J
Clin
Pathol
1982;35:1118-1121
Antinuclear
antibody-negative
systemic
lupus
erythematosus-how
common?
KC
McHARDY,*
CHW
HORNE,
JAN
RENNIE,t
From
the
University
Departments
of
Therapeutics
and
*Clinical
Pharmacology
and
tPathology,
Foresterhill,
Aberdeen
and
the
tDivision
of
Rheumatology,
City
Hospital,
Aberdeen
SUMMARY
A
review
of
five
years'
DNA-binding
antibody
results
in
a
routine
service
laboratory
revealed
38
patients
who
had
high
DNA-binding
capacity
(DNA-bc)
but
no
antinuclear
antibodies
(ANA).
On
retrospective
case
note
analysis,
22
patients
(58%)
were
thought
to
have
systemic
lupus
erythematosus
(SLE),
although
only
six
(16%)
fulfilled
the
preliminary
classification
criteria
of
the
American
Rheumatism
Association
(ARA).
Our
findings
indicate
that
ANA-negative
SLE
is
commoner
than
generally
realised
and
lead
us
to
recommend
the
measurement
of
DNA-bc
in
every
case
where
clinically
appropriate.
Current
clinical
teaching
suggests
that
ANA
are
present
in
at
least
98%
of
cases
of
SLE,'
2
and
that
the
measurement
of
DNA-bc
is
more
appropriate
as
a
secondary,
specific
diagnostic
test
on
ANA-positive
sera.
The
usefulness
of
the
ANA
test
as
a
screening
procedure
for
the
detection
of
sera
with
high
DNA-bc
has
recently
been
evaluated3
and
its
worth
established
in
this
context.
However,
it
is
possible
that
some
cases
of
SLE
which
are
not
associated
with
ANA
will
be
missed
if
the
ANA
test
alone
is
carried
out.3
Antinuclear
antibody-negative
SLE
has
been
implied,
if
not
directly
considered,
in
many
studies
of
SLE
over
the
years.46
Two
series
citing
ten7
and
eight8
cases,
respectively,
of
ANA-negative
SLE
have
drawn
more
direct
attention
to
this
entity
and
Maddison
et
al9
recently
presented
data
on
66
cases
of
SLE
with
serological
abnormalities
but
negative
ANA
tests.
In
each
study
the
cases
were
collected
over
a
number
of
years
as
being
atypical
members
of
larger
series
of
ANA-positive
lupus.
The
present
study
approaches
ANA-negative
SLE
from
a
different
angle
in
beginning
with
an
abnormal
laboratory
finding.
Review
of
DNA-bc
results
since
the
inception
of
the
assay
in
1976
as
a
routine
laboratory
test
in
Aberdeen
showed
a
number
of
sera
with
markedly
raised
titres
of
antiDNA
antibodies,
but
no
detectable
ANA.
A
high
DNA-bc
is
probably
the
most
specific
laboratory
test
for
diagnosing
SLE
"
"
although
El-Ghobarey's
study3
showed
that
only
87%
of
cases
with
DNA-bc
greater
than
30
units
(U)/ml
had
SLE.
We
present
here
our
findings
Accepted
for
publication
24
February
1982
1118
following
a
retrospective
case
note
analysis
on
38
patients
who
demonstrated
a
high
DNA-bc
in
the
absence
of
ANA.
Subjects
and
methods
The
subjects
consisted
of
38
patients
who
had
been
in
the
care
of
various
clinicians
in
Grampian
Health
Board
hospitals
and
were
therefore
drawn
from
a
population
of
approximately
500
000.
Ages
ranged
from
two
to
84
yr,
(xi
=
50
yr).
Twenty-two
were
female.
All
subjects
had
consistently
negative
ANA
tests
(x
=
4*
1
tests/subject),
and had
a
recorded
DNA-bc
of
40
U/ml
or
greater.
The
DNA-bc
assay
had
been
performed
one
to
eight
times
per
patient
(x
=
2-
64)
and
was
greater
than
40
U/ml
on
one
to
four
occasions
per
patient
(xi
=
1
77).
