Levofloxacin-resistant Streptococcus mitis Endophthalmitis: A Unique Presentation of Bacterial Endocarditis
Amreen Dinani M.D1, Nessrine Ktaich M.D.1,2, Carl Urban, Ph.D.1,2,4. David Rubin, M.D.1,2,3.
Department of Medicine, New York Hospital Queens1
Division of Infectious Diseases, New York Hospital Queens2
Department of Medicine, Weill Cornell Medical College3
Department of Microbiology, Weill Cornell Medical College4
Abstract: Endogenous endophthalmitis is a rare complication of infective endocarditis and has been decreasing due to treatment with effective antibiotics. We highlight a case of endogenous endophthalmitis due to levofloxacin-resistant Streptococcus mitis presenting as infective endocarditis. The patient was an 85-year-old Caucasian male with a medical history significant for hypertension and diabetes mellitus, who was referred to our hospital for evaluation of positive blood cultures without fever, chills, malaise, back pain or other systemic symptoms while being treated for vitritis and endogenous endophthalmitis. He was discharged on levofloxacin 750 mg PO for 14 days. After 8 days of outpatient therapy, the patient presented to our emergency department. His right eye was injected, with normal movement of extra ocular muscles and decreased visual acuity (20/160). Auscultation of the heart revealed III/VI holosystolic ejection murmur heard best at the apex. Transesophageal echocardiography revealed a 1.4cm x 1.1cm independently mobile lesion on the posterior leaflet of the mitral valve compatible with a vegetation associated with moderate mitral regurgitation. The patient was started on IV ampicillin (2gm every 4 hours), intravenous ceftriaxone (1gm every 12 hours) and intravenous gentamicin (3mg/kg once daily for two days). Upon insistence of the ID department at NYHQs, the initial alpha-haemolytic streptocoocus reported by the clinical microbiology laboratory at the first hospital was identified as Streptococcus mitis from 2/2 blood cultures. Susceptibility using Etest methodology gave the following MICs (mg/ml): penicillin, 0.125, ceftriaxone, 0.19, gentamicin, 0.24 and levofloxacin, 4.0 by Microscan testing. A PICC line was placed and the patient was discharged with total of 6 weeks of intravenous ceftriaxone (1gm every 12 hours), weekly CBC and outpatient evaluation of possible mitral valve replacement. The full recovery of our patient, including resolution of endophthalmitis, recovery of vision and absence of mitral valve vegetations on transesophageal echocardiography after completion of 6 weeks of antibiotic treatment, suggests that the endogenous endophthalmitis was an embolic manifestation of infective endocarditis. In conclusion, our patient presented with endogenous endophthalmitis with blood cultures positive for Streptococcus mitis in the setting of native mitral valve endocarditis, as the presumed source of infection. Endogenous endophthalmitis even without constitutional symptoms should be highly considered as a manifestation of septic emboli due to subacute endocarditis. This case highlights the importance of obtaining pertinent information from prior hospital admissions including antibiotic susceptibilities. Lack of timely communication between the microbiology laboratory and treating physician during the first hospital visit, patient lost to follow up or reliance of quinolone empiric therapy can lead to a dire outcome.