Sodium content of oral rehydration solutions: a reappraisal.

Proper choice of oral rehydration solution, with regard to sodium content, is a conflicting issue to general practitioners and pediatricians. World Health Organization (WHO) recommendations of oral rehydration solution containing 90mmol/1 sodium, have been effective throughout developing countries worldwide. In developed countries, however, such as England, this recommendation seems inappropriate; a recommendation of 50 - 60 mmol/1 sodium with 90 -111mmol/1 glucose is preferred. This combination will eliminate the need for free water recommended by the WHO maintenance therapy. Normonatraemia is maintained, and hyponatraemia and hypernatraemia can both be corrected. Sodium content is adequate in replacing stool loss resulting from viral and bacterial diarrheas. Iatrogenic hyponatraemia and hypernatraemia do not occur as they would with oral rehydration solutions with low (30-35) or high (90mmol/1) sodium concentration. Solutions containing 50-60 mmol/1 sodium is safer in neonates and young infants with immature renal functions, incapable of properly distributing increased sodium leads. Glucose concentration necessary to make oral rehydration with 50 - 60 mmol/1 isotonic, or hypotonic, resembles WHO - ORS, but is half that in 30-35 mmol/1 sodium solutions. In controlled clinical trials, oral rehydration solutions with 50-60 mmol/1 sodium have proven safe, and performance was compatible with WHO-ORS. For both rehydration and correction of acidosis, the solution proved effective. When oral rehydration solution sodium, concentrate is below 90 mmol/1, errors in reconstituting may be common, but offers less risk resulting in dangerous hypernatraemia. An "all purpose" physiological oral rehydration solution would make oral rehydration therapy, more economical, simpler, and safe in developed countries worldwide.     

Entgleisungen des Wasser- und Elektrolythaushaltes auf der Intensivstation

Management of critically ill patients regularly involves the treatment of water and electrolyte disturbances. Moreover, critical care itself may contribute to volume overload and electrolyte abnormalities. Initial therapy should be followed by consequent diagnostic evaluation. The shift of volume and potassium in severe pancreatitis, for example, may lead to a life-threatening situation. In brain-dead patients, successful organ donation is facilitated by careful maintenance of water and electrolyte homeostasis.     

Water and electrolyte disturbances in critical care

Management of critically ill patients regularly involves the treatment of water and electrolyte disturbances. Moreover, critical care itself may contribute to volume overload and electrolyte abnormalities. Initial therapy should be followed by consequent diagnostic evaluation. The shift of volume and potassium in severe pancreatitis, for example, may lead to a life-threatening situation. In brain-dead patients, successful organ donation is facilitated by careful maintenance of water and electrolyte homeostasis.     

Evaluation of the efficacy of oral rehydration solutions using human whole gut perfusion.

Whole gut perfusion in humans was used to compare the effect on intestinal water and electrolyte transport of the World Health Organisation oral rehydration solution (solution II, composition in mmol/l: glucose 111, sodium 90, bicarbonate 30, potassium 20; 308 mOsm/kg); a hypertonic commercial oral rehydration solution (solution III, glucose 188, sodium 50, bicarbonate 20, potassium 20 mmol/l; 335 mOsm/kg); and three experimental bicarbonate free, hypotonic oral rehydration solutions: solution IV (glucose 111, sodium 60, potassium 20 mmol/l; 260 mOsm/kg), solution V (glucose 80, sodium 60, potassium 20 mmol/l; 219 mOsm/kg), and solution VI (glucose 80, sodium 30, potassium 20 mmol/l; 177 mOsm/kg). Perfusion of the intestine with a standard cleansing solution (solution I, sodium 125, potassium 10, bicarbonate 20, sulphate 40, mannitol 80 mmol/l; 275 mOsm/kg) confirmed published data on minimal water and sodium absorption. Experimental solution VI produced maximum water absorption (mean (SE) +1660.0 (29.8) ml/h) significantly greater than solution II (+1195.3 (79.5) ml/h), III (+534.7 (140.3) ml/h), IV (+1498.0 (42.7) ml/h), and V (+1327.7 (24.4) ml/h; p less than 0.05). Sodium absorption was significantly greater with solution II (+97.4 (7.9) mmol/h) compared to VI (+43.3 (7.8) mmol/h; p less than 0.01) but not compared to IV (+67.2 (13.0) mmol/h). A hypotonic oral rehydration solution such as solution VI may provide optimal replacement treatment for patients with acute diarrhoea.     

