Post-Transplant Complications

Aerobic bacterial and fungal infections in peripheral blood stem cell transplants

Article metrics


Allogeneic and autologous peripheral blood stem cell transplants are frequently complicated by infections. This study was performed to evaluate early and late infections in 74 patients who underwent peripheral blood stem cell transplantation (PBSCT). Fifty-eight patients received allogeneic and 16 autologous PBSCT. All patients received fluconazole, ciprofloxacin and acyclovir prophylaxis. 93.1% of alloPBSCT patients and 87.5% of autoPBSCT patients developed fever. Febrile episodes were commonly seen in the week of transplantation (66%). There was a median of 3 days with fever in alloPBSCT, and 2 days in autoPBSCT. Period of neutropenia was 15 days for AlloPBSCT and 12 days for AutoPBSCT. The microbiological identification rate was 47% (32/68). Gram-positive infections dominated the early period (50%) and Gram-negative bacterial infections dominated the late period (50%). All our patients had Hickman-type catheters and 26 infections involving catheters were seen. Sixteen occurred in the early, and 10 in the late period. Ten of 14 (71.4%) late bacterial infections were catheter-related. The dominance of Gram-positive infections and high rates of methicillin resistance warranted the use of vancomycin extensively. Surveillance cultures were found to be useful in selected patients. Although slime factor is an important virulence factor, there was no difference between slime factor positive and negative coagulase-negative staphylococci isolated during infections. In conclusion, febrile episodes are the most frequent complication of PBSCT and Gram-positive microorganisms remain the main pathogen in these patients because of catheter use, mucositis and ciprofloxacin prophylaxis. Methicillin resistance is increasing and glycopeptides remain the only choice for treating such infections. Although the infection rate is high, measures taken to prevent and treat infections result in very low rates of mortality from infection in PBSCT patients. Bone Marrow Transplantation (2001) 27, 201–205.


Infections are the main cause of mortality and morbidity in bone marrow (BMT) and peripheral stem cell transplant (PBSCT) patients. These patients have defects in both humoral and cellular immunity. Bacteria and fungi cause 95% of all infections post transplant.12 In this study we investigated aerobic bacterial and fungal infections during the conditioning, transplantation, engraftment and post-engraftment period, between 9 January 1996 and 16 February 1998 at Ankara University, Ibn-i Sina Hospital, Department of Hematology, Bone Marrow Transplantation Unit. Colony-stimulating factor (CSF)-mobilized peripheral stem cells were used for all transplants. We also investigated the neutropenia period, relation of infection to neutropenia, febrile period and incidence of microbiologically defined infections.

Early diagnosis and immediate therapy is particularly important in these patients. Also, the microorganisms and their resistance vary, particularly in different hospitals. To achieve the best results in fighting infections in these patients, the dominant microorganisms should be identified and appropriate therapy should be based on their susceptibility patterns.

Patients and methods

Patients and peripheral blood stem cell transplantation

Seventy-four patients were transplanted between 9 January 1996 and 16 February 1998 at Ankara University, Ibn-i Sina Hospital, Department of Hematology, and Bone Marrow Transplantation Unit. The BMT unit has six single rooms. The rooms have UV light but no laminar airflow. Reverse isolation is used. All personnel are required to wear a mask and a gown. Savlon is used for disinfection of hands before patient examination. Patients eat sterilized, cooked food. They receive ciprofloxacin 500 mg/day from day −8 to engraftment, fluconazole 200 mg/day from day −8 to day +100, and acyclovir 1 g/day from day −8 to day +100 as prophylaxis. Trimethoprim/sulfamethoxazole (160/800 mg) is given twice a day from day −8 to day 0 and reinitiated upon engraftment until day +100. All patients have Hickman-type intravenous catheters. Growth factors are given only to patients undergoing autoPBSCT. We attempted to isolate aerobic bacteria and fungi from all patients starting from conditioning until engraftment. In addition, we attempted to isolate microorganisms from these patients if they were rehospitalized 30–100 days after transplantation. Conditioning and the pre-engraftment period were defined as the early period; the post-engraftment period was defined as the late period.

Sample collection

Blood cultures were obtained from the vein or catheter. Urine, throat and stool samples were taken twice a week. At least two blood cultures were obtained from peripheral veins half an hour apart. A Gram-positive infection was diagnosed if two consecutive blood cultures or one culture from the catheter and one from the peripheral vein taken simultaneously yielded a Gram-positive microorganism.

Isolation and identification of microorganisms

For blood cultures, the BaCT/Alert (Organon-Teknika, Durham, NC, USA) automated blood culture system was used. Other cultures were performed by conventional methods and bacteriological identification by standard methods.34 Candifast (International Microbio, Signes, France) was used for Candida typing. Antibiotic susceptibilities were defined by the Kirby-Bauer disc diffusion method5 and determined according to NCCLS.6 Congo Red was used to determine slime production.7

Definition of infections

The definitions of Center for Disease Control (CDC) for nosocomial infections were used to describe infections.8

Statistical analysis

Data were analyzed using an SPSS statistical package (SPSS, Chicago, IL, USA).


