Pediatr Surg Int (2017) 33:1001–1005 DOI 10.1007/s00383-017-4099-y

ORIGINAL ARTICLE

Indications and outcomes for tunneled central venous line placement via the axillary vein in children Allison F. Linden1,2 • Chase Corvin3 • Keva Garg3 • Richard R. Ricketts4 A. Alfred Chahine2,5

•

Accepted: 30 May 2017 / Published online: 27 June 2017 Ó Springer-Verlag Berlin Heidelberg 2017

Abstract Purpose To assess the indications, safety and outcomes of tunneled central venous catheters (CVCs) placed via a cutdown approach into the axillary vein in children, an approach not well described in this population. Methods A retrospective cohort study was performed on pediatric patients who received CVCs via open cannulation of the axillary vein or one of its tributaries between January 2006 and October 2016 at two hospitals. Results A total of 24 axillary CVCs were placed in 20 patients [10 male (42%); mean weight 7.0 kg (SD 2.9); mean age 10 months (SD 6)]. The most common indications for axillary vein access included neck or chest wall challenges (tracheostomies or chest wall wounds) (n = 18). The median duration of line placement was 140 days (IQR 146). The most common indications for removal were completion of therapy (n = 7, 39%) and infection (n = 5, 28%). There were no early complications. & Allison F. Linden [email protected] 1

Section of Pediatric Surgery, Department of Surgery, University of Chicago Medicine, 5839 South Maryland Avenue, Rm. A-426, MC4062, Chicago, IL 60637, USA

2

Department of Surgery, Medstar Georgetown University Hospital, 3800 Reservoir Road, NW, Washington, DC 20007, USA

3

Georgetown University School of Medicine, 3900 Reservoir Road, NW, Washington, DC 20057, USA

4

Division of Pediatric Surgery, Department of Surgery, Emory University School of Medicine, 1405 Clifton Road, Atlanta, GA 30322, USA

5

Division of General and Thoracic Surgery, Children’s National Health System, 111 Michigan Avenue, NW, Washington, DC 20010, USA

Long-term complications included infection (n = 5) or catheter malfunction (n = 3). Conclusions Tunneled CVC placement via a cutdown approach into the axillary vein or its tributary can be an effective alternative approach to obtain long-term vascular access in children. Outcomes may be comparable to lines placed in traditional internal jugular and subclavian vein locations. Keywords Axillary vein
Central venous line placement

Introduction Placement of long-term central venous catheters (CVC) is one of the most common procedures performed by a pediatric surgeon [1]. This procedure is technically more difficult, has a higher complication rate and involves different anatomical features compared to adults [2]. It has long been standard practice among pediatric surgeons to access the internal jugular vein and the subclavian vein, with location dependent on patient characteristics, comorbidities, rate of complications and surgeon preference [2, 3]. In children requiring long-term central venous access for such reasons as nutrition, chemotherapy, supplemental fluid or transfusions, it becomes increasingly paramount to maximize central venous access options because of need for multiple CVCs [4]. Further, in certain patients, where tunneling the lines into the chest might be a challenge or a concern, finding alternative tunneling sites is also vital. The successful use of the axillary vein as an alternative site for CVC placement has been demonstrated in adult patients [5–7]. There are few studies documenting its use in

123

1002

pediatric patients, though, and of those that exist, only one documents a series using an open approach [8–11]. The goal of our study was to review the feasibility and safety of tunneled CVC placement via a cutdown approach into the axillary vein, using the distal arm as an exit site, in pediatric patients. We will characterize the indications, complications and outcomes of this procedure and compare them to the literature regarding the traditional internal jugular and subclavian vein approach.

Methods A retrospective cohort study was performed of pediatric patients at Medstar Georgetown University Hospital (Washington, DC) and Children’s National Health System (Washington, DC) from January 2006 to October 2016 who underwent central venous access via open cannulation of the axillary vein by one surgeon (AAC). The decision to pursue an axillary approach was based mainly on challenges that existed for tunneling the lines into the standard chest position for a variety of reasons including the presence of a tracheostomy, sternotomy, or open chest wounds. Patient demographics were collected, including: gender, age and weight at time of line insertion, and diagnosis. In addition, laterality, indication for line placement and intraoperative details were obtained from dictated operative reports. Finally, duration of line, indications for line removal, early complications (within 30 days of CVC placement) and late complications (after 30 days of CVC placement) were also determined. IRB exemption was granted by the Georgetown University IRB. Procedures were all done under general anesthesia or heavy sedation in the operating room or in the NICU. Patients were placed in the supine position with a small roll under the shoulder. The arm to be cannulated was abducted at a 90° angle to assist with adequate access. A distal axillary crease transverse incision was made and extended down through the subcutaneous tissues to expose the axillary sheath, which was opened bluntly between the heads of the biceps and triceps. During this dissection, care was taken to avoid injury to the axillary artery and/or the brachial plexus. A tributary to the axillary vein (either the basilic vein or one of a few smaller unnamed tributaries) was almost always found and isolated proximally and distally. If no adequate tributary was found, the axillary vein itself was controlled proximally and distally. A counter-incision was then made on the ipsalateral anterolateral upper arm proximal to the elbow crease. A subcutaneous tunnel was created between this incision and the axillary one. A previously heparinized Broviac catheter was passed through the tunnel and the cuff placed 1 cm proximal to the distal upper arm incision skin edge. The

