sign in sign up
<<
Home , Neutropenia

Transplantation, (1997) 19, 143–147  1997 Stockton Press All rights reserved 0268–3369/97 $12.00

Infectious complications during neutropenia subsequent to peripheral stem cell transplantation

K Kolbe1, D Domkin1, HG Derigs1, S Bhakdi2, C Huber1 and WE Aulitzky1,3

1Division of , IIIrd Department of and 2Department of Microbiology, Johannes Gutenberg University Hospital, Germany

Summary: diagnostic procedures further decrease the natural resist- ance to . Finally, granulocytopenia compromises Type, severity and incidence of infection during the neu- the first line of defense against bacterial, fungal and para- tropenic period after peripheral blood stem cell trans- sitic pathogens. of unknown origin, bacteremia and plantation (PBSCT) for treatment of malignant disease pneumonia are the most frequent clinical manifestations of were studied in 66 patients treated at a single insti- infection in patients after .3 with tution. Data of 34 female and 32 male patients with a both gram-positive and gram-negative as well as median age of 43 years suffering from leukemia (12), fungi are common in these patients. In particular, staphylo- lymphoma (35), multiple myeloma (six) or solid tumors cocci, streptococci, E.coli, klebsiella, pseudomonas, can- (13) were retrospectively analyzed. All patients had dida and aspergillus species are frequently isolated from .received at least 2.5 ؋ 106 CD34-positive cells for stem infectious sites in granulocytopenic patients cell rescue after high-dose chemotherapy. Ninety-four Chemotherapeutic agents display a dose–response percent of the patients experienced at least one febrile relationship in vivo and in vitro.4,5 However, the application episode during their post-transplant course. The of high-dose chemotherapy is limited by its toxic effects, patients recovered quickly and defervesced after a in particular on the hematopoietic system. Stem cell rescue median of 4 days. The incidence of bacteremia was 39% by autologous bone marrow transplantation has been and gram-positive cocci were the predominant patho- applied to circumvent the dose-limiting hematopoietic tox- gens. In contrast, severe organ infections were rare. icities.6–8 Myeloablative with autologous bone Only 5% of the patients suffered from lung infiltrates. marrow transplantation, however, cause prolonged marrow No invasive fungal infections were observed. No trans- aplasia accompanied by severe infectious complications. plant-related deaths occurred in the 66 patients studied. Autologous bone marrow transplantation caused a trans- We conclude that the severe, but shortlasting neutro- plant-related mortality consistently in the range of penia after peripheral blood stem cell transplantation is 5–10%.6–8 Mobilization of peripheral hematopoietic stem associated with a high incidence of bacterial infection. cells by chemotherapy and growth factors and leukapher- The severity of the majority of these infections is moder- esis yields higher numbers of hematopoietic stem cells than ate. With appropriate anti-infective therapies these bone marrow harvesting.9,10 Numerous reports have mean- infections can be managed and life-threatening infec- while confirmed the feasibility and effectiveness of trans- tious complications, in particular fungal infections, are plantation with peripheral blood stem cells (PBSCT) to rare. Empirical anti-infective regimens specifically mitigate the hematotoxic effects of myeloablative ther- designed for this clinical situation should be explored. apies.11–16 Most of these studies demonstrated a shortening Keywords: infection; neutropenia; autologous peripheral of the time to leukocyte and thrombocyte recovery when blood stem cell transplantation compared to autologous bone marrow transplantation or . Very few reports, however, have focused on the type and severity of infectious complications observed in severely leukopenic patients after high-dose Infections are the leading cause of morbidity and mortality therapy and peripheral stem cell transplantation.17,18 This 1,2 in patients treated with aggressive chemotherapy. A var- study aims to analyze frequency, type and severity of infec- iety of factors contribute to the pathogenesis in these tions during the aplastic phase after peripheral stem cell patients. Direct damage of the skin and mucous membrane transplantation in a cohort of unselected patients treated at barrier facilitates the entry of pathogens. In addition, iatro- a single institution with PBSCT. genic disruption of the skin by central venous catheters or

