Search tips
Search criteria 


Logo of mayoclinprocLink to Publisher's site
Mayo Clin Proc. 2009 October; 84(10): 906–911.
PMCID: PMC2755809

Clinical Pearls in Gastroenterology

At the 2001 annual conference of the American College of Physicians, a new teaching format to aid physician learning, Clinical Pearls, was introduced. Clinical Pearls is designed with the 3 qualities of physician-learners in mind. First, we physicians enjoy learning from cases. Second, we like concise, practical points that we can use in our practice. Finally, we take pleasure in problem solving.

In the Clinical Pearls format, speakers present a number of short cases in their specialty to a general internal medicine audience. Each case is followed by a multiple-choice question answered live by attendees using an audience response system. The answer distribution is shown to attendees. The correct answer is then displayed and the speaker discusses teaching points, clarifying why one answer is most appropriate. Each case presentation ends with a Clinical Pearl, defined as a practical teaching point that is supported by the literature but generally not well known to most internists.

Clinical Pearls is currently one of the most popular sessions at the American College of Physicians meeting. As a service to its readers, Mayo Clinic Proceedings has invited a selected number of these Clinical Pearl presentations to be published in our Concise Review for Clinicians section. “Clinical Pearls in Gastroenterology” is one of them.


A 48-year-old man presents for an evaluation of abnormal findings on liver tests detected during an insurance examination. He had a blood transfusion 25 years previously but denies drug or alcohol use. He denies a history of ascites, encephalopathy, or gastrointestinal bleeding.

Findings on Laboratory Tests

  • Complete blood cell count, normal
  • Aspartate aminotransferase, 115 U/L (reference ranges provided parenthetically) (8-48 U/L)
  • Alanine aminotransferase, 137 U/L (7-55 U/L)
  • Albumin, normal
  • International normalized ratio, normal
  • Bilirubin, normal
  • Hepatis C virus antibody, positive
  • Hepatitis C RNA, 200,000 IU/L
  • Hepatitis A virus antibody IgG and IgM, negative
  • Hepatitis B surface antigen and hepatitis B surface antibody, negative
  • Human immunodeficiency virus, negative
  • Ultrasonography of the liver, normal
  • Liver biopsy specimen, consistent with hepatitis C infection without cirrhosis


Which one of the following would be the most appropriate next step?

  1. Vaccinate against hepatitis A and B
  2. Measure serum α-fetoprotein level
  3. Perform esophagogastroduodenoscopy (EGD)
  4. Refer for liver transplant evaluation
  5. Perform bone densitometry


All patients with underlying liver disease, both with and without cirrhosis, should be considered for vaccination against hepatitis A and B unless they have evidence of active infection with these viruses or are serologically immune to them (the latter indicated by hepatitis A virus antibody IgG for hepatitis A and hepatitis B surface antibody for hepatitis B). A patient with compensated liver disease from any etiology may have severe clinical deterioration if acutely infected with hepatitis A or B; therefore, the Advisory Committee on Immunization Practices advises these vaccines be administered in patients with chronic liver disease. In addition, all patients with liver disease should be offered a pneumococcal vaccine as well as yearly influenza vaccinations, barring any contraindications.1

A serum α-fetoprotein measurement may be obtained in conjunction with liver imaging in patients with cirrhosis as surveillance for hepatocellular carcinoma. On the basis of laboratory test results, findings on ultrasonography, and ultimately findings on liver histology, this patient with hepatitis C does not have cirrhosis, and therefore screening for hepatocellular carcinoma would not be indicated. In contrast, certain populations of noncirrhotic carriers of hepatitis B should be considered for surveillance (Asian men aged ≥40 years, Asian women aged ≥50 years, persons of African descent aged >20 years, and persons with a family history of hepatocellular carcinoma). Other factors, such as advancing age, persistent viral replication, and coinfection with hepatitis C or human immunodeficiency virus, should be taken into consideration regarding surveillance in chronic carriers of hepatitis B in the absence of cirrhosis.2

An EGD is indicated in patients with newly diagnosed cirrhosis to screen for esophageal varices. The presence and size of varices determine the interval for subsequent surveillance or active management. Given that this patient does not have cirrhosis, an EGD is not indicated at this time.

