Tuesday, March 31, 2009

Statins fail again

Statins have a tortured relationship with nephrology. Our patients have accelerated atherosclerosis and they die overwhelmingly of cardiovascular disease. So one of my primary jobs is to continually optimize cardiovascular risk factors to save my patients
Control blood pressure, start an aspirin, and maximize the statin are the lather, rinse, repeat of my world.
That said we have little data that this makes a whit of difference, at least in our dialysis patients.

  • No randomized trials have been done on the role of aspirin to prevent cardiovascular events among dialysis patients.
  • Aspirin was found to increase acute coronary syndrome in an unbadjusted analysis but was not significant in multivariate analysis.
  • Berger et al. (PDF), however found a dramatic reduction in 30-day mortality for patients with acute myocardial infarction given aspirin. Unfortunatly fewer dialysis patients received ASA and other standards of heart-attack care (beta-blocker and ACEi) than patients not on dialysis.
The survival of patients based on whether they received ASA for their acute MIThe use of standard therapies for acute myocardial infarction was lower among dialysis patients, even patients deemed ideal candidates for the therapy.

Blood pressure
  • Hypertension, along with cholesterol and obesity, is subject to reverse epidemiology in dialysis patients. This means that observed epidemiology trends are the opposite of what you would expect from data on non-dialysis patients. Lower blood pressure leads to high mortality, lower cholesterol leads to higher mortality, increased BMI yields better observed survival. The observational data, however, does not mean that interventions to lower blood pressure will lead to the same bad outcomes.
  • A recent meta-analysis (PDF) of 8 randomized trials of anti-hypertensive therapy gives credence to the practice of treating hypertension in dialysis patients.

  • One thing high lighted by the trial, though, is the paucity of evidence for this treatment: They were able to find only 1,679 patients. Terrible.
  • The 4D study is one of the few randomized controlled trials in dialysis patients and unfortunately did not show any improvement in mortality with atorvastatin. The study randomized 1,255 hemodialysis patients to either 20 mg of atorvastatin or placebo. After 4 years they found the statin was safe and effective in reducing the median serum LDL cholesterol level by 42%. However, the primary endpoint—cardiac death, nonfatal MI, and stroke—was reduced by insignificant 8% (P=0.37).

  • The authors found a significant increase in fatal strokes among the patients randomized to atorvastatin. (RR 2.03, P=0.04).
  • Today came word that another randomized controlled trial on statins among hemodialysis patients, AURORA, was also a bust. Published yesterday in The Journal, AURORA randomized 2,776 dialysis patients to 10 mg of rosuvastatin (Crestor) or matching placebo. The end-point was a composite of CV death, non-fatal MI, and non-fatal stroke. Average follow-up was 3.8 years and there was no difference in the primary outcome (396 outcomes with rosuvastatin versus 408 on placebo, P=0.51).

  • AURORA found no increased risk of strokes as found in the 4D study.

EKG Changes with hyperkalemia

Last week one of our second-year fellows was called into the ER for a potassium of 9.9 mEq/L. The EKG you see above was waiting for him. He arranged for emergent dialysis. In the morning the patients EKG looked like this:

Here is the time line of events:
  • 17:24 Na 128, HCO3 9, Cl 103, BUN 100, Cr 5.6 (no potassium was reported out on the initial labs)
  • 18:06 First EKG done
  • 18:28: K=9.9
  • 18:28: U/A Sp Grav 1.012, pH 5, random drug screen positive for opioids
  • 18:45: ABG 7.05/37/408/10
  • 18:45: urine Na 89, urine Cr 50.5, FENa 4.7%
  • 23:00 initiate dialysis: 2 hours on 1 K bath
  • 01:00 complete dialysis
  • 03:30 Na 140, K 5, Cl 107, HCO3 16, BUN 67, Cr 3.8, Ca 9.1, Phos 6.4, Mg 1.4, CPK 941
  • 03:30 ABG 7.22/40/117
  • 09:20 Na 142, K 4.8, Cl 111, HCO3 15, BUN 63, Cr 3.2
  • 10:00 ABG 7.20/42/96
This patient had AKI due to prolonged decreased po intake along with a loop diuretic and ACEi. The patient initially was anuric but rapidly began to recover and by the next morning was making over 100 mL of urine an hour.

His initial EKG is the best example of a sine wave from hyperkalemia I have ever seen. Below is a cardiac cycle from V4. With a quick glance it may look like a very wide QRS complex with the t wave somewhere to the right of the picture. In reality, the QRS duration is only 176 msec and the large upward thrust is the peaked T wave.

EKG Changes with hyperkalemia
  • Peaked T waves
  • Shortened QT interval
  • Widened QRS
  • Sine wave

Wednesday, March 25, 2009

Great article on the two new PSA studies in the NEJM

The New York Times has good coverage of the latest data (US Study, European study) on prostate cancer.

I loved this beat which covers the difference between relative and absolute risk and NNT:
[After discussing the 20% reduction in mortality found in the European study] But in terms of individual risk, even that is not a huge benefit. It means that a man who isn’t screened has about a 3 percent average risk of dying from prostate cancer. If that man undergoes annual P.S.A. screenings, his risk drops to about 2.4 percent.

