Close Reading Sheds New Light on a Puzzling Rash

In Dr Sean Eric Koon's Case In Point, "Puzzling Rash in an Older Woman"(CONSULTANT, April 15, 2003, page 629), I agree with his conclusion that thispatient had cutaneous leukocytoclastic vasculitis (CLV) (Figure). I also agreethat she met the American College of Rheumatology's 1990 criteria for a diagnosisof hypersensitivity vasculitis.¹ However, given the patient's history and laboratoryresults, I believe further evaluation was warranted to determine whether the medication was indeed to blame for her CLV or whether an underlying systemicdisease was responsible.Her white blood cell (WBC) count was 72,000/?L. CLV is known to producea mild leukocytosis--presumably caused by the inflammatory response of the vasculitis.Thus, one would expect to see only a slight elevation in the WBC count.Also, if the elevation had been produced by the inflammatory response of the CLV,the patient's erythrocyte sedimentation rate (ESR) would have been significantlyelevated. However, her ESR was 12 mm/h, which is essentially normal in awoman this age.The discovery of a value that is not consistent with the disease process makesme question Dr Koon's final diagnosis--or at least want to add to his differentiala disease that could be responsible for both the CLV and the level of leukocytosisseen here: hairy cell leukemia.Other facts in the case that tend not to support the conclusion that the patient'sCLV resulted from a drug reaction include the following:One would expect to see systemic symptoms, such as fever, malaise, anorexia,and/or myalgias if a drug reaction caused the CLV; this woman reportedly didnot experience any of these.Rashes associated with CLV produced by a drug reaction are generally describedby patients as pruritic, painful (sometimes significantly so), and/or associatedwith paresthesias, such as a burning or stinging sensation; this patient's rash isdescribed as "painless and nonpruritic."Thus, although an exogenous agent such as trimethoprim-sulfamethoxazole(TMP-SMX) can cause CLV, it would have been prudent in view of the findingsin this case to search for an endogenous cause, such as an underlying systemicdisease or malignancy.----Pamela Moyers Scott, MPAS, PA-C
Williamsburg, WVa
Thank you for your comments. You detected a typo; this patient's leukocytecount was actually 7200/?L. A value of 72,000/?L would indeedbe of concern and would prompt a new differential diagnosis. HerWBC count when last checked was 7300/?L.Ultimately, my determination was that the patient's rash was mostlikely triggered by a viral infection and not by the TMP-SMX. I felt that anotherhealth care provider had inappropriately treated her upper respiratory tractinfection with an antibiotic, so I stopped the medication. Because I could notcompletely rule out the antibiotic as a cause of this potentially serious condition,I recommended that she avoid it in the future. This is yet another exampleof how the treatment of colds with antibiotics can confuse the clinical pictureand possibly harm the patient.--Sean Eric Koon, MD
  &nbspFontana, Calif

References:

REFERENCE:1.

Calabrese LH, Michel BA, Bloch DA, et al. The American College of Rheumatology 1990 criteria for theclassification of hypersensitivity vasculitis.

Arthritis Rheum.

1990;33:1108-1113.