Because the patient does not have clinical or laboratory features of hepatic dysfunction or evidence of cirrhosis, referral for a liver transplant would be premature as he would not meet the standard minimal listing criteria for liver transplant, which include a Child-Turcotte-Pugh score of 7 or greater, a Model for End-Stage Liver Disease score of 10 or more, or the first occurrence of a major complication (eg, ascites, variceal bleeding, or hepatic encephalopathy).3

Osteoporosis and osteomalacia have both been found to occur in patients with chronic liver disease (referred to as hepatic osteodystrophy), with the pathophysiology likely multifactorial and not well understood. Indications for measuring bone mineral density in patients with underlying liver disease include cirrhosis, newly diagnosed primary biliary cirrhosis, at time of transplant assessment, and any other classic risk factors for metabolic bone disease (eg, corticosteroid use, fracture history, and postmenopausal status).4 The presence of hepatitis C without cirrhosis or other risk factors for bone disease would make bone densitometry unnecessary for this patient.

Clinical Pearl

All patients with chronic liver disease should be considered for vaccination against hepatitis A and B unless actively infected or serologically immune.


A 24-year-old man presents for evaluation of dysphagia. He has had intermittent symptoms for 4 years. The dysphagia occurs only with solid foods and never with liquids. He notes mild, periodic heartburn and reflux but has not lost weight. He has gone to the emergency department on 3 occasions for food impactions. Findings on 2 previous EGDs have been normal. Findings on barium esophagography were also unremarkable. The patient began receiving high-dose proton pump therapy twice daily for 8 weeks with no change in his symptoms. His medical history is remarkable for seasonal allergies, for which he takes an over-the-counter allergy medication as needed.


Which one of the following is the next best step in the management of this patient?

  1. Esophageal manometry
  2. EGD with empiric dilation
  3. Reassurance alone
  4. EGD with esophageal biopsies
  5. H2-receptor blockade at bedtime


Eosinophilic esophagitis (EE) was first reported approximately 30 years ago but has received more attention in recent years. Eosinophilic esophagitis refers to non-reflux-mediated eosinophilic infiltration of the esophagus. It is seen more commonly in men (3:1 male-to-female ratio) and in those aged 30 to 40 years, but it can present at any age. Patients commonly have a personal or family history of atopic conditions (eczema, asthma, and seasonal or food allergies). The most common symptoms of EE are chronic, intermittent solid-food dysphagia and food impactions, with approximately half of all food impactions thought to be related to EE; reflux- and heartburn-related symptoms are reported less frequently. Endoscopic features can be very suggestive of EE (eg, concentric rings, small-caliber “feline” esophagus, linear furrows, fragility, white plaques), but up to 10% of adults and 33% of children with EE have normal findings on EGD.5

Eosinophilic esophagitis is a clinicopathologic disorder. In addition to the presence of symptoms (such as dysphagia and food impactions), the diagnosis is established by obtaining multiple esophageal biopsy specimens, with 15 or more eosinophils per high power field taken from the midesophagus indicative of the condition. Patients with lesser degrees of eosinophilic infiltration may have gastroesophageal reflux disease (GERD) alone; therefore, before a diagnosis of EE is made, consensus guidelines suggest that exclusion of disorders with similar presentations, most commonly GERD, be considered.6 This patient has been receiving high-dose proton pump inhibitor therapy without a change in symptoms, making GERD alone less likely.

Although elemental diets have been useful in children, many adults require corticosteroid therapy. Swallowed aerosolized fluticasone twice daily for approximately 6 weeks is recommended, although symptoms may recur in at least half of patients. Patients should be instructed not to eat or drink immediately after dosing to prevent rapid transit of the topical medication into the stomach. Although dilation of associated strictures may be necessary, this should be considered only after treatment with topical corticosteroids has been attempted, given the reported increased risk of esophageal tears or perforation.

Esophageal manometry is useful to clarify the diagnosis of achalasia, a motility condition of the esophagus that is associated with progressive solid-food and liquid dysphagia, the latter of which this patient did not have. Food impactions would also be less common with achalasia than with EE. Esophageal manometry has no role in the evaluation of a patient with suspected EE. As stated previously, empiric dilation in a patient with suspected EE may result in serious complications; the diagnosis should first be established and therapy considered before dilation. Reassurance alone without establishing the diagnosis of EE may result in increased morbidity with recurrent food impactions and other complications. This patient was appropriately treated with high-dose proton pump inhibitor therapy with no improvement in symptoms; therefore, adding nocturnal H2-receptor blockade will likely have no supplemental benefit.

