Electrolyte abnormalities in patients with chronic renal failure receiving parenteral nutrition

Electrolyte abnormalities in patients with chronic renal failure receiving parenteral nutrition

Duerksen, Donald R


ABSTRACT. Background and Methods: Chronic renal failure frequently is complicated by elevations in serum potassium, phosphate, and magnesium. Consequently, parenteral nutrition (PN) solutions used to treat malnourished patients with chronic renal failure usually are prepared with little supplementation of these cations. Four malnourished patients with chronic renal failure and electrolyte abnormalities are reported. Results: Four patients developed significant hypophosphatemia 3 to 5 days after starting PN. Although carbohydrate infused via PN initially was not excessive (1.4 to 2.0 mg/kg/min), two patients received addi

tional dextrose through continuous ambulatory peritoneal dialysis (CAPD). Two of the four patients received insulin during PN. Other electrolyte abnormalities included hypomagnesemia (1 patient) and hypokalemia (3 patients). Conclusions: Malnourished patients with chronic renal failure receiving PN are at risk of developing electrolyte abnormalities, particularly hypophosphatemia. The electrolytes of these patients should be monitored closely when nutrition support is begun, and supplementation should be started as levels begin to fall within a normal range.

(Journal of Parenteral and Enteral Nutrition 22:102-104, 1998)

Malnourished individuals are at risk of developing rapid decreases in serum phosphorus, magnesium, and potassium when initially challenged with a significant carbohydrate load.’ The clinical result of these abnormalities has been referred to as the refeeding syndrome.l The major risk factors for developing these refeedingassociated electrolyte abnormalities are significant weight loss before nutrition supplementation, large initial carbohydrate infusions, and low supplemental cation levels in the nutritional formulation.2

Chronic renal failure frequently is complicated by elevations of several serum electrolytes including potassium, magnesium, and phosphorus. Because of this, parenteral nutrition (PN) is usually administered with a solution containing low levels of these cations. Because the serum levels of these cations frequently are elevated when PN is initiated, there is generally a low index of suspicion that significant decreases in the levels of these electrolytes typical of refeeding will develop.

We report four patients with chronic renal failure who developed significant hypophosphatemia after starting PN. Changes in magnesium and potassium also were present in some patients. Recognition of this potential complication is important to prevent morbidity related to these abnormalities.


Patient I

A 65-year-old man was admitted to hospital with an ischemic right foot, requiring a below-knee amputation. He had a past history of diabetes mellitus complicated by chronic renal failure, retinopathy, peripheral vascular disease, and severe gastroenteropathy. His chronic renal failure was managed with continuous ambulatory peritoneal dialysis (CAPD).

A nutrition history revealed poor oral intake for several weeks before admission and a weight loss of 10% of his usual body weight. His serum albumin was 12 g/L. He was intolerant of nasojejunal tube feeding and therefore PN was initiated. Pre-PN electrolytes revealed a potassium of 2.6 mmol/L and magnesium of 1.16 mmol/L. Throughout his course of treatment, 40 to 60 mmol/d of supplemental potassium was required. By day 8 of PN, magnesium had fallen to 0.79 mmol/L; no supplementation was provided. Phosphate levels are summarized in

Table I.

Patient 2

A 75-year-old man with a 2.5-year history of chronic renal failure (secondary to hypertension) and who was receiving CAPD was admitted to hospital with abdominal pain. A computed tomographic (CT) scan demonstrated an abdominal abscess secondary to a perforated sigmoid diverticulum, requiring percutaneous drainage. He subsequently developed a colocutaneous fistula. As a result of poor oral intake for 3 to 4 weeks, PN was initiated. An accurate weight before his admission was unavailable.

Because the initial potassium was 3.0 mmol/L, he required supplementation with 30 to 60 mmol/d via the PN. A pre-PN magnesium level of 0.80 mmol/L dropped to 0.54 mmol/L by day 5, but responded to supplementation with 5 mmol/d. Phosphate levels are summarized in Table I.

