This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Meikle, A. W. Articles by Terry, A. H. Search for Related Content PubMed PubMed Citation Articles by Meikle, A. W. Articles by Terry, A. H.

Pseudo-Cushing Syndrome Caused by Fenofibrate Interference with Urinary Cortisol Assayed by High-Performance Liquid Chromatography

Departments of Medicine and Pathology (A.W.M.), University of Utah, Salt Lake City, Utah 84132; Endocrine-Diabetes Center (J.F.), St. Luke’s Medical Center, Medical College of Wisconsin, Milwaukee, Wisconsin 53215; and Associated Regional and University Pathologists Institute for Clinical and Experimental Pathology (M.M.K., A.L.R., G.J.N., A.H.T.), Salt Lake City, Utah 84108

Address all correspondence and requests for reprints to: A. Wayne Meikle, M.D., Associated Regional and University Pathologists, 500 Chipeta Way, Salt Lake City, Utah 84108. E-mail: wayne.meikle{at}hsc.utah.edu.

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Urinary free cortisol (UFC) excretion over 24 h reflects the production rate of cortisol and is used commonly in the diagnosis of Cushing syndrome. We report on two patients evaluated for Cushing syndrome who had elevated UFC when analyzed by HPLC but normal values for the analysis performed by RIA and HPLC-mass spectrometry/mass spectrometry (HPLC-MS/MS). Other laboratory testing was inconsistent with the diagnosis of Cushing syndrome and raised doubts about the diagnosis. We identified a probable cause of analytical interference as coming from fenofibrate (Tricor), medication taken by the patients. Fenofibrate peak overlapped with the HPLC peak of cortisol and produced an MS/MS transition overlapping the major transition of cortisol. A second MS/MS transition was free from interference. In summary, fenofibrate administration may cause false elevation of UFC values determined by HPLC or HPLC-MS/MS in patients evaluated for Cushing syndrome. An HPLC-MS/MS method using multiple mass transitions, rather than a single transition, allows accurate quantitation of urinary cortisol in patients taking fenofibrate.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
ENDOGENOUS CUSHING SYNDROME is caused by the overproduction of cortisol throughout the day (1, 2, 3, 4, 5, 6). The measurement of 24-h excretion of urinary cortisol is the most reliable, practical assessment of cortisol secretion because cortisol excretion rises disproportionately as the plasma cortisol concentration exceeds the binding capacity of cortisol-binding globulin (6). Most physicians consider urinary free cortisol (UFC) the screening test of choice for the diagnosis of endogenous Cushing syndrome (2, 3, 7). Several methods are available for measurement of UFC, including RIAs, HPLC, and (more recently) HPLC-mass spectrometry/mass spectrometry (HPLC-MS/MS).

Though many RIA methods are useful for the diagnosis of Cushing syndrome, they are known be less specific for UFC than HPLC or HPLC-MS/MS (8, 9, 10). This is why HPLC became a method of choice for specific measurement of UFC; and, more recently, some laboratories have switched to HPLC-MS/MS. We report on two cases of patients evaluated for Cushing syndrome in which UFC was elevated by HPLC; however, salivary cortisol and dexamethasone suppression test results were inconsistent with the diagnosis of Cushing syndrome. Furthermore, testing with RIA and HPLC-MS/MS did not confirm the elevated UFC values. We also found that fenofibrate (Tricor, Abbott Laboratories, North Chicago, IL) produces a peak that overlaps with the peak of cortisol on reverse-phase HPLC column and produces an MS/MS transition similar to the major MS/MS transition of cortisol when analyzed by HPLC-MS/MS. A second MS/MS transition was found to be free of interference by fenofibrate when cortisol was analyzed by HPLC-MS/MS.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Case 1

KM is a 57-yr-old woman referred for evaluation and management of suspected spontaneous Cushing’s syndrome. She complained of lifelong history of obesity and had gained approximately 100 pounds over the past decade, with increasing facial rounding and plethora. She had an 8-yr history of diabetes mellitus, with suboptimal glycemic control on oral hypoglycemic agent therapy. She had recently started medication for hypertension and hyperlipidemia. She also had primary hypothyroidism and has been on thyroid hormone replacement. There has been no history of any neuropsychiatric problems, fractures, kidney stones, cardiovascular disease, or use of any exogenous steroids. Her family history was remarkable for diabetes mellitus in her mother and hypothyroidism in her father. She did not smoke cigarettes and rarely used alcohol.

Her medications included glyburide (5 mg daily), simvastatin (10 mg daily), L-thyroxine (100 µg daily), conjugated estrogens (0.625 mg daily), fenofibrate (200 mg daily), metformin (1000 mg daily), and aspirin (81 mg daily).

