Annals of Pediatric Cardiology
About us | Current Issue | Archives | Ahead of Print | Instructions | Submission | Subscribe | Advertise | Contact | Login 
 


 

 
     
    Advanced search
 

 
 
     
  
     Search Pubmed for
 
    -  Polat TB
Ahead Of Print
    Email Alert *
    Add to My List *
* Registration required (free)  


    Abstract
   Introduction
   Methods
   Results
   Discussion
   Conclusions
    References
    Article Figures
    Article Tables

 Article Access Statistics
    Viewed91    
    PDF Downloaded3    

Recommend this journal

 

ORIGINAL ARTICLE  
Ahead of print publication
Use of percutaneous carotid artery access for performing pediatric cardiac interventions: Single-center study


 Department of Pediatric Cardiology, Altinbas University School of Medicine, Istanbul, Turkey

Click here for correspondence address and email

 

   Abstract 


Background : Carotid artery (CA) access allows a more straight route for many left heart lesions. This has previously been achieved via a surgical cut-down approach in certain pediatric cardiac interventions. However, there are little data considering CA access in pediatric cases, percutaneously.
Aim : We hypothesized that there would be notable improvements in efficiency as well as overall success when using the CA for access in selected cases.
Methods : Between November 2016 and January 2019, records of patients undergoing attempted percutaneous CA access under ultrasound guidance for cardiac catheterization were reviewed.
Results : Thirty patients underwent 36 catheterizations; median age 17 days (range, 6 days–9 months) and median weight 3.2 kg (1.2–7.8). Procedures performed were stenting or stent redilatation of the patent ductus arteriosus in 23, stenting or angioplasty of modified Blalock–Taussig shunts in four, aortic valvuloplasty in three, angioplasty for coarctation of the aorta in four, renal angioplasty in one, and diagnostic catheterization in one case. The intended intervention was unsuccessful in two patients despite successful CA access. Follow-up imaging showed a normal carotid in 28 of 30 (94%), with mild luminal narrowing with normal Doppler velocities in two instances. No patient had clinically apparent neurological sequelae attributable to CA access.
Conclusions : Our data indicate that CA access should be employed when dealing with a select group of infants requiring vertical approach for left-sided cardiac lesions. Percutaneous CA access, even in very small preterm infants, is safe and feasible with negligible vascular injury and no neurological adverse events.

Keywords: Carotid artery, percutaneous, pediatric cardiac intervention


How to cite this URL:
Polat TB. Use of percutaneous carotid artery access for performing pediatric cardiac interventions: Single-center study. Ann Pediatr Card [Epub ahead of print] [cited 2019 Dec 11]. Available from: http://www.annalspc.com/preprintarticle.asp?id=269739





   Introduction Top


Carotid artery (CA) route achieved via surgical cutdown is rarely preferred for cardiac catheterization by pediatric interventional cardiologists because of concerns for vascular injury or potential risk for cerebrovascular accidents.[1],[2],[3],[4]

However, recent studies have shown that percutaneous CA access has a low rate of vascular complications during interventions and excellent rate of carotid patency on follow-up imaging. Moreover, there was no clinical evidence of neurological adverse events in these papers. Thus, this data exhibit encouraging safety profile of percutaneous CA access for transcatheter pediatric cardiac interventions.[5],[6],[7]

The CA may be a choice in selected cases because of the vertical angle of approach for many left heart lesions including aortic stenosis, aortic arch pathologies, modified Blalock–Taussig (mBT) shunt, and vertical patent ductus arteriosus (PDA). Considering the fact that caliber of the CA is larger in diameter than the femoral artery (FA), CA access can be performed with low risk of vascular complications in small infants.[2],[6],[8]

In this study, we initially adopted a percutaneous CA access in only small infants in order to reduce the vascular complications and unsuccessful cardiac interventions using femoral or axillary arterial access due to the vertical angle of cardiac lesions. Thereafter, we performed percutaneous CA in most aforementioned interventions that provided the ability to maneuver the catheter directly to left-sided cardiac lesions. We hypothesized that there would be notable improvements in efficiency as well as overall success when using the CA for access in selected cases.


   Methods Top


Between November 2016 and January 2019, records of patients undergoing attempted percutaneous CA for cardiac catheterization were reviewed. The retrospective review was approved by the Institutional Review Board of Altinbas University School of Medicine.

