Abstract
Advances made in the field of hematopoietic stem cell transplantations (HSCT) over the past 20 years may have had an impact on the distribution of posttransplantation infections. We sought to retrospectively analyze the epidemiology and risk factors for bacterial, fungal, and viral infections in children after allogeneic HSCT in a cohort of 759 children who underwent allogeneic HSCT in a single institution between 1990 and 2009. The association between infections and risk factors of interest at 0 to 30 days, 31 to 100 days, and 101 days to 2 years posttransplantation was evaluated using logistic regression. Difference among the subtypes within each category was studied. There were 243 matched-related donors, 239 matched-unrelated donors (MUDs), and 176 haploidentical donor transplantations. Era of transplantation (0–30 days), peripheral blood stem cell product, acute graft-versus-host disease (aGVHD; 31–100 days), and chronic GVHD (cGVHD; 101–730 days) were associated with higher risk for bacterial infections at the respective time periods. Patients with aGVHD (31–100 days), cGVHD, and older age (101–730 days) were at higher risk for fungal infections. Cytomegalovirus (CMV) donor/recipient (D/R) serostatus (0–100 days), era of transplantation, MUD HSCT (31–100 days), and cGVHD (101–730 days), influenced viral infections. Gram-positive outnumbered gramnegative bacterial infections; aspergillosis and candidemia were equally prevalent in all time periods. Haploidentical donor HSCT was not associated with an increased risk of infections. There seems to be a continuum in the timeline of infections posttransplantation, with bacterial, fungal, and viral infections prevalent in all time periods, particularly late after the transplantation, the risk affected by GVHD, CMV, D/R status, product type, older age, and use of unrelated donors.
Keywords: Infections, Children, Allogeneic, Stem cell transplant
INTRODUCTION
Allogeneic hematopoietic stem cell transplantation (HSCT) is an established treatment modality for patients with certain hematologic malignancies and for many congenital and acquired disorders of the hematopoietic system. Infections contribute significantly to the morbidity and mortality from this procedure. The manner in which the transplantation is performed has changed substantially over the past 2 decades. This may have had an impact on the type of pathogens and timeline of infections posttransplantation.
Data predominantly from the adult literature show that gram-positive (GP) and gram-negative (GN) bacteria [1], Candida and Aspergillus spp 2], are major pathogens in the pre-engraftment phase. With neutrophil recovery, bacterial infections decrease, and infections by cytomegalovirus (CMV) [3] and fungi predominate, particularly in association with graft-versus-host disease (GVHD) [4]. The risk of invasive fungal infections in the late recovery phase increases with age, CMV infection, GVHD, and in recipients of matched-unrelated donor (MUD) transplantation [5, 6]. With the use of pre-emptive therapy, disease due to CMV [3], and herpes simplex virus (HSV) [7] has shifted to the late recovery phase. Haploidentical donor transplantations have been reported to influence the risk of fungal [8] and adenoviral (ADV) [9] infections in this phase. Total-body irradiation (TBI) has been associated with an increased risk of symptomatic parainfluenza virus (PIV) [10] and ADV infection [11].
Thus, there is a complex interplay of factors including patient demographics, neutropenia, GVHD, transplantation modality, TBI, product type, CMV donor/recipient (D/R) status, and era of transplantation, contributing to infection risk. The effect of these factors on risk of bacterial, fungal, and viral infection, at 0 to 30 days, 31 to 100 days, and 101 days to 2 years posttransplantation, after adjusting for confounding variables, has not been well studied both in children and adults. This is the largest retrospective study providing a timeline of infections in young HSCT recipients.
PATIENTS AND METHODS
This retrospective cohort consisted of 786 patients who underwent a first allogeneic HSCT over a 20-year period (January 1990 through December 2009) at St. Jude Children’s Research Hospital (SJCRH). Patients who were older than 21 years of age at transplantation (19 patients) and those who underwent a cord blood transplantation (8 patients) were excluded. A total of 759 patients were included in this analysis. The study was approved by the SJCRH Institutional Review Board. Patients were followed for 10 years after transplantation or until age 18 years whichever was later. Follow-up was frequent in the first 2 years after transplantation and yearly thereafter. The mean duration of follow-up for surviving patients was 7.96 years (range, 0.21–19.91 years).
Microbiologic Methods
Microbiologic records were reviewed to identify patients with documented bacterial, fungal, and viral infections as diagnosed by culture (bacteria, fungi, viruses), direct fluorescent antibody testing, and PCR (viruses). Infection was defined as isolation or detection of an organism that was associated with symptoms or disease and included fungal and viral pathogens detected on pre-emptive screening. Colonization detected in surveillance cultures and positive blood cultures due to contamination were not considered as infections. Infections with the same organism occurring more than 14 days after the last negative culture or diagnostic test were recorded as 2 separate infectious episodes. Only the first episode of bacteremia and the first episode of GP, GN, yeast, mold, and specific viral infection in a given patient during the specified time period was included in the analyses. The day of onset of infection was defined as the day when the first positive diagnostic sample was collected. Infection as the primary cause of death was used for analyses. In patients who relapsed from their underlying disease, or those with GVHD who died from infection, infection was not noted as the primary cause of death.
Peripheral blood samples were collected once weekly starting in February 2000 to prospectively screen for CMV and starting in February and July 2002 to prospectively screen for Epstein-Barr virus (EBV) and ADV by quantitative real-time PCR, as previously described [12], using an ABI PRISM 7900HT Sequence Detection system (Applied Biosystems, Foster City, CA). Before 2000, CMV and ADV were detected by viral blood culture and EBV using endpoint PCR with detection by Southern blot. Weekly screening on whole blood for galactomannan was performed from August 2003. Preemptive treatment with ganciclovir, rituximab, and cidofovir was given based on the results of surveillance testing. Invasive fungal infection was defined according to accepted criteria [13]. Patients who met the definition of possible invasive fungal infections were not included.