123

Pediatr Surg Int (2017) 33:1001–1005

line length was estimated using the standard external surface anatomy, accounting for the fact that the catheter would be further advanced 1–2 cm when the arm was not abducted anymore. A proximal venotomy was then performed and the catheter was flushed into the central circulation through the venotomy. If back bleeding was a concern, the accessed vein was ligated and an absorbable tie was placed around the intravenous catheter proximally. In cases where the axillary vein was accessed directly, we did not ligate the distal portion of the vein but rather controlled the bleeding with a fine absorbable suture around the venotomy if needed. Intraoperative fluoroscopy was used to confirm catheter placement at the atriocaval junction. Line patency was verified and it was flushed with heparinized saline solution, the subcutaneous tissue and skin were closed at the axillary incision and the line was secured to the skin at the exit site.

Results During the study period, a total of 24 axillary venous line placements were performed on 20 patients (42% male). Six patients were lost to follow-up and three patients died while the line was still in place. Patients were a mean age at CVC placement of 10 months (SD 179; range 1–672 days old) and had a mean weight at placement of 7.0 kg (SD 2.9, range 2.9–12 kg) (Table 1). There were several indications for why the axillary location was chosen, with the most common being the presence of neck or chest wall challenges (including tracheostomies and chest wall wounds) (n = 18 patients). A majority of patients had multiple previous CVCs (n = 14 patients). The majority of catheters (67%) were placed in the right axillary vein (Table 2). Placement of a CVC via the cutdown axillary approach was successful in 100% of patients, though challenges were encountered during placement. The most common intraoperative challenges were: anatomical (difficulty isolating the axillary vein or its tributaries due to size and location variation) and mechanical (coiling of the catheter in the subclavian vein, advancement of the catheter into an incorrect position). Of the 24 CVC placements, 7 were associated with intraoperative malpositioning and were immediately corrected with fluoroscopy. In regards to outcomes, the median line duration was 140 days (IQR 146; range 10–636 days) (Table 3). Discontinuation of the CVC was due to: completion of therapy (n = 7), infection (n = 5), and catheter malfunction (e.g. broken tip, clotted segment) (n = 3). There were no early complications. Lines removed for infection occurred between 42 and 399 days post-placement (median = 56 days). One CVC was found to have a superficial

Pediatr Surg Int (2017) 33:1001–1005

1003

Table 1 Patient characteristics a

Patient characteristics

n = 24

Male gender, n (%)

10 (42)

Mean age at surgery, days (SD)

298 (179)

Mean weight at surgery, kg (SD)b

6.95 (2.86)

Indication for placementc Tracheostomy

14

Multiple previous lines

14

Chest wall skin breakdown/wounds

4

Difficult venous access—multiple failed attempts

5

Replacement of previous axillary cutdown

3

Diagnosis Short bowel syndrome Congenital heart disease

9 5

Hydrocephalus

2

Biliary atresia

1

Chronic lung disease

1

Cystic fibrosis

1

Hemophagocytic lymphoproliferative histiocytosis

1

Infantile neurofibromatosis

1

Microvillous inclusion disease

1

SCID

1

Sepsis

1

a

24 axillary lines were placed in 20 patients

b

Patient weight was not available for two procedures Many patients had multiple indications for axillary line placement

c

Table 2 Intraoperative details Intraoperative details

n = 24

Location of catheter placement, n (%) Right axillary vein Left axillary vein Median catheter size, F (IQR)

16 (67) 8 (33) 4.2 (4.2–4.2)

abscess at the skin incision made for the venotomy. This was the only incisional complication. There were no episodes of upper extremity swelling due to venous obstruction.