Patient 3

A 77-year-old woman was admitted to hospital with abdominal pain. She had a history of hypertension, peripheral vascular disease, ischemic heart disease, and chronic obstructive pulmonary disease. A laparotomy for an ischemic gut resulted in a right hemicolectomy. Her postoperative course was complicated by a prolonged ileus. PN was initiated in response to an inadequate oral intake for 2 weeks. An initial potassium of 4.1 mmol/L was supplemented with 40 to 80 mmol/d after PN was started. The pre-PN magnesium level of 0.81 mmol/L was maintained with PN supplementation of 5 mmol/d. Phosphate levels are summarized in Table I.

Patient 4

A 75-year-old man was admitted to hospital with an ischeorectal abscess. He had a history of diabetes mellitus, complicated by chronic renal failure and peripheral vascular disease. After surgical drainage of his abscess, he continued to be septic. PN was started because of a postoperative ileus associated with an inadequate oral intake of 10 days.

Serum electrolytes revealed an initial potassium level of 3.7 mmol/L, which dropped to 3.2 mmol/L after starting PN; the patient required 40 mmol/d of potassium via PN to maintain a normal serum potassium level. Serum magnesium level was initially 1.00 mmol/L. By PN day 7, the magnesium level dropped to 0.80 mmol/L, but no supplementation was given. Phosphate levels are summarized in Table I.


This report describes hypophosphatemia in four patients with chronic renal failure receiving PN. Patients with chronic renal failure have not previously been identified as being at high risk for developing hypophosphatemia when supported with PN. There are several possible mechanisms for the development of hypophosphatemia in these patients.

Serum phosphate levels depend on a balance among intake, renal excretion, and distribution between cells and extracellular fluid.3 The patients described in this article were fasting and thus had no source of intake other than what was provided parenterally. These patients did not have excess renal losses of phosphate due to medications such as diuretics or corticosteroids, renal tubular defects, or hyperaldosteronism. In fact, because of their chronic renal failure, they had decreased renal clearance of phosphate and thus were predisposed to hyperphosphatemia. Parenteral feeding does not alter the renal excretion of phosphate. Therefore, the major reason for the development of low phosphate levels in patients with chronic renal failure receiving PN is decreased intake in combination with shifts from the extracellular to intracellular compartments. Factors influencing this “shift” are discussed below.

Significant weight loss and loss of lean body mass are factors that increase the risk of developing refeedingassociated electrolyte disorders when nutritional repletion is begun. This phenomenon has been attributed to enhanced intracellular ion transport associated with carbohydrate-stimulated insulin release and cellular anabolism.l In our four cases, only one patient had a documented recent weight loss of 10%. Because of difficulty in obtaining dry weights and interpreting physical appearance, we suspect that preexisting malnutrition in the remaining three patients may have been masked by associated fluid retention and shifts in fluid balance. A high incidence of malnutrition has been reported in patients with chronic renal failure; thus, a high index of suspicion is indicated.45

The importance of carbohydrate load in causing electrolyte abnormalities was demonstrated by Weinsier and Krumdieck6 in their description of two patients who died as a result of severe hypophosphatemia and hypomagnesemia after they received in excess of 500 g of dextrose/24 h. The infusion of a dextrose load stimulates insulin release, resulting in increased intracellular transport of phosphate. Thus, it has been recommended that dextrose infusions be started gradually in the “at risk” patient, at a rate of approximately 2 mg/kg/min to meet the patient’s basal glucose needs.2 Although all of our patients received less than this amount of dextrose in the first 72 hours of PN, patients 1 and 4 exceeded 2 mg/ kg/min as a result of additional carbohydrate absorbed via their CAPD. The CAPD contributed an additional 150 and 130 g of dextrose through peritoneal dialysis in patients 1 and 4, respectively. CAPD dextrose must be considered when calculating the total carbohydrate load the patient is receiving.

