RESEARCH ARTICLE Open Access
Anesthesia management of patients undergoing
hyperthermic isolated limb perfusion with
melphalan for melanoma treatment: an analysis
of 17 cases
Heiner Ruschulte1,5, Serena Shi1, William W Tseng2,6, Kerstin Kolodzie1, Philip C Crawford3, Darren B Schneider7,
Mohammed Kashani-Sabet4, David Minor4, Christian Apfel2and Stanley PL Leong2,4,8*
Background: Hyperthermic isolated limb perfusion (HILP) is used for patients with intractable or extensive in-transit
metastatic melanoma of the limb to deliver high concentrations of cytotoxic agents to the affected limb and offers
a treatment option in a disease stage with a poor prognosis when no treatment is given.
Methods: In a retrospective chart review of 17 cases, we studied the anesthetic and hemodynamic changes during
HILP and its management.
Results: HILP was well tolerated except in one case that is described herein. We present summary data of all cases
undergoing upper and lower limb perfusion, discuss our current clinical practice of preoperative, perioperative and
intraoperative patient care including the management of HILP circuit.
Conclusion: HILP is a challenging procedure, and requires a team effort including the surgical team, anesthesia
care providers, perfusionists and nurses. Intraoperatively, invasive hemodynamic and metabolic monitoring is
indispensable to manage significant hemodynamic and metabolic changes due to fluid shifts and release of
Keywords: In-transit metastatic melanoma, Isolated heated limb perfusion, Melphalan, Anesthesia, Invasive
HILP is a cancer treatment for patients with unre-
sectable recurrent in-transit metastases of cutaneous
melanoma of the limb [1-3]. The technique delivers a
high concentration of cytotoxic agents regionally to
the affected limb and, thus, minimizes systemic tox-
icity [2-4]. Since its first description in 1950s , the
method has gained acceptance as an established treat-
ment modality with high reported complete response
rates, ranging from 40-85%, after HILP treatment [4-8].
HILP is technically complex and demands close col-
laboration amongst surgeons, anesthesiologists, perfu-
sionists and nurses. The procedure involves the use of
an extracorporeal circuit, incorporating a blood pump
and an oxygenator, to heat the perfusate and increase
the oxygen tensions before delivery to the isolated limb
[1,4,9]. During the HILP procedure, a dynamic fluid shift
exists between the systemic vascular compartment and
the vascular compartment of the isolated limb, especially
when the patient is placed upon and disconnected from
the extracorporeal circuit [2,4]. Potential life-threatening
complications, such as acute hypovolemia after HILP,
have been reported [8,10]. However, anesthetic manage-
ment of the procedure has not been much discussed,
and no standard of practice has been established [11,12].
* Correspondence: LeongSX@cpmcri.org
2Department of Surgery, California Pacific Medical Center, San Francisco, CA,
4California Pacific Medical Center (CPMC), San Francisco, CA, USA
Full list of author information is available at the end of the article
© 2013 Ruschulte et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly cited.
Ruschulte et al. BMC Anesthesiology 2013, 13:15
To gain a better understanding of hemodynamic
changes during the HILP procedure, we analyzed the
anesthetic experience during the HILP procedures at
Moffitt Long Hospital, University of California San
Francisco (UCSF), CA, from 7/1/2006 to 7/1/2009. The
information collected was used to establish a periopera-
tive hemodynamic profile so that an intraoperative
protocol during HILP could be established.
The study was approved by the Committee on Human
Research (CHR) at UCSF. We performed a retrospective
chart review study to establish the intraoperative hemo-
dynamic profiles of patients during surgery and HILP.
All consecutive patients with unresectable metastatic
melanoma to isolated limb undergoing HILP treat-
ment between 7/1/2006 to 7/1/2009 at UCSF were
included. We examined retrospectively the periopera-
tive anesthetic data and reviewed the intraoperative
hemodynamic changes in association with major stages
of the HILP procedure, fluid replacement, and medica-
Anesthesia care and monitoring
All patients were seen for preoperative evaluation at the
UCSF Melanoma Clinic including Dermatology, Surgical
Oncology and Medical Oncology. The patients were also
seen by a vascular surgeon, who participated in the can-
nulation of the vessels during the perfusion procedure.