Infection Prophylaxis
All patients received prophylaxis against Pneumocystis jirovecii for up to 1 year posttransplantation. Between 1993 and 2000, patients who were seropositive for CMV or had a seropositive donor received ganciclovir until day +120. After 2000, patients at risk for CMV or HSV reactivation received acyclovir prophylaxis until 1 year posttransplantation. Patients received antifungal prophylaxis with amphotericin B or lipid-based amphotericin formulations between 1990 and 2003. Subsequently, prophylaxis was with echinocandins until engraftment and with voriconazole thereafter. Antibacterial prophylaxis with fluoroquinolones was not given. Patients with chronic GVHD (cGVHD) received PenVK, cotrimoxazole, acyclovir, and voriconazole prophylaxis.
Era of Transplantation
The populationwas divided into 4 eras. There were 111, 218, 219, and 211 patients who underwent HSCT in the eras 1990 to 1994, 1995 to 1999, 2000 to 2004, and 2005 to 2009, respectively. The years 2000 onward represented a boundary for introduction of peripheral blood stem cell (PBSC) products, viral monitoring by quantitative real-time PCR, testing for galactomannan, and use of haploidentical donors.
Transplantation Methods
Transplantation-related variables were abstracted from a prospectively collected database that included patient demographics, underlying diagnosis, remission status, donor and product type, CMVD/R status, conditioning regimen, GVHD prophylaxis, and grading if present. The conditioning and GVHD prophylaxis have been previously described [14, 15]. A regimen based on TBI and cyclophosphamide was used for 326 of the 427 patients (76%) with acute leukemia. Cyclosporine with methotrexate or mycophenolate mofetil was predominantly used for GVHD prophylaxis. For haploidentical donor transplantations, 49 patients (28%) received TBI-based conditioning; the remaining 127 patients (72%) received fludarabine, thiotepa, and melphalanbased conditioning. Ex vivo T cell depletion of the graft was performed by anti-T cell Abs and complement in 17 patients (10%) or immune-magnetic selection using the Miltenyi CliniMACS system (Miltenyi Biotec GmbH, Teterow, Germany) in 159 patients (90%), respectively. Patients received GVHD prophylaxis with either a calcineurin inhibitor or mycophenolate mofetil. The day of engraftmentwas defined as the first of 3 consecutive days of achieving an absolute neutrophil count >500 cells/µL. Assessment of acute GVHD (aGVHD) was based on consensus criteria [16]. The aGVHD grade 3 and 4 were classified as severe. Immunization was according to standard recommendations [17] starting 1 year posttransplantation.
Statistical Analysis
The association amonginfections (bacterial, fungal, and viral) and the risk factors of interest was first evaluated using univariate logistic regression. The risk factors that were evaluated included age at transplantation, era of transplantation, donor type (haplo vs non-haplo identical), product type (PBSC vs marrow), T cell depletion, aGVHD (≥grade 3 vs others), cGVHD, TBI, myeloablative vs reduced-intensity conditioning (RIC), CMV D/R status (+/+ vs +/− vs −/+ vs −/−), and MUD vs matched-related donor (MRD) transplantation. To identify the exact sources of differences among eras and CMV status, Bonferroni correction was used to adjust for multiple comparisons, and P values were noted to be significant at level α = 0.05/6 = 0.008. The analyses for each type of infection were conducted independently. All the factors that were significant at level α = 0.15 in the univariate analyses were included in the multiple logistic regression models. The final model reports the results of all factors that remained significant at the 5% level.
The association between cumulative incidence of infectious deaths and the risk factors described above were assessed using Fine and Gray’s approach [18] within the framework of the Cox proportional hazards model. Deaths due to any other cause were treated as the competing events, and research participants alive at the last follow-up were considered as censored events.
The cumulative incidence of bacteremia within 30 days post-transplantation was defined as the time from the date of transplantation to the first episode of bacteremia within 30 days, with death within 30 days treated as a competing event. Patients who survived 30 days posttransplantation without bacteremia were treated as censored. Cumulative incidence of bacteremia within 30 days among different eras was estimated as described by Kalbfleisch and Prentice [19] and compared using Gray’s test [20].
Chi-square tests were used to test whether there was a significant difference among the proportions of different subtypes within each category of interest (ie, GP vs GN; yeast vs mold; Candidemia vs Aspergillus; HSV vs CMV vs EBV vs ADV; PIV vs influenza vs respiratory syncytial virus (RSV) within each time period. SAS version 9.2 (SAS Institute, Cary, NC) and StatXact (Cytel Corporation, Cambridge, MA) Windows version 8 were used for statistical analyses.
RESULTS
Patient Characteristics
A total of 759 patients underwent an allogeneic HSCT between1990 to 2009 at SJCRH. Demographics and patient characteristics are presented in Table 1.
Table 1.
Demographics and Characteristics of the Patients Who Underwent an Allogeneic HSCT at SJCRH Between 1990 and 2009
| Characteristic | 1990–2009 n = 759 (%) |
|---|---|
| Mean age (years) | 9.4 |
| 0–<2 | 117 (15) |
| 2–<10 | 286 (38) |
| 10–<18 | 296 (39) |
| 18–21 | 60 (8) |
| Men | 450 (59) |
| Race | |
| White | 543 (72) |
| African American | 122 (16) |
| Other | 94 (12) |
| Diagnosis | |
| ALL | 203 (27) |
| AML | 224 (30) |
| CML | 59 (8) |
| JMML | 16 (2) |
| Lymphoma | 24 (3) |
| MDS | 48 (6) |
| Solid tumors | 27 (4) |
| Hematologic disorders | 97 (12) |
| Immunologic disorders | 30 (4) |
| Metabolic disorders | 31 (4) |
| Donor type | |
| Matched related | 243 (32) |
| Mismatched related | 30 (4) |
| Matched unrelated | 239 (32) |
| Mismatched unrelated | 71 (9) |
| Haploidentical | 176 (23) |
| Product type | |
| HPC-M | 576 (76) |
| HPC-A | 183 (24) |
| CMV status | |
| D+/R+ | 212 (28) |
| D+/R− | 135 (18) |
| CMV D/R status | |
| D−/R+ | 107 (14) |
| D−/R− | 200 (26) |
| Not available | 105 (14) |
| Conditioning | |
| TBI-based | 497 (65) |
| Non-TBI-based | 262 (35) |
| Myeloablative | 597 (79) |
| RIC | 162 (21) |
| T cell depletion | |
| Yes | 365 (48) |
| No | 394 (52) |
| Time to engraftment | |
| Within 28 days | 662 (87) |
| After 28 days | 71 (9) |
| Did not engraft | 26 (4) |
| GVHD | |
| Acute | 369 (49) |
| None | 390 (51) |
| Grade I–II | 199 (26) |
| Grade III–IV | 170 (23) |
| Chronic | 129 (17) |
| None | 630 (83) |
| Limited | 79 (10) |
| Extensive | 50 (7) |
| Number of HSCT patients | |
| 1990–1994 | 111 (14) |
| 1995–1999 | 218 (29) |
| 2000–2004 | 219 (29) |
| 2005–2009 | 211 (28) |
HSCT indicates hematopoietic stem cell transplantation; SJCRH, St. Jude Children’s Research Hospital; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; CML, chronic myeloid leukemia; JMML, juvenile myelomonocytic leukemia; MDS, myelodysplastic syndrome; HPC-M/A, human progenitor cellsemarrow/apheresis; CMV D/R, cytomegalovirus donor/recipient; TBI, total body radiation; RIC, reduced-intensity conditioning; GVHD, graft-versus-host disease.
Data are number of patients (%), unless otherwise indicated.
Infections in the Posttransplantation Period
Infections were documented in 621 patients (82%). GP infections were more prevalent than GN infections in all 3 time periods posttransplantation (Table 2, Figure 1). There were 225 patients (30%) with bacteremia including 133 patients with GP and 92 patients with GN bacteremia. The median onset of bacteremiawas 87 days posttransplantation. Staphylococcus epidermidis 70 cases (31%) and viridians streptococci 16 cases (7%) were the most common GP organisms and Klebsiella spp 18 cases (8%), Escherichia coli 17 cases (8%), and Pseudomonas spp 14 cases (6%) were the most common GN organisms causing bacteremia. There were 110 patients (14%) with C difficile infection, and 13 patients (2%) with Enterococcus spp bacteremia in the entire cohort including 6 patients with E. faecalis bacteremia, which were vancomycin-susceptible, and 7 patients with E. faecium bacteremia, which were vancomycin-resistant.
Table 2.
Patients with Infection 0–30 Days, 31–100 Days, and 101 Days-2 Years Posttransplantation
| Infection | 0–30 Days Posttransplantation |
||||
|---|---|---|---|---|---|
| 1990–1994 |
1995–1999 |
2000–2004 |
2005–2009 |
1990–2009 |
|
| n = 111 | n = 218 | n = 219 | n = 211 | n = 759 | |
| Bacterial infections | |||||
| Bacteremia | 20 (18) | 12 (6) | 11 (5) | 16 (8) | 59 (8) |
| GP infections | 21 (19) | 23 (11) | 16 (7) | 13 (6) | 73 (10) |
| GN infections | 9 (8) | 5 (2) | 5 (2) | 6 (3) | 25 (3) |
| Fungal infections | |||||
| Yeast infections | 5 (5) | 10 (5) | 1 (1) | 4 (2) | 20 (3) |
| Mold infections | 3 (3) | 7 (3) | 9 (4) | 9 (4) | 28 (4) |
| Candidemia | 3 (3) | 5 (2) | 1 (<1) | 3 (1) | 12 (2) |
| Aspergillosis | 2 (2) | 6 (3) | 8 (4) | 3 (1) | 19 (3) |
| Viral infections | |||||
| HSV | 9 (8) | 31 (14) | 7 (3) | 5 (2) | 52 (7) |
| CMV | 1 (1) | 5 (2) | 17 (8) | 17 (8) | 40 (5) |
| EBV | 0 (0) | 0 (0) | 1 (<1) | 5 (2) | 6 (1) |
| ADV | 0 (0) | 0 (0) | 13 (6) | 9 (4) | 22 (3) |
| PIV | 0 (0) | 9 (4) | 4 (2) | 1 (<1) | 14 (2) |
| Influenza | 2 (2) | 1 (<1) | 1 (1) | 2 (1) | 6 (1) |
| RSV | 1 (1) | 1 (<1) | 2 (1) | 3 (1) | 7 (1) |
| Infection | 31–100 Days Posttransplantation |
||||
| 1990–1994 |
1995–1999 |
2000–2004 |
2005–2009 |
1990–2009 |
|
| n = 106 | n = 208 | n = 214 | n = 207 | n = 735 | |
| Bacterial infections | |||||
| Bacteremia | 5 (5) | 17 (8) | 21 (10) | 30 (14) | 73 (10) |
| GP infections | 19 (18) | 26 (12) | 31 (14) | 36 (17) | 112 (16) |
| GN infections | 3 (3) | 8 (4) | 9 (4) | 13 (6) | 33 (4) |
| Fungal infections | |||||
| Yeast infections | 5 (5) | 6 (3) | 1 (<1) | 6 (3) | 18 (2) |
| Mold infections | 0 (0) | 9 (4) | 4 (2) | 15 (7) | 28 (4) |
| Candidemia | 3 (3) | 1 (<1) | 0 (0) | 1 (<1) | 5 (1) |
| Aspergillosis | 0 (0) | 7 (3) | 2 (1) | 2 (1) | 11 (1) |
| Viral infections | |||||
| HSV | 4 (4) | 3 (1) | 2 (1) | 1 (<1) | 10 (1) |
| CMV | 4 (4) | 2 (1) | 29 (14) | 30 (14) | 65 (9) |
| EBV | 0 (0) | 0 (0) | 8 (4) | 15 (7) | 23 (3) |
| ADV | 1 (1) | 0 (0) | 10 (5) | 8 (4) | 19 (3) |
| PIV | 1 (1) | 4 (2) | 3 (1) | 0 (0) | 8 (1) |
| Influenza | 0 (0) | 0 (0) | 1 (1) | 0 (0) | 1 (<1) |
| RSV | 0 (0) | 5 (2) | 3 (1) | 2 (1) | 10 (1) |
| Infection | 101 Days to 2 Years Posttransplantation |
||||
| 1990–1994 |
1995–1999 |
2000–2004 |
2005–2009 |
1990–2009 |
|
| n = 79 | n = 175 | n = 179 | n = 189 | n = 622 | |
| Bacterial infections | |||||
| Bacteremia | 11 (14) | 21 (12) | 35 (20) | 26 (14) | 93 (15) |
| GP infections | 17 (22) | 21 (12) | 35 (20) | 48 (25) | 121 (19) |
| GN infections | 5 (6) | 18 (10) | 23 (13) | 16 (8) | 62 (10) |
| Fungal infections | |||||
| Yeast infections | 10 (13) | 12 (7) | 11 (6) | 6 (3) | 39 (6) |
| Mold infections | 2 (3) | 12 (7) | 6 (3) | 23 (12) | 43 (7) |
| Candidemia | 2 (3) | 0 (0) | 4 (2) | 3 (2) | 9 (1) |
| Aspergillosis | 1 (1) | 7 (4) | 4 (2) | 7 (4) | 19 (3) |
