Biostatistics for Clinicians: Making Sense of Numbers – Program Recording

We really enjoyed interpreting the published articles with the help of expert national and international faculty in this workshop. We had an overwhelming response from the delegates (370 + registrations) with interactive ‘Q and A’ sessions.

Did u miss the live workshop? Don’t worry we got you covered. Here is the recording of the program –

Day 1 –26th March 2022

Day 2 – 27th March 2022


Making sense of numbers, Interpreting statistics for clinicians

The department of nephrology, Seth GSMC and KEM Hospital, Mumbai invites you to a unique virtual CME on

⚡️Interpreting published Statistics for Clinicians⚡

Let’s learn how to make sense of those numbers published in the results & apply them to your clinical practice.

We have a grand ensemble of speakers & chairpersons from India and abroad.

Click here to register: 


All those who register will also get the recording of the program.

Registrations involving foreign transactions should be made to Mr Kishore Vardham,

Name of Bank – Saraswat bank, Kalwa Thane,
Account no. 061200100018369
Ifsc code: SRCB0000061 SWIFT code – SRCBINBBGHT

Please feel free to reach out if you have any questions at divyaa24@gmail.com

April 2022

PLEX for ANCA vasculitis

Moderate certainty evidence gathered from this updated systematic review and meta-analysis in BMJ says:

a) PLEX has no important effects on all-cause mortality, b) PLEX reduced the risk of ESKD at 12 months, and c) PLEX increased the risk of serious infections at 12 months.

These conclusions are just as definitive and helpful for clinicians as the monsoon predictions of our meteorologic department to farmers, and wise farmers take them just as seriously as the monsoon predictions based on the height of the crow’s nest on the tree.

This paper demands a major correction in the ’ what you already know on this topic’ section which reads: “clinicians use PLEX as a treatment in vasculitis based on the ’biological rationale’ and ’small clinical trials’-completely ignoring the significant practice change (of restricted PLEX USE) that followed PEXIVAS trial; the largest of the trials addressing this question.

The results of this review are largely driven by MEPEX and PEXIVAS trials. Although both were addressing the similar question, for many reasons, they aren’t comparable. Change in the standard of care over the long period of 13 years after the publication of MEPEX is likely an important issue underlying the differing conclusions of these trials: more use of IV CYP (giving the lower cumulative dose), use of lower dose steroids, and overall improvement in the care of patients with renal failure, might have left little room for PLEX in management in PEXIVAS trial.

The ancient question of ’ well-done RCT vs meta analysis’ has once again surfaced here and we prefer to side with the well-conducted large RCT instead of the meta-analysis combining studies of various sizes, designs, and strengths. PEXIVAS was a large and rigorously done RCT (reviewed by us in March 2020 post) after which it’s difficult to justify the routine use of PLEX in this disease. A recent large retrospective study also demonstrated the futility of PLEX in the contemporary population with ANCA vasculitis. Believers of PLEX point to limitations of the PEXIVAS trial: lesser severity of azotemia in its participants, the lack of data on histological severity, and possible benefit in patients with alveolar hemorrhage. Most of these arguments are held by dogmatic belief in the PLEX rather than the available evidence.

Total number of participants 137704
Number on dialysis 137140
Number with alveolar hemorrhage31191 (severe in 61)

The only major limitation of PEXIVAS may be the lack of histological data, which could have guided us about the histological correlates of response to PLEX, and the possibility of the benefit of the PLEX in the subgroup of patients with severe histological features remains unanswered.

What do we take away? A definite new message from this meta-analysis is that PLEX is an immunosuppressive strategy that can increase the risk of infections. How to trade off short-term benefits on kidney function with an increased risk of infections? This is a dilemma that clinicians are likely to face now. Bottom line: In a patient with biopsy suggestive of severe disease and poor response to the standard immunosuppression, one can engage in a discussion about the risks and benefits of the PLEX.

Hypertension and nephrologists

‘Hypertension is a disease of kidney’ Professor Hase would start his lecture on hypertension. Given the central role of the kidney in sodium and volume regulation, there is no exaggeration in saying that essential hypertension is ‘kidney failure without azotemia’. As summarised in this latest review in NEJM, seminal integrated physiology work by Guyton and colleagues, demonstrating the close relationship between sodium, volume, blood pressure and their regulation by the kidney remains valid after 50 years of its initial description and in spite of the challenges by alternative hypotheses of vascular dysfunction and immune system activation in the genesis of hypertension.

Over 80% of the patient with kidney disease have HTN, which reflects the close and bidirectional relationship of kidney function with blood pressure: with increasing severity of kidney disease, HTN gets more severe, and uncontrolled severe HTN drives the progression of kidney disease. Not surprisingly, very few specialists are confronted with ‘difficult hypertension’ like nephrologists, which makes them one of the best HTN specialists (not graded, huge personal COI).

Thanks to renalism, with the increasing severity of CKD, more and more renalism comes in and evidence guiding the hypertension treatment becomes weaker and weaker.

Agarwal et al are all set out to clear the clouds here, and after resurrection of atenolol for HTN in patients on dialysis their CLICK trial has now reported a remarkable reduction in blood pressure in a population which otherwise qualifies for exclusion in HTN trials.

160 patients with stage 4 CKD and uncontrolled HTN (most receiving 3-4 drugs at baseline to control BP, now that’s nephrologist’s HTN!), were randomized to receive chlorthalidone at an initial dose of 12.5 mg per day, with increases every 4 weeks if needed to a maximum dose of 50 mg per day, or placebo; randomization was stratified according to previous use of loop diuretics. The adjusted change in 24-hour systolic blood pressure from baseline to 12 weeks was −11.0 mm Hg (95% confidence interval [CI], −13.9 to −8.1) in the chlorthalidone group and −0.5 mm Hg (95% CI, −3.5 to 2.5) in the placebo group. The between-group difference was −10.5 mm Hg (95% CI, −14.6 to −6.4) (P<0.001). While the degree decrease in the blood pressure isn’t very different than that reported with major first line agents, its noteworthy that this was a population already on more than 3 drugs, and this effect is therefore remarkable. Significant reduction in albuminuria was also observed in patients treated with chlorthalidone group possibly due to diuretic-induced potentiation of antialbuminuric effects of renin–angiotensin system inhibitors.

Limited duration of follow up, small sample size, no inclusion of Asians, and lesser representation of women are notable limitations of the study which demands further large scale evaluation of this class of drugs in CKD HTN focusing on hard CV and renal endpoints. Nonetheless, long held dogma (based on an experimental study done in 1960s ), that thiazides are largely ineffective in treatment of HTN in advanced CKD is challenged by this study. CLICK trial is a welcome and long due addition to the limited evidence base to guide antihypertensive therapy in this population.

Sodium intake in heart failure (HF)

Like cirrhosis and nephrosis, HF is a state of poverty (reduced EABV)amidst prosperity (net salt and water excess) and one usually recommend sodium restriction of less than 3 gram per day to patients with heart failure. This recommendation is based on expert opinion and low-quality evidence. This leads ACC/AHA to recommend this:

Guideline in fact doesn’t use the word restriction and rather limits itself to ‘avoiding excess’ in the absence of any credible evidence on the benefit of sodium restriction. It is disheartening to note that one of the “cornerstone recommendations” for heart failure is based on flimsy evidence. Understandably there was a lot of excitement about the SODIUM-HF trial which attempts to answer this important question.

SODIUM-HF is an international, multicentre, open-label, RCT that enrolled 806 adult chronic heart failure who were randomly assigned (1:1) to either usual care or a low sodium diet of less than 1500 mg/day. The primary outcome was the composite of cardiovascular-related admission to hospital, cardiovascular-related emergency department visit, or all-cause death within 12 months. The investigators found the primary outcome had occurred in 60 (15%) of 397 patients in the low sodium diet group and 70 (17%) of 409 in the usual care group (hazard ratio [HR] 0·89 [95% CI 0·63–1·26]; p=0·53).  They concluded that a dietary intervention to reduce sodium intake did not reduce clinical events. 

SODIUM-HF set out to answer a vital question and based on the results, the authors emphatically claim: “study is the largest randomised clinical trial to test a strategy of dietary sodium reduction for patients with heart failure to date” and “study provides high-quality evidence to guide clinical decision making”. While the first half of the claim is true (trial is indeed the largest one on this topic), the second is far from the truth. Before getting carried away by the lofty claims of the authors, let’s examine the study in some more detail.

First, the study isn’t sufficiently powered to address the question. The authors assumed the event rate of 25% in the control arm and a relative risk reduction of 30%. The actual event rate in the trial was 15% and with this event rate, to detect a 30% reduction, we need 1722 patients. Also, over 50 patients discontinued study treatment for various reasons further compromising power. Second, the dietary recall method to check that intervention is adhered to is unreliable. In a single-blind(assessors were blinded) study, where patients were aware of the intervention, it’s likely to introduce ‘co-intervention’ where patients in the control arm can restrict the sodium intake blurring the separation in the groups. This can be inferred from the post-intervention median sodium intake in the two groups: 1658 mg/day (1301–2189) vs  2073 mg/day (1541–2900) in the intervention and control arms respectively. This wide variation in the intake of sodium in the control arm is suggestive of ‘contamination bias’. Moreover, the intervention arm was telephonically contacted more often (almost monthly), raising suspicion of the performance bias. Third, there seems to be major confusion about the amount of sodium prescribed to the participants, which differs in the protocol registered and the published paper-which everywhere wrongly equates less than 1500mg to less than 100mEq (when actually 1500mg equals 65mmol). Kudos to the reviewers of Lancet! Fourth, study protocol (supplementary appendix) talks of the primary endpoint assessment at 24 months, while paper reports it at 12 month, leaving us clueless as to the reasons of this change in the goalpost. Finally, not only the authors claim exact sodium intake based on recall, they also report the exact fluid intake (e.g.median intake 1782 ml). Ordinarily, thirst (and water intake) is inversely related to salt intake which was not observed in this study. We are also curious to learn their technique of quantifying fluid intake so accurately!

Given these issues, study only adds to the poor quality of the evidence on the salt restriction in heart failure.

February 2022

Balanced fluids versus saline: is the jury out?

“Hyperchloremia produces progressive renal vasoconstriction and fall in GFR that is independent of the renal nerves, is potentiated by prior salt depletion, and is related to tubular Cl- reabsorption. Chloride-induced vasoconstriction appears specific for the renal vessels” reads the conclusion of the seminal paper by Christopher Cox published in JCI 1983. After a prolonged quiescence of almost three decades, the harm of saline became the topic of debate and discussion following the publication of this sequential period pilot study in JAMA 2012, which reported a dramatic reduction in the risk of AKI after switching to the strategy of chloride restrictive fluid administration. Although, we often consider this study as the starting point of this ‘fluid debate’, the guys who performed a small but elegantly done 2012 pilot RCT addressing this question were a group of surgeons, with rather aberrant research interest in fluid therapy (we often joke about their knowledge of renal physiology, especially fluids, but some of them can be really cool!).

A sequential period pilot study wasn’t considered definitive enough to change the practice, for several reasons like ‘before-after design’, the possibility of Hawthorne effect, and doubtful plausibility of its rather massive effect size (40% decrease in the risk of AKI). Several large trials were subsequently published examining the harm of (ab)normal saline (if you are a tonicity freak, who like to call saline abnormal): SPLIT (2015), SALT-ED & SMART (2018), BaSICS (2021), with the latest addition of PLUS trial (2022) which will be discussed here.

In this double-blind RCT, conducted across 53 ICUs in Australia and New Zealand, involving 5037 critically ill patients, 2515 patients were assigned to the BMES (Balanced Multi Electrolyte Solution) and 2522 to the saline. The primary outcome of death from any cause within 90 days was no different in BMES vs saline [21.8 (530 of 2433 patients) vs 22% (530 of 2413 patients), 95% CI −3.60 to 3.30; P=0.90]. Similarly, secondary endpoints-mainly renal-receipt of new RRT [12.9% (310 of 2394 patients) vs 12.7% (306 of 2403 patients) 95% CI, −2.96 to 2.56] and the maximum increase in the creatinine level during ICU stay (0.41±1.06 mg/dl vs 0.41±1.02 mg/dl) showed no difference.

How do we reconcile seemingly contradictory results of SALT ED, SMART vs SPLIT, BaSIC and PLUS?

