This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Daryanani, D. Articles by Hoekstra, H.J. Search for Related Content PubMed PubMed Citation Articles by Daryanani, D. Articles by Hoekstra, H.J. Related Collections Surgery

Continuous Leakage Measurement During Hyperthermic Isolated Limb Perfusion

From the Division of Surgical Oncology (DD, RK, PHN, JHJ) and the Department of Nuclear Medicine (JTV, DAP), University Hospital Groningen, Groningen, The Netherlands.

Correspondence: Address correspondence and reprint requests to: H. J. Hoekstra, MD, PhD, Division of Surgical Oncology, University Hospital Groningen, PO Box 30001, 9700 RB Groningen, The Netherlands; Fax: 31-50–3614873; E-mail: h.j.hoekstra{at}chir.azg.nl

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Background: Continuous measurement of perfusate leakage into the systemic circulation is of the utmost importance and can be performed with the help of radioactive tracers. The purpose of this study was to assess changes in the perfusion leakage rate between two periods: 1977–1990 and 1991–2000, and to determine the factors responsible for these changes.

Methods: During the 1991–2000 period, 119 patients underwent HILP mainly for locally recurrent melanoma or locally advanced soft tissue sarcoma. HILP was performed with melphalan (33%) or in combination with TNF (65%). There were 67 iliacal, 12 femoral, 25 popliteal, and 15 axillary perfusions performed. Leakage into the systemic circulation was monitored continuously with the help of ¹³¹I-albumin and a stationary scintillation detector placed above the heart.

Results: The median maximum leakage was 2.7% (range 0%–21%) which is significantly less than the previous period (1977–1990) where leakage of 8% (range 0%–30%) was reported (P < .05). A statistical difference in leakage was detected among perfusion locations where the iliac and femoral vessels showed more leakage than the axillary and popliteal vessels (P < .05). Furthermore, there appeared to be significantly less leakage when TNF was used than when melphalan was the sole drug (P < .05).

Conclusions: Nowadays leakage from isolated perfusions into the systemic circulation is further minimized compared with the days when melphalan was the sole drug used. Increased awareness about TNF leakage, continuous external monitoring with ¹³¹I-albumin as the main isotope, flow rate regulation in the perfusion circuit, and regulation of the patient’s systemic blood pressure have all been major contributors to this improvement.

Key Words: Melanoma • Sarcoma • Tumor necrosis factor • Leakage monitoring • HILP • ¹³¹I-albumin

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
With hyperthermic isolated limb perfusion (HILP), high doses of cytotoxic agents can be administered locally in a limb without systemic toxicity. This method was introduced in 1958 by Creech and Krementz¹ and has since undergone many improvements such as the addition of hyperthermia.² HILP is employed mostly in the treatment of recurrent melanomas and locally advanced extremity soft tissue sarcomas (STS). Since Luck³ introduced the cytostatic agent L-phenylalanine (melphalan) in 1956, it has been the preferred agent in HILP for in-transit metastases of melanoma. Because chemotherapeutic agent dosage is high (up to 15–20 times the systemic tolerated dose), a considerable risk of toxic effects is possible due to perfusate leakage into the systemic circulation.⁴ Leakage of more than 15% perfusate containing melphalan into the systemic circulation may cause toxic effects such as bone marrow depression, gastrointestinal toxicity, hair loss, and pruritis.^5–7

With the introduction of recombinant tumor necrosis factor (TNF) in the early nineties, the effectiveness of HILP was increased for melanoma and especially for soft tissue sarcoma (STS) patients.^8,9 Some mechanisms for the synergism of TNF and melphalan are proposed: there can be a combined cytotoxic effect, or TNF may disrupt the tumor-associated vasculature leading to increased tumor concentrations of melphalan.^10–12 Unfortunately, leakage of TNF into the systemic circulation produces more severe side effects than melphalan alone. Even a 1% leakage may result in hypotension of the patient while a 10% leakage of TNF can cause a potentially fatal septic shocklike syndrome.^13,14 However, strict isolation of the limb is not always achievable due to anatomical variations and/or technical reasons. Therefore, minimal systemic leakage cannot always be prevented and makes continuous leakage monitoring during the perfusion procedure especially important.

Many leakage detection methods have been described recently in the literature, ranging from intermittent blood sampling to continuous external monitoring using radioisotopes.