| Viral infections | |||||
| HSV | 4 (5) | 2 (1) | 5 (3) | 3 (2) | 14 (2) |
| CMV | 1 (1) | 1 (1) | 6 (3) | 6 (3) | 14 (2) |
| EBV | 0 (0) | 0 (0) | 5 (3) | 2 (1) | 7 (1) |
| ADV | 1 (1) | 1 (1) | 11 (6) | 7 (4) | 20 (3) |
| PIV | 3 (4) | 12 (7) | 10 (6) | 5 (3) | 30 (5) |
| Influenza | 5 (6) | 7 (4) | 7 (4) | 6 (3) | 25 (4) |
| RSV | 0 (0) | 2 (1) | 5 (3) | 8 (4) | 15 (2) |
GP indicates gram-positive; GN, gram-negative; HSV, herpes simplex virus; CMV, cytomegalovirus; EBV, Epstein–Barr virus; ADV, adenoviral; PIV, parainfluenza virus; RSV, respiratory syncytial virus.
Figure 1.
Percentage of patients with bacterial, fungal, and viral infections 0 to 30 days (A), 31 to 100 days (B), and 101 days to 2 years posttransplantation (C). A, There were 73 (10)/25 (3) patients with gram-positive (GP)/gram-negative (GN; P < .0001), 20 (3)/28 (4) patients with yeast/mold (M; P = .14), 12 (2)/19 (3) patients with candidemia (Cand)/proven aspergillus (Asperg; P = .21), 52 (7)/40 (5)/6 (1)/22 (3) patients with herpes simplex virus (HSV)/cytomegalovirus (CMV)/ Epstein-Barr virus (EBV)/adenovirus (ADV; P <.0001), and 14 (2)/6 (1)/7 (1) patients with parainfluenza (PIV)/influenza (Infl)/respiratory syncytial virus (RSV; P = .49) infection of a total of 759 patients. B, There were 112 (16)/33 (4) patients with GP/GN (P < .0001), 18 (3)/28 (4) patients with yeast/mold (P = .12), 5 (<1)/11 (1) patients with Cand/Asperg (P = .13), 10 (1)/65 (9)/23 (3)/19 (3) patients with HSV/CMV/EBV/ADV (P < .0001), and 8 (1)/1 (0.1)/10 (1) patients with PIV/Infl/RSV (P = .03) infection of a total of 735 patients. C, There were 121 (19)/62 (10) patients with GP/GN (P < .0001), 39 (6)/43 (7) patients with yeast/mold (P = .57), 9 (1)/ 19 (3) patients with Cand/Asperg (P = .08); 14 (2)/14 (2)/7 (1)/20 (3) patients with HSV/CMV/EBV/ADV (P = .12), and 30 (5)/25 (4)/15 (2) patients with PIV/Infl/RSV (P = .10) infection of a total of 622 patients. Percentages are indicated in parenthesis. Yeast infections include those in blood and other sites. Mold infections include Aspergillus, non-Aspergillus molds, and probable invasive fungal infections.
Yeasts and molds were equally prevalent in all 3 time periods posttransplantation (Table 2, Figure 1). Candidemia was detected in 26 patients (3%). Candida albicans and non albicans Candida spp were equally prevalent. C. parapsilosis and C. glabrata were the most common non-albicans Candida spp isolated. In 49 patients (6%) with proven aspergillosis, Aspergillus flavus and A. fumigatus were the most frequent isolates, in 17 and 10 patients, respectively. A probable diagnosis of invasive fungal infection was made in 40 patients (5%). Candidemia and aspergillosis were equally prevalent in all 3 time periods posttransplantation (Figure 1).
The distribution of HSV, CMV, EBV, and ADV at 0 to 30 days posttransplantation was significant (Figure 1A). HSV infections were more common than ADV and EBV infections (P < .0001). Infections by EBV were also less common than CMV (P <.0001) and ADV (P <.001). The distribution of HSV, CMV, EBV, and ADV at 31 to 100 days posttransplantation was also significant (Figure 1B). CMV infections were more common than HSV (P <.001), EBV, and ADV (P <.0001). The distribution of HSV, CMV, EBV, and ADV at 101 days to 2 years (Figure 1C) was not statistically significant. Respiratory tract infections by RSV were more common than influenza virus (P = .01) 31 to 100 days posttransplantation.
CMV D/R status (−/+ vs −/−; +/+ vs −/−) increased risk for CMV viremia (P = .0003; odds ratio [OR], 3.13; 95% confidence interval [CI], 1.16–8.45; P < .0001; OR, 5.25; 95% CI, 2.14–12.84), respectively.
Infections 0 to 30 Days after Transplantation
CMV D/R status was a risk factor for viral infections (Tables 3 and 4). None of the other variables increased the risk for bacterial or fungal infections on multiple logistic regression analyses.
Table 3.