Pragmatic trials are relatively simple and easy to conduct, more aligned with the standard clinical practice, can recruit a large number of patients over a shorter time span, and are considered to have higher external validity. However, they are best suited to examine ‘whether an intervention actually works in real life’ after an explanatory trial has shown ‘if and how an intervention works (BaSIC and PLUS were needed before SALT ED and SMART). One of the 9 essential PRESIC 2 characteristics of pragmatic studies is the choice of the primary outcome which preferably should be a patient-centered hard outcome. This is especially important when trials have a very large sample size and can detect the small difference in the outcome which may be statistically significant but clinically irrelevant. In the SALT ED trial, the primary outcome of hospital-free days didn’t differ between the groups, and renal benefits (MAKE) were assessed as the secondary outcome. While SMART assessed MAKE as the primary outcome, the difference observed was small and none of the individual components of MAKE was significantly different between the groups. More flexible conduct of the trial is also reflected by the smaller difference in the chloride concentrations (which is thought to medicate the harm of saline) between the groups in these trials when compared to BaSICS and PLUS trials where a better separation of serum chloride concentration was observed. Also, possibilities of Hawthorne effect and ascertainment bias are limitations of the open-label, pragmatic trials.

Summary of evidence on fluid of choice for critically ill patients.

BaSICS trial involving over 11000 patients (largest double-blind trial on the topic so far ) across 74 ICUs, failed to show any benefit of fluid type on survival and suggested the harm of balanced fluid in patients with head trauma. BaSICS was criticized for a relatively low event rate (decreased power) and predominantly surgical patients (limiting generalizability to other critically ill). Most of these concerns were addressed in the PLUS trial, which we believe may be the best available trial on this topic. It involved critically ill medical patients, achieved better separation of chloride levels (indicating stricter adherence to the study fluids), selected hard patient-centered outcomes, and remained adequately powered even when the trial was truncated due to the COVID19 pandemic.

A 2020 meta-analysis on this topic (which was largely driven by SMART trial), concluded: “resuscitation with balanced crystalloids demonstrated lower hospital or 28-/30-day mortality compared with saline in critically ill adults but not specifically those with sepsis. Balanced crystalloids should be provided preferentially to saline in most critically ill adult patients”. Although authors of the latest 2022 meta-analysis (including BASICS and PLUS trials) aren’t willing to accept the clinical equipoise, they are less definitive in their conclusion which goes like this: “The estimated effect of using balanced crystalloids versus saline in critically ill adults ranges from a 9% relative reduction to a 1% relative increase in the risk of death, with a high probability that the average effect of using balanced crystalloids is to reduce mortality“. Authors suggest not to take the frequentist’s stand and urge us to follow the Bayesian interpretation of the direction of the effect size, ignoring the confidence intervals that cross unity. In the perennial debate of ‘well-done RCTs versus meta-analysis putting together several studies of varied design, power, endpoints’, we place far greater weight on BaSICS and PLUS here.

What do we take to the bedside? There is no ideal ‘solution’ that will suit all the problems. The balance of this fluid debate stays neutral. Saline remains a valid first-line option for fluid therapy, in most critically ill patients. Patients with worsening hyperchloremia and acidosis after initial resuscitation may be offered balanced fluids. Balanced fluids should be specifically avoided in those with cerebral edema and can be preferentially used in patients with metabolic acidosis.

Mycophenolate after rituximab in complicated childhood nephrotic syndrome

Rituximab may be effective in prolonging remission in steroid-dependent or CNI-dependent patients. Its use is recommended in patients who have either failed CNI or have received these agents for a prolonged duration. Patients with complicated FRNS/ SDNS treated with rituximab experience relapses after recovery from rituximab-induced peripheral B cell depletion. Repeated rituximab infusions may work, but are fraught with serious adverse effects.

Japanese Study Group of Kidney Disease in Children conducted a multicenter, randomized, double-blind, placebo-controlled trial to examine whether mycophenolate mofetil (MMF) administration after rituximab can prevent treatment failure (FRNS, SDNS, steroid resistance, or use of immunosuppressive agents or rituximab)

Thirty-nine patients in each group were treated with rituximab, followed by either MMF or placebo until day 505 (treatment period). The primary outcome was time to treatment failure (TTF) throughout the treatment and follow-up periods (until day 505 for the last enrolled patient).

Time to treatment failure was clinically but not statistically significantly longer among patients given MMF after rituximab than among patients receiving rituximab monotherapy (median, 784.0 versus 472.5 days, hazard ratio[HR], 0.59; 95% confidence interval [95% CI], 0.34 to 1.05, log-rank test: P=0.07).

Secondary endpoint analyses revealed that MMF after rituximab decreased the relapse rate and daily steroid dose during MMF administration by 74% (HR, 0.26; 95% CI, 0.08 to 0.48) and 57% (mean ±SD: 4.45 ±3.52 versus 10.45 ±12.49 mg/m2 per day, P=0.0004), respectively.

However, after MMF discontinuation, the relapse-preventing effect disappeared, and most patients in the MMF group presented with treatment failure. To sum up, it is unclear if any maintenance strategy after initial rituximab therapy would result in any meaningful clinical benefit. 

HIF-PHI drugs are effective, are they safe?

Daprodustat is an oral hypoxia-inducible factor prolyl hydroxylase inhibitor(HIF-PHI), which stimulates erythropoiesis and improves iron homeostasis. HIF-PHI drugs have the distinct advantage of being orally active. We reviewed PROTECT and INNOVATE trials last year.

Summary of recent trials about IHF-PHI safety

ASCEND-D and ASCEND-ND trials randomized dialysis (DD-CKD) and non-dialysis (ND-CKD) patients respectively to receive Daprodustat or an ESA. Both trials demonstrated non-inferiority of daprodustat with respect to change in hemoglobin level.

In DD-CKD, mean change in the hemoglobin from baseline to week 28 through 52 was 0.28 g/dl in the daprodustat group and 0.10 g/dl in the ESA group (difference, 0.18 g per deciliter; 95%CI, 0.12 to 0.24), which met the prespecified non-inferiority margin of −0.75 g per deciliter. Similarly, in ND-CKD the difference was 0.08g/dl, which met the non-inferiority margin.

Anemia trials have taught us to critically assess the safety of ESAs before approval. In DD-CKD, during a median follow-up of 2.5 years, MACE occurred in 374 of 1487 patients (25.2%) in the daprodustat group and in 394 of 1477 (26.7%) in the ESA group (hazard ratio, 0.93; 95% CI, 0.81 to 1.07), which also met the prespecified non-inferiority margin for daprodustat. The percentages of patients with other adverse events were similar in the two groups.

In ND-CKD, daprodustat was noninferior to darbepoetin alpha, both in efficacy and CV safety. With respect to MACE events, HR was 1.03(95% CI, 0.89-1.19).  An important point to note here is: on-treatment analysis of CV safety favored ESA over daprodustat. The trial also raised other major safety concerns with daprodustat in the non-dialysis population. Cancer-related deaths, tumor progression, tumor recurrence, esophageal and gastric erosions were significantly more in Daprodustat group.

The question of the efficacy of the drug in treating anemia of chronic kidney disease is answered. However, the question of safety is not clear. In the ND-CKD population, this result of CV safety of ASCEND-ND is in sharp contrast to PROTECT trial. In PROTECT trial, vadadustat did not meet the non-inferiority criterion for CV safety. The reason for contrasting findings of PROTECT and ASCEND-ND is not clear. Results of CV safety in the non-dialysis population are less than convincing. Also, FDA analysis of roxadustat pooled data has raised safety concerns about CV safety in the non-dialysis population. There was a strong thrombosis signal in roxadustat arm. Also, significantly more cancer-related deaths in daprodustat arm of the ASCEND-ND trial are cause for concern.

‘Trial designs’ are important when analyzing safety data. Placebo-controlled trials offer a better opportunity to assess the safety of drugs. FDA analyzed safety in ND-CKD of roxadustat versus placebo and highlighted some safety issues. Moreover, the choice of safety margins in non-inferiority trials is arbitrary. There is no reasonable data to determine and justify the margin for safety. We also do not know if non-inferiority trial designs are appropriate to answer questions of drug safety. Given all these concerns, the safety of HIF-PHI in ND-CKD will still be a question. 

December 2021

  1. KDIGO 2021 guidelines on Glomerular diseases

KDIGO glomerulonephritis guidelines for 2021 are here. Recently they were reviewed and summarised here and here. Summary provides the quick evidence based guide for the management of glomerulonephritis for practicing nephrologists.

Unlike the past version, rituximab(RTX) comes out as almost the ‘blue-eyed boy’ of the 2021 guidelines. Instead of finding indications for RTX use in GN, it may be easier to point out some of the disease categories where it is not mentioned as an option for treatment. Not only in conditions where RTX is a potential treatment option like relapsing NS, membranous nephropathy and vasculitis, RTX gets surprising mentions in the management of Steroid Resistant NS, FSGS, Immune complex mediated MPGN, lupus nephritis, lupus associated TMA-where no evidence for its use or evidence of no use exists.

According to KDIGO 2021, for management of steroid sensitive NS in children, Cyclophosphamide (CYP) is the preferred treatment option in FRNS (Frequently Relapsing Nephrotic Syndrome) but not in SDNS (Steroid Dependent Nephrotic Syndrome). For SDNS, guidelines suggest MMF or RITUXIMAB instead. What is the evidence? FRNS have poorer treatment response than SDNS (sustained remission after 2 years in 75% vs 35.3% with cytotoxic therapy). However, we are surprised by the choice of RTX for SDNS and also fail to understand the reason for preference of MMF over CYP. Evidence from trials of RTX in this condition shows it is effective to maintain remission when compared to no treatment! Yes, if you are ok with placebo/no treatment as comparator, RTX is your drug. All three RCTs evaluating RTX for this indication chose placebo (and not alternative immunosuppression) as a comparator. These trials had limited small sample size, limited duration of followup, significant heterogeneity, variety of treatment histories prior to RTX, and differences in the outcome assessment. MMF is not superior to levamisole and unlike CYP has no remnant effect-almost all patients relapse after withdrawal. Gonadotoxicity of CYP is dose dependent and is less common in women than men, and with the current dosing strategies, cumulative dose rarely exceed 167 mg/kg. Similarly, cancers which are reported with repeated courses and high cumulative dose are rare with single, short courses. Desperate cases like these (11 year boy treated with cytosine-arabinoside and chlorambucil besides CYP developing cancer or a case of life threatening viral myocarditis requiring cardiac transplant after treatment with RTX) do not represent the real world benefit-harm of these treatments in nephrotic syndrome. Commonly held belief about safety of RTX (over steroids, CYP) is not based on the data. RTX is associated with serious side effects like myocarditis, pulmonary fibrosis, pneumocystis pneumonia, serious infusion reactions, ulcerative colitis etc. Interestingly, the KDIGO workgroup considers RTX a safe option in childhood NS based on a retrospective, observational data in adults that reports RTX use in membranous nephropathy!

The Work group’s preference for RTX is inconsistent with other recommendations like Uptodate and ISPN 2021 which recommend reserving RTX as the last drug after other drugs have failed given the concerns about long term safety-efficacy. And if you choose to use RTX, here are the monitoring recommendations by KDIGO 2021: IgG levels pre and post infusion, HbsAg, hepatitis B core antibody, Quantiferon TB test (Fig 41).

Even in diseases like ANCA Associated Vasculitis(AAV) and membranous nephropathy, data is less compelling to consider RTX as the first choice. How can trial comparing RTX with placebo (in childhood nephrotic syndrome), RTX with suboptimal comparator (Cyclosporine A in membranous-MENTOR trial), or trial involving non representative patient population (AAV with serum creatinine <2mg/dl, RAVE trial) be the basis for making choice in these indications?

Take, for example, membranous GN. Guidelines create a 4th risk category of “very high risk” and advocate CYP only in such patients of Membranous GN. MENTOR trial (which brought RTX to the frontline in management of membranous GN) was reviewed by LMiN before.

MENTOR Trial is severely limited by:

A) choice of comparator [CSA performs worse than cytotoxic e.g. UK Membranous Nephropathy trial and response rates in cytotoxic trials are far better than RTX (80-90% vs 60% with RTX in MENTOR]

B) protocol of CSA use in MENTOR: CSA was tapered rapidly (over 2 months, which can be a reason for the high relapse rate in CSA arm), unlike a more gradual taper that is routinely practiced

C) unusually higher rates of side effects and discontinuation in CSA treated patients: 15/65 (23%) in CSA group vs 4/65 (6.15%) in RTX either discontinued therapy due to side effects or were unavailable for final data assessment.

