December 2017

1 ‘Normal BP’ becomes a rare disease now

Half of the globe was fast asleep then [when it was 3:15pm (PST) on 13th November at Anaheim, California]. I had just completed reading Murakami’s ‘Samsa in love’. This is an amazing story of events in the life of a Cockroach who, on one fine morning gets up metamorphosed into a human being. Upon meeting a girl, he neglects basic sexual instincts and prefers to follow the ‘neocortical attribute’ of love. After vivid dreams about ‘metamorphoses’ in early morning hours, I got up and checked my blood pressure and alas! I was no more a proud healthy young adult that went to sleep last night and found myself metamorphosed into someone with ‘elevated blood pressure’. I was one of that majority of the world’s population who was declared having ‘abnormal’ BP at the release of ACC/AHA guidelines on blood pressure. It was 122/81 on 3 consecutive measurements that need lifestyle and diet modifications and 3-6 monthly follow up to see what happens to me as per Guideline 8.1.2…to see if I get another metamorphosis into stage 1 hypertension. 

Hypertension is largely treated by primary care doctors, who are not necessarily graduates of modern medicine here whose degree can be various 4 letter permutations and combinations (heard of DEMS?). While one can doubt their qualifications, same can’t be said about their confidence, which, generally is inversely related to the knowledge, which is periodically boosted by pharma food and cocktails. Although guideline claims not advocating drugs for everyone at stage 1 hypertension (>130/80 now), this change of number is likely to lead pharma to a goldmine. Unlike ‘Samsa in love’, basic instincts ‘to prescribe’ and ‘get prescribed’ won’t be easily overcome by doctors and patients. Also, some of the experts think that hypertension is a ‘progressive vascular disease’ and not just the risk factor and therefore needs earlier and more aggressive drug treatment. Exercise, diet and lifestyle modifications were and will continue to be regarded by most doctors and patients as posthumous metaphysical experiences. 

Two good points worth putting into practice 1. better measurement of blood pressure with the emphasis on out of office records and 2. assessing CV risk beyond numbers which is a long due change in BP treatment guideline. 

Moral of the story is: don’t remain awake late night reading fictions, early morning dreams can come true. Enough of ‘coach’ing, I’m going for a walk now to improve my numbers.  


2. Association of BP lowering with mortality and cardiovascular disease across blood pressure levels


Is benefit of BP reduction same at various pretreatment BP thresholds?

Probably not. In this systematic review and meta-analysis involving 74 trials and 306273 participants association of BP-lowering treatment with major cardiovascular events was dependent on baseline systolic BP (SBP). For primary prevention (i.e. in patients without prior heart disease),  with baseline SBP 160 or above, treatment was associated with reduced risk for death (RR, 0.93; 95% CI, 0.87-1.00) and a substantial reduction of major cardiovascular events (RR, 0.78; 95% CI, 0.70-0.87). For SBP range 140-159, the association of treatment with mortality was similar (RR, 0.87; 95% CI, 0.75-1.00), but the association with major cardiovascular events was less pronounced (RR, 0.88; 95% CI, 0.80-0.96) and with baseline SBP below 140, treatment was not associated with mortality (RR, 0.98; 95% CI, 0.90-1.06) and major cardiovascular events (RR, 0.97; 95% CI, 0.90-1.04). However, for people with previous CHD and mean baseline SBP of 138 mm Hg, treatment was associated with reduced risk for major cardiovascular events (RR, 0.90; 95% CI, 0.84-0.97), but was not associated with survival (RR, 0.98; 95% CI, 0.89-1.07).

Authors caution about industry bias here, as all the trials involving patients with CHD were industry sponsored. Moreover, limited information on non-BP determinants of outcomes, underrepresentation of women, rather crude division into primary and secondary preventive trials (only 13 trials were truly primary preventive, and 17 trials truly secondary preventive), are limitations to note. Funnel plot asymmetry at highest BP levels (signifying possible publication bias-negative studies remaining unpublished)  might suggest an overestimation of benefit here.

BP lowering benefits decrease with decreasing starting BP, and risks of treatment (as a primary prevention measure) below 140 mmHg may outweigh any possible benefit. Sounds contrasting to SPRINT, however, authors feel that cutoff of 140 may not be very different than SPRINT 120, given the different method of measurement in SPRINT.  

The goal of BP treatment is not to bring down numbers but to reduce the risk of complications and this risk can be different for different individuals at same BP number. A risk-based approach is likely to be more cost-effective and avoid overtreatment in mild hypertension. This is one thing that’s worth adopting from ACC/AHA.    

3. Who develops CKD after recovery from AKI?

A six variable model using commonly used measurements -age, sex, baseline serum creatinine value, albuminuria, acute kidney injury severity, and discharge serum creatinine value was developed in 9973 patients hospitalized in Alberta and was externally validated in with data from a cohort of 2761 patients hospitalized in Ontario, Canada. The main outcome measure was advanced CKD -sustained reduction in eGFR less than 30 mL/min/1.73 m2 for at least 3 months during the year after discharge. Authors propose that this will provide an accurate but simple strategy that could be used to identify patients who can be benefited by follow up. Information on albuminuria wasn’t available in all the patients and so was the etiology of AKI, which IMHO can independently affect recovery.

Aren’t most of these factors (baseline CKD, age, sex, proteinuria) independently associated with advanced CKD, even without AKI intervening? The only conclusion from the observational data on ‘AKI-CKD interconnection’ research that can be made for now is: CKD is one of the most important risk factors for AKI and AKI  accelerates the progression of CKD. This probably can’t be extrapolated to one of the most common AKI in our practice i.e. the one associated with tropical infections, where baseline kidney function is normal.

4. Sertraline for depression in CKD

About 1 in 4 patients with CKD have depression which is 4 times higher incidence as compared to general population. Antidepressant use is common in this population although placebo-controlled trials are lacking. Here is a double-blind, placebo-controlled, parallel- design, 12-week flexible-dose randomised clinical trial (CAST)conducted at 3  medical centers comparing sertraline with placebo in patients with CKD stage 3-5 (about 50% had CKD 4,5), not on dialysis. This drug was no more effective than placebo and was associated with higher incidence of nausea and diarrhea. Similar findings were reported with use of SSRI in the treatment of depression associated with major medical conditions, which probably is different than depression in general population. 

While ‘negative results’ may add to the depression of nephrologists desperate for ‘positive trials’, some important takeaways from this research:

-almost 1/4th of the patients responded to placebo, worth trying option before anything else.

-if one is required to use SSRI for severe depression (trial didn’t include such patients), up to 200mg/day of sertraline was tolerated well except for GI intolerance.

Exclusion of dialysis population and limited duration of follow up are important limitations of the trial. 

5. Tolvaptan in ADPKD

After promising results of TEMPO 3:4 trial (only prerequisites to understand this promise are-accept kidney volume as valid surrogate and neglect high discontinuation rate due to adverse effects) in early-stage CKD-ADPKD, here is a phase 3, randomized withdrawal, multi-center, placebo-controlled, double-blind trial of this drug in late-stage CKD. The primary end point-change in the eGFR from baseline to 12 months follow-up-−2.34 ml per minute per 1.73 m2 (95% CI −2.81 to −1.87) in the tolvaptan group, as compared with −3.61 ml per minute per 1.73 m2 (95% CI, −4.08 to −3.14) in the placebo group (difference, 1.27 ml/min 95% CI, 0.86 to 1.68; P<0.001). 5.6% in treated group vs 1.2% in the placebo group developed a significant increase in ALT level.

1.27 ml per minute difference in eGFR is thought to be clinically significant by authors and Otsuka-they postulate that this will translate into postponing CKD stage 5 from 6.2 years to 9.0 years. To understand this, one needs to assume that medicine is mathematics and need to stretch your optimism to assume that drug will continue to make a difference of 1.27ml per year without acceptable side effects over several years of follow up. This trial, in its run in phase, excluded patients who didn’t tolerate tolvaptan. Longer follow-up studies with hard outcome endpoints are desperately needed now and until they are available, freely available ADH suppression i.e. liberal water intake seems enough to me. Considering variable and generally slow disease course of ADPKD, further trials also should target high-risk population where drug’s cost and side-effects will be justified better. Will precision medicine help do this for ADPKD?

6. No need to wait for few minutes to check orthostatic hypotension

Checking for orthostatic BP change in routine clinical visits is a ‘vital sign’,  especially in patients on multiple drugs that can affect blood pressure and volume. This becomes all the more relevant if you are serious about SPRINT and  ACC/AHA 2017. As a newly joined resident in medicine, I recall doing this exercise for 40-50 patients in the outpatient nephrology department and used to argue with our registrar about the time gap between supine and standing readings. We used to complete the measurement in 2 min and he insisting for a 5min gap.

In this prospective cohort study involving 11,249 middle-aged adults (mean age, 54; many with cardiovascular disease or risk factors, or on medications), orthostatic hypotension (defined as systolic drop of ≥20 mm Hg or diastolic drop of ≥10 mm Hg) was assessed at baseline, with BP readings taken with an automatic cuff every 25 seconds as many as 5 times after a patient rose from a supine to a standing position. During the median follow-up of 23 years, about 12,000 adverse events, that included falls, fractures, syncopal episodes, motor vehicle accidents, and death, were recorded and these were found most consistently and closely associated with measurements made within first 60 seconds upon standing. Waiting for more than 2 minutes may, in fact, miss some of these patients.

7. Outcomes after Angiography with Sodium Bicarbonate and Acetylcysteine

(See the excellent summary of the article by @hswapnil on NephJC here). This large randomized 2 X 2 factorial trial enrolled 5177 patients (more than any of the previous trials) who were scheduled for angiography to receive 1.26% sodium bicarbonate or 0.9% sodium chloride and oral acetylcysteine or placebo. The primary endpoint was a composite of death, the need for dialysis, or a persistent increase of at least 50% from baseline in the serum creatinine level at 90 days. The trial was stopped prematurely after a prespecified interim analysis after the primary endpoint occurred in similar proportion (about 4 to 5%) of patients in all the four groups. Contrast-associated acute kidney injury, which was defined as an increase in serum creatinine of either at least 25% or at least 0.5 mg per deciliter from baseline at 3 to 5 days after angiography occurred in about 8-9 % patients and there were no significant between-group differences in these rates.

Does it change what we practice? We have stopped using N-acetylcysteine already. We could use normal saline and sodium bicarbonate interchangeably. But do we need to?

The trial did give a very clear answer to the research question it asked. However, contrast-induced AKI is going through the existential crisis and ~4% incidence of significant renal dysfunction at day 90 in PRESERVE population may be argued as an evidence for its existence. However, as we don’t have a control group here (who didn’t receive contrast), and not all kidney events were adjudicated to contrast use, so, for now, we don’t have any definitive answer to this existential crisis.

8. One-Day Donor Assessment Model in a Living Kidney Donor Transplant Program

Being evaluated as a donor for organ transplantation is an emotionally complex journey. A thorough evaluation is required to make sure the donation doesn’t put the donor at undue risk. However, a lengthy process of donor evaluation would significantly slow the entire program of living donor transplantation. Graham and Courtney in an article published in AJKD describe their experience with a ‘1-day donor assessment pathway’ as quality improvement project that led to a sustained increase in the living donor renal transplantation rate. To decrease the donor fatigue and subsequent dropout, they introduced a pathway where the prospective donor would undergo all the investigations and consultations over one day. Over 5 years, the program assessed 431 potential donors, of which 66% went ahead for actual donation. Of course, the people undergoing the evaluation were happy, most of them rating this process as ‘very good’. There was a sustained increase in the living donation rate from <5 per million population per annum (pmp pa) before 2010 to >32 pmp pa by 2015. This is incredible!

While it is likely that this increase is the result of multiple factors, authors feel that the 1-day assessment process has been the main driving force behind this improvement.

By making appropriate changes according to the local needs, this concept is likely to work in any program.



November 2017

1. Effect of intensive glucose control on renal and patient survival  

In this Cochrane review of 11 RCTs involving 29141 diabetics, intensive control of sugars (HbA1c < 7% or FBS <120)  made little or no difference in the renal and patient survival endpoints: doubling of serum creatinine, ESRD or death as compared to the usual control (HbA1c >7 or FBS >120). Intensive control was shown to improve some surrogate endpoints like albuminuria or non-fatal myocardial infarction. Findings were consistent in various subgroups and sensitivity analyses. Similar findings were also noted in a 2012 meta-analysis.  

Some of my physician friends were disappointed by this ‘negative study’-a term I fail to understand, when, in fact, such studies further our understanding and thereby improve care. These results, therefore, should be considered as an important ‘positive’ input to guide care. 

