Acute Limb Ischemia Caused by Distal Embolization from an External Iliac Artery Pseudoaneurysm Associated with a Protruding Acetabular Screw: A Case Report

Abstract

This case report describes a 78-year-old female who presented with acute limb ischemia caused by distal embolization 23 years after total hip arthroplasty (THA). The embolic source was a pseudoaneurysm of the left external iliac artery (EIA) densely adherent to thickened soft tissue in contact with a protruding acetabular screw. The affected EIA segment was replaced with a prosthetic graft, and an above-knee popliteal-to-posterior tibial artery bypass was performed for infrapopliteal occlusion. Histopathology revealed disruption of the elastic laminae, replacement of the arterial wall by granulation tissue with prominent hemosiderin deposition, and organizing thrombus, confined to the side facing the screw, whereas the opposite wall was preserved. Vascular complications of THA are uncommon and usually occur perioperatively; however, chronic mechanical irritation from a protruding acetabular screw may contribute to delayed pseudoaneurysm formation and distal embolization. The patient remained free of recurrent pseudoaneurysm, embolization, and graft stenosis throughout 10 years of follow-up.

INTRODUCTION

Total hip arthroplasty (THA) is widely performed for hip joint disorders. Vascular complications after THA are uncommon, with severe arterial injury reported in approximately 0.2%-0.3% of reconstructive hip operations in an early series [1]. These complications are clinically important because mortality and limb-loss rates of 7% and 15%, respectively, have been reported among patients with vascular injuries [2]. The external iliac artery (EIA) is the vessel most often involved, accounting for 36 of 68 documented cases in one series [2].

Most reported vascular complications occur perioperatively and include hemorrhage and pseudoaneurysm, usually caused by direct vascular injury from THA-related devices. Here, we report a delayed complication that developed 23 years after THA, in which a pseudoaneurysm of the EIA adjacent to a protruding acetabular screw caused embolization to the infrapopliteal arteries. The patient provided written consent at the time of discharge for her information to be used for research and publication.

CASE

A 78-year-old female presented with sudden coldness and rest pain in the left lower extremity. She had undergone left THA 23 years earlier for congenital hip dysplasia. Her medical history included a treated gastric ulcer and surgically treated breast cancer with no evidence of recurrence, but no cardiovascular risk factors or arrhythmia. An orthopedic surgeon evaluated her 2 days after symptom onset and referred her to our department on the same day. Because her symptoms were improving, antiplatelet therapy was started with cilostazol 100 mg twice daily and clopidogrel 75 mg once daily, and she was followed as an outpatient at her request. Her symptoms worsened 3 days later, prompting emergency admission. Surgery was performed the following day, 6 days after onset.

On admission, the left femoral and popliteal pulses were palpable, but pedal pulses were absent. Sensation was preserved, and no motor deficit was present. Arterial Doppler signals were inaudible at the ankle, whereas venous signals remained faintly audible. The limb was therefore classified as Rutherford category IIa. The ankle-brachial index (ABI) was recorded as 0 on the left because no ankle arterial pressure was detectable, compared with 1.05 on the right. Pulse volume recording showed a flat waveform at the left ankle. The serum creatine kinase level was elevated at 925 U/L (reference range of 41-153 U/L), consistent with ischemic muscle injury.

Computed tomography angiography (CTA) showed an eccentric intraluminal filling defect causing severe stenosis of the left EIA, together with occlusion of the distal popliteal artery extending into the origin of the anterior tibial artery (ATA) and the proximal tibioperoneal trunk (popliteal trifurcation) (Fig. 1A). Characterization of the EIA lesion was limited by metal artifact from the prosthesis. An acetabular screw protruded medially into the pelvis near the left EIA and external iliac vein, although direct contact with the arterial wall could not be demonstrated (Fig. 1B, C). Transthoracic echocardiography showed no intracardiac thrombus, and computed tomography showed no shaggy aorta, making a cardiac or aortic embolic source less likely.

Fig. 1.

Preoperative computed tomography angiography. (A) Coronal maximum-intensity projection showed severe stenosis of the left external iliac artery (EIA) (arrowhead) and occlusion at the popliteal trifurcation (arrow). (B) Coronal reconstruction showed the protruding acetabular screw (arrowhead) near the left EIA and external iliac vein (EIV). (C) Oblique reconstruction showed the same screw (arrowhead) projecting medially from the acetabular component near the common femoral artery (CFA). Direct contact between the screw and the arterial wall was not demonstrable, and the arterial lesion could not be characterized because of metal artifact from the prosthesis.

