This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gamis, A. S. Articles by Woods, W. G. Search for Related Content PubMed PubMed Citation Articles by Gamis, A. S. Articles by Woods, W. G.

Alpha Hemolytic Streptococcal Infection During Intensive Treatment for Acute Myeloid Leukemia: A Report From the Children’s Cancer Group Study CCG-2891

From the Children’s Mercy Hospital, Kansas City, MO; University of Southern California School of Medicine, Los Angeles, Long Beach Memorial Hospital, Long Beach, and Children’s Hospital Oakland, Oakland, CA; and South Carolina Cancer Center, Columbia, SC; for the Children’s Cancer Group, Arcadia, CA.

Address reprint requests to Alan S. Gamis, MD, MPH, Children’s Cancer Group, PO Box 60012, Arcadia, CA 91066-6012.

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX Participating Principal... APPENDIX (cont’d) REFERENCES

 
PURPOSE: Past reports indicate that alpha hemolytic streptococcal (AHS) organisms are a common cause of infection among acute myeloid leukemia (AML) patients. This study was intended to ascertain the population incidence and rate (infections per 100 patient-days of treatment) of AHS and to identify associated risk factors.

PATIENTS AND METHODS: Patients (n = 874 with 151,350 days of risk) enrolled on the Children’s Cancer Group (CCG) protocol for newly diagnosed AML, CCG-2891, which randomly assigned intensity of induction and intensification, were prospectively evaluated for infectious complications.

RESULTS: AHS occurred in 21% of patients, was primarily blood borne (86%), made up 21% of bacteremic infections, and had a recurrent incidence of 31% during subsequent therapy. AHS was more often life-threatening (59%) than other infections (41%) (P = .001). AHS rates increased with age less than 10 years (odds ratio [OR], 2.0; P = .007), intensively timed induction (OR, 1.8 to 1.9; P = .02), and high-dose cytarabine intensification (OR, 3.7; P < .0001). Among all courses, the greatest incidence (19%) and rate (0.41) were associated with the use of high-dose cytarabine. Gastrointestinal toxicity correlated significantly with AHS bacteremia (P < .01). Infection with AHS resulted in increased hospital days (P = .0001). Only among bone marrow transplant patients were overall survival (OR, 2.8; P = .0001) and disease-free survival (OR, 2.1; P = .008) decreased after AHS bacteremia.

CONCLUSION: This study, the first to prospectively examine AHS incidence among uniformly treated patients in multiple institutions, established that as the intensity of AML therapy has increased, so has the rate of AHS. Young children, those with previous AHS bacteremias, and those receiving high-dose cytarabine are at particularly high risk of AHS bacteremia.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX Participating Principal... APPENDIX (cont’d) REFERENCES

 
INTENSIFICATION OF therapy for acute myeloblastic leukemia (AML) in recent years has improved remission induction and long-term cure.¹ The price of this success has been an increase in significant side effects. In particular, severe prolonged neutropenia (despite the use of growth factors) and severe mucositis are common. Consequently, infection has become a more frequent cause of morbidity and mortality in this population. To some extent, this has limited the potential improvement in the complete remission rate for these patients.¹ Among these infections, septicemia due to alpha hemolytic streptococcal (AHS) organisms has become particularly problematic.^2-10 During an outbreak of AHS infections at their institution, Sotiropoulos et al² analyzed a group of AML patients treated on the Children’s Cancer Group (CCG) protocol CCG-213 between 1985 and 1987. They noted that the markedly increased incidence of AHS septicemia was associated with the introduction of a high-dose (HD) cytarabine (AraC) regimen. In this multiagent chemotherapy protocol, 14 of 15 episodes (93%) of AHS septicemia were preceded by AraC administration. Subsequent studies have also implicated AraC administration as a major risk factor in the development of AHS septicemia.^3,7,9 However, intensification of AraC does improve long-term survival.^11-14

From 1989 until 1995, the CCG conducted the largest reported multi-institutional AML trial in children. This trial used cytarabine as a principal component of therapy. As part of this trial, data were collected to evaluate the incidence and type of infections that occurred during this therapy. AHS septicemia was found to be the single most common cause of bacteremia among these patients. Using this large database, we report this experience and evaluate the risk factors associated with its occurrence and the impact it had on toxicity and ultimate outcome.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX Participating Principal... APPENDIX (cont’d) REFERENCES

 
Children 21 years of age were enrolled onto CCG-2891 from 1989 until 1995. Therapy administered on CCG-2891 has been described previously.¹ Briefly, there were two randomizations resulting in three therapeutic arms during induction and three separate arms during intensification. There were initially two induction arms, standard timing versus intensive timing chemotherapy. In 1993, the intensive timing arm was closed and a third arm, intensive timing with granulocyte colony-stimulating factor (G-CSF) (beginning on day 6 of each induction course), replaced it. Six months later, the standard timing arm was closed, and all subsequent patients were enrolled solely onto the intensive timing with G-CSF arm. The induction and intensification phases are briefly outlined in Table 1. During induction, two cycles of chemotherapy made up one course, regardless of whether or not they were intensively timed (ie, induction course 1, cycles 1 and 2; course 2, cycles 3 and 4). Patients randomized to the standard timing arm were required to have hematologic recovery after cycles 1 and 3 before beginning the subsequent cycles, 2 and 4. Those patients whose bone marrow examination on day 14 of induction cycle 1 was not in remission, however, did receive cycle 2 regardless of hematologic recovery. In the intensive timing arm, patients began cycles 1 and 2 on days 0 and 10 (before hematologic recovery) of induction course 1. They were then required to have hematologic recovery before repeating the intensively timed sequence for cycles 3 and 4. Data for this study were collected and recorded by course of induction. All days during induction are numbered from the beginning of cycles 1 and 3 for courses 1 and 2, respectively. The intensification phase, for both bone marrow transplant (BMT) recipients and those randomized to chemotherapy, began after hematologic recovery from induction cycle 4. For intensification, children who did not have a matched related donor for an allogeneic BMT were randomized to either an autologous 4-hydroxycyclophosphamide–purged BMT or an HDAraC regimen, intensively timed, followed by two courses of low-dose chemotherapy. For analysis of the infectious complications of the intensification period of treatment on this protocol, patients were grouped on the basis of actual treatment received. Informed consent for participation on CCG-2891 was obtained from all patients before enrollment. All data were prospectively obtained.

Data collected regarding infections included the following: organism, body site infected, the day of each course on which the positive culture was obtained, the degree of severity (life-threatening v non–life-threatening), days of hospitalization during each course of chemotherapy, and the institution at which the patient was treated. Each patient’s baseline data were obtained and included age, race, sex, French-American-British classification, date of diagnosis, and initial total WBC count. The degree of gastrointestinal (GI) toxicity in each patient was ascertained. GI toxicity was categorized by use of the CCG toxicity and complications criteria sheets. For analysis, patients were categorized as with or without grade 3 or 4 GI toxicity by the presence or absence of one or more reports of GI toxicity during the same period of therapy that the AHS infection was detected. GI symptoms included specifically (1) stomatitis (unable to eat or required parenteral or enteral support); (2) abdominal pain (moderately or severely refractory to treatment or necessitating hospitalization and sedation); (3) constipation (severe ileus or ileus > 96 hours); (4) diarrhea (> seven stools per day or severe cramps or bloody stools and parenteral support required); and (5) nausea/vomiting (poor oral intake and > six emesis episodes per day). Standard CCG infection report forms were used prospectively. Institutions were instructed to complete a separate report form for each significant infection and to exclude positive surveillance cultures that were not associated with a diagnosed clinical infection. Streptococcal infections could be reported either nonspecifically (eg, streptococcal infection or Gram-positive infection) or more specifically (eg, Streptococcus mitis). For the purpose of this analysis, AHS was considered to be the etiologic agent if it was reported in one of the following available categories on the report form: Streptococcus sanguis, S mitis, Streptococcus viridans, Streptococcus salivarius, strep alpha hemolytic, Gram-positive cocci alpha, strep hemolytic, or streptococcus. Sites of infection were assigned by the source of the culture. A single positive culture was adequate for assignment. No distinction between bacteremia (defined as a single positive blood culture from an asymptomatic patient) or septicemia (> one positive blood culture in a symptomatic patient) could be made in the data collection. Therefore, in this report, all positive blood cultures are referred to as bacteremias. As noted above, infections were categorized as either non–life-threatening, life-threatening, or fatal to aid in the assessment of severity.