Patients, material and methods Correspondence: K Kolbe, Division of Hematology, IIIrd Department of Medicine, Johannes Gutenberg University Hospital, Langenbeckstr. 1, All patients treated at the of the Johannes 55101 Mainz, Germany 3Current address: Division of Hematology//, Robert Gutenberg University in the years 1993 to 1995 were retro- Bosch Hospital, Auerbachtra␤e 110, 70376 Stuttgart, Germany spectively analyzed. Sixty-six patients underwent high-dose Received 7 May 1996; accepted 18 September 1996 chemotherapy and autologous transplantation of peripheral Infectious complications during PBSCT K Kolbe et al 144 blood stem cells during this period. The clinical character- at least once daily in case of elevated body temperature istics of the study patients are summarized in Table 1. Ͼ38.0°C. The diagnosis of a documented bacterial infec- Thirty-four female and 32 male patients suffered from tion was based on the occurrence of at least one positive myelogenous leukemia (AML), chronic myelogenous leu- culture in the face of appropriate clinical findings. All kemia (CML), non-Hodgkin’s lymphoma, Hodgkin’s dis- medication was applied through central venous catheters ease, acute lymphoblastic leukemia (ALL), breast , throughout the in-patient period. transfusions were mutliple myeloma and testicular cancer. Only patients with administered in the case of less than an age of less than 60 years and preserved function of all 10/nl in asymptomatic patients, Ͻ20/nl in febrile patients vital organs were treated with high-dose chemotherapy or in the case of hemorrhagic diathesis being present. and PBSCT. Leukocyte-depleted erythrocyte concentrates were given to All patients had received chemotherapy for mobilization patients with a hemoglobin concentration Ͻ8 g/dl. of stem cells. Fifty-six patients received G-CSF at a dose In febrile patients a thorough physical examination and of 5–10 ␮g/kg, nine patients received sequential treatment chest X-ray were performed. All patients received i.v. anti- with interleukin-3 5 ␮g/kg plus G-CSF 10 ␮g/kg for mobil- biotics after one measurement of an elevated body tempera- ization of stem cells. Peripheral blood stem cells were ture Ͼ38.3°C or two times body temperature Ͼ38.0°C. obtained by repeated leukaphereses. Leukaphereses were Empiric therapy was performed according to pre- started after a rise of leukocytes to above 0.8/nl and more viously published guidelines.19 In 58 patients, antimicrobial than 0.5% CD34-positive cells in the peripheral blood and treatment was started with a triple combination with ␤- were repeated until a minimum number of 2.5 ϫ 106 CD34- lactam, aminoglycoside and glycopeptide. One patient positive cells/kg was harvested. Different conditioning regi- received a ␤-lactam with aminoglycoside, three patients mens were applied according to diagnosis and disease were started with ␤-lactam monotherapy. Antibiotic treat- status (Table 1). Peripheral stem cells were infused intra- ment was continued until disappearance of all clinical venously 1–3 days after discontinuation of chemotherapy. and/or microbiological evidence for infection and granulo- During PBSCT procedure patients were kept in single cyte counts above 0.5/nl. In case of persistent fever, i.v. rooms under reverse isolation. Rooms were not equipped amphotericin B was added 2–5 days after start of antibiotic with high-efficiency air filters. All patients received anti- treatment. Modifications of the antibiotic regimen were per- biotic prophylaxis with ofloxacin 2 ϫ 200 mg per day and formed, if appropriate, according to microbiologic results. antimycotic prophylaxis with fluconazole 2 ϫ 200 mg per Thirty-seven patients received 5 ␮g G-CSF s.c. beginning day. Appropriate clinical examinations were performed day 2 after peripheral stem cell transplantation, whereas the once daily and body temperature