Bacterial infections in patients in allogenic stem cell transplantation programs

Introduction: Bacterial infections are common in patients enrolled in allogenic hematopoietic stem cell transplantation. Infections are the most common cause of illness and mortality in transplanted patients. The most common pathogens are Staphylococcus and Enterobacteriaceae, causing 25% of bacterial infections, each, followed by Enterococcus and P. aeruginosa.

Aim: The aim of the paper is calculating the frequency of bacterial infections in patients enrolled in allogenic hematopoietic stem cell transplantation, at the Clinic for Hematology of the University Clinical Center of Serbia, as well as analyzing the epidemiological situation, the frequency of different clinical presentations of bacterial infections, and their effect on overall survival.

Materials and methods: This retrospective cohort study enrolled 58 patients. The presence of bacteria was determined with standard microbial cultivation from samples. The diagnosis criteria for localized bacterial infection are a positive culture and characteristic clinical presentation. Sepsis is clinically diagnosed by a combination of a positive hemoculture and the existence of systemic inflammatory response. Management of bacterial infection starts empirically and is continued in keeping with antibiogram results. Patient survival was analyzed with the Kaplan-Meier method and compared with the log-rank test.

Results: Bacterial infections were registered in 15 (25.9%) patients. The most common pathogens were Gram-negative bacteria (65.2%). The most common clinical presentation was sepsis (59.0%). Bacterial infection lethality was 60%. The Kaplan-Meier curve showed the median value for estimated patient survival in patients without bacterial infection to be 106.0 months (95% CI 85.2 – 163.3), while, for patients with bacterial infection, it was 14.0 (95% CI 8.9 – 19.1). The log-rank test showed a significant difference in the length of survival between these two groups (p < 0.001). Pseudomonas spp. was sensitive to colistin and/or meropenem. Klebsiella pneumoniae was sensitive to colistin.

Conclusion: Bacterial infections are common at our clinic and are caused by endogenous opportunistic bacteria. Therefore, emphasis should be placed on the implementation of preventive measures aimed at preventing hospital-acquired infections.

INTRODUCTION

Bacterial infections (BI) in immunosuppressed patients, which includes patients in the allogenic hematopoietic stem cell transplantation program (allo-HSCT), represent frequent and potentially life-threatening treatment complications. Infections are generally the most common cause of illness and death in transplanted patients [1],[2]. They can occur as a de novo infection, or they may be transferred from the donor, or activated from latent forms in the recipient [3].

The risk of bacterial infection development in these patients depends on numerous factors, including the type of graft, the source of stem cells, the application of the myeloablative regimens, immune reconstitution, presence of graft versus host disease (GVHD) [4], presence of mucositis, being a germ carrier, presence of vascular and other disorders. Patients in the program of allo-HSCT are extremely sensitive to bacterial infections, since, due to the damage sustained by the mucosa and caused by chemotherapeutics, the natural membrane and protection against the invasion of endogenous microorganisms is lost. During patient conditioning for allo-HSCT, bone marrow ablation occurs as well as neutropenia (whose improvement is expected upon engraftment), thus leading to the weakening of the immunological component of defense and allowing for the invasion of bacteria and the development of bacterial infection.

The portal of entry for bacterial infections is the intestinal mucosa, when it comes to Gram-negative pathogens. When it comes to Gram-positive bacteria, the most common portal of entry are vascular catheters.

Studies have shown the predominance of different pathogens in different transplantation centers, which is why determining the epidemiological situation in a given institution is of crucial importance for the prophylaxis of bacterial infections. Generally speaking, the most common causative agents of bacterial infections in patients undergoing allo-HSCT are hospital pathogens, and the infections are therefore nosocomial. Coagulase-negative Staphylococcus and Enterobacteriaceae are the most common pathogens, and they case approximately 25% of bacterial infections, each. They are followed by Enterococcus, P. aeruginosa and viridans streptococci, as well as by pseudomembranous colitis caused by C. Difficile, which can be found in 15% of the cases, more frequently in allogenic than in autologous transplantations [5].

The most frequent clinical presentations of bacterial infections are the following: febrile neutropenia, which occurs in 80% of patients undergoing allo-HSCT [6], sepsis and septic shock, gastrointestinal tract infections, pneumonia, and soft tissue infections. Less frequently, infections of the urinary tract and the central nervous system are found [6].