Multiple Logistic Regression Analyses of Factors Increasing the Risk for Bacterial, Fungal, Viral Infections, and Bacteremia 0–30 Days, 31–100 Days, and 101 Days to 2 Years Posttransplantation
| 0 to 30 Days |
31 to 100 Days |
101 Days to 2 Years |
|---|---|---|
| Variable (P value; OR, 95% CI) | ||
| Bacterial infections | ||
| aGVHD (.002; 1.98; 1.3–3.0) | cGVHD (.0006; 2.05; 1.36–3.09) | |
| Era (.0006) | PBSC (.01; 2.57; 1.25–5.31) | |
| (.004; 2.44; 1.34–4.43)* | ||
| (.0003; 3.68; 1.82–7.44)† | ||
| (.0002; 4.12; 1.94–8.73)‡ | ||
| Bacteremia | ||
| aGVHD (.006; 2.12; 1.26–3.56) | cGVHD (.0005; 2.59; 1.53–4.37) | |
| Era (.002) | PBSC (.02; 2.84; 1.2–6.71) | CMV D/R status (.04) |
| (.001; 3.50; 1.63–7.52)* | ||
| (.0006; 4.14; 1.85–9.27)† | ||
| (.02; 2.37; 1.16–4.84)‡ | ||
| Fungal infections | ||
| - | aGVHD (.0008; 3.43; 1.7–6.9) | cGVHD (<.0001; 2.95; 1.75–4.97) |
| Older age (.03; 1.72; 1.04–2.87) | ||
| Viral infections | ||
| CMV D/R status | CMV D/R status (.03) | cGVHD (.009; 1.80; 1.16–2.80) |
| (.0004) | Era§ (.001; 0.41; 0.24–0.71) | |
| MRD (.001; 0.43; 0.26–0.73) |
OR indicates odds ratio; CI, confidence interval; aGVHD, acute graft-versus-host disease; cGVHD, chronic graft-versus-host disease; PBSC, peripheral blood stem cell product; CMV D/R, cytomegalovirus donor/recipient; MRD, matched-related donor.
1990–1994 vs 1995–1999.
1990–1994 vs 2000–2004.
1990–1994 vs 2005–2009.
1995–1999 vs 2005–2009.
Table 4.
Influence of D/R CMV Status on the Risk of Viral Infections at 0–30 and 31–100 Days Posttransplantation
| CMV D/R status | P value | OR | 95%CI |
|---|---|---|---|
| 0–30 days | |||
| −/+ vs +/+ | .003 | 0.45 | 0.26–0.76 |
| −/+ vs +/− | .0003 | 0.33 | 0.19–0.6 |
| +/− vs −/− | .004 | 2.10 | 1.27–3.49 |
| 31–100 days | |||
| +/+ vs −/− | .006 | 1.92 | 1.21–3.08 |
D/R indicates donor/recipient; CMV, cytomegalovirus; OR, odds ratio; CI, confidence interval.
Infections Between 31 to 100 Days after Transplantation
Patients with severe aGVHD and those who received a PBSC product had more bacterial infections and bacteremia (Table 3). Recipients of haploidentical donor transplantations were not at increased risk for bacterial infections (P = .36) or bacteremia (P = .24). Patients with severe aGVHD had more fungal infections. CMV D/R status was a risk factor for viral infection (Tables 3 and 4).
Infections 101 Days to 2 Years after Transplantation
Patients with cGVHD had more bacterial infections (Table 3). The cGVHD and CMV D/R status were associated with an increased risk for bacteremia; cGVHD and age more than 10 years at transplantation were associated with an increased risk for fungal infections. Any cGVHD also increased the risk for viral infections. Haploidentical donor transplantation was associated with an increased risk of chronic (P = .03) but not acute (P = .15) GVHD. However, recipients of haploidentical donor transplantation and T cell-depleted donors were not at increased risk for fungal (P = .14, P = .12) or viral (P = .11, P = .31) infections, respectively. RIC transplantations were not at risk for bacterial, fungal, viral infections, or bacteremia on univariate logistic regression analyses.
Influence of Era of Transplantation on Infections
Patients who underwent a transplantation between 1990 and 1994 had a higher risk of bacterial infections and bacteremia (Table 3), 0 to 30 days posttransplantation compared to those who underwent a transplantation between 1995 and 1999, 2000 and 2004, and 2005 and 2009. There was no significant difference in GP vs GN infections (P = .43) or bacteremia (P = .55) across the 4 eras. Patients who underwent a transplantation between 1995 and 1999 had a lower risk of viral infections compared with 2005 to 2009, at 31 to 100 days posttransplantation.
Infections in Recipients of MRD vs MUD HSCT
There were 243 and 239 patients who received an MRD and MUD HSCT, respectively. Recipients of MRD HSCT had a lower risk of viral infections 31 to 100 days posttransplantation (Table 3). Bacterial and fungal infections were not significantly different in the 2 groups.
Infection-Related Mortality in Recipients of Allogeneic HSCT
Of the 759 patients in the cohort, 60 (8%) died primarily due to infections. Eight patients (1%) died within 30 days, 24 of 735 patients (3%) between 31 to 100 days, and 28 of 622 patients (4%) died 100 days to 2 years posttransplantation. Of the infectious causes, bacterial, fungal, viral, and parasitic infections caused 12 (20%), 32 (53%), 14 (24%), and 2 (3%) deaths, respectively.
Bacterial causes included 9 deaths due to GN and 3 due to GP bacteremia and sepsis. Aspergillus spp caused 24 of the 28 deaths due to mold infections A. flavus accounted for 10 of these deaths. Non-Aspergillus molds caused 4 deaths, and C. albicans and non-albicans Candida spp caused 2 deaths each. PIV accounted for 4 of the 6 deaths due to respiratory viruses; ADV and CMV led to 6 and 2 deaths, respectively. Deaths due to parasitic infections included 1 patient with toxoplasmosis and another with roundworm infestation and pulmonary artery embolization. In multiple Cox regression analyses, severe aGVHD (P = .003; OR, 2.26; 95% CI, 1.33–3.86) and age at transplantation (P = .03; OR, 1.05; 95% CI, 1.00–1.11) were risk factors for infection-related mortality.
