This 7-year-old girl was evaluated for a 3-week history of arthritis. She had been well until 3 weeks before this outpatient visit, when she developed right shoulder pain for several days that resolved. One week later, she developed a cough and throat pain; a throat culture for Group A betahemolytic streptococcus (GABS) was negative, but she was treated with oral amoxicillin for 5 days. Several days later, she developed right groin pain followed by right ankle pain, each without swelling, fever, or erythema.
For the week prior to this office visit, her left wrist had become swollen and painful. There was no history of high fever, vomiting, diarrhea, rashes, or other symptoms. Her medical history was unremarkable. There was no history of exposure to animals or foreign travelers. Family history was unremarkable.
On physical examination, she was a healthy-appearing girl. Her weight was in the 25th percentile and her height in the 90th percentile. She was afebrile, pulse rate 97, respiratory rate 16, and blood pressure 127/76. There were no rashes. HEENT exam was unremarkable save for enlarged tonsils without exudate or erythema. There was no significant adenopathy. Sl was normal. S2 was physiologically split. There were multiple premature contractions but no murmurs, rubs or gallops. The abdomen was soft and nontender, without masses or organomegaly. She had Tanner 1 breasts and perineum. She had full range of motion of all joints without pain, except for the left wrist, which had obvious swelling with minimal pain on flexion. Neurologic examination was unremarkable.
Robert Listernick, MD, moderator: There are a lot of red flags to sink our teeth into here. Let's start with the arthritis. What clues does the history provide as to its etiology?
Marisa Klein-Gitelman, MD, pediatric rheumatologist: First, let's remember that arthritis is just a clinical finding, not a diagnosis. Juvenile idiopathic arthritis (JIA), what used to be called juvenile rheumatoid arthritis, is generally a diagnosis of exclusion once all the other causes of arthritis have been excluded, such as infections, malignancy, or other rheumatologic conditions. If arthritis has persisted for a defined period of time and these diagnoses have been eliminated, the diagnosis of JIA can be established. By convention, this period of time is 6 weeks in the United States and 3 months in Europe. Other clues that I look for in the history include fever curve (looking for the twice daily high spiking fevers sometimes seen in systemic onset JIA), rashes, or hepatosplenomegaly.
Dr. Listernick: Is the physical examination helpful in determining the etiology of the arthritis?
Dr. Klein-Gitelman: The quality of the arthritis is not particularly helpful. Perhaps one thing that could be said is the good range of mobility speaks against chronicity; this goes along with the history. A complete physical examination may turn up other findings, such as rash, that may be excellent clues to the etiology of the arthritis.
Ellen Chadwick, MD, pediatric infectious disease specialist: I'd add that pyogenic arthritis is generally extremely painful; patients with septic arthritis refuse to use the joint. Also, the question of acute rheumatic fever (ARF) should be raised, since this child does have a migratory polyarthritis. However, patients with ARF classically have minimally swollen joints but exquisite pain.
Dr. Listernick: If you were to see a different child with polyarthritis but no clues as to its etiology, how do you proceed?
Dr. Klein-Gitelman: I probably would start with a simple evaluation with a complete blood count and a measure of inflammation such as the erythrocyte sedimentation rate (ESR) or C-reactive protein (CRP). Even without a history of "slapped cheeks," I might send parvovirus B19 titers, particularly if there was small joint arthritis. If there were a history of camping or travel to a Lyme-disease endemic area, I would send Lyme titers. Remember that Lyme arthritis is a late manifestation of infection, occurring months after exposure. If there were evidence of an inflammatory disease (elevated ESR or CRP) and I wished to pursue a rheumatologic diagnosis, I might send testing for antinuclear antibodies or HLA-B27 or other appropriate tests depending on the physical examination and initial laboratory evaluation.
Dr. Listernick: Would you perform an eye examination for uveitis?
Dr. Klein-Gitelman: I probably wouldn't at this point, as her arthritis is not yet chronic. If an acute rheumatic illness was suspected in which eye disease might be a prominent manifestation, such as sarcoid, an eye exam would be useful.
Dr. Listernick: Let's move on and look at the next potential clue to this child's problem, the premature contractions. If a pediatrician examines an otherwise healthy child and identifies premature contractions, what would be the appropriate next step?
Barbara Deal, MD, pediatric cardiac electrophysiologist: Occasionally, a physician will hear an exaggerated sinus arrhythmia with respiratory variation and misidentify it as premature ventricular contractions (PVC). However, if an asymptomatic child has frequent enough PVCs to be heard on exam, this is definitely abnormal and an evaluation is warranted. I would look for a history of exertional syncope or a family history of sudden death, arrhythmias, or "seizures," which might be clues to a familial electrical disorder.
Before making any decisions. I would perform echocardiography with particular attention paid to the right ventricular outflow tract as a source for the rhythm disturbance. For these reasons, I believe that any child with frequent PVCs should be seen by a pediatric cardiologist. This child's electrocardiogram did show multiple PVCs.
I know we've been skirting around the diagnosis, but the possibility of ARF needs to be entertained.
Dr. Klein-Gitelman: Given the migratory nature of the arthritis, ARF should be considered. I agree that the arthritis pain is classically quite severe with minimal findings on exam in ARF, but patients often don't read the textbook.
Dr. Listernick: What are the diagnostic criteria for ARF?
Dr. Chadwick: We continue to use the modified Jones criteria for establishing the diagnosis of ARF. The five major criteria are migratory polyarthritis, usually involving the large joints; carditis, which is most commonly endocarditis involving the valves but may also be pericarditis or myocarditis; Sydenham chorea, a very late manifestation of disease; and two manifestations we rarely see these days, subcutaneous nodules and erythema marginatum. Minor criteria include prolonged PR interval, fever, and elevated erythrocyte sedimentation rate or CRP. Required for the diagnosis is the presence of two major criteria or one major criterion and two minor criteria, in addition to evidence of an antecedent GABHS infection.
Dr. Listernick: What are the ways that we can look for evidence of an antecedent GABHS infection?
Robert Tanz, MD, general academic pediatrician: First, one can perform a throat culture. If negative, the physician can look for the presence of streptococcal antibodies. The three most commonly sought are anti-streptolysin O (ASO), antiDNAse B, and antihyaluronidase. Unfortunately, only 70% to 80% of patients who have had a GABHS infection have an elevated ASO titer 3 to 6 weeks after infection. In addition, ASO titers generally don't rise after GABHS skin infections, which are not a prelude to ARF. Therefore, it's recommended that at least two different antibody titers be measured. Unfortunately, there are fewer well-defined standards for anti-DNAse B, making the results difficult to interpret.
Dr. Listernick: How would you interpret this child's ASO titer of 120 units and anti-DNAse B titer of 340 units?
Dr. Tanz: Neither is very impressive. However, the anti-DNAse B titer tends to rise later than that of ASO. Maybe we haven't caught the anti-DNAse B titer at its peak yet.
Dr. Listernick: An echocardiogram revealed that the child had mild mitral regurgitation and trivial tricuspid regurgitation. Cardiac function was normal and there was no pericardial effusion. How should we interpret these results in the context of the patient?
Dr. Deal: The mild mitral regurgitation would support the diagnosis of ARE However, traditionally, echocardiographic results are not used in the diagnosis of ARF; evidence of valvar endocarditis should be detected with a stethoscope. This is because many normal people will be found to have trivial or mild mitral insufficiency if they undergo echocardiograms. Of course, it all depends upon who's doing the listening!
Dr. Listernick: Aren't the PVCs evidence of cardiac irritability and, therefore, of active carditis?
Dr. Deal: Yes, but the child might have had long-standing, unrelated PVC. Nevertheless, the combination of the entire clinical picture with the echocardiographic findings is highly suspicious of ARF.
Dr. Chadwick: However, playing devil's advocate, arrhythmias have never been a part of the classic description of the carditis of ARF.
Dr. Deal: True, but let's not forget that a prolonged PR interval is a manifestation of carditis and that cardiac monitoring and Holter monitors weren't used in the time when rheumatic fever was so prevalent.
Dr. Listernick: At the time I saw this patient, my reasoning was that she had ARF, although I understand that there are some arguable points. Assuming I was correct, what would have been the appropriate management?
Dr. Chadwick: First, it's standard to give a 10-day course of penicillin therapy, even if the throat culture is negative for GABHS. Generally, if there's evidence of active carditis, I would institute high-dose aspirin therapy; this should dramatically resolve the pain in patients with arthritis. In those patients with serious cardiac disease, such as lifethreatening pericarditis or valvulitis and congestive heart failure, I would treat with corticosteroids.