In the previous decades, good antibacterial prophylaxis and prompt treatment of febrile neutropenia have led to a drop in the morbidity and mortality from bacterial infections, in patients undergoing allo-HSCT. However, a new challenge has emerged – antimicrobial resistance. Namely, with time, a group of six multiresistant microorganisms has emerged, and has been labeled ESKAPE (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp.) [7]. The problem of multiresistant bacteria, especially Gram-negative ones, is of extreme importance in patients with prolonged neutropenia undergoing longterm chemotherapy regimens [8].

The aim of this paper is to determine the frequency of bacterial infections in patients undergoing allo-HSCT at the Clinic for Hematology of the University Clinical Center of Serbia (UCCS); determine the epidemiological situation and the distribution of different pathogens; determine the frequency of different clinical presentations of bacterial infections, as well as their impact on patient survival, with a special insight into septic states.

MATERIALS AND METHODS

This retrospective cohort study involved 58 patients treated at the Clinic for Hematology of the UCCS, who were in the process of allogenic bone marrow transplantation and were followed up from December 2017 until December 2020. The diagnoses of the patients were as follows: acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), Hodgkin lymphoma (HL), non-Hodgkin lymphoma (NHL), myelodysplastic syndrome (MDS), myelodysplastic/myeloproliferative syndrome (MDS/MPS).

Microbiological methos

The presence of bacteria in the isolate was determined with standard microbial cultivation from the obtained samples.

The following were used: blood samples, urine specimens, sputum samples, stool samples, skin swabs, mucosal swabs, and swabs of soft tissue lesions.

Clinical criteria

Clinical diagnosis of localized bacterial infection was established on the basis of the presence of characteristic signs, with bacteriologic confirmation of infection.

Clinical diagnosis of sepsis was established on the basis of a positive blood culture and the presence of systemic inflammatory response (Table 1).

Table 1. Criteria for systemic inflammatory response syndrome (SIRS); (presence of two or more of the presented symptoms)

Clinical diagnosis of septic shock was established on the basis of the presence of hypotension and cardiovascular instability in a patient with confirmed sepsis.

Prophylaxis and treatment

All patients undergoing allo-HSCT received antimicrobial prophylaxis. Quinolone antibiotics (levofloxacin 500 mg p.o.) and cephalosporins (cefepime i.v.) were used as antibacterial prophylaxis. Bactrim (2 x 400 mg p.o.) was used for preventing the reactivation of Toxoplasma gondii and Pneumocystis carinii. Acyclovir 3 x 400 mg p.o. was used for antiviral prophylaxis, while for prophylaxis of fungal infections, micafungin 50 mg i.v. was used until engraftment, and then posaconazole 3 x 200 mg p.o., until 90 days following engraftment.

In most cases, bacterial infections are initially treated empirically, with broad-spectrum antibiotics, until the antibiogram results are received. After that, antibiotics are administered according to the antibiogram, which was the case with our patients as well.

Surgical methods were applied in treating patients with skin infections and infections of soft tissues, while patients in septic shock received supportive therapy.

Prior to allo-HSCT, patients underwent conditioning regimens, which involve the application of chemotherapeutical agents and/or radiotherapy, prior to transplantation. Two regimens are in use – myeloablative conditioning (MAC) and reduced-intensity conditioning (RIC). MAC is more aggressive and leads to complete destruction of the bone marrow (aplasia).

Statistical data processing

Initially, a database was formed by grouping results and categorizing them in tables, according to the studied patient characteristics. Descriptive statistical parameters were expressed as the median and the distributions of relative frequencies. The overall patient survival covered the period from the time of diagnosis until the lethal outcome or until December 2020, for surviving patients. Patient survival was analyzed with the Kaplan-Meier method and compared with the logrank test. Statistical data processing was performed using the SPSS 23.0 software for Microsoft Windows.

RESULTS

The study involved 58 patients – 33 (56.9%) male patients and 25 (43.1%) female patients. Their median age was 38.40 years (range: 19 – 58). The frequency of the diagnoses for patients involved in the study were as follows: ALL (31.0%), AML (44.8%), CLL (1.7%), Hodgkin lymphoma (13.8%), MDS (1.7%), MDS/MPS (3.4%), NHL (3.4%) (Table 2).

Table 2. Frequency of hematological diseases

More of the analyzed patients, 43 (74.1%) of them, were on the MAC regimen, while a lesser number was conditioned with the RIC regimen, i.e., 15 (25.9%) patients.

