Man With Nausea, Fever, and Rash Following a Diarrheal Illness

A 52-year-old man complains of nausea, fever, and malaise following a 2-day diarrheal illness that developed at the end of a family vacation in New England. Two family members suffered a similar illness, characterized by watery diarrhea. Symptoms developed in all who were affected within 24 hours of eating hamburgers at a local restaurant.

HISTORY
The patient has mild essential hypertension, for which he takes a ß-blocker.

PHYSICAL EXAMINATION
The patient is pale and in moderate distress. Temperature is 38^oC (100.8^oF);heart rate, 78 beats per minute and regular; respiration rate, 24 breaths per minute;and supine blood pressure in the right arm, 100/60 mm Hg with no orthostatic changes. Mucous membranes are pale, and there is an extensive petechial rash on the lower extremities.

LABORATORY STUDIES
Hemoglobin level is 7.1 g/dL; hematocrit, 22%; and mean corpuscular volume, 93 fL. Leukocyte count is 16,000/μL, with 83% neutrophils and 5% band forms. Reticulocyte count is 200,000/μL, and platelet count is 8000/μL. Prothrombin time (PT) is 11 seconds, activated partial thromboplastin time (PTT) is 32 seconds, and plasma fibrinogen level is 512 mg/dL. Blood urea( nitrogen level is 78 mg/dL; serum creatinine level, 8.5 mg/dL; and serum lactate dehydrogenase, 1107 U/L. Results of a direct antiglobulin (Coombs) test are negative. Urinalysis reveals 3+ blood with gross hematuria, and a blood smear demonstrates numerous schistocytes.

Which therapy is most likely to be effective in this patient?
A. Plasmapheresis.
B. High-dose corticosteroids (eg, prednisone(, 60 mg/d).
C. Broad-spectrum antibiotics.
D. Quinine and clindamycin(.

(Answer and discussion on next page.)



CORRECT ANSWER: A
This man has microangiopathic hemolytic anemia and thrombocytopenia. These findings, together with the history of a diarrheal illness related to ingestion of hamburger (likely less than fully cooked), strongly suggest thrombotic thrombocytopenic purpura (TTP)/hemolyticuremic syndrome (HUS). HUS is very similar to TTP. In both conditions there appears to be toxin-induced endothelial cell injury, abnormally large hyperaggregant factor VIII(_VWF multimers in the circulation, and subsequent platelet thrombus formation (which causes thrombocytopenia by consumption) associated with a microangiopathic hemolytic anemia.¹ However, in HUS-unlike in TTP-there is a clear statistical association with a diarrheal illness caused by enterotoxin-producing enteric pathogens. In addition, the clinical consequences of HUS tend to be more localized to the kidneys.

Both HUS and TTP are managed by plasmapheresis with plasma infusion (choice A). The TTP variant is extremely responsive to plasmapheresis; this therapy has transformed a mortality rate higher than 80% into a survival rate of more than 80%.

New research has elucidated the biochemical pathogenesis of the syndrome: TTP develops because of the genetic absence or immune-mediated dysfunction of a cysteine( protease that normally cleaves hyper-adhesive macromultimers of factor VIII_VWF into their more physiologic sites.^2,3 The pathogenesis of HUS is more or less the same. This pathophysiology explains the efficacy of pheresis, which removes these high molecular weight multimers. Most patients with TTP/HUS who receive pheresis survive, although some degree of chronic renal insufficiency is a common sequela in patients with HUS.

Unlike in disseminated intravascular coagulation (DIC), the soluble coagulation system is not activated in HUS/TTP. Therefore, the PT, activated PTT, and fibrinogen level are normal. Broad-spectrum antibiotics for presumed sepsis (choice C) would be highly appropriate if the patient had DIC triggered by infection. However, there is no evidence of that pathophysiology here.

The negative Coombs test demonstrates that this patient's hemolysis and thrombocytopenia are not caused by autoantibodies. Thus, corticosteroids (choice B) would not be helpful.

Babesiosis, an infection endemic to New England, is associated with nonimmune hemolysis-but without microangiopathic red cell features. Thus, quinine( and clindamycin (choice D) would not be useful here.

Outcome of this case. HUS/TTP was diagnosed based on the clinical findings, and plasmapheresis was promptly initiated. By day 8, the patient had improved clinically, hemoglobin was 9.5 g/dL, platelet count was 153,000/µL, and there were only rare schistocytes on smear. Results of a stool culture were positive for a toxinproducing Escherichia coli 0157 strain. Two months after discharge, results of blood studies were normal, but the serum creatinine level remained elevated at 2.3 mg/dL.