The primary indication for initial CA approach included infants weighing <2500 g at increased risk for vessel trauma from FA catheterization,[8],[9] failed interventions via femoral or axillary arterial route due to vertical angle of cardiac lesions, stenting the vertical PDA, mBT shunt interventions, and vertical PDA stent redilatations.

Procedure

All catheterizations were performed under general anesthesia with endotracheal intubation. Heart rate, respiratory rate, saturations, and blood pressure were monitored during the procedure. Percutaneous access was obtained using a 22G needle (Radifocus® Micro puncture metallic entry needle) under ultrasound guidance (12 L linear array vascular probe, GE Medical Systems, Milwaukee, WI, USA) into the CA in all cases. We attempted to access the CA via a single puncture through the anterior wall of the vessel in the anterior triangle of the neck. The bifurcation of the common CA into the internal and external branches was visualized, and a suitable location sufficiently caudal from the bifurcation was identified (typically approximately halfway between the clavicle and the edge of the mandible). Choice of which carotid was used for access was based on intervention; we elected whichever carotid seemed more directly aligned with the structure of interest based on previous noninvasive imaging (usually echocardiography) or patient's particular anatomy as demonstrated by angiography using FA. A 0.018″ soft-tipped wire was advanced once blood return was seen in the needle hub or confirmation of needle entry into the lumen was seen on ultrasound. Fluoroscopy was then used to advance the wire into the ascending aorta and a 4 or 5 Fr sheath (Radifocus® Introducer II Pediatric Kit) was inserted and gently advanced approximately 3 cm into the CA.

All patients received intravenous 100 U/kg heparin sodium immediately after sheath placement, with redosing as needed to maintain an activated clotting time (ACT) of >200 s. Protamine was not administered to reverse heparin effect, even in the setting of supratherapeutic ACT. Hemostasis was obtained by manual compression in all cases.

All patients received routine prophylactic antibiotic therapy (cefazolin, 25 mg/kg, intravenously) during the procedure, followed by two more doses post-procedure.

Following the procedure, the infants were transferred back to cardiac intensive care unit for monitoring. An echocardiogram was performed following the procedure in all infants.

Demographic, procedural, and postprocedural variables were assessed with particular emphasis on technical success in obtaining vascular access and postprocedural vessel patency and neurologic complications.

Statistics

All data are expressed as the median (range) and n (%).


   Results Top


Thirty patients underwent 36 catheterizations via percutaneous carotid access (PCA). Thirty-five procedures were interventional, and the remaining one was diagnostic catheterization. Two patients experienced a second and two patients experienced third interventional catheterization at later times via the same vessel. The median age at catheterization was 17 days (range, 6 days–9 months) and weight was 3.2 kg (1.2–7.8) Procedure demographics are shown in [Table 1].
Table 1: Clinical characteristics of thirty patients undergoing percutaneous carotid artery access for performing pediatric interventional cardiac catheterization

Click here to view


Carotid arterial access was eventually achieved in all patients [Table 2]. A total of 34 catheterizations were successfully accomplished via percutaneous CA. Interventional catheterizations were performed in 33 patients, including the most common being stenting PDA or stent redilatation [n = 21 patients, [Figure 1], balloon aortic valvuloplasty (n = 3) and stenting or angioplasty of mBT shunts [n = 4, [Figure 2], percutaneous transluminal renal angioplasty (n = 1), and balloon angioplasty for coarctation of the aorta (n = 4). Diagnostic catheterization was performed in only one patient. The intended intervention was unsuccessful in two patients referred for ductal stenting despite successful percutaneous CA (Cases 6.2 and 9).
Table 2: Patient characteristics, stent data, and clinical outcome

Click here to view
Figure 1: Aortic valvuloplasty in a 6-day-old neonate with a weight of 1300 g, access via the carotid artery (Case 3). (a) Left anterior oblique projection direct visualization of the balloon dilatation catheter by contrast injection via introducer sheath (b) Aortic valvuloplasty (c) Postprocedural mild aortic regurgitation

