Antibiotic lock therapy for the treatment of peripherally inserted central venous catheter-related bloodstream infection in patients with hematological malignancies: a single center retrospective study

Abstract

Catheter-related bloodstream infections represent one of the most prevalent complications in patients with peripherally inserted central venous catheters (PICCs). The application of antibiotic lock therapy (ALT), particularly in patients with hematological malignancies, has not been well documented. We aim to share our experience on ALT for these patients and to evaluate its effectiveness and safety. All cases of patients with hematological malignancies who had PICC from January 2018 to October 2024 were retrospectively reviewed. Microbiologic data of PICC-related bloodstream infections (PRBSIs) were collected. A comparison was made between patients managed with ALT and those without it. Factors affecting PICC removal were also explored. A total of 45 patients experienced 67 episodes of PRBSIs, yielding an incidence rate of 2.98 per 1,000 PICC days. The median time of PRBSI onset was 42 days. Predominant pathogens included Gram-negative bacilli (49.3%) and Gram-positive cocci (35.8%). The catheter salvage rate was significantly higher at 76.5% when ALT was combined with systemic antibiotic therapy (SAT), compared to 51.5% for SAT alone (p = 0.033). 3 death events (3/34) compared with 4 death events (4/33) occurred in each therapeutic regimen (p = 0.709). Elevated procalcitonin levels (> 2ng/ml) and inadequate empirical therapy were risk factors for PICC removal; conversely, ALT served as a protective factor against it. ALT in combination with systemic antibiotics is a safe and effective approach for managing PRBSIs in patients with hematological malignancies, helping to avoid unnecessary catheter removal and could be considered in clinical practice when catheter retention is desired.

Introduction

With the development of intravenous infusion techniques, intravascular catheters are increasingly utilized in patients with hematological malignancies for medication administration, chemotherapy, transfusions, and parenteral nutrition [1]. Commonly catheter types include non-tunneled central venous catheters (CVCs), peripherally inserted central venous catheters (PICCs), and totally implanted ports [2, 3]. PICCs offer benefits of a simple insertion, versatility, and a low risk of infectious complications; consequently, their application is on the rise [4, 5].

Despite the convenience of intravascular catheters, infectious complications—such as catheter-related (CRBSI) or central line-associated bloodstream infection (CLABSI) —are emerging as significant public health concerns, leading to increased morbidity and mortality, prolonged hospital stays and excess healthcare costs [6, 7]. An international survey of 630 ICUs across 45 countries from 2015 to 2020 reported a CLABSI rate at 4.55 per 1,000 central line days [8]. In the United States, CLABSIs were identified as the most costly healthcare-associated infections at $45,814 [9], while in China the average cost was $67,923, with CRBSI accounting for $33,696 [10].

Characteristics of patients, providers, and devices are closely associated with the occurrence of CRBSI. Among patient factors, immunocompromise due to hematologic cancer, neutropenia and malnutrition increase the risk of infection [11]. Patients with hematological malignancies often experience prolonged neutropenia and multiple cycles of chemotherapy, which places them at a higher risk of CRBSI compared to other patient populations.

Systemic antibiotics and catheter removal are desirable approaches for treating CRBSI [12]. However, in patients with hematological malignancies, catheter removal or replacement is not always feasible or practical due to the need for requiring multiple infusions of chemotherapeutic agents, blood products and large amounts of fluid. Additionally, the high risk of hemorrhage and limited peripheral venous route options complicate the establishment of new vascular access in a short time to guarantee the continuation of intravenous infusions. Antibiotic lock therapy (ALT) in conjunction with systemic antibiotics is a useful strategy in terms of eliminating infections as well as retaining catheters [13]. Although the application of ALT has been reported in a number of clinical studies and recommended by some guidelines under certain conditions [12, 14–16], its efficacy and safety in patients with hematological malignancies remain unclear. In view of their degree of immunosuppression and vulnerability to severe infections, it is necessary to figure out whether catheter salvage is suitable for these individuals. The objective of our study is to investigate the feasibility of ALT for salvaging catheters concerning PICC-related bloodstream infections (PRBSIs) among patients with hematological malignancies.