In Fig 31 of guidelines, figure legend reads “There is insufficient evidence that rituximab used in standard doses prevents development of kidney failure“, however, legends unlike figures (where RTX gets privileged first mention) don’t make it to the PowerPoint slides of talks by KOLs (key opinion leaders)!

Let us now discuss RTX in AAV. RTX is considered first choice in AAV by ACR guidelines which seem justifiable based on the efficacy data from RAVE and RITUXIVAS trial involving patient population relevant to rheumatologists. However, careful interpretation of this data is needed before we extrapolate it to all the patients with renal vasculitis. Nephrologists usually deal with the severe renal involvement because of AAV (patients with serum creatinine >4 mg or those needing dialysis), and evidence of efficacy and safety of RTX for treating this patient population is scarce. RAVE trial participants had relatively preserved GFR (mean eGFR>50ml), used daily oral CYP protocol. Thankfully, KDIGO 2021, acknowledging the limitation of this data, recommend CYP in patients with severe renal involvement. Increasingly used intermittent IV doses are safer and acceptably efficacious. MAINRITSAN trial which showed superiority of RTX over AZA in maintenance therapy (after having achieved remission with IV CYP), also predominantly enrolled patients with mild renal involvement (GFR above 60ml). The trial had important baseline imbalances (more women-who are more likely to have limited disease-in RTX arm, higher baseline eGFR in RTX arm), had imprecise effect estimates (very wide confidence interval 1.56 to 27.96), and assessed surrogate endpoint (relapse).

Preferential recommendation of RTX stems from its assumed superior safety profile as compared to other immunosuppression. Interestingly, data do not support this assumption. Below, we summarize the safety data:

 Serious AE/Life threatening AE/AE needing Hospitalization (RTX vs comparator)Deaths All AEs (RTX vs comparator)Infections (RTX vs comparator)
RAVE (RTX vs oral CYP)11.1% vs 5.1% (AE Grade 4)1/99 vs 2/9831.3% vs 33.6%7% vs 7%
RITUXIVAS (RTX vs IV CYP+RTX)36% vs 36% (AE Grade3,4,5)6/32 vs 3/1176% vs 64% 36% vs 27%
MENTOR (RTX vs CSA)17% vs 31%071% vs 78%27.6% vs 29.6%
RCRNS (RTX vs placebo)42% vs 25%0100% vs 96% 96% vs 75%
MAINRITSAN (RTX vs AZA)43.8% vs 43.1%0 vs 219.2% vs 13.7%
Safety of RTX as compared to other IS drugs used in GN

Pivotal trials comparing RTX with CYP (even with higher cumulative oral CYP dosing) failed to show a better safety profile of RTX, and in fact showed more adverse events, infections and worrisome trend of higher rates of deaths, and malignancies.

Choice of induction immunosuppression in severe lupus nephritis also deserves discussion. ACR and KDIGO concur on the recommendation MMF as preferred induction agent based on ALMS trial. Mean serum creatinine of the study participants was <1.2 mg/dl, leaving open the question of the efficacy of MMF based regimen in severe disease. In fact, a systematic review of the outcomes of treating severe lupus nephritis with MMF or CYP suggested that in spite comparable efficacy in inducing remission of severe lupus nephritis (defined by renal histology, resistance to therapy, or level of kidney function at presentation), relapse rates, and risk of developing ESRD were higher for MMF compared with CYP. Question about long-term efficacy of MMF is also raised by long-term data (maintenance phase) of ALMS trial: treatment failure at 3 year was significantly higher in patients who received MMF as induction therapy 10.1 vs 4.7 events per 100 person years. Clearly, there is no evidence of superiority of MMF which is claimed based upon the subgroup analyses of original ALMS trial where Blacks and Hispanics had higher response rates. Asians had numerically lower response rates with MMF (53 vs 63%). However, KDIGO 2021 Practice Point recommends preferring MMF bases therapy in Asians as well, rationale of which is unclear.

Higher death rates in MMF arm of ALMS is a cause of concern- guidelines stop by mentioning that 7 of 9 deaths in MMF arm occurred in Asia. What does that mean? Please do not consider Asian lives less important or Asian nephrologists less competent. This may be a safety signal. Safety concerns of MMF based induction therapy were also noted in a trial from India, where most patients could not tolerate the recommended induction dose -mean dose 2.2gms, and MMF arm experienced more adverse events-64% vs 50% (particularly diarrhea) and higher death rates (10% vs 4%), even at short follow-up of 6 months. Does cost, efficacy, and safety justify MMF as induction therapy in all patients with lupus nephritis?

Fertility is important considerations in treatment choice, and one of the most important factor to preserve fertility in these women is to protect their renal function, loss of which can render them amenorrheic, or anovulatory, and if they conceive, increase the risk of pregnancy-related complications dramatically.

2. Ofatumumab in steroid dependant nephrotic syndrome in children

Response to RTX depends on disease duration and severity of nephrotic syndrome. Remissions after RTX is shorter in children who are maintained in remission with both steroids and calcineurin inhibitors, compared with those on steroids alone or mycophenolate mofetil. There is also a possibility that RTX infusion may elicit anti-rituximab antibodies. Also, following RTX injection, 75% relapse in 4 months. These patients will require either MMF or CNI to maintain remission. Given these limitations, there is a need for a therapy which was more potent and maintain a longer remission without need for additional maintenance regimen. 

Ofatumumab is a fully human anti-CD20 monoclonal IgG1(k) antibody as compared to RTX, which is a chimeric anti-CD20 monoclonal antibody. In vitro studies have shown that Ofatumumab has a more extended binding site than RTX, a higher affinity to the CD20 antigen and a more efficient complement-dependent cytotoxicity. Whether these in vitro effects translate into better clinical response is not known. 

This is the first large randomised controlled clinical trial which compared the efficacy of ofatumumab to the RTX in 140 children and young adults with SDNS who were maintained in remission with steroids and CNI. 70 subjects in each arm were randomised to ofatumumab or RTX. The primary outcome was relapse at 1 year and secondary endpoint was relapse within 24 months. The trial failed to establish superiority of ofatumumab over RTX. At 12 months, 53% participants who received ofatumumab relapsed versus 51% who received RTX (OR, 1.06; 95% CI, 0.55 to 2.06). At 24 months, 76% participants who received ofatumumab relapsed, versus 66% who received RTX (OR, 1.6; 95% CI, 0.8 to 3.3).This finding suggest that the bench effects of ofatumumab observed in vitro do not translate into clinical outcomes. 

The study was designed to detect a large effect size (50% risk reduction). Though there were no pointers of superiority of ofatumumab, a small effect size cannot be excluded. However, to detect a small effect size, a large sample size will be required. The study included patients who previously received RTX. Although RTX exposed patients were evenly distributed in both arms, it is possible that previous treatment with RTX may have enhanced the response to the intervention in the RTX arm, thus reducing the difference between the two interventions. As of now, ofatumumab does not have evidence of superiority over RTX. Welcome to the nephrology therapeutics and all the best new MAB(monoclonal antibody).

3. Intensive BP Control in Elderly

SPRINT trial showed that among patients who were at increased cardiovascular risk, targeting a systolic blood pressure of less than 120 mm Hg resulted in lower rates of major adverse cardiovascular events and lower all-cause mortality than targeting a systolic blood pressure of less than 140 mm Hg, both during receipt of the randomly assigned therapy and after the trial.

STEP trial has studied the role of intensive BP control in elderly Chinese patients. In this prospective, multicenter, randomized, controlled trial performed at 42 clinical centers in China, from January 10 to December 31, 2017, 9624 patients screened for eligibility, 8511 were enrolled. Patients of 60 to 80 years of age with hypertension (systolic blood pressure of 140 to 190 mm Hg during three screening visits or to be taking antihypertensive medication) were included. Patients with a history of ischemic or hemorrhagic stroke were excluded. At each center, eligible patients were randomly assigned in a 1:1 ratio to a systolic blood-pressure target of 110 to less than 130 mm Hg (intensive treatment) or a target of 130 to less than 150 mm Hg (standard treatment).

All the patients were required to obtain home BP readings at least 1 day per week during follow-up using validated automated home blood-pressure monitor. The blood-pressure monitor was paired to the smartphone-based app with a Bluetooth function, which then then uploaded the readings to a data-recording center. Patients were not enrolled in the trial unless they could use the app or had a family member who could use it on their behalf. During the follow up visit, office BP was measured by trial staff three times at 1-minute intervals.

As as pointed out in the editorial, it is remarkable that it they could complete the enrolment in one calendar year. Only 2.7% were lost to follow up.

The primary outcome was a composite of stroke, acute coronary syndrome (acute myocardial infarction and hospitalization for unstable angina), acute decompensated heart failure, coronary revascularization, atrial fibrillation, or death from cardiovascular causes. At the end of a median follow-up of 3.34 years, the data and safety monitoring committee recommended that the trial be stopped early on the basis of a clear cardiovascular benefit in the intensive-treatment group at two consecutive time points. Primary-outcome events occurred in 147 patients (3.5%) in the intensive-treatment group, as compared with 196 patients (4.6%) in the standard-treatment group (hazard ratio, 0.74; 95% confidence interval [CI], 0.60 to 0.92; P = 0.007).

 The incidence of hypotension was significantly higher in the intensive-treatment group than in the standard-treatment group (3.4% vs. 2.6%, P = 0.03). However, the incidences of dizziness, syncope, and fracture and the results for renal outcomes (decrease in eGFR) did not differ significantly between the two trial groups.

Happy new year!

October 2021

  1. Racing further away from truth: story of eGFR equations

Blacks have a higher serum creatinine; nobody knows for sure, why. Decreased tubular secretion, increased muscle mass, and increased generation are the speculations, but given the higher incidence and poor outcomes of kidney diseases in blacks, further research on this is urgently needed. mGFR (measured GFR by exogenous markers like inulin, iohexol- Gold standard) in blacks therefore is about 15-16% higher than whites at comparable age, sex and creatinine levels. This is the reason for ‘race coefficient’ in eGFR equations. A recent acknowledgement of the longstanding racial discrimination (better late than never) has stimulated the processes and ways of eliminating race in medical practice wherever possible, and NKF KDOQI task force was set up to address this issue of race in estimating equations.

If we simply eliminate race from the equation, blacks will be more often labelled as CKD leading to unnecessary evaluations, treatment and deferral of needed evaluations (contrast😊) and treatments (with drug cleared by kidney). How to eliminate ‘race’ from eGFR without these attendant untoward consequences? This is addressed by these two studies simultaneously published in NEJM both concluding that cystatin c may be the answer.

In the CRIC study cohort of 1248 patients with kidney disease, eliminating race underestimated GFR in blacks by about 4 ml, and compromised accuracy (P10 i.e. percentage of readings within 10% of the gold standard -mGFR). The attempts of compensating this error using various non GFR determinants (height, bioelectrical impedance analysis phase angle, bioelectrical impedance analysis–quantified fat-free mass, or urinary excretion of creatinine) were not successful in eliminating the misclassification. Although they could slightly attenuate the GFR overestimation-eGFR still remained higher than mGFR by 8%. Use of cystatin C eliminated the need of race in the formula and gave eGFR that was closest to the mGFR.

CKD EPI investigators, compared the performance of current CKD EPI equation using age, sex and race coefficient (termed as ASR equation), and evaluated two current race neutral equations [AS and AS-NB; First eliminating race coefficient and second is ASR (i.e. current CKD EPI) equations that were fit with a race term but in which the Black race coefficient was removed for computation of eGFR].

GFR in blacks was underestimated by both AS (3.6ml) and AS NB (7.1ml) equations as compared with the currently used ASR, highlighting the need of other approach to eliminate race. This was possible with CKD EPI cystatin-creatinine (new eGFRcr-cys equation) which minimized the inaccuracy for both race groups and differences in eGFR between race groups (less bias), had high P30(the proportion of eGFR within 30% of measured GFR) of over 90%, and produced little differences in estimated CKD prevalence.

Current recommendation of KDIGO about use if cystatin C goes like this “We suggest measuring cystatin C in adults with eGFRcreat 45–59 ml/min/1.73 m2 who do not have markers of kidney damage if confirmation of CKD is required. (2C)“. Cystatin C with serum creatinine is all set to become the standard marker in future KDIGO guidelines with these publications.

