Diabetes: To Biopsy or not to Biopsy - Part 2

In my experience the appetite to biopsy patients with clinical renal disease varies widely between clinician, institution and country. This is most apparent when it comes to the diabetic patient with renal disease. In my opinion there is a fear among nephrologists of biopsying patients with potential diabetic nephropathy (DN). Of course there are clinical features that increase the likelihood that a diabetic with renal disease has DN. Retinopathy in a type 1 diabetic, longstanding DM and progressive proteinuria over years preceding an elevated creatinine. However, the role of microalbuminuria and proteinuria in predicting CKD progression and the traditional course of rising albuminuria in DM1 has been challenged. In DM type 2 the picture is less clear and these patients frequently have many comorbidities. Gearoid posted on this subject about a year ago now. In CJASN (October) Sharma et al from Columbia University reviewed the characteristics and renal diagnoses in 620 diabetics who had a renal biopsy. Over 90% had type 2 diabetes. In Columbia in 2011 approximately ¼ of all biopsies were on patients with diabetes. The results of this retrospecive review were that 37% of patients had DN alone, 36% had non-diabetic renal disease (NDRD) alone, and 27% had DN plus NDRD.

In NDRD alone: FSGS (22%), hypertensive nephrosclerosis (18%), acute tubular necrosis (ATN) (17%), IgA nephropathy (11%), membranous GN (8%), and pauci-immune GN (7%) comprised 80% of diagnoses.

In DN plus NDRD: ATN (43%), hypertensive nephrosclerosis (19%), FSGS (13%), and IgA nephropathy (7%).

In multivariate analysis longer duration of DM was associated with a greater likelihood of DN and less likelihood of NDRD.

The table illustrates the features of patients at the time of biopsy. Older age, having DM2 vs DM1, short duration of DM and less proteinuria are more likely in those with NDRD alone vs DN alone in this cohort. 

I think this study highlights the importance of considering biopsy in diabetics (especially type 2). I would be interested to hear about the experience of others with respect to biopsy in diabetics.

Posted by Andrew Malone

Characteristics	DN Alone	DN Plus NDRD	NDRD Alone
Participants (n)	227	164	220
Age (yr)	59 (49–65)	63 (55–72)	63 (54–70)
Male sex	129 (56.8)	100 (61.0)	142 (64.6)
Race
Unknown	108 (47.6)	57 (34.8)	104 (47.3)
White	62 (27.3)	63 (38.4)	70 (31.8)
African American	39 (17.2)	33 (20.1)	29 (13.2)
Hispanic	12 (5.3)	7 (4.3)	8 (3.6)
Asian	4 (1.8)	4 (2.4)	7 (3.2)
Other	2 (0.9)	0 (0.0)	2 (0.9)
DM type 1	9 (4.0)	5 (3.1)	2 (0.9)
Duration of DM (yr)	13 (8–17)	10 (7–18)	5 (3–10)
Serum creatinine (mg/dl)	2.3 (1.6–3.8)	3.1 (1.7–5.2)	2.3 (1.5–4.4)
eGFR (ml/min per 1.73 m2)	31.3 (17.5–55.2)	21.4 (12.5–46.6)	32.5 (14.3–60.0)
Proteinuria (g/d)	5.0 (2.8–8.8)	5.0 (2.0–8.0)	2.9 (1.4–7.1)

The end of the road for urinary eosinophils?

Nate wrote a couple of posts in the past about the use (and misuse) of urinary eosinophils for the diagnosis of acute interstitial nephritis (AIN). Since the original report describing the use of this test in the NEJM in 1986, a number of papers and commentaries have been published that cast doubt on its true effectiveness. A paper was recently published in CJASN that looked at the accuracy of this test in a series of patients who had kidney biopsies for AKI between 1994 and 2011.

The authors examined the biopsy and urine results of 566 patients who had biopsies and a urinary eosinophil test over the course of the study. Overall 91 patients had biopsy-confirmed AIN, 73 of whom were considered to be drug-induced. 31.6% of patients had UE>1%, the traditional cut-off for the diagnosis of AIN. The majority of these did not have AIN and the distribution of positive UE was uniform across diagnoses. As a result, the sensitivity and specificity of this test for the diagnosis of AIN were poor. Using 1% UE as a cut-off, the sensitivity was 30.8% and the specificity was 68.2%. The PPV was 15.6% and the NPV was 83.7%. Using a more stringent cut-off of 5%, the sensitivity decreased to 19.8%, the specificity, PPV and NPV were 91.2%, 30% and 85.6% respectively.

One significant limitation of this study was that it was restricted to patients who had kidney biopsies and the majority of patients with suspected AIN do not have kidney biopsies. However, I would imagine that this should bias the results towards favoring UE as a test as presumably patients with more severe disease would be more likely to have a biopsy. Another issue is the potential that there was a bias towards biopsying patients who did not have UE (and so were not thought to have AIN in the initial impression). This could have the effect of reducing sensitivity.