For preoperative anesthesiological evaluation, they were
seen at the hospital’s Prepare Clinic where history and
physical were taken. Blood was drawn for CBC, electro-
lytes and liver function studies. Also, ECG and chest
x-ray were ordered. The anesthetic regimen was ex-
plained, anesthesia-relevant instruction given and con-
On the day of surgery, anesthesia was induced after
premedication with 1–2 mg of midazolam under stand-
ard monitoring. Induction agents were propofol 1–2
mg/kg body weight (= BW), fentanyl 0.5 - 1 mcg/kg BW
and rocuronium 0.6 mg/kg BW for neuromuscular
blockade. After orotracheal intubation, an arterial line, a
central venous line, a nasogastric tube and a urinary
catheter were inserted. In addition to the relatively small
(20G or 18G) peripheral intravenous cannula used for
induction, one or two more large-bore intravenous can-
nulas (14G or 16G) were placed as access for volume re-
placement. Between three and five units of packed red
blood cells (RBC) were ready for each patient.
Anesthesia was maintained with either desflurane or
sevoflurane using low flow (Aisys, Datex Ohmeda,
Waukesha, WI, USA) with an FiO2between 0.5 to 1.0.
Fentanyl was given in increments of 0.5 to 2 mcg/kg BW
per hour. If necessary for surgical nerve monitoring,
further administration of neuromuscular blockers was
avoided. Phenylephrine was occasionally administered in
the isolated limb or in the systemic circulation to
regulate pressure differentials and prevent communica-
tion between the separate “circuits”. Otherwise, standard
medications including pressors were at hand in the
All monitoring parameters and anesthesia machine
data were recorded electronically (Picis, Wakefield, MA;
USA). All anesthetics were carried out by or under the
continuous supervision of experienced and qualified
At the end of the procedure the administration of in-
haled anesthetic was stopped, and after reversal of
neuromuscular blockade, and otherwise fulfilling the cri-
teria for safe extubation, the trachea was extubated in
each patient. The patients were then transferred to an
intensive care unit (ICU) for overnight monitoring.
After induction of anesthesia, the operative site (axilla or
groin) and the entire ipsilateral limb was prepared and
draped in a sterile fashion. A radical lymph node dissec-
tion was carried out, if not previously performed. For
the upper extremity, this comprised of complete removal
of lymph nodes in levels I-III. For the lower extremity,
superficial groin dissection up to the inguinal ligament
was performed in all patients; depending on preoperative
imaging results, in selected patients, deep/pelvic dissec-
tion through a retroperitoneal exposure was also per-
formed. After lymph node dissection, the axillary artery
and vein (for upper extremity) or femoral artery and vein
(for lower extremity) were exposed and side branches ei-
ther ligated or clamped prior to cannulation. The sizes
of the catheters were adjusted according to the sizes of
the vessels. Systemic heparin (300 i.u./kg BW) was given
and the vessels were cannulated. The target ACT was
480 seconds (ACT Plus, Medtronic, Inc., Minneapolis,
MN). Heparin was administered as needed to maintain
this value throughout the perfusion. Steinman pins
were placed in the subcutaneous tissue to secure the
Esmarch tourniquet around the limb, proximal to the
Target limb-site temperatures were 40°C, with perfus-
ate temperatures maintained between 40-42°C to achieve
these temperatures. Tumor temperature was maintained
at 40°C minimum. Extremity temperature targets were
reached within 40 to 70 minutes in all cases. Upon
achieving target limb and tumor temperatures, a calcu-
lated Melphalan dosage was administered intra-arterially,
directly via the circuit reservoir. For upper extremity
13mg Melphalan/liter limb volume (by water replacement
method) were calculated, for lower extremity 10mg/ liter
Ruschulte et al. BMC Anesthesiology 2013, 13:15
Page 2 of 8
limb volume were calculated, respectively, the maximum
dose being limited to 120 mg even if the calculated vol-
ume was higher. Hyperthermic perfusion was then
maintained for 60 minutes.