Given the 6-day interval from symptom onset and the extent of the distal arterial occlusion, complete and durable revascularization by thrombectomy alone was considered uncertain. Because a suitable great saphenous vein was available, surgical bypass to the posterior tibial artery (PTA) was planned to provide reliable distal perfusion, while the EIA lesion was to be assessed and treated according to the intraoperative findings. With the patient in the supine position, the above-knee popliteal artery (AKPA) was exposed for angiography. Angiography revealed severe stenosis of the distal left EIA, with an eccentric filling defect projecting into the lumen, which was interpreted as intraluminal thrombus (Fig. 2A). Because the underlying lesion could not be characterized and no appropriate covered stent was available at that time, endovascular repair was considered unsuitable. The distal EIA was therefore exposed directly through an inguinal incision.

Fig. 2.

Intraoperative angiography. (A) An eccentric filling defect was present in the distal left external iliac artery immediately adjacent to the acetabular screws of the hip prosthesis (arrowhead). (B) The popliteal trifurcation was completely occluded, with distal vessels reconstituted through collaterals (arrowhead).

The infrapopliteal vessels were then assessed (Fig. 2B). The ATA was occluded proximally, and the peroneal artery was occluded at its midportion, whereas the PTA provided the best distal runoff to the foot. The PTA was also the most accessible target through a medial approach with the patient already supine. Therefore, the PTA was selected as the distal anastomotic site.

The protruding acetabular screw contacted thickened surrounding soft tissue, and the EIA was densely adherent to this tissue at the same site, whereas the proximal artery appeared normal (Fig. 3A). The adherent arterial segment showed aneurysmal change corresponding to the screw position, and part of the wall appeared close to rupture. After arteriotomy, organized thrombus filled the lumen, supporting this lesion as the likely embolic source (Fig. 3B).

Fig. 3.

Intraoperative findings and postoperative computed tomography angiography (CTA). (A) The external iliac artery (arrowhead) was densely adherent to thickened soft tissue at the site where the acetabular screw protruded (arrow). (B) After the external iliac artery was opened, the lumen was filled with organized thrombus (arrowhead). (C) Postoperative CTA confirmed patency of the saphenous vein bypass graft from the above-knee popliteal artery to the posterior tibial artery.

The affected segment was resected and replaced with an 8-mm GORE PROPATEN Vascular Graft (W. L. Gore & Associates) as an end-to-end interposition graft using continuous 6-0 polypropylene sutures. Because the screw lay in direct contact with the tissue adjacent to the proximal anastomosis and recurrent aneurysmal degeneration was a concern, the anastomosis was reinforced with a 1-cm polyester felt strip (FPX-100, Matsuda Ika Kogyo). The screw was not removed. A bypass from the AKPA to the PTA was then performed using the ipsilateral great saphenous vein, which measured 3.5-5 mm in diameter before dilatation, in a nonreversed configuration. The valves were disrupted using a LeMaitre Valvulotome (LeMaitre Vascular). Final graft flow was 65 mL/min. The operation lasted 8 hours 11 minutes, with blood loss of 576 mL.

Pathological examination of the excised EIA specimen demonstrated a pseudoaneurysm with adventitial fibrosis on the side adjacent to the protruding screw. The internal and external elastic laminae were disrupted, and granulation tissue with prominent hemosiderin deposition entirely replaced the arterial wall. Organizing thrombi adhered to the luminal surface of the pseudoaneurysm. In contrast, the wall opposite the screw-contact site showed only mild intimal thickening and no thrombus, with preservation of the media. No atherosclerotic plaque was identified. No abscess, purulent material, or other macroscopic evidence of infection was encountered intraoperatively. Therefore, tissue cultures were not obtained, and histological examination showed neither acute inflammatory cell infiltration nor microorganisms (Fig. 4).

Fig. 4.

