1. Introduction
Obesity is a public health problem with significant implications for individuals and societies, and its worldwide prevalence is projected to continue to increase [
1]. Obesity can lead to kidney injury through direct and indirect mechanisms and is an independent risk factor for the development and progression of chronic kidney disease (CKD) [
2,
3]. The prevalence of obesity-related glomerulopathy (ORG), which is characterized by proteinuria, glomerulomegaly, slowly progressive glomerulosclerosis and renal functional decline, rises with the prevalence of obesity [
4].
The pathways through which obesity progresses to CKD are not well understood. Besides glomerular hyperfiltration, which has been hypothesized as the main mechanism linking metabolically unhealthy obesity to renal injury [
5], studies have shown that in the setting of obesity there is also an excess tubular sodium reabsorption [
6], an overactivation of the renin-angiotensin-aldosterone system [
7] and an increased activation of the renal sympathetic nervous system, which are also likely to play a role in the disease pathology. Individuals with obesity have been found to have a significantly higher kidney weight [
8], which has been attributed to compensatory hypertrophy of individual nephrons, but also to ectopic lipid accumulation in the kidney, the “fatty kidney” [
9]. Lipid components have been shown to accumulate intra- and extracellularly in patients with obesity [
8], contributing to structural and functional changes in mesangial cells, podocytes and proximal tubular cells [
9], and are associated with ensuing obesity-related kidney disease. There is a gap in the knowledge concerning the pathways of glomerular versus tubular lesion in obesity.
Bariatric surgery, an effective option for treatment-refractory obesity, requires extensive peri-operative assessment of patients [
10]. Despite bariatric patients’ high risk of renal dysfunction given obesity-related glomerulopathy and other mechanisms, kidney function characterization in most surgical treatment of obesity programs is usually limited to pre-surgical serum urea and creatinine measurements. There are currently no practice guidelines regarding how to evaluate kidney function in patients pursuing bariatric surgery. In this study, we performed a detailed characterization of kidney function markers in patients with obesity attending a public academic center for surgical treatment of obesity. We aimed to study how common renal involvement is in bariatric surgery candidates and to evaluate clinical phenotypes with a focus on biomarkers of glomerular versus tubular lesion.
4. Discussion
Obesity has shown to play an important role in CKD development and progression independently of its etiology [
9]. Qualification criteria for bariatric surgery include BMI and obesity-related comorbidities, but kidney dysfunction is most often overlooked. In this study we aimed to depict the prevalence of kidney injury markers in candidates to bariatric surgery without diabetes or other known chronic inflammatory diseases.
Importantly, several methodological issues were considered. Measurements were based on a 24 h urine collection; creatinine clearance was not scaled for BSA to avoid the systematic underestimation of GFR that happens with BSA indexing, particularly noticeably in the population of patients with obesity [
14,
15]. A severe increase in creatinine clearance (more than 170 mL/min) was present in 27.6% of patients and it was higher than 140 mL/min in more than half of the population. These findings are consistent with previous studies showing that renal hyperfiltration is a characteristic feature in obesity-related renal disease [
16,
17] that appears in the early stages of higher than normal adiposity and even for BMIs under 30 kg/m
2 [
18,
19]. Another important observation was the weak correlation between measured 24 h creatinine clearance and the estimated GFR using the 2021 CKD-EPI Creatinine (145 ± 44 mL/min vs. 104 ± 14 mL/min/1.73 m
2) in our patient population, which supports findings from recent studies suggesting this equation might underestimate GFR in patients with obesity [
20]. As expected, both proteinuria and albuminuria were higher in patients with higher creatinine clearances [
21,
22]. Several factors have been shown to contribute to increased albumin excretion in patients with obesity [
4,
23,
24]. However, there is a gap in the knowledge related to renal tubular injury and to the qualitative composition of proteinuria. Beyond the commonplace of glomerulomegaly and hyperfiltration, obesity-related kidney disease may have tubule-centric mechanisms of injury, as it occurs in diabetes mellitus [
25].
Indeed, in our study, the rate of patients with proteinuria higher than 150 mg/day was 22.4%, and the rate of patients with albuminuria higher than 30 mg/day was 14.6%, findings that are consistent with previous studies in populations with obesity, but with higher proportions [
26,
27,
28,
29] this supports that pre-operative assessment should include a panel of renal biomarkers.
It should be highlighted that, to reduce the risk of bias, our study did not include patients with diabetes nor patients with prediabetes under metformin treatment. Nevertheless, albuminuria levels above 30 mg/day were associated with higher levels of glycated hemoglobin than in patients with albuminuria lower than 30 mg/day, and multivariate analysis showed glycated hemoglobin to significantly impact albuminuria levels. Prediabetes has been shown to be modestly associated with incident CKD [
30] and studies have shown that it may have a significant role in the development of albuminuria and hyperfiltration [
31,
32,
33]. Other studies support that some of the pathologic changes that are characteristic of diabetic kidney disease may be already present in the prediabetic state [
34,
35]. In our patient cohort, prediabetes was present in 40 (20.8%) patients and, from these, 25% had albuminuria, while among the euglycemic population (n = 152), this figure was of 11.8%. It has been suggested that there might be a synergistic effect between obesity-associated renal changes and prediabetes-associated renal changes, lowering the thresholds for the occurrence of kidney injury that is common to both pathologies in this patient population [
36,
37].
