Increased serum phosphate levels and calcium ... - Wiley Online Library

Alimentary Pharmacology & Therapeutics

Increased serum phosphate levels and calcium fluxes are seen in smaller individuals after a single dose of sodium phosphate colon cleansing solution: a pharmacokinetic analysis E. D. EHRE NPREIS

Department of Gastroenterology, Highland Park Hospital, NorthShore University HealthSystem, Highland Park, IL and Department of Medicine, University of Chicago, Chicago, IL, USA Correspondence to: Dr E. D. Ehrenpreis, Chief, Department of Gastroenterology, Highland Park Hospital, NorthShore University HealthSystem, 757 Park Avenue West, RM 3462, Highland Park, Illinois 60035, IL, USA. E-mail: [email protected]

Publication data Submitted 26 January 2009 First decision 8 February 2009 Resubmitted 18 February 2009, 24 February 2009 Accepted 24 February 2009 Epub Accepted Article 27 February 2009

SUMMARY Background Sodium phosphate containing colonoscopy preparations may cause electrolyte disturbances and calcium-phosphate nephropathy. Decreased body weight is an unexplored risk factor for complications with sodium phosphate ingestion. Aim To perform a pharmacokinetic analysis of a single dose of Fleet PhosphoSoda in smaller and larger individuals. Methods Seven subjects weighing 100 kg (Group II) consumed 45 mL Fleet Phospho-Soda. Serum electrolytes were measured. Hydration was closely maintained by monitoring weight, fluid intake and total body water. Results Marked increases in serum phosphate were seen in Group I compared to Group II. For example, mean serum phosphate at 120 min was 7.8 0.5 mg ⁄ dL in Group I and 5.1 0.8 mg ⁄ dL in Group II (P < 0.001). Normalized area under the phosphate vs. time curve for Group I was 1120 190 mg ⁄ dL*min and 685 136 mg ⁄ dL*min for Group II (P < 0.001). Twelve-hour urine calcium was lower in Group I (16.4 7.6 mg) than in Group II (39.2 7.8 mg, P < 0.001). Conclusions Increased serum phosphate occurs in smaller individuals after ingestion of sodium phosphate preparations, even with strict attention to fluid intake. Smaller body weight poses a potential risk for calciumphosphate nephropathy. Aliment Pharmacol Ther 29, 1202–1211

1202

ª 2009 Blackwell Publishing Ltd doi:10.1111/j.1365-2036.2009.03987.x

S O D I U M P H O S P H A T E : R I S K I N S M A L L E R P A T I E N T S 1203

INTRODUCTION Cleansing of the colon is required to evaluate fully for the presence of colon polyps or other lesions. A variety of cleansing methods are available as preparations for colonoscopy.1, 2 Most involve the oral administration of a potent osmotic laxative as well as additional fluids to replace fluid and electrolyte losses that occur from cathartic effects. The sodium phosphate- containing laxative, Fleet Phospho-Soda has been demonstrated to improve compliance, produce a higher preparation quality and to have fewer preparation-related side effects.3, 4 In addition, the ‘ease of use’ of Fleet Phospho-Soda was shown to be better than PEG-based preparations in 18 out of 24 comparative clinical trials.5 More than 7 million doses of Fleet Phospho-Soda preparation were sold as an overthe-counter product worldwide in 2004. Sodium phosphate colonoscopy preparations are also available in tablet form. The most important advantage of sodium phosphate-based colonoscopy preparations is the small volume of laxative consumed. Recommended dosage for the Fleet Phospho-Soda colonoscopy preparation is just two 45 mL doses, or one 45 mL and one 30 mL dose. The first dose is taken on the night prior to the procedure and the second dose is taken on the day of the procedure. The quantity of sodium phosphate provided by these regimens is either 59.4 or 48 g respectively. Despite being suggested by some as the preparation of choice for colon cleansing prior to colonoscopy, reports of electrolyte disturbances secondary to Fleet’s Phospho-soda consumption rapidly appeared in the medical literature when these preparations came into common usage.6, 7 In 2004, the first case report of calcium-phosphate nephropathy caused by Fleet’s Phospho-soda was described8 and this was followed by a larger case series reporting this phenomenon.9 Despite these reports, a number of publications stressed the rarity of calcium-phosphate nephropathy and continued to recommend sodium phosphate colonoscopy preparations.10–12 In December 2008, the CB Fleet Corporation voluntarily removed Fleet Phospho-soda from the pharmaceutical market after the FDA ruled that the product should be consumed for colonoscopy bowel cleansing only by prescription. Several presumed risk factors for calcium phosphate nephropathy and subsequent renal failure from Fleet’s Phospho-soda were suggested. These included Aliment Pharmacol Ther 29, 1202–1211 ª 2009 Blackwell Publishing Ltd

underlying renal disease, medications such as angiotensin converting enzyme (ACE)-inhibitors, angiotensin receptor blockers (ARB) and diuretics as well as longstanding hypertension.13, 14 Our group has proposed that administration of large doses of sodium phosphate to subjects with diminished body weight might increase their risk for the development of electrolyte disturbances or calcium phosphate nephropathy following sodium phosphate based colonoscopy preparations.15 Prior to the removal of Fleet Phospho-soda from the market, we conducted a formal pharmacokinetic analysis of this hypothesis. Our study demonstrates that under carefully controlled experimental conditions, smaller individuals have higher serum phosphate levels following ingestion of large doses of sodium phosphate and that phosphate elevation in this patient group produces more pronounced physiological effects.

