Subclavian central venous catheterization in critically ill children in Türkiye: a safe and feasible approach in a single-center experience Authors: Abdulrahman Özel, Esma Nur Özen, Servet Yüce, Sude Menekşe, Hazal Erdem, Merve Boyraz Published: 2026 ABSTRACT BackgroundCentral venous catheter (CVC) insertion is essential procedure in the management of critically ill pediatric patients. However, the optimal catheter insertion site remains controversial due to varying complication rates. This study aimed to evaluate the feasibility and safety of subclavian vein (SCV) catheterization compared to femoral vein (FV) and internal jugular vein (IJV) access in a pediatric intensive care setting.MethodsThis retrospective study included all patients admitted to the pediatric intensive care unit (PICU) of a tertiary hospital over a 1-year period (March 2022–March 2023) who underwent nontunneled CVC placement.ResultsAmong 174 patients included (median age 29 months, 60.3% male), 52.9% had SCV, 23.6% FV, and 23.6% IJV catheterization. The overall CVC success rate was 99.4%. There was no statistically significant difference in mechanical complications, catheter-related bloodstream infections, or thrombosis between groups; however, thrombosis occurred most frequently in the FV group (12.2%). The median age in the SCV group was significantly lower than in the other groups (20 months; interquartile range, 10–63; P<0.001), while the presence of chronic diseases was significantly higher in this group (P=0.002). In multivariate logistic regression, catheter site was identified as the only independent predictor of thrombosis, with FV catheterization associated with increased risk (odds ratio, 2.75; P=0.016).ConclusionsSCV catheterization appears to be a feasible and safe option for critically ill children, even in younger patients and in the presence of chronic comorbidities. Despite potential technical challenges, SCV access represents a viable alternative to IJV and may offer potential advantages over FV, particularly with regard to thrombosis risk. INTRODUCTION Central venous catheter (CVC) insertion is a vital procedure in the management of critically ill children, used for invasive hemodynamic monitoring, renal replacement therapy, plasma exchange, intravenous infusions (e.g., inotropes, hyperosmolar therapy, total parenteral nutrition, blood products), and blood sampling. In many cases, it becomes essential due to the inability to maintain reliable peripheral venous access during ongoing treatment in pediatric intensive care units (PICUs) [1]. There are currently no universally accepted guidelines regarding the optimal venous access site for pediatric patients. Although various studies have reported complications associated with different catheterization sites, these findings are largely influenced by the experience of individual centers. The subclavian vein (SCV), while considered safe for long-term use, has shown a wide range of complication rates in children, varying from as low as 1% to as high as 10% [2]. Compared to femoral vein (FV) access, SCV CVCs are less likely to result in thrombosis. However, the risk of pneumothorax is relatively higher than with internal jugular vein (IJV) access [3]. Other studies have found no significant differences in complication rates between subclavian and jugular approaches [4]. Despite potential technical difficulties and mechanical complications, SCV CVCs are generally well tolerated by critical patients [5,6]. Their anatomical location allows for secure fixation, which reduces the risk of accidental dislodgement and does not restrict patient mobility. Furthermore, the subclavian region is relatively isolated from bodily secretions, which may reduce the risk of catheter-related infections [6]. In hemodynamically unstable patients, the SCV is less prone to collapse compared to other veins, offering another clinical advantage [7]. The aim of this study was to evaluate the feasibility and complication profile of SCV CVC in critically ill pediatric patients, in comparison with FV and IJV approaches. We also sought to identify factors associated with catheter-related complications. MATERIALS AND METHODS This study was approved by the Institutional Ethics Committee of Bağcılar Training and Research Hospital (No. 2023/08/01/036, Date: August 8, 2023) and conducted in accordance with the Declaration of Helsinki. Due to the retrospective nature of the study, informed consent was not required. This retrospective study included all pediatric patients who were admitted to the 8-bed tertiary-level PICU of our hospital over a 1-year period (March 2022–March 2023) and who received a non-tunneled CVC. Patients were eligible if they stayed in the PICU for more than 24 hours and had a CVC placed during their admission. Only the first catheter inserted during the PICU stay was included for each patient; subsequent catheters inserted for any reason or via a different anatomical site were excluded from analysis. Demographic data (age, sex), presence of acute and chronic illnesses, number of involved organ systems, CVC insertion site, catheter dwell time, duration of PICU stay, duration of invasive mechanical ventilation, occurrence of catheter-related thrombosis, administered therapies, and Pediatric Risk of Mortality (PRISM) III scores were collected. Administered therapies included 3% hypertonic saline, antibiotic therapy, albumin, fresh frozen plasma, packed red blood cells, platelets, intravenous immunoglobulin, and inotropic agents. The study included catheters inserted via the FV, SCV, and IJV. The SCV catheterizations were performed using anatomical landmarks, whereas ultrasound guidance was used for FV and IJV insertions. The IJV was identified lateral to the carotid artery and the FV medial to the