June 2019

1) MENTORing on Rituxumab for Membranous Nephropathy

Results of the multi-center, open label, MENTOR trial comparing rituximab (RTX) with Cyclosporine (CSA) are here: “Of 130 randomised, 39 of 65 patients (60%) in the RTX group and 34 of 65 (52%) in the CSA group had a complete or partial remission (risk difference, 8 percentage points; 95% CI, −9 to 25; P=0.004 for non-inferiority) at 12 months. At 24 months, 39 patients (60%) in the RTX group and 13 (20%) in the cyclosporine group had a complete or partial remission (risk difference, 40 percentage points; 95% CI, 25 to 55; P<0.001 for both non-inferiority and superiority)”. Authors conclude that RTX is non inferior to CSA at 12 months and was superior to CSA at 24 months in inducing remission of proteinuria.

This data furthers the role of RTX in this disease, but definitely should not be considered enough to use it as first line in most high risk patients with MN. First, choice of the comparator (considered ‘clever’ by accompanying editorial), is not the standard of care for most of us. Both short and long term response rates (proteinuria and ESRD) are reported to be far better in the RCTs involving CYP (about 80-90% vs 60 % with RTX in MENTOR). Second, the strange CSA treatment protocol-in patients who responded CSA was tapered over 2 months, more rapidly than one would generally do in practice [not surprisingly, over half (18 of 34) of those in remission on CSA at 12 months, didn’t maintain the remission at 24 months) and third, an 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.

‘Biocreep’ is a trap in non-inferiority trials: if you have a standard of care,say A , based on pivotal studies, you evaluate B to see if its non inferior to A and find that it indeed is. Then you come with an option C and evaluate its non-inferiority against B and show that it indeed is. But then, if we assume that C is non-inferior to A, we end up accepting progressively worse treatments as standard of care, a phenomenon called as biocreeep.

2 Amlodipine is better than Hydrochlorothiazide in BP reduction

In this 3 arm, multi-centre RCT-CREOLE-, involving 728 black hypertensives across 6 countries in Sub-Saharan Africa, combination involving amlodipine [with hydrochlorothiazide (HCTZ) or perindopril] was better at BP lowering as assessed by 24hr ABPM at 6 months (when compared with perindopril plus HCTZ), (between group difference in the change from baseline, −3.14 mm Hg; 95%CI −5.90 to −0.38; P = 0.03; and −3.00 mm Hg; 95% CI, −5.8 to −0.20; P = 0.04, respectively). Authors propose non-BP mechanism like increased endothelial NO bioavailability (which apparently is compromised in blacks), but simpler explanation of the benefit may be: longer half life, improved adherence and better BP lowering efficacy of amlodipine. Also the thiazide arm of the trial might have had a better BP lowering with longer acting and more potent chlorthalidone.

Most important attribute of an anti-hypertensive agent is ‘BP lowering’, which often is forgotten in the noise of evidence, guidelines and propaganda of ‘pleiotropic’ effects. Similar superiority of amlodipine -not only for BP lowering but also for hard CV outcomes-was demonstrated by ACCOMPLISH trial published a decade ago. Its unfortunate to see so many practitioners here switching to fancier ‘dipins’ for the fear of edema at the cost of both rupees and BP control. I love you amlodipine.

3 DOT (Directly Observed Therapy) for apparently treatment resistant hypertension

How do we assess the compliance issue in patients referred for apparently treatment resistant hypertension? Pill counts, pharmacy refill data and direct questioning -which are commonly used tools-can be ‘non-revealing’ in a significant proportion of such patients as highlighted by this work from hypertension clinic, Ottawa, Canada. About 30% of the ‘resistant’ patient showed marked (26mmHg)BP decrease after DOT, while others had less impressive decrease (3mmHg).

Study highlights the major role of compliance in the BP management and the limitations of the conventional tools to assess it. Young hypertensives progressing from onset of CKD to ESRD within months due to uncontrolled BP is a common scenario in our practice and every attempt at controlling BP is likely to yield large benefits, DOT deserves wider exploration.

4 Ready to get TRANSFORMed? Better wait

12 month outcomes of TRANSFORM trial were reviewed previously. Here are 24 month results: “In de novo kidney transplants with low‐to‐moderate immunological risk, the EVR + rCNI regimen is a valid alternative to the standard‐of‐care regimen comprising MPA + sCNI, providing comparable antirejection efficacy, stable renal function, and low rates of mortality and
dnDSA, with an advantage of significantly reduced viral infections,
up to 2 years post-transplantation.” concluded the authors of the largest RCT (TRANSFORM) evaluating non inferiority of ‘de-novo Everolimus facilitated reduced dose CNI regimen’ as compared to standard triple immunosuppression.