DISCUSSION
This study retrospectively looked at the infectious complications of allogeneic HSCT performed over a 20-year period in a cohort of 759 patients, mostly children and adolescents (77%) with a few infants (15%) and young adults (8%). MRD and MUD transplantations were equally represented. Approximately two-thirds received TBI-based conditioning. There were a large number of haploidentical donor transplantations, mostly performed after 2000; 48% received a T cell-depleted product, and a PBSC product was given to one-quarter of the patients, all after 2000.
GP infections outnumbered GN bacterial infections in all 3 time periods posttransplantation. No significant difference was noted in the incidence of GP vs GN bacteremia across eras. Mortality attributed to bacteremia was 6%. Bacterial infections and bacteremia were significantly higher in 1990 to 1994 compared with later eras, in the 0- to 30-day time period. The decrease in bacterial infections may be attributed to earlier engraftment, better supportive care, improved patient education, and scrupulous attention to hand hygiene in later time cohorts. None of the patients received routine prophylaxis with fluoroquinolones. Increasing GN bacteremia with fluoroquinolone resistance has been reported with the use of fluoroquinolone prophylaxis in adults, with a mortality of 11.3% [21]. Prophylaxis was associated with a GP/GN ratio of 1.0, incidence of E. coli bacteremia of 62% among GN rods, and E. faecium of 9%, compared to a GP/GN ratio of 1.4, incidence of E coli bacteremia of 34% and E. faecium of 3% in our series. Higher mortality in adults may be due to increased Enterobacteriacae and E. faecium associated with high attributable mortality rates. Other factors contributing to increased mortality in adults may include presence of comorbid conditions and higher incidence of GVHD.
Severe aGVHD was associated with bacteremia and bacterial infections 31 to 100 days post-HSCT; cGVHD increased the risk for bacterial infections late after transplantation. An independent association between bloodstream infections and aGVHD has been previously observed [22, 23]. Elaboration of cytokines after bacteremia could potentially increase the risk for aGVHD. Use of immune-suppressive medications for GVHD may increase the risk of bacteremia.
Recipients of a PBSC product had a higher incidence of bacteremia and bacterial infections 31 to 100 days posttransplantation. Higher aGVHD has been shown with PBSC compared to bone marrow transplantations [24], and this may have accounted for the increased risk. Pre-engraftment bloodstream infections have also been associated with a PBSC graft in a predominantly adult series of HSCT patients [25]. The higher incidence of bacteremia in adults may be related to the increased use of PBSC grafts.
Infections by Aspergillus spp were equally represented in all time periods posttransplantation, were the dominant fungi causing infection, and were associated with high mortality. Two- thirds of cases of aspergillosis were seen after engraftment, which is consistent with adult studies [26, 27]. The shift from early to late aspergillosis may be a result of the use of PBSC grafts and RIC, with attenuated cytopenias. Non-Aspergillus mold infections were uncommon, as previously observed [5]. Increased mold infections noted from 2005 to 2009 were largely due to diagnosis of probable invasive fungal infection based on a positive assay for galactomannan. Candidemia was equally distributed among C. albicans and non-albicans Candida spp, and was equally prevalent in all time periods. Severe aGVHD and cGVHD increased the risk of fungal infections. GVHD has previously been associated with an increased risk of aspergillosis in adults [28], perhaps related to impaired neutrophil and cellular immunity and use of corticosteroids.
An increased risk for aspergillosis has also been shown in older adult patients [29]. Our data support the increased risk of fungal infections with older age. T cell receptor rearrangement excision circles (TRECs) were shown to be significantly higher in patients younger than 19 years of age as compared to older patients [30]. Low TREC values correlated with severe opportunistic infections in the first month posttransplantation and extensive cGVHD [30]. Low TREC values have also been associated with a higher incidence of CMV infection [31]. CMV disease has also been shown to be associated with an increased risk for late invasive aspergillosis [28].
CMV viremia in our study was seen in 5% of patients preengraftment and in 9% 31 to 100 days posttransplantation. Adenoviremia was detected equally in all time periods. The 16% rate of CMV detection in our cohort, in which only half the patients received T cell- replete HSCT, is comparable to the 28% rate of detection reported in a cohort of 40 children after allogeneic T cell- replete HSCT [32]. Viral DNAemia was seen after neutrophil engraftment and predominantly up to day +100 [32]. The higher rate of viral infections between 2005 and 2009 may be due to better surveillance and improved diagnostic methods. The predominance of respiratory viral infections late after transplantation is a novel observation, which needs corroboration from other studies.
Patients with CMV D+/R− status had a higher risk of all viral infections at day 0 to 30 and those with D+/R+ status at day 31 to 100. Previous studies have confirmed the high risk of CMV viremia with D/R seropositivity, as seen in our study [33]. D+/R− patients have been shown to have higher transplantation-related mortality as compared to D−/R− patients, predominantly associated with bacterial and fungal infections [30]. A bidirectional relationship has also been observed between CMV replication and aGVHD [34].
Haploidentical donor HSCT was not associated with an increased risk of infections. In the first phase II study of haploidentical donor HSCT in adults with acute leukemia conditioned with a TBI-based regimen, 26% died from infectious causes, mainly CMV (13%) and fungal infections (5%) [35]. An RIC regimen in our institution using fludarabine, thiotepa, and melphalan with T cell depletion, led to rapid immune reconstitution with less viremia and aGVHD [36]. Close monitoring for viral and fungal infections, prophylaxis, and pre-emptive therapy may have also contributed to reducing the risk of infection in these patients.
Infections 100 days to 2 years post-HSCT caused the highest mortality (4%) of the 3 time periods post-HSCT. Late infections have also been shown to be an important contributor to mortality due to infection in predominantly adult patients who undergo HSCT, with aGVHD as the most important risk factor [37], similar to our results. Other risk factors for mortality included cGVHD and CMV infection [37].
Our study has several limitations. Cord blood transplantations were excluded from the analysis, as they were few in number. This is a major limitation, as infection and immune reconstitution are very different in cord transplantations, which represent a significant donor group in the pediatric population. Further, the patient population was heterogeneous, small numbers of infections precluded analyses within each subgroup, analyses were retrospective, only microbiologically documented infections were included, antifungal, antiviral prophylaxis, and methods for detection of viremia differed across the 2 decades. Parasitic infections were few in number.
