Renal involvement in diabetes mellitus and other metabolic diseases: hyperuricemia, oxalosis
The kidney is frequently affected in systemic diseases and, in some cases, the severity of the renal damage is a determining factor for the survival of the patient. Among the diseases that we grouped here, diabetes mellitus is most important by its frequency. Other metabolic or genetic diseases will be discussed in the chapter of hereditary diseases.
Diabetes mellitus is one of the main causes of terminal renal disease. Diabetic nephropathy (DN) develops in 25-35% of patients with insulin-dependent diabetes mellitus (IDDM) and in 15-25% of non-insulin-dependent (NIDDM); renal changes in both types of diabetes are morphologically and physiopathologically similar. Proteinuria is the main marker of the DN and this disease is one of the most common causes of nephrotic syndrome.
The pathogenesis of DN is not completely understood. Nonenzymatic glycation of proteins seems to be one of the main mechanisms of glomerular injury; terminal products of the glycation can be united to amino groups of other proteins, their accumulation in the collagen and other proteins of the matrix can diminish the adhesion and replication of endothelial cells, vasoconstriction by mesangial cells, and increase lipoprotein and immune complexes adhesion to monocitos and macrophages. Hemodynamic factors that cause glomerular hyperfiltration are another mechanism of renal damage. It has been suggested that many complex pathways contribute to the pathobiology of diabetic complications including hyperglycaemia itself, the production of advanced glycation end products (AGEs) and interaction with the receptors for AGEs such as the receptor for advanced glycation end products (RAGE), as well as the activation of vasoactive systems such as the renin-angiotensin aldosterone system (RAAS) and the endothelin system. More recently, it has been hypothesised that reactive oxygen species derived from NAD(P)H oxidases (Nox) may represent a common downstream mediator of vascular injury in diabetes (Gray SP, et al. The pathobiology of diabetic vascular complications--cardiovascular and kidney disease. J Mol Med (Berl). 2014;92:441-52. [PubMed link]). Three major histologic changes occur in the glomeruli of persons with diabetic nephropathy. First, mesangial expansion is directly induced by hyperglycemia, perhaps via increased matrix production or glycation of matrix proteins. Second, thickening of the glomerular basement membrane (GBM) occurs. Third, glomerular sclerosis is caused by intraglomerular hypertension (induced by dilatation of the afferent renal artery or from ischemic injury induced by hyaline narrowing of the vessels supplying the glomeruli). These different histologic patterns appear to have similar prognostic significance (Batuman V. Diabetic Nephropathy. in: eMedicine, consulted on October 9th 2017. [See link]).
Clinical features: The earliest finding of renal involvement in DN is microalbuminuria that gradually progresses until proteinuria, usually after 15 years of the beginning of the disease. Delay in the diagnosis of NIDDM with respect to IDDM can make seem that DN is developed more early in NIDDM. DN is developed approximately in 30% of diabetic patients. DN risk seems to diminish if it has not been developed 20 years after the beginning of the diabetes, which suggests exists a group of patients with greater susceptibility to renal affectation, probably related with genetic predisposition. DN is higher in Afro-American than in Caucasians with diabetes. High systemic blood pressure and tobacco increase the risk of nephropathy in diabetes.
Microalbuminuria (30-300 mg/24 hours) is a marker of initial renal involvement and it is associated with a progressive development of more severe renal changes and greater risk of cardiovascular complications. Proteinuria >500 mg/24h indicates the beginning of DN. The severity of proteinuria is variable among patients, in 5-10% there is nephrotic syndrome, and in few of them massive proteinuria (>10 g/24h). Chronic renal damage develops progressively; damage of the renal tissue is complex and involves glomeruli, vessels (so in any other organ or system), interstitium and tubules. When progressing the chronic changes the renal functional reserve is diminished, and when renal damage exceed the critical limit of parenchymatous lesions variable degrees of renal insufficiency begin to detect progressing until terminal renal failure. DN is associated with retinopathy, and it is more frequent in patients with DMID: 90%.
Hypertension is usual in diabetics. Other organs are variably affected in the disease. In general, to greater degree of chronic renal disease more advanced will be lesions in cardiovascular system and other organs.
The progression of the initial changes in DN depends much on the control of glycemia and other risk factors, so, the time interval between initial changes and terminal renal damage is highly variable. DN can appear in renal allografts as recurrence or as de novo disease. DN is developed more quickly in the transplanted kidney that in native kidneys: in average six post-transplantation years.
