Acute kidney injury in hospitalized patients who underwent percutaneous kidney biopsy for histological diagnosis of their renal disease

Acute kidney injury (AKI) is a risk factor for the development of chronic kidney disease (CKD) and chronic end-stage renal disease (ESRD). AKI is also associated with increased short- and long-term mortality rates [1]. AKI is a complex and heterogeneous clinical syndrome observed mainly in elderly patients with several comorbidities. These patients are usually admitted to intensive care units (ICU), and AKI is primarily attributed to sepsis in such cases [2]. Percutaneous renal biopsy is usually contraindicated in these patients to prevent complications, and they are usually diagnosed with “acute tubular necrosis”.

In contrast, some kidney diseases such as glomerulonephritis and tubulointerstitial nephritis, among other such clinical conditions require histopathological analysis of kidney tissue specimens to accurately determine diagnosis, disease activity and chronicity [3, 4]. Notably, most epidemiological studies have investigated patients with AKI have not included this patient population in the analysis. For example, a meta-analysis analyzed the worldwide incidence of AKI observed that studies primarily included patients admitted to the ICU and those who underwent cardiac surgery [5].

Therefore, we investigated the prevalence and risk factors associated with AKI in hospitalized patients who underwent kidney biopsy for histopathological diagnosis.

The study included 267 patients, of which 44 (16.5%) patients were diagnosed with ESRD during admission and were excluded from the study. Of the remaining patients (N = 223), 140 (62.8%) were diagnosed with AKI during hospitalization. No significant intergroup difference was observed in sex (men, 68 [48.6%] in the AKI vs. 41 [49.4%] in the non-AKI group, p = 0.508). Patients with AKI were older than patients without AKI (41.2 ± 17.7 vs. 35.3 ± 14.4 years, p = 0.03) with similar BMI (25.7 ± 5.4 in the AKI vs. 24.5 ± 4.0 kg/m2 in the non-AKI group, p = 0.291). Most patients were white in both groups (87 [62.8%] in the AKI vs. 51 [61.5%] in the non-AKI group, p = 0.482).

No statistically significant differences were observed between the AKI and non-AKI groups, respectively in the percentage of patients with hypertension (64 [45.7%] vs. 30 [36.1%], p = 0.096), patients with CVD (9 [6.4%] vs. 3 [3.6%], p = 0.279), and patients with a history of cancer (5 [3.6%] vs. 1 [1.2%], p = 0.273). DM was more prevalent in the AKI group (17 [12.1%] vs. 3 [3.6%], p = 0.023).

A lower percentage of patients in the AKI group used ACEIs/ARBs (57 [40.7%] vs. 45 [54.2%], p = 0.038). Additionally, a larger number of patients in the AKI group received a potentially nephrotoxic antibiotic (16 [11.4%] vs. 2 [2.4%], p = 0.012). No statistically significant difference was observed between the AKI and non-AKI groups with respect to the use of diuretics (42 [30.0%] vs. 32 [38.6%], p = 0.130), statins (31 [22.1%] vs. 23 [27.7%], p = 0.227), NSAIDs (19 [13.6%] vs. 9 [10.8%], p = 0.347), and corticosteroids (34 [3.0%] vs. 28 [33.7%], p = 0.091). No patient underwent contrast examination during hospitalization.

The primary indications for renal biopsy were nephrotic-range proteinuria or nephrotic syndrome (73 [52.1%] in the AKI vs. 51 [61.4%] in the non-AKI group, p = 0.048), renal dysfunction of undetermined etiology (23 [16.4%] in the AKI vs. 5 [6.0%] in the non-AKI group, p = 0.001), and hematuria with non-nephrotic-range proteinuria (11 [7.8%] in the AKI vs. 11 [13.2%] in the non-AKI group, p = 1,000). RPGN was an indication exclusively in the AKI group (18 [12.9%] vs. 0, p < 0.001).

Primary renal disease was the most common type of disease observed in both groups (86 [61.4%] in the AKI vs. 67 [80.7%] in the non-AKI group, p = 0.125). The primary diagnoses were focal segmental glomerulosclerosis (FSGS) (24 [17.1%] in the AKI vs. 15 [18.0%] in the non-AKI group, p = 0.150) and minimal change disease (MCD) (21 [15%] in the AKI vs. 15 [18%] in the non-AKI group, p = 0.317). Secondary diseases predominated in the AKI group (54 [38.6%] vs. 16 [19.3%], p < 0.001). SLE was the main cause of secondary renal disease, with classes III and IV being the most prevalent (Table 1).