METHODS Inclusion criteria Thirteen healthy adult male and female volunteers were recruited for the study. These consisted of two groups stratified by body weight. To be included in Group I, subjects reported a body weight 2 weeks before the study (without clothing) of 55 kg (132 lb) or less. Group II had a reported body weight without clothing of greater than or equal to 100 kg (220 lb) and a BMI less than 35 kg ⁄ m2. All subjects were between the ages of 18 and 58 years. Subjects were in good health with no history of cardiac, liver or kidney disease. None was taking antihypertensive medications. None reported any of the contraindications that are listed in the Fleet’s Phospho-Soda prescribing information, including congenital megacolon, bowel obstruction, ascites, pre-existing electrolyte disturbances, renal insufficiency or general debility.

Study method All subjects took nothing by mouth after midnight prior to the study. Subjects were weighed wearing street clothing and without shoes on study arrival. Baseline body compartment analysis using Near-infrared photospectrometry (NIR) at the midbiceps position was performed using the Futrex 5500 (Futrex Inc, Hagerstown, MD). Subjects discarded their first voided

1204 E . D . E H R E N P R E I S

urine specimen. A saline lock was placed into the antecubital fossa. Baseline analysis of serum for concentrations of sodium, potassium, chloride, bicarbonate (HCO2), glucose, blood urea nitrogen (BUN) and creatinine was performed using standard methods. These were termed the basic metabolic panel (BMP). In addition, baseline calcium (Ca), phosphate (PO4) and haematocrit (HCT) were measured. Fleet’s Phospho-Soda 45 mL followed by five 250 mL glasses of clear liquid (Sprite) was consumed over 1–5 min. This is 50% or 60% of the total Phospho-soda dose typically used for bowel preparation. Subjects ingested at least 250 mL of clear liquids (water, Sprite, Diet Sprite) with intake and were told to ingest at least 250 mL of fluid hourly for 8 h after ingestion. Measurement of fluid intake and encouragement by study coordinators based on hourly ingestion was performed throughout the study. Serum was collected at 15, 30, 45, 60, 90, 120, 240, 300, 360 and 480 min after ingestion of Fleet’s Phospho-Soda. All urine produced was collected and pooled for 720 min after ingestion. Measurement of serum calcium and phosphate was performed with each draw. Serum BMP and ionized calcium was measured at 60, 120, 240, 360 and 480 min after ingestion. Serum magnesium was measured at 120, 240, 360 and 480 min after ingestion. Urine was analysed for volume, as well as calcium, phosphate, and creatinine concentrations. Subjects were weighed and body composition studies were repeated at 60, 120, 360 and 480 min. Weights were rounded to the nearest 0.5 kg. Per study protocol, any loss of weight or decreased total body water was to be brought to the attention of the study subject when it was noted, at which time further encouragement of fluid intake was performed by the study personnel. Subjects were fed a light breakfast at 180 min and lunch 240 min after ingestion of Fleet’s PhosphoSoda.

Body composition analysis The Futrex devise has an optical generating light wand that produces and detects near infrared energy (NIR). The method relies on the unique absorption, reflectance and transmission properties of near infrared light that is conveyed into body tissues. Each tissue produces a specific signature that can be quantified using this technique. Changes in the light spectra that are detected by the device after transmission reflect body chemistry also known as body composition.

Percent body fat, lean body mass and total body water are then calculated using algorithms based on spectrophotometric curves built into the device and produced as a digital readout. Detailed description of this methodology has been described elsewhere.16, 17 NIR has been favourably compared to other forms of estimates of body composition including skin fold thickness, deuterium oxide (D2O), impedance plethysmography and underwater weighing.18, 19

Data analysis Estimated creatinine clearance was performed using the Cockcroft and Gault equation as follows: Creatinine clearance = (140)age) · IBW ⁄ (Serum creatinine · 72) (·0.85 for females). Ideal body weight in kg was estimated as follows: Males: IBW = 50 kg + 2.3 kg for each inch over 5 feet. Females: IBW = 45.5 kg + 2.3 kg for each inch over 5 feet.20 Body mass index (BMI) was defined as weight (kg) divided by height (metre squared). Individual serum levels of the measured electrolytes, Hct, weight, and serum phosphate and calcium excretion were compared between the two groups using two-tailed Student’s t tests. Mean areas under the serum phosphate vs. time curves were also compared using two-tailed Student t tests. Normalized areas under the serum phosphate vs. time curves were calculated by subtracting baseline phosphate levels from additional phosphate levels and also compared by two-tailed Student t tests. Comparison of nonparametric parameters between the two groups including age, weight, height, lean weight, BMI, body fat and body water were performed using the Mann–Whitney test. These data were reported as median values with 95% confidence intervals. All study subjects were provided with an approved informed consent document and signed this prior to initiation of the study. This study was approved by the Institutional Review Board of Northwest Community Hospital, Arlington Heights, Illinois.

