| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular and Interventional Radiology |
lu, MD,
Süleyman Baykal, MD,
Ahmet Sari, MD,
Özgür Sayil, MD and
Halit Re
it Gümele, MD
1 From the Departments of Radiology (H.D., A.A., A.S., O.S., H.R.G.) and Neurosurgery (S.B.), Karadeniz Technical University Faculty of Medicine, Farabi Hospital, 61080 Trabzon, Turkey. Received August 28, 2001; revision requested October 17; final revision received April 8, 2002; accepted April 24. Address correspondence to H.D. (e-mail: hdinc@meds.ktu.edu.tr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Computed tomography (CT)–guided PD was performed in 21 patients with 26 tuberculous iliopsoas abscesses. Nineteen patients had bone involvement of two or more vertebrae. Eleven patients with spondylodiskitis had intradiskal abscesses. Five patients had bilateral psoas abscesses. Easily and safely accessible well-circumscribed abscesses larger than 3 cm were selected for PD. Catheters were inserted into the abscess cavities with Seldinger technique in all cases. In conjunction with PD, all patients had antituberculous drug therapy and underwent clinical and imaging follow-up for at least 1 year.
RESULTS: Percutaneous catheter placement was successful in all cases without procedural complications. On the basis of CT findings, complete evacuation of all abscesses was achieved initially. During follow-up, six (29%) of 21 patients had recurrences within 1 and 3 months after catheter removal. A total of 37 catheters were used; eight of the 37 catheters were inserted due to recurrences. Four patients needed two PD procedures, and two patients needed three due to recurrences. Four catheters were changed because of obstruction or dislocation. Drainage duration ranged from 5 to 36 days (mean, 14.9 days). The follow-up period was 12–52 months (mean, 24 months). None of the patients, including those with recurrence, required surgical drainage and débridement due to insufficient PD.
CONCLUSION: Image-guided PD in conjunction with antituberculous drug therapy is an effective and safe procedure in the treatment of tuberculous iliopsoas abscesses with or without spondylodiskitis.
© RSNA, 2002
Index terms: Abscess, percutaneous drainage, 33.242, 33.1262 • Tuberculosis, musculoskeletal, 33.231
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Traditionally, treatment of iliopsoas and spondylodiskitic abscesses has been either conservative (drug therapy, immobilization, external brace application) or surgical (surgical intervention in conjunction with drug therapy) (3–5). Percutaneous drainage (PD) with radiologic guidance has become established as the primary drainage procedure in the treatment of pyogenic abscesses and can be performed in most body locations, including the abdomen, thorax, and retroperitoneum (6,7). To our knowledge, there are few reports about the utility and safety of PD in conjunction with antituberculous drug therapy in the management of tuberculous iliopsoas abscesses with bone involvement (8–11).
The purpose of the present study was to evaluate the midterm results of PD performed with image guidance in 21 patients with tuberculous iliopsoas abscesses with or without spondylodiskitis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To be included in the study, patients had to have been followed up for at least 1 year after PD. A total of 26 abscesses in 21 patients were treated with PD. Abscesses were primarily located in the iliopsoas compartment (n = 24) and were more rarely located in the anterior paravertebral region (n = 2). In two patients with iliopsoas abscesses, the collections extended into the gluteal region. Five patients had bilateral iliopsoas abscesses. Seventeen patients had spondylodiskitis (an inflammatory process involving the disk and the adjacent vertebrae), and 11 of these patients had intradiskal abscesses. One patient had iliopsoas abscesses with sacroiliac joint involvement. Nineteen patients had bone involvement. Patient characteristics, abscess locations, bone involvement, and drainage characteristics are summarized in the Table. Six patients showed active or chronic tuberculous changes on chest radiographs. One patient had tuberculous lymphadenitis, one had renal tuberculosis, and one had multiple cranial tuberculomas associated with iliopsoas abscesses.