Moderate heterogeneity, underrepresentation of type 1 diabetes and fewer clinically important events like doubling of creatinine could have led to the less precise effect estimates. However, RCT on this issue, addressing these limitations, seems to be a fantasy, and this data is an important addition to the evidence guiding glucose control. While benefits of intensive control remain controversial, such attempts are associated with definite harm (often underreported)-hypoglycemia-which is the most common endocrine emergency associated with morbidly and mortality.  While I was discussing these results with my endocrine colleagues, they say, they will continue to practice ‘good (and never practiced ‘intensive’) control’ which means individualizing targets,  HbA1c between 7-8 and as close to 7 as possible without risking hypoglycemia. 

2. ACE Inhibitors and Statins in Adolescents with Type 1 Diabetes (AdDIT trial)

 443 type 1 diabetics with UACR values in the upper third of the albumin-to-creatinine ratios (below <30 mg/gm) were randomly assigned in a placebo-controlled trial of an ACE inhibitor and a statin with the use of a 2-by-2 factorial design, to evaluate its effects on change in the urine albumin/ creatinine ratio (primary outcome), and development of microalbuminuria, progression of retinopathy, changes in the glomerular filtration rate, lipid levels, and measures of cardiovascular risk assessed by carotid intima-media thickness, hsCRP, ADMA (secondary outcomes). It was a commendable job to successfully maintain this population in the trial. While the value of albuminuria as a sign of nephropathy is much more certain in type 1 than type 2 diabetes, authors proposed that even below the traditional thresholds of 30 mg/gm, rise in albumin excretion implies heightened risk based on their previous experience of Adolescent Type 1 Diabetes Cardio-Renal Intervention Trial (AdDIT) group.

Treatment of risk factor to prevent disease is called as primordial prevention; what do we call it when we treat risk markers? I don’t know, but, here is what authors of this study found: while statin did their job (of lowering lipids) ACE inhibitors didn’t perform as expected-there was no change in albuminuria assessed every 6 months over 2 to 4 years, and expressed as the area under the curve. Such attempts in type 2 diabetics were successful in achieving surrogate endpoint of reducing albuminuria but were associated with harm. In AdDIT trial, therapy was tolerated well and authors are hopeful that they might expect some ‘legacy effect’ of these interventions in future.

Hope, yes, that’s what keeps us going—to the next research article.


3. Multitarget therapy for maintenance treatment of lupus nephritis 

This open-label, multicenter study for 18 months (as an extension of the prior induction therapy trial) in 19 renal centers in China aimed to assess the efficacy and safety of multitarget (MT) maintenance therapy in patients who had responded at 24 weeks during the induction phase. Multitarget (doesn’t that sound opposite of precision?) maintenance therapy with Tac+MMF+Predni compared with AZA+Predni was associated with similar cumulative renal relapse rates (5.47% versus 7.62%, respectively; adjusted hazard ratio, 0.82; 95% confidence interval, 0.25 to 2.67; P=0.74). Serum creatinine levels and eGFR remained stable in both the groups. Authors conclude that MT therapy is an effective and safe maintenance treatment for patients with lupus nephritis. Significant dropout, the inclusion of patient only in remission (complete or partial), all Chinese participants , surrogate endpoint (proteinuria) which can be easily altered by CNI by their nonimmune effects producing ‘pseudo-remission’ and limited follow up are important concerns that make me wait to see safety and efficacy of  MT strategy to be replicated in some more trials.

Another interesting finding of the study is almost similar response rate in the ‘cyclophosphamide–>AZA’  arm at 1.5 yrs indicating the fact that some patients will remain proteinuric at the end of induction treatment and they continued to respond slowly to match those treated with MT therapy. The distinction of induction and maintenance here may, therefore, be not very relevant.

4. UK/UK+UNa as an indicator of hypovolemia in nephrotic syndrome

‘Overfill’ or ‘underfill’? Don’t worry fellows, this question about the genesis of edema bothers you only until you start actually treating nephrotic edema, where it’s ‘underfill’ if diuretics produce azotemia and overfill if they work. In those ‘difficult nephrotics,’ we often resort to albumin+frusemide infusion especially if they already have elevated BUN and creatinine.

Here is an interesting report, telling us how to identify these patients who will be benefited by the use albumin ( costly and not always the safe option). They studied UK/UK+UNa as an indicator of hypovolemia (and therefore elevated aldosterone) in a prospective study involving 44 children and 36 controls. Subjects were grouped into low, normal, and high GFR [assessed by continuous inulin and para-aminohippurate (PAH) perfusion]. In the low GFR group, significantly lower renal plasma flow (p = 0.01), filtration fraction (p = 0.01), and higher UK/UK+UNa (p = 0.03) ratio were noted. Authors conclude that a urinary potassium to potassium plus sodium ratio > 0.5–0.6 identifies patients with nephrotic syndrome who are hypovolemic suggesting benefit of albumin infusion in this subgroup.

Although small, this study involved detailed and objective assessment of GFR, renal plasma flow, and vasoactive hormone and I’m eager to check how this simple urine lytes measurement works at the bedside.

5. Levamisole in Children with Frequently Relapsing Nephrotic Syndrome

Levamisole (Dicaris) is less toxic and relatively inexpensive alternative steroid-sparing agent commonly used in childhood in nephrotic syndrome. While its use is common, evidence supporting its use till date was weak at the best. Here is an RCT that now seems to bridge this evidence gap. In an international (significant contribution from India), multicenter, placebo-controlled, double-blind, randomized clinical trial, Gruppen et al randomized 99 children with frequently relapsing nephrotic syndrome with or without steroid dependence to receive levamisole or placebo after inducing remission with prednisone. Time to first relapse (primary endpoint) was not different between the two groups in first 100 days; however, later, levamisole group had lower hazard for developing a relapse needing prednisone. Fewer children had relapse requiring prednisone over 1 year in levamisole group compared to the placebo group. More children in the levamisole group had adverse reactions (mainly neutropenia in 16%).

One child developed arthritis and antineutrophil cytoplasmic antibodies. Disseminated vasculitis has been described with levamisole use for nephrotic syndrome.  Recently, there have been several reports describing ANCA positive vasculitis in association with the use of cocaine contaminated with levamisole. A close follow-up is warranted if levamisole is used for a prolonged period.

6. Global Kidney Health and Country-specific CKD characteristics

Articles starting with ‘global kidney health’ scares me as this generally means in brief: ‘CKD epidemic’, ‘population screening’, ‘cohesive plans’, ‘implementation of this integrated comprehensive plan’ and stuff like that with minor differences of the details. This Lancet article is a report from ‘CKD summit’ of more than 85 experts around the globe, convened by the International Society of Nephrology, aimed to identify and prioritize key activities for the next 5–10 years in the domains of clinical care, research, and advocacy and to create an action plan and performance framework based on 10 themes …..(interested ones can refer to the article). Identifying CKD as ‘one disease’ is something like diagnosing someone with serious abdominal pathology as ‘chronic abdomen’, and reflects the fact that we as a specialty are probably farthest from the talk of the town -precision medicine’.

One of the 10 themes that caught my attention was “Reduce acute kidney injury—a special risk factor for CKD”. While there is no doubt that CKD is one of the most important risk factors for AKI, and AKI episodes can hasten CKD progression, considering AKI as a risk factor for de-novo CKD (barring specific scenarios like HUS, cortical necrosis, allogeneic BMT) that too of public health significance is too simplistic understanding of CKDu.

Here is a BMJ article reporting population-based data in six countries assessing if attributes of persons with CKD differ in low-income and middle-income countries compared with high-income countries.  In the USA, urban India and Moldova, 79.0%–83.9%; in China and Nepal, 62.4%–66.7% and in Nigeria, 51.6% persons with CKD fit
one of three established risk profiles (Profile 1-Diabetes, Profile 2 CVD, Profile 3 Obesity, HTN, low HDL, high TG, and Profile 4 CKD without established risk factors). So except, Nigeria, authors propose, that most other countries should be focussing on traditional risk factors for prevention of CKD. The study also highlights the need to further evaluate CKD without established risk factors in low-income and middle-income countries.

Data from India comes from two of its largest metros -Delhi and Chennai, and therefore I suspect that Profile 4 CKD might be underrepresented. Also in 3/6 countries, for the purpose of assessment, the healthy general population was invited to the center-something like a health camp where people with a known medical condition are more likely to turn up-and, therefore, might not represent actual population characteristics.

CKDu is big and it’s good that we are admitting it as an issue, better late than never.

7. PCI Strategies in Patients with Acute Myocardial Infarction and Cardiogenic Shock

Patients who have a multivessel disease and present with cardiogenic shock after acute myocardial infarction, it is unclear if PCI of all the stenotic lesions found there works better in comparison to the PCI of culprit lesion alone. In an open-label, multicentre RCT, Thiele et al randomly assigned 706 such patients to one of two initial revascularization strategies: either PCI of the culprit lesion only-with the option of staged revascularization of nonculprit lesions or immediate multivessel PCI. The primary endpoint (a composite of death or severe renal failure leading to renal-replacement therapy) occurred in fewer patients in the PCI of the culprit lesion only group (45.9% vs 55.4%; relative risk, 0.83; 95% confidence interval, 0.71 to 0.96; P=0.01). This was mainly driven by the number of deaths in multivessel PCI group; the number of episodes of AKI requiring renal replacement therapy was similar (11.6% in culprit vessel PCI group and 16.4% in multivessel PCI group; relative risk, 0.71; 95% CI, 0.49 to 1.03; P = 0.07). About one-third of the patients in either group received renal replacement therapy for uremia, defined as blood urea >50 mg/dl. The article does not mention the number of less severe episodes of AKI that are known to be associated with mortality.

Authors, strangely, attribute poor outcome in multivessel PCI group to “contrast cardiomyopathy” (if not the kidney, contrast can be of consequence to the heart!)-higher dose of contrast material that was used in the multivessel PCI group may have led to acute left ventricular volume overload and a subsequent negative effect on myocardial function and recovery. However, more logical explanation of these findings may be the prolonged duration of the multivessel PCI procedure (classical wisdom says not to treat chronic total occlusion in nonculprit vessels in this setting) may be hazardous at a time when the patient is hemodynamically compromised.

8. Glecaprevir and Pibrentasvir in Patients with HCV and Severe Renal Impairment

There is limited data on the safety of sofosbuvir-based DAA regimens in patients with CKD-this drug is almost entirely cleared renally and there is a concern of acute GFR decline during the therapy. Here is a multicenter, open-label, single-group (without a control group), phase 3 trial published in NEJM, Gane et al studied 104 patients with HCV infection and advanced CKD (stage 4 or 5), 82% on hemodialysis. They could have any genotype of HCV. Combination of glecaprevir and pibrentasvir was given for 12 weeks. Twelve weeks after the end of this treatment, sustained virologic response (SVR) was seen in 98% of the patients (102 of 104 patients; 95% confidence interval, 95 to 100). PK studies showed no significant clearance of these drugs during dialysis. Pruritus, fatigue, and nausea were the common adverse events. None of the 24 serious adverse events occurring during the study period were considered to be drug-related. So here we have a pan-genomic therapy that can be used safely in our patients with advanced CKD and on dialysis. Looking forward to having this drug combo available and accessible for our patients most of whom pay out of pocket for their treatment.

Interferon-free regimens for patients with eGFR < 30 ml/min per 1.73 m2
Genotype 1 Elbasvir- grazoprevir ± ribavirin

Ombitasvir- paritaprevir- ritonavir plus dasabuvir ± ribavirin

Glecaprevir- pibrentasvir

Genotype 2 Glecaprevir- pibrentasvir
Genotype 3 Glecaprevir- pibrentasvir
Genotype 4 Elbasvir- grazoprevir ± ribavirin

Ombitasvir- paritaprevir- ritonavir

Glecaprevir- pibrentasvir




October 2017

1 CKD after intensive BP lowering, recovery from CKD and ‘extensive’ vs ‘intensive’ BP control

A secondary analysis of SPRINT trial (BTW, how many can be done per trial?), looking at the Effects of Intensive Systolic Blood Pressure Control on Kidney and Cardiovascular Outcomes in Persons Without Kidney Disease is published in Annals of Internal Medicine. 140 of 3326 participants (4.2%) in the intensive group and 40 of 3336 (1.1%) in the standard group had an incident CKD event in SPRINT participants without CKD at baseline, at 3 years of follow-up. Intensive SBP treatment resulted in an absolute 2.6% higher risk for incident CKD and 2.2% lower risk for the composite of a primary CVD event or all-cause death. So the choice is between CVD/death or CKD! NNN=38 (number needed for nephro consult), 38 new patients will need to see nephrologist after 3.3 years if they choose to SPRINT. Here is another news for nephrologists: CKD can recover spontaneously! (25.7% in the intensive arm who were classified as ‘developed CKD’ eventually recovered).  We have discussed the limitations of generalizing SPRINT data to CKD in July. It’s no surprise that after trying to get blood pressure of an elderly hypertensive to 120 systolic, you will see an acute decline in GFR, electrolyte abnormalities, syncope, and falls more often. For those ‘classified CKDs’ where CKD is just a CV risk (SPRINT population) factor, you may discuss risks and benefits of aggressive BP lowering and for those with ‘actual CKD’ where CKD is a risk factor for renal failure, BP should be controlled and nobody knows for sure what the target should be.