There were 48 male and 26 female patients. Median age was 34 (4–48). Fifty-eight patients received alloPBSCT and 16 autoPBSCT. Patient characteristics are shown in Tables 1a and 1b.

Table 1 Patient characteristics
Table 2 Recovery times for granulocytes

Febrile episodes and survival

A total of 91 febrile episodes were seen in all patients. Sixty-eight episodes were during febrile neutropenia. Six alloPBSCT patients and two autoPBSCT patients experienced no febrile episodes. Forty-nine patients experienced one febrile episode and 17 experienced more than one. The time relation of neutropenia to 68 febrile episodes occurring in the early period is shown in Tables 1a and 1b. Overall survival was a median of 31.5 months (0.5–54) for patients undergoing alloPBSCT, a median of 9 months (0.5–50) for patients undergoing autoPBSCT and a median of 25 months (0.5–54) for all patients. Thirty-one of 58 patients are alive after alloPBSCT and five of 16 patients are alive after autoPBSCT. In the autoPBSCT group, eight of 11 patients died because of primary hematological malignancy progression, two of late graft failure and one of early aplasia. In the alloPBSCT group, 11 patients died of GVHD, eight of disease progression, three with early aplasia, two with graft failure, one after hemolytic uremic syndrome, one with VOD and one with intracranial bleeding.

Microbiologically documented bacterial and fungal infections

Of 46 febrile episodes with microbiologically documented infections, 32 were in the early period (Table 2). Eight of 32 were detected in patients undergoing autoPBSC (three bacteremia, two catheter-related bacteremia, one fungemia, one catheter-related infection, one urinary tract infection). The frequency and type of infection did not differ significantly between alloPBSCT and autoPBSCT (P > 0.05). Microbiologically documented infections documented in the early and late period are shown in Table 2. Ten of 14 (71.4%) late occurring bacterial infections were related to catheter use.

Table 3  Microbiologically documented infections in both early and late period of PBSCT

Catheter-related bacteremia, fungemia and infections

There were 26 catheter-related bacteremias and fungemias. Ten of these infections required removal of the catheter. Sixteen (61.5%) infections were treated with only antibiotics. Sixteen early period infections were caused by nine Gram-positive bacteria, four Gram-negative bacilli, two fungus and one polymicrobial agent (Table 2). Ten late period infections were caused by four Gram-positive bacteria, three Gram-negative bacilli, and three polymicrobial agents (Table 2).

Antibiotic resistance

Methicillin resistance of S. aureus and Coagulase-negative staphylococci rates were 66% and 85%, respectively. The susceptibility of Gram-positive bacteria to vancomycin, ciprofloxacin, imipenem/cilastatin, and ampicillin/ sulbactam was 100%, 38%, 20% and 16%, respectively. The susceptibility of enteric Gram-negative bacilli (EGNB) was ciprofloxacin (42%), amikacin (80%), gentamycin (47%) cefepime (71%), ceftazidime (57%), piperacillin (33%) and imipenem/cilastatin (100%). The susceptibility of nonfermentative Gram-negative bacilli (NFGNB) was ciprofloxacin (33%), amikacin (66%), gentamycin (66%), cefepime (66%), ceftazidime (66%), piperacillin (66%) and imipenem/cilastatin (100%).

Surveillance cultures

Colonization of Klebsiella in the throat was found in 10 patients and two had catheter-related septicemia with Klebsiella. In one patient MRSA colonized the throat; fever developed and responded to vancomycin. In one other patient, E. agglomerans was repeatedly found in urine cultures when the patient was asymptomatic. The patient developed fever and responded to specific therapy. Between June and July 1997, E. cloacae colonization, only susceptible to imipenem/cilastatin was found sequentially in throat, stool and urine cultures of four patients. Two of them developed a urinary infection by the same pathogen.

Slime production

The presence of slime factor was analyzed by the Congo red method in 20 pathogenic Coagulase-negative staphylococci (CNS). Slime factor was found to be present in two of five patients with bacteremia and eight of 15 patients with catheter infection/catheter-related bacteremia. There was no significant difference between the distribution of infection types and slime production of the isolates.