Pathological findings. (A) Macroscopically, thrombus was present within the resected external iliac artery specimen (arrowhead); the intima and adventitia were labeled, and the specimen was sectioned perpendicular to the arterial axis (yellow line). (B) Elastica Masson staining showed a pseudoaneurysm with organizing thrombi and adventitial fibrosis at the site adjacent to the protruding acetabular screw (arrow). In contrast, the arterial wall opposite the screw-contact site showed a relatively preserved structure with only mild intimal thickening (arrowhead). (C) At the pseudoaneurysm site (boxed area in B), granulation tissue with prominent hemosiderin deposition replaced the arterial wall. Organizing thrombus was continuous with the granulation tissue at the upper margin of the field. Scale bars=1 mm (B), 250 µm (C).

Postoperatively, the symptoms resolved completely, and the ABI improved to 0.99. CTA confirmed patency of the saphenous vein bypass graft (Fig. 3C). The creatine kinase level progressively decreased to 578 U/L on postoperative day 1 and 368 U/L on postoperative day 2. Clopidogrel 75 mg once daily was continued. Because the embolic source had been resected, anticoagulation was not administered. Graft surveillance was performed using duplex ultrasonography every 3 months for the first 2 years, every 6 months during years 3-5, and annually thereafter. During 10 years of follow-up, no pseudoaneurysm recurrence, recurrent embolization, stenosis, or occlusion of either graft occurred.

DISCUSSION

This case illustrates a delayed complication of THA: embolization to the infrapopliteal arteries from an EIA pseudoaneurysm adjacent to a protruding acetabular screw (Fig. 1B, C). The screw contacted thickened soft tissue, which was densely adherent to the EIA. Direct contact between the screw and the arterial wall was not demonstrated on imaging or intraoperatively, and the screw did not correspond precisely to the site of maximal narrowing; therefore, its contributory role is probable but not proven. However, the aneurysmal change and histological abnormalities were confined to the side of the vessel facing the screw, with disrupted elastic laminae and hemorrhagic granulation tissue on that side and preserved media on the opposite side. This spatial relationship, together with the absence of atherosclerotic change, suggests that chronic mechanical stress transmitted through the interposed soft tissue contributed to aneurysmal degeneration.

Two perioperative mechanisms of thrombosis after THA have been described: intimal injury caused by heat from cement polymerization and rupture of an atherosclerotic plaque under mechanical stress [1,3]. The present case fits neither mechanism and instead supports a third, chronic mechanical pathway. Comparable delayed cases have been reported 14 years after THA [4]; 2 years after revision THA caused by a bone spicule and presenting with acute limb ischemia [5]; and 10 years after THA with a protruding screw [6]. A further case occurred 15 years after THA, caused by medial wall penetration by a transacetabular screw; Wilson et al. [7] described this as the longest reported interval and the only late vascular injury in an aseptic THA with well-fixed components at that time. At 23 years after THA, the present case extends that interval.

The abrupt distal occlusion was segmental and considered embolic rather than acute-on-chronic ischemia. The patient had no cardiovascular risk factors, the resected EIA specimen showed no atherosclerotic change, and no cardiac or aortic embolic source was identified. Collateral filling on angiography (Fig. 2B) was therefore considered more likely to reflect the interval since the embolic event than underlying chronic occlusive process.

1) Management of the causative screw

Management of the causative screw is important for preventing recurrence. The screw was left in place because removal would have required exposure and possible revision of an acetabular component that had functioned well for 23 years. This additional procedure was considered disproportionate for a 78-year-old female already undergoing arterial replacement and distal bypass during an operation lasting more than 8 hours. We therefore prioritized control of the embolic source, reinforced the proximal anastomosis with a felt strip because the anastomosis lay against tissue overlying the screw, and retained the screw under surveillance.

This approach differs from previous reports, in which the offending structure or its protruding portion was removed in each case and removal was considered necessary to prevent recurrence in one case [5-7]. Removal itself carries risk. Wilson et al. [7] recommended excluding the arterial lesion before screw removal, citing a case in which removal caused uncontrolled hemorrhage from an unrecognized arterial perforation. In the present case, the diseased artery was resected and replaced, directly addressing the embolic source. No graft-related complication occurred during 10 years of follow-up despite retention of the protruding screw. To our knowledge, this is the longest reported follow-up for such a case and the only report in which the causative structure was retained. However, our experience should not be interpreted as evidence that retention is generally preferable.