RESULTS Thirteen subjects participated in the study, seven females in Group I and five males and one female in Group II. Characteristics of the study subjects including initial body compartment analysis are shown in Table 1. Significant differences between the groups Aliment Pharmacol Ther 29, 1202–1211 ª 2009 Blackwell Publishing Ltd


Table 1. Characteristics and body composition analysis of study subjects Lean weight (kg)

Body mass index Kg ⁄ m2

Body fat (%)

Body water (L)

59.1 60.9 55.9 58.2 56.8

37.9 37.3 37.1 45.5 38.1

23.9 26.3 22.6 22.7 24.6

29.7 29.5 30.0 34.3 293

155.0 157.5 157.5 (153.4,158.6)

50.9 64.1 60.0 (54.2, 61.8)

34.5 42.4 40.0 (35.5,42.4)

21.1 26.0 23.9 (22.1,25.6)

35.9 38.7 33.6 21.9 3 3.1 32.3 33.9 32.8 (27.9,37.6)

195.6 180.6 180.3 180.3 190.5 183.0 181.8 (178.3,191.3) P < 0.005

111.8 112.7 138.2 115.9 108.6 119.2 119.2 (107.2,131.3) P < 0.0001

88.4 81.8 85.7 75.2 79.6 74.7 80.9 (75.1,86.7) P < 0.005

29.1 34.0 42.6 35.7 30.0 38.2 34.9 (29.6,40.3) P < 0.005

20.8 27.4 38.0 35.1 26.7 41.7 31.6 (23.3,39.9) N.S.

66.6 62.2 65.9 57.4 59.6 58.1 61.6 (57.5,65.8) P < 0.005

Subject

Gender

Age (years)

Height (cm)

Group I 1 2 3 4 5

F F F F F

58 24 34 49 23

157.5 152.4 157.5 160.0 152.4

F F

23 39 38 (23,49)

M M M M M F

26 37 30 44 24 25 31 (23,39) P = N.S.

6 7 Median (95% CI) Group II 1 2 3 4 5 6 Median (95% CI)

Baseline Weight (kg)

were seen in height, weight, lean body weight and total body water, as expected based on selection criteria. The BMI was significantly higher in Group II as well. Estimated creatinine clearance was significantly higher in Group II, but calculated 12 h creatinine clearance was not statistically different between the two groups (see Table 3). A rapid rise in serum phosphate levels was seen following ingestion of Fleet’s Phospho-Soda. Most subjects had increased serum phosphate levels within 15 min of ingestion. Peak phosphate levels occurred between 1 and 3 h after ingestion, with earlier peaks and more rapid initial clearance seen in Group II (see Table 2). Figure 1 shows a typical serum phosphate concentration vs. time curve for a patient from Group I and Group II. Marked differences in serum phosphate and normalized serum phosphate levels were seen between Group I and Group II, beginning 120 min after ingestion of Fleet’s Phospho-Soda and persisting for 240 min after ingestion. Mean serum phosphate and normalized serum phosphate at 60 min were similar in the two groups. Serum phosphate at 60 min was 6.5 0.6 mg ⁄ dL in Group I and 5.8 0.8 mg ⁄ dL in Group II, (P = 0.17). Mean serum phosphate at 120 min was Aliment Pharmacol Ther 29, 1202–1211 ª 2009 Blackwell Publishing Ltd

26.5 33.1 30.3 (27.9,32.7)

7.8 0.5mg ⁄ dL in Group I and 5.1 0.8 mg ⁄ dL in Group II (P < 0.001). Mean serum phosphate at 180 min was 6.5 0.6 in Group I and 4.9 0.8 in Group II (P < 0.0005). Mean serum phosphate at 240 min was 6.2 0.4 mg ⁄ dL in Group I and 5.0 0.5 mg ⁄ dL in Group II (P < 0.001), while normalized mean serum phosphate was 2.6 0.4 in Group I and 1.3 0.3 mg ⁄ dL in Group II (P < 0.0001). Serum phosphate levels remained elevated in both groups at 480 min after ingestion, but neither phosphate nor normalized phosphate was significantly different between the two groups. The mean serum phosphate at 480 min of 5.5 0.8 mg ⁄ dL in Group I and 5.1 0.3 mg ⁄ dL in Group II (P = 0.21). The overall differences between the two groups resulted in a large disparity between the generated areas under the phosphate vs. time curves (AUC) between the two groups. AUC for Group I was 2864 180 mg ⁄ dL*min, while AUC for Group II was 2428 223 mg ⁄ dL*min (P < 0.005), while normalized AUC for Group I was 1120 190 mg ⁄ dL*min and was 685 136 mg ⁄ dL*min for Group II (P < 0.001). Table 2 shows data for serum phosphate levels. Comparisons of serum phosphate vs. time AUC are shown in Figure 2. Serum phosphate vs. time AUCs had close negative