Clinical Pearl

An EGD with esophageal biopsies needs to be considered in the patient with intermittent solid-food dysphagia to rule out EE even in the presence of normal endoscopic findings.


A 41-year-old woman presents with an 8-month history of bloating and abdominal discomfort, both relieved somewhat after a bowel movement. She denies nausea or vomiting. She has been prone to constipation most of her adult life and has a history of dysmenorrhea. The patient reports no prior abdominal surgeries, medications, or weight loss. Colonoscopy performed 1 year previously (because her father, aged 56 years, has colon cancer) was normal. Findings on examination are normal.

Findings on Laboratory tests

  • Hemoglobin, 10.7 g/dL
  • Mean corpuscular volume, 76 fL
  • Ferritin, 6 μg/L
  • Tissue transglutaminase (tTG) IgA, normal


Which one of the following is the next best step?

  1. Initiate treatment with dicyclomine
  2. Perform small bowel biopsies
  3. Perform colonoscopy
  4. Measure gliadin antibodies
  5. Perform intestinal transit studies


Iron-deficiency anemia (IDA), the most common extraintestinal manifestation of celiac disease, may be found in up to 50% of patients who are newly diagnosed as having this disease.7 In addition, IDA may be the sole manifestation of celiac disease and may be commonly overlooked in menstruating women such as this patient. Routine small bowel biopsies to evaluate for celiac disease should be performed in the patient undergoing an evaluation for IDA; the finding of other endoscopic abnormalities that could account for IDA (erosions, ulcers, vascular ectasias) should not preclude small bowel biopsies because these types of findings may be seen in up to one-third of patients with celiac disease.8 This patient's history of constipation is not enough to exclude celiac disease because up to 20% of patients with celiac disease report a history of constipation. Although findings on the tTG antibody assay were negative, the patient has experienced the onset of new intestinal symptoms and unexplained IDA, which warrants further evaluation. A tTG IgA antibody has a sensitivity ranging from 90% to 98% in detecting celiac disease, so a negative tTG does not definitively exclude the condition.

Although the patient may have features of irritable bowel syndrome (IBS), her symptoms and IDA merit further evaluation before she begins taking an antispasmodic agent such as dicyclomine. Many patients with celiac disease may actually fulfill the Rome criteria for IBS, yet IBS would not explain this patient's anemia.

A colonoscopy would be a reasonable first step to evaluate the patient's bowel symptoms and IDA, especially in the setting of her father's history of colon cancer. However, colonoscopy was performed 1 year previously with normal findings and would not need to be repeated as the next step, provided that the previous colonoscopy revealed an adequate bowel preparation and the endoscope was able to be advanced to the cecum.

Gliadin antibodies are rarely used because of the high number of false-positive results and poor specificity compared with tTG antibodies, offering little advantage over proceeding directly to small bowel biopsies.9 Intestinal transit studies can be used to assess for an underlying motility disorder; however, such a disorder is not suggested by the patient's history and would not explain the abnormal laboratory results.

Clinical Pearl

Iron-deficiency anemia is the most common extraintestinal manifestation of celiac disease; small bowel biopsies are indicated in the assessment of IDA to evaluate for celiac disease.


A 44-year-old man presents with a 6-month history of bloating and loose stools. He has 4 to 7 bowel movements daily (baseline, 1-2 daily), with nocturnal stools. His bloating is relieved after a bowel movement. There has been no hematochezia or melena. He underwent a Roux-en-Y gastric bypass 3 years previously for medically complicated obesity and has lost 54.5 kg since the operation. Although his weight loss has stabilized, he has lost an additional 4.5 kg during the previous month. His mother has Crohn disease. His child, aged 9 months, attends daycare. He takes an iron-containing multivitamin and oral vitamin B12 daily. Findings on examination are unremarkable.

Findings on Laboratory Tests

  • Hemoglobin, 10.8 g/dL
  • Mean corpuscular volume, 104 fL
  • Folate, 43 μg/L (≥3.5 μg/L)
  • Vitamin B12, 194 ng/L (200-650 ng/L)
  • Tissue transglutaminase IgA, normal

Stool cultures, including repeated cultures for ova and parasites, were negative. Findings on colonoscopy with random biopsies were normal. Esophagogastroduodenoscopy revealed post-Roux-en-Y anatomy, with otherwise normal findings. Small bowel biopsy specimens were normal and small bowel imaging was unremarkable.