Laboratory findings: Almost all the patients with DN present proteinuria, 5-10% with nephrotic syndrome. In few cases there is microhematuria, if detected we must think about the possibility of other glomerular diseases complicating the DN. Levels of serum creatinine and BUN usually are normal at the beginning of the DN, but progressively they increase. In diabetic patients there is propensity to hyporeninemic hypoaldosteronism and the patients can have hyperkalemia and metabolic acidosis. Other laboratory test will be altered in agreement with the severity of systemic alterations.
In initial phases of the DN the kidneys may be increased in size due to hyperfiltration and hypertrophy; in advanced stages there will be atrophy with diminution of size and diminution or loss of the cortico-medullary differentiation.
The more initial changes are glomerular hypertrophy, mild mesangial expansion (matrix), and thickening of the glomerular capillary walls, these changes are more evident with electron microscopy. When the disease progress there is also increase of mesangial cellularity; this mesangial matrix increase produces formation of nodules in the glomerular tuft. The nodules have variable size in a same glomerulus and affect variably the glomeruli (nodular diabetic glomeruloesclerosis). The nodules are known as Kimmelstiel-Wilson nodules. They are spherical, eosinophilic, with a central acellular area, and they can be surrounded by a ring of cells. They stain blue or green with the trichrome stain and they are positive with PAS and methenamine-silver stains. The nodules are of two types, morphologically and pathogenically: there are small nodules, mesangials (or in the intercapillary areas), and there are others with more large size and with laminated aspect when observed with the silver stain (Figures 1, 2, 3, 4). More large nodules form from expanded glomerular capillaries (microaneurisms, >35 microns of diameter) that would generate injury in the capillary wall, thromboses and collagenous matrix proliferation (Stout LC, et al., Hum Pathol; 24:77-89, 1993 [PubMed link]) (Figures 3 and 4). The small ones are seen as areas with rounded accentuation of the mesangial matrix and stain more intensely with PAS that the larger laminated nodules. The presence of these nodules: large and small, some laminates, with variable size and distribution in the glomeruli, are “virtually” pathognomonic of DN, although similar nodules have been described as “idiopathic” in non-diabetic patients (Markowitz GS et al., Hum Pathol 33:826-35, 2002 [PubMed link]; Chang CS, et al., Clin Nephrol 64:300-4, 2005 [PubMed link]; Navaneethan SD, et al., J Nephrol 18:613-5, 2005 [PubMed link] [Free full text]). The nodules seen in light chain deposit disease can be similar, but they are more homogenous in size and distribution and stain more weakly, or they are negative, with silver stain; the nodules seen in amyloidosis do not stain with silver and they are positive for Congo red. Immunofluorescence is also useful in the differential diagnosis.
In DN the glomeruli present increase (sclerosis) of the mesangial intercapillary matrix, with progressive increase of the thickening of capillary walls and later evolution to global glomerulosclerosis (Figure 5).
Figure 1. Glomeruli with Kimmelstiel-Wilson nodules. The smallest nodules can be more cellular and the greatest nodules tend to be acellular in the centre and surrounded by more cellular zones. Capillaries are seen around these nodules, sometimes adopting an aspect in garland (like in the three nodules indicated with arrows); in some cases we see microaneurisms around nodules. Notice the variability of size of the nodules (H&E, X.400).
Figure 2. Kimmelstiel-Wilson Nodules highlight with PAS stain; glomerular nodules in amiloidosis and light chain deposits disease have a more weak stain with PAS (PAS, X400).
Figure 3. The larger nodules usually have a laminated aspect (arrow) more evident with silver stain (Figure 4). These nodules have different physiopathogenic mechanism (see the text). Notice the variability in the size of nodules in this glomerulus, something that usually does not happen in amyloidosis nor in light chain deposits disease (Masson’s trichrome, X400).
Figure 4. The larger nodule in this microphotography emphasizes the prominent concentric lamination with the silver stain (arrow). This finding is very characteristic of nodular diabetic glomerulosclerosis. (Methenamine-Silver, X400).
Figure 5. In many glomeruli with DN the changes are only mesangial thickening (global or segmental) and diffuse thickening of the capillary walls. Notice in this microphotography the mesangial widening. In addition there is increase of the thickness of the Bowman’s capsule basement membrane. (Masson’s trichrome, X400).