Discussion This study characterized the indications, operative characteristics and outcomes related to the axillary cutdown approach for tunneled CVC placement in the pediatric population. Our data show that this approach may be an effective alternative for obtaining long-term central

vascular access in pediatric patients across a wide range of ages and weights. Stephens et al. were the first to describe the axillary approach for tunneled CVC placement in neonates [11]. Our study offers two technical variations on their original report. By performing a distal axillary crease incision we are almost always able to access the axillary vein via one of its tributaries, thus avoiding direct trauma to the axillary vein itself. While they describe the exit site on the lateral chest wall, we tunnel our lines into the distal arm, offering the advantage of avoiding the chest as an exit site. In addition, we have used this technique beyond the neonatal period with no increased complications. Patients in need of tunneled CVCs often have comorbidities and have had multiple other line placements resulting in their own complications. The more CVCs that are placed, the more complex each subsequent one becomes [12]. It becomes essential to carefully examine all central venous access options in light of the potential difficulties with traditional sites of placement. With a 100% success rate, our data show that an axillary approach to CVC placement should be considered a viable option for neonates through 2 years of age. Our data show that the axillary site becomes especially salient when multiple previous CVCs render traditional sites inaccessible. Additionally, it can be a preferred option in smaller children with a short neck and a tracheostomy or other challenges on the anterior superior chest wall, such as complicated wounds, that make it more difficult to access the vein or to place the exit site in its traditional or a more lateral location on the chest. Malpositioning during CVC placement in neonates and young children is a known operative anticipation. While 29% of our axillary line placements were associated with needing intraoperative fluoroscopic repositioning, this is comparable to the literature [11]. Internal jugular and subclavian placement of CVCs have a comparatively lower rate of malpositioning, but this can be anticipated [11]. Axillary CVCs have a longer track to travel and are more difficult to manipulate, especially with increased inherent anatomic challenges [13, 14]. Further, CVCs placed in the right subclavian vein are associated with over double the rate of malpositioning compared to those placed in the left subclavian [14]. Almost all of our axillary CVCs with malpositioning were placed on the right (6 out of 7). Since analyzing this data, we have switched to preferentially using the left side if possible. CVCs are also associated with significant complications after placement. Our data show that just over a third of axillary CVCs remained patent until completion of therapy. This is better than proportions described in the literature, especially compared to nontunneled CVCs [3]. Blood stream infections remain one of the costliest and most

123

1004

Pediatr Surg Int (2017) 33:1001–1005

Table 3 Outcomes a

Outcome

n = 18

\1 year old [1 year old (n = 11) (n = 7)

Median number of catheter days (IQR)

140 (146)b

54 (123)c

140 (196)

Completion of therapy Infection

7 (39) 5 (28)

5 2

2 3

Occlusion

2 (11)

1

1

Broken catheter

1 (6)

0

1

Patient death

3 (17)

3

0

Early complication (\30 days), n (%)

0

0

0

Late complication ([30 days), n (%)

8 (53)d

3

5

Reason for late complication, n (%) Infection 5 (33)d

2

3

Occlusion

2 (13)d

1

1

Broken catheter

1 (7)d

0

1

Indication for removal, n (%)a

a

Follow up data not available for six cases (5 cases \1 year old)

b

n = 15 as 3 patients died with catheter still in place

c

n = 8 as 3 patients died with catheter still in place

d

n = 15 (does not include 3 patients that died with catheter in place)

morbid complications of CVC placement [15]. While our infection rate is slightly higher than the pooled 20% found in the literature, 2 out of 5 of our CVCs removed for infection were in patients with short gut syndrome, a known risk factor for increased blood stream infections [3, 16]. The remainder of our CVCs removed due to infection were in immunocompromised patients, another known risk factor for infection. Our occlusion rate is also similar to that found in the literature [3]. Finally, while infants have been found in previous studies to be more at risk of early complications, this was not the case in our study [17]. Instead, we found a greater risk of any complication in children greater than 1 year old. This is likely due to those children having an increased duration of catheter days and subsequent increased number of catheter accesses. In conclusion, axillary CVCs do not present an increased risk of failure or complication compared to internal jugular, subclavian or saphenofemoral CVCs [18–20]. Axillary vein access with a distal arm exit site has many potential advantages over other central venous access sites. It can leave a more cosmetically acceptable scar site and allows for placement of the tunneled line in a more convenient position. Accessing the axillary vein through one of its tributaries leaves the vein available for future catheterization if needed. Further, multiple axillary vein tributaries are often present and offer more potential access

123

sites. Finally, axillary vein access does not obstruct internal jugular venous flow in those patients at potential risk of intraventricular hemorrhage. On the other hand, placement of CVCs via the axillary vein is not without its own unique difficulties and complications. This technique can present more of a technical challenge because the target vein is more deeply imbedded. Use of accurate anatomic landmarks is essential to finding the axillary vein. Ultrasound-guided percutaneous axillary vein access has been reported in the neonate PICC literature and may be considered as an adjunct for intraoperative placement [21]. Another unique complication is the risk of upper extremity edema or deep vein thrombosis due to potential venous obstruction from the catheter [11]. We did not experience this complication in our cohort, though, most likely because we accessed the vein through one of its tributaries the majority of the time. This study has many limitations. The sample size is small and the study is retrospective and performed at a limited number of institutions. While this limits its generalizability, complication rates and long-term outcomes are comparable to those found in larger systematic reviews evaluating CVCs placed in other sites [3]. A prospective study, with the comparison of intraoperative and outcome data between axillary vs. internal jugular vs. subclavian vein approach to CVC placement would greatly enhance clinical decision making.