Materials and methods

Patient population and data collection

Patients with hematological malignancies who had PICC from January 2018 to October 2024 were enrolled. All data were collected retrospectively from patient files and medical records, including demographics, diagnosis, comorbidities, catheter implantation details, microbiologic and laboratory data of PRBSI, antimicrobial treatment and clinical outcomes. PICC-related complications within 90 days from PRBSI occurrence and other adverse effects throughout the duration of ALT were also investigated. This research was in accordance with the Declaration of Helsinki, and was approved by the Ethics Committee of No. 971 Hospital of Navy (971LL-2025031601). Informed consent was obtained from patients or their relatives.

Definitions

PRBSI was defined based on the IDSA definition [15], which include clinical signs of infection with no apparent source of bloodstream infection (BSI) other than PICC, and at least one of the following microorganic criteria: (1) detection of the same organism in cultures from peripheral blood and catheter tip, and (2) positivity of blood cultures obtained from the catheter being ≥ 2 h before those obtained from a peripheral vein. The colony count criterion was not adopted due to the unavailability of quantitative blood culture in our microbiological laboratory.

As a matter of fact, the criterion for CRBSI is stringent and may have limited applicability in everyday practice. To facilitate clinical application, we adopted the definition of “probable PRBSI” according to previously published guidelines [12, 14], defined as: presence of a pathogen in blood cultures sampled from the PICC lumen or tip, but not in concomitantly sampled blood cultures from the peripheral vein, along with systemic signs of infection and no other symptomology or local signs of infection.

Multidrug-resistant (MDR) strains were defined as acquired non-susceptibility to at least one agent in three or more antimicrobial categories for Gram-negative and Gram-positive bacteria. Specifically, for staphylococcal strains, resistance to a key antimicrobial agent, methicillin, was also classified as MDR. Empirical therapy was considered inadequate when antimicrobials with in vitro activity against the causative microorganism were not administered within the first 24 h after blood culture collection.

PICC salvage rate was defined by both catheter retention and clinical resolution. Recurrence was defined as the occurrence of another PRBSI caused by the same pathogen within 90 days after the previous occurrence. Non-infectious complications after catheterization included complete or partial occlusions, malfunctions, malpositions, ruptures, phlebitis, and bleeding. Temporary occlusions that were resolved through recanalization using flushing solutions (plasminogen activators—urokinase) were excluded.

PICC insertion and management

PICCs were inserted through peripheral vessels of right or left arm. Ultrasonography was routinely used to identify the vein and insertion location before catheter placement. The position of catheter tip was confirmed by chest radiography following insertion. The device function and the local situation of insertion site were checked daily during hospital stay. Catheters were flushed with normal saline and locked with a heparin solution at a concentration of 50 U/mL during routine maintenance. The lock volume was defined according to manufacturer’s instructions.

PRBSI management and ALT

When symptoms of infections such as fever, chills, rigors, hypotension, or other signs of sepsis occurred and PRBSI was suspected, paired blood cultures were obtained simultaneously from the PICC lumen and peripheral blood. After obtaining appropriate cultures, empiric systemic antibiotic therapy (SAT) was administered based on the assessment of patients’ immune status, the severity of infection, and the local ecology of the ward. When PRBSI was confirmed, systemic antibiotics were adjusted in accordance with drug sensitivity results. PICC salvage or removal was determined by clinicians. Once the decision of retention was made, the catheter was no longer used for infusion and ALT was initiated. The lock solution consisted of an antibiotic and an anticoagulant (heparin at 50 U/mL). Antibiotic locks were prepared as follows [13, 17, 18]: vancomycin 10 mg/mL, amikacin 10 mg/mL, piperacillin/tazobactam 56.25 mg/mL, cefuroxime 150 mg/mL, cefoperazone/sulbactam 100 mg/mL, meropenem 50 mg/mL, biapenem 30 mg/mL, micafungin 5 mg/mL (Table 1). The physical compatibility of lock solutions was assessed through visual inspection during preparation, ensuring that there were no visible signs of haze, particulate formation, turbidity, color change, or gas evolution. The lock solution was prepared immediately prior to use and remained within the lumen for a duration of 24 to 48 h.