Primary endpoint of treated biopsy-proven acute rejection (tBPAR) or eGFR <50 mL at month 24 (47.9% vs 43.7%; difference = 4.2%; 95%CI = −0.3, 8.7; P = .006). You may get transformed, if you can forgo these: the higher rates of rejections, proteinuria, drug discontinuation in the everolimus treated patients. Moreover, proportion of those with eGFR<50ml at 24 months (can’t we now expect at least the better eGFR number with mTOR), was significantly higher in in everolimus treated patients. [474 (46.4%) vs 423 (41.6%) p=040].

As described earlier, Everolimus was protective against CMV and BK infections. While Novartis deserves applause for conducting the largest ever clinical trial in kidney transplantation, where do we go from here? Considering the overall poor efficacy in preventing rejections, poor tolerability, inferior graft function (eGFR) and higher risk of death shown by this meta analysis in patients treated with mTOR, there is no reason for its de-novo use. Probably, the only place for mTOR today may be a patient with documented CNI toxicity on biopsy, preserved GFR and no/minimum proteinuria after careful discussion with the patient.

5 Power of positive thinking

Nephrology community is on the verge of depression for multiple reasons: lack of innovations, workforce crisis and ‘negative trials’-to name a few. Kidney International seems to have realized this and has some lessons on ‘positive thinking’ in this issue. We are referring to the results of the three arm, open label, ATHENA trial evaluating de-novo everolimus (EVERO/TAC vs EVERO/CSA vs TAC/MMF) use in kidney transplantation. Not only did trial failed to show the non inferiority of de novo everolimus arm which was the primary end point, but also, once again, highlighted many of the previously noted serious issues with de-novo mTOR use: higher rejections, higher graft loss and one of the highest rate of study drug discontinuation due to side effects-over 50%!

Inspired by ATHENA– the Greek goddess of the war and courage-, authors fight their best and use the ‘power of positive thinking’ to go on discussing how this defeat can be TRANSFORMED into a win if you change your point of view. They attribute the failure of everolimus to higher than desired TAC levels in the everolimus treated arm (will that not further increase the risk of AR? such questions don’t bother people motivated by the power of positive thinking). This power is further manifest in an accompanying editorial by Novartis (sorry, I mean authors, commentators and sponsors) which conclude that de-novo everolimus/CNI is a viable alternative option for kidney transplant recipients in 2019. Aren’t you still feeling positive and motivated?

6 Serious Adverse Gastrointestinal Events Associated With Sodium Polystyrene Sulfonate

Sodium polystyrene sulfonate (SPS) is in use for hyperkalemia since 1950s , its use has been associated with rare but serious adverse events like colonic necrosis. While initial reports of serious GI issues like colonic necrosis were attributed to sorbitol, these were also reported when SPS was used without sorbitol. Evidence supporting this link was largely case reports and small cohort studies. Here is the largest population-based retrospective cohort study, looking at this issue.

SPS dispensed in an outpatient setting to adults of advanced age (66 years or more) was associated almost 2 fold risk of serious GI adverse events (hospitalization or emergency department visit with intestinal ischemia/thrombosis, GI ulceration/perforation, or resection/ostomy) within 30 days of initial prescription.
SPS use (n=27704) compared with non-use (n=20020) was associated with a higher risk of an adverse GI event over the following 30 days (37 events [0.2%]; incidence rate, 22.97 per 1000 person-years vs 18 events [0.1%]; incidence rate, 11.01 per 1000 person-years) (hazard ratio, 1.94; 95% CI, 1.10-3.41) and risk was consistent irrespective of eGFR, potassium level or presence of other co-morbidity like diabetes or CHF.

Although limited by retrospective design and high risk patient group (mean age of 78 years, many comorbidities,with 20% residing in long-term care facilities), this is the largest documentation of a rare but potentially life threatening complication of SPS use. Time to resort to other measures of K control: diuretics, decreasing dose of RASi, dietary K restriction. While newer agents like patiromer have shown efficacy, they aren’t yet available, cost may be a barrier and their long term GI safety remains to be established.

September 2018


1 Folly of surrogates 

validity ajkd

Visual abstract by Dr. Divya Bajpai (@divyaa24)

Whether its phosphorus, blood pressure, PTH, or albuminuria- the numbers guide us all in our practice. Being easily modifiable with treatments, surrogates are attractive outcomes not only in nephrology but also in other areas of medicines. However, more often than not, that’s not what patients want from us. They want us to improve the way they ‘feel, function and survive’. Here is something that highlights this once again for us: in a meta-analysis of BP lowering treatment trials (22 trials involving 69,642 participants), there was limited correlation between drug effects on surrogates (albuminuria, eGFR/creatinine) and risk of ESRD.

