64
ESR and CRP may be useful to monitor clinical activity in certain patients with
RA [ 32 ]; however, at least 40 % of new patients with RA identifi ed from clinical
settings over the last 20 years have normal values for each of these tests at presenta-
tion [ 33 – 36 ]. Furthermore, improvement in clinical status may be accompanied by
continued abnormal, and often unchanged, ESR, as well as CRP [ 37 ]. A larger pro-
portion of patients had abnormal values for ESR in earlier periods [ 38 ], and a
decline in the likelihood of an abnormal ESR or CRP may refl ect improved status of
RA patients over the last few decades [ 39 ]. One might suggest that the purpose of
monitoring ESR and/or CRP is as much to recognize a possible infection or malig-
nancy in an individual patient with RA as to recognize a fl are of disease activity.
In SLE , laboratory evidence does appear required, as no characteristic clinical
fi nding such as joint swelling in RA is seen in all patients. However, while the initial
screening test for antinuclear antibodies (ANA) is positive in >99 % of patients with
SLE (perhaps ANA-negative SLE exists, perhaps not), it also is positive in at least
10–15 % of women in the general population [ 40 ]. SLE occurs in about 0.05 % of
the population; therefore, about 1 in 100 people with a positive ANA has SLE, an
important problem when a non-rheumatologist (and sometimes a rheumatologist)
orders an ANA as a “screening test” [ 41 , 42 ] for individuals with any specifi c or
nonspecifi c musculoskeletal problem. A “positive ANA” is a frequent basis for a
referral to a rheumatologist, usually with no clinical fi ndings to support a diagnosis
of SLE. Nonetheless, many erroneous “diagnoses” of SLE are made in people who
have musculoskeletal symptoms and a positive ANA test.
A positive test for DNA antibodies was found in 75 % of SLE patients in 1969,
and reduction of levels of these antibodies (and ESR), along with increases in serum
complement, were recognized in association with clinical improvement [ 43 ].
However, even in the initial report, 25 % of SLE patients had negative tests for DNA
antibodies [ 43 ], a pattern that continues to this day [ 40 , 44 ]. This phenomenon is
similar to fi ndings with rheumatoid factor and ACPA, and has been overlooked in
expectation that better methods would unearth these antibodies in all patients or
other “diagnostic” biomarkers might be discovered, but that has not been the case
over more than fi ve decades.
A number of ANA subsets have been described to be associated with different
features of SLE, such as associations of anti-Smith (Sm) with renal disease and anti-
ribonucleoprotein (RNP) with mixed connective tissue disease (MCTD) [ 45 ].
However, none of these tests are more than 70 % specifi c to any clinical fi ndings or
particular diagnosis to affect clinical decisions, which are made on the basis of clini-
cal fi ndings [ 46 ]. Indeed, a follow-up study indicated that almost half of the initial
MCTD patients appeared to have different diagnoses, including RA or systemic
sclerosis, on subsequent evaluation years later [ 47 ]. This phenomenon is seen in
most rheumatic disease databases over a decade or longer, in which at least 5 % of
patients have a different primary diagnosis. Tests of ANA subsets have important
value for research concerning pathogenesis and treatment, as seen for rheumatoid
factor in RA, but appear to add expense without value in routine clinical care.
One reason that biomarkers such as laboratory tests and radiographs are empha-
sized in diagnosis and monitoring involves the biomedical model paradigm, as
noted previously, which is highly effective in many diseases [ 1 ]. However, a more
T. Pincus et al.