Largest risk reduction with BP control is achieved when baseline risk is high-i.e. those with extremely elevated BP. Rather than being ‘Intensive’, ‘Extensive’ control-by population-based low-cost interventions- may be much more effective at achieving such risk reductions. Such programmes led by nephrologists have shown a reduction in the incidence of CKD in India and JAMA article here is another excellent attempt in addressing hypertension via a community-based programme from Argentina.

2 What is the target BP for patients with preexisting CKD?

In this systematic review and meta-analysis on 18 RCTs involving 15,924 patients with CKD, Malhotra et al, evaluated the association between more intensive vs less intensive blood pressure lowering and risk of mortality. More intensive vs less intensive BP control resulted in 14.0% lower risk of all-cause mortality (odds ratio, 0.86; 95% CI, 0.76-0.97; P = .01), and this was without significant heterogeneity and appeared consistent across multiple subgroups.

This large study looking at the hard outcome of all-cause mortality is a significant addition to the evidence for treating high blood pressure in CKD. However, the definition of intensive here is very different from that of SPRINT (mean BP here was 132 vs 140 in intensive vs usual control arms). Moreover, the study addresses all patients with GFR <60 as one group which need not respond similarly to aggressive BP lowering (those with GFR <30 may experience more harm like AKI, hyperkalemia), study do not provide information on the effect of BP targets on different GFR categories. Finally, mortality may be ‘too hard’ an outcome to be affected by BP control alone- as causes of death like sepsis, infections that are common in this population.

3 What is the implication of acute decline in GFR after antihypertensive therapy in patients with baseline CKD?

One risk of such decline, that every practicing nephrologist is aware, is patient ditching you. Increase by 0.3mg or 30% from baseline creatinine, 30% drop from baseline GFR are various limits that we follow in practice; these are not based on the high-quality evidence. In this retrospective study in JASN, authors evaluated implications of such increase in creatinine for future ESRD risk in 899 AASK and 761 MDRD trial participants previously randomized to strict versus usual BP control. More than 20% decline in GFR was associated with increased ESRD risk in both intensive and usual BP control arms [ HR 3.04 (1.95 to 4.77) and  2.56 (1.60 to 4.11) in the intensive and usual control arm of AASK respectively. HR 1.57 (1.09 to 2.24) and 1.48 (1.04 to 2.1) for intensive and usual control arms of MDRD respectively].

Patients developing an increase in the creatinine after RAS blockade may not just be in ‘prerenal success‘, but they need to be identified as a patient group prone to land up into trouble. This large BMJ study also highlighted the fact that even smaller increases in creatinine may not be benign-whether this itself is the causal mediator of adverse outcomes or just marker of ‘bad kidneys’ remains to be seen.

4 Induction and maintenance treatment of proliferative lupus nephritis

This network meta-analysis of randomised controlled trials in AJKD involved 53 studies involving 4,222 participants to study the impact of induction and maintenance immunosuppression on outcomes of lupus nephritis. Authors conclude, “Evidence for induction therapy for lupus nephritis is inconclusive based on treatment effects on all-cause mortality, doubling of serum creatinine level, and end-stage kidney disease. MMF, calcineurin inhibitors, or their combination were most effective for inducing remission compared to IV cyclophosphamide while conferring similar or lower treatment toxicity. MMF was the most effective maintenance therapy.”

The superiority of MMF as maintenance therapy comes from good evidence, however, author’s conclusion- “MMF, calcineurin inhibitors, or their combination was the most effective for inducing remission” needs to be interpreted with caution. An outcome, based on surrogate endpoint at 6 months, maybe OK for a trial assessing PFS (progression-free survival) in lung cancer but not in lupus nephritis, where patients typically present in their 20s, 30s and expect much more than just remission of proteinuria at 6 months. Pivotal NIH trials have clearly shown that effect of various induction therapies may not differ from each other until several years of follow-up.  Most trials (especially those with CNI+MMF) lack information on long-term safety and efficacy and superiority of this regimen comes from 2 Chinese trials with limited follow-up.

Issues with network geometry (dots representing CNI with or without MMF aren’t well connected), conceptual heterogeneity (various definitions for response), fewer events contributing to hard outcomes like death, ESRD, doubling of creatinine (due to lack of adequate follow-up or trials not reporting it), known effect of CNI to modify proteinuria without immunological remission, era effects (different treatments used in different time frames), race/ethnicity factors affecting response- are other important issues that may limit generalizability of these results. 

5 Quantifying Post-donation Risk of ESRD in Living Kidney Donors

Based on various cohort studies, we now know that the risk of ESRD after kidney donation is low as such but higher than healthy non-donors. Here is a nice summary of available evidence by @silvishah on NephJC. We may be able to estimate the pre-donation ESRD risk in select populations (calculator here); however, these estimates are on the basis of nondonor populations, and estimate baseline ESRD risk, i.e., long-term risk of ESRD if the individual does not donate. When a patient’s family member asks you in the clinic if he/ she’d be on dialysis in the future because of the donation, you really wonder what the answer should be. Estimating the absolute risk post donation in a given potential donor is challenging.

In an effort to answer this question, Massie et al analyzed the data from 133,824 living kidney donors in SRTR database. The risk of ESRD for the median donor in the first 20 years post-donation was estimated at 34 cases per 10,000 donors. This risk varied based on certain donor characteristics. Higher BMI, male sex,  black race, first-degree biologic relatives of their recipient increased ESRD risk. Risk also increased with age in non-black donors. So huge was the variation that for the 1% of donors at the lowest estimated risk, this number was seven ESRD cases per 10,000 donors and for the for the 1% of donors at highest estimated risk, it was 256 ESRD cases per 10,000 donors. These results will certainly help to inform the potential donors of the possible risks. It is still a challenge to estimate the risk in non-white, non-black populations.

6 Effect of Sofosbuvir-Based Hepatitis C Virus Therapy on Kidney Function in Patients with CKD

Directly-acting antivirals (DAAs) have changed HCV firm a chronic disease to a curable infection. For people with eGFR > 30 ml/min per 1.73 m2, the regimen selection is the same as that for people without renal impairment. As kidney is the major elimination route, safety and efficacy of sofosbuvir-based regimens in patients with CKD are unclear. Because the DAAs tested in RCTs in CKD and dialysis population are not yet widely available, sofosbuvir-based regimens are often used by experts in this population.

 Sise et al present their real-life experience in a retrospective study of 98 patients with CKD (eGFR > 30 ml/min per 1.73 m2) who received sofosbuvir-based therapy (other drugs included simeprevir or ledipasvir with or without ribavirin). 58% had eGFR < 60 ml/min per 1.73 m2. Most of them had genotype 1. Patients with genotype 1 responded favorably with SVR 89% in 1b and 83% in 1a. Patients with more advanced CKD were more likely to be cured with direct-acting antiviral therapy. The slightly higher incidence of adverse events in the study possibly is related to ribavirin (which was used in 28 of 98 cases). The majority of the cohort had a stable renal function. There were seven events of creatinine rising >1.5 times above baseline while on therapy- these were considered unrelated to the sofosbuvir-based therapy. As mean age of the cohort was 62, and eGFR was above 45ml in 77% of the cases, the cohort may not be representative of real-world CKD in practice.

Until the DAAs tested in RCTs involving patients with advanced kidney disease become widely available, we may continue using sofosbuvir in patients with CKD.


Interferon-free regimens for patients with eGFR < 30 ml/min per 1.73 m2
Genotype 1 Elbasvir- grazoprevir ± ribavirin

Ombitasvir- paritaprevir- ritonavir plus dasabuvir ± ribavirin

Glecaprevir- pibrentasvir

Genotype 2 Glecaprevir- pibrentasvir
Genotype 3 Glecaprevir- pibrentasvir
Genotype 4 Elbasvir- grazoprevir ± ribavirin

Ombitasvir- paritaprevir- ritonavir

Glecaprevir- pibrentasvir

7 Oral Anticoagulants to Prevent Stroke in Nonvalvular Atrial Fibrillation in Patients with CKD Stage 5D: A NKF-KDOQI Controversies Report

Non-valvular atrial fibrillation is common in patients with CKD. Both CKD and atrial fibrillation increase the risk of stroke in patients on dialysis. Oral anticoagulation effectively decreases the risk of thromboembolic events in general population; however, it is unclear if oral anticoagulation prevents strokes in patients with AF who are on dialysis. Additionally, anticoagulation is associated with increased bleeding risk which in a patient who is getting heparin dialysis and intradialytic hypertension can be a nightmare.  To add to the woes of the treating physicians, the risk-stratification tools (like CHA₂DS₂-VASc Score for stroke risk and HAS-BLED Score to predict bleeding risk) do not perform well in patients with CKD. Guidelines by various societies differ in their recommendations but all of them acknowledge the clinical equipoise-I would prefer to accept the risk of embolic stroke rather than an intracranial hemorrhage if the bleeding risk is high.

NKF-KDOQI Controversies report published in AJKD briefly summarizes the extent of the problem and the available evidence about efficacy and safety of anticoagulation. More importantly, the paper enlists research questions to be answered. Adequately powered RCTs are essential; it will be interesting to have the results of RENAL-AF (apixaban vs warfarin in hemodialysis patients with AF).

Check out @hswapnil’s Tweet:


8 Taurolidine-Based Catheter Lock Regimen Significantly Reduces Overall Costs, Infection, and Dysfunction Rates of Tunneled Hemodialysis Catheters

Catheter infections and dysfunction are the two major complications of dialysis catheters and catheters lock solutions may have an effect on both these complications. Ideal locking solution: should be nontoxic, should prevent catheter thrombosis, should be biocompatible with the catheter material, and should be effective against a wide variety of microorganisms and not promote antibiotic resistance.

Taurolidine is a taurine derivative that reacts chemically to denature the bacterial cell wall components and endotoxin and microbial resistance is not a problem. In a multicenter RCT of 106 hemodialysis patients with newly inserted tunneled catheters, Winnicki et al compared taurolidine- based solutions to 4% citrate solution. Taurolidine based regimen included taurolidine-citrate-heparin500 twice a week and taurolidine-citrate-urokinase25000 once a week. Patients in taurolidine group had fewer catheter infections (0.67 vs 2.7 episodes of catheter-related infections per 1000 catheter days). They also had fewer episodes of dysfunction and required thrombolysis less often with resultant cost saving. Baseline catheter infection rate was very low in this study (reflecting adherence to guidelines for preventing CRBSI) and locking as a strategy to prevent infection is likely to work only if it is the second step after you are done with the protocols of meticulous handling of catheters in your unit.

9 Perioperative blood pressure management and its impact on postoperative organ dysfunction

‘Maintain euvolemia, avoid nephrotoxic drugs, continue antihypertensives’ is the typical physician consult for the hypertensive patient undergoing major surgery. There is more to managing blood pressure during the perioperative period and in this multicentre RCT INPRESS  (The Intraoperative Norepinephrine to Control Arterial Pressure) published in JAMA, compared the effect of ‘Individualized’ BP management (achieving BP within 10% of the reference value ie, patient’s resting SBP) vs Standard Blood Pressure Management (treating SBP less than 80 mm Hg or lower than 40% from the reference value) during and for 4 hours following surgery.

Primary outcome -composite of SIRS with at least one organ dysfunction by day 7 post surgery- occurred in 56 of 147 patients (38.1%) assigned to the individualized treatment strategy vs 75 of 145 patients (51.7%) assigned to the standard treatment strategy (relative risk, 0.73; 95% CI, 0.56 to 0.94; P = .02; absolute risk difference, −14%, 95% CI, −25% to −2%). Interestingly 48 of 147 (32.6%) in the individualized treatment group experienced AKI versus 71 of 145 (48.9%) in standard BP arm, however, individual components of the primary outcome or mortality wasn’t different between the groups.

In patients with longstanding hypertension, autoregulation curve may shift rightwards and blood flow may become pressure dependent at higher blood pressure limits. A small decrease in the blood pressure during perioperative period may, therefore, be deleterious.

