SEARCH

SEARCH BY CITATION

Keywords:

  • melanoma;
  • isolated limb infusion;
  • isolated limb perfusion;
  • regional therapy;
  • lymphoma;
  • deep venous thrombosis

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

BACKGROUND:

Isolated limb infusion (ILI) of cytotoxic agents is a regional therapy for cutaneous malignancies in a single extremity. Conventional ILI technique requires retrograde catheterization of the contralateral femoral vein. A novel modified ILI technique uses an ipsilateral popliteal venous approach. The purpose of this study was to compare the performance of ILI using the 2 different approaches.

METHODS:

Data from patients who underwent lower-extremity ILI at the authors' institution between October 2005 and June 2008 were retrospectively reviewed. The authors compared the 2 ILI approaches with regard to overall procedure time, fluoroscopy time, and the number of callbacks to the operating room (OR) for flow-related issues. The Student t test and Fisher exact test were used. Adverse events, including deep venous thrombosis (DVT) in the treated limb, were recorded.

RESULTS:

Between October 2005 and June 2008, 67 lower-extremity ILI procedures (15 using a contralateral venous access approach and 52 using an ipsilateral venous access approach) were performed in 62 patients (28 men and 34 women aged 31-82 years). The mean fluoroscopy times for the contralateral and ipsilateral groups were 17.9 and 8.3 minutes, respectively (P = .0019). No significant difference in the overall procedure time and number of callbacks to the OR for flow-related issues between the 2 groups was identified.

CONCLUSIONS:

The ipsilateral popliteal venous approach is a simplified and safe ILI technique with significantly lower overall fluoroscopy procedure times required for catheter placement and no difference in catheter-related adverse events, when compared with the conventional contralateral approach. Cancer 2010. © 2010 American Cancer Society.

Isolated limb infusion (ILI) is a relatively new technique used to treat disease primarily located in a single extremity. ILI is predominantly used to treat advanced melanoma with in-transit metastases but can also be used to treat Merkle cell tumors, diffuse B-cell lymphoma, cutaneous T-cell lymphoma, and soft-tissue sarcomas,1-3 and, occasionally, even nononcologic conditions such as refractory verrucae vulgaris.4 ILI evolved from the procedure known as isolated limb perfusion (ILP), a procedure that involved isolation of the limb's blood supply and perfusion with oxygenated blood and therapeutic agents.5 ILI is essentially the same as ILP but does not require extracorporeal oxygenation, thus reducing the amount of operating room (OR) time needed and resources used.6, 7 Traditionally, arterial and venous access catheters are introduced from the contralateral limb under image guidance with catheter tips carefully positioned distal to the level of tourniquet pressure. Chemotherapy is infused for approximately 30-45 minutes, followed by a washout of the limb. Pharmacokinetic studies have shown that there is adequate distribution of the chemotherapy in vivo,8 and pharmacodynamic studies have shown that relative to ILP, the hypoxia and acidosis induced by ILI increases the activity of melphalan by a factor of 1.5 to 3.0.9, 10 Although some have questioned whether ILI is more effective than ILP,11 ILI has similar overall response rates as ILP does, with reduced morbidity, especially among older patients.12-15

No modifications in the catheter insertion technique have been reported since ILI was initially described. By using the conventional approach, difficulty advancing the tip of the venous catheter can be encountered as the intervention requires retrograde traversal of the venous valves. Occasionally, when using the retrograde approach, it can be difficult to identify the femoral vein with a most suitable diameter. These issues can be mitigated by the use of the popliteal vein of the extremity to be treated, as the catheter is inserted under image guidance and easily advanced in an antegrade fashion, along the natural direction of flow, into the dominant femoral vein. However, it remains unknown whether ILI using a popliteal vein approach is technically superior to the conventional ILI approach. The aim of the current study was to review our experience in using the modified ILI technique to reduce the time and difficulty in placing the venous access catheter and to report its safety with regard to venous thrombosis compared with conventional ILI.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

We reviewed all ILIs performed at The University of Texas M. D. Anderson Cancer Center between October 2005 and June 2008. The institutional electronic medical records database included clinical information, as well as technical parameters related to the procedure performed, such as arterial and venous access sites, catheter size, fluoroscopy time, overall procedure time, and procedural complications. Supplemental information was obtained from handwritten patient-case logs and the Social Security Death Index.16 The institutional review board waived requirements for informed consent for this retrospective study.