Table 1.

Details of antimicrobials used in ALT [13, 17, 18]

Antimicrobials	Frequency, n (%)	Concentrations (mg/mL)	Identical agents used in SAT, n (%)1
Vancomycin	6 (17.6)	10	4 (66.7)
Amikacin	1 (2.9)	10	1 (100)
Cefuroxime	2 (5.9)	150	1 (50)
Cefoperazone/sulbactam	7 (20.6)	100	5 (71.4)
Piperacillin/tazobactam	7 (20.6)	56.25	5 (71.4)
Biapenem	6 (17.6)	30	4 (66.7)
Meropenem	3 (8.8)	50	3 (100)
Micafungin	2 (5.9)	5	2 (100)
Total	34 (100)	NA	25 (73.5)

¹Pencentage within their frequencies in ALT. ALT: antibiotic lock therapy; NA: not applicable; SAT: systemic antibiotic therapy

PICC was removed in the following situations: (1) severe sepsis or septic shock; (2) metastatic infection (septic thrombophlebitis, endocarditis, osteoarticular infection); (3) deterioration of clinical status and laboratory index after 72 h of appropriate antibiotic therapy; (4) persistence of bacteremia in subsequent blood cultures, with the same microorganism as initially identified.

Data analysis

The incidence of PRBSI was calculated as episodes per 1,000 PICC days. The proportion of different pathogens and MDR species was statistically analyzed. PICC salvage rate, PRBSI related mortality, complications during catheter use and other relevant information were compared between treatment with and without antibiotic lock. Risk factors and protective factors associated with PICC removal were also analyzed.

Quantitative variables with nonnormal distribution were presented as medians and interquartile range (IQR). Categorical variables were reported as frequencies and percentages. The chi-square or Fisher’s exact test was used for comparisons of categorical variables where appropriate. Continuous variables were compared using the Wilcoxon rank-sum test. The Kaplan-Meier method was used to estimate the cumulative incidence rate of PRBSI. Univariate logistic regression was conducted to identify predictors of PICC removal. Variables with P value < 0.1 were entered into the multivariate logistic model. The Hosmer-Lemeshow test was used to test the goodness of fit for multivariable testing, while odds ratio (OR) and 95% confidence interval (CI) were calculated. All statistical tests were 2-tailed, with a significance level of 0.05. Statistical analysis was performed with the “IBM SPSS Statistics 29.0” software.

Results

Patient characteristics and clinical outcomes

From January 2018 to October 2024, a total of 193 PICCs were inserted in 134 patients with hematological malignancies, generating a total of 22,456 PICC days. All PICCs were single-cavity devices and the predominant indication for insertion was chemotherapy delivery. The median duration of catheter use was 88 days (IQR, 35–181). Among the cohort of 134 patients included in this study, 45 individuals with 76 insertions developed 67 episodes of PRBSIs, including 49 definite diagnosis cases and 18 probable diagnosis cases. Detailed information of patients with PRBSI were described in Table 2. The incidence rate of PRBSI was 2.98 per 1,000 PICC days. The median time of PRBSI onset was 42 days (IQR, 13–90). The cumulative incidence of PRBSI was presented in the Kaplan-Meier survival curve in Fig. 1.

Table 2.