September 2017

1. Prevention of AKI in intensive care unit 

 2017 update on AKI prevention by Working Group on Prevention, AKI section, European Society of Intensive Care Medicine is here. There is no magic bullet at our disposal to do so, however, a number of factors can be modified in the course of critically ill patients to protect their kidneys, and to prevent complications once AKI develops. Authors have summarized the evidence pertaining to volume expansion, diuretics, vasopressor use, sedation use, hormonal manipulations, metabolic interventions, remote ischemic preconditioning, statin use and use of AKI bundle care.

Not delaying urgent contrast-enhanced investigations/interventions for potential preventive measures and a possible preventive role of propofol/dexmedetomidine when used for sedation in ICUs (BPSs-best practice statements which represent ungraded strong recommendations) are some of the new recommendations they propose. A 2B recommendation on the use of atorvastatin or rosuvastatin to prevent contrast-associated AKI in high-risk patients undergoing coronary angiography is controversial. While most of such patients are likely to have one or other indications to use the statin, guidelines suggest their initiation for AKI prevention probably based on the result of this 2016 meta-analysis published in Annals of Internal Medicine. However, as @hswapnil points out in a PubMed comment, the majority of these patients were at low risk of AKI and when only high-risk participants were analyzed, no benefit of statin treatment was observed. Another feature of these guidelines on AKI prevention that strikes to me is the absence of any mention about quality improvement initiatives like NINJA for AKI prevention which can be easily adopted in diverse clinical settings Good that we are realizing the correct time point of action in the course of AKI, where simple interventions and avoidance of harm are likely to be maximally effective.

2. Pharmacogenetics  of Thiazides Induced hyponatremia (TIH) in elderly

About 9% of patients initiated on thiazides develop hyponatremia and elderly are especially prone to get it, some of them develop almost an idiosyncratic response and develop severe dyselectrolytemia. TIH is unpredictable and mechanisms underlying are poorly understood however there is a susceptible subgroup of the population in whom TIH can be reproduced with a single dose of thiazide. This study in JCI was set out to explore this pharmacogenetic predisposition to TIH in two cohorts of patients admitted to the hospital with severe symptomatic TIH in the UK. Phenotype and genotype of TIH cases were compared with controls (patients on thiazides but normonatremic, and general population not on thiazides).  

Cases were typically above 70, predominantly female, had: low plasma osmolarity, inappropriately concentrated urine, more than minimal urinary sodium excretion, and normal thyroid function, low or normal ADH levels, severe hypochloremia, mild hyperglycemia, and intravascular volume expansion. GWAS (Genome Wide Association Study) was done in 48 TIH cases and 2,922 controls (from the 1958 British birth cohort) that showed an association with a variant in SLCO2A1 gene, which encodes prostaglandin transporter (PGT) in the distal nephron (co-localizing with AQP1 and AQP2). Authors hypothesize that under low ADH conditions, apical PGT in the renal collecting duct scavenges PGE2 from the lumen, resulting in AQP2 internalization and minimal osmotic water reabsorption. In the presence of the genetic defect,  apical PGT is reduced or absent, so more PGE2 reaching the lumen is able to stimulate apical EP4 (E prostanoid 4) receptors, resulting in the insertion of AQP2 and thus increased osmotic water reabsorption.

While sample size studied was small, and findings need to be confirmed in other populations, this is by far the largest phenotype-genotype documentation of TIH, and have generated an interesting hypothesis about the poorly understood entity of TIH. After seeing some of the most severe dyselectrolytemia cases with thiazides, I am scared to start someone above 60 on these drugs, and thanks to JNC 8 who have controlled enthusiasm with this class of antihypertensives.   

3. Belatacept versus Tacrolimus in Kidney Transplantation  

Here is a (very) small, single center, open label, Dutch RCT comparing belatacept with tacrolimus based immunosuppression after kidney transplant that showed a higher incidence of biopsy proven acute rejection in belatacept treated group (55% vs. 10%; p = 0.006). Authors tried exploring potential biomarkers, namely CD8+CD28-, CD4+CD57+PD1- and CD8+CD28++ EMRA T cells (all of these reported to be associated with AR in belatacept treated patients) -none of these predicted AR in this study. Also, flow-cytometric measurement of CD86 occupancy on monocytes by belatacept was assessed as a marker of the adequacy of its dose (which was found to be adequate).

Interpretation of these results is limited by very small sample size and therefore the chance of type 2 error. Belatacept is widely viewed as a promising agent that will address burning issues like chronic CNI toxicity, and antibody mediated chronic allograft injury. However, this trial and some recent observational data (here and here) highlight the need for larger trials comparing belatacept with tacrolimus based immunosuppression which is the current standard of care. Here is a 2017 review summarizing potentials and limitations of this agent. 


4. Dialysis Following Transcatheter Aortic Valve Implantation (TAVI)

This analysis of UK TAVI registry, involving 6,464 procedures performed between 2007 to 2014, reported data on predictors of post procedure dialysis and impact of pre- and post- procedure dialysis on mortality in short (30 days) and long term (4 years). Dialysis need decreased with time (6.1% to 2.3% over 7 years). Lower baseline renal function, year of procedure, impaired left ventricular function, diabetes, use of an Edwards valve, a non-transfemoral approach, need for open surgery, and moderate-to-severe aortic regurgitation after the procedure independently predicted need of dialysis after the procedure. Dialysis need after TAVI was associated with high mortality at 30 days (HR 6.44; 95% CI 4.87 to 8.53) and at 4 years (HR 3.54; 95% CI 2.99 to 4.19; p < 0.001 for all) compared with patients without dialysis requirement. This risk was probably unrelated to dialysis procedure itself as patients already on dialysis before procedure had a lower risk of death as compared to those who needed it post procedure.

This study is the largest report the incidence rate of the need for dialysis after TAVR and registry included all the procedures performed over 7 years. However, as is expected in registry data, several important risk factors like the need for prior balloon aortic valvuloplasty, contrast volume, bleeding, and recovery of renal function after dialysis might not have been included. TAVI use is increasing worldwide and be prepared to see this ‘aortorenal syndrome’ following TAVI especially if your center has started it recently, and patients have other traditional risk factors for AKI like diabetes and heart failure.

5. Renal denervation for the treatment of hypertension 

After it failed to demonstrate its utility in resistant hypertension in SIMPLICITY 3, I thought renal denervation (RDN) was dead as an antihypertensive measure. According to authors of SPIRAL HTN-OFF MED trial in Lancet, RDN is not dead and it’s the failure of ‘complete’ denervation with the previous techniques used, that underlies the lack of efficacy in large trials. So authors speculate that if enough nerves are dead, RDN may again take a breath of life in the therapy of hypertension. This is a proof of concept, double blind, multi center, sham controlled RCT involving 80 patients (38 RDN, 48 sham control) after a renal angiogram. All of them had moderate hypertension (unlike severe/resistant HTN where it was initially tried) and had to be off the drug for at least 3-4 weeks before getting randomized. The primary endpoint, the change in 24-h blood pressure at 3 months was significantly better in intervened patients 24-h SBP –5·0 mm Hg (95% CI –9·9 to –0·2; p=0·0414), 24-h DBP –4·4 mm Hg (–7·2 to –1·6; p=0·0024). 

These results should be interpreted with caution as they are the results of the third interim analysis (BTW, how many are allowed per trial?) planned to see if RDN works or is futile so that trial can be continued or stopped. It wasn’t powered to conclude on efficacy, moreover, compromised adherence to the assigned treatment (7/38 couldn’t be kept off drugs) lack of long term efficacy/safety data (renal artery stenosis is reported a complication of circumferential RDN), lack of data on patient-centered hard outcomes are important limitations of this trial. With a decrease in SBP ~5 mmHg, most patients treated with RDN  will still need medications to optimally control their BP. Nobody will be ready to expose patients to an intervention reporting some reduction in BP at 3 months. Excellent editorial putting results in context is here.

So we have evidence justifying further evaluation of this technique in large RCT and let’s hope that history doesn’t repeat itself. Let’s wait and see if subsequent SPIRAL HTN-OFF MED trials confirm this proof or challenge the concept itself as was the case with SIMPLICITY. 

6. Early versus late removal of DJ stent after kidney transplantation

Prophylactic ureteric stenting after kidney transplant reduces the risk of major ureteric complications however this benefit comes at the cost of increased risk of UTIs. This multicenter RCT of Early Versus Late removal of DJ stent enrolled 227 participants from  6 transplant centers in the UK. The intervention was early stent removal at day 5 without cystoscopy (the ureteric stent was attached to the urethral catheter using the string of the stent to allow this) or late removal at 6 weeks after transplantation with cystoscopy. Stent-related complications were significantly higher in the late as compared to early stent removal group [36 of 126 (28.6%) vs. 6 of 79 (7.6%); p < 0.001]. This was largely driven by a lower incidence of UTIs in early removal group.

However, this didn’t remain the trial of early or late stent removal alone, as those assigned to early removal were spared of cystoscopy for stent removal but, were exposed to an additional intervention of fixing stent and tip of urethral catheter together intra-op. Due to this, unique set of complications were encountered in early removal group: urethral catheter and stent fell out early (n = 2), string snapped during removal, requiring cystoscopic removal (n = 1), occlusion of the catheter balloon with the string (n = 1) required percutaneous deflation; and difficulty with catheter removal required cystoscopic removal (n = 1). Authors acknowledge the learning curve issues with this technique. Increase in the size of ureteroneocystostomy in an attempt to localize balloon of the catheter may theoretically increase the risk of reflux /ureteric stenosis; however, the study didn’t aim to study these issues. 

Less we mess with the urothelium is better and what matters most in preventing urological complications is not the duration of the stent but the experience of the operator (as suggested in this study and a previous Cochrane meta-analysis). 

7. Clinical Practice Guidelines for Screening and Management of High Blood Pressure in Children and Adolescents

American Academy of Pediatrics (AAP) has released updated clinical practice guideline for screening and management of high blood pressure in children and adolescents. A nice summary that highlights differences between the new and the old guideline (the 4th report)  is here. The document is a comprehensive review; it tries to find answers to a range of key questions: diagnosis, work up, BP goal, the role of lifestyle changes vs antihypertensive drugs, for example.

Unlike adults, hypertension in children is defined statistically rather than a BP cut-off above which there are adverse CV outcomes. Of note, the new BP tables presented in the paper exclude overweight and obese children. Obviously, the BP cut-offs to define hypertension are lower compared to those in the previous guidelines. So buckle up guys, we are going to diagnose an underrecognized problem more often and only long term prospective studies are going to tell us if we using a right cut off.

The BP tables that have BP percentiles according to gender and height percentile are not user-friendly. In fact, these scary looking tables are an impediment to the diagnosis of hypertension (here).  For busy clinicians, the guideline includes a new, simplified table for initial BP screening. These are the numbers above which one should reassess to confirm the presence of hypertension. These numbers are based on the 90th percentile BP for age and sex for children at the 5th percentile of height, which gives the values in the table a negative predictive value of >99%. Even if this could simplify things, I think it has a potential to add to the confusion. The simplest thing would be to develop an iOS/ android app that all the nurses, pediatricians and nephrologists can easily have on their phones and see if the BP reading is high for the age, gender and height!

Guidelines emphasize on the specific role of ambulatory BP monitoring in diagnosis and monitoring of hypertension. In healthy children, the committee recommends BP be measured annually in children aged 3 years or above.

BP treatment targets in the 2017 AAP CPG are: goal BP is <90th percentile for age, or <130/80 mm Hg, whichever is lower (based on office/casual BP readings). Based on the data from ESCAPE trial, the guidelines recommend that in children with CKD, BP should be monitored by ABPM, and the recommended goal BP is a 24-hour mean arterial pressure < 50th percentile.


8.Mycophenolate mofetil is inferior to tacrolimus in sustaining remission in children with idiopathic steroid-resistant nephrotic syndrome LMIN Sep

CNIs are effective in achieving and maintaining remission in children with the steroid resistant nephrotic syndrome (SRNS) side effects like nephrotoxicity are a concern in long term.   In this randomised trial by Sinha et al, 60 children with SRNS who had achieved complete or partial remission on tacrolimus and ramipril were assigned to either MMF or tacrolimus with the tapering dose of prednisolone for 12 months. The primary endpoint-the proportion of patients with a favorable outcome (sustained remission, infrequent relapses) at one year was significantly lower (44.8%) in the MMF group than in the tacrolimus group (90.3%). Replacing tacrolimus with MMF after six months of tacrolimus therapy for SRNS in children was associated with significant risk of frequent relapses or recurrence of resistance.