Arterial blood gases (ABG) were monitored through-
out HILP: Limb and systemic samples were taken
alternately. ACT values were determined every 30 mi-
nutes in both limb and systemic circulations to prevent
thrombus formation. A high extremity PaO2was main-
tained (400–550 mmHg) with attempted maintenance of
Washout of the treated limb was begun after 60 mi-
nutes of HILP after injection of Melphalan at 40°C of
the limb. Approximately 1,000 – 3,000 ml was removed
from the limb by diverting flow from the venous reser-
voir of the circuit into waste canisters. This volume was
replaced with an equal amount of crystalloid solution,
followed by 500–1,000 ml colloidal volume replacement
solution (starch or gelatine-based) and packed red blood
cells, if indicated by clinical presentation and bedside lab
testing. An attempt was made to visualize a noticeable
drop in the hemoglobin level of the discarded perfusate
with a target of a "near-clear" effluent. Tourniquet occlu-
sion was released upon completion of the washout.
At the conclusion of limb perfusion, the artery and
vein were decannulated and the arteriotomy and veno-
tomy sites, respectively, were closed using running poly-
propylene (Prolene) sutures. Protamine was given as
a test dose (0.2 – 0.5mg), and then administered (0.5 -
1 mg/100 units of Heparin) to fully reverse prior hepa-
rinization. Muscle (i.e. sartorius, for lower extremity)
flaps were rotated to provide additional soft tissue cover-
age of the exposed vessels. The operative site was irri-
gated, hemostasis was achieved, and catheter drains
(Jackson-Pratt) were left in place. The overlying fascia
and subcutaneous tissues were closed using a running,
absorbable polyglycolic acid suture. The skin was closed
using staples or sutures and local anesthetic (0.25%
bupivacaine, 10 – 15 mL) was injected subcutaneously.
Limb perfusion circuit
An integrated infant membrane oxygenator (Capiox
Baby Rx05, Terumo Cardiovascular Systems Corp., Ann
Arbor, MI) was used for its low volume prime require-
ment and high efficiency heat exchange capability. A
specially modified tubing pack (1K35R, Medtronic, Inc.,
Minneapolis, MN) with a ¼” tubing circuit was standard.
No arterial filter was used. Arterial line monitoring was
used to prevent inadvertent pressurization. A “Y” con-
nector with attached tubing was placed in the venous
line to allow for limb exsanguination during the washout
phase. A standard roller pump adjusted to occlusive
(HLM w/Perfusion Controller, Cobe Laboratories, Inc.,
Lakewood, CO) was used as the main perfusion console,
and a modified heater-cooler adjusted to exceed 42°C
(Hemotherm 400M, Cincinnati Sub Zero Products, Inc.
Cincinnati, OH) was used as the warming source.
Because of the low flow requirements of HILP, all can-
nulation was achieved by the discretionary use of ven-
ous/vena-caval type cannulae. Venous cannulae are
longer than many arterial styles, and allow for deeper
insertion in arterial vessels, providing better arterial
isolation. Lower extremity cannulation of femoral/iliac
arteries was accomplished with 12-16Fr. Cannulae (DLP
66112–66116, Medtronic, Inc., Minneapolis, MN). Lo-
wer extremity cannulation of femoral/iliac veins was ac-
complished with 14-20Fr. cannulae (DLP 66114–66120,
Medtronic, Inc., Minneapolis, MN,). Upper extremity can-
nulation of brachial/cephalic arteries was accomplished
with 10-14Fr. cannulae (DLP 66110–66114, Medtronic,
Inc., Minneapolis, MN,). Upper extremity venous cannu-
lation was accomplished with 12-16Fr. cannulae (DLP
66112–66116, Medtronic, Inc., Minneapolis, MN,).
Arterial and venous blood temperatures in the extrem-
ity sites, esophagus, tumor, and warming source (water)
were monitored. The perfusion circuit was primed with
400 ml of a balanced electrolyte solution (PlasmaLyte-A,
Baxter Healthcare, Inc., Deerfield, IL), and the perfusate
was pre-warmed to 43 C. Blood flow ranges were calcu-
lated based upon known values for normal extremity
distribution: arm at 9% of cardiac output (CO), and leg
at 18% CO. Flows for all procedures reported ranged be-
tween 300 and 1,100 ml/min (average 386 ml/min at
upper extremity, average 880 ml/min at lower extremity)
with higher flow rates in the femoral/iliac circuit.
