Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Acta Pædiatrica ISSN 0803–5253 REGULAR ARTICLE Suspected infections in children treated for ALL Sólveig Hafsteinsdóttir1, Kristján Jónasson2, Guðmundur K Jónmundsson1,3, Jón R Kristinsson1,3, Ólafur Gísli Jónsson1,3, Inga Huld Alfreðsdóttir3, Corrado Cilio4, Thomas Wiebe5, Ásgeir Haraldsson ([email protected])1,3 1.Children’s Hospital Iceland, Landspı́tali – University Hospital, Reykjavı́k, Iceland 2.Faculty of Engineering, University of Iceland, Iceland 3.Faculty of Medicine, University of Iceland, Iceland 4.Department of Paediatrics, University Hospital Malmö, Sweden 5.Department of Paediatrics, Haematology ⁄ Oncology, University Hospital Lund, Sweden Keywords ALL, Infection, Leukaemia, Treatment Abstract Correspondence Ásgeir Haraldsson, Children’s Hospital Iceland, Landspı́tali Hringbraut, 101 Reykjavı́k, Iceland. Tel: +354 543-1000 | Fax: +354 543-3021 | Email: [email protected] treated for ALL and to analyse which patients have an enhanced infection risk. Methods: Episodes of suspected or confirmed infections were evaluated during the first 12 months Received 2 November 2008; revised 8 February 2009; accepted 3 March 2009. episodes (31.8%), the most common being coagulase-negative staphylococci. The number of DOI:10.1111/j.1651-2227.2009.01286.x Aim: The aim of our study was to get epidemiological information on bacterial infections in children of treatment for childhood acute lymphoblastic leukaemia (ALL). Results: The number of patients was 73 (43 boys). The median age was 4.6 years. A total of 179 episodes occurred, varying from none in six patients to eight in one. Bacteria were cultured in 57 episodes fell significantly with increasing age for suspected and confirmed infections (p < 0.001 and p ¼ 0.03). The proportion of confirmed infections was significantly higher (p < 0.001) in the first episodes. The average number of suspected infections was higher in girls than in boys (p ¼ 0.03), but confirmed infections were not. Conclusion: Most of the serious infections occur early in the treatment and the number of suspected and confirmed infections falls with age. Suspicion of infection is more likely in girls, but the number of confirmed infections is equal in both sexes. Coagulase-negative staphylococcus was most commonly isolated, highlighting the importance of careful handling of central venous devices. INTRODUCTION Malignant diseases in children are relatively rare yet lifethreatening. In western societies, malignant diseases are the second most common cause of death in children older than one year after accidents (1,2). Despite remarkable progress in the treatment of childhood malignancies, much work still needs to be done. Understanding the main complications of the treatment is important in order to further improve the survival of children with malignancies. Survival of children diagnosed with malignant diseases has improved substantially during the last decades. This progress is mainly attributed to more accurate diagnosis, improved treatment and follow-up (2–6). The treatment is complicated, involving substantial immune suppression (2,7,8), making the children prone to severe infections, including bacteraemia and sepsis (9–13). Approximately 2–3% of children, treated for malignant diseases, die of infectious complications (9,14). In addition to the temporary immune suppression and frequent surgical procedures, these patients are prone to mucositis and almost all have indwelling catheters increasing the infection risk (15–19). Furthermore, it has become evident that some children are burdened with recurrent infections during their treatment, whereas others experience fewer complications. The reason for this still remains unknown. It is urgent to analyse which factors make children prone to infections and why some children seem to suffer from these complications more often than others. The annual incidence of acute lymphoblastic leukaemia (ALL) in children aged 0–14 years in the Nordic countries is around 4.0 per 100 000 and has remained relatively stable during the last decades (20). The total