And there is an important tradeoff. P.S.A. testing increases a man’s risk of being treated for a cancer that would never have harmed him in the first place. The European study found that for every man who was helped by P.S.A. screening, at least 48 received unnecessary treatment that increased risk for impotency and incontinence. Dr. Otis Brawley, chief medical officer of the American Cancer Society, summed up the European data this way: “The test is about 50 times more likely to ruin your life than it is to save your life.”

Tuesday, March 24, 2009

William Schwartz, the co-discoverer of SIADH has died

William Schwartz has died.

Every Monday at noon during fellowship we had the fluid and electrolyte conference with Dr. Fred Coe. Dr. Coe has extra-sensory powers for electrolytes. When you presented, you would give just the electrolytes and he would re-create the entire case from the metabolic panel. During the lecture he would ask you to explain a certain pathophysiology and then excoriate you if your thoughts were lazy and poorly organized.

If you ever blamed a poorly characterized hyponatremia on SIADH, Dr. Coe would look at you and ask "If Bartter and Schwartz were here, at this table, right now, and were looking at the same data that you have provided, would they agree that this is SIADH? Would they?"

JASN re-printed the original 1957 article describing SIADH in 2001.

Goodbye Dr. Schwartz, yours are the shoulders we stand-on in the daily grind of clinical nephrology.

iPhone and medicine in the news

Business week: Paging Dr. iPhone

PC Mag: Meet Nurse iPhone

Good website on the iPhone in medicine: MedMacs

How not to randomize a study

If you are designing a randomized study, make sure you actualize randomize your patients. Schiffl messed this up in his study of daily versus three days a week dialysis, for acute renal failure. Schiffl achieved randomization by alternating eligible patients to three days a week versus daily dialysis. One key aspect about randomization, especially in non-blinded studies, is that the investigators cannot know what arm of the study the patient will be in prior to enrolling the patient. With alternating patients every investigator knows which arm the next patient will end-up in and they can make subtle decisions on the appropriateness of the patient or in how they present the consent form to influence the composition of the study arms.

This comes up, because my fellow sent me a paper on prophylactic dialysis prior to CABG (PDF). From the paper comes this gem:

Repeat after me, "If you know what arm the patient will be in prior to enrolling the patient, you are not running a randomized trial."

NKF Spring Clinical Meeting: Meet-up

I am going to the NKF meeting and will arrive Wednesday night. If anybody is interested in grabbing a drink, caffeinated or alcoholic, drop me an e-mail or hit me on my iPhone: 248.470.8163.

Monday, March 23, 2009

Site news

I have fully committed to this blogging thing and actually dropped the $10 on my own domain name. I am now live at http://pbfluids.com

Friday, March 20, 2009

Urinary anion gap

My fellow remembers the urinary anion gap by saying:


So a negative urinary anion gap is due to gut losses as opposed to an RTA.


iPhone, Blackberry: Fight!

I have a resident in my Farmington office this month. His primary phone is a Blackberry but he also carries an iPod Touch for medical applications. It's getting to the point that the battle for the mainstream medical computer platform is the iPhone OS.

iPhone OS 3.0 is another major nail in the coffin of the Blackberry, Palm Pre and Windows. With every revision Apple fills in the holes. The critics are running out of ammunition.

Highest creatinine I have seen in acute kidney injury

We had a patient earlier this month who presented with a creatinine that was 20 mg/dL on admission and rose to 22 on the repeat. That is the highest creatinine I have ever seen in a patient with acute kidney injury. I have a seen two patients with advanced CKD with creatinines in the mid to high thirties. (34 and 37 mg/dL).

When my fellow described the patient I was sure this was going to be CKD until she mentioned, rather triumphantly, that when she examined the patient she palpated a large bladder. She had a Foley placed and the patient voided 1300 mL of urine in the next hour.

Obstructive uropathy in a woman is cervical cancer until proven otherwise. Sure enough, a subsequent CT scan of the pelvis revealed a pelvic mass which was diagnosed as cervical cancer.

The patient was discharged with a creatinine of 1.9 mg/dL.

A few aspects of the case were interesting and surprising:
  • Obstructive uropathy causes an electrogenic type 1 RTA (hyperkalemic type 1 RTA as opposed to the hypokalemic classic type 1 RTA). Because of the RTA, these patients often have hyperkalemia out of proportion to the degree of renal failure. She was not hyperkalemic and presented with a potassium of 4.6 mEq/L.
  • The patient had a pH of 7.2, bicarbonate of 4 and a pCO2 of 8, giving her a metabolic acidosis and a respiratory alkalosis (predicted pCO2 by Winter's formula is 14±2). I had been taught that patients cannot blow off CO2 below 14 mmHG. I guess she had super lungs. As best we could tell, the respiratory alkalosis was due to anxiety and resolved the following day.
  • My fellow wanted to give bicarbonate for the metabolic acidosis, but I did not. The pH of 7.2 is fine and the patient was hemodynamically stable. Her total calcium was 4.6 and her phosphorous was 10. I was worried that giving bicarbonate would correct the acidosis which at the time was essential to prevent the hypocalcemia from causing tetany or worse. The acidosis shifts bound inactive calcium to the unbound and active ionized form.