These findings contrast the NKF and USFDA’s view that proteinuria is a reliable surrogate of kidney disease progression. For decades, we have been deluded by surrogates, not surprisingly, we see dissociation between soft endpoints and patient-centered outcomes. Are CANVAS, EMPAREG, and LEADER  listening?

2 Biocompatible Solutions and Long-Term Changes in Peritoneal Solute Transport

Speaking of surrogate endpoints, if faster peritoneal solute transport rate is associated with poor outcomes(read technique failure and mortality and encapsulating peritoneal sclerosis), a fluid that preserves the transport characteristics of the peritoneal membrane would be an ideal PD solution.

Elphick et al analyzed the data from the multinational prospective Global Fluid Study to test the hypothesis that biocompatible solutions use would be associated with stable membrane function. This hypothesis proved false, because “solute transport rate, although starting slower in patients using biocompatible solutions, rose to similar levels seen in standard solutions after 2 years of treatment. After 2 years, there was a potentially beneficial effect of biocompatible solutions on solute transport rate, with abrogation of the increases in solute transport rate observed in patients using standard solutions. In addition, the increases in solute transport rate associated with peritonitis episodes were absent in patients using biocompatible solutions. The magnitude of these effects was less than the effect of using higher dialysate dextrose concentrations.” There may be a long-term benefit; only a large study with long-term follow up would tell us that.

3 Uric acid lowering and CKD progression 

febuxostat ajkd 2018final

Visual abstract by Dr. Aakash Shingada (@aakashshingada)

Monsoon is the lifeblood for Indian farmers and it regularly defies meteorological department’s predictions, so much so that some farmers recently filed a police complaint against the department and a farmers’ organization had threatened to close down their office.  We desperately need a better tool to predict rains. In absence of such a tool, some farmers look at the height at which a weaver bird builds its nest on a tree to predict the rains. I find myself in no better position when a patient with CKD (especially those with no significant HTN and proteinuria) asks me about the future course of his disease and if I would be able to change the trajectory in a better way.  Uric acid, for some of us, is that weaver bird’s nest.

In the multicenter, placebo-controlled, double-blind, FEATHER (Febuxostat Versus Placebo Randomized Controlled Trial Regarding Reduced Renal Function in Patients With Hyperuricemia Complicated by  Chronic Kidney Disease Stage 3). Compared to placebo, febuxostat did not mitigate the decline in kidney function among patients with stage 3 CKD and asymptomatic hyperuricemia. Not convinced by the futility of this therapy, the authors try and speculate on various reasons for the negative results of the trial like higher eGFR (65-year-old, eGFR, and stage 3 CKD—you know what I mean), a generally slower decline of eGFR in their patients, no significant proteinuria etc. Diastolic BP was lower in the febuxostat group (not surprising as they were more likely to be on ACEi/ARBs)

Only thing I am sure about febuxostat today is that it’s an effective urate-lowering agent and FEATHER just confirms this observation.

4 TRANSFORMing long-term outcome after kidney transplantation, really?

Everolimus sept 2018 final

Visual abstract by Dr. Aakash Shingada (@aakashshingada)

Set out to address the burning issue of chronic allograft injury, here is a report (yet another), an RCT comparing everolimus with reduced CNI exposure with conventional immunosuppression i.e. MMF with standard CNI dosing (arguably TAC trough levels of 8-12  vs 3-8 in everolimus arm will qualify for ‘high dose CNI’ strategy). Everolimus arm was non-inferior to conventional arm for the primary endpoint which was (novel according to Novartis): binary composite of tBPAR(treated biopsy-proven acute rejection) or eGFR<50 ml/min at 12 months post-transplant [48.2% (493) with everolimus and 45.1% (457) with MPA (difference 3.2%; 95% confidence interval, 21.3% to 7.6%)]. Infections (especially CMV and BKV) were less common in everolimus arm.

Are you ready to get transformed? You may if you are willing to forgo the following flaws: the fallacy of eGFR as efficacy endpoint, generous 10% noninferiority margin, two components of the binary endpoints not necessarily changing linearly, and exclusion of high-risk patients. Also please let go of the tolerability-mTOR arm was more than two times likely to discontinue medication due to side effects (23% vs 11.9%). So we are no further than this Cochrane review as of now.