number of children diagnosed annually with acute lymphoblastic leukaemia in the Nordic countries is therefore approximately 170. More than 25 years ago the Nordic countries – Iceland, Sweden, Norway, Denmark and Finland – founded the Nordic Society of Paediatric Haematology and Oncology (NOPHO) in order to improve and coordinate registration, treatment and monitoring of results for children with malignant diseases. This cooperation has been very successful. The 5-year survival of acute lymphoblastic leukaemia, standard risk or intermediate risk in the NOPHO cooperation, is continuously improving (5,21,22). Treatment of infectious complications in children undergoing chemotherapy for malignancies is complicated. Use of broad-spectrum antibiotics is often necessary but can at the same time be a confounding factor (9,23,24). In addition, the incidence, types of bacteria and their antimicrobial susceptibility, is continuously changing (17,25–28). It is therefore necessary to know the infectious agents in various places and at different times. Although long-term survival of children with malignant diseases has improved substantially, infections are still a ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1149–1155 1149 Suspected infections in ALL Hafsteinsdóttir et al. MATERIAL AND METHODS All children 0–18 years of age with acute lymphoblastic leukaemia, standard risk or intermediate risk, diagnosed in Iceland and southern Sweden during the 10-year period 1996–2005, were enrolled in the study. Diagnosis, gender, country of residence, as well as other underlying diseases or disorders were recorded from hospital records. From the patients’ registry, information was collected on blood results at diagnosis and during fever or suspected infectious episodes according to treatment protocols. The blood results retrieved were bacterial cultures, white blood cell (WBC) count, leukocyte differential count, capsular reactive protein (CRP) and tests indicating disease activity. Neutropenia was defined as the number of neutrophils in peripheral blood less than 0.5 · 109 ⁄ L. The length and type of antibiotic treatment was recorded as well as adjustments or changes in antibiotic treatment during the period. Data were collected for the first 12 months after initial diagnosis. All patients were treated according to existing NOPHO protocols for this disease (NOPHO ALL-92 and NOPHO ALL-2000) (5,21,29). Suspected infectious episodes were defined as fever of more than 38.5C during leukopenia or when the responsible paediatrician suspected bacterial infection and started antibiotic treatment according to protocol. The number of suspected infectious episodes was recorded for each patient as well as the number of positive blood or fungal cultures. The bacteria or fungi cultured from blood on all the occasions were noted and blood results during these episodes were recorded. Other suspected bacterial or fungal infections were also recorded. When blood cultures grew bacteria or fungi, the infection was regarded as confirmed. Statistical analysis was done using generalized linear models with Poisson distribution (Poisson regression) and also with binomial distribution. In addition t-tests of normalized data were used in comparison of means. A result was considered significant if the p-value was less than 0.05. Appropriate approval was obtained from the Icelandic and Swedish National Bioethics Committees and the privacy and data protection authorities in both countries and the study was approved by the local hospital authorities. RESULTS The total number of patients involved in the study was 73 in the age range 1.5)17 years, 43 boys and 30 girls, with a gender ratio of 1.43. The median age at diagnosis was 4.6 years overall, 4.4 years for boys and 4.8 years 1150 Table 1 Suspected and confirmed infectious episodes according to age and gender Average number of infections Boys Girls p-value (two-tailed) Number of cases (children) complicating factor in the treatment and a significant threat to these children. In our study of 73 children treated for ALL, we aim to get better epidemiological