- How (in)accurate is eGFR?
Precision medicine is the ‘talk of the town’ and our oncology colleagues, armed with an array of markers, genomics, proteomics, etc; often mock us for being ‘specialists’ farthest from ‘precision’. The eGFR-basic tool recommended for kidney function assessment epitomizes this imprecision. We have discussed eGFR and its limited utility in day-to-day practice in our previous post. As there is little doubt about its limited accuracy, the question is ‘how inaccurate is eGFR?’ In, what may be called one of the largest studies on mGFR-eGFR comparison, individual-level inaccuracy in eGFR was quantified in 3223 patients from 4 different patient cohorts. The results hopefully will make the nephrology community and guideline developers reexamine their overreliance on eGFR and CKD staging based on it. The following tables give a glimpse of this inaccuracy which is too large to continue relying on eGFR in practice.
| eGFR | mGFR range | ||
| 50% | 80% | 95% | |
| 60ml/min | 52-67ml/min | 45-76ml/min | 36-87ml/mim |
| 30ml/min | 27-38ml/min | 23-44ml/min | 17-54ml/min |
| mGFR>60 | mGFR <45 | |
| eGFR 45-59ml/min | 36% | 20% |
| mGFR >30 | mGFR <15 | |
| eGFR 15-29 ml/min | 30% | 5% |
The addition of cystatin C didn’t improve this inaccuracy substantially. Given these results, it wasn’t surprising that there was substantial disagreement in CKD staging by mGFR and eGFRcr. For example, you may consider referring somebody for AVF creation based on eGFR 20 ml/min, only to be embarrassed by the mGFR value of 39ml, which may be enough kidney function for patient’s lifetime.
The study was cross-sectional observational and relied on a single serum creatinine measurement, however, it clearly documents how far away eGFR is from the truth in clinical practice and argues for wider use of mGFR in practice and mandatory use of mGFR in research. mGFR may not be as cumbersome as is always thought and wider application in practice may be possible. Until then, well-done and well-read serum creatinine (along with proteinuria), is all that is needed in practice.
For population-level interventions, eGFR may be an excellent metric. Population-level differences between mGFR and eGFRcr were small; the median difference (mGFR-eGFR) was 0.6 (95% CI, 1.2 to 0.2); however, for individual patient care, it is useless and may potentially be misleading. For the labs that automatically report eGFR alongside serum creatinine, now they should also incorporate the range of possible eGFR values, which will sound something like this: Mr Bean your serum creatinine is 1.3 mg, eGFR is 60ml, and this may correspond to the actual GFR anywhere between 36-87 ml. If labs start doing this (which is warranted by this study results) patients and clinicians will eventually stop paying any attention to this number. If you are reading this eGFR review, better to know beforehand that probably, it’s time to retire Mr eGFR.
2. Surgical treatment of asymptomatic kidney stones
Small (<6mm), asymptomatic kidney stones are typically treated conservatively. In this randomized controlled trial involving 73 patients who underwent surgery for a primary stone (one producing symptom /obstruction/considered high risk for an adverse clinical event) were randomized to receive clearance of secondary stones [defined as small (≤6 mm), asymptomatic renal stones that were located in the contralateral kidney (in the case of a primary renal stone) or either kidney (in the case of a
primary ureteral stone, with the specific kidney identified before randomization]. At the end of 4.2 years of follow-up, the intervention arm had significantly lower relapse rates (hazard ratio, 0.18; 95% CI, 0.07 to 0.44), and a longer time to relapse(1631.6±72.8 days vs. 934.2±121.8 days). This exercise added only a little additional surgical time (25.6min) and was not associated with additional ED visits post-procedure.
The trial has limitations. It was open-label, small in size, had few non-white patients, and involved skilled endo-urologists, limiting the generalisability of these results in real life. Details of the medical evaluation and management to prevent recurrence aren’t available, leaving open the question if some of these recurrences were preventable. However, it’s not unusual to encounter a patient who has undergone half a dozen procedures for stones without even checking the serum calcium and phosphorus once – forget about the detailed evaluation (despite such harsh reminders). Most importantly, stone disease in our patients poses risk of life threatening sepsis, AKI and ESRD. This is unlike much of the developed world where it’s mainly the pain that bothers patients with nephrolithiasis.
Symptomatic primary stone disease in many patients is not addressed in time (cost and availability of expertise)and unfortunately remains one of the leading causes of ESRD and wider access to evaluation and intervention is urgently needed.
3. Role of steroids in Infection-related glomerulonephritis
The role of high-dose glucocorticoids in bacterial infection-related IRGN remains unproven. Bacterial IRGN can occur after a bacterial infection or in presence of an ongoing bacterial infection. Not only the utility of steroids is unproven, but their use also carries serious potential risks. This open-labelled study conducted at an academic center in India compared steroids added to supportive care versus supportive care alone.
Fifty-two patients with biopsy-proven infection-related glomerulonephritis and serum creatinine greater than 1.5 mg/dL were randomized to receive corticosteroids plus supportive care (intervention arm) or supportive care alone (control arm) and were followed for 6 months. Patients randomized to the intervention arm were given intravenous methylprednisolone, 1 g daily, for 3 consecutive days. This was followed by oral prednisolone 1 mg/kg/day for 1 month, followed by a slow taper at 5 mg/week. The primary outcome was complete renal recovery(eGFR>60ml/min) at 6 months.