Analyses evaluated the following: (1) the incidence of AHS infections overall, (2) sites of infection, (3) severity of infection, (4) the course and day in which the first AHS culture was obtained, (5) mortality due to AHS infection, (6) recurrent AHS infections in the same patient during the same or other courses, (7) the duration of hospitalization during the course in which AHS was cultured, (8) the duration of the course in which the organism was cultured, (9) the overall survival and disease-free survival for the patients in whom AHS was cultured, (10) the institution in which the AHS infection occurred, and (11) the year in which AHS infections occurred. These data were then compared with data from those patients without AHS who were treated on CCG-2891. This included analyses of all patients (n = 887) on CCG-2891 and subgroup analyses stratified by types of therapy received and year of the trial. Data were compared to determine whether there were significant differences between courses, season at the time of diagnosis, year of the trial, and therapeutic arms. Patients were grouped on the basis of actual treatment groups and not intent-to-treat groups. Frequency of AHS was determined using (1) incidence per patient (number of patients with AHS infection reported divided by the number under study) and (2) rate of infections (the incidence of AHS infection per 100 patient-days at risk). "Days at risk" was defined as the period from the beginning to the end of each course of therapy. The end of each course was always taken to be the beginning of the next course of therapy. Data on periods of neutropenia, time to neutrophil recovery, and ANC at the time of the infection were not collected as a part of CCG-2891. In addition to frequency per patient, the total number of infections (counting repeated infections in the same patient separately) was analyzed to assess the proportion of AHS among all infections reported.

Statistical analyses included a ² comparison of proportions to determine significance of differences in incidence. Cox regression was used to assess overall and disease-free survival when the date of occurrence of AHS bacteremia was used to define a time-dependent covariate.¹⁵ Poisson regression models of the infection rates were used to estimate the odds ratio for AHS infection in different patient groups.¹⁶ Analysis of the impact of AHS on hospitalization and course days used the Mann-Whitney U test to compare median values.¹⁷

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX Participating Principal... APPENDIX (cont’d) REFERENCES

 
Up to April 18, 1995, when the protocol was amended to enroll only Down’s syndrome patients, 1,097 eligible patients were enrolled onto CCG-2891. From this group, there were 210 exclusions for Down’s syndrome (n = 104), myelodysplastic syndrome (n = 79), secondary AML (n = 19), chloroma (n = 7), and no data (n = 1), leaving 887 patients available for this analysis. Another 13 patients who received mixed standard and intensive timing induction were excluded from analyses that examined the difference between induction arms. Table 2 shows the number of analyzed patients at risk of infection during each course of therapy. Overall, there were 151,350 days of risk for infection (among 874 patients).

Risk Factors for AHS Development
Baseline demographics for those enrolled onto CCG-2891 are presented in Table 4. These totals are then broken down by whether or not the patient had an identified AHS infection during induction. Among the six baseline factors analyzed, only age was found to have an association with AHS bacteremia. The youngest patients (0 to 2 years old) had a higher incidence (18%) than either the 3- to 10-year-olds (15%) or those 11 years old (11%) (P = .057). This difference was highly significant when the rate of infection per day of risk was evaluated (Table 5). When analyzed by this method, those younger than 11 years old had significantly higher AHS infection rates than the older patients (for age < 11 years, OR, 2.0; P = .007). Finally, although overall there were no racial differences, there was an unexplainable finding that children designated to be of Asian descent had no reported episodes of AHS (P = .04).

The days of highest incidence during each course of therapy were evaluated. Because of the methods of data acquisition on this study (by course and not by cycle), only intensive timing induction was considered for this evaluation. For patients during intensive induction course 1 (ie, cycles 1 and 2), the greatest period of incidence was days 11 to 15. The percentages of AHS infections occurring on days 0 to 5, 6 to 10, 11 to 15, 16 to 20, and 20+ were 3%, 19%, 36%, 20%, and 22%, respectively. The peak corresponded with the administration of cycle 2 chemotherapy. The greatest incidence period for AHS infection in the second intensive induction course (ie, cycles 3 and 4) was days 16 to 20. The percentages of AHS infections occurring on days 0 to 5, 6 to 10, 11 to 15, 16 to 20, and 20+ were 0%, 7%, 24%, 49%, and 20%, respectively. The peak came after the completion of cycle 4 chemotherapy administration. The chemotherapy arm of the intensification period, and specifically the first course that contained the intensively timed HDAraC, had its greatest period of incidence during days 16 to 20. The percentages of AHS infections occurring on days 0 to 5, 6 to 10, 11 to 15, 16 to 20, and 20+ were 0%, 2%, 20%, 66%, and 12%, respectively, following the completion of the day-7 and day-8 chemotherapy doses.

Impact of Course and Treatment Arm on AHS Bacteremia Development
AHS bacteremia incidence (defined as the percentage of patients who experience an AHS infection) increased with each subsequent course: 8.9% during induction course 1, 10% during induction course 2, and 13.6% during intensification.

The AHS bacteremia incidence per patient among each of the three induction arms is listed in Table 6. During induction, 14.5% of the patients analyzed experienced at least one AHS bacteremia (during course 1 and/or 2). Those receiving intensive timing induction, regardless of whether G-CSF was used, had a significantly greater incidence (17% to 18% v 10% in the standard arm) of AHS bacteremia (P = .007) (Table 6). Evaluation of those standard timing patients who required early initiation of cycle 2 (before hematologic recovery and due to remission failure on the day-14 bone marrow examination) revealed a higher incidence of AHS bacteremia than in those who waited for hematologic recovery (8% v 5%, respectively). When these two values are compared with the overall incidence of AHS bacteremia among the intensive induction patients during course 1 only (10%), there is a trend toward increased risk (P = .07) with increased intensive timing. The rate of AHS bacteremia also revealed similar conclusions, with odds ratios of acquiring AHS bacteremia during induction showing a 1.8- to 1.9-fold higher risk for those on the intensive arms compared with the standard arm (P = .02) (Table 7). The incidence of life-threatening AHS sepsis was also greater in these two intensive arms (P = .003) (Table 6). When examined by course of induction, the incidence of AHS was greatest for those receiving intensive timing during course 2 (cycles 3 and 4) (P = .003). When the two intensive timing arms were compared to ascertain whether G-CSF reduced AHS sepsis, no statistical differences (in a single-variable ² analysis) could be found in either course or all induction combined. However, in a multivariate regression analysis of the possible interaction of G-CSF and course (1 or 2), there were fewer AHS bacteremias during the first course of induction among those receiving G-CSF (P = .036). This suggested that there might be a beneficial effect early during induction; by course 2, however, this had disappeared.

Risk of Recurrent AHS Bacteremia
To assess risk of recurrent AHS (in a subsequent course), those patients with either a positive blood culture or central venous catheter culture for AHS were analyzed. The total number of patients enrolled onto CCG-2891 who had multiple AHS bacteremic events was quite small (6%). However, patients with at least one episode of AHS bacteremia had a recurrence risk of 14%. Excluded from analysis were those patients who, because of failure to proceed to subsequent courses, were no longer analyzable for risk. Among those patients who had AHS bacteremia and who did proceed through subsequent courses, there was a 31% incidence of recurrent AHS sepsis. This risk was greater for those who developed AHS bacteremia during the first induction course (35%) than for those whose first episode occurred during induction course 2 (21%). Those who had episodes of AHS bacteremia in both induction courses and proceeded to intensification had a 44% incidence of having a third episode of AHS bacteremia. Despite the high risks of recurrence, the majority of patients who had AHS bacteremia during induction course 2 and intensification had had no prior episodes of AHS bacteremia (69% and 77%).

