Heron E, Chatellier G, Billaud E, Foos E, Plouin P-F
The urinary metanephrine-to-creatinine ratio for the diagnosis
Ann Int Med
(Aug) 125:300-303 1996
Pheochromocytoma is distinctly rare, being less common than
renovascular hypertension, primary hyperaldosteronism and Cushing's
syndrome. Despite this, it is potentially fatal. The most common
symptoms attributable to pheochromocytoma are headache, excess
sweating and palpitations. Hypertension, either persistent or
intermittent is usually present, and multiple endocrine neoplasia
(types 2a and 2b) must be considered in all cases.
To make the diagnosis, it is necessary to perform assays of
catecholamines or their metabolites, possibly perform provocative
testing and do imaging studies. Traditionally, the use of urinary
metanephrines and plasma catecholamines have been used to screen for
the disease because they have high sensitivity and specificity.
Urinary VMA is much less sensitive and hardly used today. The authors
of this paper tried to improve the value of using urinary
metanephrines in the diagnosis of pheochromocytoma by indexing urinary
metanephrine excretion to that of creatinine. This tends to account
for errors in timing of collection and authors hypothesized that it
would improve the ability of the test to screen for pheochromocytoma.
The study was done in a hypertension referral practice in France over
a one year period. Of a total of 1831 patients, they performed
screening on 1031 (56%) patients whom they considered at risk for
pheochromocytoma (a large proportion by any criteria). They measured
urinary metanephrine and creatinine levels. They subsequently did
imaging studies on subjects with elevated levels and followed up on
slightly elevated levels.
Of the 1031 patients screened, 20 (2%) had a pheochromocytoma. Of
these 20 patients, one had a normal urinary metanephrine level but had
a high metanephrine to creatinine ratio. Of the 993 patients without
pheochromocytoma, 38 had positive urinary findings. Fifteen had high
metanephrine levels, 16 had high metanepgrine to creatinine ratios
while 7 had both elevated. Pheochromocytoma was excluded in this group
of 38 patients by repeat biochemical studies and in some, imaging
studies. Their overall results are shown below.
Urinary met Urinary met/creat
sensitivity (%) 95 100
specificity (%) 98 98
positive pred value (%) 46 47
negative pred value (%) 100 100
Hence the negative predictive value and sensitivity of the test is
excellent. Such a performance would be expected for any test used to
screen for a potential fatal disease (i.e., false negative tests
should be minimal while false positive tests are not as criticial).
As discussed by Emmanuel Bravo in the accompanying editorial, these
tests are useful but the clinician must be aware of their limitations.
Furthermore, the ability of these tests to screen patients with other
diseases which mimic pheochromocytoma is unknown. The use of the
urinary metanephrine to creatinine ratio appears to perform similarly
to urinary metanephrine in screening for this rare disorder. Another
recent paper by
Lenders et al
(Ann Intern Med 1995;123:101-109) proposed that plasma
metanephrine levels are more sensitive than measuring plasma
catecholamines or urinary metanephrines. In that paper, however
urinary metanephrine was measured by a colorimetric assay while in the
present paper liquid chromotography was used. For the practitioner who
suspects the diagnosis, the measurement of urinary metanephrines on a
24 hour sample or plasma catecholamine levels remain the tests of
choice. When the normal values for the urinary metanephrine to
creatinine ratio are better defined in the United States, then this
test may also become useful.
(George Mansoor, M.D., University of Connecticut)