A word of caution: Cystatin C is the most abundant endogenous inhibitor of cathepsins, a family of proteases involved in atherogenesis and immunomodulation. It is closely associated with obesity, inflammation and important risk markers like albuminuria, hypertension, diabetes, low HDL cholesterol, high triglycerides, high C-reactive protein, and elevated uric acid. As many of these are independently associated with renal failure, CVD, and mortality, it can be argued that perceived superiority of cystatin C in predicting outcomes at higher GFR (45-60, where 2012 guidelines recommend its use currently) may have to do with its association with diverse CV risk factors (rather than risk of kidney failure), making it a better marker for risk stratification rather than GFR.

Addition of cystatin C to serum creatinine leads to small improvement in estimating mGFR, however it comes at the cost of less accurate representation of the true association between GFR and common CKD risk factors. In other words, eGFRCr-Cys is superior for the estimation of mGFR and is also superior for the prediction of outcomes (because it contains cystatin C, that is associated with diverse risk markers). However, eGFRCr is superior for associating with most CKD risk factors similar to mGFR-the fact most relevant for diagnosis and staging of CKD. This phenomenon is termed as ” eGFRCr-Cys paradox”. Interestingly , eGFRCr was found to be a better option for characterizing the association of CKD risk factors with true GFR than either eGFRCysC or eGFRCr-CysC. Moreover, test-retest reliability, important criteria when a marker is used in clinic, of eGFR creat was superior to eGFR cystatin C and was marginally better than gold standard mGFR (Coefficient of variation-CV- for eGFR creat, mGFR and eGFR cystatin was 6.4%, 8.2% and 10.7% respectively).

If you are done with welcoming celebrations of the new eGFR equation, let me ask a question which might offend the ‘religious believers’ of the authoritative KDIGO guidance recommending reporting of eGFR whenever a serum creatinine is measured. Why is all this fuss for eGFR? Does eGFR mean anything more than serum creatinine for the clinician?

After discussing for hours how flawed a marker of filtration serum creatinine is, speakers and delegates go back to their clinics advising and interpreting serum creatinine like ever before. IMHO eGFR is one of the most hyped ‘idea’ in nephrology for the following reasons:

  1. Carefully (preferably by enzymatic method) measured and IDMS traceable creatinine, serially followed up in an individual patient, along with a nephrologist’s overall assessment, (know your marker here) is all that is needed to make clinical decisions in the management of CKD (this isn’t just common sense evidence here).
  2. Apart from broad categorisation to dose drugs cleared by kidney, it provides no additional input that is not provided by the trend of serum creatinine in an individual.
  3. Advocates of eGFR often cite the ‘need for uniformity in research inclusion and outcome criteria’ as one of the rationale of eGFR. Perhaps eGFR is particulary flawed for such usage, given the imprecise nature of this calculation. p30 (percentage of eGFR values within 30 % of the mGFR) for all current equations is less than 90%. Even with the latest CKD EPI equation, correct classification was possible only in ~70% (figure 1, agreement between measured GFR and eGFR categories of more than 90, 60 to 89, 45 to 59, 30 to 44, 15 to 29, and less than 15 ml). How can studies reporting few ml improvement in the eGFR after therapy be considered conclusive? Unfortunately, we have failed to understand that eGFR is a means to an end and have considered it as an end by itself. Arguably, this over-reliance on eGFR might be one of the precise reasons for slow progress in CKD research.

“I wrote philosophy in this grand book, the universe … It is written in the language of mathematics, and its characters are triangles, circles, and other geometric figures; ….” said Galileo. One thing in nature that is the most difficult to put in the formulas of mathematics is the human physiology and especially those mysterious beans in abdomen-filtering, adding and subtracting until delivering the final fine product-urine.

2 Treatment of hyperphosphataemia in CKD: a case of Silver Blaze bias

The human understanding, when it has once adopted an opinion (either as being the received opinion or as being agreeable to itself) draws all things else to support and agree with it.Francis Bacon, Novum Organum, 1620

Department of ‘surrogate nephrology’ aka ‘CKD MBD therapeutics’ illustrates this point most aptly, where observational studies, underpowered trials or trials with surrogate outcomes continue to guide us. Take, for example, KDIGO 2017 CKD MBD guidelines: it has nine statements on phosphate control, of which 3 are not graded, 5 are level 2C (2C means suggestion and not recommendation, based on poor quality data, essentially it means ‘to be seen later’), and 1 is level 2B. So what is the rationale of suggesting aggressive phosphate binder use with preference to non calcium based binders? Nevertheless, based on surrogate outcome of vascular calcifications and risk of hypercalcemia, KDIGO recommends “consider restricting dose of calcium-based phosphate binders in adults receiving phosphate-lowering treatment.” 

A 2018 Cochrane review concluded that the effects of binders on patient‐important outcomes compared to placebo are uncertain, making the case for a placebo controlled trial of binders versus no binders rather than a comparison of the options advocated.

LANDMARK, an open label randomized clinical trial, enrolled 2374 Japanese hemodialysis patients with hyperphosphatemia and 1 or more risk factors for vascular calcification to receive either lanthanum carbonate or calcium carbonate to achieve serum phosphate levels of between 3.5 mg/dL and 6 mg/dL. The primary outcome- a composite cardiovascular event (cardiovascular death, nonfatal myocardial infarction or stroke, unstable angina, transient ischemic attack, or hospitalization for heart failure or ventricular arrhythmia) was similar in both arms at a median follow-up of 3.16 years, 4.8 per 100 person years in lanthanum group and 4.3 per 100 person years in calcium carbonate group (HR, 1.11 [95%, CI, 0.88 to 1.41]). There were no significant differences in all-cause death or hip fracture. However, lanthanum carbonate group had an increased risk of cardiovascular death.  

A KDIGO recommendation about dietary phosphate restriction (2D) is of particular concern given the data suggesting harm of this approach. Is it not better to acknowledge the knowledge gap? Bias which arise from ‘studies which have not been conducted’ is called “silver baze bias”. By the way, one thing that binders are most effective at is producing ‘stomach upset’-constipation, diarrhoea, nausea-adding to symptom burden besides pills.  

PS :  

Gregory (Scotland Yard detective): Is there any other point to which you would wish to draw my attention? 

Holmes: To the curious incident of the dog in the night-time. 

Gregory: The dog did nothing in the night-time. 

Holmes: That was the curious incident. 

In the short story “Silver Blaze,” Sherlock Holmes solves the mystery that somebody that the dog knew committed the crime. It is from this Arthur Conan Doyle’s story, we borrow the term “Silver Blaze Bias.” 

3 Salt substitutes and cardiovascular events

When one reads current literature on salt, one is amazed by the emotional energies released by this issue. Why is this so? Undoubtedly, much of the controversy is simply due to the fact that there is much poor science in this field. E. Ritz

Against the unjust salt tax, on 12th March 1930, Mahatma Gandhi picked up a fistful of salt at Dandi, shaking the pillars of the mighty British empire. While many of Gandhi’s health beliefs might be ridiculed as unscientific today, he had a keen interest in the impact of food on health. Today, given the massive public health threat posed by hypertension in India (prevalence over 30% and dismal control rates of 10%), he would have urged us to walk back to Dandi with a fistful of salt from our kitchens to be dropped at Dandi.

Reduction in sodium intake using salt substitutes reduces BP in hypertensive patients. Salt intake in Indian diet is alarmingly higher than the prescribed safe upper limit by WHO, and attempts to curtail salt use are likely to be highly fruitful for preventing premature death, and disability. SSiIS was a double blind, randomised control trial conducted in rural India that will strengthen the evidence of beneficial effect  of salt substitutes on blood pressure. 

502 hypertensive adults were randomised to receive salt substitute(70% sodium chloride and 30% potassium chloride) or regular salt (100% sodium chloride). The participants were advised to replace all the home salt with the salt provided by the investigators. The primary outcome was a change in BP at three months follow-up favoured salt substitutes (SBP reduction of 4.6 mm of Hg). There was significant increase in potassium excretion in the salt substitute group. Surprisingly, the salt substitute group did not have reduction in sodium excretion in urine. The reason for this finding is not clear.

This is the first trial done in India to demonstrate the efficacy of salt substitute in management of hypertension. There was no problem of acceptability and usage of salt substitutes.Study was restricted to rural areas and had a short duration of follow up. Sustained benefits, compliance (due to the cost) and impact on hard end points will need to be examined in further trials.

Before you question the benefit of short term salt substitute use and dismiss this Indian study, here is another more compelling data on this issue. A meticulously designed open-label, cluster-randomized trial in rural China-SSaSS, enrolling 120,995 high risk individuals (72% had history of stroke, mean age was 65 years, 88% had hypertension). This study clarified the effect of salt substitute on patient important outcomes like stroke, cardiovascular events and death. Villages were randomised to receive either regular salt(100% sodium chloride) or salt substitute(75% sodium chloride and 25% potassium chloride). Salt substitute group had a lower rate of stroke than the regular salt group (29.14 events vs. 33.65 events per 1000 person-years; rate ratio, 0.86; 95% confidence interval, 0.77 to 0.96; P=0.006). There was also significant reduction in major cardiovascular events and all cause mortality. Incidence of hyperkalaemia was not different in the two groups.

The results of the study are impressive. The only drawback was that the study does not provide data on grade (lower or higher content of potassium) in the substitute. Except this minor drawback, we feel the study is a compelling evidence to support use of salt substitute for this high-risk group.  

These two recent publications on salt and salt substitute may, to a certain extent, exonerate us from allegation by E. Ritz, and will make us wiser in answering one of the most common question by our hypertensive patients: Can I take salt substitutes? Yes please.

May 2021

  1. Steroid for Acute Pyelonephritis in children

Acute pyelonephritis in children can lead to renal scarring, hypertension, proteinuria and ESRD, although its causal relationship with renal damage is increasingly questioned after RIVUR trial which showed that antibiotic prophylaxis in children with reflux prevents UTI, but not renal scarring. In these patients inflammation is thought to mediate renal scarring and its consequences.

So far, trials mainly focussed on preventing UTIs (with antibiotic prophylaxis), however, role of corticosteroids to prevent renal scarring by addressing renal inflammation is being explored. This was supported by benefit in animals studies and an observational study showing a decrease in urinary cytokine (IL6, IL8) with steroids in children with acute pyelonephritis. Highly awaited and the largest of the trials (NIH) evaluating role of steroid failed to conclusively answer the question as it could enrol only 79% of the planned sample size of 546 children. Trial participants had numerically lower chance of renal scarring, but it couldn’t reach statistical significance [12/123 (9.8%) vs 22/131 (16.8%), p=0.16, 95%CI − 2.2, 14.1]. This trial highlighted the difficulty of recruiting and maintaining children in trials, also showed that urinalysis is not a very specific screening method (158 febrile children with presumed UTI were enrolled but then withdrawn because of a negative urine culture). So, can we use steroids to prevent scarring in children with acute pyelonephritis?

Here is a systematic review and meta analysis of 3 RCTs involving 529 children evaluating role of adjuvant corticosteroids to prevent kidney scarring in children with acute pyelonephritis. Lower risk of renal scarring was noted with adjuvant steroids when compared with placebo (RR 0.57; 95% CI 0.36 to 0.90). No significant increase risk of bacteraemia (RR: 1.38; 95% CI 0.23 to 8.23) and hospitalisation (RR: 0.87; 95% CI 0.3 to 2.55) was observed in corticosteroid group.

While results are encouraging, we should interpret them with caution. First, overall small size of the study (NIH trials’s target sample size >total number of patients in the meta analysis) prevents generalisability of these results. Second, although the NIH trial carefully monitored and reported the possible harm of steroids, and showed no increase in the risk of sepsis, more patients in the corticosteroid arm left the study in this largest trial (73 vs 57), with no clarity on the reasons of this attrition. Third, the studies had limited follow up; we don’t know the long-term risk benefits of this strategy.

“Adequately powered RCT or meta analysis?” is a million dollar question in evidence-based medicine and I am biased for the former. Those interested can have a look at this and this. This is especially true when there are a few high-quality studies to be included in meta analysis, when ‘garbage in becomes garbage out’. So this meta analysis generates a very important hypothesis, which if tested in adequately powered multi centre studies, might change the standard of care for children with acute pyelonephritis.

