Dental Health Dental Health and Viridans Streptococcal Bacteremia in Allogeneic Hematopoietic Stem Cell Transplant Recipients
Total Page:16
File Type:pdf, Size:1020Kb
Bone Marrow Transplantation (2001) 27, 537–542 2001 Nature Publishing Group All rights reserved 0268–3369/01 $15.00 www.nature.com/bmt Dental health Dental health and viridans streptococcal bacteremia in allogeneic hematopoietic stem cell transplant recipients CJ Graber1, KNF de Almeida1, JC Atkinson2, D Javaheri2, CD Fukuda3, VJ Gill3, AJ Barrett4 and JE Bennett1 1Clinical Mycology Section, Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, 2National Institute of Dental and Craniofacial Research, 3Microbiology Service, Department of Clinical Pathology, Warren Grant Magnuson Clinical Center, and 4Stem Cell Allotransplant Unit, Hematology Branch, National Heart, Lung and Blood Institute, National Institutes of Health, Bethesda, MD, USA Summary: laxis with fluoroquinolones which have poor Gram positive activity.11 The impact of dental health as a risk factor for Viridans streptococci were the most common cause of viridans streptococcus infection is not known. bacteremia in 61 consecutive myeloablative allogeneic hematopoietic stem cell transplant (HSCT) recipients, occurring in 19 of 31 bacteremic patients (61%) during Patients and methods the period of post-transplant neutropenia. Seven of the 19 had more than one viridans streptococcus in the Patient characteristics same blood culture. Twenty isolates from 15 patients were Streptococcus mitis. Most viridans streptococci Medical charts and dental records were reviewed of 61 con- were resistant to norfloxacin, used routinely for prophy- secutive patients with hematological malignancies receiv- laxis. Comparison of the 19 patients with viridans strep- ing myeloablative allogeneic HSCT at the National Heart, tococcal bacteremia with a contemporaneous group of Lung and Blood Institute in the 21-month period from Janu- 23 allogeneic HSCT recipients with fever and neutro- ary 1997 to October 1999 (NIH protocols 93-H-0212, penia but no identified focus of infection found that 94-H-0092, 97-H-0099, 98-H-0122, 99-H-0046). patients with viridans streptococcal bacteremia were The conditioning regimen during the pre-transplant per- more likely to have severe intraoral pathology while iod was usually total body irradiation followed by two neutropenic (26% vs 0%) and slightly shorter interval infusions of high-dose cyclophosphamide (60 mg/kg i.v.). between the last dental procedure and the onset of neu- Two patients received busulfan in place of total body tropenia (11 vs 14 days). Poor underlying dental health irradiation or fludarabine instead of cyclophosphamide. and the use of norfloxacin thus appear to predispose to Cyclosporine was also given to approximately half of the viridans streptococcal bacteremia. Bone Marrow Trans- 61 cases. Norfloxacin 400 mg p.o. twice a day was begun plantation (2001) 27, 537–542. 7–10 days prior to transplantation and continued until the Keywords: viridans streptococcus; bacteremia; neutro- patient’s neutrophil count was sustained above 1000 penia; transplant; dental; resistance cells/l. Weekly trimethoprim-sulfamethoxazole prophy- laxis for Pneumocystis carinii pneumonia was started on day 30 post-transplant or upon discharge from hospital. Acyclovir was given until 30 days after transplantation as The past decade has seen a change from Gram negative prophylaxis against herpesvirus infections. Intravenous to Gram positive organisms as a cause of bacteremia in immunoglobulin 0.5 g/kg was given weekly for the first 100 neutropenic patients.1 Staphylococcus epidermidis has been days post transplant. G-CSF 5.0 g/kg/day was given from the Gram positive bacterium2 most frequently encountered, day 7 post-transplant until the neutrophil count exceeded largely due to increased use of intravascular catheters.3,4 1 × 109/l for 3 consecutive days or until day 21 post-trans- Viridans streptococci, which are oral commensals, have plant. The first two patients received infusions of donor also been reported to cause bacteremia and may cause adult bone marrow cells; the others received stem cells harvested respiratory distress syndrome, shock, or endocarditis.5–7 from peripheral blood. Factors reported to be associated with viridans strepto- coccal bacteremia have been oral mucositis,5,6,8 high-dose cytosine arabinoside,9 younger age,10 and antibiotic prophy- Blood cultures Most blood cultures were drawn from intravenous cath- Correspondence: Dr JE Bennett, Clinical Center, Room 11C304, National eters. An 8–10 ml blood sample was inoculated into an Iso- Institutes of Health, Bethesda, MD, 20892, USA lator tube (Wampole, Cranbury, NJ, USA) and 3–5 ml into Received 25 April 2000; accepted 10 December 2000 a blood culture bottle (BacTAlert, Organon Teknika, Dur- Dental