At 6 months, there was no difference in the primary outcome in both arms.17 of 26 patients (65.4%) in the intervention arm and 14 of 26 patients (53.8%) in the control arm had complete renal recovery (odds ratio 1.6; 95% confidence interval [CI], 0.5 to 4.9; P = 0.397). However, there was a significant increase in adverse events in the steroid group. Adverse events occurred in 12 patients (46.2%) in the intervention arm and 2 patients (7.7%) in the control arm (P=0.002).
Authors should be commended for conducting such trial to fill an important knowledge gap. Apart from obvious design issues like open-label and single-center study, there were several other serious limitations of the trial. The trial was not prospectively registered and was underpowered as the sample size fell short of the intended 91 patients which were calculated to provide 80% power to detect a difference of 20% in primary outcome. Though this trial is a welcome addition to define the role of steroids in IRGN, the high risk of bias in the trial will not allow any definitive conclusions regarding the role of steroids in IRGN.
4. Diamox (acetazolamide) is back
Diamox (acetazolamide as a diuretic in the presence of metabolic alkalosis), Deviry (progesterone as a respiratory stimulant), and Deriphylline (oral bronchodilator) were three famous ‘Ds’ for managing hospitalised patients with ‘cor pulmonale’ when I was resident in internal medicine. Most of them are rarely used today, but acetazolamide is all set to make a big comeback with the publication of the ADVOR trial, which evaluated whether the addition of intravenous acetazolamide to standardized intravenous loop-diuretic therapy would improve the incidence of successful decongestion among patients with acute decompensated heart failure.
Of the 519 patients randomized, successful decongestion (primary outcome defined as the absence of signs of volume overload, within 3 days after randomization and without an indication for escalation of decongestive therapy) occurred in 108 of 256 patients (42.2%) in the acetazolamide group and 79 of 259 (30.5%) in the placebo group (RR, 1.46; 95% CI, 1.17 to 1.82; P<0.001). Other efficacy (death, readmission within 3 months) and safety outcomes (worsening kidney function, hypokalemia, hypotension, and adverse events) were not different in the two groups.
This study is a welcome addition (now the largest RCT of diuretics in HF)to the previously existing limited evidence to guide therapy in these settings. It brings back a forgotten diuretic that is economical and likely to be available widely. Although death rates reported are much lower than expected and were not different between the groups, decongestion is an important treatment goal in practice. Important limitations of the trial include the inclusion of only white patients, exclusion of newly diagnosed patients with HF, the control group receiving a fixed loop diuretic regimen (without scope to use infusion/higher dose/sequential blockade that is typically used in practice), and the lack of the current standard care with SGLT2 inhibition.
[P.S. curious thing about the protocol was the daily addition of half a liter of fluid -isotonic bicarbonate with magnesium supplementation- as maintenance fluid! Such maintenance fluid in patients with volume overload maintains only a congested state and patency of the IV line, and we are not sure about the latter.]
5. DIAMOND that didn’t shine as expected
The DIAMOND trial was planned to examine if the addition of potassium binder -patiromer will allow us to keep more patients on evidence-based therapy for heart failure -MRAs (mineralocorticoid receptor antagonists), ACE inhibitors, or ARBs-which is often limited by hyperkalemia. Less incidence of hyperkalemia—>more use of evidence-based therapy for HF—->improved survival/fewer hospitalizations with HF: this was a very attractive hypothesis of the DIAMOND trial, and the sponsors hailed the results as a breakthrough:
The efficacy of this drug to lower serum potassium was already documented in the OPAL-HK trial, and the goal of DIAMOND was to test the hypothesis of whether the addition of patiromer will permit wider use of evidence-based therapy of HF (ACE inhibitors, ARBs, MRAs) and thereby will bring about CV benefits. The original study end-point was robust: time to the first occurrence of cardiovascular death or cardiovascular hospitalization-which was modified due to the slow recruitment and pandemic situations to ‘mean change in serum
potassium from the baseline’.
The pandemic was a hard time for running RCTs, and the authors should be congratulated for completing the trial, however, DIAMOND couldn’t demonstrate any of the hypothesized benefits of patiromer. At the median follow-up of 27 weeks involving 878 patients, the adjusted mean change in serum potassium was +0.03 mmol/l (95% CI–0.01, 0.07) in the patiromer group and +0.13 mmol/l (95% CI 0.09,0.16) in the placebo group, for a between-group difference of –0.10 mmol/l (95% CI –0.13, –0.07; P< 0.001). This hugely statistically significant difference in our opinion is clinically not meaningful. Authors made more than half a dozen comparisons for secondary and exploratory outcomes to make the same point of better control of hyperkalemia.
However, there are important lessons to be learnt from the results of this trial. Over 80% of the patients could continue the recommended doses of MRAs even in the placebo arm after having a history of hyperkalemia in the past, and this should caution against the practice of discontinuing the drug permanently after an episode of hyperkalemia (dose reduction or withdrawal followed by reinitiation is a better idea). Despite the clear benefit of the MRAs to reduce death and hospitalisations (RALES and EMPHASIS HF), only 15-30% of patients receive these drugs and one of the reasons may be knee-jerk reaction of stopping it permanently for hyperkalemia. As even the lower doses of these drugs are effective in reducing CV deaths and hospitalisations, DIAMOND trial showed that most patients can be maintained on MRAs even without patiromer.
More than 400 patients will need to be treated with patiromer for 3 years (which will cost about just 36000 USD) to prevent one MRA-preventable cardiovascular death or heart failure hospitalization, based on the results of the DIAMOND trial. One may wonder if wider use of other evidence-based agents like (Sacubutril/valsartan and SGLT2 inhibitors-both may attenuate the risk of hyperkalemia) may be more cost-effective than using patiromer indefinitely.
Last Month in Nephrology 