Infected chronic sinus secondary to a retained fragment of radial artery introducer sheath following percutaneous coronary intervention (PCI)

Abstract

Coronary angiography and percutaneous coronary intervention (PCI) are frequently performed procedures in the UK and the developed world, with the radial artery becoming the preferred route of access. A chronically retained macroscopic fragment of radial artery introducer sheath is a very rare complication that has not, to our knowledge, been reported. We report the case of a 62-year-old woman who underwent PCI and developed a persisting infected sinus and abscess at the cannulation site despite multiple courses of antibiotics. Surgical exploration of the forearm recovered a foreign body that was found in the brachioradialis muscle and resembled a fragment of hydrophilic sheath. In conclusion, this case highlights that it is possible to leave macroscopic fragments of hydrophilic sheaths in situ. This is likely to be encountered during difficult access, especially during arterial spasm, and it is advised that the sheath and any other vascular access device is thoroughly inspected following removal.

Background

Coronary angiography and percutaneous coronary intervention (PCI) are carried out to diagnose and manage ischaemic heart disease in a variety of presentations ranging from stable angina to acute coronary syndromes. The British Cardiovascular Intervention Society recorded 92 589 PCIs in 2013 for the UK.¹

First described by Campeau in 1989, the radial artery is becoming the preferred site of vascular access over the femoral and brachial approaches due to its lower rate of vascular complications.^1–4 The radial artery is easily accessible and it is not an end artery as the hand is perfused by the anastomosis of the ulnar and radial arteries, forming the deep and superficial palmar arches; the adequacy of the ulnar artery to perfuse the hand can be assessed through Allen’s test. Nevertheless, this approach has recognised difficulties and complications.^4–9

Arterial spasm is a recognised potential event during transradial catherisation that causes procedural difficulty and pain. Hydrophilic-coated sheaths have been developed to reduce this problem, and prophylactic calcium channel antagonists or nitrates are generally administered to help ameliorate this spasm.⁵ Hydrophilic-coated sheaths are easier to remove and cause less pain and radial artery spasm.^5–7 Bleeding, haematoma, dissection and thrombosis are the main complications and heparin is administered to reduce the risk of thrombosis.⁸ Perforation of the radial artery is also a recognised but rare complication of radial artery access that can cause several other complications such as: bleeding at a remote site to access, arteriovenous (AV) fistula, pseudoaneurysm and compartment syndrome.⁹

Infection is a rarely reported complication of radial artery access; Rathore and co-workers have reported a significantly higher incidence of late (occurring at 2–4 weeks) local infection and abscess formation at the puncture site for patients randomised to hydrophilic polymer-coated sheaths that generally resolved with antibiotics.⁷ Furthermore, Kozak and colleagues observed, for 30 patients, an association with hydrophilic sheath usage and the formation of a violet swelling at the original puncture site (at 2–3 weeks; range 3 days' to 3 months' postprocedure).¹⁰ Histopathology showed acute and chronic inflammation but with no evidence of infection; there was, however, an amorphous extracellular material that was being engulfed by macrophages, thought to be the gel hydrophobic coating and the cause of the sterile abscess (foreign body reaction). Nevertheless, there are no reports of chronic infection that is secondary to a retained macroscopic fragment of hydrophilic radial artery sheath, and to our knowledge this is the first case report of this avoidable complication.

This is a case report of a chronic infection, resulting in a persisting sinus and recurrent abscesses, secondary to a macroscopic fragment of a radial artery sheath. This patient underwent PCI for unstable angina.

Case presentation

A 62-year-old Caucasian woman (height: 152 cm; weight: 63.5 kg and body mass index: 27.4 kg/m²) was referred to us from a vascular surgeon with a suspected right forearm abscess and sinus following a right-sided transradial coronary angioplasty, performed 5 months prior at another centre for unstable angina. Her past medical history was ischaemic heart disease (three previous myocardial infarcts and coronary stenting), hypertension and hypercholesterolaemia.

Vascular access (Seldinger technique) was via the right radial artery with a 6 Fr (Terumo Europe, Belgium) introducer sheath. Five thousand international units of unfractionated heparin followed by 500 μg of isosorbide dinitrate were initially administered intra-arterially. Six French JR4 and 5 French JL3.5 catheters were selected for the procedure.