At University Hospital Groningen, continuous external leakage measurement is performed by using ¹³¹I-albumin as the radioisotope. Between 1977 and 1990, the median of the maximum percentage leakage was reported to be 8.0% in this institution.⁷ Since then, there have been several improvements in hemodynamic control during perfusion and in leakage monitoring. We therefore embarked on a retrospective study to analyze the maximum leakage encountered for patients who underwent HILP between 1991 and 2000.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Patients
Between January 1991 and March 2000, 119 patients underwent a limb perfusion at University Hospital Groningen. Of the 119 perfusions, 67 were iliac (56%), 12 femoral (10%), 25 popliteal (21%), and 15 axillary (13%) perfusions. The perfusion was carried out on 71 female (60%) and 48 male (40%) patients with a median age of 56 (range 17–79) years (Fig. 1). Sixty-two patients (52%) were perfused for intransit melanomas, 51 patients (43%) for a primarily unresectable sarcoma, and 6 patients (5%) had other types of malignancies (Fig. 2).

View larger version (42K): [in this window] [in a new window]   FIG. 1. Age distribution of all 119 perfused patients.

View larger version (33K): [in this window] [in a new window]   FIG. 2. Tumor types corresponding to sex.

Perfusion Technique
The perfusion technique used at University Hospital Groningen is based on the technique developed by Creech et al.¹ and has been described extensively by Schraffordt Koops et al.¹⁵ The surgical technique for HILP remained the same between the two time periods. All patients underwent operations under general anesthesia. After a skin incision, the iliacal, femoral, popliteal, or axillary vessels were exposed, and collateral vessels were clipped. In iliac perfusions, the hypogastric vessels were also temporarily closed by using tourniquets or clamps. The patients were heparinized (3.3mg/kg body weight) and catheters were inserted in the artery and vein; both were connected to an extracorporeal circuit. To prevent leakage via collateral vessels in the subcutaneous tissues, a rubber tourniquet was used for the iliacal, femoral, and axillary perfusions, and an arterial occlusion cuff was used for the popliteal perfusions.

The extremity was isolated from central circulation with the aid of an Esmarch bandage. The perfusate consisted of 250 ml red blood cell concentrate; 250 ml Isodex® (NPBI, Emmer-Compascuum, The Netherlands) in 0.9% NaCl; 30 ml NaHCO₃ 8.4%; and 25 mg heparin (2500 IU) (B. Braun Melsungen AG, Melsungen, Germany). All perfusions were performed under mild hyperthermic conditions (39–40°C). If the muscle reached a temperature of 38°C and if almost no leakage was detected into systemic circulation, the chemotherapeutic agents were administered to the perfusion circuit via the arterial line. For the TNF perfusions especially, almost 0% leakage was required before the TNF was added to the perfusion circuit. For the less toxic melphalan, a leakage of 5% was generally accepted. The dosage used in melphalan perfusions was 10 mg/l lower limb volume and 13 mg/l upper limb volume (Alkeran®, Glaxo Wellcome, London, UK). If TNF (Boehringer Ingelheim International GMbH, Ingelheim am Rhein, Germany) was used, the dosage was 4 mg for the lower extremity and 3 mg for the upper extremity (given 30 minutes before the melphalan). The flow rate for the perfusion fluid in the perfusion circuit was approximately 500 ml/min. After 1 hour for non-TNF perfusions and 90 minutes for TNF plus melphalan perfusions, the extremity was flushed with Isodex® in 0.9% NaCl (between 3–6 L depending on the level of perfusion) and 250 ml red blood cell concentrate. Both catheters were removed and the vessels were repaired. The heparin was then neutralized by using protamine sulfate, and a fasciotomy was performed to prevent a compartment syndrome. If required, a local excision of the melanoma and in-transit metastases was performed after the perfusion and, if necessary, a free skin graft from the contralateral limb was taken and kept under sterile conditions at 4°C and placed on the surgical wound 3 days postperfusion.

Post-HILP patients were admitted to the intensive care unit for 24 hours to monitor clinical toxicity from the TNF and/or melphalan.

Leakage Monitoring
Leakage measurement took place with the help of radioactive tracers. A low dose of ¹³¹I-albumin (0.5 MBq) and a dose of ^99mTc-albumin (10 MBq) were administered to the systemic circulation. A 10-fold higher dose of ¹³¹I-albumin (5 MBq) was administered to the isolated limb circulation. A fixed scintillation detector placed above the heart measured the radioactivity over the cardiac zone. The count rate of ¹³¹I-albumin at the start of the perfusion (t = 0) was the set point used for calculating the leakage percentage of the cytotoxic drugs. Leakage from the perfused limb into the systemic circulation results in an increase of this count rate. This increase, corrected for the blood volume ratio and the radioactivity ratio in both compartments, was a direct measure of leakage percentage. These changes were automatically converted into leakage percentages by a microprocessor connected to the detector. Radioactivity was continuously registered during the procedure (Fig. 3). The ^99mTc-albumin baseline radioactivity was used as a control for detector efficacy during the entire procedure. To avoid detrimental effects to the thyroid gland by the radioactive ¹³¹I, the patient was given iodine (15 drops of Lugol solution twice daily) 1 day before the operation.

View larger version (18K): [in this window] [in a new window]   FIG. 3. Continuous leakage monitoring during isolated limb perfusion with a scintillation detector placed above the heart.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
All perfusions were performed under standard perfusion conditions. There was no mortality related to the perfusions. For all 119 patients, the median value of the maximum leakage (leakage_max) was 2.7% (range 0%–21.0%). There was no statistical difference in leakage_max between male (2.4%, range 0%–21.0%) and female (3.0%, range 0%–12.0%) patients. Nor was there a statistically significant difference in leakage_max among the various age groups.

However, a statistical difference in leakage was detected among the blood vessels used for perfusion. Iliac and femoral perfusions had significantly higher leakage_max than popliteal and axillary perfusions. For iliac perfusions, the leakage_max was 4.5% (0%–21.0%), for femoral perfusions 3.8% (0%–15.5%), for popliteal perfusions 1% (0%– 9.0%), and for axillary perfusions 0% (0%–4.0%) (Fig. 4). The leakage_max for the groups, stratified for the different drugs used, also showed a statistically significant difference. When TNF was used, a lower leakage_max was detected compared with when no TNF was used: median 2.0% (0%– 15.5%) for the TNF group and 4.0% (0%–21.0%) for the non-TNF group (P < .05). However, there was no difference in morbidity between the patients who were perfused with TNF and those who were not. The fasciotomy and melanoma excision wounds created post-HILP, with primary closure or delayed split-skin grafting, usually healed without any major complications.

View larger version (15K): [in this window] [in a new window]   FIG. 4. Distribution of the maximum leakage and the amount of perfusions corresponding to the level of perfusion (dotted line represents the median value).

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

Several leakage detection methods are described in medical literature. Many groups have monitored the systemic levels of the drug used exclusively through the analysis of intermittent blood samples.^16,17 However, the flaws in this method are that the monitoring is not continuous and the results take at least 20 minutes before they are known. With this method, the surgeon has no possible chance to intervene rapidly if a potentially dangerous amount of perfusate leaks. Therefore, a search was initiated for a continuous, real-time method of monitoring. Stehlin¹⁸ was the first to describe a continuous method as early as 1961 when using the radioactive isotope ¹³¹I-albumin. In 1989, a handheld gamma detector was introduced by Sardi¹⁹; it was to be placed between the thigh and precordia to record the radioactivity of the ¹³¹I-albumin that had leaked into the systemic circulation. However, leakage measurement with the handheld detector is accompanied by a method error, because slight variation in probe positioning produces altered results. Nowadays, when a toxic drug like TNF is used, accurate leakage measurement is extremely important, because even 1% TNF leakage can cause hypotension and 10% leakage can cause potentially fatal septic shocklike syndrome.^13,14 With the establishment of a stationary scintillation detector placed above the heart, the previously mentioned method error was eliminated.²⁰

Despite available detection methods, the data for perfusate leakage into the systemic circulation is rarely mentioned in medical literature. In a previous article covering the period from 1977 until 1990 at this institution, a median of the maximum leakage of 8.0% (range 0%–30%) was reported, with an achieved leakage of <15% in 84% of the patients.⁷ During this period, melphalan was the drug of choice, and a maximum leakage of 15% of melphalan did not prove to cause systemic toxicity.⁶ Nevertheless, this median leakage is considered too high for currently used treatment schemes which include TNF. In the 1977–1990 period there was constant, ongoing experimenting with new monitoring techniques to achieve minimal leakage. From 1977–1982, perfusions were performed with a low flow rate of 100–150 ml/min with a mean leakage of 7.04% ± 1.04%. On the basis of Fontijne’s²¹ study, a more physiological perfusion has been implemented since 1982, with perfusion pressure derived from the patients’ arterial and venous pressure during the operation. This resulted in an increase in flow rate ranging 500–900 ml/min. However, this was accompanied with a significantly higher mean leakage: 10.43% ± 1.82% (P < .001). To reduce the complications due to this increased leakage, since 1987, cytotoxic drugs have been administered only when the observed leakage is <5% over a period of 5 minutes. Consequently, this resulted in a statistically different decrease in leakage to 7.35% ± 0.69% (P < .001) in patients who were actually perfused with cytotoxic drugs.