There seems to be a continuum in the timeline of infections posttransplantation in our institution, with bacterial, fungal, and viral infections prevalent in all time periods, particularly late after the transplantation, the risk affected by GVHD, CMV D/R status, product type, older age, and use of unrelated donors. This contrasts with the conventional teaching of increased bacterial infections and Candidemia in the pre-engraftment phase followed by aspergillosis, CMV, and respiratory virus infections postengraftment, and decreased vulnerability to infections in the late posttransplantation period. Prospective studies with inclusion of cord and haploidentical transplantations are needed to confirm these observations.
ACKNOWLEDGMENTS
The authors thank Mark Mestemacher from the clinical laboratory and Nancy L. Hooper from the Bone Marrow Transplantation and Cellular Therapy, clinical research office, at SJCRH, Memphis, TN, for assistance in data collection.
This work was supported by National Cancer Institute Cancer Center CORE Support Grant P30 CA 21765 and by the American Lebanese Syrian Associated Charities.
Footnotes
Conflict of interest disclosure: The authors declare no competing financial interests.
Presentation: This work was presented at the ASBMT tandem meeting, February 2012, San Diego, CA.
This manuscript is dedicated to the memory of Dr. Jerry L. Shenep, October 7, 1951 – July 26, 2012, who worked tirelessly for immunocompromised children with infections.
REFERENCES
- 1.Collin BA, Leather HL, Wingard JR, Ramphal R. Evolution, incidence, and susceptibility of bacterial bloodstream isolates from 519 bone marrow transplant patients. Clin Infect Dis. 2001;33:947–953. doi: 10.1086/322604. [DOI] [PubMed] [Google Scholar]
- 2.Kontoyiannis DP, Marr KA, Park BJ, et al. Prospective surveillance for invasive fungal infections in hematopoietic stem cell recipients, 2001–2006: overview of the Transplant-Associated Infection Surveillance Network (TRANSNET) database. Clin Infect Dis. 2010;50:1091–1100. doi: 10.1086/651263. [DOI] [PubMed] [Google Scholar]
- 3.Boeckh M, Leisenring W, Riddell SR, et al. Late cytomegalovirus disease and mortality in recipients of allogeneic hematopoietic stem cell transplants: importance of viral load and T-cell immunity. Blood. 2003;101:407–414. doi: 10.1182/blood-2002-03-0993. [DOI] [PubMed] [Google Scholar]
- 4.Ljungman P, Perez-Bercoff L, Jonsson J, et al. Risk factors for the development of cytomegalovirus disease after allogeneic stem cell transplantation. Haematologica. 2006;91:78–83. [PubMed] [Google Scholar]
- 5.Garcia-Vidal C, Upton A, Kirby KA, Marr KA. Epidemiology of invasive mold infections in allogeneic stem cell transplant recipients: biological risk factors for infection according to time after transplantation. Clin Infect Dis. 2008;47:1041–1050. doi: 10.1086/591969. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ochs L, Shu XO, Miller J, et al. Late infections after allogeneic bone marrow transplantations: comparison of incidence in related and unrelated donor transplant recipients. Blood. 1995;86:3979–3986. [PubMed] [Google Scholar]
- 7.Wade JC, Newton B, Flournoy N, Meyers JD. Oral acyclovir for prevention of herpes simplex reactivation after marrow transplantation. Ann Intern Med. 1984;100:823–828. doi: 10.7326/0003-4819-100-6-823. [DOI] [PubMed] [Google Scholar]
- 8.Aversa F. Haploidentical haematopoietic stem cell transplantation for acute leukaemia in adults: experience in Europe and the United States. Bone Marrow Transplant. 2008;41:473–481. doi: 10.1038/sj.bmt.1705966. [DOI] [PubMed] [Google Scholar]
- 9.Flomenberg P, Babbitt J, Drobyski WR, et al. Increasing incidence of adenovirus disease in bone marrow transplant recipients. J Infect Dis. 1994;169:775–781. doi: 10.1093/infdis/169.4.775. [DOI] [PubMed] [Google Scholar]
- 10.Srinivasan A, Wang C, Yang J, Shenep JL, Leung WH, Hayden RT. Symptomatic parainfluenza virus infections in children undergoing hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2011;17:1520–1527. doi: 10.1016/j.bbmt.2011.03.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hale GA, Heslop HE, Krance RA, et al. Adenovirus infection after pediatric bone marrow transplantation. Bone Marrow Transplant. 1999;23:277–282. doi: 10.1038/sj.bmt.1701563. [DOI] [PubMed] [Google Scholar]
- 12.Gu Z, Belzer SW, Gibson CS, Bankowski MJ, Hayden RT. Multiplexed, real-time PCR for quantitative detection of human adenovirus. J Clin Microbiol. 2003;41:4636–4641. doi: 10.1128/JCM.41.10.4636-4641.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.De Pauw B, Walsh TJ, Donnelly JP, et al. Revised definitions of invasive fungal disease from the European Organization for Research and Treatment of Cancer/Invasive Fungal Infections Cooperative Group and the National Institute of Allergy and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus Group. Clin Infect Dis. 2008;46:1813–1821. doi: 10.1086/588660. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Leung WH, Turner V, Richardson SL, et al. Effect of HLA class I or class II incompatibility in pediatric marrow transplantation from unrelated and related donors. Hum Immunol. 2001;62:399–407. doi: 10.1016/s0198-8859(01)00220-8. [DOI] [PubMed] [Google Scholar]
- 15.Leung W, Campana D, Yang J, et al. High success rate of hematopoietic cell transplantation regardless of donor source in children with very high-risk leukemia. Blood. 2011;118:223–230. doi: 10.1182/blood-2011-01-333070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Przepiorka D, Weisdorf D, Martin P, et al. 1994 Consensus conference on acute GVHD grading. Bone Marrow Transplant. 1995;15:825–828. [PubMed] [Google Scholar]