Click here to view
Figure 2: Balloon angioplasty of coarctation of the aorta in a 10-day-old neonate. (a) Lateral view of aortogram, access via the carotid artery. Preductal coarctation clarifies the failed retrograde access despite multiple attempts, retrogradely. (b) Predilatation with a 2 mm × 16 mm coronary balloon was applied to the coarctation segment. (c) 6 mm × 20 mm Thysack II balloon was positioned across the coarctation region and dilated. (d) A final aortogram shows diameter of the coarctation region and patent ductus arteriosus (prostaglandin E1 infusion could be discontinued, and the ductus arteriosus closed 2 days later, but the coarctation did not recur with a systolic gradient of 10 mmHg)

Click here to view


Initial access was attained via a different vascular access site in nine patients, including eight via FA and one via axillary artery. CA sheath placement procedures were technically successful in all. Time to complete sheath placement was 3.7 ± 1.7 min. The right carotid arteries were chosen in 13/36 catheterizations (36%). Sheath size used was a 4 Fr in 5 and 5 Fr in 31 cases.

Sheaths remained in the CA for a median of 41 min (range, 17–95 min). The median ACT after sheath placement and heparin administration was 243 s (220–440, n = 36). There were no instances of thrombus formation around the external portion of the sheath and within the catheter sheath after removal.

After the catheterization, sheaths were removed, and hemostasis was achieved by manual compression for a median duration of 9 min (range, 5–22 min).

In all cases, two-dimensional and Doppler ultrasound examination was performed within 24 h of the catheterization. Seven patients having detectable abnormalities: Two had nonocclusive thrombus (Cases 7 and 17), one had mild luminal narrowing caused by a small tissue flap at the arterial puncture site (Case 11.1), one had carotid pseudoaneurysm (Case 14), two patients had mild hematomas at the puncture site without evidence of carotid flow disturbance (Cases 2 and 8), and finally one patient with large hematoma developed at the puncture site causing respiratory distress because of upper airway compromise (Case 12).

These two patients with nonocclusive thrombus were treated with therapeutic doses of subcutaneous low-molecular-weight heparin twice daily. Case 7 with nonocclusive thrombus had restored flow and only mild luminal narrowing with normal Doppler velocities by 2 weeks, and anticoagulation was discontinued. Case 17 had nonocclusive thrombus at 24 h with restoration of flow and mild focal narrowing documented at 7 days and again anticoagulation was discontinued.

Case 11.1 had mild luminal narrowing on the ultrasound performed within 24 h, with a normal carotid by 15 days.

Two patients (Cases 2 and 8) had hematomas at the puncture site without evidence of carotid flow disturbance that resolved without incident; both patients underwent balloon angioplasty for coarctation of the aorta through a 5F sheath. Both had normal CA ultrasounds at 2 weeks follow-up.

Remaining two cases (Cases 12 and 14) required surgical intervention because of a vascular access complication: Case 14 with CA pseudoaneurysm was successfully treated by ligation and Doppler ultrasound scanning showed normal flow pattern during follow-up. Case 12 had the longest time to achieve hemostasis after diagnostic catheterization of the interrupted aortic arch and collateral arteries that arising from the ascending aorta and supply blood to the descending aorta. Pressure was maintained on the carotid puncture site for 22 min, followed by use of a pressure dressing for 3 h. However, a large hematoma developed at the site. In the course of the diagnostic workup, the hematoma enlarged, thereby causing respiratory distress because of upper airway compromise. The endotracheal intubation was lifesaving. Later, contrast-enhanced computed tomography revealed hematoma in the region of the right common CA. Emergency surgery evacuated the hematoma and repaired a pinpoint defect of the left common carotid 1 cm before bifurcation. Blood loss was minimal, and surgical control of bleeding was not required in all remaining cases.

All patients were evaluated by Doppler ultrasound scanning that encompassed the CA up to last available follow-up, ranging from 1 to 27 months after percutaneous CA (median 13 months). Carotid follow-up imaging showed a normal carotid in 28 of 30 (94%), with mild luminal narrowing with normal Doppler velocities in two instances (patients 7 and 17, described above). No patient had clinically apparent neurological sequelae attributable to PCA at any time.