information on infections of these children and an attempt is made to analyse which patients might be more prone to infections than others. The main purpose of the study is to try to define patients needing special attention or follow-up because of increased risk of infection. Average age (year) Suspected Confirmed 5.76 5.18 >0.10 2.02 3.07 0.03 0.77 0.80 >0.10 boys (43 in all) girls (30 in all) 14 12 10 8 6 4 2 0 0 1 2 3 4 5 6 7 8 Number of suspected infections within 1 yr Figure 1 Distribution of the number of suspected infectious episodes according to gender. for girls (corresponding means were 5.5, 5.2 and 5.8 years) (Table 1). The number of patients from Sweden was 65 and from Iceland eight. Two patients died during the study. The 73 patients had a total of 179 episodes with suspected infection. Of the 73 patients, 67 had one or more episodes with suspected infection; the number of suspected infectious episodes varied from none in six patients to eight in one patient (Fig. 1 and Table 2), the average number of episodes being 2.45. The number of patients with three or more episodes of suspected infections was 29, the number of periods with positive culture ranging from one to four in this group (Table 2). Of the 73 patients, 68 had already a Port-A-Cath (PAC) system or a central venous device at the time of their first episode. All patients had a central venous device, most often a PAC, during the treatment period. Seven of the 73 patients were started on antimicrobial treatment within 48 h of the first admission, either before or very soon after insertion of a central venous device. Only a few of the children received antibacterial prophylactic therapy, but 40 of the 73 children received nystatin treatment. The PAC was removed because of suspected infection in 35 of the 179 episodes. Of the 57 culture-confirmed bacterial infections, the PAC was removed on 27 occasions. The PAC was removed after 8 of the 122 episodes of suspected infections that were not confirmed with positive blood cultures. Of the 179 suspected infectious episodes, blood cultures were negative in 122 (68.2%) but bacteria was cultured in 57 (31.8%). The bacteria most often cultured was coagulasenegative staphylococcus (CNS) that was found in 23 of ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1149–1155 Hafsteinsdóttir et al. Suspected infections in ALL Table 2 Pattern of infections for children with three or more episodes of suspected infection and removal of PAC Suspected infection episode No. Gender Age 1st 2nd 3rd 4th 5th 6th 7th 8th 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Boy Girl Girl Girl Girl Girl Girl Boy Boy Girl Boy Girl Girl Girl Boy Girl Girl Girl Girl Girl Boy Boy Boy Boy Boy Boy Girl Boy Boy 4 3 7 2 4 3 4 2 3 6 3 2 4 5 4 6 4 3 15 6 4 5 2 6 3 3 7 4 6 E. coli * CNS CNS(#) CNS(#) * * CNS CNS(#) Vir strept(#) E. coli * * * E. faec CNS Proteus sp E. coli * * * * * * GNR CNS Mor(#) St aur(#) CNS * S. marc * CNS * * * * CNS * *(#) *(#) * * H. infl E. coli * CNS(#) *(#) *(#) * * * Pseud * C. par(#) * St aur(#) * CNS(#) * * CNS(#) * * CNS(#) * * * * *(#) * * CNS E. coli(#) * * * * * * CNS * * CNS(#) CNS * * * CNS(#) * * * * * * *(#) * * * * * CNS(#) * Cor sp * * * * * * * * * * * * * * * * * * CNS(#) * ¼ suspected but not confirmed infection; (#) ¼ removal of PAC; CNS ¼ coagulase-negative staphylococcus; S. marc ¼ Serratia marcescens; Vir strept ¼ Streptococcus viridance; E. faec ¼ Enterobacter faecalis; H. infl ¼ Haemophilus influenzae; Proteus sp ¼ Proteus species; Cor sp ¼ Corynebacterium species; Pseud ¼ Pseudomonas species; GNR ¼ Gram-negative rod; C. par ¼ Candida parapsilosis; Mor ¼ moraxella; St aur ¼ Staphylococcus aureus. the culture-positive cases (40.4%), followed by Staphylococcus aureus in eight cases (14.0%) and Esherichia coli was cultured on five occasions (8.8%). Pseudomonas, betahaemolytic streptococcus group A and pneumococci were cultured three times each, other bacteria were less common. In the 179 episodes of suspected infection, X-ray confirmed pneumonias were diagnosed on 32 occasions. Some of the patients had pneumonias more than once, thus the