Positron emission tomography and reframing vasculitis as a spectrum of disease when investigating a patient with a fever of unknown origin

Abstract

A retired businessman presented to the infectious diseases department with a history of ongoing fevers and myalgia and raised inflammatory markers. This continued despite adequate antibiotic treatment of an epididymo-orchitis. Extensive investigations, including bone marrow and liver biopsies and a positron emission tomography, did not reveal a cause but showed reactive change in the bone marrow. Later, he developed a vasculitic rash and vision loss due to non-arteritic anterior ischaemic optic neuropathy. High-dose steroids were immediately initiated. A temporal artery biopsy was performed, which confirmed a healing large vessel vasculitis, possibly giant cell arteritis. He has responded very well to therapy. We must better appreciate the limitations of positron emission tomography in investigating a fever of unknown origin. The case also encourages awareness of autoimmune disorders as the leading category of causative diseases for this in older age groups.

Background

Despite extensive imaging, cultures and biopsies, a cause for our patient’s fever was not found. Clinical signs during careful follow-up led to the diagnosis—eventually proven by a temporal artery biopsy. Our case highlights the importance of understanding the limitations of our most technologically advanced investigations and the need for a broad approach to conditions presenting with fever.

Case presentation

A retired businessman was referred to the infectious diseases team with a 3-week history of fevers and myalgia. This had persisted despite the successful treatment of an epidydimo-orchitis with 2 weeks of ciprofloxacin. The myalgia had started in his shoulders but predominantly affected his buttocks, thighs and right calf. He had persistently raised inflammatory markers and a mild transaminitis. Further history was unremarkable; specifically, there was no history of rashes, dysuria or visual disturbance. He maintained a monogamous relationship with his wife and they enjoyed playing music and hiking; there was no history of known tick bites. He had no significant travel history.

In his past medical history, he had a cholecystectomy and a diagnosis of localised prostate cancer with stable prostate-specific antigen levels. There was no evidence of prostatitis or prostate abscesses clinically nor on MRI. His family history revealed ankylosing spondylitis and rheumatoid arthritis on his mother’s side.

Investigations

During admission, our patient developed significantly raised liver enzymes—ultrasound and MRI only showed non-specific cysts. Liver biopsy did not show architectural or inflammatory abnormalities but did show mild steatosis. Liver vessel histology showed predominantly reactive T cells but no intravascular lymphoma. MRI of the prostate, spine and positron emission tomography scan showed ‘reactive’ bone marrow changes.

The following investigations, blood/urine markers and serology were requested and found to be negative: thyroid function; HIV; hepatitis B, C and A; cytomegalovirus; antistreptolysin O titre; chlamydia, gonorrhoea and syphilis; mumps virus; autoantibody screen including antinuclear antibody, antineutrophil cytoplasmic antibody and anti-dsDNA; myeloma; mycoplasma IgM; Epstein Barr virus IgM; Legionella Ag; Borrelia; HLA-B51; Toxoplasma; beta-2 microglobulin; liver autoantibody screen; cerebellum autoantibody screen (Yo, Hu, Ri); Coxiella burnetii; Histoplasma; QuantiFERON-TB testing; blood cultures x 13 (including prolonged sets for mycobacteria and Brucella); viral throat swabs; urine cultures x 3; echocardiogram and ultrasound right lower limb. The patient’s ferritin was raised at 1485 and his iron profile showed a low iron.

At this point, he was discharged to our outpatient follow-up. Due to his ongoing myalgia, he now required a walking stick to aid his mobility.

Differential diagnosis

We considered a drug reaction to ciprofloxacin as a plausible explanation for the liver and bone marrow changes. Most case reports of fluoroquinolone-induced vasculitis report onset of the rash within 2 weeks of starting the drug, with resolution on withdrawal.^1–3 One case report detailed a patient with ongoing symptoms 6 months after withdrawal of the drug, though this appears to be uncommon in the literature.⁴ Our patient’s case was also unusual in that his rash only developed 2 months after finishing the ciprofloxacin. We were also concerned that there was an ongoing indolent infection such as tuberculosis that was concealed by the initial use of ciprofloxacin. Other differential diagnoses included intravascular lymphoma or a rheumatological disease.

Treatment

One month after discharge, the patient developed a non-blanching rash over his body (including palms and face) which responded to steroid cream —he brought photographs with him to his next clinic visit (figures 1 and 2). Shortly before that visit, he also found that the sight in his right eye had significantly decreased. Ophthalmology review revealed non-arteritic anterior ischaemic optic neuropathy. Combined with his history, recent vasculitic rash and raised inflammatory markers, we strongly suspected vasculitis and admitted him from clinic for methylprednisolone 1 g once a day for 3 days followed by prednisolone 60 mg.