Radial artery spasm was encountered, which is not unusual, and this was managed by performing a radial artery angiogram and by administering further isosorbide dinitrate intra-arterially (1500 μg); a balance middle weight (BMW) coronary wire was also used to navigate the radial artery. Four French diagnostic and 5 French guide catheters were then used to complete the procedure. PCI was performed to the distal right coronary artery and a drug-eluting stent was deployed. The procedure was then completed in a standard fashion using a TR band (Terumo Europe, Belgium) to the right radial artery.

Five-weeks following the angioplasty, this woman presented to her general practitioner (family physician) with cellulitis over the radial puncture site and she was managed initially with oral Erythromycin (known previous penicillin allergy). Despite multiple courses of antibiotics which included oral Clindamycin (300 mg three times a day [TDS] per oral [PO]) then Linezolid (600 mg, twice daily [BD] PO), there was no significant clinical improvement and she was eventually referred to the vascular surgeons from the physicians with a suspected arteriovenous (AV) fistula on ultrasound scan; however, a repeated scan, arranged by vascular surgery, 2 months later (5 months post-PCI) excluded a fistula and demonstrated a progressing collection from the right radial artery along the veins of the forearm to the elbow. This scan also showed two linear foci consistent with foreign bodies measuring 15 mm within a vein on the lateral aspect of the proximal forearm (figure 1). Following this scan, this patient was referred to our upper limb orthopaedic team via vascular surgery for definitive surgical management of her infection.

Figure 1.

Ultrasound scan of the right forearm, demonstrating two linear luminal abnormalities approximately 15 mm in length, which are reminiscent of foreign bodies.

When she was reviewed in the upper limb clinic for the first time, she was approximately 6 months post-PCI. Clinical examination revealed a discharging sinus that was 4–5 cm proximal and volar to the radial aspect of the wrist and a non-pulsatile tender collection, consistent with an abscess, proximal to this, measuring approximately 5×5 cm, with no associated cellulitis; Allen’s test was normal. Her white cell count (WCC) was 9.3×10⁹/L (normal range 4–11×10⁹/L) and her C-reactive protein (CRP) was 7.9 mg/L (normal range 0–5 mg/L).

Investigations

An MRI scan (figure 2) guided definitive management. This scan revealed signs of severe chronic thrombophlebitis on the radial side of the forearm with distended thick-walled subcutaneous veins containing enhancing thrombus and inflammatory changes at the interface between the subcutaneous tissues and the lateral margin of the flexor muscle compartment. Moreover, there was evidence of two well-circumscribed linear non-enhancing abnormalities measuring 12 mm in length, localised 15 cm proximal to the wrist joint, and radiologically resembled a retained catheter tip. There were no signs of an AV fistula.

Figure 2.

T1-weighted sagittal MRI with contrast, demonstrating a well-circumscribed linear non-enhancing focus measuring 12 mm in length, localised 15 cm proximal to the wrist joint.

Differential diagnosis

The differential diagnosis for this case included initially an AV fistula or pseudoaneurysm, but following further investigations (ultrasound scan and MRI) these were excluded. An abscess with or without retention of a foreign body was the working diagnosis.

Treatment

Surgical exploration (under general anaesthesia) was electively planned, at just over 8 months from her PCI, and involved excising the sinus that was approximately 5 cm proximal and radial to the wrist joint, on the volar surface of the forearm. A patent tract was traced proximally from the sinus to the brachioradialis muscle, which was approximately 15 cm proximal to the radial styloid. At this point was the fragment of a radial artery sheath, measuring approximately 1 cm in length, and this was found lying in pus in the brachioradialis muscle (figures 3 and 4).

Figure 3.

Intraoperative photograph showing the hydrophilic sheath within the right brachioradialis muscle.

Figure 4.

Two surgical specimens. On the left is the hydrophilic radial artery sheath, and on the right is the excised tract of chronically inflammed tissue.

This woman had normal perfusion to her hand postoperatively with a strong radial pulse and was discharged from hospital on postoperative day one with no antibiotics.

Outcome and follow-up

The patient was reviewed in clinic 2 weeks postoperatively and the wound appeared healthy; however, at approximately 6 weeks postoperatively, she presented with a 2-day history of swelling around the scar and a further abscess was diagnosed, measuring 1×1 cm at the junction of the middle and distal thirds of the scar. An ultrasound scan confirmed a 2×0.5 cm fluctuant collection. An attempted aspiration was performed but no pus was drained.