Physical examination showed a Cushingoid-appearing, 57-yr-old woman whose blood pressure was 170/60; pulse, 80 beats/minute; height, 65.3 inches; and weight, 295 pounds (yielding a body mass index of 48.8). There was evidence of acanthosis nigricans on her neck and elbows. She had facial rounding and plethora and some increased supraclavicular fullness. There were no abnormal eye findings, her muscle strength was good, and a trace of pretibial edema was evident.

Her biochemical evaluation initially included a baseline 24-h urine free cortisol (measured by HPLC) of 166 µg/24 h (normal < 42). A low-dose 2-d dexamethasone suppression test (0.5 mg dexamethasone every 6 h for 2 d) was performed. The 24-h urine free cortisol after the dexamethasone was 218 µg/24 h.

A magnetic resonance imaging of the pituitary showed an empty sella. No discrete pituitary tumor was identified. The patient underwent repeat measurements of urine free cortisol, by an HPLC-UV-based method, results of which were 135 µg/24 h and 463 µg/24 h.

Over a 3-month period, the patient underwent several measurements of late-night (2300 h) salivary cortisol determinations (Table 1). These measurements were all well within the normal range and seemed quite discordant with the urine free cortisol determinations. An overnight 1-mg dexamethasone suppression test yielded a serum cortisol of 0.9 µg/dl with a plasma ACTH of 3 pg/ml.

Case 2

CS is a 39-yr-old woman referred for evaluation of possible Cushing’s syndrome. The patient had a long history of increasing central weight gain associated with impaired glucose tolerance and dyslipidemia. Her course had been complicated by hypertension, hyperlipidemia, polycystic ovary syndrome, obstructive sleep apnea, and bilateral carpal tunnel syndrome. There was a family history of coronary artery disease and hypertension. There was no history of alcohol use or cigarette smoking.

Her medications included fenofibrate (160 mg daily), simvastatin (20 mg daily), cetirizine (10 mg daily), bupropion (300 mg twice daily), setraline (200 mg daily), aspirin (81 mg daily), metformin (500 mg daily), glyburide, L-thyroxine, glucosamine, calcium citrate, vitamin E, and a multivitamin and pantoprazole (20 mg daily).

Physical examination revealed an obese, slightly Cushingoid-appearing woman with a blood pressure of 132/84; pulse, 80; and weight, 312 pounds. The patient had a body mass index of 44.8. She did have some facial rounding and mild plethora. There was some increase in supraclavicular fullness and facial hirsutism. She had good muscle strength, and there was no peripheral edema.

Table 1 outlines her laboratory studies over a 1-yr period of time. She had persistent elevations of urine free cortisol [UFC measured by HPLC, ranging from 80–180 µg/24 h (normal < 42)]. Over the same period of time, several 2300-h salivary cortisol measurements were all well within the normal range. An overnight 1-mg dexamethasone suppression test yielded a serum cortisol of 1.1 µg/dl.

Finally, a 24-h urine cortisol was determined using HPLC-MS/MS. The level was now well within the normal range, at 9 µg/24 h (11).

HPLC-MS/MS method

The HPLC-MS/MS analysis was performed by a method (11) using on-line sample preparation. Briefly, urine samples were aliquoted in autosampler vials, internal standard (d₄-cortisol) was added, and the vials were vortexed and centrifuged. The sample was injected into the HPLC equipped with a Luna C18 column 50 mm x 2 mm, 5-µm particles (Phenomenex, Torrance, CA). Sample purification was performed online using a C18 guard cartridge (Phenomenex). Initial mobile phase composition was 7:93 methoanol/5 mM ammonium formate. After 0.5 min, the mobile phase composition was changed to 55/45 (methanol/5 mM ammonium formate), and column effluent was directed into the tandem mass spectrometer (API 2000 Applied Biosystems, MDS Sciex, Foster City, CA). Protonated cortisol (m/z 363) was used as a parent ion, with two product ions, m/z 121 and m/z 97. The parent ion for the internal standard was m/z 367, with the same two product ions (m/z 121 and m/z 97). Calibration standards were prepared at 10, 50, 100, and 200 µg/liter in synthetic urine (11, 12). The 50-µg/liter calibrator also served to set a branching ratio for the m/z 97 and 121 product ions. Normally, the m/z 121 ion was used for quantitation; however, if the branching ratio was too low (an indication of an interference in the transition m/z 121), the m/z 97 ion was used for quantitation.