Management of Lithuanian children's acute diarrhoea with Gastrolit solution and dioctahedral smectite

OBJECTIVE: Acute gastroenteritis represents a major cause of morbidity and mortality worldwide among children, and rehydration treatment has been one of the cornerstones in the management strategy. The natural clay dioctahedral smectite (Smecta) increases intestinal barrier function and is effective against infectious diarrhoea in children. The purpose of this work was to compare the efficacy and tolerance of Lithuanian children's diarrhoea treatment with dioctahedral smectite combined with hypotonic oral rehydration solution (ORS)--Gastrolit--versus Gastrolit alone to establish the influence of Smecta on serum electrolyte balance in young children with diarrhoea and mild or moderate dehydration. METHODS: Smecta combined with ORS (study group) and ORS alone (control group) were evaluated in a multicentre, open, randomized trial in 54 children aged 6-48 months hospitalized for acute diarrhoea (mostly rotavirus aetiology) and signs of mild and moderate dehydration. The main outcomes examined were duration of diarrhoea, fever, number of vomiting episodes, and serum electrolyte balance before and after treatment. RESULTS: The mean duration of diarrhoea was significantly shorter in the study group (42.3 +/- 24.7 h) than in the control group (61.8 +/- 33.9 h). No side effects of Smecta were observed. The changes of sodium, potassium, chloride and calcium concentrations after treatment were minimal and in the normal range. CONCLUSIONS: Smecta significantly reduced the duration of diarrhoea, was safe and well tolerated, and had no impact on the adsorption of electrolytes. Smecta could be used together with ORS in children suffering from acute gastroenteritis (without uncontrollable vomiting) with mild and moderate dehydration.     

Oral rehydration solution safely used in breast-fed children without additional water

Sixty-four children of age 3 months to 2 years were treated for diarrhoea in a rural treatment centre with oral rehydration solution (90 mmol/l Na+); plain water was not supplied during rehydration. All children continued breast-feeding during the therapy. Serum sodium levels after 24 h of intake indicated that 61 children had serum sodium level below 148 mmol/l and three were above (range 150-151). None of the children developed clinical signs or symptoms of hypernatremia. The use of additional water may not be necessary during rehydration therapy of breast-fed children.     

Prednisolone in the treatment of severe malignant hypercalcaemia in metastatic breast cancer: a randomized study.

To determine the effect of prednisolone on severe hypercalcaemia in women with metastatic breast cancer, 30 patients with serum ionized calcium above 1.60 mmol l-1 (reference range 1.15-1.35 mmol l-1) entered a randomized trial. Performance status before entry to the trial and survival time after hypercalcaemia were also noted. All patients received 4 l of isotonic saline daily and 80 mg intravenous furosemide three times daily for 2 d; thereafter they received 3 l of isotonic saline daily and 80 mg furosemide twice daily for 6 d. Fifteen patients were randomized to receive prednisolone, 25 mg orally, three times daily for 8 d. Serum ionized calcium decreased significantly in both groups, but most markedly in the prednisolone group. The median difference was 0.28 mmol l-1 (95% confidence interval (CI), 0.09-0.52) on day 4 and 0.21 mmol l-1 (95% CI, 0.12-0.44) on day 8. In seven prednisolone-treated patients serum ionized calcium normalized, compared to none in the control group (Fisher's exact test; P = 0.028). No severe adverse effects were observed. Prior to detection of hypercalcaemia all patients were severely immobilized, primarily due to bone pain. Only 10 patients were still living after 3 months. Prednisolone, furosemide and rehydration is superior to furosemide and rehydration alone in severely hypercalcaemic patients with metastatic breast cancer in whom immobilization appears to be an early warning sign of life-threatening hypercalcaemia. The short survival time was not influenced by prednisolone.     

Oral rehydration therapy: efficacy of sodium citrate equals to sodium bicarbonate for correction of acidosis in diarrhoea.