1206 E . D . E H R E N P R E I S

Table 2. Serum phosphate levels in study subjects Baseline (mg ⁄ dL)

1h (mg ⁄ dL)

2h (mg ⁄ dL)

3h (mg ⁄ dL)

4h (mg ⁄ dL)

4h (norm) (mg ⁄ dL)

AUC (phos ⁄ min) (mg ⁄ dL*min)

Normalized AUC (mg ⁄ dL*min)

4.3 4.4 4.0 3.1 3.4 3.1 3.6 3.6 0.5

6.2 6.3 5.8 6.2 6.8 7.8 6.7 6.5 0.6

6.7 6.7 7.8 6.5 6.6 7.5 7.1 7.8 0.5

6.4 6.7 7.8 6.0 6.7 5.8 6.4 6.5 0.6

6.6 6.7 6.2 6.3 6.5 5.8 5.8 6.2 0.4

2.3 2.3 2.2 3.2 3.1 2.7 2.2 2.6 0.4

3090 3071 2652 2800 2969 2799 2671 2864.6 180.5

916.8 959.4 1010 1312 1337 1311 992.2 1119.8 189.7

4.1 4.0 3.4 3.0 4.4 3.1 3.7 0.6 N.S.

5.9 6.5 4.7 6.0 6.8 5.0 5.8 0.8 N.S.

4.4 6.3 4.5 4.5 6.2 4.7 5.1 0.9 P < 0.001

4.6 6.3 4.5 4.5 5.6 4.1 4.9 0.8 P < 0.005

4.8 5.5 4.6 4.6 5.7 4.5 5.0 0.5 P < 0.001

0.7 1.5 1.2 1.6 1.3 1.4 1.3 0.3 P < 0.0001

2464 2754 2279 2295 2609 2166 2427.8 223.1 P < 0.005

496.4 833.7 647.1 855.4 624.3 655.6 685.4 136.2 P < 0.001

Subject Group I 1 2 3 4 5 6 7 Mean Group II 1 2 3 4 5 6 Mean

Table 3. Results of urine studies

Subject Group I 1 2 3 4 5 6 7 Mean Group II 1 2 3 4* 5 6 Mean

12 h urine ca (mg)

Estimated Creatinine clearance (mL ⁄ min)

Calculated Creatinine clearance (L ⁄ min)

Fluid intake (mL)

710 1575 1056 1378 1620 1566 1352 1322.4 332.1

15 7 22 27 10 23 11 16.4 7.6

77 86.5 88.3 68.7 92.4 97.1 79.6 84.3

54.6 138 65.7 156.2 142.2 206.1 133.3 128.0 52.4

4220 3730 3980 3970 3720 3200 3740 3794 319

1500 2050 725 140 958 1254 1104.5 658.6 P = 0.91

44 33 29 7 47 43 39.2 7.8 P < 0.001

172.0 128.8 117.4 84.4 112.0 128.2 124 28.6 P < 0.01

268.0 180.9 79.7 – 42.2 156.3 145.5 88.7 P = 0.68

3980 3970 4960 3990 3110 3480 3915 623 P = 0.66

12 h urine Volume (mL)

12 h urine phos (mg)

1650 2250 800 2650 2000 2900 800 1864 833 2000 2050 1250 100 550 1150 1400 630 P = 0.32

* Removed.

Aliment Pharmacol Ther 29, 1202–1211 ª 2009 Blackwell Publishing Ltd

S O D I U M P H O S P H A T E : R I S K I N S M A L L E R P A T I E N T S 1207 10

Phosphate (mg/kg)

8

6

4 Subject group I

2

Subject group II 0 0

100

200 300 Time (min)

400

500

Figure 1. Representative serum phosphate vs. time curves in a subject from Group I and Group II. Both demonstrate a rapid rise in serum phosphate levels after ingestion of sodium phosphate. The subjects from Group I show higher serum phosphate and a larger area under the serum phosphate vs. time curve.

4000

correlations with body weight, lean body mass and total body water. Fluid intake was similar between the two groups as shown in Table 3. No significant differences in urine volume over 12 h or total urine phosphate excretion were seen. The 12 h excretion of calcium was significantly lower in Group I (mean 16.4 7.6 mg) compared with Group II (mean 39.2 7.8 mg), P < 0.001. Results of urine studies are shown in Table 3. Measured serum calcium levels below the normal range were seen in four subjects in Group I and three subjects in Group II. Decreased serum calcium compared to baseline levels were seen throughout the 480 min measurements in all subjects in group I and five subjects in Group II. Decreased ionized serum calcium was seen in six subjects in Group I and two subjects in Group II, with repeated abnormal measurements throughout the study in five subjects in Group I and one subject in Group II. Abnormal HCO3 levels were noted in two subjects in Group I and three subjects in Group II. No other abnormal electrolyte values were seen throughout the study. No weight loss was seen in either group of subjects, while decreased total body water at various time intervals was noted during the study in two subjects in Group I and four subjects in Group II.

phosphate AUC (mg/dL*min)

3000

DISCUSSION

2000

1000

ro up G

G

ro u

p

I

II

0

Figure 2. Comparison of mean area under the serum phosphate vs. time curve (AUC) for the two subject groups. Mean AUC for Group I was 2864.6 180.5 mg ⁄ dL*min and was 2427.8 223.1 mg ⁄ dL*min, P < 0.005.


