A Caution When Diagnosing Disease

Jonathan Kay, MD


April 15, 2015

This feature requires the newest version of Flash. You can download it here.

Hello. I am Dr Jonathan Kay, professor of medicine at the University of Massachusetts Medical School and director of clinical research in the Division of Rheumatology at UMass Memorial Medical Center, both in Worcester, Massachusetts. Welcome to my Medscape blog.

I want to share a very interesting and instructive case. The patient is an 80-year-old woman who I have been following for several years because of gout. She also has hypertension, coronary artery disease, type 2 diabetes mellitus, and glaucoma. After a fall during the summer of 2014, she experienced chest wall pain and was found to have a fractured rib. In October 2014 she noted the relatively sudden onset of left-sided temporal headache with pain in her jaw when chewing. She had no visual symptoms, no blurred vision, double vision, or transient loss of vision.

Her primary care physician appropriately thought of giant cell arteritis as a possible diagnosis. He ordered laboratory testing that revealed elevated acute phase reactants with a markedly elevated erythrocyte sedimentation rate to 105 mm/hour and a C-reactive protein above 330 mg/L. Given her clinical presentation and laboratory findings, he prescribed oral prednisone 40 mg daily and arranged for her to have a left temporal artery biopsy. For this purpose, a 4-cm piece of temporal artery was excised. The pathologist looked carefully with multiple sections and found no evidence of vasculitis or granulomatous change, but only atherosclerotic changes.

Improved but Persistent Symptoms

This woman's symptoms improved somewhat, with improvement in the jaw claudication and almost complete resolution of her headache, but she still experienced some residual symptoms and her prednisone dose was increased to 60 mg daily. When the temporal artery biopsy returned as negative, she underwent a contralateral temporary artery biopsy, which also revealed no evidence of giant cell arteritis.

With prednisone 60 mg daily, the patient's symptoms further improved, but she was now troubled by low back pain that was most symptomatic in mid-afternoon. Her primary care physician obtained MRI of her lower back, which to our surprise revealed multiple areas of increased T2 signal in her lumbar spine, consistent with bone edema or infection. She also underwent abdominal CT scanning, which showed multiple lesions in her liver and spleen. An interventional radiologist performed percutaneous drainage of one of these liver lesions; the lesion drained pus, which was cultured and grew Streptococcus intermedius; blood cultures also grew S intermedius. The patient deteriorated and required intensive care hospitalization for several weeks. While in the hospital, echocardiography showed questionable vegetation on one of her heart valves. She was treated with 9 weeks of intravenous ceftriaxone 1 g daily, improved, and eventually was discharged from the hospital.

A Lesson in Anchoring Bias

I saw the patient in follow-up recently and she looks quite good. She was overweight before this happened and she lost weight because of her illness, which probably stands her in good stead.

This is a case of anchoring bias. At presentation her jaw claudication, headache, and markedly elevated acute phase reactants suggested the diagnosis of giant cell arteritis. When the first temporal artery biopsy was negative, we appropriately went for a second temporal artery biopsy, but we were anchoring our thoughts on this initial clinical impression of giant cell arteritis and not broadening our differential diagnosis to prepare ourselves for her true diagnosis of S intermedius bacteremia and sepsis. In retrospect, this organism was probably introduced into her blood at the time that she sustained the rib fracture during her earlier fall, and the organism festered until it manifested itself with these multiple abscesses during the winter of 2014.

I now am very much aware of the pitfalls of anchoring bias, and this case was instructive not only to me but also to many of my colleagues. I hope that you find this case as interesting as I did. I look forward to seeing you again on Medscape. Thank you for your attention.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: