MCTD / overlap and blood tests

What the research found.

• One study took every patient at a hospital with a positive anti-U1-RNP antibody over twenty years, and there were 36 of them. Eighteen met a classification for this disease, and eleven of those 18 also met the 2019 criteria for lupus. Across the whole group, 23 of the 36 met lupus criteria.

  Elhani and colleagues, Rheumatology International, 2023

• One cohort followed 330 patients with this diagnosis for a median of 8 years. The Sharp criteria were met by 97.3 percent of them and the Kasukawa criteria by 93.3 percent, so the two sets agree closely. No patient in it met another classification without meeting one of those two first.

  Chevalier and colleagues, Journal of Internal Medicine, 2024

• In that cohort, 85 of the 330 patients moved on to a different connective tissue disease, which is 25.8 percent of them. An abnormal nailfold capillaroscopy at diagnosis came with an odds ratio of 2.44 for that happening. The range ranged from 1.11 to 5.58.

  Chevalier and colleagues, Journal of Internal Medicine, 2024

• In that cohort, 149 of the 330 patients reached remission, which is 45.2 percent. Interstitial lung disease occurred in 27.9 percent of them and pulmonary hypertension in 7.6 percent, over that median of 8 years. Over the whole of the follow-up, 5.6 percent of the cohort died.

  Chevalier and colleagues, Journal of Internal Medicine, 2024

• One study looked at nailfold capillaroscopy in 43 patients with this diagnosis. A scleroderma pattern was present in 46.51 percent of them, and it was mostly the early kind of pattern. Reduced capillary density was more common than in controls, at p below 0.0001, which was the strongest comparison.

  Ornowska and colleagues, Clinical Rheumatology, 2024

The antibody that defines the disease, and what it doesn't do

Mixed connective tissue disease is built around one antibody, which is called anti-U1-RNP. You can't have this diagnosis without that antibody. It sounds decisive put that way, and it isn't decisive at all, and the distance between those two things is the most useful thing on this whole page.

One study took every patient at a French hospital who tested positive for that antibody over a span of twenty years, and there were 36 of them. It then applied every relevant criteria set to each one of those 36 patients in turn. What came out of that is the rest of this page.

Half of them, and most of those had lupus too

Eighteen of the 36 met a classification for this disease, which is half of the group. Of those eighteen, eleven also met the 2019 criteria for lupus, and twelve more met lupus criteria without meeting the criteria for this disease at all. Across the whole group of 36 patients, twenty-three of them met lupus criteria.

So a positive antibody put each of those patients into a room where several diagnoses were possible. It didn't pick any one of them out for you. That's the practical meaning of a test being required for a diagnosis without being sufficient for one.

What does the picking

The examination does most of the picking here. In that study, the features that separated this diagnosis from lupus turned out to be all physical ones. Tight finger skin, swollen hands, reflux, and cold-sensitive fingers all pointed toward this disease rather than the other.

The absence of them pointed the other way entirely. When those features weren't there, the patient usually turned out to have a different definite connective tissue disease, and most often that was lupus. So a laboratory result was outranked by what a doctor sees and feels, which happens more often than people expect.

The two criteria sets, which agree

You may hear two names for the rules used to make this diagnosis, and they're the Sharp criteria and the Kasukawa criteria. One cohort of 330 patients checked both of them, where the Sharp criteria were met by 97.3 percent and the Kasukawa criteria by 93.3 percent. More usefully, no patient in that cohort met some other classification without also meeting Sharp or Kasukawa.

So which set your team prefers rarely changes the answer for any individual patient. That's unusual in medicine and it's worth knowing. Two rule sets agreeing that closely is a sign both are describing something real.

The test at the base of your fingernails

Nailfold capillaroscopy is a magnified look at the tiny blood vessels where your nail meets your finger. It takes a few minutes and nothing is taken from you at all. One study looked at 43 patients with this diagnosis, where a scleroderma pattern was present in 46.51 percent, mostly the early kind.

Giant, branched, and dilated capillaries were all more common than in controls, and so was reduced capillary density, at p below 0.0001. That was the strongest of those comparisons by a distance. The study also linked specific findings to specific problems, where reduced vessel numbers and bare patches were associated with tight finger skin, and branched vessels with high blood pressure in the lung arteries.

Why that test counts for more than it looks

The larger cohort found something that makes capillaroscopy worth asking for at diagnosis. Of the 330 patients in it, 85 moved on to a different connective tissue disease over the follow-up, which is 25.8 percent. Most of those became systemic sclerosis at 15.8 percent, or lupus at 10.6 percent.

Two things at the diagnosis predicted that happening. An abnormal nailfold capillaroscopy came with an odds ratio of 2.44, from 1.11 to 5.58, and swelling of the salivary glands came with 3.86, from 1.31 to 11.4. Neither of those is a blood test, and both are things somebody has to look for.

The rest of what that cohort found

Those 330 patients were followed for a median of 8 years, with a range of 3 to 14 years. Remission was reached by 149 of them, which is 45.2 percent, and interstitial lung disease occurred in 27.9 percent while pulmonary hypertension occurred in 7.6 percent. Over the whole period, 5.6 percent of them died.

Patients who didn't move on to another disease were much more likely to reach remission, at 51.8 percent against 25.9 percent. Read those numbers with one thing in mind, which is that the cohort came from specialist centres where the sickest patients end up. The figures in the general population of people with this diagnosis are probably better than these.

Your antibody picture can change

One more study is worth knowing about, because it explains something patients find confusing. It followed 40 patients with this diagnosis and watched for antibodies against a second, closely related target appearing over time. Thirteen of the 40 developed one, and it mostly happened within the first two years.

Those patients looked different from the rest of the group. Not one of them had skin sclerosis, against 44 percent of those without the new antibody, and after the change their arthritis became less common and so did their muscle problems. So a new antibody on a later report isn't necessarily a mistake or a new disease.

No level, no cut-off, no range

There's no antibody level here, no cut-off, and no reference range on this page. That isn't an oversight on anybody's part at all. No source behind this page measures a threshold in this disease, and not one of them shows that a particular number means worse disease or a different outcome.

So a level is a question for your team, who can see your whole picture rather than one line of it. A number read off a web page is what this site refuses to print. That rule holds everywhere on the site, and it holds here for a reason that hasn't gone away.

Common misconceptions.

Cautions specific to this condition.

• Don't take a positive anti-U1-RNP as a diagnosis. Half the people who have that antibody don't meet a classification for this disease.
• Ask about nailfold capillaroscopy. It's a magnified look at the base of your fingernails, it takes minutes, and an abnormal result at diagnosis tells your team something real.
• Ask about lung screening. Interstitial lung disease occurred in 27.9 percent of one large cohort and pulmonary hypertension in 7.6 percent.
• Don't read a number off this page. It gives no antibody level, no cut-off, and no reference range, because our sources measure no antibody level in this disease.
• Expect your picture to be reviewed over time. A quarter of patients in one cohort moved on to a different diagnosis over a median of eight years.

Questions patients ask.