Thrombosis Rate and Risk Factors in Pediatric Patients with Peripherally Inserted Central Catheters Authors: Ahmed A. Alabdulkarim, Abdullah I. Almojali, Tharaa S. Alhowaish, Dena S. Alsuhaibani, Norah I. Alshafi, Amin I. Alzahrani, Mohammed O. Alotaibi, Yousef A. Alghamdi, Hamad A. Alkhalaf Published: 2026 ABSTRACT SummaryObjective:To determine symptomatic thrombosis incidence in pediatric patients with peripherally inserted central catheters (PICCs) and identify associated patient and procedure-related risk factors in a tertiary pediatric center in Saudi Arabia.Methods:We conducted a retrospective cohort study at King Abdullah Specialized Children's Hospital, Riyadh, Kingdom of Saudi Arabia, including patients aged 0–14 years who underwent PICC insertion in the Interventional Radiology department between January 2019 and June 2023. Data on demographics, comorbidities, procedural details, and outcomes were extracted from the electronic medical record system (BestCare). Symptomatic thrombosis was defined as Doppler-confirmed thrombosis with clinical signs. Associations were tested using chi-square or Fisher's exact tests (p <0.05).Results:Among 168 patients, the incidence of symptomatic thrombosis was 16.7% (n = 28). In univariate analyses, female sex (p = 0.019), left-sided insertion (36.4% versus 13.7%, p = 0.014), nonbasilic vein access (53.8%, p = 0.001), and previous PICC exposure (p = 0.014) were associated with symptomatic thrombosis. A nonsignificant U-shaped trend was observed for indwelling time (<7 days, 35.7%; >180 days, 40.9%).Conclusion:The PICC-related thrombosis is a clinically important complication in pediatric patients. Several patient and procedure-related factors were associated with symptomatic thrombosis in univariate analyses. These findings may help inform future prospective studies and strategies aimed at optimizing vascular access practices in children. Introduction Peripherally inserted central catheters (PICCs) have become essential tools in pediatric medicine, providing reliable intermediate and long-term vascular access for patients requiring extended intravenous therapy. Their versatility allows the delivery of antibiotics, chemotherapy, parenteral nutrition, and other critical infusions without repeated peripheral cannulation [12]. In recent years, PICC use has increased globally because of advantages such as bedside placement under ultrasound guidance, reduced need for general anesthesia, lower insertion-related complication rates than centrally placed catheters, and the potential to facilitate earlier hospital discharge [123]. Despite these benefits, PICCs are associated with significant risks. Complications include mechanical issues such as occlusion or malpositioning, infectious complications such as central line-associated bloodstream infection, and thrombotic events [145]. Among these, thrombosis is of particular concern because it can impair catheter function, delay treatment, necessitate catheter removal, and, in severe cases, lead to pulmonary embolism or post-thrombotic syndrome [16]. The reported rates of symptomatic PICC-related thrombosis in children vary widely from 3% to 20%, and asymptomatic thrombosis rates may even be higher [78]. The development of thrombosis in PICC can be explained by Virchow's triad of stasis, endothelial injury, and hypercoagulability [9]. Venous stasis occurs because the catheter occupies part of the vessel lumen and reduce blood flow [10]. Endothelial injury may occur after repeated attempts at insertion and manipulation of the vessel wall [8]. Hypercoagulability results from underlying illnesses, infections, or medications [49]. In pediatric patients, additional risk factors have been identified, including a high catheter-to-vein ratio, multiple insertion attempts, nonbasilic vein access, left-sided placement, and underlying comorbidities such as malignancy or congenital heart disease [8101112]. Patient-specific anatomical and physiological differences further complicate risk prediction [8]. Most studies on PICC-related thrombosis have been conducted in North America and Europe. Notably, Menéndez et al. [8] reported that the catheter-to-vein ratio and number of insertion attempts significantly influenced the thrombosis risk, while Jaffray et al. [6] showed that PICCs carried a higher risk of venous thromboembolism than tunneled lines in children [13]. However, pediatric-specific data from the Middle East are scarce. Differences in patient demographics, healthcare delivery models, and procedural practices may influence the complication rates, underscoring the need for regional studies. This study was conducted at a large tertiary pediatric hospital in Saudi Arabia to determine the incidence of symptomatic thrombosis among children with PICCs and identify patient and procedure-related factors associated with PICC-related thrombosis. The ultimate goal was to identify potentially modifiable factors that may inform local protocols, enhance patient safety, and optimize vascular access practices in pediatric care. Methods This retrospective cohort study was conducted at King Abdullah Specialized Children's Hospital (KASCH), Ministry of National Guard Health