When will we move to patient-centered research in transplantation? Neither I nor my patients will be ready to get transformed just by good-looking numbers at 12 months. They deserve much more than that.

5 Patient-Reported Experiences of Dialysis Care

 Moving away from surrogates, here is an interesting report of patient-reported experiences of dialysis care patients from a national ESRD registry receiving in-center hemodialysis in the United States. In-Center Hemodialysis Consumer Assessment of Healthcare Providers and Systems (ICH-CAHPS) survey, (this contains questions like-how often did you feel your kidney doctors really cared about you as a person? never/sometimes/usually/ always).

2939 (59.1%) reported mean ICH-CAHPS scores. Patient experience of the dialysis centers was poor (lower ICH-CAHPS score) if the center has the following characteristics: for-profit units, stand-alone centers, fewer nurses and technicians per patient, centers belonging to LDO (Large Dialysis Organisation), had more patients with minority race/ethnicity.

Although limited by various factors (39% centers didn’t report, the interplay of medical and nonmedical factors etc), this exercise should be done more and more often to better understand the gaps in the delivery of complex treatment like dialysis. Inputs derived can potentially help improve the quality of dialysis care.

6 What happens when hemodialysis is free of cost?


Visual abstract by Dr. Divya Bajpai (@divyaa24)

While patient-reported experiences of care may be less favorable in for-profit hemodialysis units, a paper published from India enlightens us about a government-sponsored program where patients are provided maintenance hemodialysis free of cost. Andhra Pradesh, a southern state in India, implements the Rajiv Aarogyasri Community Health Insurance Scheme (RACHIS) where a private insurer provides a health insurance and is paid in full by the state government. Beneficiaries are able to utilize hospital services through a network of public and private hospitals and are covered up to INR 150,000 ($ 8876) per year.

A total of 13,118 patients received HD for ESRD during the study period (mid-2008-mid-2012). The program had a good reach and the number of patients who received HD for ESRD increased from 29.5 per million of the population in 2008–2009 to 122.2 per million of the population in 2011–2012.

Of all the subjects who started HD, 2.3% received a kidney transplantation, 17.1% were reported as dead, and 63.5% had ceased treatment of their ESRD (i.e., stopped reporting to dialysis centers).

The total cost of HD-related care was $ 63.2 million. The mean annual expenditure per patient on HD-related care was $ 4821. (This contrasts with $ 89,900 in the United States!)

Costs other than dialysis costs (commuting, loss of wages, drugs, caregiver burden) could be the reason for the high dropout rate; but what is most striking here is the high mortality rate.

Maintaining low-cost services alone is not enough; the model must have in-built checks to improve the outcomes.

In authors’ word, “In conclusion, removal of out-of-pocket of cost leads to an increase in uptake of HD, confirming a previously high unmet need. The high mortality and dropout rates suggest that insurance coverage does not address all inequities in access and the barriers to maintaining long-term care.”

[Let me thank Dr. Krishna Penmatsa (@krishnadoctor1) who gave me insights into the Rajiv Aarogyasri Community Health Insurance Scheme (RACHIS)]

Monitoring the urine flow to prevent overcorrection of hyponatremia

Urine output is probably the most important parameter to monitor when one treats hyponatremia; arguably more useful at the bedside than urine lytes and osmolality (given the typical turnaround time). Once the component of volume contraction (or whatever is the cause of high ADH level) is taken care of, the stimulus for ADH secretion is taken away and this leads to water diuresis leading to a rapid rise in serum sodium. The best strategy at our disposal is frequently checking the sodium level (and using DDAVP of course).

If you are a nephrologist, you must read (and pretend that you understood every single word of it) this article published in AJKD last month. Florian Buchkremer et al have derived on theoretical grounds a safe upper limit of urine flow, dependent only on body weight, that can be easily used at the bedside.

Edelman is the root of almost all good in nephrology. Based on the Edelman equation, authors suggest using a urine flow rate of 24 mL/kg/24 h or 1 mL/kg/h (up to a maximum of 2,400 mL/24 h or 100 mL/h) as a safe upper limit during correction of hyponatremia.

If you have been called for a hyponatremia consult and are seen doing narcissistic calculations trying to predict the next morning’s sodium and find that you are terribly wrong, you are welcome to the club! Discrepancies between actual plasma sodium changes and those predicted by the Edelman equation have been reported, but most of that is related to the things we assume. The equation stands tall! But one thing we can take from this paper is the order we are going to write- “Document total urine volume [as voided or every 2 hours in patients with catheters]. Notify the treating physician as soon as the cumulatively voided volume exceeds 6mL/kg in 6 hours.”