Patients were excluded when the ILI was performed for an upper extremity or when follow-up data were not found in the database. Patients' diseases were staged according to M. D. Anderson classification criteria.17

Before ILI, the patient's surgeon placed a radiopaque marker along the lower margin of the limb tourniquet, to guide subsequent catheter positioning on the extremity to be treated. Fellowship-trained interventional radiologists performed all catheter placements. Conventional ILI was performed as described previously with insertion of both the arterial and venous catheters from the contralateral femoral approach.6 For the simplified ipsilateral approach, the patient was initially placed in a prone position. Preprocedural ultrasonography was used to identify the popliteal vein in the extremity to be treated. After sterile preparation and local anesthesia, a 21-gauge needle and a 0.018-inch guidewire were passed into the vein under direct sonographic guidance. The needle was exchanged for a 5 French micropuncture set (Micropuncture Introducer Set; Cook Medical, Bloomington, Ind) transition dilator to accommodate a 0.035-inch guidewire. The guidewire was advanced into the femoral vein and exchanged for a 5 French multisidehole polyurethane catheter (One Step Centesis Catheter, Merit Medical, South Jordan, Utah) (Fig. 1). The venous catheter was then securely taped to the skin and connected to an infusion pump. Patients were then placed in a supine position, and the contralateral groin was prepared and anesthetized. The common femoral artery was accessed, and a 5 French catheter was inserted into the artery and advanced across the aortic bifurcation using traditional Seldinger technique. Subsequently, a guidewire was advanced into the contralateral superficial femoral artery. After catheter exchange over this wire, a 5 French multisidehole nylon straight flush catheter (Royal Flush, Cook Medical) was inserted and advanced until the tip and all sideholes were situated below the mark for the planned tourniquet. Once the catheter was considered to be in an acceptable final position, it was secured to the skin and connected to an infusion pump. Anticoagulation therapy was not used during the procedures. Patients were then transported directly to the surgical suite for the remainder of the limb infusion procedure.

thumbnail image

Figure 1. Ipsilateral approach. Anteroposterior venogram of the lower extremity shows the tip of the catheter within the popliteal vein (arrows).

Download figure to PowerPoint

After ILI, patients were monitored in the hospital for 3 to 21 days (mean, 8 days). Of the 67 cases, 37 had follow-up ultrasonography of the treated extremity to evaluate for potential DVT. Ultrasonography was performed 2 to 19 days postoperatively (mean, 4.4 days). In addition, the number of callbacks to the OR was recorded to assess the adequacy of catheter flow during therapy. Overall procedure time and catheter-related adverse events were also recorded. Particular attention to the potential for DVT of the treated limb was made as a means to determine the safety of the ipsilateral approach.

The 2 groups were statistically compared using a 2-tailed, 2-sample pooled Student t test to compare age and fluoroscopy times. A 2-sided Fisher exact test was used to compare sex, callback rates, DVT rates, and arterial and venous problems. P values less than .05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

ILI was performed 78 times at M. D. Anderson between October 2005 and June 2008 with 67 procedures (62 patients) meeting the inclusion criteria. Ten patients who underwent ILI for an upper extremity and 1 patient with inadequate follow-up were excluded from the study. The 67 remaining cases were performed in 62 patients (24 men and 38 women). Two patients in the contralateral group had subsequent procedures performed with the ipsilateral approach. Three patients in the ipsilateral group also had a second ILI procedure performed with an ipsilateral approach. For the purposes of this study, each ILI was treated as an independent event (Fig. 2).

thumbnail image

Figure 2. Schematic representations of the isolated limb infusion techniques. (A) Contralateral approach. (B) Ipsilateral approach.

Download figure to PowerPoint

Most patients had disease primarily localized to 1 extremity, although a few had distant metastases (n = 7). The majority of patients had a diagnosis of advanced melanoma stage IIIa or greater (n = 64) (grading determined according to the revised TNM staging system). Other malignancies included diffuse B-cell lymphoma (n = 1), metastatic adenocarcinoma (n = 1), and Merkle cell carcinoma (n = 1).

Overall, the patients in this study ranged in age from 31 to 84 years (mean, 62 years) (Table 1). The 6 men and 9 women who underwent the contralateral approach ranged in age from 39 to 79 years (mean, 58.1 years). Of the 52 cases performed with the ipsilateral approach, the patients ranged in age from 31 to 84 years (mean, 63.2 years). There was no significant difference in the 2 groups on the basis of age (P = .13) or sex (P = .77).

Table 1. Patient Characteristics
 Contralateral Group n=15Ipsilateral Group n=52
CharacteristicNo. of Patients (%)No. of Patients (%)
Age, y  
 Mean58.163.2
 Range39–7931–82
Sex  
 Men6 (40)24 (46)
 Women9 (60)28 (54)
Disease  
 Disease in left  lower extremity10 (67)32 (62)
 Disease in right  lower extremity5 (33)20 (38)
 Melanoma15 (100)49 (94)
 Other diagnosis03 (6)
 Distant metastases3 (20)4 (8)

ILI catheter insertion was successful in all 67 procedures. Fluoroscopy times were available for 11 of the 15 contralateral-approach procedures and for 50 of the 52 ipsilateral-approach procedures. We found a significant difference in the mean fluoroscopy times between the contralateral-approach procedures (17.9 minutes; range, 8.0 to 47.4 minutes) and the ipsilateral-approach procedures (8.3 minutes; range, 2.7 to 49.3 minutes) (P = .0019). Overall procedure times were available for 14 of the 15 contralateral-approach procedures and for 44 of the 52 ipsilateral-approach procedures. The difference in the mean total procedure times for the contralateral approach (80 minutes; range, 33 to 163 minutes) and the ipsilateral approach (86 minutes; range, 49 to 187 minutes) was not statistically significant.