Design and data collection
Anesthetic and hospital records were used to collect
demographic data and general medical background in-
formation of each subject, including age, sex, body mass
index, American Society of Anesthesiologists (ASA) clas-
sification, melanoma stages, associated medical diseases,
surgical and limb perfusion technique, perioperative
morbidities, and 3 month status.
The anesthetic and perfusion records were used to
obtain data such as anesthetic technique, monitoring,
anesthetic medication administration, intraoperative fluid
and transfusion management, and perioperative labo-
Hemodynamic indices used for the study included heart
rate (HR), systolic blood pressure (SBP), diastolic blood
pressure (DBP), mean arterial pressure (MAP), central ven-
ous pressure (CVP), urine output, temperature, estimated
blood loss, blood product transfused, and total fluid re-
placement, For statistical analysis, we calculated shock
index values using the HR values divided by SBP values
from the records.
Ruschulte et al. BMC Anesthesiology 2013, 13:15
Page 3 of 8
Hemodynamic data were collected and analyzed in
relationship to major stages of the HILP procedure,
like beginning of the surgery, clamping of infusion
vessels, initiation of extracorporeal circuit, melphalan
infusion, discontinuation of extracorporal circuit, ex-
sanguination of the operated limb and end of surgery.
A waiver of consent was requested with the CHR
application. All data were obtained in the form of a
chart review using established medical and electronic
records available at UCSF. No additional data were
created as a result of the study. All data collected
would have been already recorded by the clinicians
in line with standard clinical practice at the time of
the HILP treatment. Usual perioperative medical care
was unaffected by study participation. No further
patient contact was needed as a part of the study. No
identifiers were included in the data collection pro-
cess or included in research records. Information was
only available to research group members directly
involved in this study. The 3 month follow-up data
were collected from the charts or electronic records.
All data was kept in a secure, password-protected
Data acquired were analyzed using Excel and statistical
software STATA. Data were summarized as mean (±SD)
and analyzed using Spearman’s rank correlation test where
appropriate. Hemodynamic profiles were created by cor-
relating the mean values (±SD) of various hemodynamic
indices in relationship to major stages of the HILP proced-
ure. A p-value of <0.05 was considered significant.
From 2006 to 2009, seventeen patients with Stage III or
in-transit melanoma of the limb received HILP (see
Table 1). All patients were ASA classification II or III.
Age ranked between 42 and 84 years (average 66.06
years), 8 patients were female, 9 were male. Twelve pa-
tients were treated for lower extremity disease and five
patients for upper extremity disease. Two patients had
melanoma of an unknown primary site.
Surgeries took between 269 and 587 minutes (mean
346 minutes), mean pump time was 106 minutes. Dur-
ing limb perfusion, three patients experienced transient
hypotension and one patient went into cardiac arrest at
the time of extubation. However, the majority (14/17,
82%) of patients tolerated the procedure well without
significant intraoperative complications.
According to the anesthesia records, there was a wide
range of volume given throughout the procedure (up to
13,000 cc given in an unusual case), and losses up to
2,800 cc of blood. The volumes of fluid losses from the
exsanguinations of the perfused limbs were difficult to
Table 1 Patient demographics, site of procedure
Initital presentation Isolated limb perfusion and follow-up
complications (< 30
UnknownUnknownIII CR LE- II-Y
L UE,2.0 III CL UE- II Wound dehiscenceY
L UE, forearm2.9 III B L UE- II Atrial fibrillationN
L LE3.8 III C / IVL LE HypotensionI-Y
L LE3.1 III L LE-I Wound infectionN
L LE, footunknownIIIL LE Hypotension, arrestIII Wound infection,
L LE, calfUnknownIII C L LE HypotensionII-N
L LE, thigh4.1 III CL LE- II-Y
Unknown UnknownIII R LE- II-Y
L UE 4.3III B L UE- II Wound seromaN
L LE, thigh 4.4III CL LE- II-Y
L LE, calf1.5 III CL LE- II-Y
L UE,thumb0.5 (MIS)III CL UE- II-Y
L LE unknownIII C L LE- II IleusN
R LE, ankle 1.5III R LE- II-Y
L LE, ankle5.5 IIIL LE- II-N
(LE lower extremitiy, UE, upper extremity).