Several lipid abnormalities were shown to correlate with clinically relevant albuminuria. Lower levels of HDL cholesterol and higher levels of LDL cholesterol, VLDL cholesterol, triglycerides, non-HDL cholesterol and higher atherogenic index were associated with albuminuria higher than 30 mg/dL. In contrast to albuminuria, apart from the atherogenic index, these differences did not correlate with significant proteinuria or significant creatinine clearance. Dyslipidemia has also been associated with GFR decline [
38] and increased incidence of albuminuria [
39] in several previous studies. Alterations in lipid metabolism are being increasingly recognized as important factors in ectopic lipid accumulation in the kidney, known as “fatty kidney”, as well as increased oxidative stress, inflammation, and fibrosis [
4]. In ORG, altered lipid metabolism has been shown to promote the deposition of triglycerides and cholesterol esters in the kidney, leading to maladaptive changes in renal cells [
40,
41]. Ectopic lipid accumulation compromises glomerular integrity and has been associated with albuminuria [
42] and fibrosis [
4], with changes in mesangial cells [
9] and in podocytes [
43].
Elevated serum uric acid levels have been shown to predict the development of albuminuria in type 1 diabetes [
44], but the causal relationship between uric acid levels and CKD or the benefits of treating hyperuricemia in CKD are still a matter of debate [
45,
46]. In our study, uric acid levels were also significantly higher in patients with albuminuria over 30 mg/day. As in the previously described lipid abnormalities, this difference was noticed to be associated with increased albuminuria, but not with increased proteinuria. Higher uAPR was also associated with significantly higher uric acid levels.
The average uAPR in our patient population was 0.18 ± 0.12, with most patients showing levels between 0.1 and 0.3 (n = 114, 59.4%). To the best of our knowledge, there are no previous reports on uAPR in patients with obesity. As abovementioned, several variables were shown to have a distinctive impact on the levels of proteinuria and albuminuria. Several studies have proposed that the origin of urinary proteins can be inferred by measuring the uAPR and, in kidney diseases with significant proteinuria, a uAPR of <0.4 can predict primary tubulointerstitial, with a sensitivity and specificity of 88% and 99%, respectively [
47,
48,
49]. In patients with obesity, these qualitative differences in protein excretion between patients (
Figure 1) may reflect different types of early kidney involvement which may be related to the patient’s distinct metabolic signatures in the setting of obesity. We hypothesize that patients with significant lipid abnormalities, hyperuricemia and higher levels of glycated hemoglobin, might have a more glomerular-centered kidney injury in early disease and, therefore, present with higher rates of albuminuria, while, patients without these metabolic characteristics may develop primarily tubulointerstitial changes in early course of disease, with excretion of higher percentages of other urinary proteins, and only in later stages progress with glomerular dysfunction. While albuminuria is a marker of glomerular injury, non-albuminuric proteinuria may reflect tubular excretion and tubulointerstitial inflammation/injury [
50,
51]. Since albuminuria and proteinuria are interconnected, the biomarker urinary albumin-to-protein ratio (uAPR) is an opportunity to evaluate the proportional strength of such injury pathways. The tracks of tubule-glomerular and tubule-interstitial crosstalk in the pathophysiology of ”fatty kidney”, as well as the contribution of local fat to renal disease progression, deserve further investigation.
Several significant differences were noticed in renal function parameters between male and female participants. Male patients had significantly higher rates of proteinuria, albuminuria, uAPR and creatinine clearance compared to female patients. Obesity-related renal function differences between males and females have been reported in previous studies, showing conflicting results [
52,
53]. In our study these differences may reflect the uneven gender distribution of patients undergoing bariatric surgery, with men with obesity being less prone to seek earlier medical aid or consider surgical treatment [
54].
This study allowed us to demonstrate that a significant proportion of bariatric surgery candidates present renal hyperfiltration along with clinically relevant albuminuria and proteinuria. Moreover, the data suggest that in obesity-associated kidney disease, tubulointerstitial injury precedes glomerulopathy, which is an important addition to preexisting knowledge. The significant prevalence of kidney injury we observed in our study shows kidney function should be evaluated in bariatric surgery candidates, something which is not routinely performed in most bariatric surgery programs. Patients with abnormal kidney function may deserve a long-term follow-up targeting renal function, as is currently advised for patients undergoing bariatric surgery with other comorbidities including diabetes, obstructive sleep apnea syndrome, or hypertension.
The main limitation of our study is the fact that it was a cross-sectional study. Follow-up of these patients could provide valuable information on how renal parameters, namely creatinine clearance and hyperfiltration, evolve according to the obesity/metabolic status and specifically after bariatric surgery. Even though our sample size was limited, it is similar to other studies in this area and was sufficiently powered to depict robust statistical differences. As important strengths, our study evaluated renal function parameters through a 24 h urine collection, which increased our ability to analyze small variations in protein excretion, as opposed to the alternative use of a spot albumin-creatinine ratio, which is a good screening test for microalbuminuria, but performs poorly in determining quantitative albumin excretion rate or total proteinuria. Additionally, we used a non-indexed creatinine clearance calculation, and collected an extensive panel of biochemical variables, which allowed us to conduct a detailed characterization and broader analysis of our patient population.
In conclusion, a significant proportion of patients with obesity have clinically relevant levels of albuminuria, proteinuria and hyperfiltration, as early markers of renal lesion. In patients with obesity, the concomitant presence of prediabetes, lipid abnormalities and hyperuricemia were associated with clinically relevant levels of albuminuria, but not with proteinuria, which could translate into different types of kidney compartment injury. Our results highlight that traces of tubulointerstitial injury also occur in obesity-related kidney disease, together with or even in advance of glomerular damage, which warrants further investigation.