Bacteremia due to viridans streptococci in neutropenic patients: A review

/

~ REVIEWS

Bacteremia Due to Viridans Streptococci in Neutropenic Patients: A Review Pierre-YvesBochud, Thierry Calandra, MD, Patrick Francioli, MD, Lausanne, Switzerland

Viridans streptococci have long been considered, with the exception of the ability to cause endocarditis, as minor pathogenic agents. More recently, however, these bacteria have become a major concern in neutropenic patients undergoing a chemotherapeutic treatment. In this high-risk population, they can be responsible for up to 39% of bacteremia cases and are the most frequent cause of this type of infection. The most frequently isolated species in blood cultures are Streptococcus mitis and Streptococcus sanguis II. Viridans streptococcus bacteremia can be accompanied by serious complications, like adult respiratory distress syndrome (ARDS) (3% to 33%), shock (7% to 18%) or endocarditis (7% to 8%). Mortality rates range from 6% to 30%. Case-control studies have identified the following risk factors: severe neutropenia (C 100 neutrophils/mm3), prophylactic antibiotic treatments with quinolone or co-trimoxazole, absence of intravenous antibiotics at the time of bacteremia, high doses of cytosine arabinoside, oropharyngeal mucositis, and heavy colonization by viridans streptococci. The introduction of penicillin in prophylactic antibiotic treatments has reduced the incidence of these infections, but the long-term use of penicillin could be compromised by the emergence of resistant strains. nidans streptococci form a heterogeneous group of atypical streptococci whose classification remains controversial. They are part of the physiological flora of the oropharyngeal region,lz2 and can cause asymptomatic transitory bacteremia after dental treatment.3>4 They have long been considered as significant pathogenic agents only in the specific contexts of dental caries5 and endocarditis, for which they represent the most frequent causative agent.6-g Over the last decade, cases of viridans streptococcus bacteremia with potentially fatal complications have been reported in some immunocompromised pa-

V

Division Autonome de Medecine Preventive Hospitaliere, Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland. Requests for reprints should be addressed to P. Francioli, MD, Division Autonome de Medecine Preventive Hospitaliere, BH19, Centre Hospitalier Universitaire Vaudois, CH-1011 Lausanne CHUV. Manuscript submitted October 4, 1993, and accepted in revised form March 21, 1994.

256

September

1994

The American

Journal

of Medicine@

Volume

tients, particularly newborns,‘O-i4 and patients with oncohematologic diseases treated by chemotherapy.1s-28 The goal of this article is to review the literature concerning this last category of patients more specifically.

EPIDEMIOLOGY Gram-positive microorganisms have become the major cause of bacteremia in patients undergoing high-dose chemotherapy for the treatment of cancer (Table I). In successive studies by the EORTC (European Organization for Research and Treatment of Cancer, International Antimicrobial Therapy Cooperative Group), a significant increase in gram-positive bacteremia has been noted.2g-32 Among them, viridans streptococci were not even counted separately in the first EORTC study, but have become the organisms most frequently isolated in the latest study.31 Similar results have been observed in our institution.27 In the study by Reed et al,26 viridans streptococcus was the second most common organism isolated from blood cultures after coagulase-negative staphylococci, and represented 39% of the bacteremias. At the M.D. Anderson Cancer Center in Texas,24 the viridans streptococcus bacteremia rate has risen from 1 per 10,000 admissions in 1977 to 47 per 10,000 admissions in 1989. However, the incidence of viridans streptococcus bacteremia decreased in some centers since the introduction of penicillin for antibiotic prophylaxis.27,33

RISK FACTORS Several series of viridans streptococcus bacteremia observed in neutropenic patients were the subject of publications between 1983 and 1993. The main characteristics of these patients, detailed in 13 different studies,1525,27are reported in Table II. Four were pediatric studies and 9 concerned adults. They were mostly oncohematological patients undergoing chemotherapy. The most frequent risk factors observed were severe neutropenia, prophylactic antibiotic treatment with co-trimoxazole or quinolone, chemotherapy involving high doses of cytosine arabinoside, the presence of oropharyngeal mucositis, and a strong colonization of patients by vtidans streptococci. Among the 13 series mentioned above, 3 were the subject of case-control studies allowing a more precise evaluation of the risk factors.21~24~27 As the selection criteria for the controls were different, it is not surprising that the studies identified different risk factors (Table III). 97