When the TNF perfusions were introduced in the early nineties, no leakage was accepted of the perfusion circuit into the systemic circulation in contrast to the melphalan perfusions.

In the current series starting from 1991, the overall median of the maximum leakage was calculated at 2.7% (range 0%–21%). This is significantly lower than in the period 1977–1990 (P < .05). We think this difference is based on three aspects.

First, the previous report is based on 331 iliac and 55 femoral perfusions while the current series includes 40 axillary and popliteal perfusions out of 119 total perfusions. Univariate analysis showed that the level of isolation is a factor predicative of systemic leakage where perfusion performed on iliac and femoral vessels had a significant higher chance of leakage than that performed on popliteal and axillary vessels. This aspect has been highlighted before in the literature for which Pace²² described an optimal method for isolation at the iliac vein. His method was to temporary close the common iliac vein during perfusion, for which a mean leakage of 9% was reported. In the Groningen¹⁵ iliac perfusion technique, the external iliac, obturator, hypogastric, and the collateral veins are temporary closed.

Second, since the early nineties, the very toxic cytokine TNF is being used increasingly in the perfusion setting because it has proven to be very effective in limb salvaging for soft tissue sarcoma (STS).^8,23 These cases involve delivery, to the tumor, of TNF levels that are approximately 15–20 times the maximally tolerated systemic levels.⁴ If there is significant leakage (over 10%), the resultant systemic complications could be fatal. However, Stam et al.²⁴ recently published a report in which leakage of up to 65% TNF into the systemic circulation only caused a hypotension, which was easily corrected with either fluid administration or dopamine treatment for 2 days. Interestingly, since the introduction of TNF, there has been a significant decrease in leakage compared with when melphalan is used alone (P < .05). It is our opinion that this decrease in leakage is based on the surgeons’, anesthesiologists’, and perfusionists’ awareness of the added risk of TNF.

The third reason for a decrease in leakage is attributed to the decrease in flow rate as described by Sorkin et al.²⁵ in 1995. As mentioned before, since the early eighties at the Groningen institution, a high flow rate was used according to Fontijne.²¹ But Sorkin observed a decrease in leakage from 12.5% to 2.3% when the flow rate was decreased from 869 ml/min to 286 ml/min. Moreover, Allen²⁶ reported that a 20% decrease in flow rate will reduce leakage from the extremity into general circulation by 50%. This is a very important observation because a sufficient flow rate is required to maintain physiological blood gas values. However, the flow rate should not reach levels that cause raised venous pressure (which would subsequently lead to an increase in regional toxicity and increased systemic leakage). In our institution, a flow rate between 400 and 500ml/min was adapted for this purpose. Just recently, a new technique has been implemented where the flow rate is based on the limb volume and is expected to reduce the leakage even further.

It is interesting to note that even though the leakage rate in the latter years has decreased, we did not notice a decrease in patient morbidity between the two time periods.

If there is substantial leakage, the order of events to minimize leakage is as follows: first of all, the venous tourniquet is reapplied. If this does not result in decreased leakage, systemic arterial pressure is increased to the patient’s normal values by using dopaminergic agents. Third, the patient can be brought into a Trendelenburg position. If the leakage is still unacceptable, the flow rate in the perfusion circuitry is subsequently decreased. In case these measures have no effect in minimizing leakage, the procedure is terminated without perfusing the limb with cytotoxic drugs.

In 1993, Klaase et al.⁵ reported a cumulative systemic leakage of 0.9% (95% confidence interval is 0.7–1.1%) for their series of 438 perfusions performed at their institution. However, to our knowledge, their leakage detection method differed considerably from the one used at this institution. Their method consisted of a small systemic calibration dose and a higher limb dose of ^99mTc-albumin to measure leakage. Compare this to our method of using a small calibration dose of ¹³¹I-albumin and a higher dose of ¹³¹I-albumin for the limb.¹³¹I-albumin has a longer half-life than ^99mTc-albumin (8 days compared with 6 hours, respectively), which results in a longer systemic circulation of the isotope, which therefore enhances steady measurement.²⁷ In our setting, the use of ^99mTc-albumin serves as a control for correct measurement: because its signal to the detector is not leakage dependent, the time curve should display a decrease in radioactivity based on the half-life (t1/2) time of ^99mTc-albumin. Should an unintentional displacement of the detector occur, then this is readily reflected in a deviation from the (t1/2) curve. When only ¹³¹I-albumin is used, such a displacement of the detector might be wrongly interpreted as a change in leakage.