health and viridans streptococcal bacteremia CJ Graber et al 538 ham, NC, USA). Isolator concentrate was inoculated on to Table 1 Demographic and clinical parameters of viridans streptococ- five plates (two chocolate agar, two blood agar and one cus bacteremic patients and controls brain heart infusion agar) incubated aerobically with 5% CO . Plates and bottles were incubated for 5 days before Cases Controls 2 (n = 19) (n = 23) considered negative. Sex Male 10 13 Female 9 10 Organism identification and antibiotic susceptibility Median age at transplantation (years) 36 39 testing Underlying diagnosis Chronic myelogenous leukemia 7 13 Acute myelogenous leukemia 7 5 Alpha hemolytic or nonhemolytic streptococci were ident- Myelodysplastic syndrome 3 3 ified using a composite of phenotypic characteristics and Chronic lymphocytic leukemia 2 0 biochemical reactions included in the API 20 Strep identi- Acute lymphocytic leukemia 0 1 Multiple myeloma 0 1 fication panel (bioMerieux, Marcy L’Etoile, France) as well Median duration of neutropeniaa (days) 13 12 as additional biochemical testing as needed. The identifi- Median time from neutropenia onset to 4 2 cations were the closest individual species designations that feverb onset (days) Median duration of fever (days) 3 3 could be determined and not the broader group designations c 12 Patients with mucositis 16 18 recently proposed. All isolates were tested for antimicro- Complications from presumed sepsis or bial susceptibility using the Etest method (AB BioDisk, sepsis-related syndrome Solna, Sweden). Multiple strains of the same streptococcal Intensive care unit admission 1 1 species isolated in the same culture were differentiated by Intubation 1 0 colony morphology and subsequent antibiotic resistance Death 1 0 patterns. aThe first day of neutropenia was defined as the first day the patient’s absolute neutrophil count fell below 500 cells/l coincident with or fol- lowing the first day of the conditioning regimen. Neutropenia was con- sidered resolved when the absolute neutrophil count was greater than 500 Analysis of dental health cells/l for at least 2 consecutive days (the first of these dates was recorded as the date of neutropenia resolution). bFever was generally defined as the patient’s first recorded temperature of Fifty-six of the 61 patients were evaluated in the dental 38.0°C or above while neutropenic that was deemed to require antibiotic clinic before transplantation with a complete dental examin- therapy. Resolution of fever was defined by at least 48 continuous hours ation and panographic radiograph. Caries were diagnosed of recorded temperatures less than 38.0°C. Only the first febrile episode during the neutropenic period was included in this analysis. by clinical examination and appropriate bite wing and per- cMucositis was defined by the presence of oral lesions or erythema on iapical radiographs. Alveolar bone loss (a measure of physical examination or mouth and/or throat pain requiring parenteral cumulative periodontal disease) was calculated from pano- analgesia. Records from three control patients were insufficient to deter- graphic radiographics with the aid of a Schei ruler.13 The mine the presence or lack of mucositis. number of days between any dental treatment and the onset of neutropenia was documented. Nineteen of the 56 patients Results had bacteremia from viridans streptococcus. In order not to confound our results and ensure a relatively homogenous Causes of bacteremia control group, 14 patients with documented infection or All 61 patients in this cohort had fever during neutropenia, bacteremia during neutropenia were excluded, leaving 23 31 accompanied by a positive blood culture. Gram negative of the remaining 37 patients to serve as controls to analyze bacteria were isolated in only three episodes (two Escher- for dental risk factors for viridans streptococcal bacteremia. ichia coli and one Capnocytophaga species). In two of While the reviewer of the patients’ dental records was these episodes, Gram positive organisms were also isolated. aware of the patients’ microbiological status, possible den- Of the 30 patients who had Gram positive bacteria, 19 tal risk factors were determined prior to dental chart review. (61%) had viridans streptococci. Of these, eight had one Characteristics of patients with viridans streptococcal bac- strain of Streptococcus mitis, five had two strains of S. mitis teremia and case controls are given in Table 1. There were in the same specimen, one had S. mitis and S. sanguis I in no significant differences between the two groups. the same specimen, and one had S. mitis and S. sanguis II in the same specimen. One had a single strain of S. sanguis I; another had a single strain of S. sanguis II. Two patients had unidentified species of viridans streptococci. The Statistical analysis