Use of Veterinary Vaccines Associated with Illness in a Man

Sir—We read with interest the articles by Berkelman [1] and Weil et al. [2], and we wish to report a case of human illness associated with the use of vaccine strain Brucella mellitensis REV 1. The patient was a 76-year-old man. His medical history included arterial hypertension, herpes zoster, nonsteroidal anti-inflammatory drug– induced digestive hemorrhages, and a total knee replacement, which was performed 5 years before his current visit. The patient presented with inflammatory prosthetic knee pain associated with swelling of the joint. No other local or general syndrome involving fever was present. A nonspecific, self-limited episode of fever syndrome lasting for 1 week had occurred 45 days before presentation. The patient later reported that he had handled, sacrificed, and eviscerated 2 lambs at his farm, which is located in an area where brucellosis is endemic. Initial examination had been performed at another hospital and revealed a globular sedimentation rate (GSR) of 73 mm/h (normal value, 20 mm/h), a C-reactive protein (CRP) level of 39.9 g/dL (normal value, 10 g/dL). The Brucella titer was found to be 1:320 by agglutination testing and 1:640 by Coombs testing. No treatment had been given. On the basis of the patient’s referral data, the surgeon at our hospital performed an additional serological study, the results of which were as follows: hematometry findings were normal; GSR, 73 mm/h; CRP level, !0.5 g/dL; Brucella titer (by agglutination test), 1:1280; and glucose level, 13 mg/dL. Synovial fluid analysis revealed 50 cells/mm, of which 80% were polymorphonuclear cells. Results of blood culture were positive for Brucella species, which was identified as REV 1 (vaccine strain) Brucella mellitensis. After assessing the stability of the patient’s prosthesis, antibiotic treatment with doxycycline and rifampin was initiated, on the basis of previous experience with nonprosthetic cases. The patient’s condition improved satisfactorily, and he was discharged, subject to clinical and analytical follow-up. After receipt of 6 weeks of treatment, clinical evolution was favorable, until 1 year later, when the patient reported a painful lump on the prosthetic knee. No general fever syndrome was present. Arthrocentesis and synovial fluid culture were again performed, revealing the presence of the same species of Brucella. We contacted the Department of Animal Health of the University of Zaragoza (Zaragoza, Spain) and were told that the strain was unresponsive to aminoglycosides; thus, we decided to administer antibiotic treatment with rifampin, doxycycline, and levofloxacin, together with surgical cleansing. The patient was asymptomatic at the time of communication, 2 years later, and had a good prosthetic function. Antibiotic treatment for a 6-month period was effective. The use of modified germs for the treatment of cattle may cause certain agents to enter into the human cycle and lead to pathology. We have no experience in this field, and such situations are unknown to us. This case is an example of transmissions that could occur when live germs are used in animal vaccines. The drug susceptibility characteristics for such events are anomalous. We should remain alert and be prepared for potentially similar cases in the future.