Introduction
Interventional radiology in the thorax is widespread.
Although most people associate the practice with nonvas- cular interventions, such as lung biopsy, in fact, a huge va- riety of different nonvascular and vascular interventions are performed within thoracic structures (Tables 1, 2), and more, such as radiofrequency ablation of lung tumors, will probably come into the field. In the following, some of the most essential interventions are discussed.
Nonvascular Interventions
In the nonvascular field, CT-guided lung biopsies are the best-known and most frequently performed interventions in this part of the body. Fine-needle aspiration for cytology
and miniaturized cutting needles for histology not exceed- ing 18-20 G are used for this purpose. Automated biopsy guns have several advantages, offering excellent sampling quality and the possibility to perform repeated biopsies with a single access. Fine-needle aspiration is preferred if an object for biopsy is located close to central and vascular structures, in order to avoid major bleeding complications.
Laurent et al. [1] compared the accuracy and compli- cation rate of fine-needle aspiration vs. an automated biopsy device. The studied consisted of two consecutive series of 125 (group A) and 98 (group B) biopsies carried out using 20-22 G coaxial fine-needle aspiration (group A) and an automated 19.5 gauge coaxial biopsy device (group B). Groups A and B comprised, respectively, 100 (80%) and 77 (79%) malignant lesions and 25 (20%) and 18 (21%) benign lesions. No significant difference was found between the two series concerning patients, le- sions, and procedural variables. For a diagnosis of malig- nancy, a statistically significant difference in sensitivity was found between the results obtained with the auto- mated biopsy device and those with fine-needle aspira- tion (82.7% vs. 97.4%, respectively). For a diagnosis of malignancy, the false-negative rate of the biopsy result was significantly higher (p<0.005) in group A (17%) than in group B (2.6%). For a specific diagnosis of benignity, no statistically significant difference was found between the two groups (44% vs. 26%), but the automated biopsy device yielded fewer indeterminate cases. There was no difference between the two groups concerning the inci- dence of pneumothorax, which was 20% in group A and 15% in group B, or that of hemoptysis, which was 2.4%
in group A and 4% in group B. The authors concluded that, for the diagnosis of malignancy, automated biopsy devices have a lower rate of false-negative results and a complication rate similar to that of fine-needle aspiration.
Richardson et al. [2] surveyed 5,444 lung biopsies in the UK. Complications included pneumothorax (20.5%
of biopsies), pneumothorax requiring chest drain (3.1%), hemoptysis (5.3%), and death (0.15%). The timing of post-procedure chest radiography was variable. In centers that predominantly performed cutting-needle biopsies, the pneumothorax rates were similar to those of centers performing mainly fine-needle biopsies (18.9% vs.
IDKD 2007
Interventional Techniques in the Adult Thorax
D. Vorwerk
Department of Radiology, Klinikum Ingolstadt, Ingolstadt, Germany
Table 1.Nonvascular interventions
– Pulmonary, pleural, and mediastinal biopsy – Marker placement prior to surgical removal
– Abscess drainage from pleural, pulmonary and mediastinal origin – Breast biopsy
– Radiofrequency ablation of lung tumors and metastases
Table 2.Vascular interventions
Arterial interventions Balloon angioplasty of supra- aortic vessels
Endoluminal stent placement in- to the thoracic aorta
Transarterial embolization of bronchial arteries
Venous interventions Stents for the treatment of supe- rior vena cava syndrome Percutaneous venous catheter
placement
Catheter maintenance Foreign-body removal Pulmonary artery interventions Embolization of peritoneove-
nous shunts
Mechanical thrombectomy of pulmonary emboli
184_187_Vorwerk 28-02-2007 07:40 Pagina 184
Interventional Techniques in the Adult Thorax 185
18.3%). There is great variation in practice throughout the UK. Most procedures are carried out on a day-case basis. Small pneumothoraces are common but infre- quently require treatment.
Post-biopsy pneumothorax as a rather frequent com- plication can be treated relatively simply in most cases.
In asymptomatic patients we recommend not evacuating the pneumothorax earlier than 4 h after biopsy in order to achieve a durable success. In symptomatic patients or those with drainage failures following a single-needle ap- proach, percutaneously introduced Heimlich valves are recommended.
Minimally invasive thoracoscopic procedures have be- come increasingly popular and offer a valid alternative if the patient has only a single pulmonary nodule that can be removed for diagnostic as well as therapeutic purpos- es. In such cases, interventional radiology can be of prac- tical benefit in the procedure by CT-guided hook mark- ing of the nodule, which allows it to be easily identified during thoracoscopy and thus facilitates its removal.
Poretti et al. [3] described their experience with percu- taneous CT-guided placement of hook-wires to localize such nodules before video-assisted thoracoscopy (VATS).