Impact of AHS Bacteremia on Patient Course and Outcome
Analysis of outcome for patients who had AHS looked at days of hospitalization and duration of each course received. These analyses examined the impact on the specific course in which the AHS bacteremia occurred and compared this with patients without AHS bacteremia in the same course of therapy. Table 8 shows the median number of hospital days experienced during the induction courses by patients with and without AHS bacteremia, divided by the standard and intensive timing arms (with and without G-CSF are combined). During all courses and in both arms, hospitalization time was significantly increased (statistically and clinically) (11 to 12 days) among those who experienced an AHS bacteremia. Course duration, measured from the first day of chemotherapy of the course in question until the first day of chemotherapy for the subsequent course, was increased, although primarily in those on the intensive timing arm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX Participating Principal... APPENDIX (cont’d) REFERENCES

 
Previous single-institution studies have indicated a high risk of AHS among AML patients.^2-10 Therefore, the present analysis aimed to identify (1) whether these same rates were seen in a multi-institutional setting, (2) what the predisposing risk factors were, and (3) what impact they had on the patients’ therapy and outcome. The CCG-2891 cohort is the largest series of children treated for AML published to date. The therapeutic question looked at whether intensification of induction and consolidation would increase the disease-free survival without excessively increasing toxicity, particularly infections. CCG-2891 has shown that increased dose-intensity with intensively timed chemotherapy does improve the disease-free survival, although at an increased risk of toxicity.¹

This report represents the first large, multi-institutional, population-based analysis of the incidence and outcome of AHS infections. The data from this study indicate that children receiving present-day therapy for AML remain at great risk (21%) for the development of AHS infection. In addition, these data show that the intensive timing method significantly increases the incidence of AHS. Among all patients in the present study (standard and intensively timed), AHS accounted for 16% of all infections and 21% of all bacteremic events. The data also establish that the vast majority of AHS infections are blood borne (86%).

Several studies have examined the prevalence of AHS (14% to 28%) among bacteremic events.^3,5 Few studies have performed population-based analyses to assess incidence among patients treated as in the present study. Those that have done so were single-institution studies and significantly smaller. Cohen et al⁶ identified an incidence of 14% (10 of 72) among febrile neutropenic patients (treated for myeloid leukemias and/or receiving allogeneic BMT). Three previous studies have identified a 10% to 16% incidence rate of AHS among BMT patients.^18-20

The analysis of AHS infection among patients on this particular randomized therapeutic protocol for AML allowed the study of the impact of dose-intensity on the incidence of infection. This analysis established that the higher dose-intensity administered on the intensive timing induction arms resulted in a higher incidence of infection. Even when length of risk was factored in by examination of the rate, the risk was significantly greater (OR, 1.8 to 1.9; P = .02) than with standard timing induction. Analysis of the use of G-CSF failed to reveal a beneficial reduction in AHS bacteremia.

This study also confirms that the use of HDAraC is correlated with a high risk of AHS bacteremia (19% incidence). Previous studies have identified a variable risk of AHS after HDAraC, but these studies examined the rate among select populations (eg, febrile neutropenia) rather than the incidence or rate among all treated patients.^2,3,7,9 This study also confirms previous studies that identified the HDAraC portion of AML therapy as a particularly high-risk period compared with standard induction (OR, 3.7). Sotiropoulos et al,² in their small population, had reported that 93% of AHS septicemias were preceded by HDAraC administration within a multiagent protocol.

This study also confirms that the presence of GI toxicity is a significant risk factor in the development of AHS bacteremia (21% v 12%, P < .01). This and similar findings from previous studies^4,6,10,21 suggest that methods to reduce GI toxicity may have the added beneficial effect of reducing bacteremic infections.