Descriptive characteristics of 45 patients with PRBSI

Characteristics	Value
Age, years, median (IQR)	58 (39–67)
Sex, n (%)
Male	27 (60)
Female	18 (40)
Diagnosis, n (%)
AML	26 (57.8)
ALL	4 (8.9)
NHL	8 (17.8)
MDS or MPN	6 (13.3)
MM	1 (2.2)
Charlson Comorbidity Index (CCI), n (%)
0	16 (35.6)
1–2	16 (35.6)
3–4	7 (15.6)
≥ 5	6 (13.3)
ECOG scores, n (%)
0–1	24 (53.3)
2	8 (17.8)
3–4	13 (28.9)
Reasons for PICC insertion, n (%)
Chemotherapy or immunotherapy	33 (73.3)
TPN	5 (11.1)
Difficult venous access	5 (11.1)
Autologous stem cell transplantation	2 (4.4)
Total number of PICC insertions, n (%)
Patients with one PICC insertions	27 (60)
Patients with two PICC insertions	9 (20)
Patients with three or more PICC insertions	9 (20)
Overall 30-day mortality, n (%)	10 (22.2)

ALL: acute lymphoblastic leukemia; AML: acute myeloid leukemia; IQR: inter quartile range; MDS: myelodysplastic neoplasms; MM: multiple myeloma; MPN: myeloproliferative neoplasms; NHL: non-Hodgkin lymphoma; PICC: peripherally inserted central venous catheter; PRBSI: PICC-related bloodstream infection; TPN: total parenteral nutrition

Fig. 1.

Estimated cumulative incidence curve of PRBSI

PRBSI pathogens and antibiotic resistance

Overall, 67 organisms were isolated from blood cultures, including 28 MDR strains (41.8%). With respect to microbiology, the most prevalent pathogens identified were Gram-negative bacilli (GNB) (49.3%) and Gram-positive cocci (GPC) (35.8%), followed by fungi (11.9%) and Gram-positive bacilli (GPB) (3%). Escherichia coli (16.4%), Klebsiella pneumoniae (9%), and Pseudomonas aeruginosa (9%) were the most frequent GNB isolates. Among GPC pathogens, the most frequent isolate was coagulase-negative staphylococci (CoNS). MDR isolates accounted for 50% within GPC and 48.5% within GNB. Detailed information on the microbiology of PRBSIs was presented in Table 3.

Table 3.

Microbiology of PRBSIs

Microorganism	Number of infections, n (%)	MDR strains, n (%)3	PICC removal, n (%)3
Gram-positive cocci	24 (35.8)	12 (50)	5 (20.8)
CoNS	16 (23.9)	10 (62.5)	2 (12.5)
Staphylococcus aureus	4 (6)	2 (50)	3 (75)
Enterococcus faecalis	1 (1.5)	0	0
Other GPC1	3 (4.5)	0	0
Gram-positive bacilli	2 (3)	0	0
Gram-negative bacilli	33 (49.3)	16 (48.5)	11(33.3)
Escherichia coli	11 (16.4)	9 (81.8)	2 (18.2)
Enterobacter cloacae	2 (3)	0	0
Klebsiella pneumoniae	6 (9)	4 (44.4)	4 (66.7)
Pseudomonas aeruginosa	6 (9)	0	1 (16.7)
Acinetobacter baumannii	1 (1.5)	1 (100)	1 (100)
Stenotrophomonas maltophila	1 (1.5)	0	0
Serratia marcescens	3 (4.5)	0	2 (66.7)
Other GNB2	3 (4.5)	2 (66.7)	1 (33.3)
Fungi	8 (11.9)	0	8 (100)
Candida tropicalis	6 (9)	0	6 (100)
Candida parapsilosis	1 (1.5)	0	1 (100)
Candida albicans	1 (1.5)	0	1 (100)
Total	67 (100)	28 (41.8)	24 (35.8)

¹Other GPC: Streptococcus viridans (1), Gemella haemolysans (1), Micrococcus luteus (1); ²Other GNB: Alcaligenes xylosoxidans (1), Moraxella osloensis (1), Fusobacterium nucleatum (1); ³Percentage among their species. CoNS: coagulase-negative staphylococci; GNB: Gram-negative bacilli; GPC: Gram-positive cocci; MDR: multidrug-resistant

