Pediatr Surg Int (2013) 29:445–449 DOI 10.1007/s00383-013-3259-y
ORIGINAL ARTICLE
Management of peripherally inserted central catheter associated deep vein thrombosis in children Alessandra C. Gasior • E. Marty Knott Shawn D. St. Peter
•
Accepted: 7 January 2013 / Published online: 23 January 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Introduction No protocol exists for prophylaxis or therapeutic management of peripherally inserted central catheter (PICC) related deep venous thrombosis (DVT) in children. Currently, very few patients are provided prophylaxis for DVT. In this study, we analyzed our current practice of PICC placement to identify the frequency of DVT, analyze risk factors and determine current treatment patterns in order to determine the need for protocols. Methods The dataset was retrospectively collected from January 1, 2000 to December 31, 2011. Patients with an upper extremity PICC were assessed for subsequent DVT formation. Variables included: demographics, co-morbidities, method of DVT diagnosis, treatment course, and recurrence. Results There were 1,289 PICC placements, with 24 (1.9 %) per line events of DVT in 23 patients, of which 3 had recent surgery, 2 had sepsis, 1 had a family history of clots, and 2 had a malignancy. All but one was symptomatic. Diagnosis was made in 92 % by ultrasound, the remaining with CT. No patients had prophylaxis. Of the seven patients who underwent hypercoagulable work-up, three were positive. 15 patients were treated with enoxaparin, 5 patients were treated with heparin, 2 treated with tissue plasminogen activator and 2 were observed. 84 % were treated with long-term enoxaparin for a mean of 3.3 months with 54 % proven clot resolution and 1 patient had recurrence of UE DVT. Conclusions The risk of DVT with PICC placement is small in children and prophylaxis can probably be reserved A. C. Gasior E. Marty Knott S. D. St. Peter (&) Department of Pediatric Surgery, Children’s Mercy Hospital and Clinics, 2401 Gillham road, 64108 Kansas, MO, USA e-mail:
[email protected]
for those with previous DVT or known hypercoagulable state. Keywords
PICC DVT Pediatric Upper extremity
Introduction Peripherally inserted central line catheters (PICC) are the most common mode of central venous access in children and infants [1]. It is estimated that 8.3–33 % of patients in the neonatal intensive care units receive a PICC line [2, 3]. However, these silastic catheters are not without risk. Complication rates including breaks, occlusion and infection have been seen ranging from 20 to 50 % in children requiring long term treatment [4, 5]. With this increased use of PICC lines, a multi-center study from 2001 to 2007 reveals an increase in the diagnosis of venous thromboembolism at pediatric hospitals 34–58 per 10,000 admissions [6]. Additionally, infants less than 1 year old and adolescents account for the majority [7]. Currently, no protocol exists for prophylaxis or therapeutic treatment of deep venous thrombosis (DVT) in the pediatric population. As DVT is an escalating problem in the pediatric population where PICC lines are becoming the routine for intravenous access, we sought to determine the incidence of upper extremity PICC line DVT, examine the current management and identify risk factors for DVT formation at our institution.
Methods After Internal Review Board approval (#11-120-170) a retrospective review was conducted from January 1, 2000
123
446
Pediatr Surg Int (2013) 29:445–449
through December 31, 2011. Inclusion criteria were patients less than 21 years who underwent an upper extremity PICC line placement at our institution. Variables recorded included patient demographics, duration of PICC line, presence of clinical symptoms, diagnosis of partial or complete DVT, method of diagnosis, presence of hypercoagulable work up, initial treatment, long-term therapy, resolution of clot and recurrence rate. Partial DVT was defined as vascular filling defect with residual flow in the vein. Complete DVT was defined as complete occlusion of the vein lumen. Fisher’s exact test and unpaired t test were used, where appropriate. Significance was determined as P \ 0.05.
