Introduction
Diabetes mellitus (DM) is a public health problem encountered in various countries, including China. According to the International Diabetes Federation Atlas of 2012, the estimated diabetes prevalence was 371 million or 8.3% of the world’s adult population [
1]. Diabetes and prediabetes prevalence in a representative sample of Chinese adults were approximately 11.6% and 50.1%, respectively. Projections by sample weighting suggest that up to 113.9 million Chinese adults may suffer from diabetes and 493.4 million may experience prediabetes [
2]. As such, China has become the world’s center for DM.
Long-term DM triggers vascular changes and dysfunction, and diabetic complications are the major causes of morbidity and mortality in patients with diabetes. Diabetic kidney disease (DKD) is the most common cause of chronic kidney disease (CKD) and terminal end-stage renal disease [
3].
Kidney disease caused by diabetes is traditionally termed diabetic nephropathy (DN) [
4]. This disease is considered present if a patient presents the following conditions: suffered from diabetes for at least 7–10 years, exhibited demonstrable diabetic retinopathy (DR), showed a history of microalbuminuria without sudden-onset heavy proteinuria, did not manifest hematuria, possessed a kidney that was not abnormally small, and did not experience other renal diseases [
5]. New terms describing kidney diseases attributable to diabetes were further introduced by the National Kidney Foundation and Kidney Disease Outcomes Quality Initiative (NKF-KDOQI) in 2007 [
6]. In these guidelines, the presumptive or clinically diagnosed kidney damage caused by diabetes is defined as DKD and the definitively diagnosed biopsy-proven diabetic kidney damage is called diabetic glomerulopathy (DG). A patient with diabetes and kidney damage may also be diagnosed with one of the three pathological conditions: simple DN (also termed DG), nondiabetic renal disease (NDRD), or DN combined with NDRD. However, we have yet to determine whether the type of kidney damage can be determined using clinical data alone. The prognosis of patients with DN also significantly influences their survival and quality of life. Further studies have yet to reveal whether the clinical stage or pathological classification of DN can precisely predict its prognosis. In the present study, the clinical data of patients with biopsy-proven simple DN treated from January 1, 1998 to June 30, 2016 at the Nephrology Department in China–Japan Friendship Hospital in Beijing, China were analyzed to define the detailed clinical manifestations of patients with DN. We concluded that the clinical manifestations of this disease were nonspecific, and previous clinical DN diagnoses consequently remained questionable.
Methods
Patient selection
We examined 590 patients who suffered from type 2 DM (T2DM) and underwent renal biopsies in the China–Japan Friendship Hospital from January 1, 1998 to June 30, 2016. Patients with simple DN accounted for 18.8% (111 of 590) of these patients. T2DM was diagnosed using the criteria of the World Health Organization (WHO) [
7]. The patients with other concomitant glomerular diseases were excluded. Renal biopsies were performed after written informed consent was obtained from the patients. The study protocol was approved by the Human Ethics Review Committee of the China–Japan Friendship Hospital.
Assessments
Age, sex, diabetes duration, DR status, hematuria, 24 h urinary protein excretion level, blood pressure, and estimated glomerular filtration rate (eGFR) were the baseline clinical parameters recorded at the time of renal biopsy.
The patients were divided into five subgroups based on DM duration:<5,≥5 to<10,≥10 to<15,≥15 years, and unknown. DR severities were scored in accordance with the criteria of the Global Diabetic Retinopathy Project and the International Consensus Development Workshop (2002) [
8]. Grade I (no apparent retinopathy) is assigned when no abnormality is evident upon dilated ophthalmoscopy (DO). Grade II (mild nonproliferative diabetic retinopathy) corresponds to the presence of microaneurysms (only) evident upon DO. Grade III (moderate nonproliferative diabetic retinopathy) involves damage greater than merely microaneurysms but less severe than that of nonproliferative diabetic retinopathy that is evident upon DO. Grade IV (severe nonproliferative diabetic retinopathy) includes any of the following findings: more than 20 intraretinal hemorrhages in each of the four quadrants, definite venous beading in two or more quadrants, and prominent intraretinal microvascular abnormalities in one or more quadrants, but no sign of proliferative retinopathy. Grade V (proliferative diabetic retinopathy) exhibits one or more of the following signs: neovascularization and vitreous/preretinal hemorrhage.