She presented to the emergency department approximately 4 weeks (10 weeks postoperatively) later with signs of cellulitis (WCC was 10.41×10⁹/L and CRP was 4.0 mg/L) and an abscess that required further surgical exploration, incision and drainage. Interestingly, a further 1 cm foreign body was removed, which resembled a further sheath fragment and presumably the second fragment which was noted on the initial MRI scan. She was prescribed a course of Clindamycin postoperatively (600 mg, TDS PO). Approximately 1 month later, she was reviewed in clinic and the wound was healed with no signs of infection.

At approximately 8 weeks since she was last reviewed (~22 weeks from the first surgical exploration and 12 weeks from the second surgical intervention), she presented with a third recurrent abscess that required further surgical incision and drainage.

Prior to drainage, an MRI confirmed signs of a collection, particularly in the distal muscle belly of the brachioradialis muscle measuring 55×25×17 mm. Approximately 6 weeks later, the wound had healed without any further significant concerns of infection. This woman was discharged from our care approximately 1½ years after she was first reviewed in the upper limb clinic.

Histology from the excised tissue was consistent with an inflammatory reaction to a foreign body. Polarising light microscopy was positive for birefringent material, present with moderately abundant neutrophils, in keeping with retained plastic material, reported from the second procedure to measure 5×1×1 mm. Deep pus cultures from the first surgical intervention were positive for a mixed growth of Staphylococcus aureus, Streptococcus anginosus and Staphylococcus epidermidis (sensitive to Erythromycin). Pus cultures from the second surgical exploration were positive for Streptococcus gordonii (sensitive to Clindamycin).

Discussion

This case has demonstrated a rare complication of coronary angiography and PCI with significant implications for the patient, causing chronic infection and the need for three surgical interventions. To our knowledge, this case is the first report of a chronic infection that is likely due to a retained macroscopic fragment of hydrophilic sheath.

There have been reports of chronic inflammation associated with hydrophilic polymer coated vascular sheaths, but there are no reports of a chronically retained macroscopic fragment. Athauda-Arachchi and Dorman reported their case where traction during arterial spasm resulted in the fracture of a hydrophilic sheath from the main hub, and an attempt at removing this against spasm fractured the sheath further.¹¹ This necessitated vascular surgical intervention and retrieval of the sheath but resulted in proximal and distal ligation of the radial artery, though with no long-term complications. Kozak et al ¹⁰ reported a case series in which sterile abscesses were found in association with microscopic shedding of the sheath’s hydrophilic polymer coating. In our case however, the pus was associated with several species of Staphylococcus, implying the infection may be secondary to the chronically retained macroscopic fragment of sheath. It is difficult to establish the exact mechanism of how this fragment was noted to be extra-arterial during surgical exploration, but it appeared that vascular access was difficult, requiring advanced techniques to navigate the radial artery and a further dose of isosorbide dinitrate (1500 mg). We can suspect that the wall was unintentionally perforated through and through leading to a retained fragment of sheath, though as this procedure was performed at another centre, we can only speculate on the cause and the nature of the retrieved foreign body.

Our case report provides a significant learning point for cardiologists and other interventionists who access the radial artery using hydrophilic polymer coated sheaths, highlighting that it is possible to leave macroscopic fragments of sheath in situ. Furthermore, general practitioners and other clinicians should consider the possibility of a retained sheath for cases of chronically infected angioplasty wounds not responding to antibiotics, and these patients should be referred for ultrasonography as appropriate.

Learning points.

• This case is evidence that macroscopic fragments of hydrophilic sheath can be left in situ, particularly where arterial spasm is encountered during difficult access.

• It is advised that the end of the sheath and any other vascular access device is thoroughly inspected on its removal, particularly where there is difficult access, and where access is particularly difficult, the operator should be mindful to the fact that devices can fracture in situ.

• A retained fragment at the access site that is identified, managed and removed on the day of the procedure is unlikely to result in significant infection and the need for several surgical interventions.

• For cases of chronically infected angioplasty wounds that are failing to resolve with antibiotics, general practitioners and other medical professionals such as cardiologists should consider the possibility of a retained foreign body.