Other two glomerular lesions, called exudative lesions (similar to arteriolar hyalinosis), are capsular drop and glomerular hyalinosis. The first is a homogenous, hyaline deposit, in the Bowman’s capsule, usually it is rounded or elongated and it is highly suggestive of DN, although non-pathognomonic (it can be occasionally seen in hypertension and other idiopathic nodular glomerular lesions). Glomerular hyalinosis is formed by plasma components that are accumulated in peripheral segments of the tuft, also it is called hyaline cap or fibrin cap (Figures 6, 7 and 8). In many typical DN cases microaneurisms produced by mesangiolysis are evidenced (Figures 8 and 9).
Figure 6. Exudative lesions in DN include glomerular hyalinosis (hyaline cap) (green arrow). In this case it is accompanied by several Kimmelstiel-Wilson nodules (two of them marked with red arrows). See, in addition, some atrophic tubules with basement membranes very thickened (blue arrows). (Masson’s trichrome, X400).
Figure 7. In this glomerulus there is mild increase of the mesangial matrix. The arrow indicates a beautiful capsular drop, another one of the exudative lesions in DN, a feature “almost pathognomonic” of this glomerulopathy (see the text). In this image we see the capsular drop fuchsinophilic (red), but in other cases we can see it with a green or blue tone; compare with the capsular drops in Figure 8. (Masson’s trichrome, X400).
Figure 8. In this image we can see four alterations of DN: mesangial widening by expansion of the matrix; arteriolar hyalinosis (blue arrows); two capsular drops, one in each glomerulus (red thin arrows); and microaneurisms (green arrows), frequent changes in DN that usually are associated to focal areas of mesangiolysis. (Masson’s trichrome, X400).
In tubules there are unspecific changes: reabsorption of protein droplets, tubular damage and atrophy. The basement membranes of atrophic tubules are characteristically much thickened, usually more than in other causes of tubular atrophy; this change is another one of the alterations that can make think us about DN (Figure 10). The Armani-Ebstein change (or Armani-Ebstein cells) consists of deposits of glycogen in the tubular epithelial cells (pars straight of proximal convoluted tubule and loop of Henle); it is very rare to see it at the present time; it appears in decompensated diabetics with glycemia superior to 500 mg/dL and severe glycosuria; it is a reversible alteration without functional manifestations. In interstitium there are unspecific chronic changes.
In vessels usually there are notorious changes; the most characteristic lesion is intimal hyaline thickening of arterioles, sometimes with nodular appearance (Figures 8 and 9). If hyaline arteriolosclerosis is very prominent in young patients, it must alert us on the possibility of DN. Arteriolar lesions may involve any arteriole; if we demonstrated arteriolar hyalinosis in both glomerular arterioles (afferent and efferent) it is virtually pathognomonic of DN. In arteries there is intimal fibrosis (arteriosclerosis), but it is not different of intimal fibrosis in other diseases.
Figure 9. In this image we can see prominent arteriolar hyalinosis, with formation of nodules that replace the media of the glomerular arteriole (blue arrow). The glomerular tuft shows widening of the mesangial matrix and a capillary microaneurism in which a portion of the mesangium seems lost: mesangiolysis (green arrow). (Masson’s trichrome, X400).
Figure 10. Another characteristic lesion in DN is prominent thickening of basement membranes in atrophic tubules (and in Bowman’s capsule - see Figure 5). When we find this finding we must think about the possibility of DN, although it is not a specific finding. (Masson’s trichrome, X400).
DN is a very common disease and any type of glomerulopathy can be superposed to DN. Papillary necrosis is a complication that can appear in diabetics, mainly in patients with nodular glomerulosclerosis, significant vascular alterations and pyelonephritis.
Go to Case 47 as a good exemple of diabetic nephropathy.
The most characteristic change is the presence of linear deposits of IgG and albumin in walls of glomerular capillaries and, to a lesser extent, basement membranes of tubules and Bowman’s capsule. The immunostaining usually is slighter than in anti-GBM disease and they are not accompanied by complement deposition (Figure 11). This immunostaining is due to unspecific adhesion and it is not an immunological reaction. Hyaline lesions can have IgM and C3 as in other focal and segmental changes. Linear positivity for kappa and lambda can also be detected. Some authors prefer to call to this linear appearance “pseudolinear” staining or “accentuation of the basement membrane” instead of “linear positivity”.