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Both described cases had clinical signs consistent with Cushing syndrome and had urinary cortisol values determined by HPLC supporting the diagnosis. However, other laboratory tests of Cushing syndrome, such as late-evening salivary cortisol and dexamethasone suppression testing, were inconsistent with Cushing syndrome. When UFC is measured by HPLC, cortisol peak may overlap with an interfering peak. HPLC is an excellent test for UFC; however, if the cortisol peak is not sufficiently resolved from interference, this may lead to overestimation of the cortisol concentration, giving falsely elevated values for cortisol. Though HPLC with UV detection is a commonly accepted test for UFC, such methods are not fully specific (13). Hence, there is potential for interference by compounds having the same retention time. Cortisol measurement by HPLC-MS/MS makes it possible to detect inferring substances and allows accurate quantitation of cortisol. A considerable improvement in specificity is possible by using HPLC-MS/MS because the identification relies on more specific properties (retention time, parent ion mass, and product ion mass) rather than less specific properties used by HPLC-UV. As shown in Fig. 1, the retention time of fenofibrate and its metabolites under our HPLC conditions was within 10% of the retention time of cortisol. Furthermore, the front of the fenofibrate peak overlapped cortisol. Thus, it is likely that under different HPLC conditions, fenofribrate or its metabolites are likely to coelute with cortisol and interfere with the results, particularly if an HPLC method is not optimized to exclude this possibility. The concentrations of UFC for both cases were significantly different between the methodologies used. The discordant results between samples analyzed by HPLC and those analyzed by HPLC-MS/MS, as well as other inconsistencies already noted, raised the strong possibility that the HPLC results were likely to be subject to interference.

View larger version (26K): [in this window] [in a new window]   FIG. 1. HPLC-MS/MS MRM chromatogram of (A) typical patient sample and (B) sample from patient (case 1) showing Fenofibrate interference.

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

UFC measured by RIA, after chromatographic isolation of cortisol, provides reliable testing but is time consuming (2). RIA without isolation of cortisol is less reliable because of cross-reacting substances to the antibody (2, 4). HPLC-MS and HPLC-MS/MS allow significant improvement in specificity and speed of the analysis (9, 11, 13). Despite the significant improvement in specificity, they may not be interference free.

In the two patients described, HPLC produced elevated results of UFC, whereas results by HPLC-MS/MS were within normal range. This led us to a search of the interfering substance(s). Both patients were on multiple medications, including fenofibrate. Fenofibrate is a lipid-lowering drug containing a chlorine atom within the structure. The drug is an ester of fenofibric acid and isopropyl alcohol. Fenofibrate metabolism involves partial hydrolysis, with release of fenofibric acid that subsequently forms glucuronide conjugate.

Fenofibrate presents an interesting analytical challenge because, not only is the retention time of this compound similar to the retention time of cortisol under some HPLC conditions on reverse phase (C18) LC columns, but its molecular mass is only 2 Da lower than that of cortisol. However, fenofibrate structure contains a chlorine atom; hence, its mass spectrum contains a strong (30%) isotope peak two Da higher than the nominal molecular mass of fenofibrate. This isotope is an isobar of cortisol and produces the same molecular ion as the one selected for the MS/MS analysis. Furthermore, product mass spectra of the fenofibrate m/z 363 molecular ion also produce a strong fragment at m/z 121, a potentially very troubling interference. However, fenofibrate does not generate an m/z 97-product ion peak. By using the m/z 97 ion for quantitation (for those samples whose branching ratio of the 121 to 97 peak is high, compared with the standard), we can eliminate interference from fenofibrate. As we confirmed, the above results suggest fenofibrate or its metabolites were the actual interfering compounds in these two samples when analyzed by HPLC. The single-stage electrospray mass spectrum of the interfering peaks from the two patients’ samples was identical to that of fenofibrate, including the distinctive chlorine isotope pattern. The mass spectrum was also identical to that from a urine sample from a patient who was only on fenofibrate but not suspected of having Cushing syndrome. Carbamazepine has also been shown to interfere with UFC measurements by HPLC (9, 10). Because the molecular weight of carbamazepine is much different from cortisol, it would not be expected to interfere with assays performed by HPLC-MS/MS methods (10).

This interference with UFC by fenofibrate could lead to a false diagnosis of Cushing syndrome. Clinicians using HPLC for measurement of UFC should be cautious of the results if the patient is taking fenofibrate. If there is doubt about the results of an HPLC test for cortisol, an HPLC-MS/MS method using multiple transitions specific to cortisol can be used for the analysis. HPLC-MS/MS circumvents this problem because two independent transitions can be used for quantitation of cortisol. Whereas one transition interfered with fenofibrate, the other did not and allowed obtaining an accurate cortisol concentration.

	Footnotes

 
Abbreviations: MS/MS, Mass spectrometry/mass spectrometry; UFC, urinary free cortisol.

Received February 13, 2003.

