Uncategorized No comments yet

Can kidney function improve?

By on May 18th, 2012

The dogma has been that CKD is permanent and often progressive. A reevaluation of data from the AASK trial, with its’ 12 years of follow up data, has challenged that dogma in patients with hypertensive CKD. Of 949 patients with with follow up eGFR measurements, there were 31 (3%) who demonstrated positive slopes, on average +1.06mL/min/1.73m2, compared with negative slopes of 2.45mL/min/1.73m2 in the remaining patients. 10% of patients did not develop progressive nephropathy. Patients with improvement in eGFR had lower proteinuria at baseline (P/C ratio of 0.07 vs 0.04) and were randomized to the lower BP goal of MAP <92mmHg. The lower BP goal did fail to slow progression in the full AASK cohort, but in this subgroup of low grade CKD there may be some benefit.

Positive slopes in GFR have also been demonstrated in the MDRD study and in the Ramipril Efficacy in Nephropathy trials. Further research is needed to determine whether these findings can be replicated among racially diverse patients, as well as those with other causes of CKD.  

 

dialysis No comments yet

Does duration of dialysis affect survival?

By on May 17th, 2012

Fresenius investigators tried to answer this question in a matched cohort of nocturnal and control patients on conventional hemodialysis. Nocturnal patients received a mean of 7.85 hours per dialysis treatment while conventional treatments were an average of 3.75 hours. Nocturnal dialysis was associated with an adjusted risk reduction of 25% for mortality. They also found that nocturnal dialysis was associated favorably with phosphorus, WBC, predialysis BP, calcium, albumin, and hemoglobin. IDWG was larger on nocturnal therapy.

The HEMO study famously failed to show a benefit of higher dialysis dose, but sparked interest in evaluating other factors such as frequency (FHN trial) and duration of treatments.  There does appear to be a benefit to more frequent dialysis, but there are significant practical barriers to daily dialysis- including vascular access survival and reimbursement constraints. Increasing the duration of treatments seems to avoid some of these practical barriers, and has been associated with improved mortality and hospitalization.

The Fresenius study included 726 nocturnal dialysis patients and 2062 control patients, who were matched by a propensity score algorithm to evaluate comparative survival. The improved mortality on nocturnal dialysis was sustained over two years. Clearance was improved on nocturnal dialysis despite the use of smaller dialyzers and lower blood/dialysate flow rates. Phosphorus levels were markedly lower in nocturnal dialysis, likely due to rate-limited equilibration of phosphate from nonvascular body compartments to the vascular space. UF rate was significantly lower in nocturnal patients, due to increased time to remove accumulated gains, despite higher IDWG. UF rate of greater than 10-13m>/kg/h have been shown to predispose patients to myocardial ischemia and stunning due to hypoperfusion.

This was not a randomized controlled trial, and may have been subject to bias due to inclusion of larger patients with higher IDWG to the nocturnal dialysis prescription. However, the results are fairly encouraging despite this limitation.

In the same issue of JASN, investigators performed a retrospective cohort study evaluating intensive hemodialysis vs conventional hemodialysis. 338 patients who received intensive home dialysis in France, US, and Canada over 10 years from 2000 to 2010 were compared to 1388 patients in the DOPPS study in the same time period. They were matched by country, dialysis vintage, and propensity score, and transplant rates were similar between groups at the end of the study. The intensive group received an average of 4.8 sessions per week with treatment time of 7.4 hours per session. The conventional group received 3 sessions per week at 3.9 hours per session. Mortality was significantly lower in the intensive group at 13%, compared to 21% in the conventional group. Possible explanations for the improved survival may be improved BP, improved endothelial function, lower ventricular mass, slower UF rate, improved clearance of phosphorus. However, it is an observational study, and so causality cannot be determined, and unmeasured factors may have had an effect on the outcomes found.