2. Voclosporin: old wine in a new bottle for lupus nephritis

Visual Abstract for AURORA 1 by Divya bajpai

“Voclosporin in combination with MMF and low-dose steroids led to a clinically and statistically superior complete renal response rate versus MMF and low-dose steroids alone, with a comparable safety profile. This finding is an important advancement in the treatment of patients with active lupus nephritis” reads the interpretation section of the AURORA 1 trial evaluating role of this novel CNI in lupus nephritis. Specifically, complete renal response (composite of UPCR <0.5, eGFR>60, and no need of rescue steroids), at week 52 [73 (41%) of 179 patients vs 40 (23%) of 178 patients; OR 2·65; 95% CI 1·64–4·27; p<0·0001)]. Importantly, a reassuring finding is that the safety concerns observed in phase 2 trial of this drug (more infections and deaths in volcosporin) were not seen in this study. Voclosporin has higher molecular potency, better side effect profile and doesn’t need therapeutic drug monitoring. Welcome, voclosporin, to the lupus nephritis therapeutics.

Multi-target therapy is being discussed from quite some time in lupus nephritis and I have some reservations about its wider application beyond a specific patient subset. Concerns over long-term nephrotoxicity, ‘pseudo remission” because of non immunosuppressive effects of CNI on podocytes, safety concerns of long term “triple immunosuppression” and the cost.

Specific pitfalls of AURORA 1 include: no baseline biopsy to assess activity/chronicity, and no data on response to therapy (MMF/Steroid) prior to study participation. All patients received IV methylprednisolone before enrolment; what? If steroid sparing is the promise of multi target therapy, this doesn’t make sense. Despite improvement in the primary outcome, there wasn’t any significant difference in serological (complement, anti dsDNA) and overall disease activity, making one suspect whether it is immunological remission or pseudo-remission of proteinuria. Lastly, tacrolimus is often preferred over cyclosporin in practice for its favourable cosmetic side effect profile and also because of TAC-MMF increasing each other’s AUC by ~30%, allowing lower doses to be used. So further evaluation of voclosporin, specifically the long term safety-efficacy, its performance when compared to Tac/CSA will be needed.

3 DOACs in patients on dialysis

Lower bleeding risk, lesser laboratory monitoring, possible lower risk of fractures, preferable and more predictable pharmacokinetics, have expanded the use of DOAC s. Over last one year, we have discussed anticoagulation in CKD like never before, having seen high rates of thrombosis in patients with kidney diseases getting hospitalised for COVID 19. As usual, ‘renalism’ has left out patients with kidney diseases from most trials of DOACs, and data to support their use in dialysis population limited. This may be changing fast.

In Valkyrie study involving 132 patients on hemodialysis with atrial fibrillation and a CHA2DS2-VASc score ≥2, randomised patients to vitamin K antagonists (VKAs) with target INR 2–3, rivaroxaban 10 mg daily, or rivaroxaban 10 mg daily plus vitamin K2 2000 mg thrice weekly during 18 months. Replacement of VKAs by rivaroxaban was safe and potentially associated with less life-threatening and major bleeding. In the extended additional 18 month follow up of Valkyrie trial, the primary endpoint (composite of fatal and nonfatal cardiovascular events) occurred at a rate of 63.8 per 100 person-years in the VKA group, 26.2 per 100 person-years in the rivaroxaban group, and 21.4 per 100 person-years in the rivaroxaban and vitamin K2 group. The estimated competing risk–adjusted hazard ratio for the primary end point was 0.41 (95% CI, 0.25 to 0.68) in the rivaroxaban group and 0.34 (95% CI, 0.19 to 0.61) in the rivaroxaban and vitamin K2 group, compared with the VKA group. Life-threatening and major bleeding was less common in rivaroxaban group. These findings support a superior risk-benefit profile of DOACs versus VKAs.

Real world safety and efficacy of DOACs in dialysis population was also highlighted by Taiwan’s National Health Insurance Research Database in retrospective population-based cohort study. Patients with non-valvular atrial fibrillation and ESRD who started on rivaroxaban or warfarin (n=3358 patients 173 and 3185 patients on rivaroxaban and warfarin, respectively) were studied. The cumulative incidence of major bleeding was similar between the two groups; however, the gastrointestinal bleeding rate was lower in the rivaroxaban group than in the warfarin group. The composite risk of ischemic stroke or systemic embolism was significantly lower in the rivaroxaban group (adjusted SHR: 0.36, 95% CI: 0.17–0.79).

With this encouraging data, wider use of these agents in dialysis patients needing anticoagulation for AF is expected.

4 Which beta blocker?

After the publication of HDPAL trial, β-blockers have become one of the most commonly used antihypertensive agent in patients on hemodialysis.

While with most other classes of antihypertensives, BP lowering is a class effect, different beta blockers are recommended in different clinical settings. Cardioselectivity, additional alfa blocking action, dialysability prevents generalisability of the clinical benefits to other agents and the effect of these properties on outcomes is unclear.

Bisoprolol is β1 selective and is moderately dialyzable while carvedilol is poorly dialyzable α and β blocker.

In a nationwide population-based study using Taiwanese administrative data, Wu et al studied9305 HD patients who initiated bisoprolol and 11 171 who initiated carvedilol treatment. They compared the risk of all-cause mortality and major adverse cardiovascular events (MACEs) between carvedilol and bisoprolol users during a 2-year follow-up.

Bisoprolol users had a 34% lower all-cause mortality risk (HR = 0.66, 95% CI 0.60–0.73) and a 15% lower risk for major adverse cardiovascular events (adjusted HR = 0.85, 95% CI 0.80–0.91) compared with carvedilol users. Major adverse cardiovascular events were defined as a hospital admission for acute myocardial infarction, heart failure or ischemic stroke. These findings were consistent across propensity score-matched models and in a number of other sensitivity analyses.

What may underly the observed harm of carvedilol ? Is it the additional α-blocker property and poorly dialyzability? These are speculations which might be associated with intradialytic hypotension more frequently, and could explain, in part, these results. Retrospective nature of the data preclude any definite recommendations and only large randomised trials will be able to answer these queries. Meanwhile, atenolol which has demonstrated benefit in HDPAL remains my choice.

The editorial commentby Gregory L Hundemer, @Msood99Mand @MarkCanneysums this up:“While acknowledging the constraints of observational data, these findings may serve to inform clinicians about the preferred β-blocker agent for dialysis patients to help mitigate cardiovascular risk and improve long-term survival for this high-risk population.

5 HIF PHI inhibitors: more questions than answers?

Roxadustat and vadadustat are oral hypoxia inducible factor- prolyl hydroxylase inhibitors (HIF-PHI), which stimulate erythropoiesis and improves iron homeostasis. These drugs have distinct advantage of being orally active. The haematological efficacy of these drugs is established in various phase 3 trials in different subsets of chronic kidney disease population.

PROTECT and INNOVATE trials also demonstrated impressive efficacy of vadadustat in phase 3 randomised, open label, non-inferiority design with darbepoetin alfa as active comparator. Each published article is a pooled analysis two trials. PROTECT trials enrolled nearly 3500 non dialysis patients (one trial on ESA naive and second on ESA exposed). INNOVATE trials enrolled 3923 patients (one trial on incident and second on prevalent dialysis). To achieve sufficient power for safety endpoints, each article was based on pooled safety data of two trials. In both the studies, vadadustat was non inferior to darbepoetin alfa with respect to hematological efficacy.

The results of the trials with respect to safety endpoints were intriguing. The prespecified non inferiority margin was 1.25 for first MACE event. INNOVATE trials demonstrated non inferiority of vadadustat  with hazard ratio of 0.96(95%CI, 0.83 to 1.11). PROTECT trials found hazard ratio of 1.17(95%CI, 1.01 to 1.36) and did not meet the prespecified noninferiority margin.

The safety of vadadustat differed between dialysis and non-dialysis population. The reason for this is not clear. This reminds us of peginesatide. With respect to safety, Peginesatide (EMERALD and PEARL) was found non inferior to darbepoetin in dialysis patients but did not meet non inferiority margin in non-dialysis population. Peginesatide was withdrawn from the market due to severe adverse events observed in post marketing surveillance.

The data regarding efficacy is robust and reassuring. But critical question of safety of the drug remains unanswered, especially in non dialysis population.

Once upon a time in endocrinology

“अब तबीयत कैसी है?”(how is your health now?)– in her motherly and unique style, she would drop her hand on the patient’s shoulder. All of us wondering, why is she asking this to a patient admitted ‘just for the evaluation’ of polyuria existing since over a decade. After a pause, the patient gathered some courage and said, “जाँच पे जाँच , जाँच पे जाँच , जाँच पे जाँच हो रहा है मैडम , मगर कौनो दवा नाही मिली अब तक।” ( A battery of tests are done repeatedly but not a single medicine so far). With these words reminiscent of the Sunny Deol in famous bollywood drama, Damini, the whole unit (except madam and patient) burst into laughter. She consoled the patient and asked me to add vitamin C before moving to the next bed. Post-round tea was accompanied by her discourse, still fresh in my mind, on importance of (and lack thereof) of patient values and effective communication in medicine. Only at the end of my residency I realised that, in her preparation of vitamin C, ‘C’ stood for compassion.

After joining her unit, as houseman in medicine and endocrinology, within a few months, I was completely lost in horrible endocrine tests like ‘suppression and stimulation tests’, ‘AVS (adrenal venous sampling), IPSS (inferior petrosal sinus sampling), 2 am blood sugar, midnight cortisol, water deprivation….. The last 2 of these were ‘infamous’ as ‘sleep’ and ‘lunch’ deprivation tests, respectively, amongst housemen and were the best weapons of registrar against unruly houseman.
We had a 5 bedded endocrinology ward where patients coming from corners of the country and (no place to stay in Mumbai) were admitted mostly for diagnostic evaluations. After carefully collecting and labelling the samples of urine and blood, making sure they reach the lab in time and having collected the reports next day, I would rush to her chamber breathless, where everyone will be curiously waiting for reports. As she scanned through them, expressions will quickly start changing, which I knew indicated a confirmed diagnosis( if not I have to arrange for a repeat testing). Endocrine evaluations for us was ‘बीमारी से नहीं जाँच से डर लगता है मैडम’ (not the illness but its evaluation scares me).

Coming to evaluation of Primary Hyperaldosteronism (PA), the most common cause of endocrine hypertension. Last month, a veteran medical teacher with hypertension, was referred for optimising blood pressure control. In spite of being on ramipril 10mg/day, mild hypokalemia was seen on >2-3 occasions however Plasma Aldosterone Concentration (PAC), and Plasma Renin Activity (PRA), checked on multiple occasions were in normal range of the reporting lab. Screen for target organ damage (cardiac, eye, kidney) was negative but blood pressure remained ~140-150 systolic with occasional spikes, as high as 200 mmHg, curiously related to his virtual lectures in lockdown. Amlodipine would typically bring it down but leave him with incapacitating headache (which quickly became my own headache due to midnight calls). Clinical suspicion of hyperaldosteronism was strong, and I tried persuading him for further tests. But he was 10 times more minimalistic than me and wasn’t keen on any invasive testing (he defines CT as an invasive test) unless I fail to control BP. Having examined hundreds of medical students in his career as medical teacher, a viva was coming up for me.

He asked,” Do you wake up at 5am?”


“Do you swim daily?”


“Do you go for regular scuba diving”


“Do you participate in Mumbai marathon?”


“I am doing all of this since last 25 years…so tell if you can do your job of getting down my BP ?”

And a diagnostic test was performed……..T Spironolactone 25 mg once daily was added and blood pressure within a week was down like never before, needing dose reduction in others medicines.

While there is a lot of (justified) hue and cry for not adequately evaluating patients with hypertension for PA, part of this inertia is also related to the, ambiguous guidance in addition to the limited availability of the testing: different thresholds of PAC, PRA, used in the reported literature, years of authoritative, non evidence based and potentially harmful guidance on drug modifications before testing. Here is a systematic review and meta analysis of the available evidence on performance of ARR (Aldosterone Renin Ratio, a guideline recommended screening test for PA) using the best available 10 studies involving 4110 patients. Wide variations in the reported sensitivity  of ARR precluded recommendation of any definite cutoff, nonetheless specificity of this test was consistently high. So a normal ARR can’t reliably exclude PA, but high ARR is fairly specific feature of PA, questioning the reliance on this test as a ‘screen test’ in diagnostic algorithms.