Which one of the following is the most likely diagnosis?

  1. Irritable bowel syndrome
  2. Celiac disease
  3. Giardiasis
  4. Crohn disease
  5. Small intestinal bacterial overgrowth


Small intestinal bacterial overgrowth is one of the well-described long-term complications of bariatric surgery, with clinical features including diarrhea, bloating, excessive flatulence, and weight loss. Digestive juices, such as gastric acid, bile, and proteolytic enzymes, usually maintain a normal concentration of small intestinal bacteria. With some bariatric procedures such as a Roux-en-Y bypass, segments of bowel are excluded from the usual flow of these protective digestive juices, in which case bacterial overgrowth may occur.10 Patients with bacterial overgrowth may have diarrhea with malabsorption of fat (due to deconjugation of bile salts), protein (due to mucosal damage), or carbohydrates (due to loss of disaccharidases).11

Patients with bacterial overgrowth are also at risk of vitamin B12 deficiency because excess bacteria bind and consume vitamin B12 and cleave it prematurely from intrinsic factor. Hence, in a patient such as this who is currently receiving vitamin B12 yet has low serum values, excessive loss vs inadequate replacement should be considered. Another clue to the diagnosis, as suggested in this case, is an elevated serum folate level because excess intestinal bacteria synthesize folate. The diagnosis may be established by obtaining small bowel bacterial cultures or through breath testing; alternatively, an empiric course of antibiotic therapy could be used if confirmatory testing was unavailable.

Irritable bowel syndrome would not account for this patient's ongoing weight loss, nocturnal stools, or vitamin abnormalities, so other diagnoses need to be considered. Although celiac disease may be unmasked after bariatric surgery, this patient's negative serologic and histologic findings have essentially ruled out this condition. Although having a diapered child in a daycare environment is a risk factor for giardiasis, findings on repeated stool studies and small bowel biopsies have been normal, making this diagnosis less likely. The patient has a first-degree relative with IBS, but his negative endoscopic, histologic, and radiographic evaluations make this an unlikely diagnosis.

Clinical Pearl

Small intestinal bacterial overgrowth is a known complication of certain forms of bariatric surgery and should be considered in the postbariatric patient with diarrhea, bloating, and weight loss; laboratory features of bacterial overgrowth may include low vitamin B12 and elevated serum folate levels.


A 28-year-old man presents with recurrent, intermittent attacks of nausea and vomiting during the past few years. He feels well between episodes, but 2 to 3 times monthly he has several days of protracted nausea, vomiting, and epigastric discomfort. He is otherwise healthy, with no prior surgeries. He denies taking medications, except for antiemetic agents during attacks, which are rarely helpful. He has used marijuana daily for the past 5 years, which he thinks helps his nausea. Examinations have been unremarkable except for mild dehydration and generalized abdominal tenderness with attacks. Complete blood cell counts during his evaluations have shown mild leukocytosis. Metabolic panel results (liver biochemistries and levels of glucose, pancreatic enzymes, thyroid-stimulating hormone, calcium, and morning cortisol) have been normal. Findings on EGD, abdominal ultrasonography, computed tomographic enterography, gastric scintigraphy, and magnetic resonance imaging of the head have been unremarkable. The patient has been hospitalized for dehydration related to these attacks; the nurses report him taking hot showers for hours each day, which he states help his nausea.


Which one of the following is the most likely cause of the patient's symptoms?

  1. Psychogenic vomiting
  2. Idiopathic gastroparesis
  3. Small bowel obstruction
  4. Long-term marijuana use
  5. Chronic cholecystitis


Although delta-9-tetrahydrocannabinol (THC), the major ingredient found in marijuana, is known to be a powerful antiemetic agent, long-term use of marijuana has been associated with recurrent attacks of nausea and vomiting, known as the cannabinoid hyperemesis syndrome. It is unclear what triggers this paradoxical phenomenon in some long-term users of marijuana; because cannabinoids have a long half-life, are lipophilic, and bind to cerebral fat, the process is thought to involve alterations in the hypothalamus or limbic system. Studies have found that use of delta-9-THC can also cause secondary (rather than idiopathic) delay of gastric emptying. The attacks of hyperemesis often follow years of cannabis use and may be cyclical, occurring every few weeks to months. Patients may report colicky abdominal pain, excessive sweating, and intense thirst. Serum leukocytosis may be transiently noted. Antiemetic agents are often used but fail to produce meaningful relief for patients.12