VIRIDANS STREPTOCOCCUS/BOCHUD

ET AL

TABLE I Single Gram-Positive and Gram-Negative Bacteremias in 5 Studies of Neutropenic Patients EORTC EORTC EORTC Study Trial lzg Trial IV30 Trial Vll131 Nebraskat Years 1973-1976 1983-1985 1988-1991 Episodes of neutropenia 453 872 694 143 145 219 151 80 Bacteremias Gram-negative 103 (71%) 129 (59%) 47 (31%) 3 (4%) 42 (29%) 90 (41%) 104 (69%) 77 (96%) Gram-positive Viridans streptococci NC1 21 (10%) 43 (28%) 31 (39%) ‘Some patients were also included in the EORTC trial VIII. University of Nebraska Medical Cente? K’entre Hospitalier Universitaire Vaudois’27 ¶Not counted separately in this trial. EORTC = European Organization for Research and Treatment

of Cancer,

International

In the study by Elting et al,24 risks were evaluated relative to control patients with gram-positive bacteremia. This study clearly demonstrates that viridans streptococcus bacteremia occurs almost exclusively in highly neutropenic patients: 87% of ease patients affected with severe neutropenia (cl00 neutrophiles/ mm3) and only 20% in control patients. This was confirmed by recent studies20a21,27 in which all patients affected with viridans streptococcus bacteremia were severely neutropenic. Other studies, however most of them less precise in the evaluation of the severity of neutropenia, indicate that bacteremia can occur in patients whose neutropenia is less severe. 1g,23,25 Considering the disparity between case patients and controls in the study of Elting et al, the results of the univariate analysis are difficult to interpret and all characteristics are significantly different. Kern et al”l examined the risk factors for streptococcus viridans bacteremia relative to patients with gram-negative bacteremia and similar duration and severity of neutropenia. In our own study, control patients were also closely matched for neutropenia but were chosen from among patients without bacteremic episode. These two studies show that cytosine arabinoside administered at high doses is a risk factor. In the Kern study serieszl 43% of the case patients against 14% of the controls had received high doses of this agent. In our institution, this proportion was 64% for case patients and 14% for controlsz7 In another study,17 viridans streptococcus bacteremia represented 75% of all septic episodes among children affected with acute nonlymphocytic leukemia, after the introduction of a new protocol including cytosine arabinoside in continuous infusion or at high doses intermittently. The presence of oral mucositis or gingivitis appears in most of the studies.1j,21J3-2”,27In our institution, oropharyngeal lesions were significantly more frequent in case patients (85%) than in controls (55%).27 Moreover, 100% of the case patients had positive surveillance body cultures for viridans streptococci,

Antimicrobial

Therapy

Cooperative

CHUV*$ 1988-1991 341 76 12 (16%) 64 (84%) 25 (33%)

Group.

compared to only 72% for controls. Since the most frequently colonized area was the oropharyngeal region, it can be easily assumed that mucositis facilitates the entry of organisms causing bacteremia. Mucositis can be a consequence of chemotherapeutic treatments including cytosine arabinoside. In the study by SotiropouIos et al,17 12 case patients out of 15 receiving this agent were affected with oral mucositis or gingival bleeding. However, cytosine arabinoside is not the only cause of oropharyngeal lesions: Ringden et al3 observed a decrease in the incidence of viridans streptococcus bacteremia among the recipients of bone marrow transplants after the introduction of acyclovir for prophylaxis of herpetic stomatitis. Outside the oral cavity, Elting et al24 suggests that the stomach could be another possible point of entry: in the presence of ulcerations induced by chemotherapy, streptococcal infections could result from the growth of organisms in an environment that has become alkaline by the administration of antacids or H, antagonists. This hypothesis has been confirmed by the fact that, at the M.D. Anderson Cancer Center, the incidence of viridans streptococcus bacteremia and the use of Hz antagonists has increased in parallel. However, it was not confirmed by other studies.27 The antibiotic prophylaxis also plays a role in the development of viridans streptococcus bacteremia. Co-trimoxazole was used in the oral prophylaxis against gram-negative bacillus infections in neutropenic patients.3”-41 More recently, prophylactic quinolone treatments were administered.4247 In the study by Kern et al, 21 75% of the case patients with bacteremia received a prophylactic quinolone treatment compared with only 22% of the controls. Note that 67% of the controls had received co-trimoxazole prophylaxis; this drug is believed to also favor the development of viridans streptococcus infections. The antibiotics administered empirically for a fever can act as a protection against the development of bacteremia. In the study performed in our institution,27

September 1994 The American Journal of Medicine@ Volume 97

257

I

I

,

I

i , , , , I ,

, , I

I ,

I

!

Cohen15

li

%$-

-

-

-

-

-

-

-

-

-

60%

-

-

-

-

-

-

-

-

-

-

100%

-

30% 30%

other than viridans included. other than viridans included. among which 27 septicemias due to viridans streptococci nonlymphocytic leukemia; HSV = herpes simplex virus.

-

-

-

80%

-

93%

100% -

100% -

-

-

-

$4

-

-

occurred.