Dr. Deal: Although it's controversial, where I trained I was taught to use steroids for any patient with active carditis. Usually I use a 2- to 4week course of prednisone at a dose of 1 to 2 mg per kilogram per day.
Dr. Chadwick: The data indicate that there's no difference in the ultimate sequelae of ARF, the development of long-term mitral or aortic valve disease, when the use of steroids or salicylates are compared. I do agree that children with pancarditis and congestive heart failure should be treated with steroids, although unassailable evidence that further valve damage can be prevented is lacking.
Dr. Listernick: Wasn't bed rest one of the major forms of "therapy" in the heyday of ARF?
Dr. Deal: That recommendation has probably dwindled into obscurity.
Dr. Listernick: Moving on, I felt that she had ARF. Initial laboratory evaluation revealed a normal complete blood count; the erythrocyte sedimentation rate was 86 mm/hour. The electrolytes were normal, blood urea nitrogen was 40 mg/dL, and creatinine was 0.9 mg/dL. The urinalysis had large occult blood and 2+ protein on dipstick testing; the urinary sediment had greater than 50 red blood cells, 20 to 30 white blood cells, and two white blood cell casts per highpowered field.
Jerry Lane, MD, pediatric nephrologist: No one as yet has commented on her blood pressure; it's definitely elevated. The combination of hypertension and the abnormal laboratory findings suggests that she has active nephritis. Using Occam's Razor, it's tempting to think that she has a post-infectious glomerulonephritis, specifically post-streptococcal. Alternate diagnoses to consider would include systemic lupus erythematosus (SLE) and membranoproliferative glomerulonephritis (MPGN).
The next step would be to measure levels of C3, C4, and antinuclear antibodies (ANA). Post-infectious glomerulonephritis can be distinguished from MPGN in that the levels of C3, which are severely depressed in both diseases, should return to normal in the former within about 6 to 8 weeks, as opposed to MPGN, in which they remain low. Interestingly, we've seen some patients with post-infectious glomerulonephritis who have had systemic symptoms and arthralgias.
Dr. Klein-Gitelman: Since we're speculating here, not one of her symptoms is inconsistent with a diagnosis of SLE.
Dr. Listernick: Her initial C3 was very low and returned to normal 6 weeks later. C4 was normal and ANA was negative.
Dr. Listernick: When would a kidney biopsy be necessary in the evaluation of presumed post-infectious glomerulonephritis?
Dr. Lane: If the patient developed nephrotic syndrome or significant renal insufficiency, or if the level of C3 failed to return to normal after 8 weeks.
Dr. Listernick: How should this child be treated?
Dr. Lane: The treatment for post-infectious nephritis is purely supportive. In this case, she needs an anti-hypertensive; we chose a low dose of a calcium-channel blocker, which was effective.
Dr. Listernick: So does she have both ARF and post-streptococcal glomerulonephritis? For whatever it's worth, that's my opinion.
Dr Chadwick: There have been case reports of the two illnesses coexisting, but it is exceedingly rare. There may be a single post-infectious process going on here but, as I said before, the post-streptococcal titers are not impressive. It would have been nice to perform convalescent titers. With that said, she remarkably improved the next day after being treated with aspirin.
Dr. Listernick: Can one strain of GABHS cause both ARF and postinfectious glomerulonephritis?
Dr. Tanz: There are M protein types that are associated with ARF and those that are classically associated with post-infectious glomerulonephritis. There is some overlap, such as M 1 strains, that are associated with both diseases. Intriguingly, the strains associated with each disease in Australia are entirely different. In addition, streptococcal impetigo has been suggested to be linked possibly with ARF in Australia in the aboriginal population, an association not even suggested in the Northern hemisphere. The differences may be due to the populations' underlying genetic makeup.
Dr. Listernick: Assuming she had ARF, what would be the longterm treatment recommendations?
Dr. Chadwick: She would require lifetime penicillin prophylaxis to prevent recurrent episodes of streptococcal infection and rheumatic heart disease. There are some specialists who would argue that the highest risk of recurrent disease with cardiac involvement is within the first five years following the initial episode and would discontinue prophylaxis if there had been no recurrences by the time the child reached adulthood.
Dr. Listernick: Recognizing that it's a difficult call, we felt that there was sufficient evidence pointing to active carditis that we have kept her on penicillin prophylaxis.
Thank you, everybody.