Callbacks to the OR for the contralateral approach occurred in 2 of the 15 (13.3%) standard ILI procedures because of venous flow technical issues. In both cases, the poor flow was due to close proximity between the catheter tips and the pneumatic tourniquet. Callbacks to the OR for the ipsilateral approach occurred in 4 of the 52 (7.7%) ILI procedures because of venous (n = 1; 1.9%) and arterial (n = 3; 5.8%) technical issues. The case of poor venous flow was due to close proximity between the catheter tip and the pneumatic tourniquet. The cases of poor arterial flow were due to kinking of the catheter at the aortic bifurcation (n = 1), malposition of the catheter tip (n = 1), and presence of small fibrin clot at the catheter tip (n = 1). There was no significant overall difference in the callback rate between the 2 approaches (P = .61) or in the rate of arterial (P = .46) or venous problems (P = .12).

Seven of the 15 contralateral cases and 30 of the 52 ipsilateral cases had follow-up ultrasonography to rule out deep venous thrombosis (DVT) within 30 days of the procedure (Table 2). None of the patients in the ipsilateral group had DVTs, whereas 1 patient in the contralateral group had image findings suspicious for DVT on the basis of the presence of nonphasic flow in the popliteal vein. There was no significant difference in the DVT rate between the 2 venous approaches (P = .23). The 1 suspected DVT was due to a chronic, nonocclusive, deep-vein thrombus within the common femoral vein and nonphasic flow in the popliteal vein. The patient had no known prior history of DVT and no clinical evidence of thrombus propagation or pulmonary embolism. The patient was placed on therapeutic low–molecular-weight heparin (Lovenox) and was able to walk without difficulty before discharge. Subsequent follow-up visits did not reveal further sequelae of the DVT or any new thrombotic complications.

Table 2. Comparison Between Contralateral and Ipsilateral Approaches
 Contralateral Group n=15Ipsilateral Group n=52 
ParameterNo. of Patients (%)No. of Patients (%)P
  1. OR indicates operating room; DVT, deep venous thrombosis.

Fluoroscopy time   
 Mean, min17.98.3.0019
 Range, min8.0–47.42.7–49.3 
 No. of evaluable  patients1150 
OR callbacks   
 Overall2 (13)4 (8).6094
 Venous2 (13)1 (2).1235
 Arterial03 (6).4613
 Postoperative DVT1 (7)0.2308

Patients were also followed up between 5 and 60 days postoperatively. None had clinical evidence of DVT after discharge. Eight patients were readmitted for complications unrelated to DVT or other catheter-related issues within 60 days of the operation.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

The significantly lower fluoroscopy time needed for the combined arterial and venous catheter placement during procedures using the ipsilateral technique, resulted in a reduction in total radiation dose to the operator and the patient. Overall procedure times were not significantly different. This is likely due to the finding that any amount of time saved during fluoroscopy was offset by the time needed for ultrasound of the popliteal fossa as well as repositioning the patient from prone to supine to gain contralateral femoral access for the arterial catheter. In addition, only 1 of 52 (1.9%) patients in the ipsilateral group was called back to the OR for venous issues, whereas 2 of 15 (13.3%) patients in the contralateral group were called back. Although the callback rate was not significantly lower, it was not significantly different from the conventional ILI technique. Flow rates for catheters placed by the ipsilateral approach appeared to be subjectively higher than the rates achieved with the contralateral approach. Some of this can be attributed to the shorter (length) catheter used in the ipsilateral approach with its thinner wall and slightly larger lumen, compared with the 6 French catheter (Gensini, Cook Medical) used in the conventional contralateral approach. In addition, a catheter placed in the ipsilateral popliteal vein is positioned along the normal circulation and not against direction of the valves, thus optimizing blood flow.

One risk of using the ipsilateral venous access approach is the potential for postoperative DVT related to the indwelling catheter or the trauma of the popliteal vein at the access site. Venous thromboembolism cannot only lead to pulmonary emboli but can also cause significant chronic venous disease, increasing the morbidity of patients already in poor condition.18, 19 Although our sample sizes were relatively small, we did not find a significant difference between the contralateral and ipsilateral venous approach with regard to DVT. There were no cases of DVT in the treatment limb in the ipsilateral group, and, thus, this approach from a catheter-related complication standpoint is as safe as the contralateral approach. Moreover, placing a venous catheter in a retrograde fashion in the popliteal vein should have a lesser risk of damage to the native valves, thus decreasing the likelihood of a DVT.

This was a retrospective study in which we used unbalanced sample sizes to analyze the experience of one major cancer center. Despite these limitations, our study establishes that ILI using the ipsilateral popliteal venous access substantially decreases radiation exposure to the patient and medical staff because of the significantly lower fluoroscopy time required for catheter insertion. Given the poor prognosis of advanced melanoma, the development of proper technique facilitates delivery of novel therapies.20, 21 By using an ipsilateral, popliteal, venous access approach in ILI is safe without an increase in catheter-related adverse events.

CONFLICT OF INTEREST DISCLOSURES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES

Michael J. Wallace has received research support and a speaking honorarium from Siemens Corporation.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONFLICT OF INTEREST DISCLOSURES
  7. REFERENCES