acid/base balance, non gap metabolic acidosis No comments yet

Nongap metabolic acidosis

By on May 15th, 2012

A non gap metabolic acidosis can result from losses of sodium bicarbonate from the GI tract, addition of HCl, impairment of acid excretion,  urinary losses of organic acids replaced by chloride, or administration of chloride rich solutions during resuscitation. Distal RTA and CKD lead to impaired excretion of ammonium. A non-gap acidosis results from a limitation of acid excretion (bicarbonate regeneration) with unimpaired excretion of filtered anions (sulfate, phosphate). Bicarbonate is titrated by the hydrogen from the acids resulting in a deficit of bicarbonate and excess of acid anions. When NH4+ generation is inadequate, the anions are excreted with Na and K, resulting in Na deficit and retention of Na and Cl. This results in replacement of HCO3 with Cl. The process is similar with ketoacidosis and toluene ingestion, although NH4+ generation is increased in these situations, it can be overwhelmed by the acid anion excretion, resulting in titration of acids with Na and K.  

Iatrogenic causes of non-gap acidosis include large volume resuscitation with normal saline, TPN with cationic amino acids lysine and arginine hydrochloride, acetazolamide, prostaglandin inhibitors (produce hypoaldosteronism), and ENaC inhibitors (amiloride).

In CKD, a decrease in bicarbonate is usually observed when eGFR is <20-25%. There is a correlation of degree of CKD and the degree of metabolic acidosis. Sickle cell disease and chronic interstitial nephritis cause tubulointerstitial damage and result in a larger than expected degree of acidosis.

Disorders causing non gap acidosis can be divided by the effect on potassium levels. With hypoaldosteronism, CKD, ACE/ARB and ENaC blockers, and diseases causing intrinsic damage to collecting duct, potassium levels are elevated. When HCl is administered, potassium levels are initially high then decline due to enhanced kaliuresis from excess distal Na delivery and increased aldosterone. GI losses of bicarbonate result in hypokalemia, as do ureterosigmoidostomy and ureteroileostomy, distal and proximal RTA, toluene ingestion, ketoacidosis, and D-lactic acidosis.

If the cause of metabolic acidosis is not readily apparent, renal acidification should be assessed. Failure to excrete adequate NH4+ results in metabolic acidosis, but it cannot be readily measured. The urine anion gap can be calculated to estimate the urinary NH4+ concentration by the formula UAG=Na + K – Cl. The total positive charge of ions in the urine is equal to the total negative charge, and so the urine anion gap becomes more negative as urinary NH4+ excretion rises. (the formula ignores organic anions, calcium, and magnesium, assuming these concentrations do not change with an acid load) In general, UAG is about -20 to -50 in patients with adequate NH4+ excretion, and can be positive in patients with inadequate NH4+ excretion.

The UAG fails with anion excretion other than chloride, such as with ketonuria and hippurate excretion (toluene intoxication), where urine NH4+ excretion can be underestimated, and with states of high sodium avidity. Sodium avid states result in decreased distal Na delivery and subsequent reabsorption which decreases the lumen negative electrical gradient and creates an elevated urine pH with decreased NH4+ excretion.

The urine osmolal gap, calculated as

UOG=urine osmolality – 2(Na+K) + Urea Nitrogen/2.8 + glucose /18 

The gap (divided by 2 to reflect the contribution of the anions acompanying NH4) represents NH4 concentration in the urine. The osmolal gap can be inaccurate with changes in excretion of cations Ca and Mg, polyvalent anions, or alcohols. It cannot detect small changes in NH4+ excretion.

Urine pH can also detect impaired acidification. With an acid load urine pH should decline to less than 5.5, but can be >5.5 in patients with distal RTA. It can also be elevated in normal patients with prolonged metabolic acidosis due to buffering of H+ by increased NH3 production. Also urine pH can be low with compromised NH4+ excretion in patients with type IV RTA due to suppression of ammonia synthesis with hyperkalemia. Therefore, urine pH is only really useful information once NH4 excretion has been established using UAG or UOG.

Other tests that have been used to determine distal urinary acidification include:

Urine-Blood PCO2 test: Bicarbonate is infused to maintain serum bicarbonate of 25-26mEq/L and urine pH of 7.5. The urine-blood pCO2 difference is calculated. Avalue of >30 mmHg is found in normal individuals and less than 30mmHg in patients with impaired distal hydrogen secretion. (but false positives have been reported)

Furosemide administration: Furosemide shifts sodium distally which augments distal H+ excretion. Salt depletion and mineralocorticoid administration increases the sensitivity. 40-80mg of furosemide and 1mg fludrocortisone and collect urine for 3-4 hours. Normal patients will acidify urine to <5.3, and those with impaired hydrogen excretion fail to lower urine pH.