Conclusions Tunneled central venous line placement via a cutdown approach into the axillary vein can provide an effective alternative approach to obtain long-term vascular access in children with similar outcomes to the more traditional internal jugular and subclavian approaches. It may be best indicated for patients with anterior chest or neck challenges, such as wounds, ventriculoperitoneal shunts, tracheostomies and sternotomy incisions. This approach should be within the pediatric surgeon’s breadth of experience when dealing with challenging central venous line issues.

References 1. Somme S, Bronsert M, Morrato E, Ziegler M (2013) Frequency and variety of inpatient pediatric surgical procedures in the United States. Pediatrics 132:e1466–e1472 2. Han S, Kim S, Kim C, Kim W, Lim C, Park Y, Bahk J (2004) Comparison of central venous catheterization sites in infants. J Int Med Res 32(6):563–569 3. Ullman A, Marsh N, Mihala G, Cooke M, Rickard C (2015) Complications of central venous access devices: a systematic review. Pediatrics 136(5):1331–1344

Pediatr Surg Int (2017) 33:1001–1005 4. Yang R, Moineddin R, Filipescu D, Parra D, Amaral J, John P, Temple M, Connolly B (2012) Increased complexity and complications associated with multiple peripherally inserted central catheter insertions in children: the top of the iceberg. J Vasc Interv Radiol 23(3):351–357 5. Martin C, Eon B, Auffray J, Saux P, Gouin F (1990) Axillary or internal jugular central venous catheterization. Crit Care Med 18(4):400–402 6. O’leary R, Ahmed S, Mclure H et al (2012) Ultrasound-guided infraclavicular axillary vein cannulation: a useful alternative to the internal jugular vein. Br J Anaesth 109(5):762–768 7. Taylor B, Yellowlees I (1990) Central venous cannulation using the infraclavicular axillary vein. Anesthesiology 72(1): 55–58 8. Sandhu N (2004) Transpectoral ultrasound-guided catheterization of the axillary vein: an alternative to standard catheterization of the subclavian vein. Anesth Analg 99(1):183–187 9. Pirotte T (2008) Ultrasound-guided vascular access in adults and children: beyond the internal jugular vein puncture. Acta Anaesthesiol Belg 59(3):157–166 10. Galloway S, Bodenham A (2003) Ultrasound imaging of the axillary vein—anatomical basis for central venous access. Br J Anaesth 90(5):589–595 11. Stephens BL, Lelli JL, Allen D et al (1993) Silastic catheterization of the axillary vein in neonates: an alternative to the internal jugular vein. J Pediatr Surg 28:31–35 12. Gibson F, Bodenham A (2013) Misplaced central venous catheters: applied anatomy and practical management. Br J Anesth 110(3):333–346

1005 13. Bannon M, Heller S, Rivera M (2011) Anatomic considerations for central venous cannulation. Risk Manag Healthc Policy 4:27–39 14. Rey C, Alvarez F, De la rua V et al (2009) Mechanical complications during central venous cannulations in pediatric patients. J Intensive Care Med 35(8):1438–1443 15. Halton K, Cook D, Paterson D, Safdar N, Graves N (2015) Costeffectiveness of a central venous catheter care bundle. PLoS One 5(9):e12815 16. Miko B, Kamath S, Cohen B, Jeon C, Jia H, Larson E (2015) Epidemiologic associations between short-bowel syndrome and bloodstream infection among hospitalized children. J Pediatr Infect Dis Soc 4(3):192–197 17. Fallon S, Kim M, Fernandes C, Vasudevan S, Nuchtern J, Kim E (2014) Identifying and reducing early complications of surgical central lines in infants and toddlers. J Surg Res 190(1):246–250 18. Pinon M, Bezzio S, Tovo P, Fagioli F, Farinasso L, Calabrese R, Marengo M, Giacchino M (2009) A prospective 7-year survey on central venous catheter-related complications at a single pediatric hospital. Eur J Pediatr 168:1505–1512 19. Meland N, Wilson W, Soontharotoke C, Koucky C (1986) Saphenofemoral venous cutdowns in the premature infant. J Pediatr Surg 21(4):341–343 20. Pippus K, Giacomantonio J, Gillis D, Rees E (1994) Thrombotic complications of saphenous central venous lines. J Pediatr Surg 29(9):1218–1219 21. Panagiotounakou P, Antonogeorgos G, Gounari E, Papadakis S, Labadaridis J, Gounaris A (2014) Peripherally inserted central venous catheters: frequency of complications in premature newborn depends on the insertion site. J Perinatol 34(6):461–463

123