Elevated serum phosphate levels is an inevitable consequence of ingestion of high doses of sodium phosphate and has been demonstrated to occur with the use of these agents as colonoscopy preparations.1–5 In general, the transient, acute increases seen with these preparations have been deemed to be safe, with little or no clinical significance.3–5, 21 Secondary hypocalcaemia, sometimes with associated life threatening tetany, occasionally has been reported.22 Sodium phosphate-containing colonoscopy preparations, especially Fleet’s Phospho-Soda, have been reported to produce dialysis-requiring renal failure due to calcium-phosphate nephropathy.8, 9 Several pre-existing conditions have been assumed, but not proven, to increase the risk of calcium-phosphate nephropathy after colonoscopy preparation with sodium phosphate containing agents.8–11 These have included underlying kidney disease, longstanding hypertension, the use of angiotensin converting enzyme (ACE)-inhibitors, angiotensin receptor blockers (ARB) and diuretics

1208 E . D . E H R E N P R E I S

as well as diabetes and advanced age. Avoidance of sodium phosphate colonoscopy preparations in these risk groups has been recommended. However, at least 20% of the patients reported with sodium phosphate-induced renal failure did not have any of these risk factors. Of note, a majority of reported cases of renal failure secondary to calcium-phosphate nephropathy have occurred in females. Based on computer simulations, our group has suggested that lower body weight (with an associated decrease in lean body mass and total body water) may be as a possible risk factor for calcium phosphate nephropathy.15 Several prior studies have investigated the pharmacokinetics of serum electrolytes after a colonoscopy preparation. The first of these by DiPalma et al.23 demonstrated that elevated serum and urine phosphate with decreased serum and urine calcium and serum ionized calcium occurred in seven normal volunteers ingesting two 45 mL doses of Fleet’s Phospho-Soda separated by 12 h. These investigators were the first to demonstrate that the calcium phosphate solubility product was exceeded in their subjects and raised the possibility of nephrolithiasis as a consequence of these preparations. Caswell et al. administered two 45 mL doses of Fleet’s Phospho-Soda at 12 h intervals to twenty four subjects.24 They divided subjects into those between 30 and 64 years of age and those 65 years of age and older. They found peak serum phosphate levels at 3 h after the first dose and 2 h after the second dose, with fluctuations of serum calcium, sodium and potassium within the normal range. Subjects were persuaded to drink 480 mL liquid with the initial dose and then persuaded to drink an additional 1080 mL fluid throughout the AM after the first dose of Fleet’s Phospho-Soda. These authors concluded that electrolyte abnormalities seen in their subjects were transient and clinically insignificant. The current study differs from prior pharmacokinetic studies because only a single dose of Fleet’s PhosphoSoda was administered, ongoing monitoring of subject weight and body composition was performed to guide the study coordinators to encourage fluid intake to prevent dehydration, subjects were fed breakfast and lunch during the study and subjects were stratified according to body weight. Our study also examined phosphate levels that were normalized for initial serum phosphate levels, showing accentuated differences between smaller and larger subjects. Our data indicate that smaller individuals develop higher serum phosphate levels for a prolonged period

of time after ingesting large doses of sodium phosphate compared to larger individuals. These effects occur after a single dose of Fleet’s Phospho-Soda, even under the idealized conditions of this study, including close monitoring of weight and total body water with encouragement of increased fluid consumption after decreases in these parameters. Areas under the serum phosphate vs. time curves measured to the end of the collection period were also markedly and statistically higher in the smaller patients. These effects are demonstrated even more dramatically when serum levels are normalized for baseline phosphate levels. Smaller subjects were more likely to have abnormal ionized calcium levels of prolonged duration. Decreased urinary excretion of calcium was also seen in the smaller subjects. The clinical significance of hyperphosphatemia in smaller individuals lies in the potential for complications of this pathophysiological state, namely calcium-phosphate nephropathy. It is known that hyperphosphatemia, such as induced by ingestion of high doses of sodium phosphate, results in several occurrences that create the milieu responsible for calciumphosphate nephropathy.25, 29 Hyperphosphatemia causes marked release of parathyroid hormone (PTH),23 which in turn blocks tubular reabsorption of phosphate. This results in a prolonged increase in urinary phosphate concentrations.25 When the calcium-phosphate product is exceeded, precipitation of calcium phosphate crystals occurs, particularly in the distal tubules and collecting ducts.26, 27 It has been suggested that this phenomenon of hyperphosphatemia coupled with PTH-induced blockade of reabsorption of phosphate in the kidneys is particularly hazardous after ingestion of the second dose of sodium phosphate called for in the colonoscopy preparation.26, 30 Subjects with higher serum phosphate levels who fail to follow directions for recommended fluid intake or are unable to do so because of nausea and ⁄ or fatigue from the intensity of the preparation will add the element of dehydration to the process. Concentrated urine due to extreme gastrointestinal fluid losses coupled with poor oral replacement in these individuals could result in exponential rises in urinary calcium-phosphate product and calcium phosphate nephropathy.26, 27 The clinical use of sodium phosphate containing preparations in smaller individuals would potentially more resemble this scenario than the conditions established in our study. In general, split dosing of these agents is recommended.28 Although high fluid intake is recommended to patients Aliment Pharmacol Ther 29, 1202–1211 ª 2009 Blackwell Publishing Ltd


during and after consumption of these preparations, close monitoring and reaction to changes in body compartments, as performed in this study, cannot be accomplished at home. It is certainly feasible that if subjects take their first dose in the evening, drink a modest amount of fluid, have a large amount of diarrhoea, then fall asleep, smaller individuals are at risk for having a prolonged period of time during which their serum phosphate levels are elevated, while body fluids are markedly depleted. On the following morning, another large dose of sodium phosphate is consumed, leading to further diarrhoea and body fluid depletion. Prior studies have demonstrated that these bowel preparations, as performed at home, result in an average loss of weight of approximately 1 kg or more.25 Loss of the fluid compartment is expected to account for the majority of these losses. Lower fluid intake could also follow from gastrointestinal side effects of sodium phosphate-containing colonoscopy preparations. For example, Kastenberg et al. reported nausea 35.8%, abdominal pain in 31.1% and bloating in 47.1% of patients consuming sodium phosphate tablets for a colonoscopy preparation.21 Smaller individuals using high dose sodium phosphate colonoscopy preparations at home could thus be at very high risk for severe consequences of ingestion including calcium phosphate nephropathy due to higher than anticipated serum phosphate levels. It is our belief that these physiological occurrences in subjects with lower body weight (with associated decreased lean body mass) may represent a missing link that accounts for the calcium phosphate nephropathy and renal failure in patients with normal renal function and no other identifiable risk factors prior to exposure to sodium phosphate colonoscopy preparations. In this study, we used a device that detects near infrared energy (NIR) for body compartment analysis. We used this previously validated method18, 19 to measure percentage body fat, lean body mass and total body water. Body compartment analysis was performed for the purpose of further stratifying and comparing the differences between our two patient groups. We also used this method as a tool to prevent large fluctuations in body fluids during the post-ingestion period. To our knowledge, this is the first such use of NIR for this purpose. Further validation of NIR for following fluid fluxes in other patient groups is suggested. Measured fluid intake was similar between the two groups. Despite this, there were marked individual differences in the volume of fluid collected over 12 h, Aliment Pharmacol Ther 29, 1202–1211 ª 2009 Blackwell Publishing Ltd

with a not statistically significant larger mean fluid volume in Group I. Several explanations are possible for the wide differences in urine volumes noted. Urine was collected according to the discretion of the subjects. This added for potential for error due to difficulty of the collection process (especially in female subjects), discarding of some urine if urination occurred during defecation and subject noncompliance. In fact, one subject in Group II only turned in 100 mL urine and his urine data were removed from analysis. Alternatively, urine volume would depend in part on intestinal fluid losses. This could vary a great deal from one individual to the next. In addition, increased urine phosphate may have diuretic properties. This could potentially perpetuate the vicious cycle of hyperphosphatemia, dehydration and urinary crystallization of calcium phosphate in smaller individuals. Due to our specific criteria for study entry prior to initiation of data collection, namely weight at home less than 55 kg for Group I and greater than 100 kg for Group II, we were unable to stratify our subjects according to gender. In fact, all subjects in group I were female and five of six subjects in Group II were male. It could thus be argued that the marked differences in serum phosphate and calcium fluxes seen between the two groups could in part have been due, in part, to gender differences. Caswell et al.24 have performed detailed pharmacokinetic studies comparing electrolyte changes in a group of subjects that were divided according to age and gender. Their subjects were all of similar weight and body mass index. Their study consisted of twenty four healthy volunteers; six men and six women who were 30–64 years of age and six men and six women who were 65 years of age and older. Their subjects received two doses Fleet’s Phospho-Soda over a 12-h period. Body weight, blood pressure, electrolytes and liquid intake were monitored at frequent intervals throughout the study. They did not report significant differences in the serum phosphate or calcium levels between the four groups, but noted that mean serum phosphate levels were the highest in the elderly female group. Significant overlapping of serum phosphate and calcium levels appeared to have occurred in the rest of the subjects. All volunteers in our study were under the age of 60 and their mean ages were 35 in Group I and 31 in Group II. As can be seen from Table 1, subjects in Group I and Group II were highly separated by height, weight, lean weight and body water. For example, median baseline weight in Group I was 60 kg compared to 119.2 kg in Group

1210 E . D . E H R E N P R E I S

II (P < 0.005), while median lean body weight in Group I was 40 kg vs. 80.9 in Group II (P < 0.005). Median body water was 30.3 L in Group I compared to 61.6 L in Group II (P < 0.005). We believe that these parameters account for the marked differences in serum phosphate levels and calcium fluxes seen in the two groups. Although the study by Caswell et al. suggests that female patients should not have higher serum phosphate levels and larger calcium fluxes, as described in our study, the small number of subjects in their study does not, in fact, prove that gender differences in absorption and excretion of sodium phosphate are not present. As our study groups were not gender matched, our study also does not definitively establish that differences in serum phosphate and calcium fluxes were not due to gender differences between the two groups. Another pharmacokinetic study stratifying patients according to both age and body weight would be required to validate the concepts suggested in the current study. Several explanations may be given for higher serum phosphate levels in smaller individuals. It is well known in pharmacokinetics that serum levels of a drug after administration is dependent on that drug’s volume of distribution. Although the absorption characteristics of sodium phosphate are complex,23–26 increased serum levels in the initial portions of the serum vs. time curves are most dependent on volume of distribution. The terminal portion of the phosphate vs. time curve is dependant on clearance, which would be primarily a result of glomerular filtration rate (GFR). Of note, serum phosphate levels 480 min after ingestion of sodium phosphate were similar in the two groups, suggesting that significant differences in AUCs were due to higher initial phosphate and peak levels. Although the subjects in Group I had a lower estimated creatinine clearance compared with Group II, measured 12 h creatinine clearance in the two groups

REFERENCES 1 Nelson DB, Barkun AN, Block KP, et al. ASGE Technology Committee. Technology status evaluation report: colonoscopy preparations. Gastrointest Endosc 2001; 54: 829–32. 2 Tooson JD, Gates LK Jr. Bowel preparation before colonoscopy. Choosing the

was not statistically different. This could in part be due to issues of compliance, as the last 4 h of urine collection were performed at home. In addition, all subjects in Group I were women and 5 of 6 subjects in Group II were men. It is certainly reasonable to anticipate that less renal clearance of phosphate might occur in smaller subjects due to decreased GFR, although in this study, total phosphate excretion was the same for the two groups. This study also demonstrates that decreased serum bicarbonate, calcium and ionized calcium levels appear to be a direct result of hyperphosphatemia. Other previously described electrolyte abnormalities such as hypernatremia and hypokalemia probably occurs secondary to dehydration and should be avoidable by adequate fluid and electrolyte replacement during ingestion of the colonoscopy preparation.

ACKNOWLEDGEMENTS Declaration of personal interests: None. Declaration of funding interests: This study was funded in part by Braintree Laboratories, Inc., Grant 2008, EI, and by Arthur C. Neilsen Jr. The author (Dr Ehrenpreis) conceived this study, wrote the protocol, performed all regulatory processes with the IRB, sought funding, was present for all data collection, processed all data including statistical analysis and wrote the manuscript without assistance. Funding was utilized to pay for data collection and measurement of serum and urine chemistries only. The author has no conflict of interest regarding the performance or presentation of this study. The author would like to thank the study coordinators who made data collection possible. These include Ana Ehrenpreis, Kimberly Osimowicz, Raiza Hamm, Sonia Seidman, Maryanne Conners-Priest and Jessie Centura. Additional gratitude is expressed to Dr Bruce Kaplan for assistance in study design.

best lavage regimen. Postgrad Med 1996; 100: 203–14. 3 Vanner SJ, MacDonald PH, Paterson WG, Prentice RS, Da Costa LR, Beck IT. A randomized prospective trial comparing oral sodium phosphate with standard polyethylene glycol-based lavages solution(Go lightly) for bowel cleansing. Amer J Gastroenterol 1990; 85: 422–7.

4 Hookey LC, Depew WT, Vanner S. The safety profile of oral sodium phosphate for colonic cleansing before colonoscopy in adults. Gastrointest Endosc 2002; 56: 895–902. 5 Cohen SM, Wexner SD, Binderow SR, et al. Prospective, randomized, endoscopic-blinded trial comparing precolonoscopy bowel cleansing methods. Dis Colon Rectum 1994; 37: 689–96.


S O D I U M P H O S P H A T E : R I S K I N S M A L L E R P A T I E N T S 1211 6 Ehrenpreis ED, Nogueras JJ, Botoman VA, Bonner GF, Zaitman D, Secrest KN. Serum electrolyte abnormalities secondary to Fleet’s Phospho-Soda colonoscopy prep. Surg Endosc 1996; 10: 1022–4. 7 Beloosesky Y, Grinblat J, Weiss A, Grosman B, Gafter U, Chagnac A. Electrolyte disorders following oral sodium phosphate administration for bowel cleansing in elderly patients. Arch Intern Med 2004; 163: 803–8. 8 Markowitz GS, Nasr SH, Klein P, et al. Renal failure due to acute nephrocalcinosis following oral sodium phosphate bowel cleaning. Human Pathol 2004; 35: 675–84. 9 Markowitz GS, Stokes MB, Radhakrishnan J, D’Agati V. Acute phosphate nephropathy following oral sodium phosphate bowel purgative: an under recognized cause of chronic renal failure. J Am Soc Nephrol 2005; 16: 3389–96. 10 Tan JJ, Tjandra JJ. Which is the optimal bowel preparation for colonoscopy-a meta-analysis. Colorectal Dis 2006; 8: 247–58. 11 Wexner SD, Beck DE, Baron TH, et al. A consensus document on bowel preparation before colonoscopy: Prepared by a task force from the American Society of Colon and Rectal Surgeons (ASCRS), the American Society for Gastrointestinal Endoscopy(ASGE), and the Society of American Gastrointestinal and Endoscopic Surgeons (SAGES). Dis Colon Rectum 2006; 49: 792–809. 12 Abaskharoun R, Depew W, Vanner S. Changes in renal function following administration of oral sodium phosphate or polyethylene glycol for colon cleansing before colonoscopy. Can J Gastroenterol 2007; 21: 227–31. 13 Sica DA, Carl D, Zfass AM. Acute phosphate nephropathy-An emerging issue. Amer J Gastroenterol 2007; 102: 1844–7.


14 Lien YH. Is bowel preparation before colonoscopy a risky business for the kidney?. Nature Clin Practice Nephrol 2008; 4: 606–14. 15 Ehrenpreis ED, Varala K, Hannon B. Lower weight is a risk factor for calcium phosphate nephropathy with sodium phosphate colonoscopy preparation: a simulation study. Amer J Gastroenterol 2008; 103: S408–55. 16 Lanza E. Determination of moisture, protein, fat and calories in raw pork and beef by near infrared spectroscopy. J Food Sci 1983; 48: 471–4. 17 Norms KH. Extracting information from spectrophotometric curves. Predicting chemical composition from visible and near-infrared spectra. In: Proceedings of IUFOST Symposium on Food Research and Data Analysis. Oslo Norway: Applied Science Publishers Ltd, 1983: 95–113. 18 Conway JM, Norris KH, Bodwell CE. A new approach for the estimation of body composition: infrared interactance. Amer J Clin Nutrr 1984; 40: 1123–30. 19 Kalantar-Zadeh K, Bross R, Zitterkoph J, Colman S, Gjertson DW, Kopple JD. Comparing four body composition assessment methods for body fat measurement in dialysis patients. J Am Soc Neph 2004; 14(suppl): 591A. 20 McCauley D. Creatinine clearance calculator. http://www.globalrph.com/crcl. htm. 21 Kastenberg D, Chasen R, Choudhary C, et al. Efficacy and safety of sodium phosphate tablets compared to PEG solution in colon cleansing: two identically designed, randomized, controlled, parallel group, multicenter phase II trials. Gastrointest Endosc 2001; 54: 705–13. 22 Ehrenpreis ED, Wieland JM, Cabral J, Estevez V, Zaitman D, Secrest K. Symptomatic hypocalcemia, hypomagnesemia,

23

24

25

26

27

28

29

30

and hyperphosphatemia secondary to Fleet’s Phospho-Soda colonoscopy preparation in a patient with a jejunoileal bypass. Dig Dis Sci 1997; 42: 858– 60. DiPalma JA, Buckley SE, Warner BA, Culpepper RM. Biochemical effects of oral sodium phosphate. Dig Dis Sci 1996; 41: 749–53. Caswell M, Thompson WO, Kanapka JA, Galt DJB. The time course and effect on serum electrolytes of oral sodium phosphates solution in healthy male and female volunteers. Can J Clin Pharmacol 2007; 14: e260–74. Patel V, Emmett M, Santa Ana CA, Fordtran JS. Pathogenesis of nephrocalcinosis after sodium phosphate catharsis to prepare for colonoscopy: intestinal phosphate absorption and its effect on urine mineral and electrolyte excretion. Hum Pathol 2007; 28: 193–4. Lien YHH. Is bowel preparation before colonoscopy a risky business for the kidney? Nature Clin Practice 2008; 4: 606– 14. Asplin JR, Mandel NS, Coe FL. Evidence of calcium phosphate supersaturation in the loop of Henle. Am J Physiol Renal Physiol 1996; 270: F604–13. Rex D. Dosing considerations in the use of sodium phosphate bowel preparations for colonoscopy. Ann Pharmacother 2007; 41: 1466–75. Broadus AE, Magee JS, Mallette LE, et al. A detailed evaluation of oral phosphate therapy in selected patients with primary hyperparathyroidism. J Clin Endocrinol Metab 1983; 56: 953–61. Gonlusen G, Akgun H, Ertan A, Olivero J, Truong L. Renal failure and nephrocalcinosis associated with oral sodium phosphate bowel cleansing. Arch Pathol Lab Med 2006; 130: 101–6.