number of patients was 26. In 7 of the 32 pneumonias, Pneumocystis jiroveci infection was suspected but only confirmed in one. Respiratory syncytial virus (RSV) was confirmed twice. In 8 of the 32 episodes, a positive blood culture was obtained at the time of the pneumonia diagnosis. In four instances, CNS was cultured, once concomitantly with an RSV virus. In addition, E. coli, Candida parapsilosis, S. aureus and Enterococcus faecalis were cultured. It remained questionable if these microbes were the causative agents of the pneumonia. In three patients, cellulitis was diagnosed during these episodes and two had complicated colitis. Two patients developed abscesses causing severe complications, one of whom had a positive blood culture growing pseudomonas as was found in an abdominal abscess. Some patients might have had other complications during their treatment period with various results of bacterial cultures. The most common bacteria were coagulase-negative staphylococci, cultured in 17 patients on 23 occasions. The bacteria were most common in the induction phase when it was cultured 14 times, 6 times in the consolidation phase, once in the reinduction and twice in the maintenance phase. In 13 (56%) of these episodes, the PAC was removed and later replaced. On one occasion a C. parapsilosis was cultured. Patients with three or more episodes of suspected infection were 29. Seven patients had a total of 19 successively positive cultures (Table 2). In those cases, 12 cultures grew coagulase-negative staphylococcus and two grew S. aureus. In 9 of these 14 episodes with staphylococcus the PAC was removed. Repeated positive cultures, that is episodes when in the next episode bacteria or fungi were also grown, were analysed (Table 2). These episodes were 12; in eight of them coagulase-negative staphylococcus and in one S. aureus were grown. Five cultures grew the same bacteria as was detected in the previous episode; ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1149–1155 1151 Average number of suspected / confirmed infections in each age group Suspected infections in ALL Hafsteinsdóttir et al. 3.5 suspected confirmed 3 2.5 2 28 children 1-4yr 1.5 30 children 4-7yr 10 children 7-11yr 1 5 children 11-18yr 0.5 0 2 4 6 8 10 12 14 16 Age, yr (the dots are at average age in age group) Figure 2 Suspected and confirmed infections according to age. in two of these the central venous device had been removed. In three of the 12 cases of repeated infections the central venous catheter had been removed before the repeated infection. The number of infectious episodes by gender is given in Table 1. There were significantly more suspected infections in girls than in boys (3.07 vs. 2.02), but there was no noticeable difference in confirmed infections. Figure 2 shows the number of infections by age. Experiments with a few different canonical link functions with Poisson regression for these data gave the models: logðksusp Þ ¼ 1:844 0:629 logðageÞ and logðkconf Þ ¼ 0:651 0:595 logðageÞ, where ksusp and kconf are the expected numbers of suspected and confirmed infections. The coefficients of log (age) were both significantly negative with p ¼ 0.00011 for suspected infections and p ¼ 0.038 for confirmed infections. Thus, both suspected and confirmed infection numbers fell significantly with age. An overdispersion parameter was estimated to be 1.0, confirming that overdispersion was not present. The average number of suspected infections for the youngest age group (1–4 years) was 3.29 but 1.20 for the oldest group (11– 18 years) (Fig. 2). Corresponding numbers for confirmed infections were 1.20 and 0.40 (Fig. 2). The fraction of infections confirmed remains around onethird for all age groups (it is in fact 33, 31, 31 and 33% in the four groups). At the initial diagnosis the median total WBC count was 5.3 (mean 10.4; range 0.95–41.0), the median neutrophil count was 0.40 (mean 0.64; range 0.08–3.80), and the median number of blasts in peripheral blood was 1.30 (mean 5.46; range 0.0–36.3). In suspected infectious episodes when blood cultures grew bacteria, the average CRP value was 