Thursday, March 19, 2009

Journal Club: Dry weight adjustment in dialysis

Agarwal continues his streak of important studies on blood pressure in dialysis patients. This study shows that reducing the dry weight results in reductions in ambulatory blood pressures done between dialysis sessions. Agarwal had previously demonstrated that in-center blood pressure readings poorly correlated with ambulatory blood pressure. One of the key findings was that the systolic fell twice as much as the dialtolic blood pressure. This means they did not only reduce the blood pressure but they also reduced the pulse pressure, something which we really are unable to do with antihypertensive drug therapy (which reduces both the systolic and diastolic blod pressure and have little affect on the pulse pressure).

Personal note: Agarwal was one of my attendings when I was a resident and wrote one of my letters of recommendation for fellowship.

Tuesday, March 17, 2009

GOLDMARK: the real story.

I co-wrote a fluid and electrolyte book while in residency. During the final push to finish the book we enlisted some friends to help with proof reading and editing in a week-long proof-reading orgy. Joel Smith, a wayward cellular molecular biologist who ended up a lawyer asked, "Is there something special about the mnemonic for anion-gap metabolic acidosis, MUDSLEEPS? Is the word important? Or just the letters?"

I explained that it was a standard mnemonic along with its cousin MULEPILES. He said that's stupid and that we should make-up our own mnemonic. Five minutes later he came up with PLUMSEEDS, an exact anagram of mudsleeps, and we used that in the book. I thought it would be a marker of who used our book to learn acid-base, if they used plumseeds they were our's otherwise, not so much.

9 years have passed and I have yet to hear anyone use PLUMSEEDS.


This past September, my partner Susan Steigerwalt, put a letter on my desk she photocopied (she's old school) from Lancet. The letter described a new mnemonic for the differential of anion-gap metabolic acidosis: GOLDMARK. This reworked mnemonic had more going for it than an ego test, it was a complete reworking of the old and busted mnemonic for new hotness.

I blogged about GOLDMARK a few months ago and received an e-mail from the lead author. I have since e-mailed all three authors. Here's their story.

In May of 2008, Josh Emmett a second-year medical student at University of Texas Southwestern was having dinner with his dad Dr. Michael Emmett, Chief of Nephrology at Baylor University in Dallas. Dr. Emmett was telling Josh that he and a fellow were frustrated with MUDSLEEPS/MULEPILES/KUSMALE because of its obvious shortcomings: paraldehyde? No one uses that. DKA, Starvation and Ethanol, all of those cause ketoacidosis. Isoniazid/iron as causes of lactic acidosis? The next time I see that will be the first time I see that. Plus no D-lactic acid, no oxoproline, an issue that must have particularly rankled Dr. Emmett as he was an author on the definitive article on the subject.

Josh, looking for a distraction from his studies volunteered to help craft a new mnemonic. Dr. Emmett and a third-year IM resident, Ankit Mehta (who has subsequentky become a nephrology fellow with Dr. Emmett), came up with the letters they would use and the synonyms for different diseases:

Uremia could be U, R or K (Renal, Kidney)
Ethylene glycol could be A, E or G (Antifreeze, Glycol)
Oxoproline could be O or P (Pyroglutamic acid)
Aspirin could A or S (Salicylate)
Ketoacidosis could be K or D (Diabetes, though that is not nearly as good as ketoacidosis because there are other causes of ketosis besides DKA)

With that list in hand Josh hit the internets and plugged the letters into some mnemonic generating websites and came up with:
  • PULSE something
After a few days of vetting the possibilities they settled on GOLDMARK.

GOLDMARK has become my standard AGMA mnemonic. Bye bye PLUMSEEDS.
  • Glycol: ethylene glycol
  • Oxoproline: Pyroglutamic acid
  • L-lactic acid
  • D-lactic acid
  • Methanol
  • Aspirin
  • Renal failure
  • Ketoacidosis
UPDATE: Dr. Ankit Mehta sent me some notes from when they were trying to find mnemonic:
Hi Dr.Topf,
I was cleaning my desk over the weekend and found some papers on which i was scratching some other mnemonics for agma:
  • SMOK(ing) ALE
  • LAME SUDOK(u).
As you see none of them are as good as GOLD MARK. Also, some are a stretch of imagination!
hope this helps,
I weep that I won't ever get to pimp medical students on the meaning of SMOKing ALE

Monday, March 16, 2009

eGFR: the problem of false positives

Last week I saw a 58 year old African American woman who was referred to me for an eGFR (i.e. MDRD equation) of 58. Her insurance company notified her primary care doctor about this decreased GFR. I saw the letter that Blue Cross sent and it did not give the physician any guidance on what to do with this information. The insurance company just wanted to make sure the physician was aware that the patient was flagged as having CKD. The primary care doctor sent her to me for further evaluation.

