HDCN -- MGR -- Alkalosis in Dialysis Patient (continued)

HDCN-MGR: Alkalosis in a Dialysis Patient (continued)

Answers to Questions

1. Correct answer is (c) 38 mM
Since each 3/4 capful of Bromoseltzer (1 dose) contains 2.781 grams of NaHCO3, and the molecular weight of NaHCO3 is 84, then each dose contains 2781 mg/(84 mg/mmol) = 33.1 mmol. Twenty-three doses would equal 761 mmol of NaHCO3. In an 85 kg man with a serum bicarbonate of 20 mM, assuming a bicarbonate space of 0.5 x body wt (kg) or 42.5 L, the baseline total body bicarbonate would be 850 mmol. After ingestion of 761 mmol of bicarbonate, the total body bicarbonate would be 761 + 850 = 1611 mmol. Assuming no change in bicarbonate space, the serum bicarbonate would now be 1611 mmol/42.5L = 38 mM.

2. Correct answer is (c) Enhanced aluminum absorption
The presence of citric acid in over-the-counter medications is of concern in hemodialysis patients because of its propensity to markedly increase the gastrointestinal absorption of aluminum. If given in conjunction with aluminum-containing phosphorus binders, aluminum toxicity with resultant bone disease and encephalopathy can result. Nephrologists must counsel ESRD patients to avoid these products because of these dangers. Whereas citrate ingestion generates alkali, citric acid ingestion does not, as the alkali and acid offset one another. Citrate does chelate calcium locally, but this is of no consequence metabolically. Citrate ingestion will buffer stomach pH, and will not enhance stomach acidity.

3. Correct answer is (b) 28 mM or (c) 0 mM
Clearly, a dialysate prepared with reduced bicarbonate is indicated. Some authors have suggested that, in cases ofŒ extreme metabolic alkalosis, dialysate can be prepared without the base concentrate of a two part bicarbonate dialysis system, yielding a bicarbonate-free dialysate. The usual two part bicarbonate system contains an acid concentrate that after dilution yields 4 mM of acetic acid as well as electrolytes. The base concentrate after dilution contains 39 mM of bicarbonate. The 4 mM of acetic acid titrates 4 mM of bicarbonate, generating sufficient CO2 gas to lower pH to physiologic range and prevent calcium phosphate precipitation; however, the resulting 4 mM of acetate will subsequently generate 4 mM of bicarbonate in the body. Since our patient's metabolic alkalosis was not that severe we elected to dialyze him with a reduced bicarbonate concentration of 28 mM (this is achievable with most modern dialysis machines using standard base concentrate). Fluid removal was imperative as evidenced by his hypoxia and early pulmonary edema. No doubt the approximately 16.7 g of sodium contained in the ingested Bromoseltzer contributed to his fluid overload by obligating fluid intake in response to thirst.

Hospital course

The patient underwent hemodialysis for 3 hours using a Fresenius F8 dialyzer, over which time 3L of fluid was ultrafiltered. The bath was prepared using a potassium concentration of 4.0 mM and a bicarbonate concentration of 28 mM for the first hour followed by 30 mM for the subsequent two hours. After one hour of dialysis a blood gas drawn from the arterial port showed: pH 7.48, PCO2 46, PO2 113, HCO3 33. At the end of three hours of dialysis, values were: pH 7.44, PCO2 42, PO2 109, HCO3 29. The patient was begun on chlorpromazine for his hiccups which afforded him relief. He was treated empirically for a right lower lobe infiltrate on chest X-ray with erythromycin. His fever subsided and all cultures were subsequently negative. He was advised to avoid Bromoseltzer.

General comments

In dialysis patients, the inability of the failed kidney to excrete hydrogen ions and thus regenerate bicarbonate most frequently results in metabolic acidosis because of the obligatory daily production of acid metabolites. Dialysis solutions are usually prepared with bicarbonate concentrations of 35 mM to correct the metabolic acidosis. However, occasionally clinicians encounter dialysis patients with metabolic alkalosis which if severe can be life threatening and thus requires urgent correction. This is most effectively accomplished by hemodialysis with a reduced bicarbonate concentration in the dialysate. The most common etiology of metabolic alkalosis in this patient population is loss of acid from the upper gastrointestinal tract either through protracted vomiting or gastric suction. In this patient, the cause was an unusual one: ingestion of sodium bicarbonate contained in the BromoSeltzer.

It should be remembered that one of the causes of persistent metabolic alkalosis is renal failure, as the kidneys cannot function to excrete the retained bicarbonate. Therefore, metabolic alkalosis will only slowly correct (due to retention of endogenous acid produced by metabolism) unless a dialysate with a lower than ambient bicarbonate concentration is employed.

References

1. Lew S. Metabolic alkalosis in patients with end-stage renal disease. Contemporary Issues in Nephrology 27: 263-82, 1993.

2. Quintanilla A, Singer I. Metabolic alkalosis in the patient with uremia. Am J Kidney Dis 17: 591-5, 1991.

3. Froment DP et al. Site and mechanism of enhanced gastrointestinal absorption of aluminum by citrate. Kidney Int 36: 978, 1989.

4. Gerhardt R et al. Acid dialysate correction of metabolic alkalosis in renal failure. Am J Kidney Dis 25: 343-5, 1995

5. Kheirbek AO, Ing TS, Viol GW et al. Treatment of metabolic alkalosis with hemofiltration in patients with renal insufficiency. Nephron 9: 129-32, 1979.

That the bicarbonate space is not necessarily 50%
This is correct. The apparent bicarbonate apace (defined by the change in plasma bicarbonate concentration with bicarbonate administration) will vary according to the prevailing plasma bicarbonate concentration, i.e., it will be higher in metabolic acidosis than in metabolic alkalosis. This is because administered bicarbonate will be titrated to a greater or lesser degree by hydrogen ions that are associated with non-bicarbonate body buffers. During the development of severe metabolic acidosis, much of the retained acid is buffered by non-bicarbonate buffers; when bicarbonate is administered to correct the acidosis, much of it is consumed by hydrogen ions released from non-bicarbonate buffers: thus the apparent bicarbonate space is increased. In a recent editorial, Fernandex et al. (Kidney In 1989; 36: 747) derived a formula for bicarbonate space:

Bicarbonate space = [0.4 + (2.6/[HCO3])] x body weight (kg)

Using this formula, it can be seen that the apparent bicarbonate space will be about 100% of body wt at a [HCO3] of 5 mM, 66% of body wt at a [HCO3] of 10 mM, 53% of body wt at a [HCO3] of 20 mM, and 47% of body wt at a [HDCO3] of 38 mM.

Thus, metabolic adcidosis has a marked effect on bicarbonate space, whereas there is little change in bicarbonate space with metabolic alkalosis. We used a bicarbonate space of 50% in our calculations (mean of bicarbonate spaces at [HDCO3] of 20 and 38 mM).

That the initial electrolytes seem unlikely in a dialysis patient.

Indeed the potassium concentration was low and the bicarbonate concentration was high, reflecting the massive ingestion of alkali. Please note that all values are in mM; the corresponding values in mg/dl for BUN and creatinine are 45 and 11.3, respectively.
David J. Leehey, M.D. ()
Hines, IL USA-Saturday, March 1, 1996 at 12:54:45 (CST)