by Eric Howell, M.D.
*This case received 2nd place for Clinical
Vignette at the May 1999 meeting of the American College of
Faculty Sponsor: Joan Bathon, M.D.
- History of Present Illness
- Physical Examination
- Laboratory Studies
- Radiology Studies
- Differential Diagnosis
- Clinical Course
- Case Discussion
H.L. is a 46 year old Afro-American female, with a past medical history significant for hypothyroidism and beta-thalassemia, who was otherwise healthy until two weeks prior to admission . At that time she noted onset of a sore throat, malaise and low-grade fevers. Four days prior to admission, she experienced fever to 104oF, chills, nausea and vomiting, and arthralgias and myalgias. She presented to the Johns Hopkins Bayview Medical Center Emergency Room.
Past Medical History:
- Chronic anemia due to beta-thalassemia
- Hyperthyroidism S/P 125I treatment with resulting hypothyroidism
- History of benign systolic ejection murmur for 15 years
The patient was noted to be ill-appearing. Vital signs: Temp 98.8F, pulse 102, resp 16, BP 100/70. Her exam revealed mild pharyngeal erythema but no exudate, shoddy anterior cervical lymphadenopathy, a 2/6 systolic ejection murmur that was non-radiating, and mild diffuse abdominal tenderness but no rebound. Joint exam revealed no swelling, erythema or warmth of any of the peripheral joints. No rashes or other cutaneous abnormalities were observed.
Hematocrit 25.1%; WBC 14.2 with 51% bands; blood urea nitrogen 31. Creatinine 3.1, C3 – 145 (normal), C4 – 50 (normal), total bilirubin 3.5, alkaline phosphatase 209, AST 52, amylase 40.
Urinalysis showed 4+ protein, WBC & RBC were too numerous to count.
Chest x-ray showed no infiltrates.
EKG demonstrated sinus tachycardia and left atrial enlargement.
ASO titer-250 Todd units
Blood and urine cultures were sent.
The patient was admitted to the hospital with a presumptive diagnosis of urinary tract infection and acute renal failure. She was initially treated with levofloxacin and ampicillin. On day 2 of her hospitalization, urine and blood cultures grew a ß-hemolytic streptococcus that was identified as penicillin-sensitive Group A Streptococcus. Pharyngeal culture was negative for Streptococcus (or other pathogens) but, unfortunately, the culture had not been obtained until after the first dose of antibiotics. Because of the persistent and severe sore throat, the possibility of a retropharyngeal abscess was considered but ruled out by ENT evaluation and CT scan of the neck. Endocarditis was also considered but felt to be unlikely by the Infectious Disease consultants based on the absence of vegetations on transthoracic echocardiogram (patient could not tolerate a transesophageal study) and the rarity in which Group A Strept species cause endocarditis. The patient’s antibiotics were subsequently changed to IV Penicillin and a 2-week course was recommended.
Although the patient began to feel markedly better by day 2 of her hospitalization, her renal function continued to deteriorate with the creatinine peaking at a level of 8.0 on hospital day 5. Just as rapidly, however, her renal function subsequently improved and, by day 14, her creatinine had normalized (1.2) on antibiotics alone. Also complicating her course was the development of mild pancreatitis and cholangitis, both of which resolved with IV hydration, bowel rest, metronidazole and levofloxacin. Repeat blood and urine cultures on hospital day 4 were negative for bacteria. Despite the sepsis, the patient’s temperature was normal throughout her hospitalization except for two minor elevations – 100.3oF on day 2 and 100.1oF on day 10.
The patient was transferred to a rehabilitation unit where she continued to improve until day 11 when she complained of bilateral knee pain and swelling. The rheumatology consultant observed large, warm effusions of both knees, and a firm 6 cm mass in the proximal left calf. Deep venous thrombosis was ruled out by venous plethysmography. MRI scan of the left calf revealed a 10 x 8 x 5 cm cystic mass with debris that was “consistent with resolving hemorrhage or abscess formation”. The left knee and left calf mass were aspirated.
MRI scans of the left knee and calf is shown below (Figs. 1 and 2). A 10 x 8 x 5 cm fluid signal is demonstrated in the posterior medial aspect of the left calf. The lesion extends to the level of the left knee joint where a joint effusion is also noted. There was no evidence of underlying bone destruction.
1. [ ] Acute rheumatic fever
2. [ ] Lyme disease
3. [ ] Sepsis complicated by endocarditis, septic knees, and abscess of left calf
4. [ ] Post-streptococcal reactive arthritis
5. [ ] Systemic lupus erythematosus with glomerulonephritis and arthritis
6. [ ] Rheumatoid arthritis
7. [ ] Malignancy with infiltration of the knees and left calf
8. [ ] Ankylosing spondylitis
Aspiration of the left knee and left calf mass revealed inflammatory, but sterile, effusions. Cultures of both fluids were negative for bacteria.
Left knee fluid: cloudy, amber-colored fluid WBC 7,000 cells/mm3
Differential: 90% neutrophils, 4% monocytes, and 6% lymphocytes
RBC 4,500 cells/mm3
Protein 5.0 g/dl
Left calf mass: cloudy, amber-colored fluid WBC 13,250 cells/mm3
Differential: 92% neutrophils, 4% monocytes, and 4% lymphocytes
RBC 92 cells/mm3
Protein 5.5 g/dl
Because of the patient’s recent renal failure and low hematocrit, her arthritis was treated with low dose prednisone rather than nonsteroidal anti-inflammatory drugs (NSAIDs). Intraarticular corticosteroid injections were considered but rejected at this time because of lingering concern that her arthritis represented resolving septic arthritis. She was discharged to home but, on return to clinic one week later, her arthritis had not improved. Both knees were re-aspirated and analysis of the fluids was again consistent with an inflammatory process. Cultures of both knees were again negative for bacteria.