The patient, however, was completely freaked out. She went on the internet and started to learn about kidney disease and to her horror found (correctly) that her lisinopril and simvastatin could cause kidney disease. Since both of these medications had been started in the last few years she suspected (wrongly) that they were the cause of her kidney disease and stoppd both of them.

Her GFRs for three years before referral had been: 65, 63, 65 and 58. When I repeated her GFR it was 62.

This is a classic case of what Dr. Harold Feldman was writing about in the Feburary CJASN (PDF). Here is a patient who stopped the two most important drugs for her future health (statin, ACEi) because of a false positive eGFR.

This article uses a Markov chain Monte Carlo method to simulate use of serum Cr or serum Cr plus eGFR for CKD screening. The model they used is illustrated below:

In the model patients gets screened once a year (a cycle) from age 60 to 78. There are 6 states patients must be assigned to:
  1. No kidney disease (CKD stage 0)
  2. No kidney disease but false positive screening test (my patient)
  3. CKD, diagnosed
  4. CKD undiagnosed (false negative screening test)
  5. ESRD
  6. Death
In each cycle every patient must be assigned to a state. Dead patients must remain dead, ESRD patients can remain ESRD or die. Patients must develop CKD (state 3 or 4) at least one cycle prior to progressing to ESRD. Patients in any living state can die. Patients with CKD (state 3 or 4) can not tranition to no CKD (state 1 or 2). And according to the text but not the figure, patients without kidney disease but false positive screening (state 2) would go back to state 1 for the next cycle.

Some assumptions in the calculation:
  • Incidence of CKD 0.7% until age 65 then 2.3%
  • Mortality without CKD 0.97% from age 60 to 70, then 2.4% after age 70
  • Mortality with CKD 0.050% (why this would be half the rate of non-CKD makes no sense)
  • Mortality with ESRD age 62-67: 15%; 67-75: 19%; and 75+: 26%
  • Annual rate of progressing from CKD to ESRD 0.076%
  • Treatment of CKD had no effect on mortality
  • Treatment of CKD reduced the annual rate of progression to ESRD by 21% (from 0.076% to 0.055%)
The Baysean test characteristics for eGFR and sCr:
  • Sensitivity of eGFR: 0.924
  • Specificity of eGFR: 0.835
  • Sensitivity of serum Cr: 0.559
  • Specificity of serum Cr: 0.950
The model used the following evaluation of CKD
  • Two clinic visits with a nephrologist
  • Limited renal ultrasound
  • Renal function panel
  • U/A, urine protein, urine creatinine
The costs for the different states are outlined in the table below:

In the initial analysis eGFR was more accurate and more cost-effective than the serum Cr. Use of eGFR kept patients off dialysis (29 patients) and reduced deaths (13 patients) at the expence of an ocean of false positives:

But when you assigned a false positive CKD a slightly lower quality of life than a true negative, 0.98 versus 1.0, the serum creatinine came out more cost effective per quality adjusted life year (QALY).

Summary: the better test (as measured by the area under the curve of a receiver operator characteristics curve) loses to the worse test because of the decreased quality of life that results from a false positive reading. The false positives were so much more prevelant that they overwelmed the benefit from the decrease in death and ESRD found with the more accurate eGFR test.

This study has received tremendous publicity, likely because noone had looked at eGFR in this way before and it had a contrarian view. While the whole nephrology community has been pushing for routine eGFR reporting along with creatinine, Feldman comes and publishes a scathing indictment. Additionally, it makes good copy to say that nephrologists developed a new way to measure renal function that dramatically increases the demand for nephrology services.

My primary concearn with this study is two fold:
  1. The eGFR equation is used as a screening test in this study and in real- life. Screening tests need to be as sensitive as possible even at the expense of specificity. The thought is that the increased false positives will be picked up with secondary testing. But a screening test never wants to give patients a clean bill of health when in actuality they have smoldering unrecognized disease. In Feldman's analysis the eGFR works perfectly as a screening test by picking up nearly all of the patients with CKD but the decreased specificity results in numerous false positives. It is interesting that it is not the cost of evaluating the false positives that results in the cost ineffectiveness of eGFR but it is the decreased quality of life that results from the anxiety associated with the initially positive diagnosis of CKD. In my mind this means we need to do a better job educating patients and providers to the nature of the eGFR test.
  2. The other problem with the study is it threatens to throw out the baby with the bath water. Even if the study does show that eGFR is not cost effective, part of the problem is that the only utility given to the eGFR is in the early diagnosis of CKD to prevent ESRD. However, I more often use the eGFR to dose adjust medications, to estimate the risk of contrast nephropathy, guide the use of loop versus thiazide diuretics. All of these uses of the eGFR cannot be replaced by a serum creatinine because the serum creatinine does not account for age, gender and race.
False positive diagnosis of CKD by the eGFR are real problems and Dr. Feldman has done the nephrology community a favor by bringing this issue to light. It would be interesting for Feldman to re-run his Monte Carlo simulation with various definitions of CKD, does an eGFR of 50 ml/min reduce the false positives enough to reduce the cost below the benefits? What about 45 ml/min (sometimes called CKD 3b)? It is important for a dialog to be initiated among primary care doctors, nephrologists and payers to come up with better definitions of CKD that don't freak our patients unnecessarily while providing the best care we can.

