Rheum vs. ID II

This is a continuation of the case presented in the previous post.

Neurology determined that this was not bacterial meningitis. I went to consult for infectious disease and though we were concerned about septic arthritis, we did not feel that this was a slam dunk diagnosis. Septic arthritis presents with WBC counts into the 100,000s and the most common organisms are Staph and Strep, but those organisms are almost always seen on gram stain or culture. It is so incredibly unusual for the gram stain and culture to be negative (prior to antibiotics) that we didn't feel the most likely diagnosis was a classic gram positive joint infection. The alternative is a disseminated Neisseria gonorrhoeae infection which classically presents as a migrating large joint infection in a young sexually active patient, sometimes with a negative gram stain and culture. We recommended keeping vancomycin for gram positives including MRSA and considering ceftriaxone for gonococcus.

Rheumatology evaluated the patient and agreed that although a septic arthritis was concerning, nothing had "declared itself." The patient did not have fever or leukocytosis. They recommended stopping all antibiotics and seeing what happened clinically. Despite our conflicting advice, the primary team decided to stop antibiotics.

Clinically, she remained about the same; she didn't do better and didn't do worse. Serial taps continued to show WBCs from 40,000 to 100,000, no crystals, no organisms. Blood cultures were negative. Gonococcus swabs were negative. Hepatitis serologies were negative. Strangely, her sugars were elevated (200-300) but HgbA1C was 6.6. We began to consider more unusual infections such as TB arthritis (usually monoarticular, but large joint), rheumatic fever (ASO negative, no other symptoms), Brucella (unpasteurized milk history, but no fever, Brucella serology negative), Lyme disease (no exposure, we decided not to order titers), and rheumatologic conditions (ANA negative, RF negative).

We were stumped. Our differential diagnosis remained atypical presentation of gram positive septic arthritis, atypical presentation of disseminated gonococcal infection, atypical presentation of mycobacteria or fungus, reactive arthritis without preceding GI/GU infection, psoriatic arthritis without skin lesions, and other HLA-B27 seronegative arthritis.

Then pathology came back for a synovium biopsy done by orthopedics. It read "fibrinopurulent material." We went down to review the pathology and it seemed more consistent with infection than an inflammatory process. Yet all the stains for bacteria, mycobacteria, and fungi were negative. Nevertheless, we decided to start the patient on antibiotics (ceftriaxone for presumed DGI). We did not think there was enough to push us to start vancomycin.

Three days into the ceftriaxone and she was not getting better; we would have expected improvement by now. At that time, rheumatology gave in and started anti-inflammatories (naproxen 500mg BID). She immediately got relief from that; over the next few days, her neck pain resolved, her knees and ankle had better range of motion. Although we were stumped, she was getting better. We decided to complete a week of ceftriaxone and continue the naproxen. The only new information was an AP pelvis x-ray which found fused SI joints suggestive of a seronegative arthritis.

At discharge, the ID team still feels this is noninfectious despite the extraordinarily high WBC counts in the synovial fluids. The rheumatology team still thinks it's an infection that hasn't grown out. It's a little unsatisfying, but that's clinical medicine.