Today’s stories: The death of a definition, Diluted pee, and No trials for kids. I kill three birds with one blog to catch up with my “2012 publications” and keep my promise to be public about what I research. If you get through these, I’d appreciate any feedback on whether or not I have achieved the goal of “plain language”? If you’ve not read about what I do before you may want to check out my “I am a pee scientist” post on acute kidney injury (AKI).
The death of a definition
Nejat M, Pickering JW, Devarajan P, et al. Some acute kidney injury biomarkers are increased in pre-renal AKI. Kidney Int 2012; 81(12); 1254-1262. Open Access: http://www.nature.com/ki/journal/vaop/ncurrent/full/ki201223a.html
We think this is so important we paid US$3000 for this to be Open Acess (available to anyone). So, what’s so important? For decades there has been a syndrome that goes by various names notably “pre-renal renal failure” [ugly name!] and “pre-renal azotaemia.” It is characterized by a short duration increase in our surrogate marker for acute kidney injury (a surrogate marker “stands in the place” of a direct measurement which can not be performed) and the preservation of the kidney’s handling of sodium (important to keep our cells from either swelling or shrinking). Importantly, it has been assumed that no real damage is done.
In this paper we show that there really is injury in this syndrome. We did this by looking at 6 molecules that appear in the urine when there is damage to the kidney. We concluded that “pre-renal renal failure” is merely a mild, though clinically significant, form of AKI. We invested in making this open access as our investigation has the potential to change the paradigm and clinical practice. As these biomarkers enter clinical practice even a small injury will be recognized as needing to be taken notice of by the attending doctors. Just how they will do this is another issue.
Ralib AM, Pickering JW, Shaw GM, et al. Test Characteristics of Urinary Biomarkers Depend on Quantitation Method in Acute Kidney Injury. J Am Soc Nephrol 2012;23(2):322–33. Abstract http://jasn.asnjournals.org/content/23/2/322.abstract
How do you know if they have watered down your beer? Some of you won’t because you’re too sloshed to notice or you drink Budweiser anyway (and nothing is more watered down than that!). For the rest of you it is a vitally important question. Drinking beer also waters down your pee (I do hope you noticed!). So does drinking coffee. Doing lots of exercise has the opposite effect. When it comes to measuring the little proteins that get excreted into our urine we need to know just how much watering down has gone on. We measure the concentration of these blighters and if there is lots of water present the concentration is lower than it otherwise may have been (and vice versa) which will affect just how well these proteins detect AKI.
In this article we tested two methods of accounting for the dilution effect and compared them against not bothering to account for it. One method involved measuring the concentration of another protein, which is supposedly proportional to the water concentration, and dividing all our results by that concentrate. The other involved measuring the urine output volume over 4 hours and working out an excretion rate (amount per minute) for each of our protein biomarkers.
Suprisingly, the best way to use the biomarker as a tool to diagnose AKI was to use just the concentration and not to try and account for water at all! However, the best way to use it to try and predict those who may need dialysis or who may die was to use the concentration divided by the urinary creatinine concentration. The paper goes on to offer an explanation as to why we think this is the case.
We also came up with a means to estimate the total biomarker excreted (total mass) over a time period (in our case 24 hours) and found that this was associated with how long people are likely to stay in the intensive care and likelihood of dying in the first week. This suggests that we may use this total excretion as an outcome measure for intervention trials aimed at reducing the injury to the kidney.
No trials for kids
Endre ZH, Pickering JW. Acute kidney injury clinical trial design: old problems, new strategies. Pediatr Nephrol [Internet] 2012; On line ahead of print. Abstract http://www.springerlink.com/content/j876512550442u94/
This appeared online just last week. Professor Endre and I do a lot of work together. He’s a Nephrologist and Scientist and I do the numbers. In this case we had been asked to write a review for the journal Pediatric Nephrology. We searched for clinical trials of acute kidney injury which compared a proposed new treatment to either an existing treatment or a placebo. We included only those trials where there were more than 30 patients and where they had been assigned randomly to either the treatment or placebo group (so called Randomised Controlled Trials). We found only 49 trials – believe me, this is a very very small number. Only 1 of those trials was in children! Some of those trials tried to prevent AKI from happening by giving the patients something before the likely insult to the kidney (eg by giving them a drug or placebo before they underwent heart bypass surgery), some gave them the drug or placebo only a day or two after the insult when it was detected by the slow (normal) method of detection and a few gave it when injury was either anticipated as having just occurred, as in one trial (our own) there was an early measure of injury. Very few of the trials reported a positive outcome and none were outstanding. The trials used a wide variety of measuring just what a positive outcome meant – this is a long standing problem in nephrology with definitions of AKI and one we have been trying to work on. We went on to make several recommendations for future trial design.