Tuesday, March 10, 2009

Lead time bias

As I go through the literature on early nephrology referral, I am troubled by the possibility of lead time bias. This was a large issue in the debate surrounding the optimal time to initiate dialysis. The problem comes from measuring survival from the initiation of dialysis.

Patients with good nephrologic care regularly get started on dialysis earlier than their counterparts with poor or non-existent CKD care. This is evidenced by the lower creatinines at the time of initiation of dialysis in patients with early referral seen in multiple studies.

This is consistent with my practice where I tell my patients that we "...want to delay dialysis as long as possible, but not longer than possible..." because if they have profound malnutrition or advanced heart disease due to the delay of dialysis they will do poorly once they transition to dialysis.

Because of this skew in the initiation of dialysis it is important to account for that in any analysis of survival on dialysis. I hope this short slide show makes this clear.

Note: I do not know if lead time bias is responsible for the prolonged survival with early referral to nephrology I just know that it needs to be accounted for and most literature ignores this potential source of error.

Monday, March 9, 2009

CJSAN Article on CKD care and 1st year survival on dialysis

The Nephrology Blog has an interesting post on early nephrology referral of CKD patients and outcome on dialysis. I pulled all the references from that paper which have looked at the subject. I'll be adding my thouhts on these references over next few pages so that this post becomes a quasi review article or annotated reference list.

Hasegawa et al. Greater First-Year Survival on Hemodialysis in Facilities in Which Patients Are Provided Earlier and More Frequent Pre-nephrology Visits. Clin J Am Soc Nephrol (2009) 4 595-602. (PDF)
  • DOPPS I and II databases
  • 8500 new starts to dialysis
  • Pre-ESRD Nephrology contact defined as at least one visit prior to HD
  • PNV associated with adjusted odds ratio of 0.57! (half the risk of death for one visit! what a bargain!) p<0.001
  • adjusted for age, sex, race, primary cause of ESRD, 14 summary comorbidities (CAD, CHF, other cardiac disease, htn, dm, cerebroVD, PAD, cancer, HIV/AIDS, lung disease, neurologic disorders, GI bleeding, recurrent cellulitis/gangrene)

  • Facility level data is consistent with the individual patient data. DOPPS always tries to show this because it controls for some of the biases inherent in a retrospective study. From the ESRD NephSap:
  • When we choose to examine facility practice using statistical models, patients are assigned not to the individual treatment received but to the facility’s practice (e.g., percentage of patients receiving vitamin D or having phosphorus within guideline range). Patient- or facility-level outcomes can be used. The rationale for this method is fundamental to the DOPPS and merits additional emphasis here. In standard observational analyses, the true effect of a treatment of interest may be distorted analytically by the effect of the indication to receive that treatment (treatment-by-indication bias). This bias is a fundamental challenge to inferring the causal effect of a treatment from observational data. The DOPPS facility practice-based analytic approach is conceptually similar to instrumental variable analysis, a method embraced for decades in econometrics and now used more commonly in clinical studies to address treatment-by-indication bias (9,10). The approach seeks to identify natural experiments in which patients are nearly “randomly” assigned to a particular facility practice by factors independent of clinical characteristics, such as proximity to the patient’s residence. Ideally, this mimics randomized treatment assignment in a clinical trial. Recent publications discussed theoretical considerations and provided examples in clinical medicine outside nephrology (9 –17).
  • They found a dose effect of frequency of seeing a nephrologist. seeing a nephro doc 5 times had 28% lower mortality than if they saw one once or less.
  • The people with nephrology care were more likely to be diabetic and hypertensive but were less likely to have CHF, lung disease or cancer
  • Nephrology care was also linked to marriage, employment and college educated
  • The creatinine at initiation of dialysis was lower with CKD care 7.2 versus 7.6. [Of note: with creatinines that high, a change of 0.4 only represents a difference of 1 mL/min (9 mL/min with predialysis care, 8 mL/min without)]

Innes et al. Early deaths on renal replacement therapy: the need for early nephrological referral. Nephrol Dial Transplant (1992) vol. 7 (6) pp. 467-71.

Forty-four patients who commenced renal replacement therapy between January 1983 and January 1988 died within 1 year of starting treatment. To examine the factors influencing early mortality of patients on renal replacement therapy these patients (group A) were compared with a group of 44 age- and sex-matched subjects who started dialysis over the same period and who survived more than 1 year (group B). The interval between first presentation and dialysis was significantly shorter in group A (median 36 days) than group B (median 30 months) patients. Plasma urea and creatinine were significantly greater in group A than group B at the time of first presentation to a nephrologist but not at first dialysis. Patients in group A were more often treated first by hemodialysis. Systemic disease as the cause of renal failure did not appear to influence early death. Early death on renal replacement therapy appears to be associated with late referral to a nephrologist. Early referral may be beneficial because it allows for planning of dialysis and treatment of the complications of progressive uremia.