In their study, 19 patients with newly diagnosed intrapul- monary nodules underwent CT-guided hook-wire (X- Reidy set) localization. Subsequently, the patients under- went VATS resection of their lesions, which required a mean time of 30 min (range 10-48 min). In all patients, resection of the nodules was successful. Eight patients developed an asymptomatic pneumothorax. In four pa- tients, in whom the tumor was hit directly by the needle, local bleeding occurred. One patient experienced hemop- tysis. However, dislocation of the hook-wire system did not occur in any of the patients.
Other nonvascular interventions include abscess drainage from the lung, the pleura, and the mediastinum (Fig. 1).
Breast Biopsy
Breast biopsy is of increasing importance in daily practice, as in European countries more and more lesions detected during screening programs are either indicative of malig- nancy or at least suspicious and not clearly classifiable. A growing number of patients also request a definitive diag- nosis even for an otherwise benign-appearing lesion.
Breast biopsy may be performed by ultrasound or MR guidance, or under stereotactic mammographic guidance.
For lesions that are detectable by ultrasound, ultrasound guidance is a quick and relatively easy approach that al- lows online monitoring of the biopsy procedure.
Microcalcifications are best detected and biopsied under mammographic guidance. In addition to core biopsy using 14 G needles, vacuum aspiration biopsy using a 10 G nee- dle is recommendable, as it allows removal of larger por- tions of tissue which makes the procedure safer and the ob- tained sample more representative. Soft-tissue structures that are detectable by mammography but have no clear cor- relation in ultrasound should be examined by mammo- graphic core biopsy. MR-positive lesions that have no clear correlations in other modalities should be evaluated by MR-guided core biopsy or placement of a marker under MR guidance followed by surgical resection.
Vascular Interventions
Vascular interventions can be divided between arterial and venous interventions (Table 2).
In the former, balloon angioplasty of supra-aortic ar- teries such as the subclavian artery, implantation of tho- racic endografts, and embolization of bronchial arteries should be mentioned. Relatively rarely performed are transarterial techniques for tumor treatment, such as chemoperfusion of the lateral thoracic, mammary, and bronchial arteries in order to treat bronchial or breast cancer.
Vascular interventions involving the pulmonary artery include occlusion of arteriopulmonary fistulas, particu- larly in patients with Rendu-Osler-Weber syndrome.
Local thrombolysis or thrombodestruction of pulmonary emboli is an intervention used relatively rarely but it of- fers a promising alternative in emergency cases involving pulmonary embolism.
In the venous area, central venous stents are used to treat malignancies and, in dialysis patients, to recanalize central venous stenoses in order to allow successful drainage. In addition, stents are employed in the place- ment and maintenance of central venous catheters, fibrin- sheath stripping, and the removal of foreign bodies.
Not all these interventions can be discussed here in depth, but embolization of the bronchial arteries and treatment of malignant venous stenoses should be em- phasized, since neither is well-known but either one could be helpful in treating patients with acute symptoms.
Fig. 1. Planning CT for drainage of a mediastinal abscess. Patient is in a supine po- sition; the access tract is planned through the para- vertebral space
184_187_Vorwerk 28-02-2007 07:40 Pagina 185
Bronchial Artery Embolization
This procedure is not a simple intervention for many rea- sons. Firstly, the anatomy of the bronchial artery varies and the size of the artery is frequently very small, mak- ing cannulation difficult. Feeders to bronchial artery bleeding sources – which are mainly arteriopulmonary fistulae due to tumoral or inflammatory changes – can originate not only from the bronchial but also from many other arteries in the thorax such as the subclavian artery, the thyreocervical trunk or the mammary or costal arter- ies. Moreover, there is a variation in collaterals in the re- gion that does include the spinal arteries.
As embolization material, particles, glue, and coils can be used, but a golden rule is to seal the artery as close as possible to the bleeding source; otherwise, recurrence via collateral feeders may ensue.
Benign sources of bronchial bleeding include:
bronchiectasis, chronic bronchitis, aspergillosis, tubercu- losis, pneumonia, and abscesses. Malignant sources are predominantly bronchial cancers.
The results of percutaneous intervention are satisfacto- ry. Kato et al. [4] treated 101 patients and reported 100%
technical success, a 1-year success of 77.7% and a 5-year success of 2.5%. They described a better outcome in pa- tients with tuberculosis and bronchitis than in those with pneumonia and abscesses.
Lee et al. [5] described their experience with bronchial artery embolization (BAE) in patients with massive, life- threatening hemoptysis. In a 5-year period, 54 patients were treated for massive hemoptysis. The underlying pathology included bronchiectasis (n=31), active tuber- culosis (n=9), pneumoconiosis (n=3), lung cancer (n=2), and pulmonary angiodysplasia (n=1). Surgery was con- sidered if the patient had acceptable pulmonary reserve and a bleeding source was clearly identified. If the patient was not considered fit for surgery, BAE was attempted.