63.7 mg ⁄ L but 45.5 when cultures were negative, the difference being statistically significant (p ¼ 0.02). The total WBC count or number of neutrophils was not significantly different. The total WBC count, number of neutrophils and lymphoblasts at diagnosis or at the start of suspected infectious episodes had no significant correlation with the number of suspected infections or outcome of bacterial cultures. In the 179 episodes of suspected infection, the patients were neutropenic (number of neutrophils <0.5 · 109 ⁄ L) on 139 occasions or 78%. In the 57 episodes when bacteria were cultured, the patients were neutropenic in 48 (84%). In the 122 episodes when cultures were negative, the patients were neutropenic 91 times (75%). Bacterial infection was confirmed by a positive bacterial culture in 34 of 67 episodes (51%) in the first event after diagnosis. In subsequent episodes of suspected infections, the infections were confirmed relatively less often (Table 3). Taken together, 53% of the episodes occurred in the induction phase, 10% in the consolidation phase, 3% in the reinduction and 34% in the maintenance phase (Table 3). Some differences in these figures were noted between different risk groups of ALL and different protocols (Table 3). There were altogether 10 episodes that were the sixth, seventh or eighth episode that a patient had, and of these only one was confirmed (Fig. 3). The reduction in the proportion of confirmed infection is significant with p < 0.001 according to a binomial distribution generalized linear model. Table 3 Number of episodes (percentages) with suspected infections according to treatment protocol and phase of treatment, that is induction, consolidation, reinduction (delayed intensification) and maintenance therapy and total number of confirmed bacterial infections Induction Consolidationc Reinductiond Maintenancee Total SI (N ¼ 9)a Number of episodes II (N ¼ 15)a Number of episodes SR (N ¼ 32)b Number of episodes IR (N ¼ 17)b Number of episodes Total number of episodes Total number of confirmed bacterial infections 8 (33) 2 (8) – 14 (58) 24 18 (46) 3 (8) 4 (10) 14 (36) 39 43 (54) 12 (15) – 25 (31) 80 26 (72) 1 (3) 1 (3) 8 (22) 36 95 (53) 18 (10) 5 (3) 61 (34) 179 (100) 45 (25) 5 (3) 3 (2) 4 (2) 57 (32) a SI and II refer to patients without unfavourable features according to NOPHO-ALL 2000 standard intensive protocol (SI) and intermediate intensive protocol (II). SR and IR refer to standard risk and intermediate risk according to NOPHO-ALL 1992 treatment protocol. c Starting on day 50 for SI, II and SR but on day 106 for IR. d None in SI and SR, starts on day 120 for II and on day 169 for IR. e Starting on days 120, 211, 99 and 225, respectively. b 1152 ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1149–1155 Proportion of suspected infections which are confirmed Hafsteinsdóttir et al. Suspected infections in ALL 50% 40% 30% 20% total 67 conf. 34 total 47 conf. 11 10% 0% 1st 2nd total 29 conf. 8 3rd total 26 conf. 3 total 10 conf. 1 4th-5th 6th-8th Suspected infection episode (the numbers inside the bars give total / confirmed episode count) Figure 3 Suspected and confirmed infections according to whether it is the 1st, 2nd, 3rd, etc., time that suspicion arises and treatment is instigated. Antibiotic treatment was altered or changed in 60 (47 patients) of the 179 episodes (33.5%). DISCUSSION Bacterial infection in children treated for malignant diseases can be a complicating factor in the treatment and a serious life-threatening event (10–13). For some reason, some children experience greater number of episodes with fever and suspected infections than others. Our study attempts to identify some of the underlying factors causing this. In our study, 179 suspected infections in 73 children treated for ALL were analysed. The patients experiencing frequent episodes with suspected infections were mainly the younger patients in our study. This has been described earlier (9). The male–female ratio in our study was 1.43. Interestingly, girls had more suspected infectious episodes although positive bacterial cultures were equally common