Urine citrate: will be low in distal RTA and high in proximal RTA.

FE HCO3: Bicarbonate appears in the urine when serum HCO3 is above the threshold for reabsorption. In normal patients, this is 25-26 mEq/L, but is 15 mEq/L in patients with proximal RTA. Proximal RTA can be confirmed by demonstrating that fractional bicarbonate excretion is >15-20% when serum bicarbonate is raised to normal values with bicarbonate infusion. Glucose, uric acid, and phosphorus also usually appear in the urine, and serum levels should be checked.

Here is a nice review of nongap metabolic acidosis from CJASN.

bones, calcium/phosphate balance No comments yet

FGF23 levels and Very Low Protein Diet

By on May 15th, 2012

FGF23 is secreted by osteocytes in response to dietary phosphorus overload or increase in 1,25 dihydroxyvitamin D. It increases renal excretion of phosphorus, decreases production of 1,25-OH D, and decreases PTH levels. In CKD, FGF23 levels increase to maintain normal phosphorus balance as renal phosphorus excretion decreases. Increased FGF23 levels are associated with mortality, CKD progression, and calcification independent of phosphorus levels.

CJASN published a prospective crossover RCT in 32 patients to assess the effect of dietary phosphorus restriction on FGF23 levels in pre-dialysis CKD patients with eGFR 20-55mL/min. Half of the study patients received a very low protein diet- 0.3g/kg/day- supplemented by ketoanalogues for one week, followed by a low protein diet- 0.6g/kg/day for the second week. The other half of the patients received the low protein diet for the first week and the very low protein diet the second week. They measured levels of FGF23, and serum and urine phosphorus levels. At one week, there was a 33.5% reduction in FGF23, 12% reduction in serum phosphate, and a 34% reduction in urine phosphate.

uric acid No comments yet

Hyperuricemia is a Risk Factor for CV events

By on May 15th, 2012

Uric acid has increasingly become a point of interest in studies of renal failure and vascular disease. The proposed mechanisms of toxicity of uric acid include- mediating inflammation, inducing endothelial dysfunction, stimulating vascular smooth muscle proliferation, and increasing oxidative stress. Experimental and human hyperuricemia is associated with hypertension, diabetes, cardiovascular disease, and CKD. Hypertension, diabetes, and CVD are all risk factors for CKD, and CKD is a risk factor for CVD and complicates treatment of hypertension and diabetes. It is not clear whether uric acid is a causative factor in mortality and morbidity or a marker of other causative factors.

Uric acid has been shown to damage residual renal function in PD patients, and there have been conflicting results regarding uric acid and renal function in CKD patients. Allopurinol has been shown in a small RCT to lower uric acid level, slow progression of renal failure, and decrease CVD and hospitalization risk in moderate CKD.

CJASN published a Taiwanese study in April 2012 evaluating the association between uric acid and all cause mortality, CV events, renal replacement therapy, and rapid renal progression in more than 3000 patients with stage 3-5 CKD. They found that hyperuricemia was a risk factor for all cause mortality and cardiovascular events but not renal replacement therapy or rapid renal progression. Patients were divided into quartiles by uric acid levels, the lowest quartile was 5.7 +/-0.8 mg/dL, the highest 10.5 +/- 1.4mg/dL. The highest quartile had more hypertension, gout, DM, cardiovascular disease, higher BMI, MAP, lower GFR, lower hemoglobin, lower albumin, higher CRP, higher phosphate, and greater use of ARB, hypouricemic agent, and diuretics. In the unadjusted model, there was an association between uric acid and rapid renal progression and renal replacement therapy, but this was not significant in the adjusted model.  The relationship with mortality remained unchanged when patients were stratified by stage of CKD.