Compulsive bathing in hot water frequently relieves symptoms.13 The hot water is so therapeutic to the nausea that patients have reportedly exhausted hot water supplies or scalded themselves while bathing. Although excessive bathing is not sensitive or specific for cannabinoid hyperemesis, it is a very characteristic feature. Cessation of cannabis use leads to disruption of the cyclical attacks of vomiting (distinguishing this condition from psychogenic vomiting), whereas reintroducing cannabis in susceptible persons causes recurrent attacks.14

Although psychogenic vomiting could present with cyclical features, it would be less likely to have the protracted symptoms seen with cannabinoid hyperemesis syndrome, and the association with compulsive bathing is unlikely. Idiopathic gastroparesis implies that no underlying mechanism for the delayed gastric emptying is apparent; delta-9-THC may delay gastric emptying, but this delay would be secondary to an identifiable cause and therefore would not be idiopathic. Although recurrent small bowel obstructions can be seen in a patient with no prior abdominal surgeries or other risk factors, such obstructions would be less likely in this case because features of obstruction were not revealed on abdominal examination and computed tomography. Chronic cholecystitis can present with varying degrees of abdominal pain and nausea, but cyclical protracted vomiting would not be typical. In this patient, no features of gallbladder pathology were revealed by abdominal ultrasonography.

Clinical Pearl

Cyclical attacks of nausea and vomiting, known as the cannabinoid hyperemesis syndrome, have been reported with the long-term use of marijuana; compulsive bathing in hot water is a characteristic feature.


A 68-year-old woman presents 3 months after her general examination for a routine blood pressure follow-up. She has a history of hypertension and has been receiving combination hydrochlorothiazide-lisinopril therapy for the past 6 months. At the time of her examination 3 months previously, her electrolytes were within normal limits, including a creatinine level of 1.0 mg/dL.

Findings on Current Laboratory Tests

  • Sodium, normal
  • Potassium, normal
  • Creatinine, 3.8 mg/dL (0.7-1.2 mg/dL)

The patient takes no other medications (including over-the-counter drugs) and has had no recent illnesses or changes other than a routine screening colonoscopy 2 weeks previously that was unremarkable.


Which one of the following bowel preparations was most likely used?

  1. Magnesium citrate solution
  2. Polyethylene glycol electrolyte solution
  3. Oral sodium phosphate solution
  4. Low-volume polyethylene glycol electrolyte solution plus bisacodyl tablets
  5. Tap-water enemas


Oral sodium phosphate solution and tablets have been widely accepted by patients because they are easily administered and by physicians because of their excellent preparation quality. Although these agents were always contraindicated for use in those with renal disease or electrolyte disturbances, acute phosphate nephropathy and renal failure have now been reported with the use of oral sodium phosphate bowel preparations in patients without clinically apparent renal disease. Initially thought to be a rare entity, acute phosphate nephropathy now appears to have been widely underrecognized.15 The Food and Drug Administration has issued an alert regarding acute phosphate nephropathy with the use of oral sodium phosphate bowel preparations, including both solution and tablet form, placing a black box warning on both these products.

Potential risk factors for the condition include advancing age, female sex, hypertension, and use of an angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker. Most patients are diagnosed as having acute renal injury within a few weeks to months after the colonoscopy, but these symptoms may go unrecognized depending on when the patient next seeks medical attention. Patients typically have minimal proteinuria, bland urinary sediment, and normocalcemia. Renal biopsies in these patients have revealed nephrocalcinosis. Chronic renal insufficiency has developed in most of these patients, and some have required hemodialysis.16 Given that oral sodium phosphate solution is available over the counter, patients and physicians need to be aware of this risk if this solution is used as a bowel-purging agent.