Notwithstanding this, the fact that the distribution of positive UE was so well distributed among the various diagnoses combined with the very low sensitivity of this test would suggest that the use of this test should no longer be routine in the diagnosis of AKI. At best, a negative test helps rule out AIN in patients with a low pre-test probability while a positive test is not particularly useful. As the authors of the accompanying editorial point out, even in drug-induced AIN, the infiltrate may not have a high proportion of eosinophils, suggesting that in many cases there may not even be a plausible biologic rationale for this test.

Image of the Month - Pathology

A 60 year-old woman with a longstanding history of hypertension, scleroderma and MGUS presented to the emergency room with diarrhea, vomiting and AKI (creatinine increased from baseline of 1.5mg/dl to 3.1mg/dl). She had a distant history of membranous nephropathy diagnosed on a renal biopsy 25 years previously. She had been worked up in the renal clinic for CKD and had a renal US showing relatively small kidneys. Her medications included an ACE inhibitor.

On admission, she had no hematuria. Her BP was elevated although she had a significant postural drop. She had dipstick proteinuria and she was empirically started on oral steroids for a possible GN. A renal biopsy was performed:

The image above is of the renal cortex. A single glomerulus is seen (A) which is hypoperfused. There is significant dilatation of the tubules (B) indicating acute tubular injury. There is attenuation and degeneration of the tubular epithelial cell layer. At the lower end of the image is an atrophic tubule with a hyaline cast.

This image again shows a hypoperfused glomerulus. However, the main finding here is an extremely damaged arteriole (A). There is multilayering within wall of the vessel (that was replicated throughout the biopsy). The vascular lumen is almost occluded with endocapillary proliferation and remodeling of the vessel wall.

In some areas of the biopsy (better perfused), the glomeruli looked relatively normal. There was some mesangial expansion with irregular capillary loops but no evidence of membranous disease.

IF showed some minimal deposition of IgG in the mesangium but equal kappa and lamda light chains.

Finally, the EM showed multiple mesangial deposits (A) of uncertain significance but again, no evidence of membranous disease. Notably, there were was no evidence of active inflammation in the biopsy.

Following the report of the biopsy, the steroids were stopped. The patient's blood pressure was controlled and her renal function returned to baseline within a few days. It is likely that this acute episode was related to volume depletion and acute tubular injury exacerbated by her severe underlying vascular disease. Interestingly, she now has no albuminuria and she is back on her ACEi. The significance of the mesangial deposits remains unclear and she will be followed on an ongoing basis in the renal clinic.

We often see patients like this on consult who present with AKI following a GI illness while on an ACEi. However, we don't normally get to see the pathology in these very common case.

Click on any image to enlarge

eJournal Club - Kidney Biopsies

This month's eJournal Club concerns a paper (with an accompanying editorial) reporting the experience of renal biopsies in Norway. One issue that arises again and again is whether or not patients should undergo outpatient biopsies. The argument against this is that many complications occur more than 8 hours following a biopsy and that the complication risk is too high to allow patients to remain unmonitored at home. These recommendations were based on older data which dated from before the era of live ultrasound and the more modern biopsy needles. More recent studies have suggested that outpatient biopsies are safe and I had a conversation with one attending at the ASN last year who was surprised that everyone was not doing outpatient biopsies because they had been doing it without a problem for years.

This paper, while not specifically addressing the issue of outpatient vs. inpatient biopsies, should alleviate some of the concerns that people have. After 9288 biopsies, the rate of serious complications was 0.9% (transfusion or surgical intervention). This is similar to recently published studies. There were no deaths during the 20 years covered by the study. There was an increased risk of bleeding in patients with a low GFR, uncontrolled hypertension, older age and acute kidney injury. Notably there was no difference in bleeding complications when 14G needles were used as opposed to 18G needles although less glomeruli were obtained with the smaller needles. 

There an interesting point for discussion here - in the course of the study, there was a marked decline in the number of nephrologists performing biopsies. The majority of biopsies are now performed by radiologists. These radiologists are more likely to use smaller needles and thus get less tissue. Should nephrologists take back the renal biopsy? Should we be more assertive in insisting that larger needles are used to ensure adequate biopsies? Head over the eJC for the discussion.

To biopsy or not to biopsy - IgA

The topic of IgA nephropathy has been discussed in multiple posts here in the past. IgA is the commonest primary GN with a wide variety of presentations and prognoses ranging from isolated hematuria to a rapidly progressive GN. Over the last few years, my practice has been that if I see a patient with isolated hematuria, no adverse family history and minimal proteinuria, I will not biopsy them and reassure them that their chance of progression is very low. This is an approach that has also been supported by previous posters.

An article and accompanying editorial in this month's JASN would appear to support this approach. This was a retrospective review of the outcomes in 141 patients diagnosed with IgA nephropathy with minimal proteinuria (less than 0.5g) between 1975 and 2008 in 8 Spanish hospitals. During this period there was an aggressive policy in place with a very low threshold for renal biopsy. At the time of biopsy, only 16% were hypertensive while 17.7% had no detectable proteinuria. All patients had a normal eGFR at baseline.

