-NDNP discrimination. A sudden rise in HbA1c levels may indicate uncontrolled diabetes, while a sudden decrease in HbA1c levels may be due to reduced insulin degradation and renal gluconeogenesis caused by CKD associated with DNP. The higher frequency of patients with stable HbA1c levels in the NDNP group can therefore be interpreted in this context. Currently, there are no studies investigating HbA1c variability between DNP and NDNP. Therefore, these findings in our study reveal the importance of HbA1c variability and encourage its clinical use.
While proteinuria levels remained constant, patients diagnosed with DNP had a considerably higher incidence of albuminuria when the patient with minimal change disease (MCD) was eliminated. This may be the result of people with NDNP having a greater urine transit of non-albumin proteins, specifically multipl myeloma and MN. Although a lack of significance was shown in the multivariate analysis, it may be more advantageous to measure urine protein and albumin concurrently. A biopsy is recommended for diabetic patients who have been prediagnosed with NDNP and are experiencing rapid and growing proteinuria [30]. The length of DM has been shown as a strong predictor of DNP in prior research [39]. The effect of proteinuria duration, however, has received little research attention. We found that the duration of proteinuria in NDNP patients was considerably shorter in our study. In the multivariate analysis, however, we failed to identify any significant differences. It is critical to emphasize that clinicians may still find the duration of proteinuria to be a valuable criterion.
Our study had limitations. As our study was based on information collected from electronic medical records, we encountered missing data and difficulties in accessing the results of some previous laboratory tests, which are common limitations of retrospective studies. The main limitation of our findings relates to the retrospective design and the possibility of selection bias. The traditional criteria used to decide on KB in the diabetic patient may have increased or decreased the effectiveness of the parameters studied as risk factors. A reliable way to address this issue would be prospective studies in which renal biopsies are performed in all patients with diabetes to determine the predictive effects of clinical factors and/or laboratory tests. However, selection bias appears to be present in the few prospective studies published to date with discordant results. Finally, although pathological findings overlapped in some patients, patients were divided into two groups based on the histological findings seen in the foreground. In some publications, patients were classified as diabetic, nondiabetic and patients with combined nephropathies. In our study, it was preferred to divide the patients into two groups both in order not to reduce the statistical power by dividing the groups and because the prominent pathological findings and clinical findings could be distinguished despite overlap.
In conclusion; the absence of diabetic retinopathy, non-HbA1c variability, female gender and serology positivity were found to be risk factors in predicting non-diabetic nephropathy. In conjunction with conventional variables, gender and HbA1c variability were identified as significant determinants in distinguishing DNP from NDNP. By using these factors, the current selection criteria for renal biopsy in patients with DM can be improved, and it is possible to determine which individuals might benefit from particular treatment measures to prevent renal dysfunction caused by NDNP or DNP. Prospective studies must validate the efficacy of HbA1c variability as shown in our research.