In summary, even under the most optimal conditions, it is not always possible to achieve total limb isolation when perfusing at the iliac or femoral blood vessel. Using our monitoring technique, leakage is readily detected and seems to be superior to the technique of using ^99mTc-albumin alone. Nowadays, leakage from the isolated circuit into systemic circulation has significantly declined compared with the days when melphalan was the sole drug used. This is due to the increased surgeons’, anesthesiologists’, and perfusionists’ awareness of the very toxic TNF, which has caused them, in turn, to operate more cautiously to gain optimal isolation and to readily apply the aforementioned measures to decrease leakage. The flow rate regulation in the HILP circuit and optimal regulation of systemic blood pressure have also been major contributors to this improvement. Together, a leakage rate of 5% or less should be achievable and expected.

	Acknowledgments

 
This study was supported by the Research Foundation IJsselmonde, the Netherlands.

Received for publication September 14, 2000. Accepted for publication December 28, 2000.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Creech O, Krementz ET, Ryan RF, Winblad JN. Chemotherapy of cancer: regional perfusion utilizing an extra-corporeal circuit. Ann Surg 1958; 148: 616–32.[Medline]
2. Cavaliere R, Ciocatto EC, Giovanella BC, et al. Selective heat sensitivity of cancer cells. Biochemical and clinical studies. Cancer 1967; 20: 1351–81.[CrossRef][Medline]
3. Luck J. Action of P-di(2-chloroethyl)-amino-L-phenylalanine on Harding Passey mouse melanoma. Science 1956; 123: 984.[Free Full Text]
4. Benckhuijsen C, Kroon BB, van Geel AN, Wieberdink J. Regional perfusion treatment with melphalan for melanoma in a limb: an evaluation of drug kinetics. Eur J Surg Oncol 1988; 14: 157–63.[Medline]
5. Klaase JM, Kroon BB, van Geel AN, Eggermont AM, Franklin HR. Systemic leakage during isolated limb perfusion for melanoma. Br J Surg 1993; 80: 1124–6.[Medline]
6. Briele HA, Djuric M, Jung DT, Mortell T, Patel MK, Das Gupta TK. Pharmacokinetics of melphalan in clinical isolation perfusion of the extremities. Cancer Res 1985; 45: 1885–9.[Abstract/Free Full Text]
7. Hoekstra H, Naujocks T, Schraffordt Koops H, et al. Continuous leakage monitoring during hyperthermic isolated regional perfusion of the lower limb: techniques and results. Regional Cancer Treatment 1992; 4: 301–4.
8. Eggermont AM, Schraffordt Koops H, Klausner JM, et al. Isolated limb perfusion with tumor necrosis factor and melphalan for limb salvage in 186 patients with locally advanced soft tissue extremity sarcomas. The cumulative multicenter European experience. Ann Surg 1996; 224: 756–64.[CrossRef][Medline]
9. Lienard D, Ewalenko P, Delmotte JJ, Renard N, Lejeune FJ. High-dose recombinant tumor necrosis factor alpha in combination with interferon gamma and melphalan in isolation perfusion of the limbs for melanoma and sarcoma. J Clin Oncol 1992; 10: 52–60.[Abstract]
10. de Wilt JH, ten Hagen TL, de Boeck G, van Tiel ST, de Bruijn EA, Eggermont AM. Tumour necrosis factor alpha increases melphalan concentration in tumour tissue after isolated limb perfusion. Br J Cancer 2000; 82: 1000–3.[CrossRef][Medline]
11. Ruegg C, Yilmaz A, Bieler G, Bamat J, Chaubert P, Lejeune FJ. Evidence for the involvement of endothelial cell integrin alphaVbeta3 in the disruption of the tumor vasculature induced by TNF and IFN-gamma. Nat Med 1998; 4: 408–14.[CrossRef][Medline]
12. Sleijfer S, Le TK, de Jong S, Timmer-Bosscha H, Withoff S, Mulder NH. Combined cytotoxic effects of tumor necrosis factor-alpha with various cytotoxic agents in tumor cell lines that are drug resistant due to mutated p53. J Immunother 1999; 22: 48–53.