|
Before PD, all patients underwent radiography and CT performed with one of three scanners (Somatom ART, Siemens, Erlangen, Germany; Xvision/GX, Toshiba, Tokyo, Japan; or TCT-600, Toshiba). CT scanning was performed with a section thickness of 5–10 mm before and after the intravenous injection of a 100–120 mL bolus of contrast medium (Ultravist 300; Schering, Berlin, Germany or Omnipaque 300; Nycomed, Princeton, NJ). Before PD, 14 patients underwent additional MR imaging at 0.5 T with a Vectra unit (GE Medical Systems, Milwaukee, Wis) (n = 8) or at 1.0 T with a Magnetom Impact Expert unit (Siemens) (n = 6). Unenhanced sagittal and transverse T2-weighted spin-echo MR images (repetition time msec/echo time msec, 2,000/90) were obtained, and sagittal and transverse T1-weighted spin-echo MR images (500–600/15–20) were obtained before and after administration of 0.1 mmol of gadopentetate dimeglumine (Magnevist; Schering) per kilogram of body weight.
After CT scanning was performed, the most suitable CT scan was selected. With CT guidance lights, skin markers, and computer calculation programs available in the CT scanner, the depth, angle, and distance from the skin of each abscess were measured. After exact localization of the abscess site for primary puncture, an 18-gauge 15- or 20-cm-long sheathed needle (William Cook, Bjaeverskov, Denmark; or Boston Scientific/Medi-Tech, Rungis, France) was inserted into the most distal portion of the fluid collection. Aspiration was performed to confirm the presence of purulent material. Through the cannula or sheath, a 0.038- or 0.035-inch stiff guide wire (Amplatz extra stiff; William Cook) or a 0.035-inch superstiff guide wire (Lunderquist; William Cook) was inserted into the collection. Proper dilation was performed according to the catheter size, the drainage catheter was inserted into the collection over the guide wire, and the guide wire was removed. All drainage catheters were inserted into abscess cavities with the Seldinger technique.
The first aspirated material was sent to the laboratory for Gram and acid-fast staining and for culturing for aerobic and anaerobic microorganisms and M tuberculosis. Additional laboratory studies, including erythrocyte sedimentation rate, white blood cell count, a C-reactive protein test, a purified protein derivative skin test, and the Gen-Probe Amplified Mycobacterium Tuberculosis Direct (MTD) test (Gen-Probe, San Diego, Calif) (12) were performed for all patients. The Gen-Probe Amplified MTD test is a target-amplified nucleic acid probe test for the in vitro rapid detection of M tuberculosis complex rRNA in specimens, including bronchial specimens, tracheal aspirates, abscess aspirates, tissue and lymph node biopsy samples, and cerebrospinal fluid samples (12).
In 18 of 21 patients the MTD test results were positive. Specific microorganisms were isolated in eight of 21 cases of tuberculous psoas abscess. In three of 21 patients, the diagnosis of tuberculosis was established with the combination of purified protein derivative skin test results and results of cytologic and histologic examination of biopsy specimens. Histologic examination of biopsy specimens obtained in these three patients revealed a necrotizing granulomatous reaction compatible with tuberculosis. On the advice of an infectious disease specialist, antituberculous drug therapy with different drug combinations was continued for at least 12 months in all patients.
For the PD procedure, each of the 21 patients was placed in either an anterior (n = 4), anterolateral (n = 6), or posterior (n = 11) position. Percutaneous catheter placement was successful in all cases with no procedural complications. On the basis of CT findings, complete evacuation of all abscesses was achieved initially. Six (29%) of 21 patients had recurrences during the follow-up period. Thirty-seven drainage procedures were performed, and 37 large-lumen (17 14-F and 20 12-F) drainage catheters (Urotech, Bruckmühl, Germany; Malecot, William Cook; or vanSonnenberg sump, Boston Scientific/Medi-Tech) were used in 21 patients. For all 26 abscesses in 21 patients, a total of 25 catheters were initially inserted into the abscess cavities. Twelve additional drainage catheters were later used (eight due to recurrence, two due to obstruction, and two due to dislodgement). Ultimately, one drainage catheter was used in each of 13 patients, two were used in four patients, three were used in one patient, four were used in two patients, and five were used in one patient (Table). In four patients, two drainage catheters were placed during one session. There were no persistent sinus tracks to adjacent structures in any patient. There were no major complications such as hemorrhage, nerve trauma, septicemia, or death.