Figure 11. By immunofluorescence is frequent to find deposits of IgG that are accompanied by albumin and adopt a linear parietal pattern (or “pseudolinear”); the staining usually is more tenuous than in anti-GBM disease. This finding can be also seen in tubular basement membranes. Immunostaining with complement components usually is not seen. The linear staining with albumin helps to differentiate it from anti-GBM disease. The red arrows indicate some capillary walls with linear positivity; the clear blue arrow indicates a diabetic nodule. (Direct immunofluorescence for IgG with anti-IgG human antibodies marked with fluorescein, X400).
With this technique is more evident the mesangial expansion, it is caused by increase of the extracellular matrix; nodules have similar aspect to mesangium, with mild electron-dense appearance. GBM is seen diffusely thickened; this thickening is progressive and can be superior to 1,000 nm. GBM can be seen with fibrillary aspect and can appear laminated. GBM thickening is a consequence of extracellular matrix accumulation, with increased deposition of normal extracellular matrix components such as collagen types IV and VI, laminin, and fibronectin. Such accumulations result from increased production of these proteins, their decreased degradation, or a combination of the two. Tubular basement membranes also show this thickening. Sometimes mesangial interposition of cells in the capillary wall can be seen (focal). Hyaline deposits are electron-dense. [See image EM; notice the marked homogenous thickening of the GBM (link)]
Figure 11b. Diffuse thickening of the glomerular basement membrane. This is the earliest change that is usually seen on the kidney biopsy. It can present as an isolated ultrastructural change, without clinical or laboratory alterations. Notice the normality of the pedicels. A thicker GBM is also seen with aging or hypertension. (EM, original magnification, X6,000).
In 2010 was published a classification system containing specific categories that discriminate lesions with various prognostic severities that would be easy to use. This proposal was launched by the Research Committee of the Renal Pathology Society in 2006 in San Diego and further discussed in Leiden in September 2008, and presented as a consensus classification of DN developed by a group of international experts (Tervaert TW, et al. Pathologic Classification of Diabetic Nephropathy. J Am Soc Nephrol. 2010;21(4):556-63 [PubMed link]). In this classification glomerular alterations are graded I to IV and tubulointerstitial and vascular lesions are graded separately:
Glomerular classification of DN:
Class I: Mild or nonspecific LM changes and EM-proven GBM thickening. - Biopsy does not meet any of the criteria mentioned below for class II, III, or IV. GBM >395 nm in female and >430 nm in male individuals 9 years of age and olde. See Case 169 of our Case Series: Early diabetic nephropathy.
Class IIa: Mild mesangial expansion. - Biopsy does not meet criteria for class III or IV. Mild mesangial expansion in 25% of the observed mesangium.
Class IIb: Severe mesangial expansion. - Biopsy does not meet criteria for class III or IV. Severe mesangial expansion in 25% of the observed mesangium.
Class III: Nodular sclerosis (Kimmelstiel– Wilson lesion). - Biopsy does not meet criteria for class IV. At least one convincing Kimmelstiel– Wilson lesion.
Class IV: Advanced diabetic glomerulosclerosis. Global glomerular sclerosis in 50% of glomeruli. Lesions from classes I through III.
Mesangial expansion is defined as an increase in extracellular material in the mesangium such that the width of the interspace exceeds two mesangial cell nuclei in at least two glomerular lobules. The difference between mild and severe mesangial expansion is based on whether the expanded mesangial area is smaller or larger than the mean area of a capillary lumen.
In addition, interstitial fibrosis and tubular athropy (IFTA) is graded as 0: No IFTA; 1: <25%; 2: 25-50%, 3: >50%.
Interstitial inflammation: 0: absent; 1: Only in relation to IFTA; 2: In areas without IFTA.
Arteriolar hyalinosis: 0: absent; 1: at least one area of arteriolar hyalinosis; 2: more than one area of arteriolar hyalinosis.
Arteriosclerosis (score worst artery): 0: no intimal thickening; 1: intimal thickening less than thickness of media; 2: intimal thickening greater than thickness of media.
This classification requires further refinement of the criteria for distinguishing class IIa (mild) versus IIb (severe) mesangial disease, and reconsideration of the clinical utility of making a diagnosis of class I diabetic glomeruloscleorsis. In addition, there are concerns related to lack of specificity, reproducibility, validation, and relevance to clinical practice (Stokes MB. Classification Systems in Renal Pathology: Promises and Problems. Surg Pathol Clin. 2014;7(3):427-41 [PubMed link]).
