Acute lymphoblastic leukemia (ALL) is the most common cause of cancer in children, with a current 5-year overall survival rate estimated at >85%.¹ Because of the nature of the disease and intensive chemotherapy protocols, patients with ALL require frequent needle sticks for blood draws and intravenous access. In the last 30 years, central venous lines (CVL) have been increasingly used to manage children with ALL.² CVL provides a safe administration of intensive chemotherapy, antibiotics, blood products, and blood sampling in children with ALL. On the other hand, CVL can cause various adverse events like infections, deep venous thrombosis (DVT), and mechanical complications.^2–4

There is a clinical variation in the optimal time for the insertion of CVL in patients with ALL. Some clinicians insert the CVL at the beginning of the treatment to comfort having CVL during the remission induction. At the same time, some clinicians wait until the end of induction, the time of remission, or recovery of neutropenia to decrease the risk of CVL-related complications.^5–7 The optimal timing of CVL placement in children with ALL has been investigated in different studies. Most of the studies reported no difference in the rates of infection or thrombosis between early and late CVL placement.⁷ However, in 1 of the biggest studies performed by the Pediatric Oncology Group, 362 patients were analyzed retrospectively. They have found that the placement of CVL within the first 15 days of induction was associated with an increased risk of infectious complications.⁶

This study aimed to investigate the frequency of CVL-related complications in children with ALL according to the time of CVL insertion during induction therapy. We, therefore, hypothesized that CVLs placed during the first 15 days of induction were more likely to become infected than those inserted subsequently.

PATIENTS AND METHODS

This study was performed retrospectively in our pediatric hematology center. Fifty-two ALL patients newly diagnosed and treated according to ALL-BFM 95 or ALL-BFM 2000 study protocols between January 2006 to January 2016 with CVL were included. Data at the time of the first CVL insertion during induction chemotherapy were evaluated from the patients’ medical records. Patients’ demographics, hemoglobin level, white blood cell count (WBC), absolute neutrophil count (ANC), CVL type, time of CVL insertion, complications of CVL (infections, thrombosis, and mechanical complications), time and cause of CVL removal, type of infection, and culture results were noted. Complications developed in the first month after the first CVL insertion were defined as early complications, and late complications were evaluated until the time of removal or death of the patient with a CVL. Development of DVT in these patients was observed from the first CVL insertion to the last follow-up until January 2021. Thrombophilia testing for genetic risk factors was analyzed in patients who developed DVT.

CVLs were inserted under general anesthesia in the operating room mainly by the same surgeon. All manipulations were applied with sterile techniques. External double-lumen catheters were preferred for high-risk patients who needed intensive care or who may need bone marrow transplantation; otherwise, internal port catheters were used. We are not using PICC in our clinic. All CVLs were managed according to the international standards; a bio-occlusive dressing was applied to the insertion area and changed every third day. At the same time, the skin at the insertion area was cleaned with chlorhexidine. Perioperative prophylactic antibiotics were not used during catheter insertion.

Many of the patients with ALL are febrile at presentation, and neutropenia that occurs within a few days of starting therapy persists for 2 to 3 weeks. Our hypothesis based on clinical experience was that CVLs placed during the first 2 weeks are more likely to become infected than those inserted subsequently. Also, we reviewed the previous studies about the timing of catheter insertions. According to these results and our clinical impression, we selected the cut-off point as day 15. CVL placement before or on treatment day 15 was defined as “early insertion”, and after treatment day 15 was defined as “late insertion”.

Infections were classified as bloodstream infection and local infection. Catheter-related bloodstream infection was diagnosed when 1 of the following criteria was positive: (1) Growth of the same pathogen from blood culture of the peripheral vein and culture of CVL, (2) Growth of skin contaminant in 2 or more blood cultures, (3) Growth of skin contaminant in 1 blood culture with symptoms of fever (>38°C;), chills, and hypotension.⁸ Empirical first-line antibiotic therapy consisted of meropenem or piperacillin-tazobactam therapy. Antibiotic therapy was adjusted according to the patient’s culture results and clinical response. Catheter lock therapy was given if a catheter-associated infection was suspected.

Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS software version 21). Descriptive analyses were presented using medians, means, and SD. The proportions of patients with early CVL placement and late CVL placement experiencing infections and other complications were analyzed using the χ² test. Student T test was used to compare the total catheter days in these groups. A P value of <0.05 was considered statistically significant.

RESULTS

The clinical properties of the patients are listed in Table 1. Two patients developed pneumothorax after catheter insertion in the operating room, but they regressed spontaneously. No other surgical complications occurred.