Most of the prior studies performed evaluating hyperuricemia and risk of renal failure were epidemiologic studies of the general population, people with GFR > 60, and younger patients, and most found that hyperuricemia was associated with new onset and deteriorating kidney function. Two studies have found that hyperuricemia is associated with renal progression in IgA nephropathy.

hypertension No comments yet

Ambulatory BP monitoring in UK hypertension guidelines

By on May 14th, 2012

The National Institute for Health and Clinical Excellence released new guidelines for hypertension management in mid-2011.

1) If clinic BP is 140/90 or greater, offer ambulatory BP monitoring to confirm hypertension.

2. When using ABPM, ensure that at least 2 measurements per hour are taken during waking hours. Use the average value of at minimum 14 measurements during waking hours to confirm hypertension.

3. When using home BP monitoring to confirm hypertension, ensure that for each BP recording, two consecutive measurements are taken, at least 1 minute apart and with the person seated. Record BP twice daily in the morning and evening, and for 4 days minimum. Discard the measurements taken on the first day and use the average value of the remaining measurements to confirm hypertension.

The US has not yet supported ABPM for all newly diagnosed hypertensives. The reason for this is that hypertensive treatment recommendations have been based on office measurement of BP. In addition, ABPM is reimbursed by CMS only for patients with suspected white coat hypertension. JNC8 may address the recommendations for ABPM and home bp monitoring.

bones, calcium/phosphate balance, phosphate No comments yet

Hypophosphatemia

By on November 23rd, 2011

Phosphate is readily absorbed by the gut, stored in bones, and excreted by the kidneys. Reabsorption of phosphate is tightly regulated by the renal proximal tubules. To determine whether there is a renal cause for phosphorus losses, a 3-hour timed urine sample is collected for phosphorus and creatinine levels, and analyzed with the corresponding serum phosphorus and creatinine after an 8 hour fast. The FE phosphorus can then be calculated. If tubular reabsorption is less than 80%, a defect is present in proximal tubular phosphorus reabsorption. The test may need to be repeated several times to ensure accuracy because serum phosphorus levels and thus tubular excretion are dependent on time of day, relationship to meals, and age. However, if there is solely an extrarenal source of hypophosphatemia, tubular reabsorption should be around 100%.

FGF 23 has been the subject of much recent research on calcium/phosphate homeostasis. FGF 23 binds to fibroblast growth factor receptor 1c and its co-receptor klotho. Binding inhibits reabsorption of phosphate by reducing  expression of the sodium-phosphate cotransporters in the proximal tubule. FGF23 also decreases synthesis of 1,25 OH Vitamin D. The net result of this is to lower blood phosphate levels. Hyperphosphatemia leads to increased osteocyte secretion of FGF 23, and hypophosphatemia suppresses secretion. In hypophosphatemia, if FGF 23 is being secreted normally, 1,25(OH) vitamin D should be elevated due to suppression of FGF 23.

Causes of Hypophosphatemia:

  • Poor intestinal uptake due to Vitamin D deficiency, niacin, phosphate binders
  • Increased cell sequestration due to insulin therapy, proliferative malignancy, or bone matrix (bisphosphonates)
  • Hyperparathyroidism
  • FGF23-dependent causes: tumor induced osteomalacia or genetically linked hypophosphatemia. FGF 23 levels can be checked with a commercially available assay. Normal levels are <180 RU/mL. 
  • Alcoholism
  • Drugs
  • RTA- check bicarbonate levels
  • Fanconi’s syndrome- look for glucosuria and amino aciduria
  • Hereditary hypophosphatemic rickets with hypercalciuria
  • Malnutrition and refeeding: In a catabolic, starving state, phosphate levels are maintained but during refeeding glucose and insulin drive phosphate into phosphate-depleted cells to replenish total body stores.

There is a nice summary of causes of hypophosphatemia in this case report from NEJM.

acute renal failure, atheroembolic disease No comments yet

Renal Atheroemboli

By on November 23rd, 2011

When nephrologists are consulted following a cardiovascular procedure, the first (and often correct) thought is generally contrast nephropathy, or ATN if the patient was unstable or underwent a major surgery. However, if a patient’s clinical course doesn’t follow the typical pattern, renal atheroemboli can also cause acute kidney injury and are a relatively underrecognized complication of intravascular procedures. Compared to ATN and contrast nephropathy, presentation is usually subacute rather than acute, onset can be delayed by up to a week as opposed to 24-72 hours, and elevated creatinine lasts for a longer time period- with peak creatinine at 3-8 weeks as opposed to 1-2 weeks. Patients can develop permanent renal impairment. Atheroemboli can travel anywhere arteries travel, so the presentation depends on the location of atheroembolic infarcts and can thus be variable and difficult to diagnose.

