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.
3. COVID AKI
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.