Early studies showed that chronic kidney disease (CKD) has an approximate prevalence of 9.1%.1 In Spain, the study EPIRCE (Epidemiology of CKD in Spain) found a prevalence of CKD-3 of’ 9%, which increased to 21.4% in people over 65 years of age.2 The EROCAP study estimated the incidence of CKD-3 in primary care clinics in Spain is 21.3%.3 Subsequently, data from the ENRICA study (Study of nutrition and cardiovascular risk in Spain) reported a national prevalence of 15.1%.4 This increase in the prevalence of CKD could be partially explained by the high rates of arterial hypertension (HTN) and diabetes mellitus (DM).5

Despite the improvement in our understanding of the pathophysiological mechanisms and therapeutic strategies of CKD, a high percentage of patients will experience progression of CKD to more advanced stages that will require renal replacement therapy (RRT).

Although the rate of progression is influenced by non-modifiable risk factors, such as race and age, there are other factors that could be modifiable, such as DM, obesity, hypertension, smoking and dyslipidemia, which may have an influence on such progression.3,6–8 In this regard, improving our knowledge of all these factors may be crucial from the pathophysiological and therapeutic objective point of view.

In recent years, numerous studies have revealed the existence of new factors that may play a significant role in the progression of CKD.8–11 However, there are no studies designed with specific criteria to establish the role of these new progression factors in an integrated manner that includes clinical, biochemical and genetic factors.

CKD is not only associated with the inexorable deterioration of renal function, but also entails an increased risk of cardiovascular (CV) disease and mortality compared to the general population.12 Furthermore, the possibility of death from CV factors is greater than that of starting RRT.7 Although these facts are well-known, the increase in morbidity and mortality associated with CKD has not been well clarified, especially in the early stages of CKD.2,13–19 In the present study we have sought to examine which factors may be associated with the progression of renal and CV damage in patients with CKD stage 3, defined according to the criteria of the Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines, analyzing possible differences between patients with and without DM.

The primary objective of the study was to assess the risk factors associated with the progression of renal damage in patients with CKD-3 KDOQI, with special interest in verifying whether there were differences between the population with and without DM. The secondary objectives were to describe the hospitalization and mortality data and the related factors.

In our study, the factors that favor the progression of renal damage in CKD patients were related to parameters of bone and mineral metabolism in patients with DM (lower baseline calcium levels and higher iPTH), and probably with the presence of hypertension in patients without DM (p level close to statistical significance: p = 0.07). Traditional risk factors, such as diabetes, obesity, smoking and dyslipidemia, were not significantly associated with the progression of renal damage in the multivariate analysis. Although smoking (both current and past smoking), a history of previous cerebrovascular disease, and hypertriglyceridemia were associated with the progression of renal damage in the DM population in the bivariate analysis, they lost their significant association with the progression of renal damage as independent variables in the multiple regression analysis.

Some recent studies have attempted to elucidate which factors have the greatest impact on the development and progression of renal damage. For example, in our MERENA study,22 data on 1,156 patients with CKD stage 3 and 4 showed that several of the CV risk factors (HTN, coronary artery disease, cerebrovascular disease, heart failure, or peripheral vascular disease) are very frequently present and are more prevalent in the CKD population, regardless of the presence of DM. When we analyzed mortality after five years of follow-up in this study, it was found that hospitalizations were more frequent in the group of patients with DM (44% due to CV causes) and that in 44% of patients the CKD progressed to CKD stage 5, requiring RRT (35.9% by hemodialysis and 8.1% by peritoneal dialysis), but mortality was not significantly higher in the group of patients with DM compared to patients without DM (17.2% vs. 15.4%; p = 0.17) (data presented at the S.E.N. Congress in Seville, October 15–18, 2011).23 Similar data have been described in the PECERA study, which included 995 patients with CKD-KDOQI 4, 35% of them with DM, and with a follow-up of 3 years. In this study, the main cause of hospitalizations was CV diease, the most important being heart failure. Of these patients, 32% required RRT (27% by hemodialysis, 4% by peritoneal dialysis and 1% received a kidney transplant). During the follow-up, 15% of them died (46.2% due to CV causes).24

We believe it is important to point out that the percentage of progression in these two previously mentioned studies22,24 and in the one being reported now, although high, is not that excessively high when compared with older data from the literature,. This could be due in part to the insistence of guidelines and consensus documents on a greater control of progression factors such as hypertension, dyslipidemia, anemia or bone and mineral metabolism parameters. In this context, it is also important to note that the patients in our study did not benefit from the new antihyperglycemic drugs with potential nephroprotective and cardioprotective effects, such as SGLT2 or GLP-1 inhibitors, which were not yet introduced for clinical use in the those years when our patients were included in the study.

