Recognising Conditions Resembling Juvenile Idiopathic Arthritis

This transcript was produced by automatic speech recognition and edited for readability; it has not been fully verified word-for-word. It may contain errors — particularly with names, technical terms, and where English is a speaker's second language. It is provided as a convenience only. For anything clinical, please refer to the video recording to hear exactly what the expert said.

Introduction

Prof. Athimalaipet Ramanan: Good afternoon, everybody. My name is Ram. I'm a pediatric rheumatologist in Bristol and Bath in UK. Over the next 30 to 40 minutes, I'm going to talk to you about pediatric rheumatology, but mainly in the context of when we should suspect arthritis in children and how to discriminate between conditions which can mimic arthritis but are not quite juvenile idiopathic arthritis.

So just a few caveats for all of those. Please do not take any screenshots as these slides are copyright protected. It would be appropriate if you do not reproduce or share these without the permission of myself. We're going to talk for about 30 minutes, maybe a bit short, and then giving enough time for questions and answers, which you can communicate through the text option in your webinar.

And once you submit the questions, I will read the questions out and then I will give the answer as well so that all, including those who have not asked the question, can understand what I'm answering to. And a recording of this will be available for all those who are registered on the EIP website. So to move straight on, I think it's always appropriate to contextualize learning in the form of a case, which is how we all learn and how we see patients.

A case: the child with a swollen joint

So this is a three-year-old boy who was referred to us by his general practitioner with a history that he's got swelling of his right knee, which has been going on for a month. Now every word in this history is carefully chosen.

The fact that he's three is important because that's around the peak age at which arthritis in children manifests. Although it's twice as more common in girls, you can get boys as well. And the fact it's only in one knee, as the GP has pointed out, and it's for a month are very important because you can, with viral infections or with bacterial infections, sometimes get transient synovitis or transient viral synovitis.

The fact that this has been going on for a month and is at the right age as in a single joint should make you think when you get such a history, either by letter or by a phone call, that this might be an inflammatory arthritis. Many a time when you take history in these children between the ages of 18 months and three years, you will always get a history of trauma, and that's because three- year- old children naturally are going to be active, and when they're active they have their fair share of calls.

As soon as any parent notices a swelling of a joint, the first thing he or she will do is to look back and think, oh, has my child had a fall? Is that why it's swollen? It's usual innate human instinct, but in most cases the trauma history is a red herring. What I mean by that is if you have significant trauma, then usually that trauma will make you seek attention in the next 24- 48 hours.

It's highly unlikely that someone who's presenting with a knee swelling for one month tries to recall a history of trauma for which they didn't seek medical attention. What this means in real life is most of the time this history of trauma tends to confuse things, sometimes takes them down the route of getting x- rays, looking for tears, cartilage problems, and may delay the diagnosis of juvenile arthritis.

So the key point here is trauma which is significant will make parents or the carers seek attention within the first 24- 48 hours. So this, for example, is how a knee swelling might look in this patient. So here you can see quite a significant swelling, but it's mainly in the suprapatellar area. So I'm going to use my pointer to point out you can see the huge suprapatellar swelling.

So sometimes the swelling may not be in the knee, it might be in the suprapatellar region. So the knee looks normal, but their fusion is mainly suprapatellar. We know from work we have done in the UK that there is a delay of between six weeks to six months and sometimes up to 18 months in the diagnosis of arthritis in children.

Examining the joints properly

Now arthritis in children occurs in about one in a thousand children, so while it's not as common as adult inflammatory arthritis such as rheumatoid arthritis, it's not that uncommon as well. And why do we have this delay? Unfortunately neither undergraduate medical curriculum or postgraduate curriculum teaches examination of joints. So we are, as pediatricians, very good at examining the cardiovascular system, picking up murmurs, examining the neurological system, examining the cranial nerves, but we are not taught to examine the joints in a formal way.

So that tends to enable us not to diagnose the problems associated with joints. Orthopaedic surgeons may get it in their postgraduate teaching, pediatricians certainly don't. And this is an illustrative case I always use when I teach juniors, I ask them which ankle is swollen. And it's a trick question because the answer is that both ankles are swollen.

So normally you'll have the Achilles tendon running down like that and you should have a concavity. What you see here is a fullness and the important aspect here, which is very important when you examine children, is you need to undress, look at the joint both anteriorly and posteriorly. In case of ankle, if you do not examine posteriorly, you will miss signs of an effusion.

It's very unusual for the anterior compartment of the ankle to look swollen. It's usually posterior where there is space for you to see the effusion. This is another child where I'm demonstrating a knee effusion. Again, if you don't have a good technique of examination, it may be easy enough to miss this knee swelling. And as I pointed out, one of the main problems that are associated with delay in diagnosis of arthritis is an incomplete examination, which includes failure to examine the patient fully both anteriorly and posteriorly.