Right knee: WBC – 17,871; RBC – 2,109 cells/mm3
Left knee: WBC – 6,993; RBC – 1,221 cells/mm3
The patient’s knees were subsequently injected with a corticosteroid preparation (triamcinolone 40 mg each). Additional bloodwork was obtained: HLA-B27 antigen – negative
RF negative; ANA 1:80 (negative)
ESR 128 mm/hr
At follow-up one week later, her arthritis had completely resolved. At follow-up one month and three months later, the arthritis had not recurred.
DIAGNOSIS: Probable post-streptococcal reactive arthritis complicated by Baker’s (popliteal) cyst.
Reactive arthritis is a term used to describe a sterile arthritis occurring after a documented infection elsewhere in the body. The best-known examples in this category are the reactive arthritides associated with bacterial dysentery caused by Salmonella, Shigella, and other enteropathic strains that occur in HLA-B27 positive individuals (originally known as Reiter’s syndrome). The inflammatory reaction in the joints is believed to be due to immune complex deposition in, rather than direct infection of, the joints. That a reactive arthritis might also occur in response to a streptococcal infection was first suggested in the 1950s by Friedberg (ref 1) and later by others.(refs. 2-5) These patients were believed to represent an entity distinct from acute rheumatic fever (ARF) based on the following observations:
- they had no evidence of carditis;
- the latent period between the antecedent streptococcal infection (usually pharyngitis) and the arthritis was shorter (1-2 wks) than the 3-4 wks usually seen with ARF;
- the arthritis was more severe and prolonged and unusually resistant to treatment with salicylates, in contrast to the migratory, salicylate-sensitive arthritis generally observed with ARF;
- some patients had accompanying post-streptococcal glomerulonephritis that is only rarely seen in conjunction with ARF.
We present a patient whose clinical features are more consistent with post-streptococcal reactive arthritis than ARF. ARF typically attacks several joints in quick succession, each only for a brief period of time (1-3 days), resulting in the typical pattern of migratory polyarthritis. In contrast to this, but similar to the patients described above, our patient’s arthritis was severe and persistent and unresponsive to low dose steroids. [Interestingly, her arthritis was further complicated by rupture of the synovial effusion into the calf, resulting in a Baker’s (popliteal) cyst.] In addition, she lacked evidence of carditis but developed post-streptococcal glomerulonephritis (a clinical diagnosis based on the rapid reversal of her renal failure). Inconsistent with either a diagnosis of ARF or post-streptococcal reactive arthritis, however, was the fact that this patient had documented streptococcal sepsis, in addition to (presumed) streptococcal pharyngitis.
This case demonstrates the complexity of attempting to categorize musculoskeletal complications of streptococcal infections into discrete non-overlapping entities. Some authors have argued, for example, that post-streptococcal reactive arthritis is not a distinct entity but rather an adult variant of ARF.(ref 6) One could argue, in fact, that our patient fulfills Jones criteria for the diagnosis of ARF (arthritis + fever + elevated ESR + evidence of streptococcal infection). However, this is debatable since her arthritis was very atypical and her temperature elevations were minor, occurring only twice and ten days apart. It has also been suggested that diagnoses of ARF and post-streptococcal reactive arthritis should be discarded if the patient is demonstrated to be septic.(ref 7) Given the positive blood cultures in our patient, one could argue that her arthritis was septic in origin. The negative cultures of her joint aspirates do not definitively rule out this diagnosis (septic arthritis) since they were obtained while on, and after completing, a course of penicillin. However, septic arthritis seems highly unlikely, since the arthritis did not appear until 10 days after antibiotics were initiated and she was already proven to be culture negative.
Other non-streptococcal causes of inflammatory arthritis are unlikely in this patient in light of her documented streptococcal sepsis. She had no antecedent rashes such as erythema chronicum migrans, making Lyme arthritis unlikely. Although systemic lupus erythematosus causes arthritis and glomerulonephritis, her ANA was negative and her renal failure was atypical for lupus in that it reversed rapidly without immunosuppressives and was not associated with hypocomplementemia. Malignancy with involvement of the calf and knees is unlikely since tumors more typically cause arthralgias than arthritis, and the MRI of the left leg did not disclose a sarcoma or bony tumor. Rheumatoid arthritis is unlikely since it usually begins in the small joints of the hands and feet, and the onset is usually (although not always) more insidious. HLA-B27 associated reactive arthritis was ruled out with the negative B27 antigen.
In summary, although definitive categorization remains uncertain, the most satisfying diagnosis for this patient is streptococcal sepsis and pharyngitis with post-streptococcal reactive arthritis and glomerulonephritis. A diagnosis of acute rheumatic fever can also not be definitively excluded.
3. Arnold MH and Tyndall A. Poststreptococcal reactive arthritis. Ann Rheum Dis 48:686, 1989
4. Powell RJ and Jenkins S. Poststreptococcal reactive arthritis. Ann Rheum Dis 49:270, 1990
5. Gutierrez-Urena S, Molina J, Molina JF, et al. Poststreptococcal reactive arthritis, clinical course, and outcome in six adult patients. J Rheum 22:9, 1995