femoral artery, with the ultrasound probe held in a transverse position. Sedation and analgesia were provided with fentanyl (1 µg/kg), midazolam (0.1 mg/kg), or ketamine (1–2 mg/kg), and prilocaine was used for local anesthesia if necessary. Patients were continuously monitored for cardiovascular and respiratory parameters during the procedure. All catheterizations were performed by a general pediatrician with over 4 years of experience in critical care and more than 150 catheter placements per year. Due to the retrospective design of the study and the absence of a standardized institutional protocol, the choice of CVC insertion site was based on the treating clinician’s discretion. Strict aseptic technique and hand hygiene protocols were followed for all procedures. The insertion site was sterilized with 10% povidone-iodine. For FV and IJV catheterizations, a handheld ultrasound device with a 7.5 MHz linear probe (Vscan Extend, GE Healthcare) was used. Catheters were placed using the Seldinger technique. Catheter tip position was confirmed by posteroanterior chest x-ray for SCV and IJV access, with optimal positioning defined as the tip located between the distal superior vena cava and the right atrial junction. Right-sided insertion was preferred in all cases. Catheter size was selected according to body weight: 4-Fr double-lumen (Beijing Target Medical) for patients <5 kg; 5-Fr double-lumen for 5–20 kg; and 7-Fr triple-lumen (Guangdong Baihe Medical Technology) for those >20 kg. According to unit protocol, routine catheter care included flushing twice daily with a mixture of 1 mL unfractionated heparin diluted in 0.9% saline, and dressing changes every two days. Mechanical complications were identified based on routine post-procedural imaging, including chest radiography performed after SCV and IJV catheterization to confirm catheter position and detect complications such as pneumothorax or malposition. Ultrasonography was performed when patients exhibited signs of catheter dysfunction, swelling, redness, or edema near the catheter site. Thrombosis was defined as visualized thrombus or absence of intraluminal flow. Catheter-related bloodstream infection (CRBSI) was diagnosed when the same microorganism was isolated from both the catheter and a peripheral blood culture in the presence of clinical and laboratory signs of sepsis. RESULTS During the study period, a total of 202 CVC placements were performed in the PICU. After excluding 27 repeated CVCs (as only the first insertion per patient was included), 175 patients underwent initial CVC placement at their admission. The procedure was unsuccessful in only one case, which was subsequently managed via transhepatic venous catheterization under interventional radiology guidance. Accordingly, the overall success rate of CVC placement was 99.4% (174/175). Notably, all catheters were inserted on the right side. A total of 174 patients were included in the final analysis (Figure 1). Of these, 60.3% (105/174) were male, and the median age was 29 months (interquartile range [IQR], 13–108 months). The catheter sites were the SCV in 92 patients (52.9%), FV in 41 patients (23.6%), and IJV in 41 patients (23.6%). The median catheter dwell time was 8 days (IQR, 4–15 days) (Table 1). Multisystem involvement was present in 67.2% (117/174) of patients, and chronic diseases were documented in 43.1% (75/174). Neurological disorders were the most common chronic condition, affecting 25.3% (44/174) of patients. Chronic illness was most frequently observed in the SCV group (52.2%, 48/92), which was statistically significant (P=0.002). Among acute conditions, sepsis and pneumonia were the most common diagnoses (each 22.4%, 39/174), followed by trauma (16.7%, 29/174) and pediatric acute respiratory distress syndrome (6.3%, 11/174). While Table 2 presents the primary acute diagnoses, multiple acute conditions were identified in 13.8% (24/174) of patients, with sepsis (58.3%, 14/24) and post-cardiopulmonary resuscitation events (33.3%, 8/24) be­ing the most frequent combinations. The most common indication for CVC insertion was the need for multiple infusions (61.5%, 107/174), followed by lack of peripheral venous access (36.2%, 63/174). In the FV group, multiple infusions were the most frequent indication (87.8%, 36/41), whereas in the SCV group, the same indication accounted for 47.8% (44/92) of cases (P<0.001). When reasons for CVC removal were compared across CVC sites, the most frequent reason was that the CVC was no longer needed. No statistically significant difference was observed between groups. Catheter-related thrombosis occurred in 12.2% of patients with FV catheters, 7.3% with IJV catheters, and 2.2% with SCV catheters, with no significant difference between groups (P=0.064). Similarly, the incidence of CRBSIs was comparable across groups (P=0.979). Mechanical complications were not significantly different between groups. The most frequent mechanical complication was catheter malposition (P=0.069). Comparisons of catheter site and illness severity revealed no statistically significant differences in PRISM III scores or the presence of organ dysfunction. However, invasive mechanical ventilation was more frequently required in the IJV group (80.5%) and less commonly in the FV group (53.7%) (P=0.011) (Table 3). A multivariate binary logistic regression analysis was performed to identify independent predictors of catheter-related thrombosis. Catheter location emerged as a statistically significant factor. Patients with FV catheters had a 2.75-fold higher risk of developing thrombosis compared to other sites (odds ratio [OR], 2.745; 95% CI, 1.202–6.264; P=0.016). Although not statistically significant, transfusion of any blood product was associated with a trend toward reduced