Pyogenic infection of the hip joint as a complication of a femoral artery vascular access for percutaneous coronary intervention

Introduction

The transcutaneous endovascular approach via the femoral artery is a common procedure and provides measurement of arterial blood gases as well as the ability to conduct many diagnostic and/or therapeutic procedures with percutaneous coronary intervention (PCI) being the most frequent. The increasing number of such procedures has resulted in an increase in the frequency of complications that range from common to rare.

One such rare complication, which is particularly serious, is a pyogenic infection of the ipsilateral hip joint. We present a case with this complication. The case illustrates that diagnosis of rare complications is difficult and the resulting delay in diagnosis will significantly limit treatment options. In this case, the delay resulted in a complete destruction of the hip joint before the correct diagnosis and treatment could be initiated.

Case report

The patient in this case gave his written permission to release his data for publication. Furthermore, the ethics committee at the Medical University of Warsaw was consulted and waived any approval process in light of the signed patient consent. No special authorization to publish this case was necessary within the legislature.

A 68-year-old patient with myocardial infarction (MI) underwent PCI with stent implantation via the right common femoral artery (December 2002). At 5 days after the procedure, during inspection of the access site, no arterial damage was noted. A typical compression time of 10 min was routinely used. However, the patient was found to have developed a femoral artery aneurysm at the access site measuring 30 mm in diameter. The aneurysm was closed by the direct local application of thrombin.

At 2 months after this revision, the patient suffered another MI. A subsequent PCI with additional stenting via the left femoral artery was performed. Another arterial aneurysm was found in the left femoral arterial access site 6 days later and required surgical intervention.

In April 2005, the patient had a third MI and repeated PCI with stenting again using the right femoral artery site. A subsequent hematoma in the inguinal fossa was found and treated appropriately. Soon after this intervention, the patient began to complain of hip pain on the right side. Initially, it was thought that this pain was caused by the hematoma and that the pain was explained by age-adjusted degeneration of the hip. The patient’s pain continued to intensify even at rest and began to limit the hip’s range of motion. This pain was also accompanied by intermittent febrile episodes. Prior to the endovascular interventions, the patient had never experienced such symptoms and was able to walk without difficulty or limitations.

Radiographic imaging studies revealed changes that were atypical for such degenerative defects and these were not considered to be degenerative changes of the musculoskeletal system.

In May 2005, radiographic imaging demonstrated degenerative changes in the acetabulum and femoral head (Figure 1a). Additional imaging in June 2005 (Figure 1b) and July 2005 (Figure 1c) revealed rapidly progressing degenerative destruction of the hip joint and acromium. In addition, the head of the femur was initially flattened, and later changes consistent with fragmentation and torsion were observed.

OPEN IN VIEWER

Figure 1.

(a) X-ray of the pelvis May 2005; (b) X-ray of the pelvis June 2005; (c) X-ray of the pelvis July 2005; (d) computed tomography (CT) scan showing destruction of the hip; (e) magnetic resonance imaging (MRI) scan showing destruction of the hip; (f) bone scintigraphy.

In June 2005, the patient was admitted to the internal medicine department of the hospital with an admitting diagnosis of anemia, hip-joint deterioration, and surrounding tissue with significant elevation of the erythrocyte sedimentation rates (ESR) of unknown cause.

Prior to admission, neuroarthropathy, as well as metabolic arthropathy, and adrenal Cushing’s syndrome-induced osteonecrosis were ruled out. Prolonged iatrogenic corticosteroid use and immunosuppression were also ruled out.

At this point, it was suspected that these changes in bone structure were caused by a bone marrow malignancy and/or a metastatic lesion. As a result, an extensive work up for an underlying malignancy was performed (Appendix 1).

Computer tomography (CT) (Figure 1d) and magnetic resonance imaging (MRI) (Figure 1e) demonstrated destructive changes in the acetabulum and femoral head with fragmentation. Moreover, there was a large amount of fluid in the synovial joint space as well as numerous pockets of fluid in the surrounding soft tissue.

Bone scintigraphy did not reveal any pathologic changes except for irregularities in the right hip joint and surrounding tissue (Figure 1f).

During this hospitalization, the patient complained of severe pain in the affected joint requiring narcotic analgesia. The patient was also found to have intermittent low-grade fevers.

At this point in the diagnostic process, the patient was transferred to the orthopedic surgery ward. The same day, a fluoroscopy-guided bone biopsy was performed with fluid aspiration of the joint space, and the fluid sent for bacterial culture. The joint space was irrigated and a drain was placed. At this point, intravenous ciprofloxacin was initiated. Ankle traction was applied with the purpose of stabilizing the hip joint and minimizing the strain on the joint.

The bacterial cultures revealed methicillin-susceptible Staphylococcus aureus. Based on the results of the antibiogram, ciprofloxacin was discontinued and intravenous clindamycin was started. After initial improvement in the clinical status of the patient, an anterolateral approach arthroplasty of the right hip joint was performed during which a necrotic femoral head was removed with irrigation of purulent fluid and surrounding inflamed and necrotic tissue. Afterwards, gentamycin-infused surgical sponges were placed in the joint space.

During the pre- and postoperative course, the patient required eight units of packed red blood cells as well as two units of fresh frozen plasma. The patient was treated with intravenous clindamycin for a total of 4 weeks with monitoring of complete blood counts for leukocytosis, ESR, and levels of C-reactive protein (CRP).