All catheters were left in place until (a) drainage stopped or was less than 10 mL within 24 hours for 2 consecutive days, (b) clear clinical improvement was observed, and (c) blood tests for inflammatory signs showed definitive improvement and follow-up CT and/or ultrasonography revealed total or near-total resolution of the abscess cavity. Drainage duration ranged from 5 to 36 days (mean, 14.9 days). The duration of follow-up was between 12 and 52 months (mean, 24 months). The first follow-up radiologic examination was performed within 1–3 months after the catheter was removed. The follow-up radiologic examinations, including radiography and CT, were performed at 6-month intervals for 2 years and once every year thereafter. During the follow-up period, contrast material–enhanced MR imaging was also performed in 14 patients with spondylodiskitis.
Two experienced radiologists (H.D., A.A.) evaluated the follow-up studies in consensus. Radiographs were evaluated for bone destruction and sclerosis, vertebral collapse, disk space narrowing, paravertebral swelling and calcification, and spinal deformities such as kyphosis and scoliosis. Contrast-enhanced CT and MR images were used for diagnosis of abscesses and phlegmon. Follow-up CT and MR images were used for investigation of abscess recurrence and bone changes such as destruction and sclerosis. MR images were used for the evaluation of epidural areas, the spinal canal, and the spinal cord and for assessment of whether dural sac compression and spinal deformities such as kyphosis and scoliosis were present. In patients with spondylodiskitis, contrast-enhanced MR images were also assessed for T1 and T2 signal intensity alterations seen in osteomyelitis and diskitis. In three of 17 patients with spondylodiskitis, MR imaging could not be performed (two patients had claustrophobia, and one refused to undergo MR imaging). The diagnosis of spondylodiskitis in these three patients was rendered on the basis of radiographic and CT findings.
In addition to radiologic examinations, all patients underwent routine clinical and neurologic examination and blood testing for signs of inflammation. Clinical follow-up and neurologic examinations were performed by an experienced neurosurgeon (S.B.). Radiologic findings were compared with clinical status and laboratory results.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
Percutaneous catheter placement was successful in all cases, with no procedural complications (Figs 1, 2). On the basis of CT findings, complete evacuation of all abscesses was achieved initially. None of the patients, including those with recurrence, had to undergo drainage surgery because of ineffective PD. A total of 37 drainage catheters were used in 21 patients. Eight of 37 catheters were placed due to recurrences. Four catheters (two 14-F catheters and two 12-F catheters) were changed due to obstruction or dislodgement. A sinogram was obtained in 17 of 21 patients just before withdrawal of the catheter; no sinus track to any adjacent structure was found in any patient.
|
|
|
Drainage duration ranged from 5 to 36 days (mean, 14.9 days). Drainage volume ranged from 85 to 1,450 mL (mean volume, 321 mL). The duration of follow-up was between 12 and 52 months (mean, 24 months).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The iliopsoas compartment begins at the T12 vertebra and extends to the lesser trochanter of the femur. It is in close contact with the retroperitoneal organs, pelvis, and thigh. Infections of the spine can spread into the iliopsoas compartment, pelvis, and thigh (15,16).
The thoracic and thoracolumbar segments are the most commonly affected regions (2,13). In the present study, a total of 40 vertebrae were involved; approximately 58% of bone lesions were in the lumbar region. One patient (5%) had involvement of the sacroiliac joint. Sacroiliac joint involvement is relatively rare compared with vertebral involvement (8). Sacroiliac joint involvement has been reported in up to 10% of patients with skeletal tuberculosis (8,17). None of our patients had involvement of the cervical vertebrae. The disease usually is confined to one interspace, with involvement of two vertebrae, although involvement of multiple contiguous or noncontiguous vertebrae may occur (13). In the present study, 84% of patients with bone lesions had involvement of two vertebrae and one disk space.