Subgroup analysis between patients managed with and without ALT

34 episodes of PRBSI were treated with ALT combined with SAT, while 33 episodes of PRBSI received SAT alone. The antimicrobials used as lock solutions included cefoperazone/sulbactam (n = 7), piperacillin/tazobactam (n = 7), vancomycin (n = 6), biapenem (n = 6), meropenem (n = 3), cefuroxime (n = 2), micafungin (n = 2), and amikacin (n = 1). In 25 out of 34 episodes (73.5%), the same antimicrobial was used in both ALT and SAT (Table 1). The median duration of ALT was 7 days (IQR, 5–9). Among the 34 episodes managed with adjunctive ALT, 8 PICCs required removal; in contrast, among the 33 episodes managed without ALT, 16 PICCs were pulled out. The catheter salvage rate was 76.5% or 51.5% when ALT was added or not (p = 0.033). 3 PRBSI related death events (3/34) compared with 4 death events (4/33) happened in each therapeutic regimen (p = 0.709). In 43 episodes of PICC retention, recurrence rates of PRBSI were 7.7% (2/26) and 5.9% (1/17) in management with and without ALT. PICC related non-infectious complications within 90 days from PRBSI occurrence included occlusion, malfunction and thrombophlebitis; however, no complications or adverse effects was observed during ALT (Table 4).

Table 4.

Comparison of patient groups with and without adjunctive ALT

Characteristics and outcomes	No ALT in therapeutic management(n = 33)	ALT in therapeutic management(n = 34)	P value
Status of underlying disease, n (%)			0.54
Remission	15 (45.5)	18 (52.9)
Non-remission	18 (54.5)	16 (47.1)
PICC salvage, n (%)	17 (51.5)	26 (76.5)	0.033
Days from PICC insertion to PRBSI, median (IQR)	47 (14–84)	40 (12–96)	0.985
WBC (×109/L) at time of PRBSI, median (IQR)	0.34 (0.11–1.05)	0.43 (0.14–1.97)	0.629
Neutrophils (×109/L) at time of PRBSI, median (IQR)	0.03 (0.01–0.29)	0.04 (0.01–0.71)	0.825
Duration of neutropenia, days, median (IQR)	11 (5–15)	13 (0–22)	0.471
PCT at time of PRBSI, ng/ml, median (IQR)	0.89 (0.29–4.05)	1.05 (0.22–6.89)	0.91
CRP at time of PRBSI, mg/L, median (IQR)	169 (79.7–204)	166.75 (46.7–212)	0.783
Multiple PRBSIs, n (%)	11 (33.3)	10 (29.4)	0.729
Prior corticosteroid exposure (30 days), n (%)	7 (21.2)	15 (44.1)	0.046
Prior antibiotic therapy (30 days), n (%)	14 (42.4)	19 (55.9)	0.271
SOFA score, n (%)			0.126
< 2	23 (69.7)	29 (85.3)
≥ 2	10 (30.3)	5 (14.7)
Pathogens, n (%)			0.303
GPB	1 (3)	1 (2.9)
GPC	9 (27.3)	15 (44.1)
GNB	17 (51.5)	16 (47.1)
Fungi	6 (18.2)	2 (5.9)
MDR isolates, n (%)	13 (39.4)	15 (44.1)	0.695
Inadequate empirical therapy, n (%)	15 (45.5)	12 (35.3)	0.397
PRBSI related mortality, n (%)	4 (12.1)	3 (8.8)	0.709
Complications within 90 days from the onset of PRBSI, n (%)	2 (6)	3 (8.8)	> 0.99
Occlusion	1 (3)	2 (5.9)
Thrombophlebitis	0	1 (2.9)
Malfunction	1 (3)	0
Recurrence of PRBSI, n	1	2	> 0.99