The treatment regimen resulted in an amelioration of the clinical and laboratory markers indicating infection. The patient’s hip pain subsided and the weight-bearing function of the lower extremity returned. The patient was satisfied with the medical outcome, however, he complained of gait abnormality while ambulating. At this point, prosthetic reconstruction of the hip joint was considered. Nevertheless, this was not an easy decision given the difficulty of the operation secondary to the existing destruction of the hip joint and the possible risk of further re-infection and sepsis. Moreover, in the setting of this prolonged soft-tissue infection, the resulting difficulty of balance and muscle strength could have negatively affected the stability of the hip joint, thereby further increasing the risk of joint sprain and prosthetic instability and malfunction.

In October 2007, using a posterior approach, a prosthesis was placed without fixation in the hip joint. However, three additional operations were required to properly stabilize the prosthesis. In July 2008, an anti-luxation device was placed during the last of the three procedures.

Since then, the joint prosthesis has remained stable. For the 3 years following the last procedure, the hip joint continued to function properly without any infection, loosening, or damage to the prosthesis (Figure 2c). The patient is able to walk without pain and limitation.

OPEN IN VIEWER

Figure 2.

(a) X-ray of the pelvis 1 year after intervention; (b) three-dimensional computed tomography (CT); (c) X-ray of the pelvis 3 years after implantation of a hip prosthesis.

Discussion

In contrast to Mediterranean countries such as France and Spain, where the radial approach has been preferred for quite some time, the femoral approach is still more common in central Europe. The advantage of choosing the femoral approach is that the superficial anatomic location of the common femoral artery allows an easy vascular access for endovascular treatment. However, there are disadvantages to using this approach. One such disadvantage is regaining hemostasis after the femoral sheath is removed. It is also important to point out that in Poland, primarily due to financial or reimbursement reasons, closure devices are not routinely used.^1,2 Hemostasis is achieved via placement of direct pressure dressings. This pressure dressing frequently is inadequate however in the obese patient population with a femoral artery in more profound tissue layers, patients with vessel wall calcification, or if the access site is superior to the inguinal ligament,.

As the most common late complication, pseudoaneurysms occur in 2–8% of cases.^3,4 Usually, prolonged pressing therapy or, if not efficient, percutaneous thrombin injection is the basic strategy to combat this complication, but sometimes surgery may also be required. ⁵ Unfortunately, we could not establish a theory as to why we observed pseudoaneurysms in our case. This of course would certainly help to elucidate the overall traumatic infection process.

During the endovascular approach with femoral vascular access, one should remember that this access lies in close anatomic proximity to the anterior surface of the hip joint and the only separation between the two spaces is the iliopsoas muscle. Within this space, there is extensive bacterial colonization of the skin, which increases the risk of site infection. Furthermore, the risk of joint infection significantly increases when access to the femoral artery is suboptimal and subsequently requires numerous attempts. Most frequently, this occurs in obese patients, patients with low arterial pressure in whom the femoral pulse is difficult to palpate, and those patients with significant arterial friability, which does not allow for the femoral sheath to remain fixed. PCI is frequently carried out during medical emergencies, which require rapid intervention. In such situations, the urgency of the procedure has a higher priority than a strict aseptic technique and the time to administer antibiotic for prophylaxis.

In the aforementioned scenarios, it is frequently difficult to avoid penetrating the hip joint with the cannula for vascular access. The risk of infection is further augmented when occasionally profuse bleeding follows the insertion. In the event of a hematoma and its infection, it can easily spread into the joint space. Such hematomas are examples of a complication that cannot be eliminated, particularly in patients receiving antiplatelet therapy and/or anticoagulants.

Theoretically, any one or a combination of the following can cause infection of the hip joint:

In this case, the infection was most likely caused by contiguous spread from an infected hematoma or because of direct penetration with a nonsterile surgical instrument.

In the case of pathologic symptoms from the hip, rapid diagnosis of possible complications is critical.

Aside from measuring clinical and laboratory markers (leukocytosis, ESR, CRP), ultrasonographic evaluation of the joint space should also be obtained. If there is site swelling, warmth, and edema, it may be indicative of infection of the joint space. This requires joint fluid aspiration and fluid analysis and culture. Furthermore, in every case of suspected infection with pending bacterial cultures, broad-spectrum antibiotics should be initiated.

In the initial phases of infection, X-ray changes may not be observed. CT and bone scintigraphy are also of limited diagnostic value. CT is the preferred imaging only with an advanced infection. An MRI should be obtained, which can be more sensitive to early developing infectious changes within the joint space and soft tissue.

Generally, it is rare to see rapidly progressing bone destruction involving the acetabulum and/or femoral head during the course of a joint-space infection. ⁶ In our case, the bacterial infection was superimposed on a prior unknown degenerative hip defect that promoted the rapid destruction of the hip joint.

This case illustrates the diagnosis of a hip-joint infection because of an endovascular procedure using femoral artery access. The diagnosis of such complications may not be easy, particularly if one does not consider such a complication. Moreover, a delay in the diagnosis and appropriate treatment may result in irreversible complications. It is important for all physicians who conduct endovascular procedures to recognize signs and symptoms of common as well as rare complications so that rapid interventions and/or revisions can be carried out to avoid more severe complications.

On a different note, it must be permissible to propose the radial approach in patients with acute MI or in the emergency setting, which would have avoided the pyogenic hip-joint infection and the femur reconstruction in the first place. According to the latest guidelines (class IIa recommendation), radial access must of course be performed by an experienced operator. ⁷

Our final advice is that a patient complaining of serious hip-joint pain should undergo additional diagnostics to rule out local infections in case they had undergone prior emergency PCI via the femoral approach.