   Discussion Top


It is recognized that access via the CA provides a more direct route for many pediatric left heart interventions. There have been few articles published on the safety of percutaneous puncture of the CA without surgical cutdown.[6],[7],[10] Vida et al. reported forty infants <3 months of age, who underwent primary percutaneous balloon dilatation with a transcarotid approach by direct puncture of the right CA for severe aortic valve stenosis. No complications were related to the transcarotid approach, and a right CA Doppler at discharge was normal in all patients.[5] Recently, Justino and Petit reported 42 patients underwent 47 catheterizations during which ultrasound-guided percutaneous CA access was attempted, with five patients undergoing a second percutaneous catheterization at a later time via the same vessel. At follow-up, 40 of 42 patients (95%) had a normal CA, with two instances of mild stenosis. There were no neurological sequelae attributable to percutaneous CA.[6]

In our series, we performed preliminary percutaneous CA cautiously, at first only in small neonates who have the highest risk of femoral arterial injury (Cases 1, 2, and 3). Given that the neonatal CA is larger in diameter than the FA, CA access may play an important role in preventing the complications of FA thrombosis when applied to this higher risk infant population. After this initial encouraging experience, we began to use percutaneous CA after failure of the FA or axillary arterial approach in older children. Finally, we performed percutaneous CA in all interventions that provided the ability to maneuver the catheter directly to left heart structures with potentially less technical difficulty.

In this context, balloon aortic valvuloplasty was performed in three cases via right CA (Cases 1, 3, and 22). All these three cases were preterm infants weighted <1500 g. Furthermore, the potential benefit of the carotid approach is the ability to maneuver the catheter directly to the left ventricle. Particularly in very small preterm infants, because of close proximity and relatively straight course of the sheath while performing balloon aortic valvuloplasty, direct visualization of the aortic valve by contrast injection via introducer sheath and over the wire ballooning may be achieved possibly shortening cardiac catheterization time and reducing catheter-based complications [Case 22, [Figure 1]].

Similarly, we opted for the left CA access in three neonates with severe aortic coarctation. All four cases were hemodynamically frail at the time of the coarctation intervention. The interventions were performed within 5 min in three very low birth preterm after direct visualization of lesion via sheath placed in the left CA (Cases 2, 16, and 28). In the remaining cases, severe coarctation lesson could not be accessed retrogradely due to preductal location (Case 8). In this case, guidewire was always accessed across the PDA despite multiple attempts. Thus, coarctation dilatation procedures were performed via left CA [Figure 2].

It is shown that access via the CA allows a more straight route for interventions targeting at some mBT shunts, specifically those mBT shunt that arise from the common CA itself.[2],[6],[10] Similarly, mBT shunt arose from the common CA in three of four cases, the decision was made to intervene on the mBT shunt using the CA after multiple attempts were made through FA access (Cases 6.3, 10, and 20) [Figure 3]. In the remaining cases, mBT shunt arose from the innominate artery (Case 4.3). However, directing the stent through the mBT shunt was impossible via FA approach because of the tight angle between the shunt and the innominate artery. In all cases, access was obtained from the CA approach during the same catheterization. Subsequent successful mBT intervention was achieved in all.
Figure 3: Modified Blalock–Taussig shunt stenting in an infant with pulmonary atresia-ventricular septal defect and cyanosis. Ductal stenting and stent redilatation were performed previously (a) The right carotid artery was used for angiography (Case 4), demonstrating the region of narrowing at the systemic arterial end of the modified Blalock–Taussig shunt (black arrow) and occluded ductal stent (white arrow) (b) The stent was passed across the site of stenosis and positioned along the length of modified Blalock–Taussig shunt (c) Postexpansion angiogram, resolution of the shunt stenosis required only 12 min