Laboratory data and measurements of renal length are presented in Table 2. Statistically significant differences were observed between the AKI and non-AKI groups with respect to the following variables: serum hemoglobin (11.7 ± 2.7 vs. 13.4 ± 2.1 g/dL, p < 0.001), serum sodium (137 ± 3.7 vs. 138 ± 2.6 mEq/L, p = 0.047), serum urea (76.0 [55.0–115.0] vs. 57.0 [39.0–77.3] mg/dL, p < 0.001), serum bicarbonate (21.4 ± 5.5 vs. 25.7 ± 4.0 mEq/L, p < 0.001), and serum HDL-c (43.5 ± 16.2 vs. 56.3 ± 28.5 mg/dL, p < 0.001). No statistically significant intergroup difference was observed in renal length.

Table 3 shows an intergroup comparison of serum creatinine levels across different time-points in the study. We observed that the AKI group showed higher serum creatinine levels than the non-AKI group (p < 0.001) across the entire follow-up period. Compared with peak levels during admission, we observed a decrease in serum creatinine levels in the AKI group after 12 months (p < 0.05).

Among patients diagnosed with AKI (62.8%), 91 (65.0%) were classified as showing stage I disease, 19 (13.6%) as stage 2, and 30 (21.4%) as stage 3 disease.

Dialysis was performed in 19 (8.5%) patients in the AKI group and of these 19, 9 (4.0%) remained dependent on dialysis therapy at the time of discharge. All patients were dialysis-free at the end of 12 months. Notably, 8 patients (3.6%) died and 5 of them died of infections during hospitalization. Table 4 shows the multivariate analysis of factors associated with the development of AKI.

This study showed a high prevalence of AKI in hospitalized patients who underwent kidney biopsy for renal disease primarily caused by glomerulonephritis. Most patients showed primary renal disease, and the most common indications for renal biopsy were nephrotic-range proteinuria or nephrotic syndrome. Among the primary diseases, FSGS showed the highest prevalence. Among all secondary diseases, lupus glomerulonephritis was the most common.

Epidemiological studies investigating AKI should necessarily describe patients’ baseline serum creatinine levels. In this study, we defined the lowest serum creatinine level during hospitalization as an estimate of baseline renal function, given the lack of previous values. Retrospective baseline serum creatinine levels (assuming GFR = 75 mL/min/1.73 m²) based on KDIGO criteria may miss the occurrence of AKI in the community [7]. Moreover, our study group comprised patients predominantly with glomerulonephritis, which tends to demonstrate an insidious onset and progression. Occasionally, in a few cases, AKI can be diagnosed in clinical practice only by analyzing temporal changes in serum creatinine levels. Our findings concur with this observation in that progressive improvement in renal function occurred in the AKI group throughout the study period.

We could not rule out the existence of a pre-renal component of AKI, particularly in patients with less severe AKI (stage 1). Despite the long-term protective effect of ACEI/ARB in renal diseases in patients with proteinuria, their use may cause acute and reversible renal dysfunction, particularly when used concomitantly with diuretics and in patients with other conditions causing hypovolemia [8, 9]. Interestingly, we observed a higher percentage of patients using ACEI/ARB in the non-AKI group. In contrast to data reported in the literature, the effect of AKI on glomerular hemodynamics (decreased filtration pressure) and the potential negative effect on renal function was not evident in the present study [8, 10]. For example, a recent study investigating patients with hypertension using ACEI/ARB and diuretics showed AKI was associated with CKD and poor cardiac performance [11]. Another study involving new users of ACEI/ARB reported a low incidence of AKI and showed AKI was more likely to be associated with the individual clinical characteristics than with the use of the medication itself [12].