In the diagnosis of spinal infections, radiography, CT, MR imaging, and radionuclide studies can be used (3,5,18–20). CT is excellent for diagnosis and radiologic evaluation in patients with iliopsoas abscesses and areas of bone destruction. MR imaging with gadolinium enhancement has become the standard modality for diagnosis and follow-up of diskitis and osteomyelitis of the spine, especially in patients with epidural and intradural infections and neurologic symptoms (4,18–20). In our study, MR images were obtained before and after the administration of contrast material in 14 patients with diskitis, intradiskal abscesses, and extensive bone destruction.
Because CT is an excellent modality for showing needle or catheter localization, the exact location and extent of abscesses, and the relationship of abscesses to nearby organs, CT was used in all patients for percutaneous interventions. Radionuclide scanning can be used for diagnosis of tuberculous spondylodiskitis, especially in defining the extent of disease involvement. Gallium 67 (67Ga) scintigraphy has been recommended for diagnosing extrapulmonary tuberculosis and monitoring treatment response. Unfortunately, radionuclide scanning is not sensitive for tuberculous infections, with technetium 99m bone scanning reportedly negative in 35% of cases and 67Ga scanning negative in 70% of cases in one series (3).
For definitive diagnosis of tuberculous abscesses, tissue culture and histologic examination are always necessary. But cultures for M tuberculosis are positive in only 50%–60% of patients (3,4,21). In some studies, negative cultures have been obtained in as many as 50%–85% of patients (10,14,22). In a study by Gupta et al (10), the diagnosis of tuberculosis was presumptive and was rendered on the basis of clinical presentation features, radiographic findings, and radiologic evidence of response to antituberculous therapy in 14 (52%) of 27 patients. In the remaining 13 patients, the diagnosis was rendered on the basis of the presence of acid-fast bacilli in aspirate or lymph node biopsy positive for epithelioid granulomas (10). Gupta et al did not report their positive culture rate. Because 2–8 weeks are required for the growth and identification of organisms, histologic studies showing granulomatous tissue compatible with tuberculosis and positive skin tests are sufficient to begin therapy (20,22). Adequate response to antituberculous chemotherapy is another recognized method of establishing the diagnosis (10,22,23). In our series the positive culture rate was 38%. The high rate of negative cultures in our series can be explained by the fact that six patients in our series had previously undergone antituberculous drug therapy for pulmonary tuberculosis.
Despite many comprehensive studies, treatment recommendations for spinal osteomyelitis are not uniform throughout the world (5). Prior to the advent of modern antibiotics, vertebral osteomyelitis had a poor prognosis, with a reported mortality rate of 25%–70% (3,4,18). Today, in mild and moderate cases, conservative treatment, including drug therapy, immobilization with or without bed rest, brace application, and physical therapy, is sufficient in 80%–98% of cases (3,4,14). In patients with neurologic deficits, reported rates of improvement have been 69%–94% in those treated with surgery and 79%–87% in those who underwent chemotherapy alone (5). Indications for surgery are failure of medical treatment, spinal deformity or instability secondary to either pathologic fracture or advanced bone destruction, worsening neurologic status, and spinal cord compression with deficit (4,5). Because the anterior elements of the spine are most often involved, anterior decompression, drainage, and débridement supplemented with posterior fusion with instrumentation are the traditional surgical procedures. The mortality rate for spinal tuberculosis surgery is between 2% and 11%; the relapse rate is 10%–20% (3,4).