Affairs, Riyadh, Kingdom of Saudi Arabia. A tertiary pediatric referral hospital with a fully integrated electronic medical record system. The study period spanned from January 2019, to June 2023, with 6-month follow-up period. Children aged 0–14 years who underwent PICC insertion in the Interventional Radiology department during the study period were included. Patients with PICCs inserted in non-IR departments and those with drug-induced thrombophilia were excluded. The patient selection process is summarized in Fig. 1. Flow chart visualizing the patient's selection process. N: total number, PICC: peripherally inserted central catheter. Patient variables included age (in months), sex, weight status (for patients below 2 years of age using the weight percentile, while 2 years and above using the body mass index percentile to categorize weight, underweight, normal weight, and overweight), and comorbidities (whether or not they had a comorbidity). Procedural variables included access vein (basilic, brachial, cephalic, axillary, or other central, such as femoral or internal jugular, side (right/left), catheter size (3 Fr, ≥4 Fr), any previous history of PICC insertion and indwelling time (days; categorized as <7, 7–<30, 30–<90, 90–<180, >180). The outcome variable was the occurrence of symptomatic thrombosis, defined as the presence of compatible clinical signs and symptoms (such as limb swelling, erythema, or pain) with confirmation by Doppler ultrasonography. The reasons for PICC removal (clinical resolution, infection, malfunction, and death). Data were extracted from BestCare electronic medical records into a standardized Excel form with predefined categories and operational definitions. We used consecutive samples of all eligible PICC insertions to reduce selection bias and excluded patients with drug-induced thrombophilia to reduce confounding factors. The number of eligible cases determined the sample size over a 5-year period. Statistical analysis The data were entered into Microsoft Excel 2019 and analyzed using the Statistical Package for the Social Sciences, version 25 (IBM Corp., Armonk, NY, USA). Categorical variables are presented as frequencies and percentages, and continuous variables are presented as means and standard deviations. Associations with symptomatic thrombosis were evaluated using the chi-square or Fisher's exact tests. A p value of <0.05 was considered statistically significant. Ethical considerations. The study was approved by the KAIMRC Institutional Review Board (IRB/0252/24; NRC24R/058/01; 18 February 2024). Results A total of 168 pediatric patients, who underwent PICC line insertion in the Interventional Radiology department, were included in this study. The population was predominantly infants, with 41.1% under 12 months of age, followed by 20.2% aged 12–24 months, with smaller proportions in older age groups. The sample was divided nearly equally into males (52.4%) and females (47.6%). From a nutritional standpoint, more than half of the patients had a normal weight (52.4%), 38.1% were underweight, and 9.5% were overweight. Notably, the majority (74.4%) of patients were medically free of comorbidities, suggesting that PICC placement is frequently indicated in otherwise healthy children. The basilic vein was the preferred access site (59.5%), with the brachial vein used in nearly one-third of the cases (30.4%). Only a small proportion (2.4 %) had PICCs inserted via the cephalic vein, whereas the remaining 7.7% were inserted into other central veins. A clear preference for right-sided insertions was noted, which accounted for 86.9% of the cases. Most catheters were of 3 French size (65.3%), consistent with pediatric guidance to use the smallest feasible PICC (typically 3–4 Fr in children) and to maintain a catheter-to-vein ratio ≤ 45% [89]. Two-thirds of the PICC lines (67.8%) remained in place for 7–90 days. Only a minority were in situ for less than 7 days (8.3%) or more than 180 days (13.1%). Only 13.1% of the patients (Table 1). Table 1.Demographic and clinical characteristics of the study subjects (N = 168).CharacteristicsLeveln%Age groupLess Than 12 Months6941.112 months to 2 years (<24 months)3420.22 years to <5 years2716.15 years and older3822.6SexMale8852.4Female8047.6WeightUnderweight6438.1Normal8852.4Overweight169.5ComorbiditiesNo12574.4Yes4325.6Location of PICC insertionBasilic10059.5Brachial5130.4Cephalic42.4Others137.7Side of PICC insertionLeft2213.1Right14686.9PICC line size310965.34 and larger5834.7PICC indwelling time1 day to <7 days148.37 days to <30 days5733.930 days to <90 days5733.990 days to <180 days1810.7>180 days2213.1Previous history of PICC insertionYes2213.1No14686.9Thrombosis (symptomatic)Yes2816.7No14083.3Thrombosis resolution in first 7 daysN/A13982.7Yes74.2No2213.1N: total number, PICC: peripherally inserted central catheter. Demographic and clinical characteristics of the study subjects (N = 168). N: total number, PICC: peripherally inserted central catheter. Symptomatic thrombosis occurred in 28 patients (16.7%). Female sex was associated with a significantly higher incidence of thrombosis (p = 0.019). Likewise, nonstandard insertion sites (femoral, subclavian, or internal jugular) showed