was measured at least remaining patients were not treated with recombinant three times daily. Routine cultures from stool, urine and growth factors during the post-transplant period. No differ- mouth swabs were examined twice weekly for contaminat- ences regarding leukocyte recovery, thrombocyte recovery ing pathogens. In addition, two blood cultures were drawn and incidence of infection were observed between these two patient groups. All data were analyzed with descriptive statistical Table 1 Clinical characteristics of PBSCT patients methods. Statistical significance of differences between diagnostic groups was calculated using ␹2 test. Probability No. 66 Age of events (onset of fever, leukocyte recovery, response to Median 43 (range 18–59) ) were analyzed using Kaplan–Meier product limit estimates and groups were compared applying the log Sex Female 34 (51%) rank test. Male 32 (49%) Diagnosis AML 8 Results ALL 2 NHL 31 Breast cancer 12 Hematopoietic reconstition and support with blood CML 2 products in patients undergoing high-dose chemotherapy Multiple myeloma 6 and peripheral blood stem cell transplantation Hodgkin’s disease 4 Testicular cancer 1 All patients treated became severely neutropenic with a Conditioning regimen median nadir of 0.1/nl (Table 2). The duration of neutro- TBI + 14 penia was moderate ranging from 4–21 days (median 10 Busulfan + cyclophosphamide 8 days) with a leukocyte count less than 1/nl. All patients BCNU-VP16-Cy-Ara-C 21 recovered from neutropenia within 3 weeks. No major dif- Cy-Mtx-thiotepa 10 Cy-Mtx- 2 ferences in duration of were observed between BEAM 3 patient groups with different underlying diseases. In Melphalan 7 addition, no clear correlation between numbers of infused VP-16-carboplatin 1 CD34-positive cells and duration of neutropenia was Stem cell support observed (data not shown). 106 CD34-positive cells/kg 5.5 (2.4–55.3) Patients also became severely thrombopenic and 64 out 66 patients required at least one platelet transfusion during Infectious complications during PBSCT K Kolbe et al 145 Table 2 Hematopoietic recovery and supportive care of PBSCT and Strep. sanguis in one) were isolated from blood cul- patients tures. In two patients gram-negative bacteria (E. coli, Acinetobacter) were detected in blood cultures. No. 66 Severe organ infections were rarely observed in patients Leukocyte nadir Median (range)/nl 0.1 (0.1–0.5) treated with high-dose chemotherapy. Lung infiltrates were Days leukocytes Ͻ0.5/nl recorded in three out of 66 patients (4.5%). In two of these Median (range) 8 (3–16) patients these abnormalities were transient and the patients Days leukocytes Ͻ1.0/nl recovered quickly. In one patient interstitial infiltrates were Median (range) 10 (4–21) observed. These radiographic changes disappeared upon Days thrombocytes Ͻ50/nl Median (range) 11 (2–26) hematopoietic reconstitution and empiric treatment with Red cell transfusion (units) erythromycin. No pathogens were identified in the patients Median (range) 4 (0–10) with lung infiltrates. Bronchoalveolar lavage was perfor- Platelet transfusion (units) med in two patients but failed to provide a positive Median (range) 3 (0–11) Days on antibiotics microbiological result. One patients suffered from a herpes Median (range) 10 (0–20) zoster infection which resolved after institution of acyclovir Days on Ampho B treatment. No deaths were observed during the first 3 Median (range) 0 (0–16) months after transplantation. In the majority of patients, 35 Days in patient care out of 66 (53%) patients the febrile episode was classified Median (range) 23 (15–52) as . The response to antimicrobial therapy was prompt in the majority of patients. Forty-three out of 62 febrile patients the post-chemotherapy period (Table 2). The thrombocyte defervesced within 5 days after the start of empiric anti- counts recovered with a similar kinetics as leukocyte biotic treatment. Accordingly, only 18 out of 62 patients counts. In 90% of the patients the thrombocyte counts reco- required empirical treatment with intravenous amphotericin vered to Ͼ50/nl within 20 days. The remaining patients all B. Sytemic fungal infections were not observed in these recovered to safe thrombocyte levels