Click here to view


Among the nonconventional vascular sites, carotid and axillary arteries are the most expedient routes for stenting PDA with complex morphology. Axillary artery route was the favored approach in our earliest cases.[11] However, it is a slightly less direct route to the PDA and the puncture site at the apex of the axilla is little distance away from the pleural cavity. For the PDA morphology in question, the most direct approach that would eliminate acute curves of the guidewire for stent delivery is via the common CA. The angle of approach from the CA is much straighter than a FA or axillary artery approach and this facilitates wire and stent passage across tortuous anatomy and provides more back up support when needed. Stenting or stent redilatation of the ductus arteriosus was the most common intervention in our series [n = 23 patients, [Figure 4]. The intended intervention was unsuccessful in two patients despite successful percutaneous CA. Reintervention was performed in Case 6.2 due to extreme ductal stent stenosis, but the stent could not be accessed. In Case 9, stent migration occurred while retrieving the guidewire. Additional manipulation intended to improve the stent apposition, but migrated stent could not be accessed. Thus, inadequate flow occurred soon after the procedure, due to incomplete stent coverage of the ductus. In both cases, the right-sided mBT shunt was performed soon after intervention. Case 9 died 2 days after operation due to cardiac decompensation (low cardiac output state).
Figure 4: Ductal stent redilatation in an infant with pulmonary atresia with ventricular septal defect and cyanosis (Case 11) of the ductus arteriosus. (a) In-stent stenosis: layer of neointimal proliferation (black arrow); (b) stenosis of unstented segment of a long vertical ductus arising from the left subclavian artery; (c) Postexpansion angiogram

Click here to view


Case 18 has the left renal artery stenosis diagnosed with digital subtraction angiography. The patient has malignant systemic hypertension being measured in all the four limbs. Levels of rennin, human atrial natriuretic peptide, and brain natriuretic peptide were all markedly elevated in the plasma. At an initial attempt to perform percutaneous transluminal angioplasty of the stenotic artery through the right FA. Through a 5F Judkins right-guiding catheter, the stenotic lesion could not be crossed with a 0.014-inch floppy wire because of the steep angle of branching of the renal artery from the descending aorta. Thus, the second attempt at transluminal angioplasty was performed via the left CA. A 0.014 floppy guidewire was easily inserted into the narrowed left renal artery. Balloon dilatation was then performed using 1.5-mm fixed balloon [Figure 5]. The use of an antegrade approach for renal artery interventions overcome most of the problems, particularly the small size of the renal arteries, and their steep angle of branching from the descending aorta, faced when attempting percutaneous transluminal angioplasty in small infants with renovascular hypertension.[12],[13] Hence, percutaneous renal artery angioplasty is reasonable via CA access in infants. This approach provides a more direct route to the renal artery.
Figure 5: Percutaneous transluminal angioplasty in a 3-month-old infant with left renal artery and malignant systemic hypertension (Case 18) (a) The right femoral artery was used demonstrating the short stenotic lesion, almost complete (95%) when compared to the diameter of the adjacent healthy segment of the left renal artery (white arrow) and steep angle of branching of the artery from the descending aorta. (b) The catheter advanced from the right carotid artery, mounting a 1.5 mm balloon, was passed across the site of stenosis and inflated until the waist of the balloon disappeared. (c) After the angioplasty, the ratio of stenosis was reduced from 95% to 30%

Click here to view


Removal of a percutaneously placed sheath with manual compression of the carotid puncture site is safe and successful at achieving hemostasis, and this occurred without reversal of heparin. Only two cases including one with pseudoaneurysm and the other with a large hematoma required surgical intervention. Both patients had high blood pressures during catheterization. Doppler ultrasound scanning showed normal flow pattern during follow-up. There were no documented neurologic sequelae attributed to CA access in the immediate postprocedure period; however, a detailed evaluation has not been performed.

Limitations

Ultrasound-guided vascular access is not a method of choice for many pediatric heart centers. As with any new practice, there was a learning period where these new methods required more time to be performed. In our clinic, ultrasound guidance has been used for FA access in most cases for 4 years. This provided for this study a practical experience in the arterial access. This is a small cohort from a retrospective single-center study. Thus, the results may not be necessarily generalizable to every Pediatric Heart Center. In addition, it would be early to state that CA access is safe in the absence of data looking at long-term outcomes, including subtle changes such as cognitive functions. Nevertheless, potential occlusion of CA by the arterial cannula does not appear to increase the risk of neurologic injury in infants undergoing venoarterial extracorporeal membrane oxygenation in recent studies.[14],[15]


   Conclusions Top


Ultrasound-guided percutaneous CA, access even in very small preterm infants, is feasible and should be employed when dealing with a select group of cases required vertical approach for left-sided cardiac lesions. Carotid follow-up imaging showed a normal flow velocity in all cases and negligible vascular injuries in few cases.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.