12%

-

9%

-

-

-

100% -

56% 100%

-

10%

30%

-

r

10% -

-

-

St%

7%

7.3 -

children

-

1.4

43.5 2.3

20% 70%

adults

adults

loox -

children -

-

-

-

-

45%

43%

11%

19% 70% -

100% -

-

-

1.0

45

adults

-

-

-

-

-

32%

-

90% -

-

23% -

50% 20% 7%

7.3 1.2

children

100%

50%

-

-

66%

0%

-

-

100% I -

T

GO%

17% 83% -

27.5 -

adults

-

-

59% 65%

-

39%

-

-

70% 17% -

87% -

50% 54%

15%

&I

0.6

adults 41.3

Characteristics of the Patients With Viridans Streptococcus Bacteremia in 13 Clinical Studies Sotiropoulos17 Menichettils Villablancalg Valteau*O Kern*l Weisman** Classen Elting24 77-87 76-88 79-88 86-88 72-89 81-88 83-88 15 123 34’ 55+ 6 46 31

-

100%

children -

82-83 10

Hensle16

‘Four episodes of bacteremia due to streptococci +Ten episodes of bacteremia due to streptococci ‘Characteristics for 143 episodes of neutropenia ALL = acute lymphocytic leukemia; ANLL = acute

<500

Years No. of episodes 10 evaluated Epidemiology Population adults Mean age 32.4 Male/female 2.3 ratio Underlying disease ALL 90% ANLL Lymphoma, 10% myeloma Other Bone marrow transplant Neutropenia (cell/mm31

TABLE II

-

-

36% -

-

78%

-

-

-

-

$6

33%

28% 25% 47%

54 1.9

adults

Burden*5 88-90 33

-

-

-

-

-

-

-

47%$ 53%t

-

15%X 94w

70%i

&%

36t 0.9*

Reed*6

100%

72%

41% 32%

9%

82%

64%

4%

11%

85%

100% -

ii

58% 19% 23%

adults 41.3 1.9

Bochud27 88-91 26

I

VIRIDANS STREPTOCOCCUS/BOCHUD

ET AL

TABLE III Risk Factors for Viridans Streptococcus Bacteremia and Criteria Used for the Selection of the Control Patients in Three Case-Control Studies Kernzl Eltingz4 Bochud27 Study characteristics Years No. of patients No. of controls Statistical analysis Controls selection Type of controls Duration and degree of neutropenia Underlying disease Age, sex, observation period Antimicrobial prophylaxis Significant risk factors (P value) Profound neutropenia (~100 cell/mm31 High-dose cytosine arabinoside Oropharyngeal mucositis Viridans streptococcus colonization (body surveillance cultures) Upper gastrointestinal toxicity plus treatment with antacids or H, antagonists Quinolone or co-trimoxazole antimicrobial prophylaxis Quinolone antimicrobial prophylaxis Previous systemic antibiotherapy (see text)

81-88 40 36 univariate

72-89 46 92 multivariate

88-91 22 22 multivariate

gram-negative bacteremia M M NM NM

gram-positive bacteremia NM NM NM NM

no bacteremia M M M M

P = 0.0001

NA

P = 0.01

(P <“0401, (P = 0.02) NA

(P .!A0011 NA

P 10.02 (P <0.05)

NA

P <0.0001

NS

NA (P 50.01) NA

P <.OOOl (P 10.004) NA

NA NA P
M = characteristic matched for cases and controls; NM = characteristic NS = not significant; 0 = significant only in univariate analysis.

not matched

59% of the controls had received, in addition to quinolone prophylaxis, an empirical antibiotic treatment for a fever that was not caused by viridans streptococcus infection. This treatment administered early during the period of neutropenia probably provided them with protection against the development of such infections. Most of the episodes of viridans streptococcus bacteremia were associated with the first neutropenic fever and thus the patients had not received previous empirical antibiotic therapy. This was also observed by the study by Reed et al,26where most the of the episodes of viridans streptococcus bacteremia were associated with the first neutropenic fever.

MICROBIOLOGY The classification of viridans streptococci into different species remains difficult. Recent genetic studies@ have not been able to clarify the situation. While awaiting the results of further studies, Facklam and Washington4g have suggested a classification in 5 species/groups based on a series of 6 laboratory tests. Each species/group is comprised of several genospecies.48,50-56 In the literature, the species responsible for bacteremia in neutropenic patients have always been typed using commercial systems that identify the 7 most common species.57 The species most frequently identified by these procedures are Streptococcus mi-

for cases and controls;

NA = not assessed;

tis and Streptococcus sanguis II (Table Iv>. ln our institution, out of 26 episodes of bacteremia, these systems identified 15 S mitis (58%) and 11 S sanguis II (42%).27 A new typing system using more recent techniques5s was available for 15 episodes, identifying 11 S mitis (73%), 3 Streptococcus oralis (20%) and 1 S sang&s I,! (7%) [unpublished results]. These 3 species produce immunoglobulm A, subclass 1 (IgAl) proteases, which could explain their capacity to impair local immune defenses when they colonize certain surfaces of the oropharyngeal cavity.2a5g Moreover, severe complications caused by bacteremia, like shock or adult respiratory distress syndrome (ARDS), could be associated with particular strains. Regarding this point, a study cited S orab’mitis biovar 1.60In Elting’s studyti and in our institution,27 the cases of ARDS are frequently associated with streptococci identified as S mitis by commercial systems.57 We were able to reidentify 3 strains out of 4 associated with cases of ARDS using more recent methods,58 and all were confirmed as being S mitis [unpublished result]. In the absence of a clear classification of the different strains of viridans streptococci, these results remain difficult to interpret. CL’NlCAL EVALUATloN The main clinical characteristic of viridans streptococcus bacteremia is fever (33% to 100%) associated

September 1994 The American Journal of Medicine@ Volume 97

259

5 (45%)41 3 (27%)¶

-

7% 7%

10% 10%

‘Forty-nine strains tested. +One hundred thirty-nine strains tested. ‘Four episodes of bacteremia due to streptococci other than viridans included. §Ten episodes of bacteremia due to streptococci other than viridans included. Eensitivity evaluated with minimal inhibitory concentrations (MICs). ARDS = adult respiratory distress syndrome.