ALT: antibiotic lock therapy; CRP: C-reaction protein; GNB: Gram-negative bacilli; GPB: Gram-positive bacilli; GPC: Gram-positive cocci; IQR: inter quartile range; MDR: multidrug-resistant; PCT: procalcitonin; PICC: peripherally inserted central venous catheter; PRBSI: PICC-related bloodstream infection; SOFA: sequential organ failure assessment; WBC: white blood cell

Risk factors of PICC removal

We performed a univariate regression analysis with the following variables: days from PICC insertion to infection, multiple insertions, WBC count, neutrophils count, duration of neutropenia, multiple PRBSIs, prior corticosteroid exposure, prior antibiotic therapy, procalcitonin (PCT) level, C-reaction protein (CRP) level, GNB infection, MDR bacterial infection, adequacy of empirical antibiotic therapy and management with ALT. The results of the univariable analyses were used to identify variables for the multivariable analyses. In the multivariate logistic analysis, a high PCT level > 2ng/ml (OR, 4.687; 95% CI, 1.195–18.39; p = 0.027) and inadequate empirical therapy (OR, 5.509; 95% CI, 1.483–20.473; p = 0.011) were risk factors for PICC removal. On the contrary, management with ALT (OR 0.276; 95% CI, 0.082–0.927; p = 0.037) was a protective factor against it (Table 5).

Table 5.

Univariate and multivariate analysis of risk factors for PICC removal

Factors	Univariate analysis					Multivariate analysis
	OR	95%CI		P value		OR	95%CI		P value
Days from PICC insertion to infection	1.005	0.998	1.013	0.157
Multiple insertions	0.853	0.288	2.529	0.774
WBC count	1.041	0.964	1.125	0.304
Neutrophils count	1.049	0.956	1.153	0.313
Duration of neutropenia	0.987	0.937	1.04	0.629
Multiple PRBSIs	1.154	0.396	3.362	0.793
Prior corticosteroid exposure within 30 days	1.036	0.358	2.994	0.948
Prior antibiotic therapy within 30 days	1.359	0.499	3.701	0.548
PCT > 2ng/ml	3.053	1.075	8.667	0.036		4.687	1.195	18.39	0.027
CRP level	1.006	1	1.012	0.062		1.005	0.998	1.013	0.157
GNB infection	0.808	0.297	2.198	0.676
MDR bacterium infection	1.687	0.614	4.639	0.311
Inadequate empirical therapy	3.231	1.142	9.143	0.027		5.509	1.483	20.473	0.011
ALT in therapeutic management	0.327	0.115	0.93	0.036		0.276	0.082	0.927	0.037

ALT: antibiotic lock therapy; CI: confidence interval; CRP: C-reaction protein; GNB: Gram-negative bacilli; MDR: multidrug-resistant; OR: odds ratio; PCT: procalcitonin; PICC: peripherally inserted central venous catheter; PRBSI: PICC-related bloodstream infection; WBC: white blood cell

Discussion

CRBSI is a critical problem among patients with hematological malignancies, frequently causing treatment delays, lower doses of chemotherapies, and, consequently, suboptimal treatment [19]. Pathogens colonize within the lumen of the catheter, forming the so-called biofilm, in which enables them to resist systemic antimicrobial therapy at standard concentrations [20]. The biofilm is considered to be the main cause of treatment failure and infection recurrence [21]. The lock solution used in ALT contains a high concentration of antibiotics (usually 100 to 1,000 times above the minimum inhibitory concentration of a pathogen) to completely fill the catheter lumen during a certain time, with the aim of directly eradicating microbial biofilms and finally achieving sterilization [17]. This retrospective study investigated the efficacy and safety of ALT in patients with hematological malignancies who had PRBSI, as well as the incidence and causative microorganisms in our department.