13. Thom AK, Alexander HR, Andrich MP, Barker WC, Rosenberg SA, Fraker DL. Cytokine levels and systemic toxicity in patients undergoing isolated limb perfusion with high-dose tumor necrosis factor, interferon gamma, and melphalan. J Clin Oncol 1995; 13: 264–73.[Abstract/Free Full Text]
14. Tracey KJ, Beutler B, Lowry SF, et al. Shock and tissue injury induced by recombinant human cachectin. Science 1986; 234: 470–74.[Abstract/Free Full Text]
15. Schraffordt Koops H, Oldhoff J, Oosterhuis JW, Beekhuis H. Isolated regional perfusion in malignant melanoma of the extremities. World J Surg 1987; 11: 527–33.[CrossRef][Medline]
16. Hafstrom L, Hugander A, Jonsson PE, Westling H, Ehrsson H. Blood leakage and melphalan leakage from the perfusion circuit during regional hyperthermic perfusion for malignant melanoma. Cancer Treat Rep 1984; 68: 867–72.[Medline]
17. Alexander C, Omlor G, Berberich R, Gross G, Feifel G. Rapid measurement of blood leakage during regional chemotherapy. Eur J Nucl Med 1993; 20: 187–91.[Medline]
18. Stehlin J, Clark R, Dewey W. Continuous monitoring of leakage during regional perfusion. Eur J Cancer 1961; 83: 165–71.
19. Sardi A, Minton JP, Mojzisik C, et al. The use of a hand-held gamma detector improves the safety of isolated limb perfusion. J Surg Oncol 1989; 41: 172–6.[Medline]
20. Sandrock D, Horst F, Gatzemeier W, et al. Leakage measurement during selective limb perfusion using a gamma probe. Eur J Nucl Med 1996; 23: 534–8.[Medline]
21. Fontijne WP, Mook PH, Schraffordt Koops H, Oldhoff J, Wildevuur CR. Improved tissue perfusion during pressure regulated hyperthermic regional isolated perfusion. A clinical study. Cancer 1985; 55: 1455–61.[CrossRef][Medline]
22. Pace M. Optimization of the vascular isolation in isolated regional perfusion of the limbs. Regional Cancer Treatment 1990; 3: 90–7.
23. Eggermont AMM, Schraffordt Koops H, Klausner JM, et al. Limb salvage by isolated limb perfusion (ILP) with TNF and melphalan in patients with locally advanced soft tissue sarcomas: Outcome of 270 ILPs in 246 patients. Proc Am Soc Clin Oncol 1999; 18: 535a.
24. Stam TC, Swaak AJ, de Vries MR, ten Hagen TL, Eggermont AM. Systemic toxicity and cytokine/acute phase protein levels in patients after isolated limb perfusion with tumor necrosis factor-alpha complicated by high leakage. Ann Surg Oncol 2000; 7: 268–75.[Abstract]
25. Sorkin P, Abu-Abid S, Lev D, et al. Systemic leakage and side effects of tumor necrosis factor alpha administered via isolated limb perfusion can be manipulated by flow rate adjustment. Arch Surg 1995; 130: 1079–84.[Abstract]
26. Allen RE, Jr. Systemic leakage and side effects of tumor necrosis factor alpha administration via isolated limb perfusion can be manipulated by flow rate adjustment. Arch Surg 1996; 131: 220.[Medline]
27. Terril J. Radiological Health Handbook. Rockville, MD: US Dept of Health, Education and Welfare, 1970.

This article has been cited by other articles:

				L. B. Been, A. J.H. Suurmeijer, P. H. Elsinga, P. L. Jager, R. J. van Ginkel, and H. J. Hoekstra 18F-Fluorodeoxythymidine PET for Evaluating the Response to Hyperthermic Isolated Limb Perfusion for Locally Advanced Soft-Tissue Sarcomas J. Nucl. Med., March 1, 2007; 48(3): 367 - 372. [Abstract] [Full Text] [PDF]

				J. F. Thompson and J. H. W. de Wilt Isolated Limb Perfusion in the Management of Patients With Recurrent Limb Melanoma: An Important but Limited Role Ann. Surg. Oncol., August 1, 2001; 8(7): 564 - 565. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Daryanani, D. Articles by Hoekstra, H.J. Search for Related Content PubMed PubMed Citation Articles by Daryanani, D. Articles by Hoekstra, H.J. Related Collections Surgery