30% 10%

-

-

-

-

13% 20%

-

5 (50%)

10

0 (0%) -

10

-

-

-

-

111

-

-

47% 7% 47%

41%’ 33%’ 26%’

6% 6%

-

15% 8% 15%

-

2 (12.5%) 4 (25%)

16

47%+ 28%+ 25%+

12% -

3% -

18% 6%

2 (9.5%)

21

3 (9%) 14 (41%)

34

30%$ 30%$ 40%$

22w 18W

16%§

lOO%§ 5%§ 7%§ 31%§

-

-

0 (0%) -

6

47%5 18%5 35%5

-

-

29%

13 (76.5%)¶

1741

-

-

-

11% 7% 11% 11%

33% -

83% -

41% 37% 11%

-

-

-

-

42% 22% 36%

Bactermia Elting24 72-89 46

3 (50%)

6

100% -

Blood Culture Results, Clinical Evaluation and Outcome of Neutropenic Patients With Viridans Streptococcus Cohen15 Hensle16 Sotiropoulos17 Menichettil* Villablancalg ValteauzO Kern*l Weisman2* Classen 82-83 77-87 76-88 79-88 81-88 83-88 86-88 59 31 6 10 10 15 10 123 343

Years No. of episodes Blood cultures S mitis 70% S sanguis II 10% Other viridans 20% streptococci Penicillin sensitivity No. of strains 9 tested Resistant 3 (33%) Intermediate Co-trimoxazole sensitivity No. of strains 8 tested Resistant 7 (87.5%) Clinical evaluation Fever 100% Chills Hypotension 10% Shock 10% Pneumonia (after bacteremia) ARDS 10% Endocarditis Outcome Deaths (total) 10% Deaths related 10% to the bacteremia

TABLE IV

12% -

6% -

6% 6%

-

10% 3%

-

12% 3%

-

39% -

23

63% 7% 30%

-

-

-

39% 52% 9%

in 13 Clinical Studies Burden25 Reed26 88-90 33 27

12% 4%

15% 8%

96% 50% 12% 15%

-

2 (8%P 1 (4%P

26

58% 42% -

Bochud27 88-91 26

VIRIDANS STREPTOCOCCUS/BOCHUD

or not with several complications: hypotension (5% to 370/o), shock (7% to 18%), pneumonia (3% to 31%) ARDS (3% to 33%) or endocarditis (7% to 8%) (Table IV).1j27 Fever caused by this type of bacteremia usually occurs quite late in the course of neutropenia. In our institution,“7 fever appeared after an average of 9.8 days of neutropenia (
September

ET AL

by a cytotoxic effect on alveolar pneumocytes and endothelial capillary cells. The triggering role of infections, especially viridans streptococcus bacteremia, seems quite clear. Tjon A Tham et alss showed that among 15 patients who developed a respiratory failure considered as a pulmonary reaction to cytosine arabinoside, 12 had viridans streptococcus baeteremia shortly before. Other similar observations21>22B”>6g-70 suggest that viridans streptococci could complete the action of cytosine arabinoside in the development of ARDS. Guiot et al33 have observed a simultaneous decrease in the incidence of viridans streptococcus bacteremia and in respiratory complications after the introduction of a prophylaxis with oral penicillin. Weisman et al22 showed that the incidence of respiratory complications was more frequent during viridans streptococcus bacteremia than during bacteremia involving other microorganisms (29% versus 10.3%). The mechanisms by which baeteremia can trigger development of ARDS are unknown. According to Guiot et al,33 in a lung which has already been damaged by cytosine arabinoside, the infection could increase capillary permeability via mediators of inflammation, causing albumin penetration into the interstitium.62~71~72This hypothesis has been corroborated by antileukoprotease assays, a parameter of pulmonary capillary leakage.72 Ognibene et al’j3 also evoke the role of direct endothelial cell lesions through complement or other mediators. While cytosine arabinoside and viridans streptococcus bacteremia probably play a role in the development of ARDS, it should however be noted that ARDS has also been observed in patients with viridans streptococcus bacteremia who had received agents other than cytosine arabinoside,15J6)22,25and that microorganisms other than viridans streptococci can also induce ARDSGs