Figure 1.

The patient’s vasculitic rash.

Figure 2.

Closer view of the vasculitic rash affecting the palms of the hand.

Outcome and follow-up

Our patient’s inflammatory markers and liver enzymes have all normalised. He is now independently mobile and his vision has improved.

Temporal artery biopsy taken following the initiation of high-dose steroids showed a cross-section of artery with focal destruction of the internal elastic lamina and mild intimal proliferation. Although there was no evidence of giant cells, these features are in keeping with the healed phase of temporal arteritis. He is now under the care of the rheumatologists and being treated for systemic vasculitis.

Discussion

Autoimmune disease is an important differential diagnosis in fever of unknown origin, particularly in the older patients where it is the more common causative group of diseases.⁵ Giant cell arteritis is a large and medium-sized vessel often associated with polymyalgia rheumatica.^{6 7} It classically presents as headache, jaw, scalp or tongue claudication, vision loss, fever and myalgia.^{6 8} Some 40% of cases are thought to present atypically.⁸ This represents a diagnostic challenge, in part because there is a lack of awareness among most clinicians that non-classical presentations can be so common. In our patient’s case, the slightly more classical symptomatology that presented later helped us to achieve his diagnosis of systemic vasculitis.

A 2003 review recommends temporal artery biopsy in older patients with fever of unknown origin.⁹ Recent data suggest that autoimmune inflammatory disorders are increasingly found to be the leading cause of fever of unknown origin, confirming this 2003 recommendation.¹⁰ Temporal artery biopsy is the gold standard of diagnosis for giant cell arteritis, but false negatives occur due to the presence of ‘skip’ lesions.⁷ Without typical symptoms, there may be no prompt to perform the test.

18F-fluoro-deoxyglucose positron emission tomography scan has been shown in a number of studies to be a useful tool in the diagnosis of vasculitis. One study found 18F-fluoro-deoxyglucose uptake by large vessels yielded a sensitivity of 66.7% and a specificity of 100% for giant cell arteritis diagnosis.¹¹ However, these patients either had a known diagnosis of the condition or were already suspected of having it. Two review articles of this modality of scanning for diagnosis of large vessel vasculitis and giant cell arteritis could not calculate sensitivity and specificity if temporal artery biopsy was negative. This is because there is no gold standard for comparison and no standardised interpretation technique.^{7 12}

Our patient’s scan showed widespread bone marrow oedema not confined to the classical polymyalgia rheumatica areas and with no typical signs of vasculitis. Additionally, MRI and bone marrow and liver biopsies all showed ‘reactive’ changes. A number of case reports have identified a possible link between myeloproliferative/myelodysplastic diseases and giant cell arteritis.^{13 14} One case report diagnosed giant cell arteritis from a bone marrow biopsy based on the presence of an affected vessel.¹⁵ It is unclear overall though, what the effect of vasculitis is on bone marrow.

In the absence of typical signs and symptoms and due to his unusual ophthalmology examination, our patient had a significant delay in his diagnosis. This probably resulted in some loss of vision. There is more work to be done to find investigations that help clinicians more readily make this difficult diagnosis.

Patient’s perspective.

After 7 weeks of illness (5 of which had been spent undergoing tests in hospital), it had still not been possible to make a diagnosis of autoimmune disorder leading to vasculitis. The key evidence emerged at 10 weeks (rash) and 13 weeks (headache, pain in the eyes and some sight loss). With regard to the possibility that my condition might have been a rheumatoid problem/autoimmune disorder, I think some signs were there from the outset. At times, it felt as though all life had been drained from me. It seems, however, that symptoms didn’t come in the ‘usual’ order to indicate one thing or another. I certainly wish I had been told that if any pain started around or in the eye, I should make immediate contact, since there is such a short time before serious eye damage can occur. It seems to me that everyone from the local general practitioner to the doctors in hospital might be more open to the possibility of a rheumatoid problem/autoimmune disorder, especially if up to 40% of cases present atypically. What is ‘usual’ appears to be the central point.

Learning points.

• Positron emission tomography is often used to find a diagnosis in patients with a fever of unknown origin. It is important that the clinician understands the limitations of this test for the various differential diagnoses being considered.

• Autoimmune diseases are a leading category of causative diseases in fever of unknown origin, particularly in older age groups.

• Giant cell arteritis is a large vessel vasculitis that can present ‘atypically’ in as many as 40% of cases. Large vessel vasculitis should therefore be considered as a spectrum of disease disorders to avoid missing ‘atypical’ cases.

• All fevers of unknown origin should be considered for temporal artery biopsy if all other tests are inconclusive, with the caveat that there is a chance of a false-negative result.

• Ciprofloxacin should be considered as a possible causal agent of systemic vasculitis.