Results During the study period 1,289 upper extremity PICC lines were placed. Of those PICC lines, there were 24 (1.9 %) per line events of DVT formation occurring in 23 patients. There was no difference in age, gender, or duration of PICC (Table 1). In five cases, the primary diagnosis for hospital admission was the DVT. Risk factors included recent operation in three patients, malignancy in two patients, prior thrombosis in one patient, and family history of thrombosis in one patient. No patients were on DVT prophylaxis. One patient had a pulmonary embolism (PE). The mean duration of PICC line before DVT diagnosis was 9.7 days. 92 % of patients were clinically symptomatic from the DVT, presenting with upper extremity pain, edema, errythema, or even facial swelling. 92 % were diagnosed by doppler ultrasound; the remainders were diagnosed by CT scan. 19 patients had occlusive thrombi and 5 patients had non-occlusive thrombi. Nearly 80 % of the clots extend into the subclavian vein by US exam. 88 % of the PICC lines were removed after diagnosis. Of the three patients who did not have the PICC removed after diagnosis, one patient was a neonate with no other access. The PICC was left in place with a low molecular weight heparin drip therapy. The patient was followed by daily ultrasounds, which showed the clot to decrease in size. The patient’s affected extremity was monitored clinically and over time the swelling and erythema decreased Table 1 Comparison of DVT per PICC line events No DVT (n = 1265)
DVT (n = 24)
Age (years)
7.0
7.9
0.56
Gender
M 50.2 %
M 33.3 %
0.15
Duration of PICC (days)
18.8
9.7
0.22
123
P value
significantly. In another patient, the PICC line was unable to be removed due to the clot. After 3 days of systemic heparin therapy, the PICC line was able to be dislodged. Interestingly, this patient was also the one patient in the study with a PE. The fourth patient had a terminal illness and the parents chose not to treat the thrombosis and to keep the PICC in place. Only seven patients underwent a hypercoagulable workup, and three were positive. Treatment was initiated in 22 of the 24 PICC thrombi, 15 with enoxaparin, 5 with heparin drip, and 2 with tissue plasminogen activator. 19 patients were treated with long-term anticoagulation with a mean duration of 3.3 months, of which 16 patients were treated with enoxaparin and 3 were treated with warfarin. Of the five patients not treated with long-term anticoagulation, one patient was not treated due to an intraventricular hemorrhage after 5 days of lovenox therapy. She was followed with ultrasounds and found to have resolution of the clot. Another patient was not given anticoagulation due to a recent history of intracranial hemorrhage and transplant kidney hematoma formation after her transplant surgery. A third patient was not treated with anticoagulation at the parents request due to terminal illness, as mentioned above. A fourth patient was treated with removal of PICC alone, due to a history of a bleeding disorder. The fifth patient did not receive anticoagulation as the clot had resolved after removal, as seen on a followup ultrasound obtained 2 days after the catheter was withdrawn. Recurrence of UE DVT occurred in one patient 6 months after the initial diagnosis. The recurrence occurred on the contralateral UE following another PICC line placement. This patient did not undergo a hypercoagulable workup at the time of the initial thrombus and she was not placed on prophylactic anticoagulation prior to the recurrence. two other patients had subsequent lower extremity DVT. Follow-up imaging with ultrasound in 20 cases documented resolution of clot in 13. The mean follow-up ultrasound that demonstrated resolution of clot was 2 months. Complication of anticoagulation occurred in one patient who developed an intraventricular hemorrhage 5 days after starting enoxaparin, which was subsequently stopped. No patients had any long-term sequelae from the DVT. When the patients were grouped based on age distribution between neonates, 1-to12-years-old and teenagers, the peak incidence was found in patients aged 1–12 years (Fig. 1). Most infants had the primary diagnosis of infection, whereas 1-to12-year- old and teenagers had other as the primary diagnosis. Only the 1-to12-year-old were found to have a positive hypercoagulable state. Additionally, the 1-to12-year-olds and teenagers were less likely to have resolution of clot (Table 2).
Pediatr Surg Int (2013) 29:445–449
447
Fig. 1 Age distribution of UE DVT
Table 2 Age distribution and risk factors in UE PICC DVT Infants (n = 5)
1–12 years (n = 14)
Teenagers (n = 5)
Fig. 2 DVT prophylaxis protocol
Primary diagnosis Infection
4
3
–
Recent surgery
1
–
–
Cancer
–
1
–
Others
–
10
5
0
3
0
Initial treatment
4
14
4
Clot resolution
4
7
2
Hypercoagulable state
Discussion The reported incidence of traditional central catheter related DVT in the pediatric population varies from 11 to 50 % with the highest risk at 4 days after insertion [8, 9]. The incidence of symptomatic PICC line DVT ranges from 0 to 5.3 % [10]. Neurological ICU patients were reported to have a PICC line associated UE DVT rate of 8.4 % [11]. These data suggest that PICC lines have a lower rate of thrombosis than centrally inserted catheters. However, the reported incidence of DVT is thought to be underestimated as only the symptomatic patients undergo imaging studies and are subsequently diagnosed with clots [12]. Moreover, the incidence of DVT is as high as 9.3 % when all patients with a PICC line were imaged, with only 1 out of 20 patients presenting with clinical indications of a DVT [13]. Our study shows that with upper extremity PICC lines, 92 % of the patients were symptomatic from the DVT with swelling, pain and erythema. However, we expect to have
underestimated the incidence in our population given the majority of patients were imaged only based on clinical symptoms. Our institutional incidence of UE PICC related DVT is 1.9 % per line event, with a mean PICC line duration of 9.7 days at DVT diagnosis. Others have shown more complications are typically seen in low body weight (\12.5 kg) infants [14]. This may be expected as the catheter to vein size ratio is more similar to standard central lines. However, our average age of diagnosis was 8 years old and no patients were\2.5 kg. When the ages were distributed into neonates, 1-to12-years old and teenagers, ages 1–12-year-old were found to be the most common age group in our study. One report documented the use of enoxaparin has increased over the past decade to 53 % [7]. Our institution reflects this shift in DVT management as well, with 63 % of the patients being treated with enoxaparin. This is likely due to the ease of subcutaneous administration compared to continuous infusion as well as simplicity of titration with less monitoring. Only 29 % of our patients with a DVT underwent a hypercoagulable work-up, nearly half of which were positive. Perhaps, a standardized protocol for patients with DVTs should include hypercoagulable studies to determine future risk for DVT in our young patients. The average duration of long-term anticoagulation therapy in our study was 3 months, however for patients who underwent follow-up ultrasound, the average date of
123
448
Pediatr Surg Int (2013) 29:445–449
Fig. 3 Upper extremity PICC DVT treatment protocol
clot resolution was 2 months. Perhaps an ultrasound at 2 months post-diagnosis demonstrating resolution of clot would terminate anticoagulation therapy sooner.