Fourteen
patients
had
a
single
result
over
40
U/ml;
the
assay
had
not
been
repeated
in
five
of
these.
Antinuclear
antibody
tests
were
carried
out
by
the
method
of
Beck
12
using
an
indirect
immuno-
fluorescence
technique
on
a
substrate
of
rat
liver'
slices
at
a
serum
dilution
of
1/16.
The
DNA-binding
capacities
were
measured
using
the
antiDNA
kit
(RC
Amersham)
based
on
the
method
of
Wold
et
al,
3
which
is
a
specific
radioimmunoassay
using
ammonium
sulphate
precipitation
of
'251-labelled,
double-stranded
DNA.
The
manufacturers
suggest
a
normal
range,
subject
to
interlaboratory
variation,
of
0-10
U/ml,
and
25
U/ml
is
the
level
above
which
the
diagnosis
is
"largely
confined"
to
SLE.
14
After
initial
perusal
of
the
case
histories
we
eliminated
those
in
which
there
was
insufficient
Antinuclear
antibody-negative
systemic
lupus
erythematosus-how
common?
evidence
to
support
a
diagnosis
of
SLE.
The
clinical
details
in
the
remainder
were
then
compared
to
the
American
Rheumatism
Association
(ARA)
preliminary
classification'5
criteria
for
SLE,
substituting
a
high
DNA-bc
for
positive
LE
cell
test
and
"Albustix
(Ames)++
"
for
"proteinuria
>
3
5g/
day".
The
14
ARA
criteria
were
devised
to
allow
more
objective
classification
of
patients
for
the
purpose
of
clinical
trials,
population
surveys
etc,
and
a
patient
who
exhibits
four
or
more
criteria,
simultaneously
or
serially,
is
said
to
have
SLE.
Their
use
is
widespread
but
of
limited
value
in
that
many
well
established
features
of
SLE
are
completely
omitted
while
others
appear
in
a
very
precise
and
over-restrictive
way.
It
has
been
repeatedly
stressed
that
fulfilment
of
four
ARA
criteria
is
not
necessary
to
make
a
diagnosis.'6
To
allow
a
clinically
more
comprehensive
assessment
of
our
cases
we
further
modified
the
original
ARA
list
to
that
shown
in
Table
1,
and
compared
the
clinical
details,
in
turn,
to
these
amended
criteria.
Table
1
Expanded
list
of
SLE
features
based
on
ARA
list
*1
Rash
(inci
butterfly,
livedo
9
False
+
ve
syphilis
serology
reticularis,
vasculitis).
*
10
Albustix
+ +
2
Discoid
lupus
11
Granular
urinary
casts
3
Raynaud's
phenomenon
12
Pleurisy/pericarditis
4
Alopecia
*
13
Neuropsychiatric
disorder
5
Photosensitivity
14
Cytopenia
6
Oral/nasal
ulceration
*
15
Fever
7
Arthritis
*
16
Ocular
disease
*8
DNA-bc
>
40U/ml
*Denotes
changes
from
original
ARA
list.
Table
2
Clinicalfeatures
of
SLE
in
22
patients
DNA-bc
>
40U/ml
22
(100%)
Raynaud's
3
(14)
Arthfitis
15
(68)
Albustix
++
3
(14)
Neuropsychiatric
disorder
14
(64)
Photosensitivity
2
(9)
Haematological
disorder
8
(36)
Alopecia
2
(9)
Fever
8
(36)
Granular
casts
2
(9)
Pleurisv/pericarditis
7
(32)
Oral/nasal
ulcers
1(5)
Ocular
problems
7
(32)
False
+ve
syphilis
-
Rashes
5
(23)
serology
Discoid
lupus
-
Table
3
Clinical
pattern
in
patients
with
only
three
listed
features
of
SLE
Patient
1
2
3
4
5 6
Raynaud's
-
-
+ -
-
-
Arthritis
+
+
+ +
-
Highest
DNA-bc
(U/mI)
54
74
52
57
41
71
Proteinuria
-
-
-
-
-
+
Pleurisy
-
-
+
-
Neuropsychiatric
disorder
-
-
+ +
+
Cytopenia
+ +
Results
Review
of
the
case
notes
suggested
that
22
(58%)
of
our
cases
had
SLE
and
the
clinical
features
in
this
group
are
shown
in
Table
2.