And I highlight again the importance of examining posteriorly. This is a patient who's got what looks like a Baker's cyst and could well be a Baker's cyst, but in this case it's a loculated posterior effusion of the knee. So if you try to examine without undressing the child properly, without looking at the posterior aspect of the knee, you could very easily miss this loculated synovial effusion in this case, but it could also be a ruptured Baker's cyst.

And this demonstration is again to show in a patient which we were sent by the GP with a knee swelling, but actually had this wrist swelling as well. And again, this is a normal wrist and here you can see. While in this picture none of you is going to miss which is the affected wrist and which is a normal wrist, it's very easy in real life, if you're not used to seeing inflammatory arthritis in children all the time, to miss this sort of a swollen wrist, which if left untreated for months can cause localized joint destruction.

Another problem in arthritis diagnosis is the inability or the failure to examine the skin properly. Now you might wonder why is the skin important in rheumatology. It is exactly for this. What I'm demonstrating here is a psoriatic rash. Now psoriasis can be seen in multiple forms. It can be seen as plaque psoriasis, in which case they report a history.

But many a time in children, psoriasis can be quite subtle. It can be seen only in the gluteal cleft as it is here, or a small patch behind the ear, in teenage children just in the scalp, or just in the umbilicus. And if you don't examine the skin properly, you may not pick this up. And the relevance of this skin lesion is when a child presents with a swollen toe, a dactylitis like this, and you see that skin rash, you can put the two together and know clearly what you're dealing with here is a psoriatic arthritis rather than an infected toe, which would lead you to do unnecessary tests and take you down a different path.

Age of onset and mimics of arthritis

So we talked about the peak age of onset of arthritis, which is about 1 in 1, 000 in the Western world as being in the toddler age group, an age group where it's very difficult to examine the children, leave alone examine the musculoskeletal system. But here are three pointers to what are the presenting features in the toddler.

One is a child who is walking, now refusing to weight- bear. A child who is walking normally, suddenly starting to walk with a limp. And most importantly, in a child who hasn't started walking, in the differential diagnosis of delayed motor milestones, one of the diagnoses to consider would be whether that's a simple mechanical problem, such as a swollen knee or an ankle, which is preventing this child from starting to walk.

And it's because it happens at the same age group as when they start to walk, it can easily be wrongly diagnosed as a neurological problem. One of the most common things in general practice or in primary care is when arthritis is dismissed by either healthcare professionals or doctors as growing pains. So let's look at what exactly is growing pains.

Growing pains is an important cause of children getting symptoms and parents bringing these children to the attention of general practitioners or primary care pediatricians. It's actually a discrete entity. Usually happens in children between the ages of 4 to 8. It can happen in either leg. These children will wake up in the middle of the night, claim that they have pain in the leg.

They'll point in a nonspecific fashion. It will last for 5 to 10 minutes, and they'll feel better. During the day, they will have no limitation, and they will be perfectly normal functionally in terms of doing curricular or extracurricular activities. If it's not any of these, if they have associated systemic features, if they consistently point to one joint, one leg, if they point to pain in the joint, if it happens at daytime, or if they have associated limp, or when you examine, you see physical signs, that is not growing pains.

For growing pains, all you need to do is reassure them and tell them that even though we've been taught in medical school that things which wake up children at night is sinister, in this case, it is no wonder they have to worry about it. So in this child, going back to the case, we've got the three- year- old with a swollen knee.

The pitfalls of autoimmune tests

What do you do? People would have done blood tests usually, and a few pointers if you do blood tests. ESR and CRP is not elevated if you've got only one or a couple of joints swollen. So absence of elevation of the inflammatory markers does not exclude arthritis. Inflammatory markers tend to get elevated only when you have arthritis in multiple joints.

Another huge problem is poor understanding of autoimmune tests. You frequently get calls saying, I've done the ANA, I've done the autoimmune serology, I've done the autoimmune profile. They sometimes cause more anxiety among patients and parents when they resolve the anxiety. So let's look at the anti- nuclear antibodies. These are antibodies against the nuclear components, both the protein and the nucleic acid.

And these are done in dilutions usually. You will have a 1 in 20 dilution of the patient's serum, and then you will keep diluting it. There are a couple of important things to remember. In the past, the substrate which we used to use for ANA might still be used in some countries was rat liver. In that test, the false positivity was very low, but you could have a false negativity of almost 5%.

These days we use HEP2 or human epithelial cell. With these, you have a very high false positivity. Almost 10, even 15% of normal children could have low-titre ANA positivity. So it's important to not test ANA if you're not sure if it's going to help in that particular context. In a child with nonspecific aches and pains, in whom you haven't seen any swelling, it is not appropriate to check the ANA because a positive ANA is not a diagnostic value.

