Antinuclear Antibodies (ANA) Save

Welcome to Advanced Practice RheumNow. Hi, I'm Jack Cush. RheumNow. In this review, we're going to cover anti-nuclear antibodies and auto-antibodies in general. Anti-nuclear antibodies or ANA are basically equated with the diagnosis of lupus, but they really don't equate. Meaning that while everyone with lupus is going to be ANA positive, 99.9% of people, finding a positive ANA does not mean that you have lupus. And we'll talk about why. So the test is very sensitive but poorly specific.

The converse should also be considered. Could you have lupus and be ANA negative? That was very popular back in the 70s and 80s, it's really no longer possible now. It's virtually extinct. And so while you may be able to come up to me and tell me a case and point one out, it's like 0.1% of all the ANAs ever done. So trying to make a diagnosis in someone who's ANA negative, a diagnosis of lupus, you're barking up the wrong tree. So again, it is really a test that tells you you're maybe dealing with — not inflammatory — autoimmune diseases. And can be used as a confirmatory test. As I said, it's very, very sensitive. So much so that upwards of either 20 to 30 million Americans have a positive ANA at any titer. But only less than 300,000 are going to have lupus. That means about one in 100 ANAs will be due to lupus. Now it can be due to other things and we'll talk about that.

How is an ANA done? It used to be done on a slide with cells and they would take the patient's serum that has auto-antibodies in it maybe, and the antibodies would bind to nuclear antigens on the cells and then they would put on a secondary fluorochrome that would bind to the antibodies and they would read it under fluorescence. Now they do it by faster methods including ELISA, which makes the test a lot more sensitive, meaning it's a lot more positive.

Now to define the cutoffs of what's positive, they take this assay and they apply it to the normal population where you live and they make the cutoff at roughly five, maybe 6%. When you apply that cutoff to normal healthy people, five or 6% are going to have an ANA with basically zero risk of having lupus. However, you apply the same cutoffs to an elderly population over 65, a hospitalized sick population, it's going to be a three-fold higher rate of positivity. And that doesn't necessarily mean that they're going to have lupus. I mean, it just goes along with immunologic activation.

So you need to know it's a very ineffective screen for lupus being talked about without criteria or arthritis or arthralgia in general. You need a good reason to order the test. As we said, this may be found in other conditions. Most commonly, no condition at all. Second most common in my experience is autoimmune thyroid disease — Graves' disease, Hashimoto's, autoimmune thyroiditis in general. There are many other causes that are so autoimmune disease in general: myositis, polymyositis, dermatomyositis, systemic sclerosis, antiphospholipid syndrome, autoimmune hemolytic anemia, and I could go on and on. Women who are pregnant have a doubling or tripling of the risk. It's 9 to 18% of pregnant women are going to be ANA positive but not have lupus. Same thing for chronic renal disease, especially chronic liver disease. Chronic liver disease is notorious for giving you not just ANAs but lots of auto-antibodies including double-stranded DNA. It's associated with neoplasia. So the differential diagnosis is quite wide on any positive ANA.

The significance of the test rests with a few things. Especially number one, the history. It's not significant if they don't have a history that suggests a diagnosis of lupus or another autoimmune disease or liver disease or lung disease or etc. The significance also rises with titers — titers of 1 to 40, 1 to 80, 1 to 160, I don't care about — they're truly non-specific. Titers of greater than 1:280, greater than 1:640, greater than 1:1280, whatever — those you've got to take a little more seriously. And then next it's the pattern.

What we do know about the use of the ANA applied to the general population is poorly useful. It has really little predictive value. There's a study back in 1972 in Annals of Internal Medicine. It showed if you had a few lupus criteria, the odds of getting lupus only went up until you had three or four criteria or more. However, if you add an ANA to that equation, same thing — you have to have at least three criteria to make an ANA provide added diagnostic value in making the diagnosis of lupus.

So the indications — we said the causes — age is a cause, three-fold increased risk. Drugs are a cause and there are drugs that just cause ANA positivity and drugs that also cause drug-induced lupus. They are not the same. Drugs that cause ANA positivity are more frequent than drugs that cause drug-induced lupus. That's only a subset. It used to be that it was drugs like procainamide
and hydrazine. More recent years it's thorazine, lithium, quinidine, tetracycline and TNF inhibitors are notorious for causing positive ANAs. The next most common reason is no reason at all. The ANA is out there and you can just follow it and do nothing about it. The next most common reason is not having lupus. Again, no reason at all. And then finding an ANA — an ANA or an ANA referred to you has a small chance of having lupus, and then it goes up when these other things are in play: the history supports the diagnosis, the titer is very very high, there's maybe that you do because the ANA — you do other autoantibodies. So what are the indications for ordering an ANA? There's only one: a clear suspicion of lupus based on criteria and clinical presentation. After that you could order an ANA to evaluate the possibility of another autoimmune disease like scleroderma, like myositis, like Sjögren's. So in systemic sclerosis you know 60–70% are going to be ANA positive, in myositis about 50% are ANA positive, in Sjögren's it's up to 80%, in rheumatoid arthritis 40% of rheumatoid arthritis patients have ANA positivity as well.