TREATMENTAND EVOLUTION Most of the empirical treatments proposed for neutropenic patients are adequate for viridans streptococcus bacteremia in terms of the sensitivity of the strains.30,31,73,74Broad-spectrum antibiotic treatment should be simplified, once the microorganism has been isolated and its sensitivity to antibiotics is kr~own.~~~~~ This type of treatment usually results in the disappearance of both fever and infection. In some eases, the fever only drops when the neutrophil count returns to normal.21,27 In institutions with high rates of viridans streptococci resistant to beta-laetams, the addition of vancomycin should be considered for empiric treatment of fever.26 The mortality rate ranges from 6% to 30%.16~18~20,21123~24 However, bacteremia can not always be held respon-

1994

The American

Journal

of Medicine@

Volume

97

261

VIRIDANS STREPTOCOCCUS/BOCHUD

ET AL

sible for death.1sa21>27 When it is responsible for death, it is usually associated with the presence of respiratory complications or shock.15J7Jg,24In the study by Elting et al,24 the 5 patients who had developed ARDS died after shock and acute renal failure.

PREVENTION Oral administration of penicillin in prophylaxis has been used successfully to decrease the incidence of viridans streptococcus bacteremia.27,33 Other antibiotics, such as vancomycin16a23 or roxithromycin75 have also given satisfactory results. The same goes for prophylactic antibiotics administered intravenously. Guiot et al76 compared intravenous co-trimoxazole administration to administration of penicillin G. Among the 35 patients who received penicillin G, only one developed streptococcal bacteremia compared to 7 out of 40 patients who received co-trimoxazole. Likewise, empirical antibiotic therapies, administered early in the course of an episode of neutropenia, also have a protective effect because viridans streptococcus bacteremia often occurs quite late in the course of an episode of neutropenia.27 However, in the study by Reed et al,26 patients receiving oral ampicillin prophylaxis had an incidence of viridans streptococcus bacteremia similar to patients who did not receive ampicillin. The effectiveness of penicillin or other antibiotics both in the treatment and the prevention of viridans streptococcus bacteremia could be compromised by noncompliance, poor drug absorption, or by the emergence of strains resistant to these antibiotic agents. Viridans streptococci are usually sensitive to beta-lactams, vancomycin, rifampicin, macrolides, lincosamides, and aminoglycosides.74 Strains resistant to penicillin, responsible for numerous infections, have however been isolated.13,15)23a25,77 The resistance mechanisms are probably linked to the alteration of penicillin-binding proteins.78-80 Prolonged exposure to penicillin citn favor the development of such resistances. In our institution,27 3 strains of penicillin-resistant streptococci have caused bacteremia in patients receiving an oral prophylactic quinolone and penicillin treatment. In the study by Reed et aI, only 61% of the viridans streptococcus strains were sensitive to penicillin, partially explaining why this antibiotic failed to prevent bacteremia.

CONCLUSION Because of the growing incidence, the potentially fatal complications, and the emergence of penicillinresistant strains, viridans streptococcus bacteremia is a major problem for neutropenic patients undergoing chemotherapeutic treatment. Viridans streptococci have become the most frequent cause of bac-

262

September

1994

The American

Journal

of Medicine@

Volume

teremia in this type of patient and the mortality rate associated with these infections is not insignificant. Although several risk factors have been identified, including severe neutropenia, qtiolone and co-trimoxazole prophylactic antibiotic treatments, the absence of intravenous antibiotic administration before the onset of bacteremia, high doses of cytosine arabinoside, oropharyngeal mucositis, and the colonization by viridans streptococci, the pathophysiological mechanisms that can lead to severe complications like ARDS or shock remain poorly known. Further studies, and especially a clarification of the taxonomy of these microorganisms will be necessary for a more appropriate prevention and a better understanding of the pathophysiological mechanisms involved in complications associated with this infection.