It's purely a prognostic value. This is a lupus patient, a nice picture demonstrating the ANA fluorescing very nicely, what we call as a homogenous pattern because of the double- stranded DNA. So what proportion of children who are normal and adults who are normal can have positive ANA? So 1 in 40, you can see in some adult populations, almost a third of the population is positive.

13% when you dilute it further to 1 in 80. 5%, 1 in 60, and 5% of adults don't have arthritis. 3% of 1 in 320. The importance here is to remember that false positivity of ANA is common, but as the T tier goes up, the relevance of the test increases. So it's a test which has a poor positive predictive value, and here I'm going to point out two nice studies.

One in a pediatric rheumatology clinic where they took 245 patients, and they found that almost half of them had a positive ANA with no rheumatic disease.

And another study, they took 110 children, again, these were children sent to a rheumatology clinic with a positive ANA, and two- thirds had no rheumatic disease. Many a time, a referral to a pediatric rheumatologist is because of aches and pains, positive ANA. It's perfectly appropriate to send a patient with aches and pains if you think they may have an underlying rheumatological disease, but do not do the ANA unless you have a strong reason to do so.

And I'm encapsulating that in this slide. In children presenting with aches and pains, if you do not see swollen joints, if there's no clear signs of rheumatic disease, do not do the ANA. And if the ANA is done prior to referral and positive, then it adds a complexity, but you need to look at the titer, the history, and the context in which the patient's presenting.

There is no clear evidence that positive ANA predicts arthritis development. In children with signs of rheumatic disease, be that cutaneous manifestations, rashes, the AMA can be predictive of lupus, can be useful in children's, in mixed connective tissue disease, and in scleroderma. Negative ANA doesn't always exclude lupus. About 1% of children with the way we test lupus these days or ANA these days can be ANA negative.

In the older methods, you may read that almost 5% of lupus is ANA negative. I would say these days it's 1%. So let's look at the auto- inflammatory diseases which are associated with a positive ANA. Lupus is the most important one. So if you see a teenage girl with fevers, hair loss, mouth ulcers, a malar rash or nonspecific fatigue, an easy way to exclude lupus would be to do an ANA.

And as I said, these days almost 99 percent of them with the human epithelial cell should be positive. Sjogren's syndrome, extremely rare in children, but half of them are ANA positive. Systemic sclerosis, incredibly rare, probably one in a couple of million and almost 90 percent of them will be ANA positive. JIA, only half of them are ANA positive.

But if you're ANA positive and you have arthritis, you're more likely to have uveitis. Dermatomyositis, again, five per million condition, two- thirds will be ANA positive. In the past, when we didn't identify the specific antibodies, we used to use the pattern of ANA staining to help us diagnose who's got lupus, who's got scleroderma. These are less important now.

Double- stranded DNA is very specific for lupus. Antibody called anti-Ro/La is very specific for Sjogren's, also has relevance in congenital heart block in babies born to mothers with lupus or Sjogren's. And some of the more rarer antibodies, which I won't go into. So coming back to our three- year- old, we know if you've done the blood tests, they will all be normal.

ESR and CRP does not need to be elevated. The diagnosis of JIA is purely clinical. Autoantibodies are of prognostic significance, not of diagnostic significance. And most importantly, you need six weeks of signs and symptoms. So if you see a child within two weeks, give them naproxen, ibuprofen, or paracetamol and ask them to come back. And this is because bacterial and viral arthritis can give you fleeting arthritis or arthralgia.

Why JIA diagnosis is delayed, and dangerous mimics

So why is the diagnosis of JIA delayed? Narrow history- taking. We know the concept of preferred pain. Children may complain of hip pain when the joint involved is the knee, or of knee pain when the joint involved is the hip. The inability of pediatricians to examine the musculoskeletal system, even though they're very good at examining other systems.

The concept of uveitis, which can happen in children with silent arthritis. Sometimes, temporomandibular joint arthritis presents as an earache. Infections which can masquerade as arthritis. Malignancy which can masquerade as arthritis, I'll come to that later. Discitis which can come and present as back pain. Inflammatory bowel disease can sometimes present as arthritis initially. So we talked about what delays the presentation of arthritis.

These are the conditions. Malignancy, which can masquerade as arthritis, and very important not to miss them. Storage disorders like mucopolysaccharidosis can present as arthritis, particularly the milder forms. And in the older children, connective tissue diseases such as lupus, the first presentation might be joint symptoms. And in some families, particularly where there's a high prevalence of consanguinity, you can get syndromes which will present as joint contractures and swelling, but they're not inflammatory arthritis, they are syndromes.