Insulin therapy was required in two of the four patients. Because insulin promotes the intracellular transport of phosphate,7 its use may have influenced the development of hypophosphatemia in these patients. Alkalosis also may stimulate an intracellular shift in phosphate.3 None of the cases had overt evidence of a metabolic alkalosis with an elevated Tco2. In critically ill patients, respiratory alkalosis is frequent, but measurements of arterial blood gas levels were not done in our patients.

None of the four patients studied received supplemental phosphate (other than the obligatory phosphate found in lipid preparations) during the first 3 days of PN. Two of the four patients were hyperphosphatemic before starting PN. When phosphate levels dropped below normal between day 3 and 5 of starting PN, supplemental phosphate was started. Severe hypophosphatemia could have been prevented by initiating supplementation as levels moved downward.

Hypomagnesemia and hypokalemia also have been associated with PN and refeeding syndrome. Only one of our patients developed hypomagnesemia. Two of the patients were hypermagnesemic at baseline; although their serum levels decreased with PN, they did not develop below normal levels despite not receiving supplemental magnesium. The fourth patient received 5 mmol/ 24 h as supplemental magnesium with the PN and maintained a normal magnesium level. Three of the four patients were hypokalemic and required PN supplementation of potassium during their course of PN.

Fortunately, these patients did not develop any morbidities associated with their electrolyte abnormalities and did not develop fluid overload, cardiac arrhythmias, or respiratory failure, complications that have been associated with hypophosphatemia.8-10

What recommendations can we make on the basis of these cases? First, despite frequent elevations in serum phosphate at baseline, all patients with chronic renal failure who begin receiving PN should be considered to be at risk for developing significant electrolyte abnormalities, especially hypophosphatemia. Patients may develop these complications despite conservative initial dextrose infusions and high baseline serum electrolyte levels. During the first week of therapy, phosphate, magnesium, and potassium should be monitored judiciously and supplemented to the PN solution as the levels move downward, before the development of significant serum depletion.


Solomon SM, Kirby DF: The refeeding syndrome: A review. JPEN 14:9097, 1990

Apovian CM, McMahon MM, Bistrian BR: Guidelines for refeeding the marasmic patient. Crit Care Med 18:1030-1033, 1990 Knochel JP, Agarwel R: Hypophosphatemia and hyperphosphatemia. IN Brenner and Rectors The Kidney, Brenner BM (ed). WB Saunders Company, Philadelphia, 1996, pp 10861133

Markman P: Nutritional status of patients on hemodialysis and peritoneal dialysis. Clin Nephrol 29:7578, 1988 Young GA, Kopple JD, Lindholm B, et al: Nutritional assessment of continuous ambulatory peritoneal dialysis patients: An international study. Am J Kidney Dis 27:462471, 1991

Weinsier RL, Krumdieck, CL: Death resulting from overzealous total parenteral nutrition: The refeeding syndrome revisited. Am J Clin Nutr 34:393-399, 1981

Brooks MJ, Melnik G: The refeeding syndrome: An approach to under standing its complications and preventing its occurrence. Pharmacotherapy 15:713-726, 1995

Newman JH, Neff TA, Ziporin P: Acute respiratory failure associated with hypophosphatemia. N Engl J Med 296:1101-1103,1977 O’Connor LR, Klein KL, Bethune JE: Effect of hypophosphatemia on myocardial performance in man. N Engl J Med 297:901-903, 1977 Venditti FJ, Marotta C, Panezai FR, et al: Hypophosphatemia and cardiac arrhythmias. Miner Electrolyte Metab 13:19-25, 1987

From the Departments of *Medicine and ?Nursing, St Boniface General Hospital, University of Manitoba, Winnipeg, Canada

Received for publication, May 27, 1997. Accepted for publication, August 6, 1997. Correspondence: Donald R. Duerksen, MD, Department of Medicine, Division of Gastroenterology, C5120 409 Tache Avenue, St Boniface Hospital, Winnipeg, Manitoba R2H 2A6, Canada.

Copyright American Society for Parenteral and Enteral Nutrition Mar/Apr 1998

Provided by ProQuest Information and Learning Company. All rights Reserved