Hemoptysis ceased with conservative management in seven patients (13%) only. Of the 27 (50%) patients who underwent surgical resection the procedures included lobectomy (n=21), bilobectomy (n=4), and pneumonecto- my (n=2). The in-hospital mortality after surgery was 15%. Postoperative morbidity occurred in eight patients, including the need for prolonged ventilatory support, bronchopleural fistulae, empyema, and myocardial in- farction.
Twenty-one patients who could not be treated surgi- cally underwent BAE, which was successful in 17 pa- tients without any complications.
In the treatment of bleeding from bronchial carcino- ma, Witt et al. [6] performed BAE using platinum coils with Dacron fibers in 30 consecutive patients. The aim was to compare immediate results of bleeding cessation, recurrence, and survival rates in patients treated with BAE vs. those managed conservatively. Active bleeding stopped immediately in all patients. In the two groups, the cessation of first-time hemoptysis (BAE 100% vs.
non-BAE 93%) and the rates of bleeding recurrence
186 D.Vorwerk
(BAE 50% vs. non-BAE 47%) were similar. Regarding recurrent bleeding, repeated BAE led to a definite ces- sation of pulmonary hemorrhage in every case. In con- trast, all patients with recurrent hemoptysis without a re- peated BAE (8 patients, 27%) and all patients with bleeding recurrence in the non-BAE group died from pulmonary hemorrhage (8 patients, 53%). The mean sur- vival time of the BAE group was significantly longer than that of the non-BAE group. The authors therefore concluded that consistent BAE proved beneficial in tu- morous pulmonary bleeding.
Malignant Venous Obstruction (SVC Syndrome): Implantation
Another percutaneous procedure that has gained in pop- ularity is the placement of metallic stents to treat superi- or vena cava syndrome (Fig. 2). Unlike emergency radi- ation, stenting offers rapid relief of symptoms within a few hours or even immediately. The technique is relative- ly simple and can be achieved from either a brachial or a transfemoral approach. If the obstruction is associated with thrombus, some clinicians prefer to combine stent- ing with thrombolysis.
Lanciego et al. [7] used stent placement as the treat- ment of choice for the relief of symptoms. Wall-stent prostheses (n=73) were inserted in 52 cancer patients whose were diagnosed with superior vena cava syn- drome, as confirmed by cavography or phlebography. A single stent was sufficient in 37 patients, two stents were
Fig. 2a, b. Malignant obstruction causing a superior vena cava syn- drome (SVCS). a Subtotal stenosis of the vena cava due to tumor and partial thrombosis. b After placement of a 16-mm-wide self-ex- panding stent and subsequent balloon dilatation, patency is restored
a b
184_187_Vorwerk 28-02-2007 07:40 Pagina 186
Interventional Techniques in the Adult Thorax 187
required in 11, three stents in two, and four stents in an- other two patients. Contraindications for the procedure were severe cardiopathy or coagulopathy. Resolution of symptoms was achieved in all patients within 72 h. At follow-up, six obstructions, one partial migration to the right atrium, two incorrect placements, and four stent
‘shortenings’ were noted. All were successfully resolved by repeated stenting. Symptom-free survival ranged from 2 days to 17 months (mean: 6.4 months). The authors concluded that the wall-stent vascular endoprosthesis is effective in the initial treatment of superior vena cava syndrome of neoplastic origin.
Conclusions
Of the numerous vascular and nonvascular procedures in the thorax, some are only rarely performed while the use of others is confined to a few centers. However, at least a general knowledge of all of them and of the benefits they have to offer is essential for optimal patient management.
References
1. Laurent F, Latrabe V, Vergier B, Michel P (2000) Percutaneous CT-guided biopsy of the lung: comparison between aspiration and automated cutting needles using a coaxial technique.
Cardiovasc Intervent Radiol 23:266-272
2. Richardson CM, Pointon KS, Manhire AR, Macfarlane JT (2002) Percutaneous lung biopsies: a survey of UK practice based on 5444 biopsies. Br J Radiol 75:731-735
3. Poretti FP, Brunner E, Vorwerk D (2002) [Simple localization of peripheral pulmonary nodules - CT-guided percutaneous hook-wire localization]. Rofo 174:202-207
4. Kato A, Kudo S, Matsumoto K et al (2000) Bronchial artery embolization for hemoptysis due to benign diseases: immedi- ate and long-term results. Cardiovasc Intervent Radiol 23:351-357
5. Lee TW, Wan S, Choy DK et al (2000) Management of mas- sive hemoptysis: a single institution experience. Ann Thorac Cardiovasc Surg 6:232-235
6. Witt Ch, Schmidt B, Geisler A et al (2000) Value of bronchial artery embolisation with platinum coils in tumorous pul- monary bleeding. Eur J Cancer 36:1949-1954
7. Lanciego C, Chacon JL, Julian A et al (2001) Stenting as first option for endovascular treatment of malignant superior vena cava syndrome. Am J Roentgenol 177:585-593
184_187_Vorwerk 28-02-2007 07:40 Pagina 187