Renal manifestations:

  • renal infarct
  • acute renal failure, presentation often subacute and delayed by days to weeks after the procedure
  • proteinuria, which can be nephrotic
  • hematuria
  • secondary FSGS
  • hypertension

Non-renal manifestations

  • fever, myalgias, nausea, anorexia, weight loss
  • livedo reticularis
  • digital ischemia (blue toes)
  • TIA or CVA, altered mental status
  • amaurosis fugax
  • retinal emboli
  • neuropathy
  • spinal cord infarct
  • abdominal pain or GI bleed due to intestinal infarct
  • pancreatitis
  • hepatitis
  • myositis

Laboratory findings (some of which also depend on the location of atheroembolic infarcts):

  • elevated BUN/Creatinine
  • elevated amylase
  • elevated CK
  • abnormal LFT
  • leukocytosis with eosinophilia
  • anemia
  • thrombocytopenia
  • elevated ESR
  • low complements
  • urinary eosinophils
  • pyuria, hematuria
  • proteinuria- not usually prominent, but nephrotic proteinuria with secondary FSGS has been reported

Treatment is supportive. Steroids have been used but have not consistently shown benefit. ACE/ARB and statin are beneficial, and nutritional support is also recommended. Further vascular intervention should be avoided. It is also important to remember that these are not blood clots and anticoagulation should be avoided as it can result in further atheroemboli. Here is a review article from the Lancet on atheroembolic renal disease.

transplant, viral infections 1 comment

Fluoroquinolones for prevention of BK viremia

By on October 21st, 2011

And uses for fluoroquinolones continue to appear.

The BK virus causes tubulointerstitial nephritis and ureteral stenosis in renal transplant patients due to a tropism for the genitourinary epithelium. Most people have been exposed to BK virus in the past, but with immunosuppression, it can reactivate. BK nephropathy can present with allograft dysfunction causing an acute or progressive decline in kidney function.  In transplant patients, the incidence of BK viremia has been estimated at up to 30%. There is a recognized progression from viremia to nephropathy, with viremia occurrring 1-12 weeks prior to the onset of nephropathy. The peak incidence of viremia is at 3 months post transplant, and the onset of nephropathy occurs on average at 9-12 months post transplant. BK nephropathy has been estimated to occur in about 10% of transplant patients.

Quinolones inhibit the BK virus DNA topoisomerase and polyomavirus associated large T-antigen helicase, inhibiting BK virus DNA replication. Investigators analyzed 25 patients who received one month of quinolones following transplantation and compared them to 160 patients who did not. 40 of those patients did receive quinolones at some point in the year of the study, and they were sub-analyzed. They found that one month of quinolone therapy resulted in a lower rate of BK viremia at one year (4% vs 22.5%), with few side effects.

electrolytes 2 comments

Rare causes of hypercalcemia

By on September 10th, 2011

When patients present with hypercalcemia, the cause is not usually difficult to determine with a careful dietary and activity history, imaging of the chest and abdomen as indicated to evaluate for granulomatous disease, and a laboratory workup including PTH, PTHrP, 25-OH Vitamin D, 1,25-OH Vitamin D, TSH, cortisol, and SPEP. But then I saw a 60 year old patient with markedly elevated serum calcium, and the above workup was completely unremarkable. He had experienced several episodes of acute symptomatic hypercalcemia in the past year, and a negative workup had been repeated each time. He did not take in excess oral or dietary calcium, and did not take Vitamin A or other supplements. This article was extremely helpful in identifying more rare potential causes of hypercalcemia in the setting of a normal/negative laboratory workup. This article identified 16 patients with advanced chronic liver disease who suffered hypercalcemia with otherwise normal laboratories, apparently as a complication of liver disease, which I believe may be the cause of the hypercalcemia in my patient.