In an NIH sponsored RCT of children and adults with steroid-resistant focal segmental glomerulosclerosis, although statistically non-inferior, MMF with high dose dexamethasone was only able to achieve complete or partial proteinuria remission in 33% of patients, compared with the 48% of CNI-treated patients. Moreover, in this trial, it’s difficult to separate the effect of high dose dexamethasone from MMF. So it looks like CNIs are the best available therapy for inducing and maintaining remission in SRNS.


9. Furosemide/Fludrocortisone Test and Clinical Parameters to Diagnose Incomplete Distal Renal Tubular Acidosis (dRTA) in Kidney Stone Formers

Ammonium chloride loading test is considered the gold standard to diagnose incomplete dRTA whose only feature may be nephrolithiasis. However, many patients do not tolerate ammonium chloride because of GI side effects.

Furosemide/fludrocortisone test could be an alternative. Fludrocortisone stimulates electrogenic sodium absorption in the collecting duct via ENaC channel. Furosemide delivers sodium from the thick ascending limb to the collecting duct, stimulating the epithelial sodium channel–driven, luminal-negative potential difference. A person without distal RTA can increase H+ excretion and urine pH would decrease below 5.3.  In this prospective cohort study of 170 stone formers, by Dhayat et al  8% were diagnosed to have incomplete dRTA based on ammonium chloride loading test. Considering this as the gold standard, sensitivity, and specificity of the furosemide/fludrocortisone test were 77% and 85%, respectively, yielding a positive predictive value of 30% and a negative predictive value of 98%. In other words, if a person can acidify urine in furosemide/fludrocortisone test, distal RTA is almost ruled out.

Ok, now if you do not want to perform a ‘provocative test’,  a fasting morning urinary threshold pH < 5.3 with a plasma potassium threshold > 3.8mEq/L may be good enough screening test (negative predictive value of 98% with a sensitivity of 85% and a specificity of 77% for the diagnosis of incomplete dRTA). In busy clinics, simply checking urine pH and serum K may help detect dRTA and this diagnosis may change the life of that recurrent stone formers.


10. Liraglutide and renal outcomes
It seems everyone else than nephrologists is interested in protecting kidneys of diabetics. After CANVAS and CANVAS R that we reviewed in detail in July, prespecified secondary analysis of LEADER trial in this weeks NEJM concludes that liraglutide offers renoprotection: renal outcome occurred in fewer participants in the liraglutide group than in the placebo group (268 of 4668 patients vs. 337 of 4672; hazard ratio, 0.78; 95% confidence interval [CI], 0.67 to 0.92; P=0.003). And what is that renal outcome benefit exactly?-fewer patients on this drug progressed to dipstick positive proteinuria :161 vs. 215 patients; hazard ratio, 0.74; 95% CI, 0.60 to 0.91; P=0.004). Almost all the shortcomings of CANVAS hold true for LEADER as well and I’m not willing to buy this renoprotection benefit until it is replicated in well-designed trials looking at outcomes of patient interests.


In an interesting report, nephrologists from Cape Town, South Africa used 1000 mg of rabbit ATG (yes, some grafts are extraordinarily precious),  10 mg/ day of tacrolimus,  2 gm/ day of mycophenolate, injections of methyl prednisone (500, 250, 250 mg) followed by  30 mg/ day of prednisone for 3 months for induction immunosuppression. Arterial thrombosis needing re-exploration and re-anastomoses, urethrocutaneous fistula needing a suprapubic catheter, catheter occlusion due to bladder clots, fungal infection due to Alternaria alternata and finally a successful outcome! This is not a kidney transplant of a high immunologic risk recipient (although nephrologists were in charge of immunosuppression), but a sensational story of first successful penile allotransplantation done in South Africa making the place as original research with an editorial in the Lancet.

The superhuman physical strength and endurance, that this 21-year-old recipient showed, is nothing less than that of Hercules. Many African young men starting their sexual life face penile amputation as a consequence of ceremonial traditional circumcision done under all ‘septic care’-something that Hercules didn’t have to face, except once when his opponent wrestler Diomedes caught hold of his penis as Hercules was about to throw him down [and on a second occasion posthumously when  people started stealing penis of Hercules statue in Paris leading government to decide to put a removable one there!). 

Penile transplant recipient is happy at the end of 24 months, however, unfortunately, couldn’t become a father due to intrauterine fetal death at the term. The result of this transplant might have opened a new chapter in penile reconstruction, but I hope that some resources are also allocated for education to prevents complications arising out of such rituals and a good perinatal care to save mothers and newborns.

August 2017

Glomerulonephritis and Hypertension

 TESTING Corticosteroids in the treatment of IgA nephropathy

TESTING  was a multicenter, double-blind, randomized clinical trial which planned to enroll 750 participants with IgA nephropathy (proteinuria greater than 1 g/d and estimated glomerular filtration rate [eGFR] of 20 to 120 mL/min/1.73 m2) after at least 3 months of blood pressure control with renin-angiotensin system blockade. Steroid (oral methylprednisolone 0.6-0.8mg/kg/day for 8 weeks followed by monthly taper by 8mg) was compared with placebo.  After 1/3rd of the sample size (n=262) was reached, the significantly higher incidence of serious adverse events, in particular, infections in the steroid arm,  led to the decision to terminate the trial early.  Improvement in proteinuria and rate of GFR decline was reported in the treatment arm, however, due to premature termination of the trial, results can’t be considered conclusive. Limited power due to premature termination of the trial might have overestimated both benefits and harms of immunosuppression and this is for sure not the time to abandon immunosuppression in IgA nephropathy. However, several lessons are learned from recently published trials:

One, steroid treatment reduces proteinuria and may slow down the rate of GFR decline. This systematic review of literature summarizes the evidence on corticosteroid use and concluded that steroids have beneficial effect in reducing proteinuria and slowing GFR decline.  Reduction in proteinuria (without an effect on renal function) was also evident in recently published STOP IgA trial which also raised safety concern due to a higher incidence of side effects. Second, lack of systematic reporting of steroid side effects might have underestimated the harm of prolonged steroid therapy in previous trials. Both STOP IgA and TESTING call for a careful assessment of risk-benefit before choosing to immunosuppress. The role of aggressive RASi (as shown by STOP IgA) is underestimated and this should be offered to all the patients. Current guidelines of prolonged steroid treatment in IgA nephropathy with subnephrotic proteinuria will need to be reviewed in the light of  STOP IgA and TESTING trials. We need better tools to identify right patients, the degree of proteinuria alone may be too crude a measure. Third, whether the risk-benefit profile of immunosuppression can be improved by adding steroid sparing agent, antibiotic prophylaxis, incorporating pathology features that predict response, testing for galactose deficient IgA1 in serum remains to be seen. Targeted release formulation of steroids showed promise in NEFIGAN trial(reviewed here) and let’s hope that these results will replicate. Finally, as suggested by authors, caution is needed in the interpretation of eGFR in patients on corticosteroids as it can be affected by steroids (due to sarcopenia and  hyperfiltration)  

We know that some patients with IgA nephropathy need immunosuppression, however, several issues like patient selection, appropriate drug, and dosage, remains to be sorted out.  

 Redefining Lupus Nephritis

Discussion on lupus case by Prof Hase will never be complete without this quote-“The course of the disease is highly variable and unpredictable-this is the only certain thing about lupus nephritis.” A patient who is in remission for years can come with such bad flare that will lead to end stage renal disease. A patient with class IV disease may quickly go into remission and same class IV in an another patient may manifest as renal failure needing dialysis and may recover slowly, often incompletely over months. 

ISN/RPS classification was an improvement over previous WHO classification, however, it does not consider various important histopathological lesions that are increasingly recognized as predictors of outcome. This excellent review in Nature Reviews Nephrology discusses molecular heterogeneity and pathologic complexity of this disease and proposes several modifications in ISN/RPS classification of lupus nephritis. Failure to understand this heterogeneity may underly failure of seemingly promising treatment options in trials. Authors review the evidence showing that two subclasses of class IV disease behave differently. Class IV‑S lesions tend to have more glomerular fibrinoid necrosis and fewer immune deposits than IV‑G lesions, whereas IV‑G lesions behave as a typical immune complex-mediated glomerulonephritis, suggesting pathogenic differences. They propose a treatment algorithm based on histological features that are identified as important predictors of outcome. This article in KI published online studied 32 cases of lupus nephritis who had concomitant MPO ANCA antibodies and compared them with 222 ANCA negative lupus nephritis cases.  ANCA-positive patients had anti-MPO antibodies and were characterized by more class IV-S with glomerular necrosis, higher creatinine, lower C4, higher anti-dsDNA titer. There was no significant difference in the outcome between the two groups, however, as authors conclude, ANCAs appear to influence the histopathologic pattern. Another cohort study in CJASN ( editorial) reports a cohort of 105 patients with lupus nephritis (80% class III or IV) where clinical and histological characteristics associated with renal outcomes were studied. Authors conclude that ISN/RPS classes had a poor association with the clinically relevant outcomes, and prognostication in LN may benefit from the specific assessment of clinical variables and lesions currently obscured in the classification. Awaiting validation of a formal index, authors suggest that at least fibrinoid necrosis, fibrous crescents, and IF/TA warrant explicit independent scoring to assess the risk of progressive renal dysfunction in conjunction with clinical findings. All these are interesting developments, and raise the hope for better assessment of risk and better treatment selections in patients with lupus nephritis. 

 Treating Hypovitaminosis D in children with nephrotic syndrome

Almost every child on steroids for nephrotic syndrome gets calcium and vitamin D supplementation with the hope that it will help minimize steroid induced bone loss. In an RCT published in Pediatric Nephrology involving 48 children with relapse of steroid sensitive nephrotic syndrome, authors studied the effect of vitamin D3 (60,000 IU orally, weekly for 4 weeks) and calcium supplements on bone mineral content and bone mineral density over 6months. In addition, the effect of this intervention on relapse frequency (considering pleiotropic effects) was evaluated. Only thing Vit D supplementation did was to increase blood levels of vitamin D. There was no difference in bone mineral content or density between the groups, in addition, hypercalciuria was noted in a significantly higher number of treated patients which is a safety issue. The practice of looking at serum levels and loading doses can be tempered.  

Hypertension after Hypertensive disorders of  pregnancy 

Pregnancy is a stress test of the vascular health of a woman and 5% of all pregnancies test positive by manifesting preeclampsia. Accumulating evidence suggests that these women continue to carry this ‘vascular risk’ years after delivery as shown in this large cohort study of over 90000 women demonstrating an increased risk of morbidity due to atherosclerotic complications in long term and interestingly odds ratio for renal hospitalisations following preeclampsia (OR 3.7; 95% CI 2.2 to 6.0) was higher than cardiovascular events (OR 2.4; 95% CI 2.2 to 2.8). Here is nationwide cohort study in BMJ from Denmark which adds to our understanding of this risk. 23,235 of 4,82,972 women had a hypertensive disorder of pregnancy in their first pregnancy, and 16 611 developed hypertension during follow-up. 14-32% of women with a hypertensive disorder of pregnancy in their first pregnancy developed hypertension in the decade after delivery, compared with 4-11% of women with normotensive first pregnancies. Rates of post- pregnancy hypertension in women with a hypertensive disorder of pregnancy in the most recent pregnancy was 12-fold to 25-fold higher in the first year post-partum and up to 10-fold higher in the decade after delivery. Risk persisted long after delivery-remained doubled more than 20 years later. So follow up should begin soon after delivery and continue for at least 2-3 decades.

What can be done to modify this risk? This prospective cohort study reports data from Nurses Health Study II on 54,588 parous women aged 32 to 59 years having data on reproductive history. They studied various risk factors for the development of chronic hypertension including 4 lifestyle factors-post pregnancy BMI, DASH diet, physical activity and sodium/potassium intake. Being overweight or obese was the only lifestyle factor consistently associated with higher risk of chronic hypertension. It’s difficult to separate the effect of first 3 lifestyle factors though. So whatever way you achieve it, help them maintain a healthy weight after pregnancy. 

If I’d been born a woman and given a chance, I’d choose Denmark. For millions of less fortunate women preeclampsia is not just a stress test or an opportunity to address metabolic risk; for them, surviving through a preeclamptic pregnancy is still a difficult challenge. For them to get admitted to the casualty, convulsing, with severe hypertension, intrauterine fetal death, and multi-organ dysfunction is not a rare scenario.


What is AKI risk attributable to contrast after intra-arterial contrast exposure in the primary percutaneous coronary intervention(pPCI)?