in both sexes. The higher number of suspected infections in girls has not been reported earlier and the reason for this is unclear. However, one study found that girls were at a higher risk of treatment-related death, mostly because of infection (9). In 68% of the episodes with suspected infections in our study, no bacteria could be cultured from the blood. However, one must bear in mind that on many occasions only one blood sample could be drawn for culture before initiating antibiotic treatment. One must also keep in mind that these children tend to be severely ill, not allowing for series of blood cultures to be drawn. It can therefore be expected that the study underestimates the number of bacteraemias in the patient group. In addition, several other factors may cause infections in this group of patients, including viral and fungal infections. Immunological disturbances as well as activity of the underlying disease in addition to drugs may also cause clinical symptoms and fever, leading to a suspicion of infections. Only six patients had no episode of suspected infection. The all remaining patients had suspected infections and 29 had three or more episodes. On several occasions, the same bacteria were cultured repeatedly, in particular coagulasenegative staphylococcus, sometimes despite removal of the central venous catheter. This does not come as a surprise but underlines the need for careful handling of these devices. Nevertheless, although removal of PAC almost always requires a new central venous catheter, the reappearance of CNS in some patients is a concern and might raise the question of the effect of the procedure. Coagulase-negative staphylococci were the most common bacteria cultured. This is in agreement with other studies reporting an increasing incidence of Gram-positive bacteria in children with malignant diseases, in particular coagulase-negative staphylococci (10,15,17,25–28,30). This is related to more intensive chemotherapy and increased use of indwelling central venous catheters (26–28). Although coagulase-negative staphylococci can frequently be regarded as contamination during blood sampling, this is certainly not as simple in children receiving cancer treatment. Infections caused by coagulase-negative staphylococci must be taken into consideration as potential pathogens in this patient group. Our study also indicates that most of the serious infections involving life-threatening pathogens occur early in the treatment period. This may be partly explained by the fact that the patients are the most immunocompromised during induction treatment. In addition, patients with recurrent episodes of fever with suspected infections tended to have negative blood cultures in the second half of their treatment period. The cause of this remains uncertain but may reflect increased awareness of severe infections in this subgroup of patients. In our study, infection parameters from blood, that is WBC count, number of neutrophils and CRP, were recorded when antibiotic treatment was started. The WBC and number of neutrophils were frequently low as was expected in this patient group. The ratio between neutropenic and nonneutropenic patients was similar in patients with positive or negative cultures. We found that CRP was significantly higher in the febrile periods when blood cultures revealed bacteria compared to those that did not. CRP has been studied extensively in cancer patients in order to detect bacterial infections (18,31). Although studies, including ours, find higher CRP in patients with proven bacterial infections, these results should be evaluated with caution. CRP can certainly be raised due to other reasons than bacterial infections and, more importantly, low CRP does not exclude severe infections. The initial antibiotic treatment in the two study centres is somewhat different as is often the case in different countries (15,30,32,33). In Iceland, the initial treatment is piperacillin and aminoglycoside, whereas the empirical treatment in southern Sweden is usually imipenem or ceftazidime. The treatment needed to be adjusted or altered in one-third