 
   References Top

1.
Robinson BV, Brzezinska-Rajszys G, Weber HS, Ksiazyk J, Fricker FJ, Fischer DR, et al. Balloon aortic valvotomy through a carotid cutdown in infants with severe aortic stenosis: Results of the multi-centric registry. Cardiol Young 2000;10:225-32.  Back to cited text no. 1
    
2.
Davenport JJ, Lam L, Whalen-Glass R, Nykanen DG, Burke RP, Hannan R, et al. The successful use of alternative routes of vascular access for performing pediatric interventional cardiac catheterization. Catheter Cardiovasc Interv 2008;72:392-8.  Back to cited text no. 2
    
3.
Fischer DR, Ettedgui JA, Park SC, Siewers RD, del Nido PJ. Carotid artery approach for balloon dilation of aortic valve stenosis in the neonate: A preliminary report. J Am Coll Cardiol 1990;15:1633-6.  Back to cited text no. 3
    
4.
Maeno Y, Akagi T, Hashino K, Ishii M, Sugimura T, Takagi J, et al. Carotid artery approach to balloon aortic valvuloplasty in infants with critical aortic valve stenosis. Pediatr Cardiol 1997;18:288-91.  Back to cited text no. 4
    
5.
Vida VL, Bottio T, Milanesi O, Reffo E, Biffanti R, Bonato R, et al. Critical aortic stenosis in early infancy: Surgical treatment for residual lesions after balloon dilation. Ann Thorac Surg 2005;79:47-51.  Back to cited text no. 5
    
6.
Justino H, Petit CJ. Percutaneous common carotid artery access for pediatric interventional cardiac catheterization. Circ Cardiovasc Interv 2016;9:e003003.  Back to cited text no. 6
    
7.
Choudhry S, Balzer D, Murphy J, Nicolas R, Shahanavaz S. Percutaneous carotid artery access in infants <3 months of age. Catheter Cardiovasc Interv 2016;87:757-61.  Back to cited text no. 7
    
8.
Glatz AC, Shah SS, McCarthy AL, Geisser D, Daniels K, Xie D, et al. Prevalence of and risk factors for acute occlusive arterial injury following pediatric cardiac catheterization: A large single-center cohort study. Catheter Cardiovasc Interv 2013;82:454-62.  Back to cited text no. 8
    
9.
McMahon CJ, Price JF, Salerno JC, El-Said H, Taylor M, Vargo TA, et al. Cardiac catheterisation in infants weighing less than 2500 grams. Cardiol Young 2003;13:117-22.  Back to cited text no. 9
    
10.
Ligon RA, Ooi YK, Kim DW, Vincent RN, Petit CJ. Intervention on surgical systemic-to-pulmonary artery shunts: Carotid versus femoral access. JACC Cardiovasc Interv 2017;10:1738-44.  Back to cited text no. 10
    
11.
Polat TB. Stenting the vertical ductus arteriosus via axillary artery access using “wire-target” technique. Congenit Heart Dis 2017;12:800-7.  Back to cited text no. 11
    
12.
Takigiku K, Satomi G, Yasukochi S. Successful transluminal angioplasty of renal arterial stenosis using the transcarotid approach. Cardiol Young 2002;12:589-91.  Back to cited text no. 12
    
13.
Courtel JV, Soto B, Niaudet P, Gagnadoux MF, Carteret M, Quignodon JF, et al. Percutaneous transluminal angioplasty of renal artery stenosis in children. Pediatr Radiol 1998;28:59-63.  Back to cited text no. 13
    
14.
Lamers LJ, Rowland DG, Seguin JH, Rosenberg EM, Reber KM. The effect of common origin of the carotid arteries in neurologic outcome after neonatal ECMO. J Pediatr Surg 2004;39:532-6.  Back to cited text no. 14
    
15.
Wagner K, Risnes I, Berntsen T, Skarbø AB, Ramberg B, Vandvik IH, et al. Clinical and psychosocial follow-up study of children treated with extracorporeal membrane oxygenation. Ann Thorac Surg 2007;84:1349-55.  Back to cited text no. 15
    

Top
Correspondence Address:
Tugcin Bora Polat,
Department of Pediatric Cardiology, Altinbas University School of Medicine, Istanbul
Turkey
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/apc.APC_26_19



    Figures

  [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
 
 
    Tables

  [Table 1], [Table 2]



 

Top
Previous Article Next Article