In this report from Israel, 2025 patients with ST-segment–elevation myocardial infarction who underwent pPCI (cases) and 1025 patients receiving fibrinolysis or no reperfusion and were not exposed to contrast material during the first 72 hours of hospital stay (control group) were studied. AKI rates were similar in the pPCI and control groups (10.3% versus 12.1%, respectively; P=0.38). Even after propensity score matching, AKI rates were not significantly different with and without PCI (8.6% versus 10.9%, P=0.12). 

In addition to general limitations of the propensity score matching, some important limitation of this data should be noted. The study spanned over 15 long years and indication for pPCI changed over time–the possibility of patients at high risk of AKI getting preferentially treated by thrombolysis remains. Over years, awareness and practices might have improved and this could well be the reason for better outcomes after pPCI (after 2007 all cases were treated with pPCI). Finally, it’s not clear if missing data on creatinine favored pPCI group. Nevertheless, this study represents the first to include a control group of patients who were not exposed to contrast in the setting of STEMI.  

 So, intra-arterial contrast associated AKI  is going through the same difficult time as intravenous contrast AKI. “Oh, come on, the risk may not be great, but it exists, there is a strong basic science support to this entity. What is the harm in being just cautious?” is another argument. A couple of weeks ago, we treated a young female on maintenance dialysis who got admitted with dengue hemorrhagic fever, massive hematemesis, hematochezia, requiring multiple pack cells and platelet transfusions. Upper and lower gastrointestinal endoscopies didn’t reveal any bleeder and it was decided to go for an angiogram and endovascular intervention. Radiology fellow who initially just refused to carry out the scan got convinced only after we threatened to wake up his professor in the middle of the night. How can we blame him-this is what ‘scared to marrow’ means. I know of one of the largest cardiac centers in this region which as a policy denies any diagnostic/therapeutic procedure involving contrast in all patients with serum creatinine>2mg/dl. This harm of being ‘just cautious’ will never ever be reported in the literature. Another problem is the disproportionate academic and research space that this entity occupied for years in nephrology-was it worth? 1/3 of AKI guidelines talk about contrast AKI. Are not obstetrics, tropical and cardiac surgery associated AKI specific enough entities? Serum creatinine may increase around the time of contrast administration but is not worth the scare, space, priority.  Disclosure: I am neither married to a cardiologist nor have joined hands with the radiologist (as some of my colleagues are suspicious). 

APOL1 and CKD: is mystery over?

What came into being as survival advantage against sleeping sickness, –polymorphisms in the APOL1 genes (that encode apolipoprotein L; G1 and G2 being risk variants and G0 the reference allele)– is now one of the most important risk factors for kidney disease that is peculiar to African-Americans i.e. non-diabetic CKD, FSGS, and HIV associated nephropathy. While epidemiological observations and genetic studies (reviewed here) have strongly suggested the role of APOL1 in mediating kidney disease in this population, not all individuals and animal models carrying this risk genotype develop kidney disease –suggesting the role of a ‘second hit’ for disease expression. suPAR was reported to be the ‘circulating permeability factor’ causing FSGS but is now considered to be independently associated with incident chronic kidney disease and an accelerated decline in the eGFR. 

To study molecular mechanisms involved in kidney disease expression by APOL1, authors studied two large, unrelated cohorts [Emory Cardiovascular Biobank (EmCAB) and the African American Study of Kidney Disease and Hypertension (AASK)] for APOL1 genetic variants and plasma suPAR levels.  The apol1-related risk was attenuated in patients with lower suPAR and strengthened in those with higher suPAR levels. Using surface plasma resonance (a powerful technique to monitor affinity and selectivity of biomolecular interactions) experiments in mice, they noted that the presence of AOPL1 risk alleles (G1, G2) augments alfa5,beta3 integrin activation on podocytes leading to proteinuria in the presence of high suPAR concentrations. Authors conclude that the synergy of circulating factor suPAR and APOL1 G1 or G2 onalfa5,beta3 integrin activation is a mechanism for CKD in this population. suPAR is a marker of inflammation and immune system activation-it can originate at podocytes or circulation. While these are novel and exciting findings, what are the upstream events leading to elevated suPAR? This keeps open possibilities of ’ second hit’ before clinical manifestation of APOL1 risk alleles. Let’s hope that we get to see a replication of these findings soon and more information on interactions between this tripartite.  

Proton Pump Inhibitors and CKD

PPI and kidney disease has been a topic of discussion and debate recently – this review in Indian Journal of Nephrology by Malavade and Hiremath summarizes the evidence and briefly discusses the shortcomings of observational data. While large observational studies have found that PPI use is associated with a higher risk of CKD,  limitations like unmeasured confounding, residual confounding, confounding by indication, outcome misclassification, inability to account for different eras when drugs were used-all should be carefully considered and therefore PPI-CKD relationship can’t be considered causal. A host of other complications like hip fracture, community acquired pneumonia, Clostridium difficile infection, and dementia have been linked to PPI use, and, as the authors point out, findings could relate to the same sources of bias. Large RCTs to answer this question are unlikely, and this is an opportunity to learn basics of pharmacoepidemiology, and various pitfalls of observational research excellently reviewed in this NDT article. PPI will be definitely indicated to prevent ulcer bleedings and avoid surgeries in some patients, however, unnecessary use has to be actively recognized and discouraged. The article provides a clinically useful algorithm to decide if PPIs can be deprescribed in an individual case. Huge scope for deprescribing here. 

  Dialysis and Transplantation  

Morbidity and mortality after parathyroidectomy (PTX) 

Approximately 10% of the US patients on dialysis undergo PTX for unremitting hyperparathyroidism-this remains so even with the advent of calcimimetics. This large retrospective national database review of over 21000 PTXs, give important insight about surgical complications following this procedure. Morbidity and mortality were significantly higher when PTX was done for secondary versus primary HPT (26.8 versus 4.9%). Pre operative renal function, hypertension, and alkaline phosphatase were predictors of complicated postoperative course. Patients with tertiary hyperparathyroidism faired better (morbidity and mortality 21.8%) probably because their operative risk was modified by RRT (dialysis or transplant). Most of the complications were respiratory, related to neck hematoma and airway compromise indicating the need for intensive post operative monitoring. Although retrospective design and lack of information on outcomes after 30 days are important limitations of the study, this is an important piece of evidence that calls for a more careful assessment of risk-benefit before resorting to PTX. So PTX in secondary HPT is not as safe a procedure as it is in primary HPT. In real life, identifying the right patient for surgery may be an easier task than finding right hand and center to do PTX -not many surgical centers have required interest, surgical skills, and expertise to manage postoperative medical and surgical issues.

Treating the itch

Almost 50 percent of patients on dialysis complain of itching and in some of them, it may interfere with sleep and quality of life. This systematic review of the literature (RCTs) summarizes the evidence on treatment of this poorly understood symptom and found that whatever best evidence we have supports the use of gabapentin for itch.

I was amazed to know that 39 different treatment options are available and they were studied in 44 randomized trials-very few other ‘issues’ that our patients face having been studied so extensively. Most studies that compared active treatment against placebo reported some benefit. Overall RCTs examining this question are characterized by all those features that a good RCT shouldn’t have: the small sample size, high risk of bias, flawed methodology. Study heterogeneity didn’t allow for a meta analysis, however, this review highlights the need for a large well designed RCT which will not only assess itch more objectively but also will report on more relevant outcome measures like sleep and quality of life.

Gabapentin/Pregabalin can put patients to sleep, produce dizziness, imbalance and falls, and I try to reserve them for those few not happy with coconut oil/other topicals. A timely NEJM perspective on use and abuse of gabapentin is here.

RAS inhibition in patients on dialysis 

Survival of patients on dialysis remains poor, and nephrology community is desperately (and rightly) in search of treatments that will improve the same. Statins, more dialysis,  phosphate binding, PTH lowering, anemia correction—nothing is working as expected. CV disease being common in the dialysis population, some believe that strategies like RAS inhibition may work as it does in general population, however, there is a lack of good quality data from large RCTs to specifically answer this question. This meta-analysis in BMC Nephrology tries to compile the available evidence from RCTs looking at the effect of RAS blockade on CV events, residual renal function (RRF), and mortality. Authors concluded that RAS blockade protected residual renal function in PD, however, was largely ineffective in reducing the CV events/mortality. ARB may have a favorable effect of reducing heart failure.  

While the debate about ‘utility of large RCT versus meta-analysis’ in guiding therapy decisions will continue infinitely, quality of meta-analysis (nice CJASN review here) will be largely dependent upon the quality of trials that it contains-garbage in and garbage out. Unfortunately, most of the RCTs addressing this important questions are underpowered, and therefore inconclusive. Larger, and blinded trials were more likely to be negative, and benefit on ‘heart failure’ is questionable if this endpoint is not adjudicated (as is the case with RCT showing benefit). In a survey that I conducted on

TwitterTwitter pole

(responders n=87, this also happens to be the combined sample size of two RCTs showing the efficacy of ramipril and valsartan in preserving RRF!) to understand the prevalent practice,  44% believed that RASi offers BP reduction and that’s all. I am with this majority and manage a dialysis population that has been traditionally practicing ‘incremental’ dialysis -well before this terminology appeared in nephrology literature a few years ago. Some may go for ‘decremental’ dialysis-where patient for various reasons may decrease dialysis frequency to 2 per week/1 per week/1 dialysis whenever needed. Those nightmares aren’t rare when patients land up into casualty with potassium report reading “too high” and an ECG that resembles drawing of a ‘sleeping line’ that my 4-year old daughter makes. So we are not in any better position to decide on the use of RASi in dialysis after this meta-analysis. Cardioprotection from hyperkalemia is a priority and given the available evidence, we currently aren’t obliged to follow the recommendation to use these agents preferentially for the treatment of HTN in patients on dialysis if others are serving the purpose.

Continue reading “August 2017”

July 2017

1. Single-Nephron Glomerular Filtration Rate (snGFR) in Healthy Adults

Urinary creatinine clearance 59 ml/min, DTPA GFR  75 ml (gates), 65 ml (plasma sampling), CKD-EPI 90 ml/min, CKD-EPI using cystatin C 102 ml/min–these are ‘GFR’ values by various methods for one of our perfectly healthy 58-year-old donor under evaluation. Which of these should I rely upon? This should probably be one of the top research priorities  here and this NEJM article tells us a way how to. In this study of 1388 live donors, (mean age 44, 55% female), authors studied glomerular number (cortical volume of both kidneys by CT multiplied by glomerular density on kidney biopsy), single nephron GFR (measured GFR -mGFR-by iothalamate divided by glomerular number so actually esnGFR) and its determinants. Total mGFR and snGFR were 115 ml/min (SD 24)  and 80 nl/min (SD 40) respectively. snGFR didn’t vary according to age (until <70), sex or body height (until< 190cm). High snGFR was associated with large nephrons, glomerulosclerosis, height >190cm, obesity, family history of ESRD. Predominantly white population, selection of healthy donors, and a possibility of variation in the glomerular volume distribution within the biopsy are important limitations of this study. Nevertheless, it is an important clinico-patho-physiological exercise telling us — estimating nephron number and snGFR measurement is feasible (even before death and without sacrificing animals), an increase in snGFR is an initial adaptive mechanism that precedes kidney disease. Also, the age-associated decline in GFR is a feature of normal nephron senescence and usually not accompanied by increased snGFR. Are we right then in assigning CKD status based on GFR alone in elderly? This makes case for age calibration in CKD definition and staging and this single intervention can potentially halt the epidemic of CKD! With serum creatinine, only the trend can give you the faintest idea of what’s happening to those 860,000±370,000 per kidney tiny filters. Why then the delusion “eGFR”? Especially, in populations where none of the estimating equations have been formally evaluated.


Renoprotection by SGLT2 inhibitors: hope or hype? 

‘Pancreatic islets are like kidney, initial hyperfunction as an adaptive response becomes maladaptive to give rise to diabetes, similar to initial compensatory hyperfiltration by nephron’——Prof Hase taught us during internal medicine grand rounds a decade ago. Never imagined then that a drug would act along both of these pathways. CANVAS program (CANVAS+CANVAS R) compared Canagliflozin with placebo to assess the effect on CV outcomes in a randomized controlled trial involving 10,142 participants. While effect on primary CV outcome was modest (as compared to EMPA REG OUTCOME trial) , renal outcomes were significantly improved as assessed by :progression of albuminuria (hazard ratio, 0.73; 95% CI, 0.67 to 0.79) and the composite outcome of a sustained 40% reduction in the estimated glomerular filtration rate, the need for renal-replacement therapy, or death from renal causes (hazard ratio, 0.60; 95% CI, 0.47 to 0.77). Significant improvement in renal end point was also demonstrated in a previous EMPA REG OUTCOME trial involving empagliflozin. Both reviewed in depth by NephJC here and here.