The reasons to order it do not include arthralgia. Arthralgia is not a reason to order an ANA. You'll be having to spend a lot of time to explain why the ANA is positive in someone who just has arthralgia. Family history is not a reason and the primary care doctor thinking that it should be ordered is definitely not a reason. As I said, it's non-specific, and it is three-fold higher positivity rates just like rheumatoid factor when you apply it to an elderly population over the age of 65.

So the sensitivity of an ANA is you know 98%. The specificity however is only about 57%. This compares poorly to other more specific autoantibodies for lupus like double-stranded DNA which is found in 50 to 60% of people but the specificity is 97%. Also highly specific anti-Sm antibodies — 97% specific but only about 30% of lupus patients will have that positive, that would be the sensitivity. RNP is not sensitive nor specific for lupus. Okay. So again, you need to think about sensitivity and specificity when ordering these tests. So just like the CCP antibody is a better, more specific test for rheumatoid when compared to the rheumatoid factor, double-stranded DNA and anti-Sm antibodies are a much more specific test for lupus compared to the ANA alone.

There's an article in the Journal of Rheumatology this year in 2025 that looked at multiplex ANA assays done on 2500 individuals in one hospital system. The vast majority of these tests — these are ELISA tests done in bulk numbers — were ordered by primary care doctors and next by other medical subspecialties for hundreds of reasons. Of all the tests that were done, only 1.6% of the ANA negative group ended up having an autoimmune rheumatic disease, most often rheumatoid arthritis or spondyloarthritis. 1.6%. Right? That's one or two out of a hundred. In the ANA positive group, out of again 2500, 13.7% were positive. And I'm sorry — all of them were positive. This is 271, I'm sorry. So let's do this again. The ANA positive group: 271 of the 2500 patients — of that group, less than 14% were diagnosed with a new rheumatic disease or autoimmune disease, most often Sjögren's syndrome, lupus, or undifferentiated connective tissue disease. Bottom line is that 97% of ANAs ordered by the general prescribing practitioners do not lead to an autoimmune or rheumatic diagnosis. So again, be selective in what you order.

The important thing is to note not just the titer but also the pattern. Most ANAs come back with a speckled pattern and that's horribly non-specific, and the antigens that are being recognized are many — RNP, that's been associated with MCTD; Sm, very specific for lupus; SSA or Ro, associated with Sjögren's; SSB or La, associated with Sjögren's but also other conditions; PM1, Jo — myositis kind of things — but speckled is very non-specific and you can order an extractable nuclear antigen or individual titers of RNP, Sm, SSA, SSB, Scl-70, PM1, Jo1, etc.

Also non-specific is a diffuse pattern of ANA staining that's seen in almost nothing. It's due to deoxyRNP and histones — antibodies to those particular substrates — and it's very non-specific, could be associated with drug-induced lupus. Most specific as we said is double-stranded DNA and that used to give us a lot of what we call peripheral or rim patterns. We don't see much of that anymore.

The things I want you to remember about patterns: one, speckled is most common. Two, you often get reports of nucleolar patterns or centromeric patterns. Nucleolar patterns are found in 40% of patients with systemic sclerosis — the diffuse kind — and that's due to antibodies against RNA polymerase I and a few other antigens. The centromeric pattern is seen in CREST syndrome and seen in 75% of patients with limited systemic sclerosis, and that's due to antibodies to the kinetochore. You can sometimes
get cytoplasmic staining and that can be sometimes confusing. So when you're using an ANA to diagnose lupus and you think of systemic lupus erythematosus, that's usually we're talking about the acute forms of lupus skin disease — the malar rash, the alopecia, bullous LE and photosensitivity and oral ulcers. Such patients are always ANA positive; they may or may not have double-strand DNA and SM. There are other forms of lupus — there's the chronic um or discoid form of lupus. And you know what? Only 50% of them are going to be ANA positive and not have um risk of organ disease like um the acute LE rash and the ANA double-strand DNA patients will. And in between is subacute cutaneous lupus. They often have antibodies against the ANA, but also you can prove that with um SSA or Ro positivity.