Four subgroups were formed on the basis of the number of red blood cells per high-power field (HPF):<3,≥3 to<10,≥10 to<50, and≥50 cells/HPF. Four subgroups were also prepared in terms of 24 h urinary protein excretion level:<0.3,≥0.3 to<1.0,≥1.0 to<3.5, and≥3.5 g/d. Three subgroups were further established in accordance with the WHO/International Society of Hypertension (ISH) Guidelines on the Management of Hypertension (2003) [
9]: grade 1, systolic blood pressure (SBP) 140–159 mmHg or diastolic blood pressure (DBP) 90–99 mmHg; grade 2, SBP 160–179 mmHg or DBP 100–109 mmHg; and grade 3, SBP≥180 mmHg or DBP≥110 mmHg. The eGFRs were calculated using the CKD–EPI (Epidemiology Collaboration) formula [
10] as eGFR [mL/(min·1.73 m
2)] = 141 × min (SCr/k, 1)
a × max (SCr/k, 1)
−1.209 × 0.993
Age × [1.018 if female] × [1.159 if black], where k is 0.7 for females and 0.9 for males, a is-0.329 for females and -0.411 for males, min is the lesser of SCr/k or 1, and max is the greater of SCr/k or 1. Five subgroups were created in accordance with the baseline eGFR [
11]: CKD1, glomerular filtration rate (GFR)>90; CKD2 (mild), GFR 60–89; CKD3 (moderate), GFR 30–59; CKD3a, GFR 45–59 combined with CKD3b, GFR 30–44; CKD 4 (severe), 15–29; and CKD5 (renal failure), GFR<15 mL/(min·1.73 m
2).
DN was clinically staged using Morgensen’s criteria developed in 1987 [
12]. Grade I patients show no clinical manifestation of a kidney disease in the glomerular-filtration-positive period. In this period, renal hypertrophy develops and GFR increases, but the urinary albumin excretion rate (UAER) remains normal. Grade II patients enter the phase of normal albuminuria, display unapparent clinical manifestation, experience general diabetes for>5 years, present microalbuminuria, and yield a normal UAER (<20 µg/min) at rest but possess a slightly increased UAER under stress. Grade III (microalbuminuria phase) patients exhibit persistent microalbuminuria, a UAER of 20–200 µg/min, and a 24 h urinary albumin level of 30–300 mg. However, their urine does not usually contain proteins, and their GFR is normal. Grade IV (clinical proteinuria) patients have persistent urinary protein, a 24 h urine protein level>0.5 g, a UAER>200 µg/ min, and a gradually decreasing GFR. Clinically, these patients manifest nephrotic syndrome (NS), a reduced creatinine clearance rate, and high blood pressure. Grade V (terminal renal failure) patients experience a further decline in GFR (to<15 mL/min) or undergo dialysis. Clinically, these patients often develop massive proteinuria or chronic renal failure.
All of the tissues were routinely processed for light microscopy, immunofluorescence, and electron microscopy. Renal tissues were divided into three portions. One portion was fixed in buffered formalin, processed into paraffin blocks for light microscopy, and stained with hematoxylin and eosin, periodic acid–Schiff (PAS), silver methanamine, and Masson trichrome. The second portion was frozen for direct immunofluorescence studies by using fluorescein isothiocyanate conjugated antibodies detecting IgG, IgA, IgM, C3, C4, C1q, and fibrinogen. The third portion was fixed in Trump’s EM fixative and processed into resin blocks; afterward, ultrathin sections were stained with uranyl acetate and lead citrate and subjected to transmission electron microscopy [
13,
14]. On the basis of the DN criteria of the Research Committee of the Renal Pathology Society (2010) [
15], we classified the tissues with evident DN. Class I refers to glomerular basement membrane thickening, in which thickening is isolated and only mild and nonspecific changes observed in light microscopy do not satisfy the criteria for Classes II–IV. Class II is designated as mesangial expansion, that is, mild (IIa) or severe (IIb). In this class, the glomeruli exhibit mild or severe mesangial expansion but without nodular sclerosis (Kimmelstiel–Wilson lesion) or global glomerulosclerosis in more than 50% of the glomeruli. Class III is defined as nodular sclerosis (Kimmelstiel–Wilson lesion) with at least one glomerulus with a nodular increase in mesangial matrix (Kimmelstiel–Wilson lesion) and without changes in class IV. Class IV includes advanced diabetic glomerulosclerosis with more than 50% global glomerulosclerosis accompanied by other clinical or pathological evidence, which indicates that sclerosis is attributed to DN.
Follow-up
A total of 53 patients who underwent renal biopsies in China–Japan Friendship Hospital from 1998 to 2016 were followed up. Renal survival was defined as the time from renal biopsy to chronic renal replacement therapy (hemodialysis or peritoneal dialysis) or to the time of the last follow-up (June 30, 2016). Clinical data were analyzed, and differences in the clinical parameters and pathological classifications of the patients who had and had not progressed to dialysis were compared.