within 26 days. Sixty- patients. Neither proven nor probable fungal infections two out of 66 patients required substitution of erythrocyte were diagnosed in the 66 patients. concentrates. Only minor differences in type, frequency and duration of infection were observed in patients with different under- lying diseases (Table 4). Most of the patients became Type and incidence of infection in patients treated with febrile during the neutropenic episode. The onset of fever high-dose chemotherapy and peripheral blood stem cell was rapid in all disease groups within a few days after the transplantation onset of leukopenia. The pattern of clinically documented Sixty-two out of 66 (94%) of patients experienced at least infection was also not different. Pneumonia was rare in all one febrile episode during the neutropenic phase sub- patients irrespective of the diagnosis. The incidence of bac- sequent to peripheral stem cell transplantation (Table 3). teremia was slightly decreased in patients with malignant A median time to defervescence of 4 days was observed. lymphoma compared to the other patient groups. This dif- Microbiologically and/or clinically documented infection ference, however, did not reach statistical significance. occurred in 32 patients (48%). Bacteremia was the pre- Similarly, the response of all patient groups as measured dominant type of infection in this patient group and was by the median time to defervescence to antimicrobial ther- documented in 26 out of 66 patients (39%). Gram-positive apy was similar. pathogens, mostly coagulase-negative staphylococci, were the most frequent cause of bacteremia and were recovered from blood culture in 23 patients (34%). In five patients Discussion streptoccoci (Viridance strep. in two, Strep. mitis in two, Infections were observed at a very high frequency in patients treated with intensive chemotherapy and peripheral Table 3 Features of clinically and microbiologically documented infection blood stem cell transplantation. More than 90% of the patients experienced at least one infectious espisode during No. (%) chemotherapy-induced neutropenia. However, when appro- priate supportive care measures were applied the severity Fever 62 (94) of these infections was moderate. Bacteremia predomi- Febrile days nantly with low pathogenic gram-positive bacteria was the Median (range) 4 (1–18) most frequent infectious syndrome in these patients. No Pneumonia 3 (5) Bacteremia 26 (39) septic shock was observed and severe organ infections Coagulase negative-staphylococci 18 including pneumonia were rare. No infection-related death Streptococci 5 occurred in this cohort of 66 patients. Gram-negative 2 This pattern of infection was different from that observed Other 4 in other patients with chemotherapy-induced leukopenia. Fever of uknown origin 35 (53) Bacteremia was detected in 39% of the patients and was more common after PBSCT than in patients with aggressive Infectious complications during PBSCT K Kolbe et al 146 Table 4 Type, severity, duration and treatment of infection in PBSCT patients according to underlying disease

Leukemia Lymphoma Multiple myeloma Solid tumor

No. 12 35 6 13 Febrile days 2.5 (0–9) 4 (0–18) 5.5 (0–16) 4 (2–14) Pneumonia 1 (8%) 2 (6%) 0 0 Bacteremia 6 (50%) 12 (34%) 3 (50%) 5 (38%) FUO 4 (30%) 20 (57%) 2 (30%) 6 (46%) No infection 1 (8%) 2 (6%) 1 (16%) 0 Days i.v. amphotericin B Median (range) 10 (0–20) 9 (4–15) 8 (0–10) 11 (6–19) Days amphotericin B Median (range) 0 (0–16) 0 (0–6) 0 (0–5) 0 (0–5)

chemotherapy for treatment of solid tumors or acute leuke- marrow transplant recipients.15,16 It can be concluded that mias. Large studies in those patient groups revealed an inci- appropriate antimicrobial therapy protects the PBSCT dence of bacteremia in the range of 10–30%.3,20–22 Simi- patients from development of severe organ infection for a larly, bacteremia was observed in 84 out of 396 limited time period irrespective of the severity of the pre- chemotherapy cycles (21%) in ceding neutropenia. This analysis of incidence and severity patients treated at our institution during the same time per- of infection in an unselected series of patients treated with iod and according to the same guidelines as