Transacetabular screws are conventionally placed within the posterior acetabular quadrants, which are generally considered safer for screw placement. The anterior quadrants are avoided because the external iliac vessels run directly against the medial wall. Retrospective review of the imaging suggested that the implicated screw had most likely been placed anteromedially, outside this safe zone, and was longer than appropriate for that position.

Although screw removal is technically straightforward in isolation and the cup depends on screw fixation only until bone ingrowth is established, removal in this setting would not have been an isolated procedure. Access would have required incision of the healed pseudocapsule, dislocation, and exchange of the polyethylene liner. A liner compatible with a 23-year-old design might no longer have been available, in which case cup revision would also have been required. Disrupting healed soft tissue would have increased the risk of dislocation. In addition, a screw retained in situ for two decades may be fractured or have damaged threads and therefore may be irretrievable. Because no loosening was evident, revision was considered unwarranted, and retention was considered reasonable. However, a third option was not considered at the time. Dividing only the protruding tip with a high-speed burr during the vascular procedure might have removed the offending prominence without disturbing the arthroplasty. Resection of the protruding portion of a screw has been described previously [6]. In retrospect, this option may merit consideration in comparable cases.

Routine radiographic follow-up may be appropriate for an asymptomatic retained screw, whereas the vascular imaging interval should be determined by graft surveillance requirements. The proximity of the graft to the retained screw also warrants attention during long-term follow-up.

The screw implicated in the case reported by Wilson et al. [7] also lay in the anterior region. When a medial screw is necessary for cup orientation, its length should be limited to the minimum required. Anteroposterior radiographs have low sensitivity for detecting intrapelvic penetration, whereas a three-view fluoroscopic series has been shown to detect 5 mm of penetration in the anterior quadrants with high sensitivity and may be used intraoperatively to confirm screw position [8].

2) Revascularization strategy

Thrombectomy was not attempted for the infrapopliteal occlusion. By that point, 6 days had elapsed since symptom onset, and the embolic occlusion involved the distal popliteal and tibial arteries, with distal vessels filling through collaterals. Given the delayed presentation, the possibility that the embolic material had become partially organized could not be excluded. Complete thrombus clearance by thrombectomy alone was therefore considered uncertain, and incomplete thrombus retrieval could have caused distal migration and compromised the runoff vessel required for anastomosis. Catheter-based aspiration devices were not available in Japan in 2015, and because a suitable great saphenous vein was present, bypass was considered the more reliable option. Aspiration thrombectomy using this system has since been reported as safe and effective for acute limb ischemia of less than 14 days’ duration [9] and became available in Japan in 2023; therefore, a hybrid approach would be reasonable today. In retrospect, thrombectomy could have been attempted first.

Open repair of the EIA was selected because the lesion could not be characterized preoperatively, rather than because endovascular treatment was contraindicated. Metal artifact prevented adequate assessment of the lesion, and angiography showed only an intraluminal filling defect. The aneurysm, including a portion that appeared close to rupture, became evident only after exposure. Covered stents have been used successfully to treat pseudoaneurysms after THA [10,11] and might have been considered if the diagnosis had been established preoperatively. However, the GORE VIABAHN Endoprosthesis (W. L. Gore & Associates) had not yet been approved in Japan, and the distal landing zone would have lain beneath the inguinal ligament. Open exposure also permitted histopathological confirmation. Because cultures were not obtained, occult low-grade infection, which is a recognized cause of pseudoaneurysm in this setting [12], cannot be formally excluded. If recurrence occurs, the optimal approach remains uncertain. A covered stent would avoid reoperation in a scarred field but could remain exposed to the same mechanical stress if the screw is retained. Therefore, treatment should be individualized, with consideration of the screw as well as the arterial lesion.

Most vascular complications of THA occur perioperatively. However, chronic mechanical irritation from a protruding acetabular screw may cause delayed pseudoaneurysm formation and subsequent distal embolization, even decades after the index procedure. In patients with a long-standing hip prosthesis who present with acute limb ischemia of unclear origin, the iliac arteries adjacent to acetabular screws should be evaluated as a potential embolic source. Orthopedic consultation should also be sought when a protruding screw is identified.

Funding Statement

FUNDING The authors thank the Asahikawa Medical University Surgical Educational Support (AMUSE) Organization for its support with English-language editing costs.