- 17.Ljungman P, Engelhard D, de La Cámara R, et al. Vaccination of stem cell transplant recipients: recommendations of the Infectious Disease Working Party of the EBMT. Bone Marrow Transplant. 2005;35:737–746. doi: 10.1038/sj.bmt.1704870. [DOI] [PubMed] [Google Scholar]
- 18.Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94:496–509. [Google Scholar]
- 19.Kalbfleisch JD, Prentice RL. The statistical analysis of failure data. New York: John Wiley; 1980. pp. 163–188. [Google Scholar]
- 20.Gray RJ. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat. 1988;16:1141–1154. [Google Scholar]
- 21.Mikulska M, Del Bono V, Raiola AM, et al. Blood stream infections in allogeneic hematopoietic stem cell transplant recipients: reemergence of gram-negative rods and increasing antibiotic resistance. Biol Blood Marrow Transplant. 2009;15:47–53. doi: 10.1016/j.bbmt.2008.10.024. [DOI] [PubMed] [Google Scholar]
- 22.Poutsiaka DD, Munson D, Price LL, Chan GW, Snydman DR. Blood stream infection (BSI) and acute GVHD after hematopoietic SCT (HSCT) are associated. Bone Marrow Transplant. 2011;46:300–307. doi: 10.1038/bmt.2010.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Atkinson K, Storb R, Prentice RL, et al. Analysis of late infections in 89 long-term survivors of bone marrow transplantation. Blood. 1979;53:720–731. [PubMed] [Google Scholar]
- 24.Eapen M, Logan BR, Confer DL, et al. Peripheral blood grafts from unrelated donors are associated with increased acute and chronic graft-versus-host disease without improved survival. Biol Blood Marrow Transplant. 2007;13:1461–1468. doi: 10.1016/j.bbmt.2007.08.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Almyroudis NG, Fuller A, Jakubowski A, et al. Pre- and postengraftment bloodstream infection rates and associated mortality in allogeneic hematopoietic stem cell transplant recipients. Transpl Infect Dis. 2005;7:11–17. doi: 10.1111/j.1399-3062.2005.00088.x. [DOI] [PubMed] [Google Scholar]
- 26.Neofytos D, Horn D, Anaissie E, et al. Epidemiology and outcome of invasive fungal infection in adult hematopoietic stem cell transplant recipients: analysis of multicenter prospective antifungal therapy (PATH) alliance registry. Clin Infect Dis. 2009;48:265–273. doi: 10.1086/595846. [DOI] [PubMed] [Google Scholar]
- 27.Kontoyiannis DP, Marr KA, Park BJ, et al. Prospective surveillance for invasive fungal infections in hematopoietic stem cell transplant recipients, 2001–2006: overview of the transplant-associated infection surveillance network (TRANSNET) database. Clin Infect Dis. 2010;50:1091–1100. doi: 10.1086/651263. [DOI] [PubMed] [Google Scholar]
- 28.Marr KA, Carter RA, Boeckh M, Martin P, Corey L. Invasive aspergillosis in allogeneic stem cell transplant recipients: changes in epidemiology and risk factors. Blood. 2002;100:4358–4366. doi: 10.1182/blood-2002-05-1496. [DOI] [PubMed] [Google Scholar]
- 29.Baddley JW, Stroud TP, Salzman D, Pappas PG. Invasive mold infections in allogeneic bone marrow transplant recipients. Clin Infect Dis. 2001;32:1319–1324. doi: 10.1086/319985. [DOI] [PubMed] [Google Scholar]
- 30.Lewin SR, Heller G, Zhang L, et al. Direct evidence for new T-cell generation by patients after either T-cell-depleted or unmodified allogeneic hematopoietic stem cell transplantations. Blood. 2002;100:2235–2242. [PubMed] [Google Scholar]
- 31.Clave E, Rocha V, Talvensaari K, et al. Prognostic value of pretransplantation host thymic function in HLA-identical sibling hematopoietic stem cell transplantation. Blood. 2005;105:2608–2613. doi: 10.1182/blood-2004-04-1667. [DOI] [PubMed] [Google Scholar]
- 32.Schönberger S, Meisel R, Adams O, et al. Prospective, comprehensive, and effective viral monitoring in children undergoing allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2010;16:1428–1435. doi: 10.1016/j.bbmt.2010.04.008. [DOI] [PubMed] [Google Scholar]
- 33.Nichols WG, Corey L, Gooley T, Davis C, Boeckh M. High risk of death due to bacterial and fungal infections among cytomegalovirus (CMV)- seronegative recipients of stem cell transplants from seropositive donors: evidence for indirect effects of primary CMV infection. J Infect Dis. 2002;185:273–282. doi: 10.1086/338624. [DOI] [PubMed] [Google Scholar]
- 34.Cantoni N, Hirsch HH, Khanna N, et al. Evidence for a bidirectional relationship between cytomegalovirus replication and acute graftversus- host disease. Biol Blood Marrow Transplant. 2010;16:1309–1314. doi: 10.1016/j.bbmt.2010.03.020. [DOI] [PubMed] [Google Scholar]
- 35.Aversa F, Terenzi A, Tabilio A, et al. Full haplotype-mismatched hematopoietic stem-cell transplantation: a phase II study in patients with acute leukemia at high risk of relapse. J Clin Oncol. 2005;23:3447–3454. doi: 10.1200/JCO.2005.09.117. [DOI] [PubMed] [Google Scholar]
- 36.Chen X, Hale GA, Barfield R, et al. Rapid immune reconstitution after a reduced-intensity conditioning regimen and a CD3-depleted haploidentical stem cell graft for paediatric refractory haematological malignancies. Br J Haematol. 2006;135:524–532. doi: 10.1111/j.1365-2141.2006.06330.x. [DOI] [PubMed] [Google Scholar]
- 37.Bjorklund A, Aschan J, Labopin M, et al. Risk factors for fatal infectious complications developing late after allogeneic stem cell transplantation. Bone Marrow Transplant. 2007;40:1055–1062. doi: 10.1038/sj.bmt.1705856. [DOI] [PubMed] [Google Scholar]