The diagnosis you do not want to miss in a child is leukemia. And the importance and the association between leukemia and arthritis is that one- fifth of all leukemias in children will present with musculoskeletal symptoms. So the presenting feature of leukemia is the musculoskeletal features. This nice study from Denmark shows that in a 10- year period in a tertiary pediatric rheumatology unit they looked at 10 of the 1, 254 patients referred actually had malignancy.

And I refer to those of you who are interested in this area to this beautiful paper which was in Archives of Diseases in Childhood in September 2015. This again is from Scandinavia. They looked at all acute lymphoblastic leukemia, the most common leukemia in children, between a 10- year period, 1992 to 2013.

And they found, as we know, a fifth of these patients presented with joint pain, a fifth of acute lymphoblastic leukemia patients presented with joint pain, and half of them were joint pain and arthritis. So 19 of these 53 had reactive arthritis, 9 thought to have osteomyelitis, and 8 thought to have JIA. Most importantly, children with joint symptoms had less hematological signs of the leukemia.

And those who had joint symptoms also had a delayed in diagnosis compared to those who didn't have the joint symptoms. And this is also showing the event-free survival in children with ALL, with or without joint involvement, and the overall survival in those with and without joint involvement over a 15-year period. Again, if you're interested, we also wrote an editorial saying, are you missing leukemia?

A case: not missing leukaemia

Because what we don't want to do is miss leukemia. We'd rather delay the diagnosis of arthritis and be careful, cautious, and confident. So I'm going to finish off with a couple of less common entities before I take some questions. This is a prototypical patient you might see in a clinic, 15-year-old girl with fatigue. You're wondering if she's got an organic condition, if she's got chronic fatigue.

And when you look at the blood results, she's got low white count, low hemoglobin. In this sort of a patient, you must always consider lupus. And in this context, an ANA is extremely helpful because 99% of the patients with lupus these days will be ANA positive. And we also know that a lot of children at this age can get chronic fatigue.

And again, that's why the ANA is very helpful. We see, on average, one child every couple of weeks with a teenage girl presenting with Raynaud's. And the story is cold hands, particularly with cold weather. Very, very rarely, the Raynaud's is the first sign of scleroderma. So the only test you need to do is check the ANA. And if it's positive, ask them to do the centromere antibody.

After that, you say to the family and the patient in question about keeping warm, reassuring them this is primary Raynaud's, and that it will settle with time. The rare patient in whom the Raynaud's is a manifestation of scleroderma, you pick up on the ANA. The symptoms you want to watch out for is joint pain associated with the Raynaud's, photosensitivity, malar rash with lupus, hair loss with lupus, dry eyes and mouth, particularly in the older children who can articulate this, which could be in keeping with showings.

I'm going to leave you with this last slide. And this is a patient we've seen for a long time who was initially thought to just have a bruise and then was in orthopedics. And then to dermatology, this is a boy with a linear scleroderma, which eventually if left untreated will result in significant wasting. And you can already start seeing the loss of bulk in one leg compared to another.

Questions and answers

I think I'm going to stop there, take your questions, which you can send by text. I'll read out your question, and then I will read out my potential answers to that as well. So the webinar is now open for questions. So someone's asked a question here, which we've already addressed. Should one check ANA in a child with aches and pains?

As I said before, if there is no evidence of rheumatic disease, if there is no swelling of the joints, please don't check an ANA because there's a high degree of false positivity of ANA in children. So another question, which is a very good one. Can we label a child as systemic onset JIA without joint involvement, but with fever and raised inflammatory markers after ruling out malignancy?

Yes, you can. In such patients, it's very important to do a bone marrow. We are in the process internationally of changing the criteria for diagnosis of systemic onset JIA. And I think in the future you will have the ability, even as per classification criteria, to diagnose somebody with just fever and rash with appropriate raised inflammatory markers, raised ferritin, as systemic onset JIA without the need for arthritis.

Increasingly, that's how these patients present. Someone's asked, what's the best treatment for growing pains? There is no treatment. Most people will give you all sorts of treatments on Google. The truth is reassurance, reassuring the family, particularly the parents, that waking up in the middle of the night, in this case, is not sinister and really that would settle with time.

There is no evidence base for any other treatment. So someone's just texted, bone marrow done, normal. I presume it's in the context of a systemic JIA. Absolutely right. If you've done a marrow, then go ahead and consider a systemic JIA. You do not need to have arthritis because a proportion of patients do not have arthritis at presentation.

So another question, does a double- stranded DNA have to be positive to make a diagnosis of lupus? And the answer is no. Double- stranded ANA is positive only in about 50% to 70% of patients with lupus in pediatric populations, depending on the study you look at. It's very unusual to be ANA negative, but it's not uncommon at all to be double- stranded ANA negative.

I'm led to believe there are no more questions. We're probably 10 minutes ahead of schedule. I wish you all a pleasant rest of the day. And I hope you find the webinar to be of some value.

Thank you very much.

‍