Statistical analysis
Statistical analysis was performed using SPSS version 17.0 for Windows (IBM SPSS Statistics, Armonk, NY, USA). Data were expressed as means±standard deviations. Kaplan–Meier and Cox’s regression analyses were conducted to determine the potential risk factors for poor renal survival.P<0.05 was considered statistically significant.
Results
Baseline characteristics
We retrospectively analyzed data from 590 patients with T2DM. A total of 111 patients (18.8%) were diagnosed with simple DN by renal biopsy, and their baseline characteristics are shown in Table 1. We evaluated 80 (72.1%) males and 31 (27.9%) females aged 26–74 years (mean 49.9 years). Of the 110 patients whose disease duration was known, 27 (24.3%) lived with the disease for<5 years, 40 (36.0%) suffered from this disease for 5–10 years, 18 (16.2%) coped with the disease for 10–15 years, and 18 (16.2%) manifested this disease for>15 years. Thus, short-duration diabetes did not indicate that a patient would unlikely progress to DN. Moreover, long-duration diabetes did not necessarily predict DN progression.
The proportions of patients with grade I (none apparent), grade II (mild), grade III (moderate), grade IV (severe), and grade V (proliferative) DR were 18.9% (21), 13.5% (15), 29.7% (33), 15.3% (17), and 9.9% (11), respectively. Therefore, DN likely occurs if a diabetic patient with kidney damage suffers from DR. However, the data did not suggest that patients with DN would manifest DR.
The proportions of patients with hematuria of<3, 3–10, 10–50, and>50 cells/HPF were 51.4% (57), 30.6% (34), 14.4% (16), and 3.6% (4), respectively. Of the total number of patients, 48.6% (54) experienced hematuria, and this finding suggested that the condition was prevalent among patients with simple DN.
Of the patients with available daily urinary protein excretion levels, 2 (1.8%) excreted<0.3 g/d, 2 (1.8%) released 0.3–1.0 g/d, 33 (29.7%) secreted 1.0–3.5 g/d, and 69 (62.2%) produced>3.5 g/d. Although more than half of the patients with DN exhibited severe proteinuria, only a small proportion suffered from mild proteinuria or even lacked overt proteinuria.
Only 30 (27.0%) patients yielded a normal blood pressure. Furthermore, 29 (26.1%) patients were grade 1, 26 (23.4%) patients were grade 2, and 26 (23.4%) patients were grade 3. Most of the patients with DN manifested hypertension.
The proportions of patients with eGFRs of>90 (CKD1), 60–89 (CKD2), 45–59 (CKD3a), 30–44 (CKD3b), 15–29 (CKD4), and>15 mL/(min·1.73 m2) (CKD5) were 18.9% (21), 28.8% (32), 20.7% (23), 11.7% (13), 16.2% (18), and 3.6% (4), respectively. Furthermore, the proportions of patients with eGFRs of≥60 (CKD1–2) and<60 (CKD3–5) mL/(min·1.73 m2) were both 50%, suggesting that half of the patients with DN already suffered from severe renal dysfunction at the time renal biopsy was performed.
On the basis of DN pathological type, we treated 0 cases of class I, 26 (23.4%) cases of class IIa, 14 (12.6%) cases of class IIb, 65 (58.6%) cases of class III, and 6 (5.4%) cases of class IV. More than half of the patients were found to have Kimmelstiel–Wilson lesions in their glomeruli upon renal biopsy.
Follow-up characteristics
Among the 111 patients with DN, 58 (52.3%) were lost to follow up. Of the remaining 53 patients, the average follow-up time was 3.31±2.06 years, and the average age at renal biopsy was 49.98±12.16 years. However, we found no difference in sex, age, glomerular classification, renal survival duration, or DM duration between the two groups (Table 2). Of all the patients with DN, 93.5% (29/31) belonged to Morgensen stage IV before kidney biopsy was performed but exhibited different prognoses (19 progressed to dialysis, but 34 did not). The 24 h urinary protein excretion levels were significantly higher in the patients who progressed to dialysis than in the patients who did not have to undergo dialysis (P = 0.032), but hematuria and eGFR did not differ significantly between the two groups.
Kaplan–Meier analysis revealed that renal survival was not significantly associated with the glomerular classification (Fig. 1). Cox’s regression analysis (Table 3) indicated that survival was significantly influenced by sex (b = 1.394, P = 0.038), hematuria (b = 0.036,P = 0.029), and eGFR (b = -0.039, P = 0.002) but not by age, 24 h urinary protein excretion, or glomerular classification (P>0.05). After adjusting for sex, age, hematuria, 24 h urinary protein excretion, and eGFR, we observed through Cox’s regression analysis that renal survival was not significantly associated with glomerular classification (P>0.05).