Forty patients with moderate degrees of dehydration and acidosis because of acute watery diarrhoea were successfully treated randomly with either WHO recommended oral rehydration solution containing 2.5 g sodium bicarbonate or an oral solution containing 2.94 g sodium citrate in place of sodium bicarbonate per litre of oral rehydration rehydration solution. Efficacies were compared by measuring oral fluid intake, stool and vomitus output, change in body weight, hydration status, and rate of correction of acidosis during a period of 48 hours. Seventy five per cent (21 cases) in the citrate group and 83% (19 cases) in the bicarbonate group were successfully rehydrated (p greater than 0.05). There were no significant differences in intake, output, gain in body weight, fall in haematocrit and plasma specific gravity, and correction of acidosis between the two groups of patients within 48 hours after initiation of therapy. The solution with sodium citrate base was as effective as WHO-oral rehydration solution for management of diarrhoea. This study shows the efficacy, safety, and acceptability of citrate containing oral rehydration solution for rehydration and correction of acidosis in diarrhoea.     

Gum arabic promotes rat jejunal sodium and water absorption from oral rehydration solutions in two models of diarrhea

BACKGROUND & AIMS: We have shown that addition of gum arabic (GA) to a 90 mmol/L sodium-111 mmol/L glucose oral rehydration solution (ORS) enhances its effectiveness for water and electrolyte absorption in normal rats. The present study extends these observations on GA in ORS to two rat models of diarrheal disease. METHODS: Juvenile rats were either treated for 1 week with magnesium citrate-phenolphthalein to produce chronic osmotic-secretory diarrhea or luminally exposed to 10 mmol/L theophylline to induce jejunal secretion. In both models jejunal perfusion was used to assess absorption. RESULTS: Addition of 2.5 or 5.0 g/L GA to ORS increased roughly twofold absorption of sodium, potassium, and water in the model of chronic osmotic-secretory diarrhea. Rats perfused with GA-supplemented ORS showed an expansion of the basolateral intercellular spaces between villus absorptive epithelial cells and the lamina propria, reflecting enhanced water and sodium absorption. Similarly, addition of 2.5, 5.0, or 10.0 g/L GA to the ORS neutralized theophylline-induced abolition of net sodium and potassium absorption and reversed water and glucose malabsorption. CONCLUSIONS: These experimental studies in models of diarrhea suggest that GA may be a useful additive to ORS for the potentiation of water and electrolyte absorption. (Gastroenterology 1997 Jun;112(6):1979-85)     

Cholera

Intestinal infection with Vibrio cholerae results in the loss of large volumes of watery stool, leading to severe and rapidly progressing dehydration and shock. Without adequate and appropriate rehydration therapy, severe cholera kills about half of affected individuals. Cholera toxin, a potent stimulator of adenylate cyclase, causes the intestine to secrete watery fluid rich in sodium, bicarbonate, and potassium, in volumes far exceeding the intestinal absorptive capacity. Cholera has spread from the Indian subcontinent where it is endemic to involve nearly the whole world seven times during the past 185 years. V cholerae serogroup O1, biotype El Tor, has moved from Asia to cause pandemic disease in Africa and South America during the past 35 years. A new serogroup, O139, appeared in south Asia in 1992, has become endemic there, and threatens to start the next pandemic. Research on case management of cholera led to the development of rehydration therapy for dehydrating diarrhoea in general, including the proper use of intravenous and oral rehydration solutions. Appropriate case management has reduced deaths from diarrhoeal disease by an estimated 3 million per year compared with 20 years ago. Vaccination was thought to have no role for cholera, but new oral vaccines are showing great promise.     

LACK OF EFFECT OF ALDOSTERONE ON INTRACELLULAR SODIUM AND POTASSIUM IN MONONUCLEAR LEUCOCYTES FROM PATIENTS WITH PSEUDOHYPOALDOSTERONISM

In three patients with pseudohypoaldosteronism the effects of aldosterone on intracellular sodium and potassium were studied and compared with normal controls in whom aldosterone prevents the loss of sodium and potassium in vitro. Mononuclear leukocytes were incubated with or without aldosterone (1.4 nM) in RPMI-1640 for 1 h at 37 degrees C. After two washes in isotonic MgCl2 the wet cell pellets were weighted and intracellular sodium and potassium determined by flame photometry, results are expressed as mmol/kg wet cells. In the patients intracellular sodium fell from 18, 23 and 29 mmol/kg to 14, 18 and 11 mmol/kg, respectively, in the absence of aldosterone. With aldosterone added to the incubation medium sodium was not different from values obtained without aldosterone (15, 20 and 13 mmol/kg). Corresponding values for potassium were 89, 48 and 75 mmol/kg before and 68, 32 and 51 mmol/kg after incubation without and 69, 36 and 54 mmol/kg after incubation with aldosterone. Thus, incubation with aldosterone did not show an effect on intracellular sodium and potassium as seen in normals. Baseline values of sodium and potassium before the incubation were within the normal range. From these results it is concluded that in patients with pseudohypoaldosteronism the absent or decreased number of mineralocorticoid receptors in mononuclear leucocytes are accompanied by a lack of response of intracellular sodium and potassium to aldosterone in vitro. However, normal baseline intracellular electrolyte concentrations in these patients may indicate that mineralocorticoids are not involved in the maintenance of normal levels of intracellular sodium and potassium.     