The
remaining
16
cases
did
not
have
enough
evidence
to
support
this
diagnosis.
Comparison
with
the
ARA
list,
modifying
only
the
criteria
regarding
LE
cells
and
proteinuria,
showed
that
only
six
cases
(16%)
had
four
or
more
features
of
SLE,
while
the
remaining
16
(42%)
whom
we
thought
to
have
SLE
had
either
two
or
three
of
the
features
listed.
When
the
clinical
data
on
these
22
cases
were
compared
to
the
expanded
list,
16
had
between
four
and
nine
(3x
=
5
1)
features
of
SLE
and
the
remaining
six
had
three
features
each.
All
in
this
latter
group
had,
by
definition,
markedly
raised
DNA-bc
and
their
other
clinical
features
are
shown
in
Table
3.
Discussion
A
condition
with
manifestations
as
protean
as
SLE
does
not
readily
lend
itself
to
objective
analysis.
The
ARA
criteria
go
part
of
the
way
towards
allowing
comparison
of
different
groups
but
it
is
intellectually
disquieting
to
accept
that
a
patient
with
a
butterfly
rash,
arthritis,
positive
WR
and
raised
DNA-bc
has
the
very
same
disease
as
another
with
alopecia,
Raynaud's
phenomenon,
haemolytic
anaemia
and
recurrent
mouth
ulcers,
while
conversely,
a
third
patient
with
arthritis,
DNA
binding
antibodies,
mild
proteinuria,
mononeuritis,
vasculitic
rash
and
thrombocytopenia
cannot
be
classified
as
a
case
of
SLE.
These
examples
underline
the
need
to
apply
the
ARA
criteria
sensibly
and
to
accept
their
limitations.
In
their
review
of
the
use
of
the
ARA
criteria
Canoso
and
Cohen
16
state
that
a
standard
set
of
diagnostic
(cf
classification)
criteria
for
SLE
is
unlikely
to
appear
until
a
great
deal
more
is
known
of
the
aetiology
and
pathogenesis.
In
trying
to
evaluate
our
cases
we
have
based
assessment
on
the
ARA
criteria
but
have
tried
to
broaden
these
to
increase
their
clinical
application.
Initially
two
of
the
original
criteria
were
modified.
First,
substitution
of
high
DNA-bc,
the
most
specific
laboratory
investigation,
for
the
outdated
LE
cell
test
is
widely
accepted.
Second,
proteinuria
greater
than
3
5
g/day
was
replaced
by
"Albustix++",
representing
approximately
1-5
g/day
of
urinary
protein.
This
semiquantitative
method
was
employed
as
accurate
quantification
of
proteinuria
is
seldom
carried
out;
the
lesser
amount
of
protein
loss
was
chosen
as
considerable
degrees
of
renal
damage
can
occur
without
a
urinary
protein
loss
ever
approaching
the
demands
of
the
ARA
criteria.
This
point
has
been
1119
McHardy,
Horne,
Rennie
previously
raised'6
17
and
is
supported,
amongst
others,
by
Pollak
and
colleagues'8
whose
patients
with biopsy-proven
lupus
nephritis
had
proteinuria
in
excess
of
200
mg/day
in
only
48
of
77
cases.
The
ARA
criteria
with
these
modifications
gave
six
patients
with
the
requisite
four,
or
more,
criteria
and
a
further
16
patients
with
a
formally
inadequate
two
or
three
features.
The
discrepancy
between
clinical
and
classification
assessments
was
clearly
due
to
the
exclusion
from
the
ARA
list
of
a
number
of
well
recognised
features
of
SLE
which
were
exhibited
by
some
of
our
patients.
Fever
has
been
described
in
80-100%1920
of
SLE
patients
and
was
seen
in
a
number
of
ours.