R UE1.1 III CR UE-I-N
Ruschulte et al. BMC Anesthesiology 2013, 13:15
Page 4 of 8
assess and were not recorded. Usually the amount of
blood loss can be estimated from the volume of blood
being given to the patient either systemically to maintain
adequate MAP or through the perfusion circuit.
Changes in hemodynamics
The adjusted MAP remained comparably stable in most
stages of the procedure. Before unclamping the isolated
and exsanguinated limb, pressors and colloid infusions
were ready to be given in order to avoid hypotension.
The adjusted shock index (HR/SBP) decreased during
the cases. However, there were no significant differences
to be observed (Figure 1).
The patients received between one and six units of
packed red blood cells. In retrospect, both fluid balances
and hematocrit changes did not appear to be suitable for
statistical analysis due to incomplete data material;
hematocrit values showed a trend towards slight de-
creases, but remained above 30%.
One patient (ASA III) who had congestive heart failure
(CHF) due to coronary artery disease and arrhythmias
before, and whose HR was regulated with an auto-
mated implanted cardioverter-defibrillator went into
severe hypotension at the end of his surgery and re-
quired immediate pharmacologic resuscitation and
chest compressions. He responded instantly to the
taken, and recovered without any neurological deficit.
Anaphylaytic reaction to protamine was unlikely as the
protamine test dose was well tolerated. His shock index
as compared to other individuals was lower as showed
in Figure 1. This patient also suffered from obesity, dia-
betes, chronic renal insufficiency and anemia.
Postoperative complications occurred in six patients
and were mostly related to surgical wound infections
(see Table 1). Except for one patient (no. 6), regional
limb toxicity in all other patients consisted of only mild
limb erythema and edema (Grade I-II) , without sig-
nificant sensorimotor deficits. No patient developed
compartment syndrome or required surgical debride-
ment or amputation for tissue loss (Grade IV-V). With
respect to oncologic outcome, at 3 months, nine of the
seventeen patients (53%) were completely free of disease.
Cutaneous melanoma is the sixth most common cancer
in the United States and an increasing health care prob-
lem worldwide . Approximately half of all newly di-
agnosed cases of melanoma occur in the upper or lower
limbs. HILP is a valuable therapeutic modality for mel-
anoma patients with heavy disease burden confined to
an isolated limb. The technical aspects of limb perfusion
were first described by Creech et al. in the 1950s ;
the procedure was subsequently modified to include
hyperthermia, which enhances the cytotoxic effect of
melphalan . Although to date there have been no
randomized controlled trials for in-transit disease of the
limb which have shown prolonged survival with HILP,
this procedure provides locoregional disease control in a
significant percentage of patients that would otherwise
require amputation [16,17].
Other treatments may include isolated limb infusion
(ILI) or intralesional injection with substances such as
interleukin-2 [18,19] and other agents . ILI is less in-
vasive, allows for shorter surgery and perfusion times
and uses lower temperatures and might be an option for
patients in otherwise limited medical and cardiac condi-
tion. Beasley (23) et al.  report an overall response
Figure 1 Shown are the adjusted schock indices of 16/17 patients who tolerated the treatment well throughout all periods of the
procedure. The course of the patient who went into circulatory arrest due to poor cardiac function, fluid imbalance, masked by a paced
heartrate is displayed in comparison. All patients (n=16+1*).
Ruschulte et al. BMC Anesthesiology 2013, 13:15
Page 5 of 8
rate of 61% (33% complete response, 28% partial
The hemodynamic changes in these procedures are im-
portant, and are known from major surgeries involving the
pathophysiology of ischemia and reperfusion. The under-
lying cardiovascular condition with diminished cardiovas-
cular reserves in some of the patients contributed further
to the challenging anesthetic management.
In the setting of HILP, however, it is crucial for the
anesthesia team need to understand that for a while two
separate circulations exist. When compared to vascular
or orthopedic surgery, circulation of the perfused limb
requires special attention due to the ongoing perfusion.