Bacterial infections occurred in 15 (25.9%) patients, of whom three patients had polymicrobial infection. This is why the total number of isolated bacteria was 23. The most frequent pathogens were Gram-negative opportunistic bacteria (15; 65.2%), while Gram-positive pathogens caused infections in 34.8% of the cases. The most common pathogens were Pseudomonas spp. and Acinetobacter baumannii, while the least common were Enterococcus faecium, Providencia stuartii, and Bacillus spp. The frequencies of the isolated pathogens have been presented in detail in Table 3 and Graph 1 (Table 3 and Figure 1).

Table 3. Frequency of bacterial pathogens

Figure 1. Frequency of bacterial pathogens

The following samples were taken for bacteriological testing: blood samples, urine specimens, stool samples, sputum samples, and swabs taken from the skin, mucosa and soft tissue lesions. The total number of samples taken for laboratory testing was 17. Sampling was most frequently done from blood (47.1%), followed by swabs taken from the skin, mucosa, and soft tissue lesions (23.5%), urine (11.8%), stool samples (11.8%), and finally sputum (5.9%).

The total number of clinical presentations of the infections was 22. Of that number, the most common was sepsis, with or without septic shock (13 patients; 59.0% of infections), followed by infections of the skin, mucosa, and soft tissues (18.0%), pseudomembranous colitis (14.0%), and finally, infections occurring individually – pneumonia and enterocolitis (4.5%, each) (Table 4).

Table 4. Frequency of bacterial infection clinical presentations

The number of patients who died due to bacterial infections was 9, of whom 8 clinically presented with sepsis. Therefore, lethality of bacterial infections was 60.0%, while the mortality rate related to bacterial infections amongst patients involved in the allo-HSCT program was 15.5%.

Based on the Kaplan-Meier survival curve, the median of survival in patients without proven bacterial infection was 106.00 months (95% CI 85.2 – 163.3), while, for patients with bacterial infections, it was 14.0 months (95% CI 8.9 – 19.1). Based on the log-rank test, a statistically significant difference was registered in the survival of these two groups of patients (p < 0.001) (Figure 2).

Figure 2. Frequency of bacterial pathogens

Fisher’s exact test was used to compare the frequency of bacterial infections in relation to the conditioning regimen (MAC or RIC). The test did not show a statistically significant difference in frequency (p = 0.308).

Descriptive data

Of a total of 13 patients with sepsis, two patients had septic shock.

The four infections of skin, mucosa, and soft tissues were as follows: mucositis, phlegmon of the earlobe, phlegmon of the vulva, and phlegmon of the perianal region.

Polymicrobial infection occurred in three cases, with the following pathogens:

• Acinetobacter baumannii, Klebsiella pneumoniae, Enterobacter spp.
• Pseudomonas spp., Enterobacter spp.
• Klebsiella pneumoniae, coagulase-negative Staphylococcus, Providencia stuartii.

We had an antibiogram for 11 pathogens, and the antibiogram results were as follows: in all the laboratory results, Pseudomonas spp. was sensitive to colistin and/or meropenem; Staphylococcus aureus was sensitive to amikacin, vancomycin, and rifampicin; Enterococcus faecium was sensitive to vancomycin and meropenem; in all cases, Klebsiella pneumoniae was multiresistant, but sensitive to colistin.

DISCUSSION

The frequency of bacterial infections in patients involved in the program of allo-HSCT, at the Clinic for Hematology of the UCCS, falls within the range reported in other publications. At our hospital, the frequency of bacterial infections has proven to be somewhat smaller than reported by the Polish Pediatric Group for Hematopoietic Stem Cell Transplantation, in pediatric patients involved in the allo-HSCT program, where the frequency was 40% [9], while it was 25.9% in our study.

A study by Balletto and Mikulska [5] showed that Gram-positive pathogens were more frequent in patients undergoing allo-HSCT (60%), as compared to Gram-negative bacteria (40%). However, the frequencies in different hospitals are different. These frequencies range from 85%/15% to the advantage of Gram-positive bacteria to 74%/26% to the advantage of Gram-negative bacteria, which is in keeping with our result, which was 65.22%/34.78% to the advantage of Gram-negative pathogens.

It can be noted that the bacteria isolated in our study are often opportunistic and that they are in the group of pathogens causing hospital-acquired infections, which is why the state of immunosuppression is not a key factor in their pathogenesis. Also, in the prevention of these infections, measures for preventing nosocomial infections are of the utmost importance.

There are data in literature on the occurrence of opportunistic mycobacterial infections and infection with Nocardia asteroides in transplanted patients; however, in our experience so far, this has not been the case, [3],[11]. A significant factor in their development is immunosuppression, and the infections are chronic.