Discussion
With lifestyle changes and increasing obesity prevalence, T2DM and its complications have become a pandemic [
16]. A patient who suffered from diabetes for at least 7–10 years, exhibited demonstrable DR, experienced prior microalbuminuria without sudden-onset heavy proteinuria, lacked hematuria, possessed a kidney that was not abnormally small, and did not manifest other renal diseases was previously considered to have progressed to DN [
5].
Some authors believed that a patient with long-duration DM develops microalbuminuria or proteinuria, shows an increase in blood pressure or serum creatinine level, and suffers from other DM complications, including retinopathy, should be clinically diagnosed with DN. In this regard, the prognosis is known; hence, renal biopsy is diagnostically unnecessary. However, renal biopsy must be conducted to confirm the diagnosis if NDRD or DN combined with NDRD is suspected.
According to NKF-KDOQI guidelines published in 2003 [
9], patients with DKD should be subjected to renal biopsy under the following circumstances: absence of DR, low or rapidly decreasing GFR, rapidly increasing proteinuria or NS, refractory hypertension, presence of active urinary sediment, signs or symptoms of other systemic diseases, or>30% reduction in GFR within 2–3 months after initiation of angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker therapy.
DN or DM combined with simple NDRD is difficult to diagnose correctly by considering clinical manifestations and laboratory results only. We found that the clinical and laboratory characteristics of our biopsy-proven patients with DN varied (Table 1). However, most of the patients with DN suffered from hypertension and DR, and this observation was consistent with previous data. Many of the patients with DN experienced hematuria or presented NS, whose symptoms include>3.5 g protein/d in urine, but this finding was inconsistent with those in previous reports.
We found that short-duration diabetes was not associated with low DN incidence. Moreover, long-duration diabetes was not invariably associated with DN. Of the total number of patients, those with diabetes for less than 10 years constituted 60.3% of all the patients with DN. Of all the patients with DN, 48.6% experienced hematuria, and this observation suggested that the incidence of this condition was high in patients with simple DN. Therefore, hematuria is not a specific marker of DN or NDRD. Furthermore, hematuria possibly plays a pathogenic role in promoting kidney disease progression, especially in patients with NDRD. In our study, hematuria was associated with the renal outcomes of the patients with simple DN. Although more than half of the patients with DN exhibited severe proteinuria, only a small proportion had mild proteinuria, and overt proteinuria was absent in some instances. Our data and previous findings revealed that neither hypertension nor renal dysfunction is a specific marker of DN or NDRD [
13]. However, we observed that a patient with diabetes and kidney damage should be strongly suspected of DN if DR is also evident, and this recommendation is consistent with previous data [
13,
14].
The prognosis of patients with DN significantly influences their survival and quality of life. For example, Chinese patients of different glomerular classifications exhibit 5-year renal survival rates of 100% (class I), 90.1% (class IIa), 75.4% (class IIb), 39.0% (class III), and 15.3% (class IV) [
17]. Glomerular classifications are also significantly correlated with renal outcomes [
17–
19]. However, we found that glomerular classification was not associated with renal outcome (progression to dialysis or not) and the Morgensen stages of our patients did not differ. Of the total number of patients who were followed up, 94.3% (50/53) were classified as Morgensen stage IV before kidney biopsy was performed, but their prognosis differed (progression to dialysis or not). These results were inconsistent with those of previous studies because of the following factors. (1) In addition to clinical and pathological stages, DN progression and outcome are frequently affected by various clinical parameters, including age, DM duration, and blood glucose level. Therefore, we experienced difficulty in precisely predicting DN prognosis based on clinical stage and pathological classification only. (2) The limited sample size of our study also weakened the strength of the observed associations. (3) This study was retrospective in nature, and only the patients treated from 1998 to 2016 were included. The proportion lost to follow-up was high (52.3%), and this finding possibly introduced some bias. (4) The 1987 Morgensen criteria might not match well with our enhanced understanding of DN. Thus, some investigators recommended that a pathological scoring system combined with clinical parameter evaluation should be employed to predict renal outcomes in patients with DN [
18].
Conclusions
Our work emphasized that the clinical manifestations of DN varied and could not be easily summarized. Neither clinical Morgensen stage nor pathological classification indicated the DN outcomes, such as progression to dialysis. In the future, a prospective multi-center study would be conducted to explore the association of DN prognosis with pathological and clinical parameters and to overcome the limitations of a retrospective study and a small sample size. This study also aimed to improve the prediction accuracy of DN prognosis. We recommend that patients with diabetes and renal damage should undergo renal biopsy if this procedure is not contraindicated.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(128KB)