The rate of AHS infection was higher among those receiving autologous BMT than those receiving allogeneic BMT, despite similar preparative regimens (OR, 4.3; P < .001). No reason was apparent for this effect, although more antimicrobial prophylaxis may have been administered to these patients during the transplant period. The use of prophylactic agents during BMT was not collected in the data forms, and thus the reason for this reduced rate remains speculative. The incidence during the autologous BMT phase was similar to that during the intensive timing induction phases, during which time AHS prophylaxis also would not have been routinely used.

The use of TMP-SMX has been implicated as a factor that increases the risk of AHS bacteremia developing.^3,4,6,22 In one study, patients receiving TMP-SMX prophylaxis had an OR of 50.3 for the development of AHS.⁴ TMP-SMX has been a recommended prophylactic agent for pediatric oncology patients and in particular for AML patients. Because its daily use was recommended in the present study and compliance was not assessed, it may have confounded the results of this analysis by elevating the risk of AHS.

Several risk factors of AHS during induction present at diagnosis of the patient’s AML have not been examined previously. Among the analyzed characteristics present at diagnosis, age of the patient correlated with risk of AHS bacteremia during induction. There was a higher incidence of AHS among children 2 years of age (18%, P = .06) than among children age 3 to 10 years (15%) and older than 10 years (11%). When the rate of infection was examined, there was a significant increase in rate in the younger patients ( 10 years) compared with those older than 10 years of age (OR, 1.6 to 1.9; P = .007). Tooth exfoliation and new tooth eruption common at these ages may contribute to the increased risk, because this organism constitutes a majority of the normal oral flora; to date, however, no one has explored this association.

Finally, the present study found a high incidence of recurrence of AHS in subsequent cycles of myelosuppressive chemotherapy. When patients received subsequent chemotherapy after an episode of AHS, there was a 31% risk of having a second episode of AHS infection. Weisman et al³ and Sotiropoulos et al² found similar high recurrence risks among their populations of pediatric patients (46% and 71%, respectively). This implies that patients with a history of AHS bacteremia should be strongly considered for either prophylaxis or close monitoring of recurrence.

The AHS bacteremias identified in this study exhibited a higher proportion of life-threatening infections than non-AHS infections (59% v 41%, P = .001). This study also revealed that hospitalization time during induction was significantly increased in patients with AHS bacteremia compared with all other patients on protocol (69 v 51 days, P = .0001). Prior studies have also identified that this organism is associated with higher morbidity, particularly shock, respiratory distress, and CNS symptomatology. Overall, the incidence of either shock or respiratory complications because of AHS ranges from 10% to 38%.^2-4,6,8,23 Of significant concern is the rapidity of fatal sequelae of AHS.²⁴ In one study, in patients in whom intravenous vancomycin therapy was delayed 48 to 72 hours after admission until a positive culture was found (compared with a group of patients in whom vancomycin was started empirically), there was a significant excess mortality among AHS patients (14% v 0%, P = .004). In these same reports, the incidence of death as a result of AHS ranged from 0% to 31%. The mortality rate specifically reported to be due to AHS in our study was not found to be higher than the rate of non-AHS infections (2% v 3%).

This is the first study to examine the impact of AHS infection on therapy for pediatric AML. Fortunately, no difference in overall survival between those who had an AHS bacteremia and those who did not could be found. However, when the analysis was restricted to just the BMT patients (autologous and allogeneic), there was a significant increase in the risk of worse overall survival (OR, 2.8; P = .0001) and disease-free survival (OR, 2.0; P = .008) from the end of induction. This is in concordance with earlier studies examining BMT patients with AHS^18,23 that showed significant morbidity in this population.

This report, which used a well-defined population, identifies the high risk of AHS among children with AML. The greatest risk is among children younger than 11 years old and after the use of HDAraC. The more intensively timed therapies currently used significantly increase the risk of AHS infection. AHS infections have a high degree of associated toxicity and a high rate of recurrence during subsequent episodes of intensive chemotherapy. The CCG is currently pursuing a prophylactic study of oral penicillin in this high-risk population. This approach has been used in the past with significant reduction of AHS infections.^19,22,25-27 Physicians caring for these children must maintain a high index of suspicion for AHS bacteremia and must be prepared to institute early empiric antibiotic coverage for possible AHS infections.