We reported a PRBSI incidence rate of 2.98 per 1,000 PICC days. According to other studies, the incidence rate of CRBSI/CLABSI in patients with hematological malignancies ranged from 0.59 to 6.61 per 1,000 PICC days [4, 5, 22–30]. This variation may be attributed to differences in CRBSI/CLABSI definitions, catheter types, PICC insertion methods, management procedures, participants demographics, and blood culture methodologies. It should be highlighted that the definition of catheter related infection varies among these studies and the terms “CRBSI” and “CLABSI” are frequently interchanged. A challenge with the strict CRBSI definition in hematologic populations is that the majority of patients receive intravenous antibiotics both prior to catheter insertion and throughout its usage. As a result, many blood cultures might yield false-negative results. A more recent definition—suspected or probable CRBSI as we used in this manuscript, has been suggested in recent expert consensus-based clinical practice guidelines [12, 31]. Therefore, we argue that a broader definition such as probable CRBSI, despite lowering the specificity, increases the sensitivity and may be better suited to patients with hematological malignancies.

Historically, the etiology of CRBSI has predominantly been GPC (mostly CoNS and S. aureus) [32]. However, A significant decrease in rates of GPC and an increase in rates of GNB and candidemia have been documented in patients with hematological malignancies over time, coinciding with an increasing number of GNB-MDR infections [33–35]. In our study as well, the proportion of PRBSI due to GNB reached as high as 49.3%. The most frequently isolated GNB were E. coli, P. aeruginosa and K. pneumoniae, which comprised up to 69.7% of all GNB isolates. Meanwhile, a considerable proportion of MDR isolates was identified among both GNB and GPC. It is hypothesized that the translocation of GNB from the gastrointestinal tract due to mucosal injury associated with highly intensified chemotherapy may be an explanation for the high rate of GNB related CRBSI in patients with hematological malignancies. Our findings highlight the fact that GNB, including MDR isolates, are common causes of PRBSI.

Numerous studies have proven the efficacy of adjunctive ALT for CRBSI [36]. A previous systematic review found that the combination of systemic antibiotics and culture-guided ALT was superior to systemic antibiotics alone (OR 0.20, 95% CI 0.10–0.39), with 10% of locked patients requiring replacement compared to 33% of subjects without locks [37]. A recent meta-analysis of children showed that the addition of ALT improved successful catheter salvage rates (77% vs. 68%) and reduced CRBSI recurrence (5% vs. 18%) [38]. The overall therapeutic efficacy of ALT combined with systemic treatment was 50–90% depending on different studies and causative microorganisms in most cases [13, 39–42]. The discrepancies in success rates between studies may be due to differences in the definition of treatment failure, treatment protocols of CRBSI and patient characteristics. However, in terms of patients with hematological malignancies, whether the addition of ALT is superior to SAT alone has not been well described. The outcomes of our research showed that ALT in combination with SAT significantly reduced the rate of PICC removal. ALT was also proved to be a protective factor in subsequent logistic analysis. Patients with hematological malignancies are often in great need of central line catheters; thus, incorporating ALT into conventional therapeutic regimen may help them avoid repetitive aspiration and catheterization. Regarding the safety of ALT, there is theoretically a potential for catheter malfunction or systemic exposure to anticoagulants and antibiotics. In this study, we only observed occlusion in two episodes and thrombophlebitis in one episode with ALT. No apparent trend towards increased incidence was observed compared with SAT alone. Our findings further demonstrate the safety of ALT in patients with hematological malignancies.