123
Guidelines from the American College of Chest Physicians (ACCP) give a strong recommendation for no chemoprophylaxis for children with central venous catheters
Pediatr Surg Int (2013) 29:445–449
due to the scarcity of supporting evidence, as well as increased risk for bleeding complications [15]. The risk of a bleeding complication, although small, is present. Our one patient complication was an intraventricular hemorrhage after 5 days of enoxaparin therapy, which was immediately stopped. The ACCP gives weak recommendation that treatment be individualized, based solely upon extrapolation of adult data. While the risk for DVT in the pediatric population is low, our study suggests that DVT prophylaxis should be utilized for patients with previous history of thrombosis or known hypercoagulable state (Figs. 2, 3). A protocol for placement, work-up and treatment may streamline care and improve resource utilization.
References 1. Braswell LE (2011) Peripherally inserted central catheter placement in infants and children. Tech Vasc Interv Radiol 14:204–211 2. Chien L, Macnab Y, Aziz K et al (2002) The Canadian neonatal network. Variations in central venous catheter-related infection risks among Canadian neonatal intensive care units. Pediatr Infect Dis J 21:505–511 3. Van den Hoogen A, Brouwer MJ, Gerards LJ et al (2008) Removal of percutaneously inserted central venous catheters in neonates is associated with the occurrence of sepsis. Acta Paediatr 97:1250–1252 4. Matsuzaki A, Suminoe A, Koga Y et al (2006) Long-term use of peripherally inserted central venous catheters for cancer chemotherapy in children. Support Care Cancer 14:153–160
449 5. Barrier A, Williams DJ, Connelly M et al (2012) Frequency of peripherally inserted central catheter complications in children. Pediatr Infect Dis J. 31:519–521 6. Raffini L, Huang YS, Witmer C et al (2009) Dramatic increase in venous thromboembolism in children’s hospitals in the United States from 2001 to 2007. Pediatrics 124:1001–1008 7. Monagle P (2000) 1, Adams M, Mahoney M, et al. Outcome of pediatric thromboembolic disease: a report from the Canadian childhood thrombophilia registry. Pediatr Res 47:763–766 8. Beck C, Dubois J, Grignon A et al (1998) Incidence and risk factors of catheter-related deep vein thrombosis in a pediatric intensive care unit: a prospective study. J Pediatr 133:237–241 9. Glaser DW, Medeiros D, Rollins N et al (2001) Catheter-related thrombosis in children with cancer. J Pediatr 138:255–259 10. Revel-Vilk S, Yacobovich J, Tamary H et al (2010) Risk factors for central venous catheter thrombotic complications in children and adolescents with cancer. Cancer 116:4197–4205 11. Wilson TJ, Brown DL, Meurer WJ et al (2012) Risk factors associated with peripherally inserted central venous catheterrelated large vein thrombosis in neurological intensive care patients. Intensive Care Med 38:272–278 12. Male C, Kuhle S, Mitchell L (2003) Diagnosis of venous thromboembolism in children. Semin Thromb Hemost 29: 377–390 13. Dubois J, Rypens F, Garel L et al (2007) Incidence of deep vein thrombosis related to peripherally inserted central catheters in children and adolescents. CMAJ 177:1185–1190 14. Goutail-Flaud MF, Sfez M, Berg A et al (1991) Central venous catheter-related complications in newborns and infants: a 587-case survey. J Pediatr Surg 26:645–650 15. Monagle P, Chan A, Massicotte P et al (2004) Antithrombotic therapy in children: the Seventh ACCP conference on antithrombotic and thrombolytic therapy. Chest 126:645S–687S
123