thrombosis risk (OR, 0.211; 95% CI, 0.041–1.084; P=0.062). Other variables, such as CVC dwell time, organ involvement, and use of inotropic agents, were not significantly associated with catheter-related thrombosis in this model (Table 4). DISCUSSION SCV catheterization may present anatomical and technical challenges that deter clinicians from selecting this route, particularly in pediatric patients. Reported success rates for SCV catheterization using the landmark method in children vary between 87.4% and 100% in the literature [8,9]. In our study, the success rate was 99.4%, with only one unsuccessful attempt. This case involved a patient with multiple prior PICU admissions and repeated CVC insertions, which likely contributed to the difficulty encountered. Improved understanding of pediatric critical care and advances in imaging technology have contributed to increased success rates in central venous access procedures. In its 2011 guidelines, the Centers for Disease Control and Prevention recommended the use of ultrasound guidance during catheter insertion to reduce the number of attempts and procedural complications [10]. While ultrasound guidance is more commonly used for IJV and FV access in critically ill children, it is less frequently employed for SCV catheterization due to anatomical constraints. Nevertheless, studies investigating ultrasound-guided SCV catheter insertion in pediatric patients have been reported in the literature [6]. In our cohort, all SCV catheterizations were performed using anatomical landmarks without the aid of ultrasound guidance. Further randomized controlled trials are needed to determine whether ultrasound guidance can improve the safety and success rates of SCV catheterization in the pediatric population. Alongside physician experience, patient-specific factors play a significant role in determining the safest and most sustainable vascular access [11]. In our study population, the most common acute diagnoses were sepsis and pneumonia (each 22.4%). Chronic illnesses were present in 43.1% of patients, with neurological diseases being the most frequent (25.3%). These findings are consistent with previous studies. For example, Erdoğan et al. [12] identified neurological disorders and sepsis as the most common acute diagnoses, while Jarraya et al. [2] reported that 76.1% of patients presented with active infection or high fever, with bronchiolitis being the leading acute diagnosis; additionally, 86.6% of their patients had underlying comorbidities. CVCs are often required in critically ill pediatric patients for various reasons, including hemodynamic monitoring, blood sampling, administration of medications, securing access when peripheral veins are unavailable, and extracorporeal therapies [13]. In our study, the most common indication for CVC was the need for multiple infusions (61.5%), followed by difficulty in obtaining peripheral venous access (36.2%), and less frequently, total parenteral nutrition (2.3%). This is consistent with other studies, which have reported the most common indications as inability to obtain peripheral access [2], long-term venous access [9], total parenteral nutrition [12] and multiple infusions [14]. With regard to mechanical complications, our cohort showed no statistically significant differences between catheter insertion sites. Malposition was the most frequently encountered complication. While some studies in the literature have reported varying rates of mechanical complications among FV, SCV, and IJV catheterizations [6,15], others have found no significant association between the insertion site and complication rate [4]. Overall, however, SCV catheterization is generally considered to carry a lower risk of thrombosis but a higher risk of pneumothorax compared to IJV or FV access [3]. These discrepancies are likely influenced by operator experience and the single-center design of many studies. A closer look at previous reports reveals that most studies disproportionately represent one or two catheter sites, with limited data on the third site, thereby restricting generalizability. To obtain more robust evidence, future multicenter randomized controlled trials and systematic reviews with meta-analyses are warranted. In our cohort, the most frequent mechanical complications in the SCV group were malposition (7.6%) and pneumothorax (3.3%). Although catheter tip malposition—such as migration to the contralateral SCV or jugular vein—can have serious implications, this issue can generally be corrected by repositioning the catheter [16]. In our study, of the seven malpositioned SCV catheters, five were found to have migrated into the IJV and two into the contralateral SCV; all were successfully revised without further complications. According to the study by Jarraya et al. [2], which included 134 pediatric SCV catheterizations, the most common mechanical complications were bleeding and hematoma (7.5%), with no cases of pneumothorax reported. Vafek et al. [9] also found malposition (9.1%) to be the most common complication in 55 SCV catheterizations, while Isguder et al. [17] reported a pneumothorax rate of 3.3% in 105 SCV insertions. During SCV catheterization, using ultrasound to visualize the guide tip in the superior vena cava from the supraclavicular fossa may reduce the risk of malposition [18]. Although no statistically significant differences were observed between catheter insertion sites in terms of overall mechanical complications, clinically important events such as malposition and pneumothorax were numerically more frequent in the SCV group. Pneumothorax, in particular, should not be considered a minor complication in pediatric patients, especially in younger age groups. While these events did not result in severe adverse outcomes in our cohort, their