REFERENCES 1. Carlsson J. Presence of various types of non-hemolytic streptococci in dental plaque and in other sites of the oral cavity in man. Odontof Revy. 1967;18:55-74. 2. Fransden E, Pedrazzoli V, Kilian M. Ecology of viridans streptococci in the oral cavity and pharynx. Oral Microbial Immunol. 1991;6:129-133. 3. Berry FJ, Yarbrough S, Yarbrough N, et al. Transient bacteremia during dental manipulation in children. Pediatrics. 1973;51:476-479. 4. Speck W, Spear S, Krongrad E, et al. Transient bacteremia in pediatric patients after dental extraction. Am J Dis Child. 1976;130:406-407. 5. Nyvad B, Fejerskov 0. Structure of dental plaque and the plaque enamel interface during experimental caries. Caries Res. 1985;23:151-158. 6. Bayliss R, Clarke C, Oakley C, et al. The teeth and infective endocarditis. Br Heart J. 1983;50(6):506-512. 7. Von Reyn C, Levy B, Arbeit R, et al. Infective endocarditis: an analysis based on strict case definitions. Ann Intern Med. 1981;94:505-518. 8. Bradenburg RO, Giuliani ER, Wilson WR, Geraci JE. Infective endocarditis-a 25.year overview of diagnosis and therapy. J Am Coil Cardiol. 1983;1:280-291. 9. Sussman JI, Baron EJ, Tenenbaum MJ, et al. Viridans streptococcal endocarditis: clinical, microbiological and echocardiographic correlations. J Infect Dis. 1986;154(4):597-603. 10. Broughton RA, Krafka R, Baker CJ. Non-group D alpha-hemolytic streptococci: new neonatal pathogens. J Pediair. 1981;99(3):450-454. 11. Spiegelblatt L, Saintongue J, Chicoine R, Laverdiere M. Changing pattern of neonatal streptococcal septicemia. Pediatr Infect Dis J. 1985;4(1):56-58. 12. Fraser JJ, Marks MI, Welch DF. Neonatal sepsis and meningitis due to alpha-hemolytic streptococcus. South Med J. 1983;76:401. 13. Goldfarb J, Wormser GP, Glaser JH. Meningitis caused by multiple antibiotic-resistant viridans streptococci. J Pediatr. 1984;105:891-895. 14. Cox RA, Chen K, Coykendall AL, et al. Fatal infection in neonates of 26 weeks’ gestation due to Streptococcus miller;: report of two cases. J Clin Pathol. 1987;40(2):190-193, 15. Cohen J, Donnely JP, Worsley AM, et al. Septicemia caused by viridans streptococci in neutropenic patients with leukemia. Lancet. 1983; 2:1452-1454. 16. Henslee J, Bostrom B, Weisdorf D, et al. Streptococcal sepsis in bone marrow transplant patient. Lancet. 1984;1:393. 17. Sotlropoulos SV, Jackson MA, Woods JM, et al. Alpha-streptococcal septicemia in leukemic children treated with continuous or large dosage intermittent cytosine arabinoside. Pediatr Infect Dis J. 1989;8:755-758. 18. Menichetti F, Del Favero A, Guerciolini R, et al. Viridans streptococci septicemia in cancer patients: a clinical study. fur J Epidemiol. 1987; 3:316-318. 19. Villablanca JG, Steiner M, Kersey J, et al. The clinical spectrum of

97

VIRIDANS infections with viridans streptococci in bone marrow transplant. Bone Marrow Transplant. 1990;6:387-393. 20. Valteau D, Hartmann 0, Brugieres L, et al. Streptococcal septicemia following autologous bone marrow transplantation in children treated with highdose chemotherapy. Bone Marrow Transplant. 1991;7:415-419. 21. Kern W, Kurrle E, Schmeiser T. Streptococcal bacteremia in adult patients with leukemia undergoing aggressive chemotherapy. A review of 55 cases. Infection. 1990;18:138-145. 22. Weisman SJ, Scoopo FJ, Johnson GM, et al. Septicemia in pediatric oncology patients: the significance of viridans streptococcal infections. J Clin Oncol. 1990;8:453-459. 23. Classen DC, Burke JP, Ford CD, et al. Streptococcus mitis sepsis in bone marrow transplant patients receiving oral antimicrobial prophylaxis. Am J Med. 1990;89:441-446. 24. Elting LS, Bodey GP, Keefe BH. Septicemia and shock syndrome due to viridans streptococci: a case-control study of predisposing factors. Clin Infect Dis. 1992;14:1201-1207. 25. Burden AD, Oppenheim BA, Crowther D, et al. Viridans streptococcal bacteremia in patients with hematological and solid malignancy. Eur J Cancer. 1991;27:409-411. 26. Reed EC, Woods GL, Vaughan WP, et al. Characteristics of gram-positive septicemia in patients with neutropenic fever. In: Dicke KA, Armitage J, eds. Autologous Bone Marrow Transplantation Proceedings of the Fifth fnternational Symposium. 1991:267-272. 27. Bochud PY, Eggiman P, Calandra T, et al. Bacteremia due to viridans streptococcus in neutropenic patients with cancer: clinical spectrum and risk factors. Clin Infect Dis. 1994;18:25-31. 28. Venditt M, Baiocchi P, Santini C, et al. Antimicrobial suscephbrlrbes of Streptococcus species that cause septicemia in neutropenic patients. Antimicrob Agents Chemother. 1989;33:580-582. 29. EORTC Antimicrobial Therapy Project Group. Three antibiotic regimen for the treatment of infection in febrile granulocytopenic patients. J Infect Dis. 1978;137:14-29. 30. EORTC International Antimicrobial Therapy Cooperative Group. Ceftazidime combined with a short or long course of amikacin for empirical therapy of gram-negative bacteremia in cancer patients with granulocytopenia. NEJM. 1987;317:1692-1698. 31. EORTC International Antimicrobial Therapy Cooperative Group. Single daily dosing of amikacin and ceftriaxone is as efficacious and no more toxic than multiple daily dosing of amikacin and ceftazidime. Ann Int Med. 1993; 119:584-593. 32. EORTC International Antimicrobial Therapy Cooperative Group. Grampositive bacteremia in granulocytopenic cancer patients. Eur J Cancer. 1990;26:569-574. 33. Guiot HFL, Peters WG, Van den Broek PJ, et al. Respiratory failure elicited by streptococcal septicemia in patients treated with cytosine arabinoside, and its prevention by penicillin. Infection. 1990;18:131-137. 34. Ringden 0, Heimdal A, Lonnqvist B, et al. Decreased incidence of viridans streptococcal septicaemia in allogenic bone marrow transplant recipients after the introduction of acyclovir. Lancet. 1984;1:744. 35. Wieser B, Lange M, Fialk MA, et al. Prophylactic trimethoprim/ sulfamethoxazole during consolidation chemotherapy for acute leukemia. A controlled trial. Ann Intern Med. 1981;95:436-438. 36. Wade J, DeJongh C, Newman K, et al. Selective antimicrobial decontamination as prophylaxis against infection during granulocytopenia: trimethoprim-sulfamethoxazole vs. nalidixic acid. J Infect Dis. 1983;147:624-634. 37. Wade JC, Schimpff SC, Hargadon MT, et al. A comparison of trimethoprim/sulfamethoxazole plus nystatin with gentamycin in the prevention of infection in acute leukemia. NEJM. 1981;304:1057-1062. 38. Starke ID, Donelly JP, Catovsky D, et al. Cotrimoxazole alone for prevention of bacterial infection in patients with acute leukemia. Lancet. 1982;1:5-6. 39. Kaufmann CA, Liepman MK, Bergman AG, Midduszewski J. Trimethoprimsulfamethoxazole prophylaxis in neutropenic patients: reduction of infections and effect on bacterial and fungal flora. Am J Med. 1983;74:599-607. 40. Henry SA, Armstrong D, Kempin S, et al. Oral trimethoprimSeptember