One may label this as a negative review overall, but this well done exercise offers important insights regarding lab evaluation of PA. Most original studies evaluating ARR significantly and systematically overestimated its sensitivity due to the reasons like selection bias (studies being done at hypertension referral centre, on resistant HTN, or in those with other clinical clues like low potassium), verification bias (many older studies restricted confirmatory testing to ARR screen positives), and development of cut-offs post hoc (after the results of study). Diurnal variation in the aldosterone secretion (which is more pronounced than cortisol), further compromises the reliability of spot ARR. This review is also an eyeopener on how tests with limited formal validation are accepted religiously for decades, which may miss a large proportion of, easily treatable subset of hypertension population. PA increases cardiovascular risk over and above that mediated by high blood pressure, so identification of PA for appropriate therapy (medical in most instances, and surgical in a few) can potentially prevent significant morbidity and mortality.

So, what is the bottom line? There is none. Increasingly, PA is considered as a spectrum of renin independent excess aldosterone secretion; adrenal hyperplasia, autonomous nodule and eventually an adenoma. Laragh approach of managing ‘essential’ hypertension based upon the predominant pathophysiological mechanism (renin versus volume) might well see a wider appeal in coming days. Authors make a crucial point to move away from ‘chasing sensitivity of the screening test to label someone as PA’ to a more simplified approach based upon only measuring plasma renin as a screen, and then proceeding with 24-hour urinary aldosterone measurement if its positive. Also, only a minority of the total PA population (1-5%) will have surgically correctable lesions i.e. unilateral adenoma or hyperplasia, but most other will be benefited by mineralocorticoid receptor antagonist. The benefits of MRAs (although benefits are not predicted by renin level nor are they completely explained by Na and volume loss ) in trials may be ‘a proof of the concept’ that hyperaldosteronism is a bigger player in mediating high BP and its complications. Studies of wider application of MRAs in hypertension management are needed.


April 2021

Liver and Kidney

I once overheard a hepatologist counselling the family of a patient with HRS-1, “Treatment of liver disease is already on, now only hope for him is recovery of kidney!” I mean, are you out of your mind (or liver rather)? Stop blaming poor kidney, my dear hepatologist, if you give me that kidney out, it may well start diuresing in a patient with intact liver. Smart nephrologists knew about this ‘hepatic nephropathy’ 50 years back. AKI here identifies patients at highest risk of death, and without liver transplant-which, of course, is not an option for many, median survival is 2 weeks, with almost all patients dying within 3 months.

Especially after publication of ANSWER trial, which claimed the survival benefit and disease modifying role of long-term albumin administration, Albumin (like Vit D,C, statins……) is believed to have pleiotropic effects (briefly defined as the effects which can’t be proven scientifically) via which it can control inflammation which, besides systemic vasoconstriction, is a major pathogenic mechanism mediating complications of cirrhosis.

ATTIRE and CONFIRM trials published in NEJM address two important issues in the management of cirrhosis.

Open label ATTIRE trial randomised 777 patients with cirrhosis and decompensation event to receive either daily albumin to increase serum albumin to >3) or standard of care (who could still receive albumin if indicated for SBP, HRS, large volume paracentesis). This ‘aggressive albumin strategy’ resulted in 10 times more albumin use and was not only ineffective in improving primary outcome (a composite of infection, renal failure, or death-occurred in ~30% in each arm), but also proved counterproductive -higher life-threatening serious adverse events, especially pulmonary edema and infections.

CONFIRM trial (which was designed to confirm the role of terlipressin in HRS 1 reversal and to seek its approval in USA) randomised 300 patients with HRS 1 either terlipressin plus albumin or albumin alone administered for 14 days. Given the central role of splanchnic vasodilatation and resultant reduced renal perfusion, terlipressin was clearly effective in HRS 1 reversal, (defined as two consecutive serum creatinine measurements of <1.5 mg/dl at least 2 hours apart and survival without dialysis for at least 10 days after the completion of treatment). However, terlipressin arm experienced higher side effects and mortality, especially because of pulmonary complications and sepsis.

What does the result mean for the nephrologists called to see this one of the most challenging AKI consult? Permissive ‘hypercreatininemia’ and hypoalbuminemia may be an acceptable scenario, however, at the same time these studies don’t rule out the role of albumin and vasoconstrictor therapy in properly selected patients. AFFIRM is a clear caution against overzealous albumin use outside the currently used indications. What about CONFIRM? We typically begin vasoconstrictor therapy (for me it’s cheaper and easily available noradrenaline, unlike my US colleagues, I can use it outside ICUs, which I believe is an acceptable alternative). Patients in CONFIRM trial had higher serum creatinine (than what scares our hepatologists and makes them call us), and therefore the benefits may be underestimated. What about harm? More pulmonary complications may be partly driven by higher albumin usage (83% terlipressin arm and 91% in albumin alone arm) ; albumin is known to precipitate pulmonary edema, especially in patients with severe AKI in the face of vasoconstrictor therapy.

Therapies here also should serve as a bridge to the liver transplant, and interestingly terlipressin arm of the study had lower liver transplant rates-likely because of reduction of the MELD score (renal component) or possibly because of recipients rendered unfit by pulmonary complications and sepsis.

AKIKI-2 trial: what do you mean by timing?

Consensus toward ‘delayed start’ of dialysis in AKI is finally emerging. One of the important outcomes of this ‘strategy of waiting’ is that a significant number of patients would recover without ever needing dialysis: about 38 to 40% in delayed arm didn’t need dialysis in AKIKI, IDEAL-ICU and START-AKI trial, and 17% in our experience of community acquired AKI. So those who waited longer would gain more RRT-free days.

Would a further delay result in the gain of more RRT free days? AKIKI-2 tried answering this important question in a multi-centre RCT involving 278 patients with KDIGO stage 3 AKI. In the ‘delayed strategy arm’ (n=137) RRT was initiated if participants met one of the 2 original AKIKI trial’s delayed arm criteria: blood urea nitrogen level higher than 112 mg per deciliter, or oliguria for over 72 hours. ‘More delayed strategy arm’ (n=141) received dialysis only if mandated by hyperkalaemia or metabolic acidosis or pulmonary oedema or until blood urea nitrogen concentration reached 140 mg/dl. ‘More delayed strategy’ didn’t result in more RRT-free days (12 days vs 10 days, in the ‘delayed strategy’ and ‘more-delayed strategy’ respectively p=0·93). It also was associated this potential harm [hazard ratio for death at 60 days was 1·65 (95% CI 1·09–2·50, p=0·01)

I hate the term ‘delayed start’, which gives an impression of missing your bus or train. Prof Hase, who taught us the art and science of conservative management of AKI, suggested the terms “prophylactic start” and “whenever needed start” of dialysis in AKI, which I believe more appropriately convey the nature of these interventions. (But most journals weren’t happy with these terms, and we had to settle for “Earlier start” versus “Usual start”; finally avoided delay!).

Timing based on the ‘clock time’ (hours or days after onset or admission) or based on pre-specified level of azotemia (as in AKIKI-2) fails to consider a very important characteristic of AKI i.e. complications (which may need dialysis) correlate neither with time nor with the degree of azotemia. For example, uremic bleeding or CNS dysfunction has a very poor correlation with the degree of azotemia. Also, primary outcome in AKIKI 2 involved clinician’s judgment to discontinue RRT, which might have introduced subjectivity in the assessment of this parameter. Successful discontinuations of the RRT, spontaneous improvement in azotemia, or improvement in the urinary creatinine clearance are more objective parameters to assess the recovery. To what extent, this might have affected the primary outcome is uncertain.

Definitely, no need to start early, and let’s continue to deliver ‘just when needed’. And for this don’t just rely on blood urea nitrogen and creatinine; close monitoring of volume status, taking into consideration the kinetics of solute accumulation (a muscular man with polytrauma with acceptable parameters on morning round can get a life threatening hyperkalemia in the evening), need of transfusions in the setting of oliguria or positive water balance, will help you choose the right time for an individual patient. With ‘watchful waiting’ (remember its ‘wait and watch’ and not ‘wait and sleep’!) one needn’t bother about azotemia, the highest that I remember managing with Prof Hase was a lady with rifampicin induced AKI who recovered (without needing dialysis) after peaking to BUN of 188mg/dl and serum creatinine of 17mg/dl.

Tenapanor for phosphate binding (Dr Manjunath Kulkarni)

Elevated phosphorus in patients with ESRD is associated with increased mortality. Sometimes, it is difficult to lower phosphorus despite using multiple phosphate binders, diet control and adequate dialysis. Phosphate binding in the gut has been the main strategy so far, a novel approach of controlling phosphorus by inhibiting paracellular phosphate absorption (by tenapanor) is being explored.

236 maintenance haemodialysis patients (with hyperphosphatemia which was not controlled with phosphate binders) were randomised in AMPLIFY trial to receive either oral tenapanor 30 mg twice a day or placebo for 4 weeks. The primary endpoint was change in phosphorus concentration from baseline to week 4 was better with tenapanor (phosphorus reduction of -0.84 versus -0.19 mg/dl in placebo). Approximately 40% of patients in intervention arm could achieve serum phosphorus levels of less than 5.5 mg/dl as compared to 20% in the placebo group. There was also a significant reduction in FGF 23 levels in patients who received tenapanor. The most common adverse event was diarrhoea. The investigators concluded that dual mechanism treatment using both tenapanor and phosphate binders improved control of hyperphosphatemia in patients on maintenance haemodialysis.  

The selection of study participants reflects a real world difficulty in managing elevated phosphorus levels. Short duration of follow-up and use of a variety of phosphate binders at baseline (making comparison between 2 arms difficult because they have a different capacity of P binding) are important limitations to note. Though the authors claim that the phosphate binders were balanced in both the arms of the trial, using the concept of phosphate binding equivalent dose (PBED) would have simplified comparison of phosphate binders and given a clear picture of how much phosphate binders each group was receiving. 

Data on long term and efficacy of tenapanor is not yet available. If NORMALIZE trial clarifies these issues in the future, tenapanor might be a useful weapon in armamentarium of managing hyperphosphatemia in dialysis patients. Constipation is not an uncommon problem in our patient population, and given this agent’s efficacy in management of constipation in irritable bowel, this might well serve dual purpose.

Whither cinacalcet? 
(Dr Manjunath Kulkarni)

Another important candidate in the department of surrogate nephrology is parathyroid hormone. A large EVOLVE trial found no impact of PTH lowering on cardiovascular events and death. So if not meaningful outcomes, whether lowering one surrogate (PTH) affect another surrogate (vascular calcifications) is very tempting idea. This small but well conducted RCT questions the role of cinacalcet even to reduce calcifications in patients with secondary hyperparathyroidism (iPTH > 300 pg/mL, calcium > 8.42 mg/dL). Regression of LVMI and CIMT was seen (P = 0.03 and P = 0.001) and was significantly associated with change of phosphate on multi-factorial analyses.

Now, please welcome on the board of surrogates in nephrology, a novel marker T50 – a nanoparticle based assay to measure the propensity of calcification in the serum. In a sponsored trial, which compared effect of etelcalcetide and maxacalcitol on serum calcification propensity in secondary hyperparathyroidism among hemodialysis patients, it was found that etelcalcetide was more effective in increasing T50 than maxacalcitol. A shorter T50 is surrogate marker for poor outcomes due to calcification stress (read the paper to understand what is T50). 

P.S. Higher up the crows built their nests, better the rainfall. This is the wisdom that many farmers used to predict rains. Every year IMD (Indian meteorological Department) gives a monsoon forecast, accuracy of which is no better than that wise farmer’s prediction. But so far, no farmer has attempted to alter this height of the crow’s nest for modifying the rains, as farmers are wise men and know that it’s a crude surrogate, unlike us, the desperate physicians, who are easily lured by surrogates, and run after serum creatinine and albumin to change outcomes of the patients with HRS-1, phosphorus, PTH, vascular calcifications and so on.

Why do trials which use surrogate markers as outcome variable end up with a positive result compared to those which use hard clinical outcomes as endpoints? The use of surrogate markers to get a positive result reminds us of Dr Trisha Greehalgh’s tongue in cheek tips to pharmaceutical industry (in her book how to read a paper). She says “think of a plausible physiological mechanism why the drug works and become slick at presenting it. Preferably, find a surrogate endpoint that is heavily influenced by the drug”. 

March 2021

  1. Immunosuppression for Steroid Dependent Nephrotic Syndrome (SDNS) Contributed by Dr Manjunath Kulkarni

Corticosteroids are the cornerstone in the management of childhood nephrotic syndrome. Resolution of the anasarca after treatment is one of those rare ‘aha!’ moments in nephrology. While most children respond well to steroids, in about 30-40% of them, course is complicated by either frequent relapses or steroid dependence. Long-term prognosis of renal function is good in these patients as long as they remain steroid sensitive. Significant long-term complications are related to side effects of the therapy. After the landmark ISKDC study, there was a long quiescence before we started seeing a number of RCTs addressing several important issues in the management.

