Paediatric percutaneous catheters have improved the care of children requiring intravenous lines for prolonged therapy (Marcy, 2008). However, patients with cancer who have had a history of central venous access and/or chemotherapy may have an increased risk of difficult cannulation with high rates of catheter-related complications, as they often present with thrombocytopenia, aplasia, and immunosuppression (Barrera et al, 1996). However, critically ill children with multivisceral deficiencies may be at a higher risk of catheter-related morbidity (Aftab et al, 2021). A high rate of nosocomial bloodstream infections was previously reported in our country (Jaballah et al, 2007).
The aim of this study was to compare the difficulty and the morbidity related to the left-sided insertion of percutaneous central subclavian venous catheters in children with cancer with critically ill paediatric patients and to investigate risk factors for complications.
Materials and methods
After obtaining local ethics committee approval and parents' informed consent, a prospective observational study was conducted. We collected data from infants and children referred to our institution's paediatric surgery department for percutaneous infraclavicular subclavian vein catheterization under general anesthesia in the period lasting from July 2021 to July 2022, with a follow-up visit one month later.
In this study, we included all patients aged 3 months to 14 years admitted to the paediatric surgery department for infraclavicular subclavian central vein catheter insertion. We excluded cases of venous catheter insertion failure and patients who died before the removal of the catheter.
The variables included demographic parameters (age, age <1 year, weight, weight <10 kg, comorbidities, ASA class, history of central venous access, and context of catheter insertion). We assessed the difficulty of catheter insertion by evaluating the duration of the insertion procedure, the number of punctures, and the occurrence of an accidental arterial puncture. We also assessed the catheter-related complications (infection, bleeding, pneumothorax, vein thrombosis, and mechanical complications like dislodgement, and malposition). The duration of a catheter was defined by the length of the period from the catheter's insertion to its removal.
All patients had the same protocol of central catheter insertion. All catheters were inserted in the paediatric surgery operating room under general anesthesia by the same team of experienced anesthetists (more than five years of experience in the field of paediatric anesthesia), using an ultrasound-guided technique for infraclavicular subclavian vein catheterization.
In the event of failure of the procedure, a Broviac tunneled catheter or peripherally inserted central catheter (PICC) line was inserted surgically on the left side also.
Patients were divided into two groups: Group 1 included children with malignancies (hematology or oncology patients); and Group 2 included critically ill paediatric patients without malignancies (paediatric critical care unit).
All statistical analyses were achieved using the SPSSv23.0 statistical package. The comparison between the two groups was achieved by the Student's t-test for continuous variables and the chi square test for categorical variables. The Fisher exact test was used when the chi square test was not applicable. A univariate and multivariate logistic regression was used to investigate the risk factors for complications and to seek an association between cancer and catheter-related complications. The significance level was set at P<0.05.
Results
Of 67 patients admitted for subclavian venous catheter insertion, two patients were excluded. We excluded one patient for insertion failure in a four-month-premature boy and a second patient who died while being operated on for intestinal volvulus. The success rate of central catheter insertion was 98.5%. Twenty-eight patients were enrolled in Group 1 (19 with haematologic malignancies and nine with solid tumours) and 37 in Group 2 (without cancer).
Demographic parameters were comparable in both groups (Table 1). The procedure of ultrasound-guided infraclavicular subclavian vein catheterization was more difficult in children with cancer (Table 2). The incidence of complications was 46.4% in Group 1 versus 21.6% in Group 2 (P=0.032). The children with cancer presented more catheter-related infections and more bleeding events with P=0.035 and P=0.03, respectively (Table 3). However, the catheters' lifetime was higher in Group 1. In this study, we noted no complications in the removal of the catheters.
Table 1. Demographic parameters
| Group 1 With cancer n=28 | Group 2 Critically-ill n=7 | P value | |
|---|---|---|---|
| Age (years) | 4.14 ± 4.1 | 2.87 ±3.6 | 0.196 |
| Age <1 year | 11 | 12 | 0.377 |
| Weight (kg) | 18.3± 12 | 13.7±10 | 0.110 |
| Weight <10 kg | 11 | 16 | 0.474 |
| With comorbidities | 25 | 31 | 0.398 |
| ASA Class (I/II/III/IV) | 3/12/13/0 | 6/21/10/0 | 0.266 |
| History of central venous catheterization | 14 | 15 | 0.306 |
| Emergency | 7 | 8 | 0.488 |
Table 2. Difficulty with infraclavicular subclavian venous catheterization
| Group 1 with cancer n=28 | Group 2 critically-ill n=37 | P value | |
|---|---|---|---|
| Duration of the procedure (minutes) | 17.2±4.7 | 12.5±3.9 | <0.001 |
| Number of attempts | 2.07 ± 0.9 | 1.29 ± 0.5 | <0.001 |
| Single shot | 10 | 27 | 0.003 |
| Accidental arterial puncture | 3 | 1 | 0.209 |
| Thrombocytopenia <50000 | 13 | 0 | <0.001 |
Table 3. Complications of infraclavicular subclavian venous catheterization
| Group 1 with cancer n=28 | Group 2 critically-ill n=37 | P value | |
|---|---|---|---|
| Complicated catheters | 13 (46.4%) | 8 (21.6%) | 0.032 |
| Catheter-related infection | 10 | 6 | 0.035 |
| Pneumothorax, air embolism, chylothorax | 0 | 0 | - |
| Bleeding and or haematoma | 4 | 0 | 0.03 |
| Mechanical complications | 1 | 2 | 0.605 |
| Vein thrombosis | 2 | 0 | 0.104 |
| Duration of the catheter (days) | 14.7 ±8.8 | 8.6 ± 4.7 | 0.001 |
| Removed when not needed | 19 | 28 | 0.337 |
Predictors of catheter-related complications were severe thrombocytopenia (aOR=5.29; 95%CI: 0.69–40.6), comorbidities (aOR=4.8; 95%CI: 0.44–53.3), and haematology/oncology patients (aOR=2.95; 95%CI: 0.63–13.8; Table 4).