The type of conditioning regimen is mentioned in literature as a significant risk factor for the development of bacterial infection [5]. The MAC regimen is believed to be linked with more frequent infections; however, our study did not establish this link (the p value of Fisher’s exact test was greater than 0.05). This result could be attributed to the small size of the sample, but also to the inequality amongst the groups of patients conditioned with different regimens. It is interesting that the patients which were on the MAC regimen had a 6.5 times higher frequency of bacterial infections (MAC/RIC = 13/2).

The profile of the antimicrobial resistance of the bacteria found in our isolates speaks in favor of the existence of significant resistance, which is also indicated by the fact that Pseudomonas spp. was sensitive only to colistin and/or meropenem, which are in the reserved group of antibiotics. It is interesting to note that the presence of vancomycin-resistant Enterococcus (VRE) was not registered. These data should be taken with certain reservations, due to the limited size of the sample, which is primarily the result of the lack of data in the health information system and the electronic medical records. The antibiogram for C. difficile was not performed, due to the predictable result and good empirical therapy (metronidazole or vancomycin p.o.).

The distribution of clinical presentations of bacterial infections in our study is in keeping with other publications; the most common clinical presentation in these patients is sepsis. In our population, it presented in 59% of bacterial infections, while in a study involving 313 subjects who were followed up until day 100 after hematopoietic stem cell transplantation, a positive hemoculture occurred in 64.1% of the patients [11].

In long-term follow-up, Pseudomonas spp. has shown to be the most common pathogen that leads to bacterial infection in patients undergoing allo-HSCT [12], which has proven to be true in our study as well.

A study from 2019, involving a group of patients with multiple myeloma following bone marrow transplantation, determined that sepsis caused by enterococcus significantly affected overall patient survival [13]. Our study can provisionally be compared with this study, but it also shows, based on the Kaplan-Meir curve, a significant drop in survival in patients with bacterial infection, with the predominant BI being sepsis.

In our study, lethality due to bacterial infections was 60%, which is significantly higher than the range of expected lethality of 6% to 40%, reported in two studies which had the most extreme results [14],[15].

Our study has several limitations. The group of patients was heterogenous; the study is retrospective; the number of subjects was limited due to the fact that this is a new transplantation center. As already stated, the antibiogram results were lacking for some of the patients, due to technical difficulties, which were outside of our control. Due to a small number of samples, the number of identified bacterial infections was also small, which is why certain parameters were reported descriptively.

CONCLUSION

At our clinic, in patients undergoing allo-HSCT, bacterial infections are common, but are not only the result of immunosuppression, rather the combination of immunosuppression and the effect of hospital and endogenous opportunistic bacteria. This is why its is necessary to direct the prevention of these diseases towards comprehensive measures of preventing hospital-acquired infections and towards antimicrobial prophylaxis of endogenous, primarily Gram-negative pathogens. With this study we have shown, to a certain degree, the epidemiological situation at our Center, which, going forward, may significantly influence the planning of prophylaxis. The fact that bacteria with an extreme spectrum of resistance, such as carbapenem-resistant Pseudomonas, methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus (VRE), have as yet not been registered on our ward, is encouraging.

The most frequent bacterial infections, described in this study, presented in the form of sepsis, with or without septic shock. Their treatment is complex, often necessitating a multidisciplinary approach, and they can cause complications and significantly worsen the patient’s prognosis. Knowing the profile of antimicrobial resistance enables the introduction of more specific empirical therapy before the antibiogram is obtained, which may significantly improve outcomes. It is quite evident, however, that when Gram-negative bacteria (Pseudomonas spp., Klebsiella pneumoniae) are suspected, standard empirical antibiotic treatment has its limitations, due to resistance.

LIST OF ABBREVIATIONS AND ACRONYMS

BI – bacterial infections
allo-HSCT – allogenic hematopoietic stem cell transplantation (bone marrow transplantation)
GVHD – graft versus host disease
ALL – acute lymphoblastic leukemia
AML – acute myeloid leukemia
CLL – chronic lymphocytic leukemia
HL – Hodgkin lymphoma
NHL – non-Hodgkin lymphoma
MDS – myelodysplastic syndrome
MDS/MPS – myelodysplastic/myeloproliferative syndrome
SIRS – systemic inflammatory response syndrome
p.o. – per os (administered orally)
i.v. – intravenous
MAC – myeloablative conditioning
RIC – reduced-intensity conditioning