Repeated ANA testing after negative
result provides little diagnostic value
Repeated antinuclear antibody testing after a negative result has limited use for the diagno- sis of ANA-associated rheumatic conditions, according to data from
a multicentre, retrospective analysis
that considered a 7-year period.
For more than 7,875 repeated ANA
tests in 4,887 patients, “the vast majority of results didn’t change,” Dr. Ai
Li Yeo, a PhD candidate, rheumatologist, and infectious disease fellow at
Monash University, Melbourne, reported at the European Congress of
ANA tests were repeated between
2 and as many as 45 times in individual patients, she reported, but
the results of 79% of these tests
remained unchanged – 45% of tests
were persistently negative and 34%
persistently positive using a cutoff
titer of 1:160.
“Our study showed that there was
a very low yield in repeating an ANA
test for the diagnosis of ANA-associated rheumatologic conditions unless
there was evidence of evolving multi-system clinical features,” Dr. Yeo said.
Indeed, the positive predictive value was just 0.01. “So for a hundred
patients starting off with a negative
ANA result that on repeat testing
became positive, the probability is
that one patient will have a new ANA-associated rheumatic condition diagnosis,” Dr. Yeo said.
“ANA testing is frequently performed and is part of the classification criteria for autoimmune
conditions such as lupus and scleroderma,” she observed. However, the
test provides no information on the
severity or activity of the disease,
and the value of serial monitoring for
such conditions is unclear.
“Minimising unnecessary tests is
a global health economic priority,” Dr.
Yeo said. She noted that there are
multiple initiatives in place to try to
open a dialog about using healthcare
resources most effectively, such as
“Choosing Wisely” set up by the
American Board of Internal Medicine
The aim of the present analysis
was to calculate the cost of repeated
ANA testing and to see if any change
in the ANA result was associated
with new diagnoses of ANA-associated rheumatic conditions.
The analysis considered more than
36,700 tests that were performed
on samples from more than 28,800
patients within the Monash Health
tertiary health network between 2011
and 2018. Of these, 22,657 (62%) had
given a negative result and 14,058
(38%) had given a positive result.
“Not surprisingly, the age of those
who tested positive was significantly higher than those who tested
negative,” Dr. Yeo said (52.6 vs. 48.9
years; P less than .001). There was
also a higher number of women than
men tested, and women more often
Around one-fifth of tests performed
were repeat tests, of which 511 ( 6.5%)
changed from being negative to positive over a median of 1.71 years.
“A small percentage of people
alternated between results,” Dr. Yeo
acknowledged, with 9.4% of people
going from a positive to a negative
result, 10.5% moving from a negative
to a positive result, and 1.9% going
from positive to negative to positive.
With repeated tests, just five new
diagnoses of ANA-associated rheu-
matic conditions were made: two
cases of systemic lupus erythema-
tosus, one case of scleroderma, and
two cases of undifferentiated con-
nective tissue disease. There was a
range of ANA titers and patterns and
evolving clinical features of a multi-
Based on the direct costs of ANA
testing in her health care system,
not performing repeated tests could
yield significant savings, Dr. Yeo said,
a 21.4% reduction, in fact, based
on this analysis. The cost of an ANA
test in Australia ranges from 15 to 46
euros, making the cost of all tests in
this analysis 564,745 euros. Taking
away the cost of all the single ANA
tests performed (443,209 euros)
gives a potential cost saving of more
than 121,000 euros, she said.
“We now have an opportunity to
prevent unnecessary ANA testing, Dr.
Yeo said. “Ultimately, our aim is to
change behaviour at the start of the
ordering cycle by educating medical
students and doctors about inappro-
priate test ordering.”
The majority of repeated tests had
been ordered by nonrheumatologists
(82% of cases), and Dr. Yeo said that
rheumatologists ordered repeat tests
in 11% of cases. However, there
was little information available in this
retrospective analysis as to why the
tests had been repeated.
The research was picked as one of
the six best clinical abstracts at the
congress, out of a total of almost
5,000 submitted abstracts.
Dr. Yeo reported having no conflicts
Dr. Ai Li Yeo