Although it is a goal to keep the perfused limb isolated
from systemic circulation, volume shifts into systemic cir-
culation can occur: melphalan can cause nausea, allergic
reaction and cardiac arrest, and affect the hematopoietic
system from the unavoidable leak. The influence of
mediators such as cytokines that might be generated
during HILP and their possible effects in systemic circu-
lation have not been quantified. While HILP allows for
delivery of higher amounts of melphalan to the disease-
affected limb, regional limb toxicity is seen. In the
1980s, Wieberdink et al. described a grading scale (I-V)
for toxicity associated with limb perfusion . Based
on larger published series, mild erythema or edema
(Grade I-II) is seen in most patients [22,23]; our ex-
perience was similar. Higher grade toxicity requiring
fasciotomy (for compartment syndrome) or amputation
(for tissue loss) has been reported to occur in up to
1-2% of patients undergoing HILP [21,22,24].
Tumor necrosis factor alpha (TNFα) has been exam-
ined both as an additive to HILP with melphalan in sar-
coma surgery , and also its effects on the formation
of cytokines such as interleukin 6 and interleukin 8 ,
the latter being more pronounced when TNFα was
added for therapeutic purposes. Christoforidis et al. 
(2003) found hemodynamic responses similar to septic
shock when TNFα was used during HILP procedures.
TNFα has no approval for perfusion in the United
States. The effects of “natural” TNFα effects may how-
ever contribute to the shock-like phenomena observed.
For the procedure of the patient who went into arrest, we
suspect that his limited cardiac condition was the major
cause. With the cardioverter device running at a fixed rate
the limited hemodynamic response to volume shifts may
have been masked. The overall worse adjusted shock index
of this patient as compared to the remaining 16 patients
indicates the hemodynamic instability during the case.
Other patients with comparable profiles of coexisting dis-
eases tolerated the procedure and the anesthetic well. How-
ever, in a similar situation we would advocate the use of
transesophageal echocardiography to better monitor the
patient’s cardiac function and fluid status. In retrospect,
due to his limited condition, choosing ILI instead of HILP
might have been better tolerated.
Data were retrieved in retrospect. Literature regarding
the anesthetic management of this procedure is sparse. We
identified milestones events in the procedure that could be
found in each case. Because of the wide range of parame-
ters, we displayed the hemodynamic data as adjusted values
which, with a low total n, did not show significant results,
but trends that deserve further consideration.
For HILP a team approach is crucial during which each
member of the surgical team is aware of the particular as-
pects of each stage of the procedure. The changes in the
volume status, have to be monitored invasively and require
proactive and prompt reaction. We would nonetheless
propose a protocol of blood tests performed before, during
and after the surgery with respect to the surgical milestones
to have a more standardized approach (Table 2) which also
is useful for further studies.
Patients with preexisting cardiovascular disease should
have a thorough preoperative cardiological evaluation.
Continuous intraoperative assessment of cardiac function
and fluid status using an echocardiography probe might
be helpful. Using an indicator dilution technique might
be questionable, as the algorithms for hemodynamic
Table 2 Proposed protocol for perioperative and intraoperative blood testing
CBC Hb/Hctliver function ABG Coagulation (PTT, INR)ACT
PREOP prepare clinicxxx
INTRAOP After incisionxxx
During perfusion (q 30)xxx
After decannulation/reestablished circulationxxx
End of surgeryxxx
POSTOP Arrival PACU/ICUxxx (x)
Ruschulte et al. BMC Anesthesiology 2013, 13:15
Page 6 of 8
calculations with a PICCO catheter or a Swan-Ganz
catheter use size and weight. With the clamped extremity
and abnormal temperatures these might lead to wrong
It is important to be careful about blood, urine and
other body fluids secretions. The washout from the ex-
tremity, the pump filling and the circuit were put into
cytostatic waste cans. As a routine, we used special-
safety trash cans for every secretion and urine during
the first 24 hours in recovery and in the ICU.
HILP is a treatment option in an otherwise desperate
situation for patients with invasively growing melanoma
in an extremity. The procedure requires thorough pa-
tient preparation, a dedicated, experienced surgical, per-
fusionist, nursing and anesthesia team, a perioperative
setting that allows for intensive monitoring and care
throughout the entire procedure and the postoperative
period. High risk patients with significant cardiac com-
promise should probably not be candidates for HILP.
We suggest a standard treatment scheme of monitoring,
blood tests and record keeping especially with respect to
hemodynamic and volume changes.
No author had any competing interests.