In an open label, randomized controlled trial in JAMA Paediatrics by Basu et al from Kolkata (India), 120 consecutive children aged 3 to 16 years with SDNS not previously treated with corticosteroid-sparing agents, were randomized to receive either Rituximab (RTX) or Tacrolimus (TAC). At the end of one year, relapse free survival (primary outcome) was significantly better with RTX as compared to TAC (54 [90.0%] vs 38 [63.3%] children; P < .001; odds ratio, 5.21; 95% CI, 1.93-14.07). RTX also fared better in terms of lower number of patients with over one relapse, lower cumulative steroid dose, and longer time to first relapse. Those who relapsed in RTX group showed earlier recovery of B cells, suggesting potential utility of this measurement in identifying patients for repeat RTX infusions. Also, FSGS on histopathology predicted relapse irrespective of the treatments used.

All the patients were “steroid sparing treatment naïve” and thus don’t represent the real world SDNS who cannot maintain remission with alternate daily low dose steroid-which currently is probably the most common first line strategy. So the benefits in practice may not be as dramatic as reported in trial. Another limitation of the trial is the relatively short duration of follow-up; course of SDNS typically spans several years, and long-term safety and efficacy of RTX remains to be established. Levamisole and shorter courses of Cyclophosphamide (CYP)-another common strategy that many of us use-may not be so toxic as feared. Gonadotoxicity (which is not associated with the typical dose of less than 167mg/kg) and cancer are rare with short single course of CYP. (1 case of malignancy is reported till date when a desperate physician treated a 11 year boy with cytosine-arabinoside and chlorambucil besides CYP).

An RCT set out to compare RTX with low dose MMF for SDNS was stopped early (after enrolling just 30 patients) for high relapse rate in MMF arm, so leaving us clueless whether MMF was bad or RTX did better. Until we see more data and long-term outcomes of RTX, strategy of steroid minimisation (rather than withdrawal) seems appropriate, for which evidence base in getting stronger. Here is another important Indian RCT supporting this strategy, randomising 117 children to a short regimen (40 mg/m² of prednisolone for 2 weeks) or standard 4 week regimen for relapse of nephrotic syndrome. Proportion of children developing frequent relapse or steroid dependence at 12 months of follow-up was similar in the two groups (24% vs 23%, risk difference, −1%; 95% CI−15 to 16; P=0.90) as were the secondary outcomes like time to relapse, relapse rate, and steroid related side-effects during the follow-up period. As expected, cumulative steroid dose was lower in the shorter treatment arm. Although trial didn’t reach the non-inferiority criteria (sample size fell short because of lower than predicted relapse rates), it generates the important hypothesis, and steroid minimisation deserves evaluation in adequately powered trials.

2 Asymptomatic Bacteriuria (ASB) after kidney transplant

Contributed by Dr Divya Bajpai

Urinary tract infection (UTI) is the most common infection after kidney transplant and is associated with sepsis, hospitalization, rejections, allograft dysfunction, and death. With the emergence of multi-drug-resistant organisms, it is becoming increasingly difficult to treat. With the hope of early diagnosis and treatment, some of us screen for bacteriuria and then face the dilemma of treatment versus observation when the patient is asymptomatic. Asymptomatic Bacteriuria (ASB) is common and evidence so far doesn’t support treatment, however, this was limited by a few good trials and trials with a high risk of bias. Check the summary of evidence to date.

Here is a welcome addition to the existing literature on this issue: a multicenter (7 French and 6 Belgian), open-label, superiority, parallel-group, randomized controlled trial (BiRT). 199 adult KTRs (≥2 months post-transplantation) with ASB were randomized to receive either 10 days of antibiotics or no therapy and were followed up for 1 year with repeat urine cultures. Treatment did not significantly reduce the incidence of symptomatic UTIs (Primary outcome – 27% in antibiotic group vs 31% in no therapy group; hazard ratio = 0.83, 95% CI 0.50–1.40; P = 0.49). There was no difference in the primary outcome in any of the pre-specified subgroups (</> 6 months post transplantation, baseline eGFR </> 40 ml/min/1.73 m2, presence/absence of resistance organism at baseline). There was no significant difference in any of the secondary outcomes (symptomatic pyelonephritis, bacteremia, hospitalization with UTI, all-cause mortality, graft rejection, graft loss, change in serum creatinine). Antibiotics did not lead to significant improvement in graft function.

However, therapy did what it was expected to do–it significantly reduced the prevalence of ASB at 1 month (29%, versus 66%, p < 0.001) and fostered the emergence of antibiotic resistance in the treatment group (18% versus 4%, p 0.003). Findings are consistent with the recent recommendations made by the Infectious Diseases Society of America (IDSA), the American Society of Transplantation, and the European Association of Urology against systematic antibiotic use in kidney transplant recipients with ASB.

While consensus favouring ‘observation’ is emerging, some caution in the interpretation of this data is needed. Early post-transplant (<2 months) and those with a history of recurrent UTIs were excluded from the BiRT trial, a population at the highest risk of symptomatic UTI (for reasons like stent, higher immunosuppression, structural or functional abnormality of urinary tract). Of course, this doesn’t justify the treatment of ASB in these patients, but cautious individualisation of screening will enable us to identify patients for close monitoring, early diagnosis, and treatment.

The authors also published a meta-analysis of four trials (including BiRT) which doesn’t support treating ASB in KTRs. Thus, we have adequate evidence not to increase the financial and pill burden of our recipients who are in the later post-transplant period and are at low risk for UTIs. As a corollary, we should also stop screening them routinely for ASB.

3 Steroid Free Immunosuppression for Kidney Transplant

A couple of decades back, (when patients often received dual immunosuppression because of exorbitant cost of calcineurin inhibitors) you could identify kidney transplant recipients in outpatient clinics by the obvious Cushingoid look, a limp due to necrosed femur head and warts all over the body. Thanks to the generic tacrolimus and mycophenolate, which made triple immunosuppression of Tac/MMF/Pred a standard of care for the most renal allograft recipients today. As steroid minimization is adopted by most centers, the incidence of dreadful steroid toxicity has dramatically gone down. However, metabolic complications and attendant elevated CV risk remains a concern.

Steroid withdrawal is strongly advocated by some experts, although the rationale for this strategy is not strong enough IMHO. This report in JAMA Surgery is a long term follow up of a previously reported RCT comparing steroid withdrawal before day 7 versus a continuation of low dose steroid. Original trial reported higher risk of acute rejections (Banff 1A), similar patient and graft survival, and an improvement in some (not all- notably NODAT, total cholesterol, LDL did not differ significantly) metabolic risk factors like weight gain and serum triglyceride.

This follow up report presents the outcomes (derived from OPTN registry) of trial participants until 2018, after trial got over in 2002, and showed that risk allograft failure from any cause including death was 0.83 (95% CI, 0.62-1.10; P = 0.19) and for allograft failure, censored for patient death was 0.78 (95% CI, 0.52-1.19; P = 0.25) and did not differ between the patients assigned to withdraw from corticosteroids vs assigned to continued corticosteroids.

Surgeons usually are in disagreement with nephrologists for managing kidney transplant recipients. Fortunately, at our centre, the recipient is handed over to the nephrology team as soon as he/she is out of anaesthesia after surgery, so disagreement ends with the end of the surgery. Both the surgeons (authors of the paper and accompanying editorial) question the wisdom of most nephrologists continuing steroid in kidney transplant recipients. So here is the answer to our surgical KOLs in this area. We already have adopted steroid minimization as standard of care, sparing recipients of the dreaded steroid toxicities. Proposed metabolic benefits are not realized to the extent one would expect from a steroid free strategy in these trials, raising the possibility that rescue pulse therapy for higher rates of acute rejections might have neutralized the expected metabolic benefits. Steroid withdrawal also entails use of ATG in immunologically low risk population, which may not be a fair deal considering the cost of the drug and that of management of the side effects. Finally, study group is called as “Astellas Corticosteroid Withdrawal Study Group”. I have to make a note of this small font paragraph at the end of the paper: “The funding organization and original trial sponsor (Astellas) had role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication”. That’s all for now.

4 Rituximab versus Cyclophosphamide for Membranous Nephropathy -RI-CYCLO trial

Although the cyclic regimen of corticosteroid and cyclophosphamide is the first-line therapy for membranous nephropathy, there are concerns about bone-marrow suppression, gonadal toxicity, and malignancies. 

In a pilot, multi-site, open label randomized controlled trial, the RI-CYCLO investigators randomly assigned 74 adults with membranous nephropathy and proteinuria >3.5 g/d to rituximab (1 g) on days 1 and 15, or a 6-month cyclic regimen with corticosteroids alternated with cyclophosphamide every other month. The primary outcome (complete remission of proteinuria at 12 months) was achieved in six of 37 patients (16%) randomized to rituximab and 12 of 37 patients (32%) randomized to the cyclic regimen (odds ratio 0.4; 95% CI, 0.13 to 1.23)

There was a statistically non-significant trend for lower complete remission rates at 12 months in the rituximab arm for men, people   < 55 years old, and for those with more severe proteinuria or lower serum albumin.

The probabilities of achieving remission during a longer follow-up were also similar by study arm. (At 2 years, complete remission in 39% vs 38% in rituximab and cyclical regimen, OR 1.04, CI 0.35-3.08).

Combined complete or partial remission at 2 years did not differ in the groups: 87% in the rituximab group and 79% in the cyclical group, OR 1.72, CI 0.4-8.86. 

Adverse events occurred in 32 participants during the study, evenly distributed by arm (43%). As expected, infusion-related reaction was a concern in the rituximab arm and leukopenia and pneumonia were concerns in cyclophosphamide arm.

Authors point out that the slow recruitment rate may make a sufficiently large noninferiority trial difficult to complete. Also, considering the lower initial remission rates, it will be difficult to justify designing a superiority trial. What we need is long term safety and efficacy data of rituximab compared to cyclophosphamide. We also need to identify the group of patients who are unlikely to respond to rituximab and will be served well with cyclophosphamide.  

An ongoing trial is testing low dose cyclophosphamide vs the standard dose in The modified Ponticelli regimen. If successful, this will further reduce the cumulative dose of cyclophosphamide and hopefully allay the fear of gonadal toxicity and potential oncogenesis.


Placed at the backwaters of the City of Lavieville, was the Nephrona Waste Disposal and Recycling System Co. at work 24×7. Everyone marveled at the intricate system of pipes, filters and the flawless job it did, which was vital for the smooth functioning of the Lavieville. Sometimes filters used to get leaky and, if not addressed in time, get clogged, leading to havoc in the city. Cyclophos, trained under Master Pontic, was a long trusted City engineer, reputed as a man who knew his job. Even in his 80s, he was performing just as well as ever before whenever called for breakdowns. He didn’t like to be bothered by minor leakages which Rasi, an in-house plumber, was capable of handling well. Everything was going well until University of Anythingzumab was opened, started training a cop Rituxan. He was originally designed for anti insurgency squad, and earned the reputation of encounter specialist who managed to keep a dreaded terrorist-Blympho in check. After this success he was advertised in the city as man of all seasons and reasons. He was young, smart, had very cordial manners and most importantly the had godfathers in city politics. No surprise, he quickly became the chocolate boy of city folks. Before formally getting appointed as city Engineer, (he being a cop without engineering background), he had to clear various exams by certifying boards. He appeared for a couple of exams: MENTOR where he competed and apparently excelled over Siclops-another engineer trained under venerable Dancatra. We have previously analyzed the results of this exams and have pointed out the unfair nature of this competition. In another strange examination called GEMRITUX (strange because he was the only candidate to appear), he was further found to be competent for job. Impressed by these results, leading municipal authorities assigned him the job of chief engineer of the Nephrona Waste Disposal and Recycling System Co. and dismissed Cyclophos from the post. Cyclophos was still considered by many as the most dependable man when it came to serious leakages in the system. Most examinations those days were conducted by sponsors who would fund the candidate for participation (who will subsequently repay this debt over his lifetime). No sponsor was ready to negotiate with Cyclopshos, as he didn’t promise a substantial repay. Finally we are seeing Cyclophos back in exams. Preliminary results of RI-CYCLO and STARMEN examinations showed that Cyclophos is best suited for the job, and why municipal authorities should not replace him back is an enigma.