potential clinical impact warrants careful consideration. Moreover, the relatively small sample size and low number of complication events may have limited the statistical power to detect true differences between groups. Therefore, these findings should be interpreted with caution. Arterial injury is another well-recognized complication of CVC placement, particularly in critically ill patients, and the use of ultrasound has been shown to reduce the incidence of vascular injury [15,19]. In our study, arterial injury was the second most common mechanical complication. While this complication is generally more frequent with FV catheterization and least common with SCV access [19], bleeding from the carotid or femoral arteries can often be managed with compression. In contrast, hemorrhage from the subclavian artery is anatomically more difficult to control due to its location beneath the clavicle. In our cohort, the frequency of arterial injury did not significantly differ between catheter sites. This finding may be explained by the use of ultrasound guidance during IJV and FV catheterizations, which likely reduced the risk of vascular injury. Furthermore, the SCV group included younger patients with smaller vessel diameters, limiting the available anatomical space for catheterization and potentially increasing the risk of arterial trauma. In the literature, SCV catheterization has been identified as a relatively safe option regarding CRBSIs, whereas FV access is considered the highest-risk site [3,14,16]. In our study, the incidence of CRBSI did not significantly differ between catheter locations, consistent with some previous reports [4,20]. Several factors may account for this, including the relatively smaller number of FV and IJV cases compared to SCV CVC insertions, and differences in baseline patient characteristics across groups. Despite the SCV group including younger patients with more comorbidities, the infection rate was not significantly higher, suggesting that SCV access may not be at a disadvantage in this regard. Another important contributor to the development of this complication may be suboptimal adherence to catheter care protocols following insertion. In our unit, nursing staff regularly receive training on central line maintenance in an effort to reduce CRBSI rates. The overall rate of catheter-related thrombosis in our cohort was 5.7%, consistent with previously reported rates ranging from 2% to 26%. Although the difference between groups was not statistically significant, half of the thrombosis cases occurred in patients with FV catheters. Previous studies have similarly reported a higher incidence of thrombosis associated with FV access [3,14]. In our logistic regression model evaluating thrombosis risk factors, SCV emerged as the safest option, while FV posed the greatest risk. The administration of infusions through the catheter was not associated with increased thrombosis risk. However, our relatively small sample size and the uneven distribution of acute and chronic illnesses limited our ability to assess clinical predictors of catheter-related complications. In hospitalized pediatric patients, CVC, systemic infections, intensive care admission, mechanical ventilation, and prolonged hospital stay are known to be the most significant risk factors for venous thromboembolism [21]. Many previous studies have underrepresented SCV catheterizations, making it difficult to draw definitive conclusions about the relative risks. In our unit, suspected catheter-related thrombosis is investigated using Doppler ultrasonography in cases of limb edema, impaired perfusion, or catheter dysfunction (e.g., inability to aspirate blood or resistance to infusion). Early diagnosis, prompt catheter removal, and the initiation of anticoagulant or thrombolytic therapy help mitigate adverse outcomes associated with thrombosis. This study has several limitations. It was conducted at a single center with a retrospective design and a relatively small sample size. In addition, catheter site selection was based on clinician preference without standardized criteria, which may introduce selection bias. There was also a methodological asymmetry between groups, as ultrasound guidance was used for IJV and FV catheterizations but not for SCV access, potentially influencing complication rates. Furthermore, all procedures were performed by a single experienced operator, which may limit the generalizability of the findings to settings with different levels of operator expertise. These factors should be considered when interpreting the results. In conclusion, SCV catheterization in critically ill pediatric patients appears to be a feasible and reliable approach with a high success rate. When compared with other central venous access sites, SCV catheterization demonstrated comparable rates of mechanical and long-term complications. While it may offer potential advantages in selected clinical scenarios, particularly regarding thrombosis risk, these findings should be interpreted cautiously given the study limitations. SCV access should therefore be considered a viable alternative option, with the choice of catheterization site guided by clinical context and operator experience. KEY MESSAGES ▪ Subclavian vein (SCV) catheterization is a feasible and reliable option in critically ill pediatric patients, with a high success rate and acceptable complication profile. ▪ SCV catheterization may be considered a valuable alternative approach, particularly in selected patients depending on clinical context and operator experience. ▪ Although femoral vein catheterization provides easy vascular access, it carries a higher risk of catheter-related thrombosis during prolonged use, and should be approached with caution in long-term management.