A
range
of
neuropsychiatric
problems
far
wider
than
"psychosis
and
convulsions"
is
now
recognised2'
22
and
it
was
of
particular
interest
that
depression
which
Bennett
and
his
associates23
describe
as
the
commonest
disorder
of
mental
function
in
SLE
led
to
four
of
22
in
our
clinical
SLE
group
requiring
formal
psychiatric
assessment.
A
variety
of
ocular
problems
is
recognised24
and
the
32%
incidence
of
eye
disorders
in
our
patients
compares
well
with
overall
figures
in
previous
studies.20"5
In
assessing
cutaneous
manifestations
we
included,
in
addition
to
facial
erythema,
rashes
such
as
livedo
reticularis,
vasculitis
and
maculopapular
eruptions,
all
of
which
may
be
useful
clinical
pointers
to
a
diagnosis
of
SLE.2026
Using
our
broadened
list
of
clinical
features
of
SLE
(Table
1)
we
now
have
16
patients
with
four
or
more
features
and
a
further
six
(Table
3)
each
with
three
features.
Some
additional
points
of
interest
in
this
latter
group
are:
haematological
disorder
in
patient
1
included
temporally
distinct
thrombocytopenia
and
Coombs'-positive
haemolysis;
patient
2
had
a
spontaneous
abortion
and
a
puerperal
exacerbation
of
arthritis;
patient
3
had
episodes
of
non-specific
chest
pain,
required
sedatives
for
"nerves"
and
had
a
period
of
weight-loss,
diarrhoea
and
abdominal
pain
for
which
extensive
investigation
gave
no
diagnosis;
patient
4
had
recurrent
paraesthesiae
and
a
depressive
illness
culminating
in
a
drug
overdose,
in
addition
to
mild
renal
impairment
for
no
apparent
reason;
patient
5
had
periungual
erythema,
a
diminished
gas
transfer
factor
with
exertional
dyspnoea
and
a
history
of
musculoskeletal
pains;
patient
6
had
recurrent
myalgia
and
stiffness,
headaches,
steroid-responsive
strokes,
telangiec-
tases,
proteinuria
of
up
to
five
grams
per
day
with
impaired
creatinine
clearance
and
a
brother
who
during
the
last
few
months
has
developed
SLE
with
ANA.
Thus
we
support
a
diagnosis
of
SLE
in
22
cases.
What
of
the
other
16
patients
with
high
DNA-bc
but
not
thought
to
have
SLE?
In
some
the
information
in
the
case
notes
was
simply
inadequate
to
allow
any
conclusions.
Three
patients
had
liver
disease,
which can
be
associated
with
high
DNA-bc.27
Some
may
be
evolving
cases
of
SLE
in
whom
sufficient
clinical
features
to
allow
diagnosis
will
only
develop
after
a
number
of
years.
Of
our
six
cases
who
fulfilled
the
ARA
criteria,
only
four
would
have
done
so
at
the
time
of
first
detection
of
a
high
DNA-bc.
Late
development
of
ANA
has
been
highlighted
by
Bohan28
who
described
five
cases
of
active
SLE
developing
ANA
for
the
first
time,
ten
months
to
seven
years
after
the
initial
diagnosis.
From
the
outset
we
eliminated
from
further
study
any
patient
who
had
ever
had
a
positive
ANA
result.
The
38
patients
included
in
the
study
had
remained
ANA-negative
over
follow-up
periods
ranging
from
nine
to
40
months.
Detection
of
cytoplasmic
antibodies
(Ro
and
La)929
may
be
an
important
serological
feature
of
ANA-negative
SLE
and
we
are
hoping
to
study
these
in
our
patients.
More
extensive
studies
will
be
necessary
before
it
can
be
determined
whether
or
not
ANA-negative
SLE
is
clinically
distinct
from
the
seropositive
variety.
The
prominent
characteristics
in
Fessel's8
and
Maddison's9
groups,
respectively,
differed
from
each
other
and
were
also
at
variance
with
the
original
ARA
group.