a notably higher thrombosis rate (53.8%, p = 0.001) than basilic or brachial insertions. However, as central access is commonly pursued after failed peripheral attempts, this association may reflect confounding by indications from multiple attempts, rather than intrinsic site risks. Left-sided PICCs were associated with a higher incidence of thrombosis (36.4%) than right-sided PICCs (13.7%, p = 0.014), and prior PICC insertion history was also associated with symptomatic thrombosis (p = 0.014) (Table 2). Table 2.Association between symptomatic thrombosis and the study variables (N = 168).ThrombosisNo CharacteristicsLeveln%n%P - valueAge groupLess Than 12 Months1115.95884.10.09412 months to 2 years823.52676.52 years to <5 years725.92074.15 years and older25.33694.7SexMale910.27989.8.019*Female1923.86176.3Weight%Underweight1523.44976.60.139Normal1213.67686.4Overweight16.31593.8ComorbiditiesNo2116.810483.20.937Yes716.33683.7Location of PICC insertion‡Basilic1313.08787.00.001*Brachial815.74384.3Cephalic00.04100.0Others753.8646.2Side of PICC insertion†Left836.41463.60.014*Right2013.712686.3PICC line size32119.38880.70.2364 and larger712.15187.9PICC indwelling time‡1 day to <7 days535.7964.30.1087 days to <30 days610.55189.530 days to <90 days58.85291.290 days to <180 days316.71583.3>180 days940.91359.1Previous history of PICCYes836.41463.60.014*insertion†No2013.712686.3P - values were calculated using the Pearson's Chi-square test,(†):Fisher's exact test,(‡):Linear-by-Linear Association,*:significant p - value at <0.05, N: total number, PICC: peripherally inserted central catheter. Association between symptomatic thrombosis and the study variables (N = 168). P - values were calculated using the Pearson's Chi-square test, Fisher's exact test, Linear-by-Linear Association, :significant p - value at <0.05, N: total number, PICC: peripherally inserted central catheter. Although age, weight, comorbidities, line size, and catheter dwell time were not statistically significant, the data suggested a consistent pattern. Infants demonstrated a higher frequency of thrombosis than older children did, although this trend was not statistically significant. Likewise, thrombosis occurred more often when PICCs were in place for very short durations (<7 days, 35.7%) or for prolonged periods (>180 days, 40.9%) (p = 0.108). These non-significant trends indicate potential age- and dwell-related vulnerabilities, which warrant further evaluation in larger cohorts. The most common clinical indication for PICC line use was prolonged medication or antibiotic infusion (39.3%), followed by parenteral nutrition (33.3%), and pediatric intensive care unit admission (11.3%) (Fig. 2). When these lines were removed, clinical resolution was the primary cause (69.0%). Other causes included PICC infection (11.3%), malfunction (7.7%), and death (7.1%) (Fig. 3). Reported indications for PICC lines insertion (N = 168). N: total number, PICC: peripherally inserted central catheter. Reported reasons of PICC lines removal (N = 168). N: total number, PICC: peripherally inserted central catheter. Discussion Approximately 1 in 6 children developed symptomatic PICC-related thrombosis [8]. Female sex, left-sided placement, nonbasilic access, and previous PICC exposure were associated with symptomatic thrombosis in univariate analyses [1415]. A U-shaped dwell-time signal suggests early procedure-related and late maintenance-related vulnerability windows. Comparison with the literature Our incidence aligns with pediatric reports (3–20%) [78]. Previous studies have identified several factors associated with PICC-related thrombosis in children, including catheter to vein ratio, number of insertion attempts, catheter tip position, and underlying patient characteristics. Unfortunately, these variables were not systematically captured in our retrospective dataset and therefore could not be evaluated. Nevertheless, our findings support the importance of further prospective studies incorporating these variables to better characterize thrombosis risk in pediatric patients [1011]. Our findings suggest that vein selection and insertion side may be important considerations in pediatric vascular access practice and warrant further prospective evaluation [316]. Clinical implications Our findings suggest that vein selection, insertion side, and previous PICC exposure may be important considerations in pediatric vascular access planning. However, these observations should be considered exploratory and require confirmation in prospective studies incorporating multivariable analyses before definitive clinical recommendations can be made. Future studies may also evaluate the potential role of targeted surveillance strategies in pediatric patients with characteristics associated with PICC-related thrombosis. Future directions Prospective validation of a risk-stratified insertion algorithm, routine assessment of the catheter-to-vein ratio, and evaluation of ultrasound screening in selected high-risk pediatric subgroups [11]. The association between prior PICC exposure and new thrombosis may reflect residual vessel injury or altered flow patterns around prior insertion sites. Pediatric data suggested that