of the episodes; usually antibiotics with extended spectrum toward Gram-positive bacteria were often added to the treatment. Colony-stimulating factors are at this time not included in the empirical therapy, but beneficial effects of this treatment has been reported (15,30,33). Apparently, the treatment regimens in both centres seem to be appropriate considering the types of bacteria cultured. Prophylactic antibacterial treatment was rarely used in our cohort, in particular in southern ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1149–1155 1153 Suspected infections in ALL Hafsteinsdóttir et al. Sweden, but prophylactic treatment against fungi was common. There is clearly a difference between various centres in terms of antimicrobial prophylactic therapy in this patient group. Awareness is needed for increasing number of Gram-positive bacteria, but at the same time complications of over usage or wrong use of antibiotics must be kept in mind (23,24). Our study provides important information regarding patients with recurrent episodes of suspected bacterial infections. The results are helpful in deciding on initial antibiotic treatment in this patient group. The study also provides interesting background for further research into suspected infections in patients treated for malignant diseases. References 1. Longo DL. Approach to the patient with cancer. In: Kasper DL, Braunwald E, Fauci AS, Hauser SL, Longo DL, Jameson JL, editors. Harrison’s principles of internal medicine. 16th ed. New York: McGraw-Hill, 2005. 2. Nachman JB, Abelson HT. Oncology. In: Behrman E, Kliegman R, editors. Nelson essentials of pediatrics. 4th ed. Philadelphia: WB Saunders, 2002. 3. Hallbook H, Gustafsson G, Smedmyr B, Soderhall S, Heyman M. Treatment outcome in young adults and children >10 years of age with acute lymphoblastic leukemia in Sweden: a comparison between a pediatric protocol and an adult protocol. Cancer 2006; 107: 1551–61. 4. Saarinen-Pihkala UM, Gustafsson G, Carlsen N, Flaegstad T, Forestier E, Glomstein A, et al. Outcome of children with high-risk acute lymphoblastic leukemia (HR-ALL): Nordic results on an intensive regimen with restricted central nervous system irradiation. Pediatr Blood Cancer 2004; 42: 8–23. 5. Gustafsson G, Schmiegelow K, Forestier E, Clausen N, Glomstein A, Jonmundsson G, et al. Improving outcome through two decades in childhood ALL in the Nordic countries: the impact of high-dose methotrexate in the reduction of CNS irradiation. Leukemia 2000; 14: 2267–75. 6. Pui CH, Boyett JM, Rivera GK, Hancock ML, Sandlund JT, Ribeiro RC, et al. Long-term results of Total Therapy studies 11, 12 and 13A for childhood acute lymphoblastic leukemia at St Jude Children’s Research Hospital. Leukemia 2000; 14: 2286–94. 7. Allison AC. Immunosuppressive drugs: the first 50 years and a glance forward. Immunopharmacology 2000; 47: 63–83. 8. Kristinsson V, Kristinsson J, Jonmundsson G, Jónsson O, Thorsson A, Haraldsson A. Immunoglobulin class and subclass concentrations after treatment of childhood leukemia. Pediatr Hematol Oncol 2001; 18: 167–72. 9. Christensen MS, Heyman M, Mottonen M, Zeller B, Jonmundsson G, Hasle H, et al. Treatment-related death in childhood acute lymphoblastic leukaemia in the Nordic countries: 1992–2001. Br J Haematol 2005; 131: 50–8. 10. Auletta JJ, O’Riordan MA, Nieder ML. Infections in children with cancer: a continued need for the comprehensive physical examination. J Pediatr Hematol ⁄ Oncol 1999; 21: 501–8. 11. Finberg R. Infections in patiens with cancer. In: Kasper DL, Braunwald E, Fauci AS, Hauser SL, Longo DL, Jameson JL, editors. Harrison’s principles of internal medicine. 16th ed. New York: McGraw-Hill, 2005. 12. Shaw PJ. Suspected infection in children with cancer. J Antimicrob Chemother 2002; 49(Suppl 1): 63–7. 1154 13. Urrea M, Rives S, Ofelia C, Navarro A, Garcı́a JJ, Estella J. Nosocomial infections among pediatric hematology ⁄ oncology patients: results of a prospective incidence study. Am J Infect Control 2004; 34: 205–8. 