'5
Our
patients
were
different
again,
displaying
a
relatively
low
incidence
of
skin
manifestations,
alopecia,
photosensitivity
and
cellular
urinary
casts
(Table
2).
It
remains
to
be
seen
how
large
the
ANA-negative
SLE
group
will
become
once
it
is
actively
sought.
The
prevalence
of
SLE
is
not
known
but
using
a
suggested
figure
of
one
in
2000,30
our
half-million
catchment
population
should
contain
around
250
cases
of
SLE.
The
22
ANA-negative
cases
detected
to
date
would
represent
8-
9%,
a
much
larger
figure
than
the
one
or
two
per
cent
which
we
are
led
to
expect.
Furthermore,
we
have
excluded
levels
of
DNA-bc
less
than
40
U/ml
from
our
study
thus
leaving
out
an
unknown
number
of
cases
with
significantly
raised
DNA-bc.
The
diagnosis
of
SLE
is
often
difficult;
only
seven
of
our
22
cases
had
been
diagnosed
by
their
attending
physician.
However,
the
potential
benefits
of
making
the
correct
diagnosis
at
an
early
stage
are
obvious
and
we
therefore
recommend
the
measurement
of
DNA-bc
in
every
case
where
clinically
appropri'ate,
regardless
of
the
presence
or
absence
of
ANA.
We
would
like
to
thank
the
local
Area
Committee
of
Physicians
for
permission
to
include
its
members'
patients
in
our
study.
References
I
Mowat
AG.
Laboratory
investigations
for
joint
diseases.
Medicine
Magazine
3rd
ser
1978;12:590-5.
1120
Antinuclear
antibody-negative
systemic
lupus
erythematosus-how
common?
2
Hughes
GRV.
Immunological
tests
in
the
rheumatic
diseases.
In:
Connective
tissue
diseases
2nd
ed.
Oxford
Blackwell
Scientific
Publications,
1979:256.
El-Ghobarey
AE,
Sloane
DJP,
Whaley
K.
The
clinical
signifi-
cance
of
the antinuclear
antibody
test
as
a
screening
procedure
for
DNA
antibodies
in
SLE.
Scott
Med
J
1980;25,4:293-8.
Estes
D,
Christian
CL.
The
natural
history
of
systemic
lupus
erythematosus
by
prospective
analysis.
Medicine
(Baltinore)
1971
;50:85-95.
Bartholomew
BA.
Antinuclear
antibody
tests
as
a
clinically
selected
screening
procedure.
Am
J
Clin
Pathol
1974;61:495-9.
6
Lee
P.
Urowitz
MB,
Bookman
AAM,
et
al.
Systemic
lupus
erythematosus:
a
review
of
110
cases
with
reference
to
nephritis,
the
nervous
system,
infections,
aseptic
necrosis
and
prognosis.
Q
J
Med
1977;46:1-32.
7
Fessel
WJ.
ANA-negative
systemic
lupus
erythematosus.
Am
J
Med
1978;64:80-6.
Gladman
DD,
Chalmers
A,
Urowitz
MB.
Systemic
lupus
erythematosus
with
negative
LE
cells
and
antinuclear
factor.
J
Rheumatol
1978;5:142-7.
Maddison
PJ,
Provost
TI,
Reichlin
M.
Serological
findings
in
patients
with
"ANA-negative"
systemic
lupus
erythematosus.
Medicine
(Baltimore)
1981;60:87-94.
0
Pincus
T,
Schur
PH,
Rose
JA,
Decker
JL,
Talal
N.
Measurement
of
serum
DNA-binding
activity
in
systemic
lupus
erythematosus.
N
Engl
J
Med
1969;281:701-5.
Hughes
GRV,
Cohen
SA,
Christian
CL.
Anti-DNA
activity
in
systemic
lupus
erythematosus:
a
diagnostic
and
therapeutic
guide.
Ann
Rheum
Dis
1971;30:259-64.
"
Beck
JS.
Variations
in
the
morphological
pattern
of
auto-immune
nuclear
fluorescence.
Lancet
1961
;i:
1203-5.