In the NEFRONA study, which included 4,137 patients with CKD stages 3–5 without previous CV history and 843 control subjects without known kidney disease, the factors associated with worse CV prognosis were the progression of carotid atheroma plaque, smoking, the presence of DM or systolic hypertension, low levels of 25-OH vitamin D, as well as the absence of treatment with phosphorus binders.25 A subsequent subanalysis of the same study in a cohort of 1,152 patients showed that the factors for progression of kidney damage were higher levels of P in CKD- 3 to 5, decreased levels of 25-OH vitamin D, and i-PTH levels higher than 110 pg/mL in CKD 4 and 5.26

In the RIACE (Renal Insufficiency and CV Events) study, a multicenter, observational study that included 15,733 patients with type 2 DM, the most striking data were the high number of patients without albuminuria (the predominant phenotype), who also showed a high prevalence of CV risk factors, the poor concordance between the stage of CKD and the presence of diabetic retinopathy, as well as the administration of oral hypoglycemic agents not recommended with eGFR <30 mL/min/1.72 m2.27

A meta-analysis published in 2016 evaluated twelve studies of CKD progression with more than 13,000 patients and showed an average progression to CKD-5 of 40 events/1,000 patients/year, a mortality of 41 episodes/1,000 patients/year, with CV episodes of 29/1,000 patients/year. It should be noted that the authors observed a great variability of results between different countries.28

We find it interesting to highlight that the estimation of GFR using equations that include plasma creatinine may show variability in the the classification y the stage of CKD in these patients. When we began the present study, the equation available in the laboratories of the different centers was the MDRD in its different formulas, with 4, 6 and MDRD-IDMS as the most used, the latter being the one selected to classify the patients included. New equations have been subsequently introduced, the most used being the CKD-EPI. We therefore had to recalculate the GFR both at baseline and throughout the study and at three years of follow-up.

As a result of this re-estimation, 19 patients were reclassified as CKD-4. No other changes were detected in the reclassification of CKD stage. This data may be important when estimating the degree of alteration in renal function and the possible consequences of the use of medications which dose is modified according to the GFR, to avoid possible nephrotoxicity or other adverse effects. However, the analysis of patients with progression renal damage, as well as the risk factors associated with it, was carried out in the multivariate analysis, and no differences were detected with the use of the MDRD and the CKD-EPI formula.

We must recognize several limitations in our study. The first is that we did not manage to include the number of patients initially proposed due to difficulties in recruitment, so the closing period had to be extended significantly. Second, the fact that we did not reach the proposed “n” value could have influenced the failure to demonstrate statistical significance in the multivariate analysis of some of the progression factors proposed in the bivariate analysis. This same fact could explain the lack of statistical significance of albuminuria and proteinuria in the multivariate analysis, a fact aggravated by the loss of these two results in some patients.

Although the study has these limitations, especially the lack of detection of risk factors in the clinical variables, this observation is not infrequent, especially in patients from large studies, since they usually have adequate control of risk factors (hypertension, dyslipidemia, etc.) and the event rate is usually low. A recent example is the EMPA-KIDNEY study,29 which analyzed the effect of empagliflozin in 6,609 patients with CKD and risk of progression of kidney damage, including diabetics and non-diabetics, where 65.8% had CKD stage 3 and 34.2% had CKD stage 4. The mean eGFR was 37.4 ± 14.4 mL/min/1.73 m2 (in our study the mean eGFR was 43.9 ± 7.9 mL/min/1.73 m2 and the percentage of normo, micro and macroalbuminuria was 52%, 28% and 20%, very similar to the data from our study: 41.9%, 31.8% and 26.3%). The rate of events related to the combined primary endpoint (reduction of eGFR ≥40%, reaching an eGFR of 10 mL/min/1.73 m2 or starting dialysis) was only 18% in patients with proteinuria and 6% and 7% in patients with normo- and microalbuminuria. This may be a common aspect with our study. That is why the search for new biomarkers in the samples stored in the biobank may become important in the near future.

Despite these limitations, we believe that the strengths of our study lie in the fact that it included a broad representation of patients all over Spain (from14 of the 17 autonomous communities), the follow-up period was relatively long (three years), and the sample included a significant number of patients with DM. The fact that that patients did not receive treatment with SLGT2 inhibitors or GLP-1 RAs, may be used to compare the evolution of our patients as a control group for future studies that evaluate the progression of kidney damage and CV damage with that of other patients treated with these drugs, versus those treated as classically with renin-angiotensin-aldosterone system blockers, in addition to standard treatments. An important additional strength is the availability of biological samples in the biobank, particularly urine and cell, which may allow us to evaluate new biomarkers that may help to identify factors related with progression of kidney damage and establish new therapeutic targets.

A good example of the benefit of saving biological samples for future analysis is the study recently published in Nephrology, Dialysis and Transplantation, in which patients from the PROGRESER cohort were analyzed together with patients from the PRONEDI study.30 The authors identified the urinary protein Dickkopf-3 as a new marker of progression and death in patients with CKD stages 2 and 3.31 This study highlights the importance of having biological samples in translational research that assess not only aspects related to traceability, but also the connection with clinical characteristics of the patients, which is fundamental for future research.32

We believe that the progression of kidney damage must be interpreted in a multifactorial context. It is necessary to search for new biomarkers, different from the traditional ones, in order to establish therapeutic strategies able to prevent the progression of CKD.

Our study has not revealed new factors of progression of kidney damage in relation to the factors already known. The study has shown in the multivariate regression analysis that factors such as hypertension may be related to the progression of kidney damage in the population with non-diabetic CKD, and that low calcium levels and high intact PTH levels were correlated with progression of kidney damage in patients with DM and CKD-stage 3 KDOQI. We have not found significant differences in terms of the progression of CKD in patients with and without DM, nor in the factors related to such progression.

The project received funding from Laboratorio Esteve, Barcelona, and the SENEFRO Foundation.