During their collaboration at UCSF, HR and SL had the idea to analyse the
cases. They wrote introduction, results and discussion and edited the entire
manuscript. SS, KK and CA collected and analysed the data material. WT and
SL wrote the section of the surgical methods and provided clinical data
about the clinical results. PC and DS provided methdological information
regarding vascular access and extracorporeal circulation. MKS and DM
contributed to the discussion section regarding therapeutic options and
outcomes. All authors read and approved the final manuscript.
This study was supported by departmental funds from the Department of
Anesthesia & Perioperative Care, UCSF. The authors gratefully acknowledge
support of publication-related expenses by the German Research Council
(Deutsche Forschungsgemeinschaft, DFG).
1Department of Anesthesia & Perioperative Care, California Pacific Medical
Center, San Francisco, CA, USA.2Department of Surgery, California Pacific
Medical Center, San Francisco, CA, USA.3UCSF Adult Perfusion Service,
California Pacific Medical Center, San Francisco, CA, USA.4California Pacific
Medical Center (CPMC), San Francisco, CA, USA.5Department of
Anesthesiology & Intensive Care Medicine, Hannover Medical School Medical
Center, Hannover, Germany.6Department of Surgical Oncology, University of
Texas, M.D. Anderson Cancer Center, Houston, TX, USA.7Division of Vascular
and Endovascular Surgery, Weil Cornell Medical College, New York, NY, USA.
8Center for Melanoma Research and Treatment, California Pacific Medical
Center, California Pacific Medical Center Research Institute, 2340 Clay Street,
2nd Floor, San Francisco, CA 94115, USA.
Received: 12 November 2012 Accepted: 11 July 2013
Published: 17 July 2013
1. Thompson JF, De Wilt JH: Isolated limb perfusion in the management of
patients with recurrent limb melanoma: an important but limited role.
Ann Surg Oncol 2001, 8:564–565.
2.Thompson JF, Kam PC: Current status of isolated limb infusion with mild
hyperthermia for melanoma. Int J Hyperthermia 2008, 24:219–225.
3. Beasley GM, Ross MI, Tyler DS: Future directions in regional treatment
strategies for melanoma and sarcoma. Int J Hyperthermia 2008,
4. Schaadt J, Crowley R, Miller D, Kavanah M: Isolated limb perfusion:
a literature review. J Extra Corpor Technol 2002, 34:130–143.
5. Creech O Jr, Krementz ET, Ryan RF, Winblad JN: Chemotherapy of cancer:
regional perfusion utilizing an extracorporeal circuit. Ann Surg 1958,
6. Klopp CT, Alford TC, Bateman J, Berry GN, Winship T: Fractionated intra-
arterial cancer; chemotherapy with methyl bis amine hydrochloride; a
preliminary report. Ann Surg 1950, 132:811–832.
7. Sanki A, Kam PC, Thompson JF: Long-term results of hyperthermic,
isolated limb perfusion for melanoma: a reflection of tumor biology. Ann
Surg 2007, 245:591–596.
8. Kroon HM, Moncrieff M, Kam PC, Thompson JF: Outcomes following
isolated limb infusion for melanoma. A 14-year experience. Ann Surg
Oncol 2008, 15:3003–3013.
9. McDermott P, Lawson DS, Walczak R Jr, Tyler D, Shearer IR: An isolated
limb infusion technique: a guide for the perfusionist. J Extra Corpor
Technol 2005, 37:396–399.
10.Yam PC, Ashcroft P: Acute hypovolemia after weaning from isolated limb
perfusion. J Cardiothorac Vasc Anesth 1995, 9:722–724.
11.Goldberg ME, Rosenblum HM, Seltzer JL, Rosato FE: Isolated regional
perfusion; anaesthetic technique, monitoring and blood replacement.
Can Anaesth Soc J 1984, 31:552–558.
12. Kam PC, Thompson JF: Anaesthetic experience with isolated limb
perfusion with melphalan for melanoma. Perfusion 1996, 11:395–399.
13. Wieberdink J, Benckhuysen C, Braat RP, van Slooten EA, Olthuis GA:
Dosimetry in isolation perfusion of the limbs by assessment of perfused
tissue volume and grading of toxic tissue reactions. Eur J Cancer Clin
Oncol 1982, 18:905–910.
14. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ: Cancer statistics, 2009.
CA Cancer J Clin 2009, 59:225–249.
15. Cavaliere R, Ciocatto EC, Giovanella BC, Heidelberger C, Johnson RO,
Margottini M: Selective heat sensitivity of cancer cells. Biochemical and
clinical studies. Cancer 1967, 20:1351–1381.
16.Coleman A, Augustine CK, Beasley G, Sanders G, Tyler D: Optimizing
regional infusion treatment strategies for melanoma of the extremities.
Expert Rev Anticancer Ther 2009, 9:1599–1609.
17.Kroon BB, Noorda EM, Vrouenraets BC, van Slooten GW, Nieweg OE:
Isolated limb perfusion for melanoma. Surg Oncol Clin N Am 2008,
18.Huismans AM, Kroon HM, Kam PC, Thompson JF: Does increased
experience with isolated limb infusion for advanced limb melanoma
influence outcome? A comparison of two treatment periods at a
single institution. Ann Surg Oncol 2011, 18:1877–1883.
19. Weide B, Derhovanessian E, Pflugfelder A, Eigentler TK, Radny P, Zelba H:
High response rate after intratumoral treatment with interleukin-2:
results from a phase 2 study in 51 patients with metastasized
melanoma. Cancer 2010, 116:4139–4146.
20.Leong SP, Gershenwald JE, Soong SJ, Schadendorf D, Tarhini AA, Agarwala
S: Cutaneous melanoma: a model to study cancer metastasis. J Surg
Oncol 2011, 103:538–549.
21.Beasley GM, Caudle A, Petersen RP, McMahon NS, Padussis J, Mosca
PJ: A multi-institutional experience of isolated limb infusion:
defining response and toxicity in the US. J Am Coll Surg 2009,
22. Vrouenraets BC, Eggermont AM, Hart AA, Klaase JM, van Geel AN, Nieweg
OE: Regional toxicity after isolated limb perfusion with melphalan and
tumour necrosis factor- alpha versus toxicity after melphalan alone. Eur J
Surg Oncol 2001, 27:390–395.
23.Pace M, Gattai R, Matteini M, Mascitelli EM, Bechi P: Toxicity and morbidity
after isolated lower limb perfusion in 242 chemo-hyperthermal
treatments for cutaneous melanoma: the experience of the Tuscan
Reference Centre. J Exp Clin Cancer Res 2008, 27:67.
Ruschulte et al. BMC Anesthesiology 2013, 13:15
Page 7 of 8
24. Kettelhack C, Hohenberger P, Schulze G, Kilpert B, Schlag PM: Induction of
systemic serum procalcitonin and cardiocirculatory reactions after
isolated limb perfusion with recombinant human tumor necrosis
factor-alpha and melphalan. Crit Care Med 2000, 28:1040–1046.
Bonvalot S, Laplanche A, Lejeune F, Stoeckle E, Le Péchoux C, Vanel D,
Terrier P, Lumbroso J, Ricard M, Antoni G, Cavalcanti A, Robert C, Lassau N,
Blay JY, Le Cesne A: Limb salvage with isolated perfusion for soft tissue
sarcoma could less TNF-α be better? Ann Oncol 2005, 16:1061–1068.
Ferroni P, Di Filippo F, Martini F, Spila A, D'Alessandro R, Cavaliere F, Anzà
M, Garinei R, Aloe S, Carone MD, Gazzaniga PP, Guadagni F: Effects of
isolated limb perfusion with tumor necrosis factor-alpha on circulating
levels of proinflammatory cytokines. J Immunother 2001, 24(4):354–362.
Christoforidis D, Chasso PG, Mosimann F, Lienard D, Brunstein F, Bejko D,
Lejeune FJ, Chiolero R: Isolated limb perfusion: distinct tourniquet and
tumor necrosis factor effects on the early hemodynamic response. Arch
Surg 2003, 138:17–25.
Cite this article as: Ruschulte et al.: Anesthesia management of patients
undergoing hyperthermic isolated limb perfusion with melphalan for
melanoma treatment: an analysis of 17 cases. BMC Anesthesiology
Submit your next manuscript to BioMed Central
and take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at
Ruschulte et al. BMC Anesthesiology 2013, 13:15
Page 8 of 8