Severe normotensive metabolic alkalosis in a 2-month-old boy with hyperekplexia

Summary A 2-month-old infant with hereditary hyperekplexia, umbilical and bilateral inguinal hernias and history of poor feeding was noted to have severe normotensive metabolic alkalosis: sodium 132 mmol/L, potassium 3.4 mmol/L, chloride 77 mmol/L, pH 7.55, carbon dioxide tension 56.3 mmHg and bicarbonate 48.0 mmol/L. After parenteral rehydration and treatment with clonazepam, laboratory parameters normalized. Communicating editor: Georg Hoffmann Competing interests: None declared     

Assessment of the effectiveness of oral rehydration therapy against severe diarrheal dehydration.

A hospital based case-control study for assessing the effectiveness of oral rehydration therapy (ORT) preparation against severe dehydration due to diarrhea was conducted at the Infectious Diseases Hospital, Jakarta, Indonesia. A total of 202 children aged 24 months or less who attending the hospital were suffering from acute watery diarrhea were recruited in the study. Those who were severely dehydrated as assessed by WHO criteria were accounted as cases; those who were non-severely dehydrated were accounted as controls. There were 59 cases and 143 controls. A questionnaire was used to interview all study subjects' mothers about ORT usage and various risk factors. Mothers who used ORT were asked to show how they prepared either oral rehydration solution (ORS) or sugar salt solution (SSS). Effectiveness of ORT against severe diarrheal dehydration was based on the formula for assessment of vaccine efficacy by using the odds ratio (OR). With the use of the logistic regression method, an adjusted OR was obtained after controlling various confounders. The effectiveness of ORT against severe diarrheal dehydration was 72.1% for proper ORT preparation and was decreased to 63.2% when ORT was improperly prepared.     

Electrolyte disorders following oral sodium phosphate administration for bowel cleansing in elderly patients

BACKGROUND: Oral sodium phosphate is currently used for colon preparation prior to colonoscopy or barium enema. Sodium phosphate induces hyperphosphatemia, hypocalcemia, and hypokalemia. Elderly patients are at an increased risk for phosphate intoxication due to decreased glomerular filtration rate, medication use, and systemic and gastrointestinal diseases. We investigated these electrolyte disorders and their correlation with creatinine clearance, coexistent diseases, medications, and functional status. METHODS: Thirty-six hospitalized patients were included in the study. On day 1, patients were administered 2 doses of oral sodium phosphate. Venous blood samples for electrolyte determination were obtained at 7 AM on days 1, 2 (the procedure day), and 3. Urine samples were obtained from 10 patients. RESULTS: An increase in serum phosphorus level was correlated with a decreased creatinine clearance (R = -0.52; P =.001). Hypocalcemia and hypokalemia were present in 21 (58%) and 20 (56%) patients, respectively. Patients with a serum potassium concentration of 3.5 mEq/L or less on day 2 had a lower serum potassium concentration on day 1 vs those with a serum potassium concentration greater than 3.5 mEq/L on day 2 (P =.03). Five (dependent patients) had a serum potassium concentration of 3 mEq/L or less and 2 had severe diarrhea, necessitating treatment. There were more demented patients with hypokalemia compared with normokalemic patients (P<.05). Urinary fractional excretion of phosphorus tripled on day 2 (P =.01). Potassium and sodium fractional excretion remained unchanged. CONCLUSIONS: Sodium phosphate induces serious electrolyte abnormalities in the elderly. The frequency and severity of hypokalemia is due to intestinal potassium loss associated with inadequate renal potassium conservation and is apparently more prevalent in frail patients. Assessment of serum electrolytes, phosphorus, and calcium prior to sodium phosphate preparation is advised, and in selected patients, postprocedural assessment and correction may be required.     