The March 2021 Infographic

Infographic by Dr Divya Bajpai

July 2020

1. Uric acid lowering doesn’t fix CKD progression 

Uric acid has long been viewed as a risk marker of  CV disease, hypertension, CKD, metabolic syndrome and has been a target of treatment (more so after generic febuxostat appeared in the market) without clear evidence of benefit. Almost alternate patient in our CKD clinic is on urate lowering therapy (in their first visit, because by second or third visit it can be safely deprescribed in most cases IMHO). Studies supporting febuxostat use for retarding CKD progression are small, single centre, and of too short a duration to conclude on such an effect. A larger controlled FEATHER trial showed no such benefit.

Things that you cannot explain by the laws of science are called ‘supernatural’ and when believers in medicine talk about them they are called as ‘pleiotropic effects’. So, what about pleiotropic benefits of allopurinol? Two large trials in NEJM should inform us now that urate lowering is ineffective as an intervention to retard CKD progression. CKD FIX randomised  369 patients with CKD 3&4 (with albuminuria and at least 3ml eGFR decline in last 1 year), to allopurinol or placebo. After a followup of over  2 years, the change in eGFR did not differ significantly between the allopurinol group and the placebo group (−3.33 ml/min per year [95%CI −4.11 to −2.55] and −3.23 ml/min/per year [95% CI, −3.98 to −2.47], respectively). Insufficient power because of incomplete enrolment and ~30% discontinuation rate in allopurinol group are the major limitation of the trial, however between the group difference in the urate level was well maintained and a futility analysis showed that even if they had enrolled 1006 patients (needed to maintain sufficient power, with drop out accounted for, they would have had a relatively low probability of finding a significant effect of the drug). 

In another double-blind RCT PERL involving 570 patients with type 1 diabetes and mild-to-moderate kidney disease (age ~51, GFR baseline 67-68ml/min, UACR ~40mg), after 3 years of treatment with allopurinol, iohexol measured GFR (kudos, very few trials have dared to do this), was no better in treatment group: decreased by −3.0 ml/min/year with allopurinol and −2.5 ml/min/per year with placebo (between-group difference, −0.6 ml per minute per 1.73 m2 per year; 95% CI, −1.5 to 0.4). The mean UACR was 40% (95% CI, 0 to 80) higher with allopurinol than with placebo. Taken together, both these trials showed the futility of urate lowering with allopurinol in CKD stage 1 to 4. 

However, accompanying editorial doesn’t seem to be convinced, and falls back on in vitro cell culture studies (inflammation, endothelial dysfunction…….. and do I need to elaborate further?), subgroup analysis of FEATHER trial (whose main results showed futility of febuxostat), and their own small study (n=60) in young hypertensives where uric acid lowering normalised blood pressure. Cognitive biases are just as difficult as retarding CKD progression, and it would be no surprise, if folks using unnecessary aggressive urate lowering will continue to do so, putting everything aside and highlighting just the one finding of these trial to their benefit -allopurinol was well tolerated with no major adverse effects. 

2. RAS inhibition and COVID 19

Corona virus spike (S) protein binds to ACE2, which serves as the cell membrane receptor for SARS-CoV-2, but that’s not enough. S protein needs priming by the action of the serine protease TMPRSS2 (transmembrane protease serine S2).  Thus, the host enzymes, ACE2 and TMPRSS2, act in concert to facilitate viral entry, which subsequently leads to the development of COVID-19. Actions of ACE and ACE2 are opposite of each other, ACE generates angiotensin 2 (vasoconstrictor) while ACE2 catalyzes degradation of angiotensin 2 to angiotensin 1-7 (vasodilator). ACE inhibitors and ARBs upregulate expression of ACE2, and ever since the first report of the epidemic, hypertension, RAS blockade and their association with risk and severity of COVID 19have been a matter of debate and discussion. Given the higher risk of COVID 19 in patients with hypertension and CV disease, it was no surprise that the use of RAS blockade was higher in these patients, but does association mean causation?  
Four different  observational studies involving large database analyses have largely settles this question (put together over 40000 patients with COVID 19 from New York, Lombardy, Florida, Ohio and Denmark). Authors used various statistical methods to address the imbalance in the characteristic in the groups at baseline, which probably was the reason for the observed ‘association of RAS blockade with COVID 19. Results of all the four studies were consistent and showed no association of use of RAS blockade with either development or severity of COVID 19. This is reassuring news and can now inform us about how to answer those panic calls from patients with hypertension.

We have always been hypercritical about the observational research in nephrology here, but rarely, such studies are so important, welcome and comforting. At the same time, let’s not buy the hypothesis that RAS blockade protects against COVID 19, which is another observed association, that should be similarly scrutinised.

For a change, we are seeing NEJM editorial office busy handling observational research apart from large RCTs.


One of the large early report of COVID 19, noted overall AKI incidence of 1.6% which increased to 4.3% in severe cases.  More recent reports show that this might be an  underestimate and the  largest report of COVID 19 associated AKI-estimating it to be as high as 34%. Of 5,449 patients admitted with COVID-19, AKI developed in 1,993 (36.6%), importantly, 53.5% of the AKI cases were stage 2 or 3 and 14.3% required dialysis, so it’s definitely more than minor bumps in serum creatinine.

Of 1993 patients with AKI, 694 died (35%), 519 (26%) were discharged, and 780 (39%) were still hospitalized. Respiratory failure and mechanical ventilation were closely associated with the development of AKI:  276/285 (96.8%) of patients requiring RRT were on ventilators. Considering the high mortality (60-80%) of AKI in ventilated patients in ICUs in general, 35% mortality in this cohort is likely to be significantly lower than the actual-as about 39% of their patients are still in the hospital and death rate in this population will add to the final estimate of the entire cohort’s mortality.

Several mechanisms have been postulated about COVID 19-AKI, like direct viral cytotoxicity, interleukins, endothelial dysfunction, micro thrombosis, and VIKI (stands for Ventilator Associated Kidney Injury-please don’t remove ventilator for this!). The largest autopsy study looking at renal pathology showed tubular necrosis as the dominant feature, with electron microscopic invasion of tubular epithelial cells and podocytes. 

NEJM features a case of COVID 19-AKI that would fit in the characteristic of the AKI cohort reported in KI, notable features being difficulty preventing filter clotting in CRRT-this has also been observed before. It’s not a time to debate about the relative superiority of one modality over another and given the logistic challenges of coping with the sudden increase in the demand of RRT, whatever is feasible, available and working locally should serve the purpose including peritoneal dialysis. A review in Lancet Infectious Diseases discussing the management of AKI in COVID, speculates that early, and aggressive CRRT (considering the role of cytokines -haven’t we learnt from sepsis?)-to be potentially useful.

These speculations don’t make our lives easy, and it’s clear that AKI identifies most critical of the ICU patients with high risk of death and every effort should be made to tackle usual suspects to prevent the development and progression of AKI, minimise the complications once it develops like: minimising nephrotoxin exposure, sepsis control, hemodynamic stabilisation, restrictive fluid strategy.

Remdesvir and dexamethasone are the only agents currently recommended outside trial settings, and former was found to be effective in reducing time to recovery. It has sulfobutylether-β-cyclodextrin (SBECD) carrier -similar to intravenous voriconazole-which might accumulate in those with GFR<30ml/min. However, if the patient is a suitable candidate for remdesvir, this review in JASN should ease the anxiety over its use in patients with renal impairment -after informed consent.

4. Kidney transplant and COVID

Although speculations are made about possible protective effect of calcineurin inhibitors on cytokine storm, this doesn’t appear to be the case in real life and high mortality in kidney transplant recipients is the consistent feature of several reports including this largest cohort in NEJM. At least 2 other published reports (here and here) reported similar high mortality, ranging from 25 to 30%. Majority of the patients in these series were recipients of deceased donor kidneys (probably received higher cumulative immunosuppression including ATG), were over 50 yrs of age, had diabetes and hypertension- explaining higher mortality-this appears to be lower ~20% in our cohort (unpublished observations). General principals of treating severe viral pneumonia should apply to patients presenting with severe COVID 19-withholding MMF, decreasing/withholding Tac in those with ARDS, and supportive care. An interesting report showing the benefit of switching Tac to CSA here, however, it’s difficult to be sure of the antiviral benefit of CSA from this, as a change to CSA essentially lowers overall immunosuppression as well and this finding needs to be confirmed in further studies before this strategy can be considered useful. Utility of antivirals, immunosuppressive and other interventions is largely unknown in this population.  

5. Selfie time: Timing of RRT initiation in AKI

Individual patient data meta analysis of 10 randomised controlled trials involving 1879 patients with severe acute kidney injury, proportion of patients who died by day 28 did not significantly differ between the delayed RRT group (366 [44%] of 837) and the early RRT group (355 [43%] of 827; risk ratio 1·01 [95% CI 0·91 to 1·13], p=0·80), corresponding to an overall risk difference of 0·01 (95% CI −0·04 to 0·06). Notably, 390 (42%) of 929 patients allocated to the delayed RRT group and who had available data did not receive RRT.

There was no heterogeneity across studies (I2=0%; τ2=0), and most studies had a low risk of bias. Result are robust and were consistent across multiple sensitivity analyses and should inform decision making of dialysis initiation in AKI.  

START AKI trial which has completed enrollment of over 3000 patients (more than all the ‘timing trials’ so far put together) should hopefully put to the rest one of the most controversial issues in nephrology practice. 

6. Cardiorenal conflict in times of COVID

Nationwide lockdown was announced, and I thought of making courtesy calls to my friends whom I haven’t talked for long. I called my cardiologist friend (why did I?)

“Hi,  how are you doing?”

Unfortunately, he misunderstood me, and thinking I’m teasing him about slack in the practice because of lockdown, he put down the phone without answering.

He is one of the leading interventional cardiologists in his city and doesn’t read our blog and nor does he waste time on debates surrounding contrast AKI. Cardiologists are straightforward guys and generally have very little spare time. He has a formula (that he uses to explain his patient when he doesn’t want to intervene) to calculate dialysis risk after cardiac cath: Risk=serum creatinine x10.

“There is 50% chance of you needing dialysis after procedure,” he told one of my patients with CKD stage 4 due to type 1 diabetes. This lady had angina on minimum exertion and when I called up to explain my point, he calmly replied “your patient will anyway land up with dialysis in few months or year, why you want to bring the blame on me?” This is cardiorenal syndrome 6, where patient dies because of a coronary event as cardiologist fears contrast. 

She was keen to enjoy the dialysis free life and stopped seeing me until she developed choking sensation after more than routine exertion.

“If you don’t want to wake up (or not wake up) in the middle of the night with a heart attack, go for angiogram at the earliest. Don’t consider dialysis as the end, it can be a fresh beginning.” I was trying to be loud and affirmative at the same time. 

After having made financial and mental preparation for cardiac cath, she came back after 3 months, this time asymptomatic, asking me to connect to the cardiac colleague. “Ok, but before I do that, let me tell you something. Life is not all the same after lock-down. We have results of ISCHEMIA-CKD trial now.”

Kudos to the investigators of ISCHEMIA CKD trial, who enrolled and maintained 777 patients with advanced CKD (57% on dialysis) and CAD (a positive stress test) in this trial for over 2 years. Patients received either an initial invasive strategy (CAG followed by revascularisation if needed) or initial conservative management and intervention if needed. Primary outcome (death or non-fatal MI) occurred in 123 patients in the invasive-strategy group and in 129 patients in the conservative-strategy group (estimated 3-year event rate, 36.4% vs. 36.7%; adjusted hazard ratio, 1.01; 95% confidence interval [CI], 0.79 to 1.29; P = 0.95). Invasive strategy group experienced a higher risk -disability (stroke), dialysis and death.

ISCHEMIA and ISCHEMIA-CKD trials are robust, multi-centric, adequately powered and should caution against the initial invasive strategy in both general and CKD population with stable CAD-such strategy is highly common in the former group but less common in the latter because of a scare of contrast. These results, however, should not foster the therapeutic nihilism- renalism-as deferring interventions is not advisable in unstable CAD, severely symptomatic cases, those with depressed LV function and heart failure-ISCHEMIA-CKD excluded such patients. Although event rates were less than expected (which might decrease the power of the study), ISCHEMIA-CKD is now the first and the largest RCT to inform decision making in this group of patients with CAD.

I called my friend, thanked him for deferring the procedure, “you are nut” was his response-case of serious misunderstanding once again. 

%d bloggers like this: