Sodium Phosphate Tablets and Acute Phosphate Nephropathy

We report 10 cases of acute phosphate nephropathy (APN), associated with acute renal failure (ARF), following administration of sodium phosphate tablets (SPTs) that were used as a bowel preparation before colonoscopy. SPTs were first approved by the US Food and Drug Administration (FDA) in September 2000 for bowel cleansing at a dose of 60 g (40 tablets) in two divided doses with adequate hydration (trade name Visicol). The proposed mechanism of action of SPTs is an osmotic effect. The Visicol product was subsequently reformulated, because it was found that one of the excipients, microcrystalline cellulose, was leaving a residue in the colons of some patients that impaired visibility during their colonoscopy (1). In addition, the manufacturer of this product provided data demonstrating that patients’ bowels were effectively cleansed with the use of 48 g (32 tablets) of sodium phosphate (SP) in divided doses with adequate hydration. With these doses, a new product (trade name OsmoPrep) was approved by the FDA in March 2006. Both products are marketed by Salix Pharmaceuticals at a dose of 1.5 g SP per tablet. Note that APN also has been associated with the use of SP oral solution for bowel cleansing (2). This article addresses the SPT dosage form only. (SP oral solution is no longer available over the counter for the indication of bowel cleansing. CB Fleet is applying for prescription status (3).) From July 2006 to September 2008, 10 cases of APN associated with SPT use for bowel cleansing were reported to the FDA’s Adverse Event Reporting System (AERS) database (Table 1; patients 1–7 took Visicol and patients 8–10 took OsmoPrep). Note that two cases (patients 5 and 6 in Table 1) also have been reported in the literature; no additional cases have been published (2,4). Renal biopsy in these 10 patients revealed nephrocalcinosis (calcium phosphate crystal deposition in the distal tubules and collecting ducts). Decreased intravascular volume caused by bowel cleansing may also contribute to an increase in the phosphate concentration in renal tubular fluid. The patients in this case series all had at least one underlying risk factor for ARF, and most had more than one (Table 1). In addition to decreased intravascular volume as described above, other risk factors for ARF include preexisting renal insufficiency, hypertension, diabetes mellitus, advanced age, underlying electrolyte imbalance, and concomitant medications that affect renal perfusion or function (e.g., angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blocker (ARB) drugs, nonsteroidal anti-inflammatory drugs (NSAIDs), or diuretics) (5,6). Patients with chronic inflammatory bowel disease may have enhanced absorption of SP (7). One retrospective case-controlled study using records from the University of Pennsylvania health-care system found a significant association between ACEI/ARB use and acute kidney injury in patients with baseline serum creatinines of ≤1.5 mg/ dl who received SP for bowel cleansing before colonoscopy (8). It has been documented in the literature that maintaining adequate hydration and dividing the two SP doses by 12 hours are important to prevent intravascular volume depletion and thereby minimize the risk of APN and ARF (9–11). The product labels for Visicol and OsmoPrep recommend that patients take a total of at least 112 ounces and 64 ounces of liquid, respectively, and that divided doses be taken in the evening and morning before colonoscopy (7,12). In this case series, there is little reported information on whether patients took SPTs correctly or incorrectly (information of this type is typically not provided in adverse-event reports and is difficult to obtain on follow-up inquiry); patient 9 separated the two SP doses by 3 hours only. Four patients in this case series required dialysis. To date, there have been no reported deaths or kidney transplantations associated with SPT use. From 1 January 2001 through 30 September 2008, 2,297,958 and 1,329,497 prescriptions were dispensed for Visicol and for OsmoPrep, respectively (Figure 1). These data were obtained from the SDI Vector One database, which measures retail dispensing of prescriptions. In addition to the 10 cases of APN, the AERS has received 31 cases reported as ARF associated with SPT use. These patients did not have a renal biopsy to document nephrocalcinosis; therefore, the mechanism of renal failure with SPT exposure could not be verified. Of these 31 cases, 12 patients used Visicol, 4 patients used a foreign SP product (1 g per tablet, total dose of 50 g in divided doses with adequate hydration), and 15 patients used OsmoPrep. A total Sodium Phosphate Tablets and Acute Phosphate Nephropathy

We report 10 cases of acute phosphate nephropathy (APN), associated with acute renal failure (ARF), following administration of sodium phosphate tablets (SPTs) that were used as a bowel preparation before colonoscopy. SPTs were first approved by the US Food and Drug Administration (FDA) in September 2000 for bowel cleansing at a dose of 60 g (40 tablets) in two divided doses with adequate hydration (trade name Visicol). The proposed mechanism of action of SPTs is an osmotic effect. The Visicol product was subsequently reformulated, because it was found that one of the excipients, microcrystalline cellulose, was leaving a residue in the colons of some patients that impaired visibility during their colonoscopy (1). In addition, the manufacturer of this product provided data demonstrating that patients' bowels were effectively cleansed with the use of 48 g (32 tablets) of sodium phosphate (SP) in divided doses with adequate hydration. With these doses, a new product (trade name OsmoPrep) was approved by the FDA in March 2006. Both products are marketed by Salix Pharmaceuticals at a dose of 1.5 g SP per tablet. Note that APN also has been associated with the use of SP oral solution for bowel cleansing (2 Table 1) also have been reported in the literature; no additional cases have been published (2,4). Renal biopsy in these 10 patients revealed nephrocalcinosis (calcium phosphate crystal deposition in the distal tubules and collecting ducts). Decreased intravascular volume caused by bowel cleansing may also contribute to an increase in the phosphate concentration in renal tubular fluid.
The patients in this case series all had at least one underlying risk factor for ARF, and most had more than one ( Table  1). In addition to decreased intravascular volume as described above, other risk factors for ARF include preexisting renal insufficiency, hypertension, diabetes mellitus, advanced age, underlying electrolyte imbalance, and concomitant medications that affect renal perfusion or function (e.g., angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blocker (ARB) drugs, nonsteroidal anti-inflammatory drugs (NSAIDs), or diuretics) (5,6). Patients with chronic inflammatory bowel disease may have enhanced absorption of SP (7). One retrospective case-controlled study using records from the University of Pennsylvania health-care system found a significant association between ACEI/ARB use and acute kidney injury in patients with baseline serum creatinines of ≤1.5 mg/ dl who received SP for bowel cleansing before colonoscopy (8).
It has been documented in the literature that maintaining adequate hydration and dividing the two SP doses by 12 hours are important to prevent intravascular volume depletion and thereby minimize the risk of APN and ARF (9)(10)(11). The product labels for Visicol and OsmoPrep recommend that patients take a total of at least 112 ounces and 64 ounces of liquid, respectively, and that divided doses be taken in the evening and morning before colonoscopy (7,12). In this case series, there is little reported information on whether patients took SPTs correctly or incorrectly (information of this type is typically not provided in adverse-event reports and is difficult to obtain on follow-up inquiry); patient 9 separated the two SP doses by 3 hours only.
Four patients in this case series required dialysis. To date, there have been no reported deaths or kidney transplantations associated with SPT use.
From 1 January 2001 through 30 September 2008, 2,297,958 and 1,329,497 prescriptions were dispensed for Visicol and for OsmoPrep, respectively ( Figure 1). These data were obtained from the SDI Vector One database, which measures retail dispensing of prescriptions.
In addition to the 10 cases of APN, the AERS has received 31 cases reported as ARF associated with SPT use. These patients did not have a renal biopsy to document nephrocalcinosis; therefore, the mechanism of renal failure with SPT exposure could not be verified. Of these 31 cases, 12  nine laboratory values were measured in samples obtained up to 3 months before sPt administration with the following exceptions: time frame not specified for patient 1, and baseline for patient 4 was taken 4-5 months before sPt administration.

The red secTion 1905
association of APN and SP product use for bowel cleansing on its Web site (14). Since May 2006, the FDA has received additional cases of APN and ARF associated with SPT use. As a result of this new safety information, the FDA has required the manufacturer of prescription SPT products (Visicol and OsmoPrep) to strengthen product labeling to include a boxed warning, provide a medication guide for patients, and implement a risk evaluation and mitigation strategy to ensure that the benefits of SP products outweigh the risks. The FDA is also requiring that a postmarketing clinical trial be conducted to further assess the risk of acute kidney injury with use of these products. For additional information, see the December 2008 FDA Alert at the FDA's website (15). The patients in this case series had potentially confounding medical conditions or were taking concomitant medications that put them at increased risk for nephrotoxicity; however, a causal or contributory role of SPTs cannot be ruled out. It appears that some patients, especially those with risk factors, developed APN with doses as small as 30 g of SP.
Clinicians should be aware when prescribing SPTs that some patients who developed renal injury did not present with symptoms of APN for up to several months after using SPTs. Until further information is available (e.g., a recently proposed clinical trial by the FDA (15)), there is insufficient information to make global recommendations regarding standard pre-and postprocedure renalfunction testing for patients who may be at risk. In addition, the importance of taking SPTs correctly (i.e., with adequate hydration and splitting the two SP doses by 12 hours) should be stressed in order to reduce the risk of developing APN and ARF.

CONFLICT OF INTEREST
Guarantor of the article: Ann Corken Mackey, RPh, MPH. Specific author contributions: Each author contributed to the draft and review of this article. Financial support: None. Potential competing interests: None.
reported for 6 patients). Baseline creatinine values were reported for 13 patients (these were within normal limits except for one patient with preexisting renal insufficiency); postprocedure creatinine values were reported for 20 patients and ranged from 1.2 to 10 mg/dl (mean 5.7 mg/dl). Two patients' creatinine values were reported as "elevated"; postprocedure creatinine values were not included in reports for 9 patients.
The AERS is a spontaneous, voluntary surveillance system that collects reports of adverse events for US marketed medicinal products. Manufacturers holding New Drug Applications or Biologic Licensing Applications must submit adverse-event reports to the AERS according to the Code of Federal Regulations. Data from the AERS cannot be used to estimate true incidence rates of events, because the numerator is underestimated and the denominator can only be estimated. The AERS is subject to underreporting; fewer than 10% of adverse events are reported to the FDA, so there may be additional cases of APN associated with SPT use for bowel cleansing (13).
The manufacturer of SPTs added APN to the warnings section of the labels for its products in March 2006. At that time, the FDA posted information about the of 27 cases were reported from domestic sources, and 4 cases were reported from foreign sources. The patients ranged in age from 45 to 77 years (mean 65 years; age not reported for 1 patient); 22 patients were female, and 9 patients were male. The doses used ranged from 30 to 60 g (mean 49 g; n=23; dose was not reported for 8 patients). It was reported that 21 patients used a split-dose regimen (3 of the 21 patients split their dose with an interval of only 3-4 hours); 12 patients were reported to have followed hydration directions (information on dose and hydration was not reported for the remaining patients). Of the 31 patients, 4 patients used additional laxatives with SPTs (i.e., bisacodyl, magnesium citrate). The time to ARF detection after SPT use ranged from 0.5 to 60 days (mean 6.3 days; n=25; information was not provided for 6 patients). The following risk factors for renal dysfunction were reported (not mutually exclusive): hypertension (22 patients), diabetes mellitus (9 patients), preexisting renal insufficiency (2 patients), inflammatory bowel disease (2 patients), ACEI/ARB use (20 patients), diuretic use (8 patients), NSAID use (7 patients), and contrast dye use (2 patients). At least 25 patients were hospitalized for ARF, including 4 patients who required dialysis (outcome was not