the study high-dose chemotherapy and PBSCT further confirms that patients (unpublished data). The pathogens responsible for peripheral blood stem cells are an effective measure for bacteremia in PBSCT patients were gram-positive bacteria prevention of severe infections associated with prolonged in 88% of cases, whereas highly pathogenic gram-negative aplasia. However, the empirical anti-infective therapy bacteria were rarely cultured. Several mechanisms might should be modified according to the specificities of the neu- explain this high incidence of gram-positive bacteremia tropenic period after PCSCT. In particular, start of empiri- after PBSCT. Central venous lines have been reported to cal treatment with a combination including a glycopeptide be the most common site of entry for gram-positive patho- antibiotic or delay of the start of empirical amphotericin B gens.23 However, intravenous devices are part of the stan- treatment seems rational in view of the epidemiology of dard care both in acute leukemia and PBSCT patients. Thus infection in PBSCT patients. Such interventions should be it seems unlikely that use of the same type of catheter dur- further explored in randomized clinical trials. ing a neutropenic episode of comparable duration causes highly different rates of gram-positive infection. High-dose chemotherapy, however, almost invariably leads to severe Acknowledgements mucositis during the neutropenic phase. The gram-positive bacteria identified in blood cultures in the PBSCT patients This study was supported by a research grant from ‘Deutche Kreb- are all saprophytes harbored in the oral cavity in healthy shilfe’ (project number 10368). individuals.24 Thus, severe mucosal damage might be at least in part responsible for the high frequency of gram- positive bacteremia in PBSCT patients. However, the present data do not prove this hypothesis. References In contrast to the high incidence of gram-positive bacter- emia, severe clinically and/or microbiologically docu- 1 Pizzo PA, Meyers JD, Freifeld AG, Walsh T. Infections in mented infections are rare after PBSCT. In particular, pneu- cancer patients. In: DeVita VT, Hellman S, Rosenberg SA monia occurred at a low frequency and was diagnosed in (eds). Cancer. JB Lippincott: Philadelphia, 1993, pp 2292– 2336. less than 5% of the PBSCT patients. In contrast, Masch- 25 2 Pizzo PA. Management of fever in patients with cancer and meyer et al reported an incidence of pneumonia in the treatment-induced neutropenia. New Engl J Med 1993; 328: range of 20% in a large multicenter study on patients with 1323–1332. chemotherapy-induced neutropenia. Other authors recorded 3 Wade JC, Schimpf SC. Epidemiology and prevention of infec- a comparable frequency of lower respiratory tract infection. tion in the compromised host. In: Rubin RH, Young LS (eds). Similary, pneumonia was observed in 93 out of 396 (23%) Clinical Approach to Infection in the Compromised Host. acute leukemia patients treated in parallel at our institution Plenum Press: New York, London, 1988, pp 5–40. (unpublished data). None of the PBSCT patients suffered 4 Hryniuk W, Bush H. The importance of dose intensity in from systemic fungal infection; this despite the fact that chemotherapy of metastatic breast cancer. J Clin Oncol 1984; patients were kept in single room reverse isolation without 2: 1281–1288. 5 Frei EI. Pharmacologic strategies for high dose chemotherapy. an effective air filter system. Prolonged bone marrow apla- In: Armitage JO, Antman KH (eds). High Dose Cancer sia has been implicated in particular in the pathogenesis Chemotherapy. Williams & Wilkins: Baltimore, 1992; pp 3– 25 of pulmonary infiltrates and fungal infection. Neutropenic 14. episodes in PBSCT patients are severe, but of shorter dur- 6 Biermann PJ, Armitage JO. Autologous bone marrow trans- ation than in acute leukemia patients and autologous bone plantation for non-Hodgkins lymphoma. In: Forman SJ, Infectious complications during PBSCT K Kolbe et al 147 Blume KG, Thomas ED (eds). Bone Marrow Transplantation. ments of autologous recovery phase peripheral blood stem cell Blackwell Scientific: Boston, 1994, pp 683–695. transplants, autologous bone marrow transplants and allo- 7 Yeager AM. Autologous bone marrow transplantation for geneic bone marrow transplants. Bone Marrow Transplant acute myeloid leukemia. In: Forman SJ, Blume KG, Thomas 1992; 9: 277–284. ED (eds). Bone Marrow Transplantation. Blackwell Scien- 17 Watanabe T, Takaue Y, Abe T et al. Early infectious compli- tific: Boston, 1994, pp 709–730. cations after peripheral blood stem cell autografts in children. 