  • 44 patients who died in the first year of dialysis (group A)
  • Compared to 44 patients (age and sex matched) who survived the first year of dialysis (group B)
  • Association found with early death and late referral to a nephrologist. Average time between first visit to nephrologist and initiation of dialysis 36 days in group A and 30 months in group B.
  • no difference in BUN or Cr at initiation of dialysis
Sesso et al. Late diagnosis of chronic renal failure. Braz J Med Biol Res (1996) vol. 29 (11) pp. 1473-8.

BACKGROUND: Recent observations in our country have shown that late diagnosis of chronic renal failure (CRF) is an important cause of late referral and late commencement of maintenance dialysis. We prospectively investigated the influence of late diagnosis of CRF on patient mortality during dialysis therapy. METHODS: Among 184 consecutive patients with nondiabetic end-stage renal disease starting chronic dialysis at the Federal University Hospital in the city of São Paulo, 106 had a late diagnosis of CRF (less than 1 month before starting dialysis) and 78 had an early diagnosis. During the first 6 months of dialysis treatment, patient survival was compared in the two groups, using the Kaplan-Meier method and the Cox proportional hazards model. RESULTS: Six-month patient survival rate was lower in the late than in the early diagnosis group (69% versus 87%, P less than 0.01). In the late diagnosis group, the hazard ratio of mortality was 2.77 (95% CI, 1.36-5.66) times that of the early diagnosis group. In a multivariate analysis, after adjusting for age, comorbid illness, and serum biochemical measurements, time of diagnosis did not remain significantly associated with mortality risk. In this analysis, age, pulmonary infection, and low serum albumin were significant predictors of mortality. CONCLUSIONS: Patients with a late diagnosis have a higher mortality risk during the first 6 months of maintenance dialysis. This increased risk is related to comorbid conditions, some of which could be prevented by predialysis care. Interventions to promote early diagnosis of CRF and adequate predialysis follow-up need to be evaluated if the survival of patients with chronic renal failure is to improve.

  • 184 new dialysis patients
  • Lack of pre-ESRD CKD care defined as initiation of dialysis within 1 month of the first visit to the nephrologist
  • 6-month rather than 1 year analysis
  • raw mortality was 31% for late dx and 13% for early diagnosis
  • HR for death in the first 6 months of dialysis with late diagnosis was 2.77
  • Despite a very high hazard ratio with late diagnosis, this was not an independent association such that after controlling for age, comorbid illness and biochemical measurments the time of diagnosis was not significant.

Jungers et al. Longer duration of predialysis nephrological care is associated with improved long-term survival of dialysis patients. Nephrol Dial Transplant (2001) vol. 16 (12) pp. 2357-64. (PDF)
  • 1057 consecutive patients
  • pre-dialysis nephrology care less than 6 months in 258 patients (193 of them initiated within a month of the first nephro encounter.)
  • 6-35 months in 267 patients
  • 36-71 months in 227 patients
  • over 71 months in 307 patients
  • mean age 54
  • only 13% diabetics (French study), little hypertension
  • half of the late presentation group had been referred earlier but had neglected or refused nephrological care
  • DM, PVD, HTN, CVD all were more prevelant with late referral
  • Cr was higher in late referral [We need to worry about systemic lead time bias]
  • Dramatic changes in the rates of catheter use and the length of hospital stay associate dwith the initiation of dialysis. Importantly the authors excluded 60 patients who had either AKI or RPGN which precluded prolonged pre-dialysis care.
  • one-year mortality was 13.6% for less than 6 months pre-dialysis nephrocare and 7.4% (6-35 mo), 7.2% (36-71 mo) and 2.5% (over 71 mo) for longer care
  • With multivariate analysis only the longest duration of care (over 71 mo) was significantly and independantly associated with survival with a RR of death of 0.56 p=0.002.
  • The two shorter periods barely missed signifigance: 6-35 mo RR=0.73, p=0.058 and 36-71 mo RR=0.71, p=0.066.
  • Age, DM and prior CVD were each significant with multivariate analysis

Kinchen et al. The timing of specialist evaluation in chronic kidney disease and mortality. Annals of Internal Medicine (2002) vol. 137 (6) pp. 479-86. (PDF)
  • 828 new on-set ESRD from 81 centers
  • time from first visit to initiation of dialysis
  • Late: less than 4 months
  • Intermediate: 4-12 months
  • Early: over 12 months
  • 213 patients were eliminated because of a lack of definitive medical records
  • Significantly associated with late referral was Black ethnicity, not attending college, not having insurance, PD, not having DM, less urine output, lower renal function, less exercise, more anemia, lower albumin, less ESA, less vascular access.
  • Late referral was associated with death
  • Notice how quickly the Late and Early curves digress. By 10 months it looks like the association is maximal and the curves are roughly parallel after that.
  • The association with timing of referral and mortality was robust and remained significant when examined with 5 recipes for controlling different factors:

  • An interesting finding was that late referral was worse for patients with diabetes and hypertension, growing segments of the ESRD population
  • Sicker patients were referred later begging the question does the sickness prevent the consultation or does lack of nephrologic care cause other co-morbid diseases to become more severe (i.e. better use of diuretics in a heart failure patient, sometimes the nephrologist is the only one still willing to use ACEi with a GFR of 18)