	APPENDIX Participating Principal Investigators: Children’s Cancer Group

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX Participating Principal... APPENDIX (cont’d) REFERENCES

 

	APPENDIX (cont’d)

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX Participating Principal... APPENDIX (cont’d) REFERENCES

	ACKNOWLEDGMENTS

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION APPENDIX Participating Principal... APPENDIX (cont’d) REFERENCES

 
1. Woods WG, Kobrinsky N, Buckley JD, et al: Timed-sequential induction therapy improves postremission outcome in acute myeloid leukemia: A report from the Children’s Cancer Group. Blood 87:4979-4989, 1996[Abstract/Free Full Text]

2. Sotiropoulos SV, Jackson MA, Woods GM, et al: Alpha-streptococcal septicemia in leukemic children treated with continuous or large dosage intermittent cytosine arabinoside. Pediatr Infect Dis J 8:755-758, 1989[Medline]

3. Weisman SJ, Scoopo FJ, Johnson GM, et al: Septicemia in pediatric oncology patients: The significance of viridans streptococcal infections. J Clin Oncol 8:453-459, 1990[Abstract]

4. Elting LS, Bodey GP, Keefe BH: Septicemia and shock syndrome due to viridans streptococci: A case-control study of predisposing factors. Clin Infect Dis 14:1201-1207, 1992[Medline]

5. Burden AD, Oppenheim BA, Crowther D, et al: Viridans streptococcal bacteraemia in patients with haematological and solid malignancies. Eur J Cancer 27:409-411, 1991

6. Cohen J, Worsley AM, Goldman JM, et al: Septicaemia caused by viridans streptococci in neutropenic patients with leukemia. Lancet 2:1452-1454, 1983[Medline]

7. Kern W, Kurrie E, Vanek E: High risk of streptococcal septicemia after high dose cytosine arabinoside treatment for acute myelogenous leukemia. Klin Wochenschr 65:773-780, 1987[Medline]

8. Devaux Y, Archimbaud E, Guyotat D, et al: Streptococcal bacteremia in neutropenic adult patients. Nouv Rev Fr Hematol 34:191-195, 1992

9. Peters WG, Willemze R, Colly LP, et al: Side effects of intermediate- and high-dose cytosine arabinoside in the treatment of refractory and relapsed acute leukemia and non-Hodgkin’s lymphoma. Neth J Med 30:64-74, 1987[Medline]

10. Richard P, Amador Del Valle G, Moreau P, et al: Viridans streptococcal bacteraemia in patients with neutropenia. Lancet 345:1607-1609, 1995[Medline]

11. Woods WG, Ruymann FB, Lampkin BC, et al: The role of timing of high-dose cytosine arabinoside intensification and of maintenance therapy in the treatment of children with acute nonlymphocytic leukemia. Cancer 66:1106-1113, 1990[Medline]

12. Ravindranath Y, Steuber CP, Krischer J, et al: High-dose cytarabine for intensification of early therapy of childhood acute myeloid leukemia: A Pediatric Oncology Group study. J Clin Oncol 9:572-580, 1991[Abstract]

13. Wells RJ, Woods WG, Lampkin BC, et al: Impact of high-dose cytarabine and asparaginase intensification on childhood acute myeloid leukemia: A report from the Children’s Cancer Group. J Clin Oncol 11:538-545, 1993[Abstract/Free Full Text]

14. Mayer RJ, Davis RB, Schiffer CA, et al: Intensive postremission chemotherapy in adults with acute myeloid leukemia. Med 331:896-903, 1994[Abstract/Free Full Text]

15. Cox DR: Regression models and life tables. J R Stat Soc B 30:89-99, 1972

16. Frome EL, Checkoway H: Use of Poisson regression models in estimating incidence rates and ratios. Am J Epidemiol 121:309-323, 1985[Abstract/Free Full Text]