After the initial introduction of these drugs in the management of diabetes, my endocrine colleagues were worried that they will end up bothering nephrologists more often for urinary tract infections and AKI. Gears have shifted now, and they wonder ‘what Nephrologist are left to do’ with diabetic nephropathy after this ‘ breakthrough’?. Industry hailed it as the magic pill for nephropathy in diabetes– ‘Cure for diabetic nephropathy found’ read one of the headlines in an online medical news portal. The renoprotective benefit appears almost double than what could be achieved in pivotal RAAS inhibition trials (risk reduction 40% with Canagliflozin vs 16% with losartan in RENAAL, 20% with irbesartan in IDNT trial).  These dramatic results are widely received with great optimism by the scientific community- NEJM reversing section title to ‘ basic implications of clinical research, and JAMA coining the term metabolodiuretic in praise. They will continue to influence journal editors and we will get to read a lot more speculations on how this class of drug saves heart and kidney.  Before we sign off diabetic nephropathy to Boehringer Ingelheim, Eli Lilly, and Janssen, let’s have a closure look at the data to put results in perspective.

Both EMPA-REG OUTCOME and CANVAS  studied renal outcomes as secondary endpoints which should only generate hypothesis and not conclusions.  Renoprotection benefit in CANVAS R was largely driven by softer end points like microalbuminuria (or moderately increased albuminuria), and progression/regression of microalbuminuria- a marker that is not unique to CKD especially in patients with preserved GFR and other metabolic risks. Mean GFR at baseline was >70ml  and <8% of the total study population had albuminuria >300mg/d. The risk of CANVAS population developing kidney failure in remaining lifespan is very low (0.03% at 5 yrs by this risk calculator). Hence, although relative risk reduction appears massive, absolute risk reduction is much smaller when compared to pivotal ARB trials [as baseline risk of renal failure there -RENAL/ IDNT- is about 600 times more(20.67% and 15.5% at 5yrs)]. So you need to treat  54 patients with losartan and 11 with irbesartan to prevent one adverse renal outcome versus 278 patient needing treatment with canagliflozin. Moreover, 31% discontinued treatment, reflecting challenge in handling these drugs in real life. Important safety issues are keto-acidosis (documented by this report telling us how important post-marketing data will be), amputations (1 per 346 patients treated), genitourinary infections (1 per 20 patients treated), hypovolemia (1 per 136 patients treated).

It’s not an easy trade-off and more colors are yet to be filled in this beautiful appearing CANVAS. I will wait for replication before application.

3. Fast renal decline to end-stage renal disease: an unrecognized feature of nephropathy in diabetes  

This review in KI reports data from Joslin Clinic on the rate of GFR decline in a cohort of 364 ESRD patients with T1DM. Almost half of these cases progressed at an extraordinarily rapid rate (GFR decline of 5-70ml/min/yr). No clue as to which of the kidney compartment (glomerulus, vessels, tubulointerstitium) is mediating this progression.Very fast and fast renal decline developed in a significant proportion of patients with only mild to moderate albuminuria.

Authors noted elevated levels of circulating TNFR1 and TNFR2 (TNF receptor) as strong predictors of fast renal decline to ESRD. Whether high levels of circulating TNFR1 and TNFR2 are causal factors or only markers of this course remains unknown. Kidney disease in diabetics is much more complex than we think; this highlights the fact that widely held -“normoalbuminuria->albuminuria->GFR loss”- theory of diabetic nephropathy progression may be too simple to be true in all the cases.  This also raises questions whether albuminuria is a valid surrogate endpoint in nephropathy trials.

Rapid progression is rather underrecognized and less acknowledged feature of kidney disease type 2 diabetes as well. Every one of us must have also faced type 2 diabetics with a relentless decrease in kidney function in your clinic like this-I know how painful, helpless and embarrassing it can be.

4. Everolimus in kidney transplantation: forget nephron, spare the patient

HERAKLES is a multicentric German randomized trial evaluating the use of everolimus in kidney transplantation. 499 de novo kidney transplant patients were randomized at Month 3 to (i) remain on standard CNI (cyclosporine) therapy with mycophenolic acid, (ii) convert to everolimus with mycophenolic acid or (iii) start everolimus with reduced CNI and no mycophenolic acid. Primary efficacy endpoint was the effect of everolimus conversion on eGFR at 12-month follow-up which was better by ~5ml/min in everolimus arms.

Authors conclude, “Everolimus initiation with CNI withdrawal at Month 3 after kidney transplantation achieves a significant  improvement in renal function at 12 months, with a similar rate of acute rejection.” Although it didn’t bother Novartis and authors much, composite efficacy failure (biopsy-proven acute rejections, graft loss, death, drug discontinuation due to lack of efficacy or adverse events) was significantly more likely with everolimus. This is in line with previous 3 meta-analyses that showed increased risk of acute rejections, side effects necessitating discontinuation, graft loss and death associated with mTOR use and its tough to find a patient who will be willing to bear with such cost and risks. For what? a non-melanoma skin cancer which only rarely kills and few ml increase in that number called eGFR. Forget the nephron, spare the patient.

5. AV Fistula creation for dialysis: novel ways to the lifeline 

Novel Endovascular Access Trial (NEAT) is a prospective, single-arm, multicentre study evaluating endovascular approach (using everlinQ endoAVF System, TVA Medical) to create AV fistula for dialysis. Successful creation of AVF was possible in 98% of the cases, functional usability was 64% in those who received dialysis and 12-month primary patency rate was 69% -comparable to surgically created fistulas. Functional usability is much lower than previously reported by the authors in a pilot study (64% vs 96%) reflecting various challenges for success in real world settings (learning curve, patient comorbidities etc). Follow-up is limited to 12 months, and a significant number of patients were not yet on dialysis for whom clinical usability data isn’t available. Even then, results are encouraging and hopefully, this technique will offer an additional option for creation of the lifeline. Another interesting report in this month’s Indian J of Nephrology on an alternative approach to brachiocephalic AVF creation (sparing cephalic vein for future use) and commentary here.

While I was discussing this potential option of endoAVF creation by nephrologist with my interventional radiology colleague, he mocked me by saying,”what’s new there, nephrologists keep creating AVF in the neck while cannulating IJV, and in the kidney, while doing the biopsy!”. I said, “I can understand your pain.” I know few nephrology colleagues here who surgically create beautiful AVFs themselves (although they don’t call themselves Interventional Nephrologists). Given the interest and priorities of our surgery colleagues, this skill demands urgent multiplication among budding nephrologists. What’s the fun of being an Interventional Nephrologist if you can’t do an intervention (much simpler although less sexy than plasties and stenting) which can actually add years to patient’s life?

6. Speculations on salt and the genesis of arterial hypertension

This excellent review in KI challenges our widely held beliefs on the relationship of salt, volume and blood pressure. We know kidney as central in the genesis of hypertension in response to high salt diet since ages, however relationship of dietary sodium intake and blood pressure may not be as simple. Authors review here human and experimental data supporting ‘’vasodysfunctioner” theory of salt-sensitive hypertension according to which mere increase in the ECF volume and cardiac output is not enough to elevate blood pressure.  Normal physiological response to salt loading is, in fact, vasodilatation and failure of this vasodilation (or vasodysfunction) underly salt-induced increase in the blood pressure. They also review their previous seminal work challenging our simplistic belief about sodium balance i.e. salt excretion parallels intake and increased salt intake leads to increased thirst and so increased water intake. All these years we have completely neglected the third body compartment i.e. interstitial fluid which is now considered to be osmotically active. JCI paper by same authors is reviewed in detail by NephJC here. High salt diet increased urinary osmolyte (read urinary sodium) excretion but reduced free water clearance (meaning water was held back in the body). This extra water suppressed thirst leading to decreased water intake. So the water intake was actually least in the high salt intake phase. They also observed rhythmic mineralocorticoid and glucocorticoid surges. Mineralocorticoid-coupled increases in free water reabsorption were counterbalanced by rhythmical glucocorticoid release, with excretion of endogenous osmolyte and water surplus by relative urine dilution. Higher salt intake elevated rhythmic glucocorticoid secretion and suppressed rhythmic mineralocorticoid excretion. These rhythmic patterns of mineralocorticoid and glucocorticoid secretion were not daily; they are weekly or half-weekly. In other words, if you collect a 24-hour sample to assess daily salt intake, you are probably wrong. It’s time to ask for a 7-day urine collection if you go by the book! This is really getting interesting.

7. CKD-MBD guidelines once more

What should guideline expert do when there is little or no evidence? They can ask their nephro friends for sure, who have developed an unique expertise in creating guidelines based upon imperfect evidence.  The 2017 KDIGO clinical practice guideline update on CKD-MBD published last month has very little to add what we already know (or don’t know!) about the mineral bone disease in CKD. As was the case with the last version, most of the statements are either not graded or are 2C/2D. ISN webinar by Geoffrey Block explaining the basis of this update is here. What’s new: 1) BMD testing to assess the risk of fractures in patients with CKD G3a–G5D with evidence of CKD-MBD and/or risk factors for osteoporosis (2B) based on the observational data suggesting BMD predicts fracture risk across the spectrum of CKD. 2) In absence of data supporting benefits of maintaining phosphate in the normal range in CKD G3a–G4 patients and because of safety concerns with phosphate binders, the group suggests lowering elevated phosphate levels toward the normal range rather than attempting to maintain in the normal range. 3) As inorganic phosphate is a constituent of food preservatives,  patients should be encouraged to have fresh and homemade foods, rather than processed foods, to avoid additives. About 40% to 60% of phosphate in animal proteins is absorbed, as against 20%–50% with plant-based phosphate, because of associated phytates. So, rigorous restriction of plant-based proteins may not be warranted and may not be possible without harmful protein restriction

8. Complement Disorders and Malignant Hypertension

A young adult coming with: blood pressure >200systolic, papilledema, subnephrotic proteinuria, markedly reduced kidney function, all for the first time in his life, is not an unusual case scenario here. ‘Malignant hypertension’ can give rise to thrombotic microangiopathy (TMA) that is evident only on kidney biopsy i.e. without other clinical features of TMA. In this interesting report, Timmermans et al hypothesize alternative complement pathway (AP) dysregulation as an unrecognized, but treatable cause of hypertension-associated TMA. To test this hypothesis, they analyzed AP in 9 patients with severe hypertension who had histological criteria of TMA on renal biopsy.  Six out of nine patients had mutations in C3, CFI, CFH, and CD46 with complement activation evident in the plasma (raised C5b-9 level) and in the biopsy (deposits of C3c and C5b-9 along the vasculature and/or glomerular capillary wall). Eight patients progressed to ESRD. Four patients received six allografts and TMA recurred in four allografts (patients with C3 and CFH mutations) -leading to the graft loss in three cases. Hypertension-induced shear stress possibly activates complement pathway leading to the formation of C5b-9 on activated endothelial cells. Eculizumab was started in one patient with CFI mutation who was on dialysis for three months and his renal function improved (estimated glomerular filtration rate: 38 ml/min) within a 12-month treatment period. These patients, for sure, fall in the spectrum of complement-mediated TMA. We do not routinely biopsy patients with severe hypertension, rising creatinine, and not-so-high proteinuria. Challenge is how to identify people who are likely to have complement dysregulation. Wider availability of tests to evaluate complement pathway and having a low threshold for kidney biopsy in such a case may be the way to go.


The prespecified subgroup analyses of outcomes in SPRINT (Systolic Blood Pressure Intervention Trial) participants with baseline CKD got published online ahead of print in JASN. 2646 participants (28.3%) had GFR <60 ml/min/1.73m2 at baseline. Primary composite cardiovascular outcome occurred in 112 intensive group and 131 standard group CKD participants (hazard ratio [HR], 0.81; 95% confidence interval [95% CI], 0.63 to 1.05). The intensive group also had a lower rate of all-cause death (HR, 0.72; 95% CI, 0.53 to 0.99). Treatment effects did not differ between participants with and without CKD (P values for interactions $0.30).

As expected, the intensive group had a higher incidence of ≥ 30% GFR loss in the first 6 months. However, the rate of change in eGFR using the values at 6 months after randomization as the baseline was slightly higher in the intensive group. [-0.47 ml/min per 1.73 m2 per year in the intensive group and -0.32 ml/min per 1.73 m2 per year in the standard group (P <0.03)]—SPRINT-CKD doesn’t tell us what is its implication after 3.3yrs. Hypokalemia, hyperkalemia and AKI were more often seen in the intensive group, but unlike the main study population, hypotension, syncope, hyponatremia and orthostatic hypotension were similar in the two groups.