8 Philips GL. Transplantation for Hodgkin’s disease. In: Forman Med Pediatr Oncol 1991; 19: 100–105. SJ, Blume KG, Thomas ED (eds). Bone Marrow Transplan- 18 D’Antonio D, Iacone A, Pierelli L, Bonfini T. Patterns of tation. Blackwell Scientific: Boston, 1994, pp 696–708. recovery phase infection after autologous blood progenitor cell 9 Pettengell R, Woll PJ, Thatcher N et al. Multicyclic, dose- transplantation in patients with malignancies. The Gruppo Ital- intensive chemotherapy supported by sequential reinfusion of iano di Studio per la Manipolazione Cellulare in Ematologia. hematopoietic progenitors in whole blood. J Clin Oncol 1995; Eur J Clin Microbiol Infect Dis 1995; 14: 552–556. 13: 148–156. 19 Highes WT, Armstrong D, Bodey GP. Guidelines for the use 10 Nademanee A, Sniecinski I, Schmidt GM et al. High-dose of antimicrobal agents in neutropenic patients with unex- therapy followed by autologous peripheral-blood stem-cell plained fever. A statement by the Infectious Diseases Society transplantation for patients with Hodgkin’s disease and non- of America. J Infect Dis 1990; 161: 381–396. Hodgkin’s lymphoma using unprimed and col- 20 Link H, Maschmeyer G, Meyer P et al. Interventional anti- ony-stimulating factor-mobilized peripheral-blood stem cells. microbial therapy in febrile neutropenic patients. Study Group J Clin Oncol 1994; 12: 2176–2186. of the Paul Ehrlich Society for Chemotherapy. Ann Hematol 11 Liberti G, Pearce R, Taghipour G et al. Comparison of periph- 1994; 69: 231–243. eral blood stem-cell and autologous bone marrow transplan- 21 De Pauw BE, Deresinski SC, Feld R et al. Ceftazidime com- pared with and tobramycin for the empiric treat- tation for lymphoma patients: a case-controlled analysis of the ment of fever in neutropenic patients with cancer. A multicen- EBMT Registry data. Lymphoma Working Party of the ter randomized trial. The Intercontinental Antimicrobial Study EBMT. Ann Oncol 1994; 5 (Suppl. 2): 151–153. Group. Ann Intern Med 1994; 120: 834–844. 12 Siena S, Bregni M, Di Nicola M et al. Durability of hemato- 22 Freifeld AG, Walsh T, Marshal D et al. Monotherapy for fever poiesis following autografting with peripheral blood hemato- and neutropenia in cancer patients: a randomized comparison poietic progenitors. Ann Oncol 1994; 5: 935–941. of ceftazidime versus imipenem. J Clin Oncol 1995; 13: 13 Coiffier B, Philip T, Burnett AK, Symann ML. Consensus 165–176. conference on intensive chemotherapy plus hematopoietic 23 Uderzo C, D’Angelo P, Rizzari C et al. Central venous cath- stem-cell transplantation in malignancies: Lyon, France, 4–6 eter-related complications after bone marrow transplantation June, 1993 (see comments). J Clin Oncol 1994; 12: 226–231. in children with hematological malignancies. Bone Marrow 14 Lamy T, Drenou B, Grulois I et al. Improvement of hematol- Transplant 1992; 9: 113–117. ogic recovery after high-dose intensification using peripheral 24 Isenberg HD, D’Amato RF. Indigenous and pathogenic micro- blood progenitor cells (PBPC) mobilized by chemotherapy organisms of humans. In: Lennette EH, Balows A, Hausler and GM-CSF. Ann Hematol 1994; 69: 297–302. WJ, Shadomy HJ (eds). Manual of Clinical Microbiology. 15 Beyer J, Schwella N, Zingsem J et al. Hematopoietic rescue American Society for Microbiology: Washington, 1985, after high-dose chemotherapy using autologous peripheral- pp 24–36. blood progenitor cells or bone marrow: a randomized compari- 25 Maschmeyer G, Link H, Hiddemann W et al. Pulmonary infil- son. J Clin Oncol 1995; 13: 1328–1335. trations in febrile patients with neutropenia. Risk factors and 16 To LB, Roberts MM, Haylock DN et al. Comparison of outcome under empirical antimicrobial therapy in a ran- haematological recovery times and supportive care require- domized multicenter study. Cancer 1994; 73: 2296–2304.