In the early period after BMT, neutropenia is the major factor predisposing to infection, as is the case in leukemia patients receiving conventional chemotherapy. Recently, peripheral blood stem cells have been increasingly used as a source of stem cells and carry the advantage of faster neutrophil and platelet recovery although an increased risk of chronic GVHD remains a possibility.91011 In a previous study conducted in our center, neutrophil counts remained below 1.0 × 109/l for a median of 17 days after alloBMT, 11 days after autoBMT and 11 days after autoPBSCT.12 In our present study, these medians were 15 days for alloPBSCT and 12 days for autoPBSCT, respectively. There was a median of 3 days of fever after alloPBSCT, and 2 days after autoPBSCT. Short periods of fever can be attributed to the shortened period of neutropenia and prompt initiation of antibiotics. Febrile episodes are commonly seen in the week of transplantation (66%). Late febrile episodes are less common, mainly because of the short period of neutropenia. Catheter-related infections were primary responsible for the late period infections (71.4%). Prolonged immunosuppression for GVHD and inadequate catheter maintenance in the outpatient setting explain the high rate of late catheter infections in our patients.

In different studies, the rate of microbiological diagnosis varies between 32.5 and 48%. In this study the rate is 47% (32/68), comparable with the literature.1314 Although Gram-negative microorganisms were the main cause of infections in patients with febrile neutropenia, the EORTC Group showed an increase in Gram-positive infections from 29% to 67% and decrease in Gram-negatives from 71% to 31%.15 The GIMEMA group repeated these results and showed an increase in Gram-positive microorganisms from 64% to 89%.16 This trend was also demonstrated in Turkey.1217 This can be explained by the routine use of catheters, severe mucositis and ciprofloxacin prophylaxis in transplant patients. In this study the rate of Gram-positive infections was 50%, Gram-negative 40.6%, fungal infections 6.2% and polymicrobials 3.1%. If only the early period was taken into account, the rate of Gram-positive infections was 66.6% in this study. Staphylococci predominated and the most frequent isolate was S. epidermidis. Although Klebsiella and E. coli were the most common pathogens in previous studies in the same center, this study showed an increase in the rate of Enterobacter spp., mainly E. cloacae, indicating a change in the Gram-negative pathogens identified.

The incidence of fungal infection in transplant patients is approximately 22–25%. Although Candida albicans is the main pathogen, the incidence of C. parapsilosis and C. glabrata is increasing, related to catheters. Fluconazole-resistant C. krusei seems to be increasing, related to fluconazole prophylaxis.1819 We had two fungemias, C. krusei, which was resistant to fluconazole, and C. parapsilosis, in the early period.

All early aplastic deaths and deaths due to graft failure were related to infections. No death related to bacterial and fungal pathogens occurred in engrafted patients.

In our study, colonization by Klebsiella in the throat was found in 10 patients and two had catheter-related septicemias with Klebsiella. In one patient MRSA was found to colonize the throat; fever developed and responded to vancomycin. In one other patient, E. agglomerans was repeatedly found in urine cultures when the patient was asymptomatic. The patient developed fever and responded to specific therapy.

In a multicenter study in Europe, methicillin resistance of S. aureus was found to be between 20 and 60% and of Coagulase-negative staphylococci 40%.20 In our study these rates are 66% and 85%, respectively. A similar study was performed in Turkey in nine centers and a similar resistance pattern was observed in 981 Gram-negative isolates.21 In the study, imipenem/cilastatin, amikacin, ciprofloxacin, cefepime and ceftazidime were the most effective agents against Gram-negative isolates, which is consistent with our results.

Slime factor is considered a pathognomic factor for Coagulase-negative staphylococci and S. aureus in various studies and was found to be between 15 and 57%.2223 In our study slime factor was found to be present in 10 of 20 pathognomic CNS. There was no significant difference between the distribution of infection types and slime production of the isolates. This finding can be explained either by the ability of low virulence microorganisms to produce severe infections in immunosuppressed patients or by the fact that slime production in vivo is not always associated with increased virulence.24

We can state that, as in many other centers, there is an increase in Gram-positive microorganisms and an increase in multidrug resistance in our BMT center. An increase in MRSA and Enterobacter spp. is in part due to prophylactic use of ciprofloxacin. Thus, vancomycin should be the preferred agent for Gram-positive microorganisms. Surveillance studies may be helpful in determining the cause of the infections and may help to develop strategies against certain agents. There was no difference between slime factor positive and negative Coagulase-negative staphylococci isolated.