Wold
RT,
Young
FE,
Tan
EM,
Farr
RS.
Deoxyribonucleic
acid
antibody:
a
method
to
detect
its
primary
interaction
with
deoxyribonucleic
acid.
Science
1968;161:806-7.
4
Holian
J,
Griffiths
ID,
Glass
DN,
Maini
RN,
Scott
JT.
Human
anti-DNA
antibody:
reference
standards
for
diganosis
and
arrangement
of
systemic
lupus
erythematosus.
Ann
Rheum
Dis
1975;34:438-43.
s
Cohen
AS,
Reynolds
WE,
Franklin
EC,
et
al.
Preliminary
criteria
for
the
classification
of
systemic
lupus
erythematosus.
Bull
Rheum
Dis
1971;21:643-8.
16
Canoso
JJ,
Cohen
AS.
A
review
of
the
use,
evaluations
and
criticisms
of
the
primary
criteria
for
the
classification
of
systemic
lupus
erythematosus.
Arthritis
Rheum
1979;22:
917-21.
"
Davis
P,
Atkins
B,
Josse
RG,
Hughes
GRV.
Criteria
for
classification
of
SLE.
Br
Med
J
1973;iii:88-9.
Pollak
VE,
Pirani
CL,
Schwartz
FD.
The
natural
history
of
the
renal
manifestations
of
systemic
lupus
erythematosus.
J
Lab
Clin
Med
1964;63:537-50.
Shearn
MA,
Pirofsky
B.
Disseminated
lupus
erythematosus:
analysis
of
34
cases.
Arch
Intern
Med
1952;90:790-807.
20
Dubois
EL,
Tuffanelli
DL.
Clinical
manifestations
of
systemic
lupus
erythematosus.
J
A
M
A
1964;190:104-1
1.
21
Johnson
RT,
Richardson
EP.
The
neurological
manifestations
of
systemic
lupus
erythematosus.
Medicine
(Baltimore)
1968;
47:337-69.
22
Gibson
T
and
Myers
AR.
Nervous
system
involvement
in
systemic
lupus
erythematosus.
Ann
Rheum
Dis
1976;35;398-406.
23
Bennett
R,
Hughes
GRV,
Bywaters
EGL
and
Holt
PJL.
Neuropsychiatric
problems
in
systemic
lupus
erythematosus.
Br
MedJ
1972;iv:342-5.
24
Lachmann
S,
Hazleman
B.
Rheumatic
disorders
and
the
eye.
Update
1976;13:367-87.
25
Harvey
A
McG,
Shulman
LE,
Tumulty
PA,
Conley
CL,
Schoenrich
EH.
Systemic
lupus
erythematosus:
review
of
the
literature
and
clinical
analysis
of
138
cases.
Medicine
(Baltimore)
1954;33:291-437.
26
Winkelmann
RK.
Diagnosis
and
treatment
of
lupus
erythematosus,
dermatomyositis
and
scleroderma,
with
emphasis
on
cutaneous
findings.
J
Chronic
Dis
1961;13:401-10.
27
Jain
S,
Markham
R,
Thomas
HC,
Sherlock
S.
Double
stranded
DNA-binding
capacity
of
serum
in
acute
and
chronic
liver
disease.
Clin
Exp
Immunol
1976;26:35-41.
28
Bohan
A.
Seronegative
systemic
lupus
erythematosus.
J
Rheumatol
1979;6:534-40.
29
Provost
TI,
Reichlin
M.
Antinuclear
antibody-negative
systemic
lupus
erythematosus:
Anti-Ro
(SSA)
and
Anti-La
(SSB)
antibodies.
JAm
Acad
Dermatol
1981;4:84-9.
30
Fessel
WJ.
Systemic
lupus
erythematosus
in
the
community.
Arch
Intern
Med
1974;134:1027-35.
Requests
for
reprints
to:
Professor
CHW
Home,
Department
of
Pathology,
University
Medical
Buildings,
Foresterhill,
Aberdeen
AB9
2ZD,
Scotland.
I1121I