Magnesium citrate can be used as a bowel-purging agent before colonoscopy and is more commonly used in pediatric patients. Higher doses of magnesium citrate can result in hypermagnesemia, which can be associated with cardiac conduction abnormalities and electrolyte disorders (hypocalcemia, hypophosphatemia, and hyperkalemia). Patients with underlying renal disease are more at risk of developing hypermagnesemia, so magnesium citrate should clearly be avoided in these patients as a bowel-purging agent before colonoscopy. Polyethylene glycol electrolyte solution, in either full- (4-L) or low-volume (2-L) formulations, is the most commonly endorsed bowel-purging agent because of its safety profile. Tap-water enemas can be used in the management of constipation but are not endorsed as bowel-purging agents before colonoscopy; frequent use of high-volume tap-water enemas may lead to substantial electrolyte disturbances and should be avoided.

Clinical Pearl

Oral sodium phosphate solution and tablets have been associated with acute phosphate nephropathy; they should be used cautiously if at all as bowel-purging agents.

Supplementary Material

Author Interview:


See end of article for correct answers to questions.

Correct answers: Case 1: a, Case 2: d, Case 3: b, Case 4: e, Case 5: d, Case 6: c


1. Advisory Committee on Immunization Practices Recommended adult immunization schedule: United States, 2009. Ann Intern Med. 2009;150(1):40-44 [PubMed]
2. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology 2005;42(5):1208-1236 [PubMed]
3. Murray KF, Carithers RL., Jr AASLD Practice Guidelines: Evaluation of the patient for liver transplantation. Hepatology 2005;41(6):1407-1432 [PubMed]
4. Collier J. Bone disorders in chronic liver disease. Hepatology 2007;46(4):1271-1278 [PubMed]
5. Yan BM, Shaffer EA. Primary eosinophilic disorders of the gastrointestinal tract. Gut 2009. May;58(5):721-732 Epub 2008 Dec 3 [PubMed]
6. Furuta GT, Liacouras CA, Collins MH, et al. First International Gastrointestinal Eosinophil Research Symposium (FIGERS) Subcommittees Eosinophilic esophagitis in children and adults: a systematic review and consensus recommendations for diagnosis and treatment. Gastroenterology 2007. October;133(4):1342-1363 Epub 2007 Aug 8 [PubMed]
7. Farrell RJ, Kelly CP. Celiac sprue. N Engl J Med. 2002;346(3):180-188 [PubMed]
8. Grisolano SW, Oxentenko AS, Murray JA, Burgart LJ, Dierkhising RA, Alexander JA. The usefulness of routine small bowel biopsies in evaluation of iron deficiency anemia. J Clin Gastroenterol. 2004;38(9):756-760 [PubMed]
9. Rostom A, Dubé C, Cranney A, et al. The diagnostic accuracy of serologic tests for celiac disease: a systematic review. Gastroenterology 2005;128(4, suppl 1):S38-S46 [PubMed]
10. Decker GA, Swain JM, Crowell MD, Scolapio JS. Gastrointestinal and nutritional complications after bariatric surgery. Am J Gastroenterol. 2007. November;102(11):2571-2580 Epub 2007 Jul 19 [PubMed]
11. Schiller LR. Evaluation of small bowel bacterial overgrowth. Curr Gastroenterol Rep. 2007;9(5):373-377 [PubMed]
12. Allen JH, de Moore GM, Heddle R, Twartz JC. Cannabinoid hyperemesis: cyclical hyperemesis in association with chronic cannabis use. Gut 2004;53(11):1566-1570 [PMC free article] [PubMed]
13. Chang YH, Windish DM. Cannabinoid hyperemesis relieved by compulsive bathing. Mayo Clin Proc. 2009;84(1):76-78 [PMC free article] [PubMed]
14. Chepyala P, Olden KW. Cyclic vomiting and compulsive bathing with chronic cannabis use. Clin Gastroenterol Hepatol. 2008. June;6(6):710-712 Epub 2008 May 5 [PubMed]
15. Khurana A, McLean L, Atkinson S, Foulks CJ. The effect of oral sodium phosphate drug products on renal function in adults undergoing bowel endoscopy. Arch Intern Med. 2008;168(6):593-597 [PubMed]
16. Markowitz GS, Stokes MB, Radhakrishnan J, D'Agati VD. Acute phosphate nephropathy following oral sodium phosphate bowel purgative: an unrecognized cause of chronic renal failure. J Am Soc Nephrol. 2005. November;16(11):3389-3396 Epub 2005 Sep 28 [PubMed]

Articles from Mayo Clinic Proceedings are provided here courtesy of The Mayo Foundation for Medical Education and Research