Accepted April 14, 2003.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Liddle GW 1960 Tests of pituitary-adrenal suppressibility in the diagnosis of Cushing’s syndrome. J Clin Endocrinol Metab 20:1539–1559
2. Meikle AW, West CD 1986 Laboratory tests for the diagnosis of Cushing’s syndrome and adrenal insufficiency and factors affecting those tests. In: DeGroot L, Cahill G, Potts J, Nelson DH, Winegard A, Martini L, eds. Endocrinology. New York: Grune, Stratton, Inc.; 1157–1178
3. Mengden T, Hubmann P, Muller J, Greminger P, Vetter W 1992 Urinary free cortisol versus 17-hydroxycorticosteroids: a comparative study of their diagnostic value in Cushing’s syndrome. Clin Investig 70:545–548[Medline]
4. Meikle AW 1993 A diagnostic approach to Cushing’s syndrome. Endocrinologist 3:311–320
5. Findling JW, Doppman JL 1994 Biochemical and radiologic diagnosis of Cushing’s syndrome. Endocrinol Metab Clin North Am 23:511–537[Medline]
6. Meikle AW, Takiguchi H, Mizutani S, Tyler FH, West CD 1969 Urinary cortisol excretion determined by competitive protein-binding radioassay: a test of adrenal cortical function. J Lab Clin Med 74:803–812[Medline]
7. Contreras LN, Hane S, Tyrrell JB 1986 Urinary cortisol in the assessment of pituitary-adrenal function: utility of 24-hour and spot determinations. J Clin Endocrinol Metab 62:965–969[Abstract]
8. McWhinney BC, Ward G, Hickman PE 1996 Improved HPLC method for simultaneous analysis of cortisol, 11-deoxycortisol, prednisolone, methylprednisolone, and dexamethasone in serum and urine. Clin Chem 42:979–981[Free Full Text]
9. Turpeinen U, Markkanen H, Valimaki M, Stenman UH 1997 Determination of urinary free cortisol by HPLC. Clin Chem 43:1386–1391[Abstract/Free Full Text]
10. Findling J, Pinkstaff S, Shaker JL, Raff H, Nelson J 1998 Pseudohypercortisoluria: spurious elevation of urinary cortisol due to carbamazepine. Endocrinologist 8:51–54
11. Kushnir M, Rockwood A, Nelson G, Terry A, Meikle AW 2003 Urinary free cortisol analysis by LC-MS/MS. Clin Chem 49:965–967[Free Full Text]
12. Henry J 1996 Clinical diagnosis and management by laboratories. 19th ed. Philadelphia: Saunders Company
13. Taylor RL, Machacek D, Singh RJ 2002 Validation of a high-throughput liquid chromatography-tandem mass spectrometry method for urinary cortisol and cortisone. Clin Chem 48:1511–1519[Abstract/Free Full Text]
14. Mengden T, Hubmann P, Muller J, Greminger P, Vetter W 1992 Urinary free cortisol versus 17-hydroxycorticosteroids: a comparative study of their diagnostic value in Cushing’s syndrome. Clin Investig 70:545–548
15. Nieman LK 2002 Diagnostic tests for Cushing’s syndrome. Ann NY Acad Sci 970:112–118[Abstract/Free Full Text]
16. Flack MR, Oldfield EH, Cutler Jr GB, Zwerg MH, Malley JD, Chrousos GP, Loriaux DL, Nieman LK1992 Urine free cortisol in the high-dose dexamethasone suppression test for the differential diagnosis of the Cushing syndrome. Ann Intern Med 116:211–217
17. Kaye TB, Crapo L 1990 The Cushing syndrome: an update on diagnostic tests. Ann Intern Med 112:434–444
18. Findling JW, Raff H 1999 Newer diagnostic techniques and problems in Cushing’s disease. Endocrinol Metab Clin North Am 28:191–210[CrossRef][Medline]

This article has been cited by other articles:

				L. K. Nieman, B. M. K. Biller, J. W. Findling, J. Newell-Price, M. O. Savage, P. M. Stewart, and V. M. Montori The Diagnosis of Cushing's Syndrome: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1526 - 1540. [Abstract] [Full Text] [PDF]

				J. W. Findling and H. Raff Cushing's Syndrome: Important Issues in Diagnosis and Management J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3746 - 3753. [Abstract] [Full Text] [PDF]

				M. M. Kushnir, A. L. Rockwood, W. L. Roberts, E. G. Pattison, W. E. Owen, A. M. Bunker, and A. W. Meikle Development and Performance Evaluation of a Tandem Mass Spectrometry Assay for 4 Adrenal Steroids Clin. Chem., August 1, 2006; 52(8): 1559 - 1567. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Meikle, A. W. Articles by Terry, A. H. Search for Related Content PubMed PubMed Citation Articles by Meikle, A. W. Articles by Terry, A. H.