Roderick et al. Late referral for end-stage renal disease: a region-wide survey in the south west of England. Nephrol Dial Transplant (2002) vol. 17 (7) pp. 1252-9. (PDF)

  • 361 of 411 in England
  • Late within 4 months
  • Late' within 1 month
  • 35% within 4 months
  • 23% within 1 month
  • Again late referrals were older, sicker, had less accesses
  • 33% Six-month mortality with less than 1 month versus 16% for longer than a month
  • A late referral occured in 60% of patients with an established diagnosis of CKD for more than year (preventing the physcian or patinet from excusing themselves by stating "I didn't know.")
  • Again the cr was higher at initiation (10.7 versus 9.4 mg/dl) for late referrals.
  • Time of referral was not an independent predictor of survival at 6 months in regression analysis (p=0.293)
  • The only interesting figure comes from the discussion where the authors graph the rates of Late referrals in the medical literature and make a case that it is getting better.

Kessler et al. Impact of nephrology referral on early and midterm outcomes in ESRD: EPidémiologie de l'Insuffisance REnale chronique terminale en Lorraine (EPIREL): results of a 2-year, prospective, community-based study. Am J Kidney Dis (2003) vol. 42 (3) pp. 474-85 (PDF)

Stack. Impact of timing of nephrology referral and pre-ESRD care on mortality risk among new ESRD patients in the United States. Am J Kidney Dis (2003) vol. 41 (2) pp. 310-8. (PDF)

Winkelmayer et al. A propensity analysis of late versus early nephrologist referral and mortality on dialysis. J Am Soc Nephrol (2003) vol. 14 (2) pp. 486-92. (PDF)

Kazmi et al. Late nephrology referral and mortality among patients with end-stage renal disease: a propensity score analysis. Nephrol Dial Transplant (2004) vol. 19 (7) pp. 1808-14. (PDF)

Khan et al. Does predialysis nephrology care influence patient survival after initiation of dialysis?. Kidney Int (2005) vol. 67 (3) pp. 1038-46.

BACKGROUND: Early nephrology referral of patients with chronic kidney disease (CKD) has been suggested to reduce mortality after initiation of dialysis. This retrospective cohort study of incident dialysis patients between 1995 and 1998 was performed to address the association between frequency of nephrology care during the 24 months before initiation of dialysis and first-year mortality after initiation of dialysis. METHODS: Patient data were obtained from the Centers for Medicare & Medicaid Services. Patients who started dialysis between 1995 and 1998, and were Medicare-eligible for at least 24 months before initiation of dialysis, were included. One or more nephrology visits during a month was considered a month of nephrology care (MNC). RESULTS: Of the total 109,321 patients, only 50% had received nephrology care during the 24 months before initiation of dialysis. Overall, first-year mortality after initiation of dialysis was 36%. Cardiac disease was the major cause of mortality (46%). After adjusting for comorbidity, higher mortality was associated with increasing age (HR, 1.04 per year increase; 95% CI, 1.03 to 1.04) and more frequent visits to generalists (HR, 1.009 per visit increase; 95% CI, 1.003 to 1.014) and specialists (HR, 1.012 per visit increase; 95% CI, 1.011 to 1.013). Compared to patients with >/=3 MNC in the six months before initiation of dialysis, higher mortality was observed among those with no MNC during the 24 months before initiation of dialysis (HR, 1.51; 95% CI, 1.45 to 1.58), no MNC during the six months before initiation of dialysis (HR, 1.28; 95% CI, 1.20 to 1.36), and one or two MNC during the six months before initiation of dialysis (HR, 1.23; 95% CI, 1.18 to 1.29). CONCLUSION: Nephrology care before dialysis is important, and consistency of care in the immediate six months before dialysis is a predictor of mortality. Consistent nephrology care may be more important than previously thought, particularly because the frequency and severity of CKD complications increase as patients approach dialysis.

Schwenger et al. Late referral--a major cause of poor outcome in the very elderly dialysis patient. Nephrol Dial Transplant (2006) vol. 21 (4) pp. 962-7. (PDF)

Schmidt et al. Early referral and its impact on emergent first dialyses, health care costs, and outcome. Am J Kidney Dis (1998) vol. 32 (2) pp. 278-83. [negative trial] (PDF)

Roubicek et al. Timing of nephrology referral: influence on mortality and morbidity. Am J Kidney Dis (2000) vol. 36 (1) pp. 35-41. [negative trial] (PDF)

Avorn. ( )

Zhao et al. Physician access and early nephrology care in elderly patients with end-stage renal disease. Kidney Int (2008) vol. 74 (12) pp. 1596-602 (PDF)

Sunday, March 8, 2009

Comment on the latest MDRD article and the state of clinical nephro research

Last Thursday in journal club we reviewed the latest data on protein restriction and progression of CKD.