  1. 1

    Kolbe K, Domkin D, Derigs HG et al. Infectious complications during neutropenia subsequent to peripheral blood stem cell transplantation Bone Marrow Transplant 1997 19: 143–147

  2. 2

    Kirk Jl, Greenfield Ra, Slease B, Ebstein RB . Analysis of early infectious complications after autologous bone marrow transplantation Cancer 1988 62: 2445–2450

  3. 3

    Baron EJ, Peterson LR, Finegold SM . Diagnostic Microbiology. Bailey and Scott's 9th edn CV Mosby Co: St Louis 1994 pp 321–405

  4. 4

    Murray PR, Baron EJ, Pfaller MA et al. Manual of Clinical Microbiology 6th edn ASM Press: Washington DC 1995

  5. 5

    Munro S . Disk diffusion susceptibility testing In: Isenberg HD, Hindler J (eds). Clinical Microbiology Procedures Handbook, Vol. 1 ASM Press: Washington DC 1992

  6. 6

    NCCLS Performance Standards for Antimicrobial Disk Susceptibility Tests Vol. 17, document M2-A6, 6th edn Approved Standard Documents: Pennsylvania 1997

  7. 7

    Freeman DJ, Falkiner FR, Reene CT . New method for detecting slime production by coagulase-negative staphylococci J Clin Pathol 1989 42: 872–874

  8. 8

    Garner JS, Jarvis WR, Emori TG et al. CDC definitions for nosocomial infections Am J Infect Cont 1988 16: 128–140

  9. 9

    Üstün C, Arslan Ö, Beksaç M et al. A retrospective comparison of allogeneic peripheral blood stem cell and bone marrow transplantation results from a single center, a focus on the incidence of GVHD and relapse Biol Blood Marrow Transplant 1999 5: 28–35

  10. 10

    Bensinger WI, Weaver CH, Appelbaum FR et al. Transplantation of allogeneic peripheral blood stem cells mobilized by recombinant human granulocyte colony-stimulating factor Blood 1995 85: 1655–1668

  11. 11

    Talmadge JE, Reed E, Ino K et al. Rapid immunologic reconstitution following transplantation with mobilized peripheral blood stem cells as compared to bone marrow Bone Marrow Transplant 1997 19: 161–172

  12. 12

    Akan H, Koç H, Arslan Ö et al. Febrile neutropenia in a bone marrow transplantation unit Int J Antimicrob Agents 1997 16: 127–130

  13. 13

    Çelebi H, Akan H, Aķçağlayan E et al. Febrile neutropenia in allogeneic and autologous peripheral blood stem cell transplantation and conventional chemotherapy for malignancies Bone Marrow Transplant 2000 26: 211–214

  14. 14

    Sable CA, Donowitz GR . Infections in bone marrow transplant recipients Clin Infect Dis 1994 18: 273–281

  15. 15

    Klastersky J, Zinner SH, Calandra T et al. Empirical antimicrobial therapy for febrile granulocytopenic cancer patients: lessons from four EORTC trials Eur J Cancer Clin Oncol 1988 24: S35–S45

  16. 16

    Del Favero A, Bucaneve G, Menichetti F . Empiric monotherapy in neutropenia: a realistic goal? Scand J Infect Dis Suppl 1995 96: 34–37

  17. 17

    Akova M . Kanser hastalarnda bakteriyel infeksiyonlara yaklasm ve empirik antimikrobiyal tedavi (approach to the bacterial infections in cancer patients and empirical treatment) Flora 1998 3: (Suppl. 1) 3–13

  18. 18

    Wingard JR . Importance of Candida species other than C. albicans as pathogens in oncology patients Clin Infect Dis 1995 20: 115–125

  19. 19

    Goodrich JM, Reed EC, Mori M et al. Clinical features and analysis of risk factors for invasive Candidal infection after marrow transplantation J Infect Dis 1991 164: 731–740

  20. 20

    Verbist L . for the International Study Group Epidemiology and sensitivity of 8625 ICU and hematology/oncology bacterial isolates in Europe Scand J Infect Dis Suppl 1993 91: 14–24

  21. 21

    Gür D, Ünal S ve çalyşma grubu . Yoğun bakym ünitelerinde izole edilen Gram negatif bakterilerin çeşitli antibiyotiklere invitro duyarliliklari (the in vitro susceptibility of Gram-negative bacteria to antibiotics isolated in ICU) Flora 1996 3: 153–159

  22. 22

    Votava M, Woznicova V . Production of slime by staphylococcal isolates from blood cultures Cent Eur J Public Health 2000 8: 18–20

  23. 23

    Arciola CR, Montanaro L, Baldassarri L et al. Slime production by Staphylococci isolated from prosthesis-associated infections New Microbiol 1999 22: 337–341

  24. 24

    Perdreau-Remington F, Sande MA, Peters G, Chambers HF . The abilities of a Staphylococcus epidermidis wild-type strain and its slime-negative mutant to induce endocarditis in rabbits are comparable Infect Immun 1998 66: 2778–2781

Download references

Author information

Correspondence to H Akan.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Aksu, G., Ruhi, M., Akan, H. et al. Aerobic bacterial and fungal infections in peripheral blood stem cell transplants. Bone Marrow Transplant 27, 201–205 (2001) doi:10.1038/sj.bmt.1702739

Download citation


  • infections
  • peripheral blood stem cell
  • transplantation
  • bacterial
  • fungal

Further reading