Electrolyte disturbances in patients with chronic, stable asthma: effect of therapy

OBJECTIVE: To determine the prevalence of electrolyte disturbances in patients with chronic, stable asthma, and to assess whether the therapeutic agents used to treat chronic asthma have an effect on abnormal electrolyte levels. DESIGN: Prospective, hospital-based, cross-sectional study. SETTING: University teaching hospital in Jeddah, Saudi Arabia. PATIENTS: Patients with chronic, stable asthma. METHOD: Ninety-three consecutive patients with chronic, stable asthma were involved in the study. On the day of the visit to the asthma clinic, particulars such as age, sex, duration of asthma, and details of drug therapy were obtained from each asthmatic patient. Serum potassium, magnesium, phosphorus, calcium, and sodium levels were measured. Normal values were as follows: potassium, 3.5 to 5 mmol/L; magnesium, 0.74 to 1.2 mmol/L; phosphorus, 0.8 to 1.4 mmol/L; and calcium, 2.1 to 2.6 mmol/L. RESULTS: Electrolyte disturbances were found in 43% of the patients; 85% of the patients had one electrolyte disturbance, 10% had two electrolyte disturbances, and 5% had three electrolyte disturbances. The highest proportions were for magnesium (26.9%) and phosphorus (15.1%) [serum levels were 0.69 +/- 0.04 mmol/L and 0.64 +/- 0.09 mmol/L, respectively], the lowest proportions were for potassium (5.4%) and sodium (4.3%) [serum levels were 3.3 +/- 0.01 mmol/L and 133 +/- 0.01 mmol/L, respectively], and no patient had a calcium disturbance. Logistic regression analysis showed no statistically significant association between the therapy used and electrolyte disturbances. CONCLUSION: Hypomagnesemia and hypophosphatemia were found to be the two most common electrolyte disturbances in patients with chronic, stable asthma. Therapeutic agents used to treat patients with chronic asthma have no effect on abnormal electrolyte levels. The underlying cause still remains unclear.     

Diabetic ketoacidosis. Induction of hypocalcemia and hypomagnesemia by phosphate therapy.

A nine year old boy with previously undiagnosed diabetes mellitus presented with severe ketoacidosis. His hyperglycemia (plasma glucose = 786 mg/dl), acidosis (arterial pH = 6.86), dehydration and coma responded well to therapy with intravenous fluids, bicarbonate and insulin. Potassium supplementation was given as a phosphate salt.Despite marked clinical and biochemical improvement, 28 hours after therapy was initiated he was found to have profound hypocalcemia (2.6 meq/liter), hypomagnesemia (0.8 meq/liter) and hyperphosphatemia (9.2 mg/dl). All three electrolyte levels had been normal upon admission, and they were gradually corrected with appropriate supplementation of calcium and magnesium and discontinuation of the intravenous phosphate. We interpret these iatrogenic electrolyte abnormalities in the patient described to have been the result of the massive phosphate load administered, resulting not only in hypocalcemia, but also in hypomagnesemia that inhibited parathyroid hormone release.Current recommendations suggest replacement of the potassium losses in diabetic ketoacidosis with the phosphate salt to compensate for depleted stores of 2,3-diphosphoglycerate. We caution physicians that such a regimen can result in severe electrolyte disturbances which potentially may be life threatening. Judicious use of potassium phosphate as an adjunct to traditional potassium chloride therapy, and close monitoring of serum calcium, magnesium and phosphorus, appears to be a preferable therapeutic regimen than potassium phosphate alone.     

Absorption of a hypotonic oral rehydration solution in a human model of cholera.

The development of oral rehydration solutions (ORSs) has been one of the important therapeutic advances of this century. The optimal formulation, however, of ORSs for both cholera and other infective diarrhoeas is still debated. Part of the problem in developing ORSs has been the lack of adequate test systems for the assessment of new formulations before clinical trial. We have developed a jejunal perfusion, cholera toxin induced, secretory model in humans and have compared net water and solute absorption from a hypotonic ORS (HYPO-ORS: sodium 60 mmol/l, glucose 90 mmol/l, osmolality 240 mOsm/kg) and the British Pharmacopoeia recommended ORS (UK-ORS: sodium 35 mmol/l, glucose 200 mmol/l, osmolality 310 mOsm/kg) in six healthy volunteers. A plasma electrolyte solution (PES) was also perfused in all subjects to confirm a secretory state. Only HYPO-ORS reversed sodium secretion to absorption (p < 0.01). Both ORSs promoted net water absorption but this was greatest with HYPO-ORS (p < 0.01). Glucose and potassium absorption rates were similar for both ORSs whereas chloride absorption mirrored sodium absorption and was greatest from HYPO-ORS (p < 0.05). These results, in a biologically relevant model of secretory diarrhoea, suggest it may be possible to achieve improved rates of rehydration by the use of hypotonic ORS with mid range sodium concentrations.     