PD has become established as the primary drainage procedure for intraabdominal abscesses, but there is limited information in the literature on PD of tuberculous psoas abscesses (8–11,14). In the largest study to date (to our knowledge), in which Gupta et al (10) evaluated 27 patients with iliopsoas abscesses of tuberculous origin, three patients were treated with percutaneous aspiration and the remaining 24 patients were treated with PD. In that study, initial aspiration or PD was successful in all 27 (100%) patients. In eight patients (29.6%), recurrence was seen. Six recurrences occurred in the 24 patients who underwent only PD; these recurrences were managed with needle aspiration (n = 4) or repeated catheter drainage (n = 2). The reason for recurrence was irregular or incomplete antituberculous treatment in most (ie, five) of the patients. The three patients who were treated with percutaneous aspiration also experienced recurrence while they were being treated with regular antituberculous therapy (10).
Pombo et al (11) reported results in six patients with seven tuberculous abscesses that were treated with PD and chemotherapy. They observed only one relapse, which was probably related to an irregular or incomplete course of antituberculous medication. In a study conducted by Staatz et al (14), seven patients with tuberculous abscesses were successfully treated with PD. It is unusual for PD to be successful in patients with tuberculous abscesses without recurrence. The successful results of Staatz et al (14) can probably be explained by the fact that all of their patients had mild to moderate bone lesions that were limited to one or two vertebrae and one disk space. They also placed the drainage catheters into the intradiskal space in patients with intradiskal abscesses (14). We did not specifically place drainage catheters within the intradiskal area in patients with intradiskal abscesses. Because intradiskal collections usually have communication with iliopsoas abscesses, we placed drainage catheters within the largest pocket of the iliopsoas abscess in our patients.
In the present study, the recurrence rate was 29%. The reasons for recurrence were as follows: (a) Resistance to antituberculous drugs (n = 1). In endemic areas the incidence of isoniazid resistance is above 4% (5,16). (b) Misdiagnosis of abscess as pyogenic in origin due to secondary infection (n = 1). (c) Early cessation of antituberculous medication (n = 1). Antituberculous drugs should be used for at least 12 months in the treatment of tuberculous osteomyelitis (24). (d) Presence of large abscesses extending to different areas with multiple bone lesions that were associated with intradiskal abscesses (n = 3). Because the medical treatment of tuberculous spondylitis requires a long period of time, in some cases antituberculous drug therapy may not be efficient enough to prevent recurrence after the drainage catheter is removed. The chronic nature of tuberculous spondylitis and the persistence of an active tuberculous focus in the spine after catheter removal may be responsible for the recurrences in the latter three of our patients. Most recurrences are not due to the PD procedure but are related to inadequate antituberculous drug therapy. Therefore, close contact with the patient and regular clinical and radiologic follow-up are important for reducing the recurrence rate.
In conclusion, 26 iliopsoas tuberculous abscesses in 21 patients were initially drained successfully in conjunction with drug therapy. Four patients needed a second and two patients a third PD procedure. During the 12–52-month follow-up period, no recurrences were observed. Image-guided percutaneous drainage in conjunction with drug therapy is safe and effective in the treatment of tuberculous iliopsoas abscesses in patients with or without spondylodiskitis.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, H.D., S.B., A.A.; study concepts, H.D., A.A., A.S.; study design, H.D., A.S., S.B.; literature research, S.B., A.S., O.S.; clinical studies, H.D., S.B., A.A.; data acquisition, H.D., S.B., A.S.; data analysis/interpretation, H.D., S.B., H.R.G.; manuscript preparation, H.D., S.B., A.A.; manuscript definition of intellectual content, H.D., S.B., A.S., O.S.; manuscript editing, H.D., S.B., A.A.; manuscript revision/review and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H.T. Sanal, M. Kocaoglu, A. Sehirlioglu, and N. Bulakbasi A Rare Cause of Flank Mass: Psoas Abscess due to Extensive Primary Thoracolumbar Tuberculous Spondylodiskitis. AJNR Am. J. Neuroradiol., September 1, 2006; 27(8): 1735 - 1737. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Melissano, E. Civilini, M. Papa, R. Del Guercio, and R. Chiesa Antalgic Flexion of the Lower Limb: An Unusual Presentation of Aortoiliac Infection with Psoas Muscle Abscess: Four Case Reports Vascular and Endovascular Surgery, May 1, 2005; 39(3): 287 - 292. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