Accurate prediction of catheter removal through validated risk factors can help to identify those patients who are most likely appropriate for salvage therapy. A 14-year retrospective analysis of pediatric CLABSI found that bone marrow transplant status, neutrophil count and line type were correlated with CVC removal [43]. Results from pediatric patients with cancer and hematological disorders indicated that ALT was associated with fewer catheter removals [44]. Early ALT initiation, dwelling 12–24 h per day and use in permanent catheters might increase the rate of catheter preservation as well [45]. A study conducted by Freire et al. identified ALT at the start of treatment as a protective factor for initial treatment failure, while a high SOFA score at diagnosis of infection and being in palliative care were recognized as risk factors [41]. In compliance with their findings, our results similarly showed that ALT was a protective factor, whereas a high PCT level and inadequate empirical therapy were associated with a high rate of PICC removal. It has been demonstrated that PCT, as a prognostic biomarker for sepsis, has its elevated value associated with adverse outcomes including mortality [46]. A high PCT level often means severe infection which enables the catheter to be removed imperatively. Likewise, inappropriate empirical antibiotic therapy frequently causes treatment delay and increases mortality in patients with febrile neutropenia and hematological malignancies [47]. Consequently, it should be cautious to choose empirical antibiotics and make the decision of ALT application in patients with elevated PCT levels.

Another important finding in our study was that the addition of ALT did not increase the risk of recurrent PRBSI. The reason might be that a high proportion of cases used the same antibiotics in ALT and SAT, implying little impact on inducing resistant isolates. An important characteristic that an ideal lock solution should possess is having a low potential for resistance. The leakage or flushing of lock solution from the lumen is likely to cause systemic exposure to antibiotics [48]. As a result, repeated exposure may induce drug resistance in pathogens. An increased incidence of drug-resistant bacteria following ALT with corresponding antibiotics has been described in previous literature [49, 50]. However, in a treatment modality, this concern will be alleviated if the same systemic antibiotic is also used concurrently as a component of the lock solution [17, 51]. In our study, nearly three-fourths (25/34) of cases received identical antimicrobials in ALT and SAT, yielding a high rate of catheter salvage as well as low rate of recurrence. No drug-resistant bacteria were found following ALT either. In our opinion, the identical agent selected in systemic treatment together with ALT offers a feasible approach for patients with hematological malignancies, in terms of avoiding resistance, facilitating clinical practice and minimizing product waste.

Our study has some limitations. First, there were selection biases of patients due to the retrospective design, which may allow confounders to affect the outcomes and could not be generalized to all populations. Second, the patients in this study were treated at a single center, so it could not truly reflect the epidemiology and etiology of PRBSI across the region. Third, the blood culture system of out lab was semiquantitative, which may influence the accurate diagnosis of PRBSI. In addition, the small sample size with 34 episodes receiving ALT made it impossible to incorporate more variables into multivariate analysis. Nevertheless, this research represents an analysis evaluating the efficacy and safety of ALT for patients with hematological malignancies who experienced PRBSI. Additionally, it provides details on the administration of ALT, which may serve as an optional approach for those in urgent need of PICC retention.

Conclusion

In conclusion, our results support ALT as a catheter salvage strategy for PRBSI in patients with hematological malignancies. Specifically, when used in addition to systemic antibiotics, ALT is safe and efficient, and it is associated with an increased chance of PICC salvage in the majority of cases. Additional high-quality prospective trials are needed to compare ALT with conventional antibiotic therapy and to identify the factors that affect ALT outcomes.

Author contributions

Qin Zhang collected and analyzed the data, and wrote the original draft of the manuscript. Yujia Huo and Chengfei Li were responsible for data collection, statistical analysis and visualization. Qinggang Sun contributed to patient care and management. Xi Xi and Rui Sun contributed to data collection and patient follow-up. Qingju Sun and Meijuan Jiang conducted laboratory testing and verification. Guang Li designed and supervised this study and guided the article revision. All authors contributed to the manuscript and approved the submitted version.

Funding

No funding was received to assist with the preparation of this manuscript.

Data availability

The datasets that support the findings of this study are available on reasonable request from the corresponding author. The data are not publicly available due to privacy restrictions.

Declarations

Ethics approval and consent to participate

This research was in accordance with the Declaration of Helsinki, and was approved by the Ethics Committee of No. 971 Hospital of Navy (971LL-2025031601). Informed consent was obtained from patients or their relatives.

Competing interests

The authors declare no competing interests.