The best summery of the results are provided at the end of the paper:
...a very low-protein diet increased the risk of death in long-term follow-up of the MDRD Study, but had no impact on delaying the progression to kidney failure...
Imagine that the primary results had been different. Imagine for a moment that the MDRD study, rather than being one of the first of the large, NIH-sponsored, negative clinical trials in nephrology, was instead a great success. Imagine that the very-low-protein diets resulted in a delay of dialysis of 20% compared to a low protein diet and that a low-protein diet resulted in a 25% delay in progression compared to a normal-protein diet. Imagine a universe where protein restriction is the ACE inhibitors of our universe.

Now imagine if this most recent analysis came out in that universe. The above quote in this imagined universe would read something like:
...a very low-protein diet increased the risk of death in long-term follow-up of the MDRD Study, despite successfully delaying the progression to kidney failure...
How would we as a nephrology community come to terms with the fact that our primary intervention that we were advocating in a thousand CKD clinics across the land, was actually killing our patients after they start dialysis. Imagine the hand wringing as we start to realize that we were able to delay dialysis from 12 months to 18 months but at the cost of a doubling of their first year mortality from 22% to 40%.

I would be horrified and stop advocating it in my clinic but lots of my patients would adopt the low protein strategy, essentially play the lottery that this radical change in diet would allow them to escape their fate.

We as the nephrology community need to demand better research. This study stands alone (nearly? or completely?) by looking at a pre-dialysis intervention but measuring the outcome in dialysis. This study goes over the wall separating chronic kidney disease research from dialysis research. We need a name for this x-ray vision of looking through the artificial barrier between CKD and dialysis. I propose transitional research.

We need to demand that our CKD research does this. This distinction is less important when looking at CKD 3 where only 1% go on to dialysis; but when looking at CKD4 patients we need to know how that intervention affects dialysis survival. In CKD 4, 18% of patients will end up on dialysis in 5 years. (D. Keith's data, PDF)

Which of today's avant garde treatment of CKD results in a doubling of dialysis mortality?
  • Use of active vitamin D to treat secondary hyperparathyroidism
  • Treatment of anemia with ESAs
  • Use of phos binders, calcium based or otherwise
  • Bariatric surgery
  • Aggressive control of blood sugar
None of these "standard" therapies has been examined with an eye on total mortality before and after initiation of dialysis. We need the definitive studies so at some time in the future we don't have to tell a patient's family that the pills we have been prescribing may actually have caused the stroke or heart attack or cancer or...

Friday, March 6, 2009

Typo in NephCalc

My fellow noticed that the calculation for FE Urea is described as FENa in the Results.

Cryptogenic acute renal failure and emboli

Acute renal failure with the creatinine doubling everyday.
AKIN 3 on hospital day three.
Past medical history: a-fib

HPI: includes flank pain, but no hematuria.
PE: fluid overload
Low FEUrea

U/S is unremarkable, Mag3, LDH pending.

This study is afoot.

ABG questions and answers

I have been editing our consult month syllabus adding the links for the study materials as we go.

Today we fell off the syllabus. The acid-base lecture on Wednesday revealed a chasm of ignorance. We need to really pound on the basics. So today, instead of doing a Powerpoint style lecture on non-anion gap metabolic acidosis, we sat in a conference room with a couple of calculators and pounded through 23 ABG problems.


Acid Base Machine Qs

Acid Base Machine
While we were going through the questions, part of me wanted some tunes. Just saw Slum Dog last weekend.

Who Wants To Be A Millionaire? - TV Themes

Tuesday, March 3, 2009

MedCalc for the iPhone

I had never heard of MedCalc for the iPhone but MedCalc had been my medical calculator on the Palm Platform since 1999. I noted that this iPhone version had one of the same authors form the Palm Version, Mathias Tschopp.
So one more medical calculator review: MedCalc

For the iPhone version, Mathias has added a second partner, Pascal Pfiffner. This duo has put together a sharp and innovative medical calculator.
The program has created the best system for finding the equations you use. First off the equations are grouped into specialties. Then the program keeps a list of recently used programs and finally it allows you to generate a list of favorites. Mathias and Pascal have created a new system for marking favorites which is not modal. This makes it unconventional for the iPhone but it also makes generating a favorite list less of a chore. Whenever you are browsing through the equations and you see one you want to add to your favorites all you need to do is double tap the equation and it gets marked by a star. Neat. In addition to categories, favorites and recents, MedCalc also allows the user to search for equations.

The other innovation for the calculator is how the user interacts with the formula. The program does not have separate panels for entering data. all of the values can be seen on one page. The number pad is a panel which can be slid out of the way and is slightly transparent. The whole package is very slick. The units for the variables can be switched by selecting the appropriate unit on the right.

The other cool feature is that the program indicates required variables by shading the background pink. Optional data has a white background. MedCalc swithce from the 4 to the 5 or 6 variable formula on the fly depending on what variables the user inputs.

Additionally for the osmolar gap it includes the optional ethanol level. In terms of information supplied with the equation the program always supplies adequate data for the user to interpret the formula. Some of the references are not the ones I would have selected but overall its pretty slick. In terms of weaknesses, there's no FE Urea. Oh well.

Related Posts Plugin for WordPress, Blogger...