STREPTOCOCCUS/BOCHUD

ET AL

sulfamethoxazole in attempt to prevent infection after induction chemotherapy for acute leukemia. Am J Med. 1984;77:663-666, 41. Gualtieri RJ, Donowitz GR, Kaiser DL, et al. Double-blind randomized study of prophylactic trimethoprim/sulfamethoxazole in granulocytopenic patients with hematologic malignancies. Am J Med. 1983;74:934-940. 42. Goldstein EJC. Norfloxacin, a fluoroquinolone antibacterial agentclassificahon, mechanism of action and in vitro activity. Am J Med. 1987;82 (suppl 6B):3-17. 43. Bow EJ, Rayner E, Louie TJ. Comparison of norfloxacin wrth co-trimoxazole for infection in acute leukemia. Am J Med. 1988;84:847-854, 44. Maschmeyer G, Harlambie E, Gaus W, et al. Ciprofloxacin and norfloxacin for selective decontamination in patients with severe granulocytopenia. Infection. 1988;16:334-339. 45. Wlnston DJ, Ho WG, Nakao SL, et al. Norfloxacin versus vancomycrn/polymyxin for prevention of infections in granulocytopenic patients. Am J Med. 1986;80:884890. 46. Cruciani M, Concia E, Navarra A, et al. Prophylactic co-trimoxazole versus norfloxacin in neutropenic patients - a prospective randomized study. Infection. 1989;17:65-69. 47. Karp JE, Merz WG, Hendricksen C, et al. Oral norfloxacin for prevention of gram-negative bacterial infectron in patients with acute leukemia and granulocytopenia. Ann Intern Med. 1987;106:1-7. 48. Coyckendall AL. Classification and identification of the viridans streptococci. Clin Microbial Rev. 1989;2:315-328. 49. Facklam RR, Washington JAI. Streptococcus and related catalase-negative gram positive cocci. In: Balows A, Hausler WJ, Herrman KL, et al, eds. Manual of Clinical Microbiology. American Society of Microbiology: Washington, DC; 1991:238-257. 50. Whiley RA, Russell RRB, Hardie JM, Beighton D. Streptococcus downei sp. nov. for strains previously described as Streptococcus mutans serotype h. Int J Syst Bacterial. 1988;38:25-29. 51. Whiley RA, Hardie JM. Streptococcus vestibularis sp. nov. from the human oral cavity. Int J Syst Bactenol. 1988;38:335-339. 52. Kilian M, Nyvad B, Mikkelsen L. Taxonomic and ecological aspects of some oral streptococci. In: Hamada S, et al. eds. Molecular Microbiology and lmmunobiology of Streptococcus mutans. Amsterdam; Elsevier Science Publishers BV; 1986. 53. Kilpper-Bdlz R, Williams BL, Lutticken R, Schleifer KH. Relatedness of “Streptococcus miller?‘: with Streptococcus anginosus and Streptococcus constelfatus. Syst Appl Microbial. 1984;5:494-500. 54. Kilpper-Balz R, Wenzig P, Schleifer KH. Molecular relationship and classification of some viridans streptococci as Streptococcus oralis and amended description of Streptococcus oralis (Bridge and Sneath 1982). Int J Syst Bacterial. 1985;35:482-488. 55. Knight RG, Shlaes DM. Physiological characteristics and deoxyribonucleic acid relatedness of Streptococcus intermedius strains. fnt J Syst Bacterial. 1988;38:19-24. 56. Schmidhuber S, Kilpper-Balz R, Schleifer KH. A taxonomic study of Streptococcus mitis, S. oralis and S. sanguis. Syst Appl Microbial. 1987; 10:74-77. 57. Facklam RR, Carey RB. Streptococci and aerococci. In: Lennette EH, Balows A, Hausler WJ, Shadomy HJ, eds. Manual of Clinical Microbiology. American Society of Microbiology: Washington, DC; 1985,154-175. 58. Beighton D, Hardies JM, Whiley RA. A scheme for the identification of vrridans streptococci. J Med Microbial. 1991;35:367-372. 59. Kilian M, Reinholdt J, Nyvad B, et al. IgAl proteases of oral streptococci: ecological aspects. lmmunol Investig. 1989;18:161-170. 60. Vael C, leven M, Vercauteren E, Pattyn SR. Species identification and penicillin susceptibility of viridans streptococci. Program and Abstracts of the ICAAC. 1992. Abstract 1033. 61. lngbar DH, White DA. Acute respiratory failure. Crit Care Clin. 1988;4: 1 l-40. 62. Braude S, Krausz T, Apperley J, Goldman JM. Adult respiratory distress syndrome after allogenic bone-marrow transplantations: evidence for a neutrophil-independent mechanism. Lancet. 1985;1:1239. 63. Ognibene FP, Martin SE, Parker MM, et al. Adult resprratory distress syndrome in patients with severe neutropenia. NEJM. 1986;315:547-551. 1994