14. Hargrave DR, Hann II, Richards SM, Hill FG, Lilleyman JS, Kinsey S, et al. Progressive reduction in treatment-related deaths in Medical Research Council childhood lymphoblastic leukaemia trials from 1980 to 1997 (UKALL VIII, X and XI). Br J Haematol 2001; 112: 293–9. 15. Blijlevens NMA, Donnelly JP, de Pauw BE. Empirical therapy of febrile neutropenic patients with mucositis: challenge of risk-based therapy. Clin Microb Infect 2001; 7: 47–53. 16. Dourd MC, Arlet G, Leverger G, Paulien R, Waintrop C, Clementi E, et al. Quantitative blood cultures for diagnosis and management of catheter-related sepsis in pediatric hematology and oncology patients. Intensive Care Med 1991; 17: 30–5. 17. Pizzo PA. Fever in immunocompromised patients. Curr Concepts 1999; 341: 893–9. 18. Santolaya ME, Alvarez AM, Becker A, Cofre J, Enriquez N, O’Ryan M, et al. Prospective, multicenter evaluation of risk factors associated with invasive bacterial infection in children with cancer, neutropenia, and fever. J Clin Oncol 2001; 19: 3415–21. 19. Pizzo PA. Management of fever in patients with cancer and treatment-induced neutropenia. N Engl J Med 1993; 328: 1323–32. 20. Hjalgrim LL, Rostgaard K, Schmiegelow K, Soderhall S, Kolmannskog S, Vettenranta K, et al. Age- and sex-specific incidence of childhood leukemia by immunophenotype in the Nordic countries. J Natl Cancer Inst 2003; 95: 1539–44. 21. Gustafsson G, Kreuger A, Clausen N, Garwicz S, Kristinsson J, Lie SO, et al. Intensified treatment of acute childhood lymphoblastic leukaemia has improved prognosis, especially in non-high-risk patients: the Nordic experience of 2648 patients diagnosed between 1981 and 1996. Acta Paediatr 1998; 87: 1151–61. 22. Forestier E, Johansson B, Gustafsson G, Borgstrom G, Kerndrup G, Johannsson J, et al. Prognostic impact of karyotypic findings in childhood acute lymphoblastic leukaemia: a Nordic series comparing two treatment periods. Br J Haematol 2000; 110: 147–53. 23. Ammann RA, Hirt A, Luthy AR, Aebi C. Identification of children presenting with fever in chemotherapy-induced neutropenia at low risk for severe bacterial infection. Med Pediatr Oncol 2003; 41: 436–43. 24. Klastersky J. Empirical treatment of sepsis in neutropenic patients. Int J Antimicrob Agents 2000; 16: 131–3. 25. Cohen J, Brubn-Buisson C, Torres A, Jorgensen J. Diagnosis of infection in sepsis: an evidence-based review. Crit Care Med 2004; 32(Suppl 11): 466–94. 26. Hengartner H, Berger C, Nadal D, Niggli FK, Grotzer MA. Port-A-Cath infections in children with cancer. Eur J Cancer 2004; 40: 2452–8. 27. Philpott DC, Georg RH. Central venous catheter-related septicaemia in peadiatric cancer patients. J Hosp Infect 1997; 36: 67–76. 28. Rosato AE, Tallent SM, Edmond MB, Bearman GML. Susceptibility of coagulase-negative staphylococcal nosocomial boodstream isolates to the chlorhexidine ⁄ silver sulfadiazine-impregnated central venous catheter. Am J Infect Control 2004; 32: 486–8. ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1149–1155 Hafsteinsdóttir et al. 29. Ruud E, Holmstrom H, de Lange C, Natvig S, Albertsen BK, Wesenberg F. Thrombotic effects of asparaginase in two acute lymphoblastic leukemia protocols (NOPHO ALL-1992 versus NOPHO ALL-2000): a single-institution study. Pediatr Hemat Oncol 2006; 23: 207–16. 30. Haslam DB. Managing the child with fever and neutropenia in an era of increasing microbial resistance. J Pediat 2002; 140: 5–7. 31. Fleischhack G, Kambeck I, Cipic D, Hasan C, Bode U. Procalcitonin in paediatric cancer patients: its diagnostic relevance is superior to that of C-reactive protein, interleukin 6, Suspected infections in ALL interleukin 8, soluble interleukin 2 receptor and soluble tumor necrosis factor receptor II. Br J Haematol 2000; 111: 1093–102. 32. Chuang YY, Hung IJ, Yang CP, Jaing TH, Lin TY, Huang YC. Cefepime versus ceftazidime as empiric monotherapy for fever and neutropenia in children with cancer. Pediat Infect Dis Jl 2002; 21: 203–9. 33. Kline RM, Baorto EP. Treatment of pediatric febrile neutropenia in the era of vancomycin-resistant microbes. Pediatr Blood Cancer 2005; 44: 207–14. ª2009 The Author(s)/Journal Compilation ª2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1149–1155 1155