Fluid and electrolyte absorption and renin-angiotensin-aldosterone axis in patients with severe short-bowel syndrome.

In eight patients who had received long-term parenteral nutrition because of short-bowel syndrome the need for parenteral supply of fluid, sodium, and potassium was estimated by balance studies. Six patients had jejunostomies. In two, most of the colon was preserved. Jejunostomy patients had a huge stool mass (1710--5270 g, median 2530 g/day) with fixed concentrations of sodium (92 +/- 10 mmol/l) and potassium (15 +/- 4 mmol/l). In contrast, two patients with massive small-bowel resection but with more than half of the colon intact showed almost normal sodium absorption and considerably smaller stool mass (170--510 g/day). Despite apparently good health and normal plasma electrolytes, urea, and haematocrit, four of six jejunostomy patients were sodium-depleted with low plasma volume, low sodium excretion in the urine, and increased plasma renin activity and, in the three most severe cases, increased aldosterone. Even in case of sodium depletion the sodium loss from jejunostomy effluents remained high and presumably unaffected by salt-retaining hormones. The study confirms the importance of preservation of part of the colon for maintenance of fluid and electrolyte balance in patients with extensive bowel resection. Jejunostomy patients who are eating normally may need large parenteral saline supply. Assessment of water and electrolyte homeostasis in these patients requires determination of the urinary sodium excretion and is supported by measurements of plasma renin activity and plasma aldosterone concentration.     

Rice-powder salt solution in the treatment of acute diarrhea in young children.

Dehydration is the most common cause of death in diarrheal patients. Early oral rehydration therapy (ORT) can prevent or reverse dehydration from diarrhea in almost almost all cases. Shortages of oral rehydration salt (ORS) packets in certain areas remain a major problem of the Diarrheal Diseases Control Program of Thailand. To find an effective solution that can be prepared locally, a randomized trial of oral rehydration solutions was conducted. A rice-powder salt solution containing rice-power 30 g/l and salt 3.5 g/l (RPSS) was evaluated in a group (n = 23) of infants and young children aged between 4 months and 5 years with mild or moderate dehydration from acute watery diarrhea, and the results were compared with those who received WHO recommended glucose electrolyte solution (WHO-ORS) (n = 21), and glycine supplemented WHO-ORS (G-ORS) (n = 20). The efficacies of WHO ORS and G-ORS were found to be similar. The RPSS was found to be more effective than WHO-ORS and G-ORS as shown by a significantly lower stool frequency, lower rate of stool output, a significantly shorter duration of diarrhea, and a smaller intake of rehydration fluid. Promotion of the effective rice-salt solution could increase early implementation of ORT in many rural communities.     

Bowel preparation for colonoscopy in very old patients: a randomized prospective trial comparing oral sodium phosphate and polyethylene glycol electrolyte lavage solution

BACKGROUND: Colonoscopic preparations are defectively tolerated among many very old patients and the quality of preparation is often inadequate. METHODS: Seventy-two patients aged 80 years or more were randomized to receive either NaP (37) or PEG (35) preparation. Patients' and endoscopists' opinions of the preparation and colonoscopy were ascertained, the adverse effects were recorded, clinical indicators of dehydration such as tongue dryness, upper body muscle weakness and confusion were assessed and laboratory tests were taken before and after the preparation. RESULTS: Patients' evaluations showed no significant difference regarding the tolerance of preparations. Only the frequency of nausea was higher in the NaP group (16%/9%, P = 0.01). Assessment of the indicators of dehydration showed a difference only in tongue dryness in the NaP group (P = 0.02). The quality of preparation was evaluated by endoscopists as good or excellent in 81% of cases in the NaP group and in 77% of cases in the PEG group. The median values of potassium decreased from 4.0 mmol/L to 3.7 mmol/L (P < 0.05) and of sodium increased from 142 mmol/L to 145 mmol/L (P < 0.05) after NaP preparation. CONCLUSION: Sodium phosphate and PEG preparations are almost equally tolerated and effective in very old inpatients. NaP preparations cause more changes in the levels of potassium and sodium. For those elderly patients who are vulnerable to complications caused by electrolyte disturbances we recommend PEG preparation.