Over a mean follow-up of 9 years, no patients were treated with immunosuppressive agents and 41% were treated with RAAS blockade. 5 patients (3.5%) had an increase of 50% from baseline creatinine while 1 patient had a doubling of creatinine (following pregnancy). 14.9% of patients developed proteinuria (more than 0.5g) while it increased above 1g in 6 patients. The only factor predicting increased proteinuria or an increased creatinine after multivariable analysis was the presence of FSGS (S1 by the Oxford Classification).

These results are very reassuring and for me reinforce the idea that we should not be routinely biopsying patients who present like with isolated hematuria. It is notable that most of the hospitals in this study have stopped this aggressive biopsy policy and it is unlikely that a similar study will be done in the future. However, there is no doubt that they should be followed long term because there is a (small) risk of progression. It should be stated that these results are not necessarily generalizable to other populations - studies in Asian populations have found a much higher rate of progression and the indication for biopsy might be different in this group. It would be interesting to know what the genetic factors underlying this difference in outcomes might be.

Still mysterious: the elusive circulating factor for FSGS

Important new findings were recently published in relation to proteinuria and FSGS, which are definitely of interest to our community.

First, the punch line:

There is new evidence for a “circulating factor” in recurrent FSGS in a fascinating case of a re-transplanted kidney (here)

BUT

There is growing evidence that suPAR is a non-specific marker of kidney disease and therefore not likely to be the “circulating factor.”(here)

In fact, it appears that it is non-specifically found in CKD, and correlates with a declining GFR.

Now for some details:

The re-transplanted kidney

A letter to the NEJM editor (here) describes an amazing case of resolution of recurrent FSGS after re-transplantation. 

A 27 year old patient with primary FSGS receiving a kidney from his healthy 24 year old sister developed proteinuria in the nephrotic range (up to 25 g/day!) within 2 days of transplantation, and had no improvement after plasmapheresis and standard immunosuppressive treatment. A renal biopsy confirmed foot process effacement, the first hallmark of recurrent podocyte damage heralding recurrent FSGS. Incredibly, with all appropriate consents and institutional approval, the transplant team removed the allograft from Patient 1 and re-transplanted it into another patient who had ESRD due to diabetes. Within 3-4 days, the proteinuria resolved and a repeat biopsy showed resolution of foot process effacement and re-establishment of a normal podocyte architecture. Eight months later, Patient 2 is reported to be doing very well, with good allograft function and no proteinuria.

This case demonstrates in a remarkable way that recurrent FSGS results from an elusive “factor” rapidly produced by the recipient (with primary FSGS), and that the allograft itself can remain fully functional if removed from the influence of this “factor” and placed in another patient.

suPAR is not suPER specific

What may have seemed to be exciting news in 2011, namely the notion that soluble uPAR may be predictive of recurrent FSGS (here), appears to be unfortunately evolving into yet another unsuccessful attempt to identify the ever elusive circulating factor.

Recent work published in Kidney International by Maas et al. (here) confirms that suPAR is not able to distinguish between idiopathic FSGS, secondary FSGS or minimal change disease. 

This is actually not surprising, because a closer look at the clinical data in Wei et al. (here) reveals that the admittedly arbitrary cut-off for separating primary FSGS from all other glomerular disease (3000 pg/ml) did not hold up when tested among their patient cohorts with idiopathic, recurrent versus non-recurrent FSGS (all had suPAR> 3000 pg/ml, thus suPAR could not predict the recurrent from the non-recurrent cases). 

The second figure in the Maas et al. paper may help explain this conundrum: they show a negative correlation between suPAR and eGFR, meaning that as GFR drops, suPAR levels rise, which essentially means that suPAR is simply a marker of CKD.

Future work will no doubt continue to address these issues, but the apparent lack of specificity of suPAR for FSGS casts serious doubt on its proposed role as the circulating factor.

So, the search is still on!!!

Glomerular Density and Predicting Progression in IgA Nephropathy

IgA nephropathy (IgAN) is an unpredictable beast, and identifying patients who will go on to develop progressive disease can be challenging. The classic predictors of progression, namely heavy proteinuria, CKD, glomerular sclerosis and tubulointerstitial fibrosis, are themselves just markers of advanced disease. There is a real need for better early histopathologic predictors of renal prognosis.

One promising prospect is glomerular density (GD), i.e. the number of nonsclerotic glomeruli per mm2, which was the focus of this recent study from Japan. The investigators performed a retrospective analysis of almost 100 IgAN cases with preserved renal function at the time of biopsy (average GFR 90 ml/min). Interestingly, they found that the GD varied hugely between patients, from 1 to 8 glomeruli per mm2. In multivariate analysis, only GD, and the presence of a cellular/fibrocellular crescent, were found to be significant predictors of progression. Patients with a low GD experienced more rapid decline in renal function, and this enhanced risk for progression was increased if a cellular/fibrocellular crescent was also present. Interestingly, the classic predictors of progression listed above did not predict prognosis, presumably due to the early stage of disease.

Of course, these findings will need to be replicated in other ethnic groups. Nonetheless, when a case of IgAN next appears at your biopsy conference, you should pipe up “what’s the GD?”. Just try not to look too smug when you do it.