The strength of the SPRINT CKD data is its size which is larger than MDRD, AASK and REIN-2 combined and thus represent one of the largest existing evidence to treat high blood pressure in CKD. They support CV benefit of aggressive BP lowering in elderly, nondiabetics with a mild decrease in eGFR. Before you start SPRINTING with your patients, let’s look at the population here. This is not a CKD progression trial. The typical patient here is a nondiabetic, elderly about 72 years old (about 3 yr older than the main SPRINT), with baseline GFR ≥ 45 ml/min per 1.73 m2 (in 66%of the patients), having mean urinary albumin excretion of ~80mg/gm. Very few physician colleagues of ours are obliged to refer us such cases (of CKD3a, A2)-that SPRINT CKD is talking about. We need to choose the right population to use this information. Also if you measure the BP the way SPRINT did and want reduce CVD/ deaths notwithstanding the higher pill burden and side effects, and want to get down to 120 (and diastolic BP remains recordable then) here is advice  by Richard Lehman-“If you are consulted by an individual who fits its criteria, you could have an informed conversation about their risk-lowering options, of which tighter BP control might be one.”

June 2017


1. Targeted-release formulation (TRF) of Budesonide for IgA nephropathy

RCT in Nephrology is that rare occasion to celebrate especially if it is positive. Here is a multi-centre, double-blind, placebo-controlled trial (NEFIGAN) of Targeted-release formulation ((TRF) budesonide versus placebo in patients with IgA nephropathy. Gut mucosal B lymphocytes synthesise abnormally galactose deficient IgA which after complexing with IgG antibodies get deposited in the mesangium and initiate sequence of downstream events: stimulation of cell proliferation, release of inflammatory mediators that promote proteinuria, and fibrotic remodelling, ultimately leading to loss of renal function. High dose systemic corticosteroids have poor safety-efficacy profile especially in patients with sub-nephrotic proteinuria and preserved GFR, best illustrated by STOP IgA and TESTING  (results presented at ERA-EDTA 2016-prematurely terminated for safety concerns) trials )Targeting gut mucosal immune system by TRF-budesonide 8 and 16mg capsules (which delivered drug at distal ileum where density of Payers patches is maximum) was compared with placebo in total of 150 patients who were on optimal RAS blockade at the start. TRF Budesonide group had significantly reduced proteinuria, better GFR and lesser microhematuria at the end of 12months than placebo. Short follow up (to conclude on effect of TRF Budesonide on GFR), and predominantly white population are the limitations of this study.  IgA nephropathy in this part of the world is characterised by marked severity -a disease type that hasn’t been represented in any of the IgA nephropathy trials so far. Nonetheless TRF Budesonide appears promising and larger studies involving diverse patient populations are highly desirable.


2. Accepting kidneys from HCV positive donors

Discovery of Directly Acting Anti-viral agents (DAAs) for hepatitis C will be rated as pathbreaking research similar to HAART or penicillins. After initial studies of DAAs, considering exorbitant cost, one of the BMJ bloggers  wrote,”now all the billionaires in the world will be cured of hepatitis C!”  Thanks to the generics and competition in Indian pharmaceutical industry, DAAs are within the reach of most patients needing it today (~INR 30000 for 3 month therapy).  DAAs may also give a new lease of life to the kidney transplant recipients by allowing usage of kidneys from deceased donors with HCV infection (which otherwise would be discarded). THINKER study in NEJM reported, outcomes of 10 HCV-negative adults (of blood group A, B or O, who were undergoing dialysis and who had long anticipated waiting times for a kidney transplant) after deceased donor kidney transplant from a donor with HCV genotype 1 infection. Immunosuppression consisted of IV Steroids and rabbit ATG followed by tacrolimus, MMF, and prednisone. Once the virus was detected in recipient’s blood (which was seen on 3rd post-op day in all the recipients), therapy with elbasvir–grazoprevir was started and continued for 12weeks. All recipients were cured of HCV-defined as a sustained virologic response 12 weeks after the end of treatment. One recipient had delayed graft function, transiently elevated aminotransferase levels developed in two recipients. One patient had proteinuria with kidney biopsy showing FSGS. No graft loss was reported. Although small, this data is highly encouraging and we may be able to recycle most of the kidneys from HCV positive donors in future. 

3.  Looking beyond one year post transplant

Dramatic improvement in short-term results after kidney transplantation today, is in sharp contrast to the static long-term outcome figures. This comprehensive review in Lancet summarises strategies  to improve long-term outcomes after KTR. (unfortunately many potential and few actual). A lot of what was once thought to be -IFTANOS(Interstitial Fibrosis and Tubular Atrophy Not Otherwise Specified)- is attributable to chronic ongoing immune injury to the graft and we are far from precisely diagnosing and treating it. In addition to DSA monitoring which is already the part of standard of care, more sensitive matching techniques like epitope matching done by computer programmes like HLAMatchmaker may add to the precision in assessment of immunological risk. Simple intervention like behavioural change training (specifically for patients with history of non-compliance, wide variations in trough CNI concentrations), minimising pill burden (a risk factor for non adherence), and treatment of common CV risk factors can help improve outcomes. Tolerogenic immunomodulation, revolution in omics technologies and bioinformatics to improve precision in matching and monitoring, and novel tolerance induction methods ….are all in the ‘pipeline’-length of which I am not aware of.

4. Management of patients at risk of AKI

Yes, I agree, we need to start very early in AKI and I am not talking about dialysis. While nephrologist are busy debating timing, mode and dose of dialysis in AKI, this review in Lancet -Management of patients at risk of acute kidney injury  identifies the correct time of action in the course of AKI where simple intervention may potentially improve the prognosis of AKI in ICU (which remains almost unchanged over years). Being used to reading RIFLE, AKIN, KDIGO definitions and classifications, this review is probably the first of its kind in daring to ignore all of them!  Interventions early in the course of AKI (electronic alerts, restrictive fluid strategy after initial resuscitation, avoidance of additional insults, and close monitoring) are simple, not necessarily delivered by nephrologist and are more likely to be fruitful and may be more important than RRT timing, dose, modality. We probably have reached the the limit of mortality benefit by dialysis strategies in AKI.

5. Update on Oxford classification of IgA Nephropathy 

Oxford Classification of IgA nephropathy 2016—the role of crescentic lesion: an update from the IgA Nephropathy Classification Working Group is ready with their recommendation to add C (Crescent) to the earlier MEST score. So now its MEST-C (C0 no crescents, C1 crescent in at least 1, C2 crescent in >25% of the glomeruli). I am not sure how it will exactly alter the way you treat your IgA nephropathy patients, but it will definitely affect score of fellows preparing for their examinations. As nephrologist, we know crescent as the sign of severe injury to the glomerular capillary wall and such patient will need closer monitoring, may be better immunosuppression and will carry a poorer outcome. Whatever said, only biopsy finding that changes treatment plan definitely is “all glomeruli sclerosed with frequent foci of IFTA” — and finally its the old friends who come for help: proteinuria, and GFR trajectory.

6. Liver as a guard for kidney 

Transplantation in sensitized recipients is a nightmare for transplant nephrologist. This thought provocative work by Taner et al from Mayo clinic may be that ray of hope for highly sensitized recipients who are either not transplanted or experience a poorer graft outcomes after transplant. In a case control study by the same group last year, they demonstrated markedly improved graft outcome in 14 recipients of simultaneous liver-kidney transplants (SLK) who had DSA at the time of transplant when compared with 28 recipients of kidney transplant alone (KTA) having DSA (ABMR 46 vs 7%, TG 54 vs 0%, combined endpoint of allograft loss or >50% decrease in eGFR 20 vs 7%). In this paper in KI, authors studied molecular signatures of this protective effect, and found that liver offers more than just DSA adsorption.In some way liver favourably modified B and T cell immunity-SLK recipients expressed less pro inflammatory transcripts, they expressed more transcripts encoding metabolic processes involved in the maintenance of tissue integrity.  This raises hopes whether liver, part of liver, or molecules from liver will optimise outcomes for highly sensitised recipients. With the help of transcriptomics used by authors, there is potential for early identification of antibody mediated injury to graft before histologic changes of TG (which is characterised by relentless progression with or without treatment ) .

7. Disease eradication before therapy getting discovered -story of contrast AKI

I used to wonder if I was missing Contrast Induced Nephropathy (CIN) in our AKI patients. I am yet to see a patient of AKI needing dialysis that can be solely attributed to contrast (at a centre where AKI caseload is about 100 per month). If you still believe that AKI following intravenous contrast is a major problem, this issue of ACKD is for you. Scared to the marrow: pitfalls and pearls in renal imaging, Intravenous contrast: friend or foe?, Intravenous contrast induced nephropathy-the rise and fall of a threatening idea, all of them will help clear the delusion of CIN after IV contrast. CIN after iV contrast will be a unique example of a disease getting eradicated even before a therapy coming into being. Risk of CIN after IV contrast appears much lower (if not nil) than previously emphasised- first suspected by radiologist who were questioning its very existence. Finally nephrologists are in agreement. Absence of appropriate control group, attribution of all rises in creatinine to contrast, random variations in measurement of creatinine especially if high at baseline, lack of data on clinically meaningful outcomes (length of hospital stay, morbidity, mortality etc) call into question the observational data linking contrast with AKI. Same may not be true with intra-arterial contrast use and general measure (that every wise clinician would take-correcting hypovolemia, minimising amount being used, withholding other drugs like diuretics, RAS blockers around the time of procedure) which are considered good clinical practice around any preoperative period should apply here too. Here is an excellent blogpost summarising  the issue.

NSF (Nephrogenic Systemic Fibrosis) reminds me of Gary Cooper’s masterpiece movie High Noon. A vicious outlaw (who is sent to Jail by town Marshal) is getting released and coming back to take revenge. Without a single scene of violence, director succeeds in creating the atmosphere of terror. Finally outlaws come. NSF seems to be that outlaw who isn’t just coming back. Terror continues. No new case of NSF is registered over last several years to the NSF registry.

This article discusses Potential use of MRI in noninvasive assessment of kidney fibrosis looks promising interesting.

8. Comparing safety of Rituximab with Cyclophosphamide in membranous nephropathy

In this retrospective observational study from Italy in JASN, rituximab (RTX) had better safety profile than cyclophosphamide (CP). Adjusted hazard ratios (95% confidence intervals) between RTX and ST-CP groups were 0.27 (0.16 to 0.44) for any first adverse event, 0.32 (0.15 to 0.68) for serious adverse events, and 0.23 (0.13 to 0.41) for nonserious adverse events.

Study’s primary objective was to compare safety of RTX with CP and I find no reason to agree with authors on their bold conclusions regarding efficacy which goes like this: “RTX might be considered as an alternative to ST-CP–based immunosuppression as first-line therapy in patients with IMN and persistent nephrotic syndrome, despite conservative treatment”.

Really? Centre where CP was used was different from that where RTX was used, CP was used in unusually high dosages (36 gm-believe me this isn’t a typo), and time period during which these treatments were used differed considerably (CP from1995 and 2012 and RTX 2001 onwards) producing ‘ era effect’ ; all of these make it almost impossible to conclude anything definitely about safety and efficacy. Although adjustments for baseline confounders was made and sensitivity analyses were performed, patients in CP group were older, had more severe disease at baseline-higher proteinuria, lower albumin, higher cholesterol, lower albumin level, higher statin use! Given the CP dosing protocol in this study, its no surprise that incidence of side effects was considerably higher (when compared to previous studies by Ponticelli and Jha). Authors attribute this difference to the fact that ICH-GCP guidelines (of adverse effect reporting) came into being after pivotal trials and their late adoption in India. I am not convinced by this argument as RCT design is much more robust than observational studies to record all clinically meaningful side effects.

Overall this retrospective, observational study is nothing less than a race of dead horse versus sheep, and I sincerely hope that the RTX will soon find an appropriate disease indication in the field of glomerulonephritis.

9. IgM Nephropathy -a precursor to FSGS?

If you are a believer in IgM nephropathy, then this retrospective cohort study in NDT is for you. Presence of mesangial deposits on electron microscopy and exclusion of systemic disease were required to define the cases in addition to dominant IgM staining. Response rate of 40% with steroid is similar to previously reported series, and predictors of loss of GFR were: presentation with eGFR 20% of glomeruli, interstitial fibrosis and tubular atrophy affecting >20%. 8 of 10 patients undergoing repeat biopsy in this series developed FSGS, with disappearance of mesangial deposits on EM in all eight despite continued positivity for IgM. Authors speculate that IgM nephropathy progresses to FSGS.