The American

Journal

of Medicine@

Volume

97

263

VIRIDANS STREPTOCOCCUS/BOCHUD

ET AL

64. Sivan Y, Mor C, Al-Jundi S, Newth CJL. Adult respiratory distress syndrome in severely neutropenic patients. Ped Pulm. 1990;8:104-108. 65. Andersson BS, Cogan BM, Keating MJ, et al. Subacute pulmonary failure complicating therapy with high-dose ara-C in acute leukemia. Cancer. 1985;56:2181-2184, 66. Donehower RC, Karp JE, Burke PJ. Pharmacology and toxicity of high dose cytosine arabinoside in 72 hours continuous infusions. Cant Treat Res. 1986;70:1059-1065. 67. Haupt HM, Hutchkins GM, Moore GW. Ara-c lung: noncardiogenic pulmonary edema complicating cytosine-arabinoside therapy of leukemia. Am J Med. 1981;70:256-261. 68. Tjon A Tham RTO, Peters WG, de Bruine FT, Willemze R. Pulmonary complications of cytosine-arabinoside therapy: radiographic findings. Am J Roentgenof. 1987;149:23-27. 69. Dybedal I, Lamvik J. Respiratory insufficiency in acute leukemia following treatment with cytosine-arabinoside and septicemia with Streptococcus vfridans. Eur J Haematol. 1989;42:405-406. 70. Peters WG, Willemze R, Golly LP, Guiot HFL. Side effects of intermediateand high-dose cytosine arabinoside in the treatment of refractory or relapsed acute leukaemla and non-Hodgkin’s lymphoma. Neth J Med. 1987;30:64-74. 71. Fleck A, Raines G, Hawker F, et al. Increased vascular permeability: a major cause of hypoalbuminuria in disease and injury. Lancet. 1985;i: 781-784. 72. Pepe PE, Potkin RT, Reus DH, et al. Clinical predictors of the adult

264

September

1994

The American

Journal

of Medicine@

Volume

respiratory distress syndrome. Am J Surg. 1982;144:124-130. 73. Klastersky J. Empiric therapy for febrile granulocytopenic patients: no longer a challenge? Ann Oncol. 1991;2:454-455. 74. Bourgault AM, Wilson WR, Washington JAI. Antimicrobial susceptibility of species of viridans streptococci. J Infect Dis. 1979;140:316-321. 75. Rozenberg-Arska M, Dekker L, Verdonck L, Verhoef J. Prevention of bacteremia caused by a-hemolytic streptococci by roxithromycin (RU 28 965) in granulocytopenic patients receiving ciprofloxacin. Infection. 1989; 17:240-244. 76. Gutot HFL, van der Meer PJ, van der Broek PJ, et al. Prevention of viridansgroup streptococcal septicemia in oncohematologic patients: a controlled comparative study of the effect of penicillin G and cotrimoxazole. Ann Hematol: 1992;64:260-265. 77. Quinn JP, Di Vicenzo CA, Lucks DA, et al. Serious infections due to penicillin-reslstant viridans streptococci with altered penicillin-binding proteins. J Infect Dis. 1988;157:764-769. 78. Farber BF, Yee Y. High-level aminoglycoside-modifying enzymes among viridans streptococci. J Infect Dis. 1987;155:948. 79. Farber BF, Eliopoulos GM, Ward JI, et al. Multiply resistant viridans streptococci: susceptibility to p-lactams antibiotics and comparison of penicillin blnding protein patterns. Antimicrob Agents Chemother. 1983;24:871. 80. Farber BF, Eliopoulos GM, Ward JI. Resistance to penicillin-streptomycin synergy among clinical isolates of viridans streptococci. Antimicrob Agents Chemother. 1983;24:871.

97