cumulative device exposure and repeated attempts can increase endothelial trauma, tracking our signal of prior catheter history [812]. The U-shaped relationship between dwell time and thrombosis events with very short and long indwelling intervals has face validity. Early events may be driven by procedural trauma or a suboptimal tip position, while late events may reflect maintenance factors, biofilm formation, and infection-related inflammation [145]. Interventions aimed at both ends warranted maintenance and timely removal [12]. Therefore, a risk-stratified follow-up plan may be beneficial. Children with prior PICCs, left-sided or non-basilic insertions, or difficult multi-attempt placements could undergo closer clinical review in the first week (when early events cluster) and a low threshold for ultrasound if symptoms arise [138]. The higher thrombosis rate observed in left-sided PICCs may reflect procedural practice rather than true laterality risk. At our center, operators typically attempted right-sided access first, with left-sided placement pursued only after multiple unsuccessful right-sided attempts with difficult venous access. These repeated attempts could increase endothelial injury and potentially contributed to the higher thrombosis rates associated with left-sided insertions. Study strength The strengths of this study include, to our knowledge, the largest cohort on this topic in the region, resulted from pediatric patients enrolled consecutively over multiple years at a tertiary children's center, thereby enhancing clinical relevance and generalizability. Standardized variable definitions and uniform data abstraction reduce misclassifications and improve reliability. The analysis focused on potentially modifiable procedural characteristics, including insertion side, access vein, previous PICC exposure, catheter size, and dwell time to provide concrete targets for prevention. However, because Doppler ultrasonography was performed only in patients with clinical suspicion of thrombosis, asymptomatic thrombosis events were not captured. Therefore, the reported incidence likely underestimates the true burden of PICC-related thrombosis. Due to the retrospective, single-center nature of this study, several limitations warrant consideration. The absence of multivariable analysis represents an important limitation. Because of the retrospective design and the limited number of thrombosis events, we were unable to determine whether the observed associations remained independently associated with thrombosis after adjustment for potential confounders. Therefore, the identified factors should be considered exploratory associations rather than definitive independent risk factors. We lacked data on anticoagulation therapy (prophylactic or therapeutic), which could have significantly affected the observed incidence. Details of the insertion technique, including routine ultrasound guidance, operator experience, catheter-to-vein ratio, number of insertion attempts, catheter tip position, and adherence to standardized protocols, were not systematically captured and may have influenced thrombosis risk. The PICC maintenance practices, such as flushing and dressing regimens, were not evaluated. In addition to excluding drug-induced thrombophilia, we did not assess hereditary or acquired thrombophilic disorders that may predispose some children to PICC-related thrombosis. Our analysis focused on short-term events and did not capture long-term sequelae, such as post-thrombotic syndrome or recurrent thrombosis. Finally, we did not compare PICCs with alternative central venous access devices, such as tunneled catheters or implanted ports, thereby limiting the interpretation of the relative safety of PICCs in pediatric patients. In conclusion, PICC-associated thrombosis is frequent in pediatric patients. Several patient and procedure-related factors were associated with symptomatic PICC-related thrombosis. These findings should be considered hypothesis generating and require validation in larger prospective studies with multivariable adjustment. Future work should prospectively capture the catheter-to-vein ratio, first-pass success, and real-time maintenance adherence and evaluate targeted ultrasound screening in high-risk subgroups. Given the multiple factors associated with thrombosis identified in this study, the development and validation of a comprehensive risk-prediction tool could help guide personalized approaches to PICC insertion and management. Funding statement The authors declare that no funds, grants, or other support were received during the preparation of this manuscript. Data availability The data that support the findings of this study are available on request from the corresponding author. Acknowledgments The authors would like to thank Editage proofreading services and for the English language editing. AI disclosure statement This manuscript was written without the use of AI tools. The authors confirm that no AI technologies were utilized in the writing, creation of images, or in the collection and analysis of data. Disclosure The authors declare no conflicts of interest and that this work received no funding or other support.