17. Armitage P: Statistical Methods in Medical Research. Oxford, United Kingdom,Blackwell Scientific Publications, 1971

18. Steiner M, Villablanca J, Kersey J, et al: Alpha-streptococcal shock in bone marrow transplantation patients. Blood 72:409a, 1988 (abstr 1548) (suppl 1)

19. International Antimicrobial Therapy Cooperative Group of the European Organization for Research and Treatment of Cancer: Reduction of fever and streptococcal bacteremia in granulocytopenic patients with cancer: A trial of oral penicillin V or placebo combined with pefloxacin. JAMA 272:1183-1189, 1994[Abstract]

20. Heimdahl A, Mattsson T, Dahllöf G, et al: The oral cavity as a port of entry for early infections in patients treated with bone marrow transplantation. Oral Surg Oral Med Oral Pathol 68:711-716, 1989[Medline]

21. Groot-Loonen JJ, van der Noordaa J, de Kraker J, et al: Alpha-hemolytic streptococcal septicemia with severe complications during neutropenia in childhood cancer. Pediatr Hematol Oncol 4:323-328, 1987[Medline]

22. Guiot HFL, van der Meer JWM, van den Broek PJ, et al: Prevention of viridans-group streptococcal septicemia in oncohematologic patients: A controlled comparative study on the effect of penicillin G and co-trimoxazole. Ann Hematol 64:260-265, 1992[Medline]

23. Villablanca JG, Steiner M, Kersey J, et al: The clinical spectrum of infections with viridans streptococci in bone marrow transplant patients. Bone Marrow Transplant 6:387-393, 1990

24. Elting LS, Rubenstein EB, Rolston K, et al: Impact of vancomycin (vanc) use on duration of fever and mortality in neutropenic patients with gram positive bacteremia (GPB). Proc Am Soc Clin Oncol 15:543a, 1996 (abstr 1765)

25. 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 18:25-31, 1994[Medline]

26. de Jong P, de Jong M, Kuijper E, et al: Evaluation of penicillin G in the prevention of streptococcal septicemia in patients with acute myeloid leukaemia undergoing cytotoxic chemotherapy. Eur J Clin Microbiol Infect Dis 12:750-755, 1993[Medline]

27. Broun ER, Wheat JL, Kneebone PH, et al: Randomized trial of the addition of gram-positive prophylaxis to standard antimicrobial prophylaxis for patients undergoing autologous bone marrow transplantation. Antimicrob Agents Chemother 38:576-579, 1994[Abstract/Free Full Text]

Submitted July 6, 1999; accepted January 10, 2000.

This article has been cited by other articles:

				L. Sung, B. J. Lange, R. B. Gerbing, T. A. Alonzo, and J. Feusner Microbiologically documented infections and infection-related mortality in children with acute myeloid leukemia Blood, November 15, 2007; 110(10): 3532 - 3539. [Abstract] [Full Text] [PDF]

				U. Creutzig, M. Zimmermann, D. Reinhardt, M. Dworzak, J. Stary, and T. Lehrnbecher Early Deaths and Treatment-Related Mortality in Children Undergoing Therapy for Acute Myeloid Leukemia: Analysis of the Multicenter Clinical Trials AML-BFM 93 and AML-BFM 98 J. Clin. Oncol., November 1, 2004; 22(21): 4384 - 4393. [Abstract] [Full Text] [PDF]

				D. K. H. Webb, G. Harrison, R. F. Stevens, B. G. Gibson, I. M. Hann, and K. Wheatley Relationships between age at diagnosis, clinical features, and outcome of therapy in children treated in the Medical Research Council AML 10 and 12 trials for acute myeloid leukemia Blood, September 15, 2001; 98(6): 1714 - 1720. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Gamis, A. S. Articles by Woods, W. G. Search for Related Content PubMed PubMed Citation Articles by Gamis, A. S. Articles by Woods, W. G.