So there are rules for using ANAs. Number one, don't order an ANA unless lupus criteria are present. Number two, don't repeat the ANA over and over and over again. It doesn't matter if it goes from negative to 1 to 40, back to negative to 1 to 80. It's still non-specific and kind of useless, right? And the ANA does not go up or down with disease activity. Don't order sub-serologies unless the ANA is positive. If the ANA is positive, go right ahead and check those boxes for double-stranded DNA, SM, RNP, SSA, SSB, ENA — basically testing extractable nuclear antigens. And maybe that'll give you further insight as to what's going on. As I said, there is no such thing as ANA-negative lupus anymore, um unless you're an expert rheumatologist dealing with thousands of patients. ANA sensitivity is variable in other autoimmune diseases, other connective tissue diseases, ranging from about 40% to 80% positivity. I would tell you do not refer patients to rheumatology for a titer of 1 to 40 or 1 to 80 without other supportive features that would suggest an autoimmune disease. The antibody by itself is meaningless at those titers and is seldom pathogenic.

Now there are other autoantibodies, and the ANA is often very associated with the antiphospholipid syndrome. Here the triad is um having an ANA — usually ANA positive — and they have one of the antiphospholipid antibody tests: either a biologic false positive RPR test for syphilis, or anticardiolipin antibodies, or the lupus anticoagulant. And lupus anticoagulant can be tested in many different ways. What we see with the antiphospholipid syndrome — you don't have the syndrome if you just have the antibody. Then you're just antiphospholipid positive with a very low risk of having any of the triad manifestations. However, if you have a positive antiphospholipid test — RPR, lupus anticoagulant, anticardiolipin antibodies — you're now looking for the manifestations that go along with that. The main ones, the triad, is recurrent thrombotic events or major thrombotic events, spontaneous abortions and miscarriages, and refractory thrombocytopenia. There are other features that go along with this: migraine, Libman-Sacks or sterile endocarditis, mitral regurgitation, transverse myelitis, and strange neuropathies. So these antibodies are found in lots of autoimmune diseases, lots of connective tissue patients, and they mean nothing unless they have thrombotic events, spontaneous abortions, and thrombocytopenia. It's found in 30% or more of lupus patients. If you want to know if the test is going to be pathogenic, well, look for those clinical features I just mentioned or order a beta-2 glycoprotein antibody, and that along with an antiphospholipid antibody test makes it much more likely to be pathogenic. If you find this, these patients need to be treated not with anti-inflammatory or immunosuppression. They need to be on chronic anticoagulation.

Lastly, I'm going to talk about antineutrophil cytoplasmic antibodies or ANCA. ANCA, as you know, is a test that was associated with Wegener's granulomatosis, which is now called granulomatosis with polyangiitis, GPA. It's found in 50 to 90% of GPA patients when you find a C-ANCA or cytoplasmic staining of a neutrophil nucleus — um a neutrophil cytoplasm. So that's why I call it C-ANCA. It stains the cytoplasm in a very sort of diffuse fashion. The antibody that is directly associated with a C-ANCA finding is antibodies to proteinase 3. Again, that's the test that you would order if you're considering GPA. Sometimes the ANCA test comes back as a P-ANCA at a certain titer. The higher the titer the more you worry. The P-ANCA means that the staining of the cytoplasm has a perinuclear distribution and this is less specific. It's found in patients with microscopic polyangiitis, EGPA — the eosinophilic granulomatosis with polyangiitis, what was previously called Churg-Strauss syndrome — and sometimes patients with lupus and lupus nephritis. The antibody that goes best with P-ANCA is the myeloperoxidase antibody.

Now there are a lot of other autoantibodies that I could cover. I'm not going to. I do want to
make a point that I brought up in the last session on uh tests for inflammation. I want to give you the American College of Physicians recommendations about the use of the ESR. They sort of apply here. You got to be smart about how you use these tests and how you interpret them. In 1986, the Annals of Internal Medicine published these guidelines from the American College of Physicians. Number one, the sed rate should not be used to screen asymptomatic people. Number two, the sed rate should be used selectively and interpreted with caution. Extreme elevation of the sed rate seldom occurs in people who have no evidence of serious disease or inflammatory disease. Number three, if there's no immediate explanation for an increased sedimentation rate, you should repeat the test at a later interval, some months later, rather than undertake a long, expensive, and worrisome search for some occult illness. Number four, the sed rate is useful in the diagnosis and monitoring of patients with polymyalgia rheumatica and giant cell arteritis. Number five, in diagnosing um and treating and monitoring patients with rheumatoid arthritis, the sed rate could be used to resolve conflicting clinical evidence. I don't know what to do. Is the manifestation inflammatory or non-inflammatory? Sed rate or CRP could help you make up your mind. And lastly, the sed rate may be helpful in monitoring and treating patients with Hodgkin's disease. That's it. Tune in for more of these great