NEPHROPATHY IN HYPERURICEMIA
Hyperuricemia can involve the kidney producing hyperuricemic nephropathy, chronic nephropathy by urates and uric acid renal calculi. Hyperuricemic nephropathy is found in cytotoxic therapy for leukemias, lymphomas, and other malignant tumors. Chronic nephropathy by urates is found in persistent or recurrent hyperuricemia (as a metabolic disease); and uric acid renal stones also are evidenced in some patients with persistent or recurrent hyperuricemia, the renal alterations are consequence of the obstruction.
In acute hyperuricemia nephropathy (by massive tumor lysis) there are precipitation of monosodic urate forming crystals in collecting ducts, destruction of the epithelium and reaction of foreign body type.
Chronic nephropathy by urates is not universally accepted like a specific disease; when histologic characteristic findings appear isolated, without other alterations, usually there is not renal failure, this one is developed mainly if it is associated to hypertension or poisoning by lead. In most of the cases it is not an injury that causes renal failure.
Macroscopically yellowish striae can be seen, they correspond to crystal deposits. Microscopically the lesions characterize by deposits of crystals in tubules and collecting ducts and can break the basement membranes and leave to the interstitium generating inflammation, gigantocellular reaction and fibrosis. The crystals are soluble in water, reason why processing in nonwatery solutions like alcohol is required. They have needle form, although also they can see rectangular or amorphous (Figure 13); they are birefringent with polarized light; they are seen in small irregular groups in the medulla, surrounded by giant cells and histiocytes (Figure 12).
Figure 12. Medullary interstitial urate crystal deposits in chronic nephropathy by urates. The crystals usually are lost in the routine processing of the tissue, but the shadow of the needle crystal remains surrounded by multinucleated giant cells. The particular grouping that they adopt in the tissue is very characteristic. This finding is relatively frequent in kidneys from older people; these deposits usually do not produce renal failure. (Masson’s trichrome, X400).
Figure 13. Uric acid crystals. They have needle form. Crystales of this image were obtained by aspiration of a gouty tophus in the forearm. This crystals are soluble in water, reason why processing in nonwatery solutions like alcohol, or direct observation is required. (Direct observation with polarized light, X400)
Oxalosis is the pathological deposition of oxalate in tissues; it can happen in the kidney and other organs. The toxicity can take place by hyperoxalemia and hyperoxaluria; the main cause is gastrointestinal hyperabsorption (in celiac sprue, ileal resection, pancreatic insufficiency, short intestine syndrome, and intestinal inflammatory disease) associated to malabsorption of biliary salts (the greater amount of biliary salts that arrive at the colon increases the absorption of oxalate). Hyperoxalemia can be also associated to ethyleneglycol poisoning. Also a genetic disease exists, in this there are enzymatic deficiencies or hyperabsorption of oxalate.
Acute hyperoxalemia produces calcium oxalate crystal deposition in tubular lumina, the crystals are irregular, laminates or in fan form; they are colourless with routine stains and birefringent with polarized light. Chronic lesions relate to tubular basement membranes rupture and interstitial deposits with inflammatory reaction and fibrosis. These crystals are conserved in the tissue routinely processed (Figure 14).
In any case of acute renal failure (of any cause) these crystals can be deposited, thus they are not pathognomonic of the disease. According to some authors, the only histologic criterion that would allow to confirm that it is a primary oxalosis (genetic), would be crystal deposits in the wall of small renal arteries (Colvin RB. Renal transplant patholgy. In: Heptinstall's Pathology of the Kidney, 5º ed.; Lippincott-Raven, Philadelphia, 1998, p. 1.509).
Figure 14. Nephrectomy of renal allograft by chronic rejection. There are extensive deposits of oxalate with variable size and form; they occupy mainly distal tubules and, in fewer amounts, they can be seen in the interstitium, with gigantocellular reaction. The asterisks indicate proximal tubules, which usually do not contain these crystals. This finding, thus severe, does not indicate oxalosis; there are frequent and unspecific deposits in cases of renal failure in native or transplanted kidneys. (H&E seen with polarized light, X200).
Acknowledgements: Many of the microphotographies of this page have been obtained from biopsies of the Department of Pathology of the Hospital Clínico San Carlos and Universidad Complutense, Madrid, Spain, thanks to the kindness of Dr. Julia Blanco.