Table 4. Predictors of catheter-related complications
| OR [95% CI] | aOR [95% CI] | P value | |
|---|---|---|---|
| Age <1 year | 1.6 [0.55–4.69] | 0.60 [0.07–5.2] | 0.643 |
| Weight <10kg | 1.44 [0.50–4.12] | 1.65 [0.17–15.5] | 0.493 |
| Thrombocytopenia <50000 | 5.6 [1.4–20.9] | 5.29 [0.69–40.6] | 0.027 |
| Emergency | 2.25 [0.686–7.37] | 1.95 [0.35–10.6] | 0.181 |
| Comorbidities: ASA class > I | 4.4 [0.51–38.1] | 4.8 [0.44–53.3] | 0.050 |
| Haematology/oncology | 3.14 [1.06–9.24] | 2.95 [0.63–13.8] | 0.034 |
Discussion
In this study, we showed that children with cancer can have a more difficult central venous catheter (CVC) insertion procedure, as the time taken for the procedure and the number of attempts were higher in these patients. Central venous catheter complications were present 46.4% of the time in children with cancer compared with 21.6% in critically ill children (P=0.032). Furthermore, malignancies were associated with an increased risk of complications (aOR=2.95; 95%CI: 0.63–13.8).
The high rates of catheter-related complications in children with cancer are not surprising, given that cancer will naturally make children more susceptible to complications.
However, the lifetime of the catheter was higher in this population, which seems to be related to patients with cancer needing these catheters for a longer term compared with those in critical care settings. The incidence of complications in our population was higher than in the literature (van den Bosch et al, 2022). This may be explained by nurses' experience, nursing skills, and the quality of health care in developing countries (Jarraya et al, 2014). Despite new guidelines and improved anaesthetist skills in using ultrasound guided techniques (Crocoli et al, 2022; Gualtieri et al, 1995), the difficulty of percutaneous CVC insertion has been reported in the paediatric population, particularly in children with onco-haematological conditions. Evaluating this difficulty using objective parameters, such as the duration of the procedure and the number of attempts, can have a clinical implication if some precautions are taken before CVC insertion in children with cancer. The use of ultrasound by experienced paediatric anaesthetists explains the high success rate in this study (de Souza et al, 2018). Parenteral nutrition in children with cancer has been implicated in catheter-related infections and occlusions (Shenep et al, 2017). Moreover, chemotherapy can lead to severe thrombopenia and leucopenia, which can increase the risk of bleeding, haematoma, and infections (Yacobovich et al, 2015; Viana Taveira et al, 2017). We also believe that difficult central venous access can result in multiple punctures with the added risk of asepsis and non-compliance with guidelines, which can be directly implicated in catheter-related bloodstream infections. (Bishop et al, 2007). We should mention that some severe and rare adverse events can occur during the removal of the catheter. Previous studies reported massive hemothorax (Collini et al, 2002), air embolism (Brockmeyer et al, 2009), and severe hemorrhage during subclavian venous catheter removal (Paskin et al, 1974).
Predictors of percutaneous central catheter-related complications are very interesting, as they can help physicians in selecting risky patients who should be managed by an interprofessional team approach. In the literature, several studies have reported the risk factors for PICC line complications in children with cancer (Doganis et al, 2013), but there is little data about risk factors related to complicated percutaneous subclavian venous catheters in paediatric population.
Limitations
The main limitation of this study is that we assessed the risk of complications related to the percutaneous central catheters, which are primarily used for critical care patients, not for haematology and oncology patients requiring prolonged intravenous treatments, and we did not compare them with other devices, such as Broviac catheters, portacaths (Zouari et al, 2018), and PICC lines, commonly used in paediatric populations suffering from cancer (Beck et al, 2019; Kammoun et al, 2022). We also should mention that the risk related to anaesthesia (inhalation sedation with Sevoflurane, mandatory for the insertion of the catheter in children) is not negligible (Jarraya et al, 2021).
Conclusion
The paediatric population is known to have a high risk of difficult venous access, particularly in children with oncohaematological conditions requiring prolonged intravenous therapy. However, there is an increased risk of complications among children with cancer. It seems that this population can be at risk for venous catheter-related morbidity and needs specific management based on experienced anaesthetists and skilled nursing care. Finally, we suggest specific and standardized written protocols for the management of difficult venous access to improve the quality of health care for these children.