Proteinuria and low serum albumin levels are risk factors for AKI [13, 14]. This is particularly evident in patients diagnosed with MCD, notably elderly patients with hypertension (showing arteriolar nephrosclerosis) and patients with more severe degrees of nephrotic syndrome [15]. In our cohort, no significant differences were observed in 24-h proteinuria, serum albumin levels, and the number of patients diagnosed with major proteinuria-causing diseases (FSGS, MCD and Membranous Glomerulonephritis) between the AKI and non-AKI groups. Previous studies have shown that DM is also an independent risk factor for AKI, mainly secondary to microvascular dysfunction [16, 17]. Although the AKI group included a greater number of patients with DM, it was not an independent risk factor in our study. Our study included younger patients than those studied previously, and owing to the multiple disease diagnosed, AKI could be attributed to other mechanisms as well.

Pre-existing kidney dysfunction serves as an independent risk factor for AKI in conditions such as sepsis and contrast-induced nephropathy, as well as in patients undergoing cardiac surgery or solid organ transplantation, and in patients admitted to the ICU [18‐22]. This factor was also relevant in our study. Patients with CKD are at a higher risk of AKI secondary to the role of several etiopathogenetic contributors such as activation of transforming growth factor beta, action of hypoxia-inducible factors, mitochondrial and endothelial dysfunction, oxidative stress, chronic inflammation, and alterations in renal blood flow autoregulation observed in this patient population [23‐25].

Anemia and AKI are occasionally associated. A few studies have shown anemia serves as an independent risk factor for AKI [26‐28]. Previous studies have shown anemia was a risk factor for contrast-induced nephropathy in patients undergoing coronary angiography [29], postoperatively in patients undergoing hip arthroplasty [30], in patients admitted to the ICU [31], and postoperatively in patients undergoing cardiac surgery [32]. In these cases, AKI associated with anemia could be attributed to a reduction in the oxygen supply to renal tissues and the consequent aggravation of pre-existing ischemia occurring in hospitalized patients [32‐34].

In our study, we observed a higher level of serum hemoglobin was associated with a lower incidence of AKI. Several studies have shown an association between anemia and AKI; however, it is unclear whether higher levels of serum hemoglobin could reduce the incidence of AKI. It could be deduced that early intervention, for example, through the use of recombinant human erythropoietin, could reduce the risk of AKI in specific diseases, such as in glomerulonephritis. Previous studies have shown the role of erythropoietin in patients admitted to the ICU and in those undergoing cardiac surgery was controversial and in a few cases the benefit was independent of hematocrit and serum hemoglobin levels [35].

Altered lipid metabolism is common in patients with renal disease, particularly in those with glomerulonephritis and nephrotic syndrome. Hypercholesterolemia and hypertriglyceridemia are the most common findings in such cases; however, low serum HDL-c levels may occur. Resolution of nephrotic syndrome usually tends to reverse these changes [36].

HDL-c possesses antioxidant properties, thereby reducing endothelial damage and the risk of atherosclerosis [37]. The MDRD study showed a low serum HDL-c level was an independent risk factor associated with a faster decline in GFR. In experimental AKI models, HDL-c showed anti-inflammatory actions and reduced ischemia and reperfusion injury [38, 39]. Roveran Genga et al. showed low levels of serum HDL-c in patients with sepsis were associated with a higher incidence of sepsis-associated AKI [40]. Another study reported by Smith et al. demonstrated higher levels of preoperative serum HDL-c decreased the incidence of AKI after cardiac surgery [41]. Arora et al. observed low levels of serum HDL-c were associated with AKI after revascularization surgery for chronic limb ischemia [42].

In our cohort, we observed an association between higher levels of serum HDL-c and lower incidence of AKI. We inferred perhaps intensive drug therapy for dyslipidemia in addition to exercise and dietary strategies could improve patients’ lipid profile and consequently reduce the incidence of AKI, particularly in patients with glomerulonephritis.

The following are the limitations of our study: [1] The lack of outpatient baseline serum creatinine values is the main limitation, since some cases of non-recovering AKI could be missed. [2] Because it is a retrospective and single-center study, a bias cannot be ruled out. [3] Histopathological heterogeneity and dropouts during follow-up are other limitations. [4] We do not have information regarding the treatment for each specific disease diagnosed by biopsy. Despite these limitations, the high prevalence of AKI observed in these subjects is concerning because this patient population is often excluded from epidemiological studies investigating AKI.

We observed a high prevalence of AKI in hospitalized patients who underwent kidney biopsy to investigate their renal disease, particularly glomerulonephritis. Higher levels of hemoglobin and serum HDL-c were associated with lower risks of AKI.