British Journal of Anaesthesia 101 (5): 61826 (2008)

doi:10.1093/bja/aen237 Advance Access publication August 8, 2008

CARDIOVASCULAR
Management of patients undergoing multivalvular surgery for
carcinoid heart disease: the role of the anaesthetist
J. G. Castillo1*, F. Filsoufi1, D. H. Adams1, J. Raikhelkar2, B. Zaku1 and G. W. Fischer2
1
Department of Cardiothoracic Surgery, Mount Sinai Medical Center, 1190 Fifth Avenue, Box 1028,
New York, NY 10029, USA. 2Department of Anesthesia, Mount Sinai Medical Center, One Gustave L. Levy
Place, Box 1010, New York, NY 10029, USA
*Corresponding author. E-mail: javier.castillo@mountsinai.org
Background. The management of patients with carcinoid heart disease poses two major

Downloaded from http://bja.oxfordjournals.org/ at departemen anestesi on December 2, 2014

challenges for the anaesthetist: carcinoid crisis and low cardiac output secondary to right
ventricular (RV) failure. Carcinoid crises may be precipitated by the administration of catechol-
amines and histamine-releasing drugs.
Methods. We analysed a series of 11 patients [six males, median (range) age 60 (42 73) yr]
with severe symptomatic carcinoid heart disease who underwent multivalve surgery (right-
sided valves, n8; right- and left-sided valves, n3) between 2001 and 2007.
Results. All patients received octreotide intraoperatively [650 (300 1050) mg] to prevent
carcinoid symptoms and vasoplegia. Those patients on a greater preoperative octreotide
regime required additional intraoperative octreotide [median (range) dose 320 (300 850) vs
750 (650 1050) mg]. Similarly, the use of greater doses of aprotinin (.5 KIU) was associated
with greater requirements for octreotide [475 (300 700) vs 750 (320 1050) mg] and higher
glucose levels (8.5 mmol litre21). Catecholamines were generally required in those patients
who presented with a worse New York Heart Association functional class. Overall mortality
was 18% (n2) and only one episode of mild intraoperative carcinoid crisis was observed.
Conclusions. Carcinoid crisis and RV failure still remain the primary challenges for the
anaesthesiologist while managing patients with carcinoid heart disease. Our study supports the
administration of catecholamines to wean patients off cardiopulmonary bypass, particularly in
the presence of myocardial dysfunction. Those patients on higher octreotide dosages may
require close intraoperative glucose monitoring. Despite high operative mortality, surgical
outcome has been improved potentially due to earlier patient referral and better perioperative
management.
Br J Anaesth 2008; 101: 61826
Keywords: complications, carcinoid syndrome; pharmacology, 5-HT antagonists; surgery,
cardiovascular
Accepted for publication: June 7, 2008

Carcinoid tumours are rare slow-growing tumours derived
from enterochromaffin cells that make up the APUD
(amine precursor and decarboxylation) system. They are
most commonly located in the gastrointestinal tract and
the bronchopulmonary system. The incidence of carcinoid
tumours ranges from 3 to 4 per 100 000 yr in the USA.1
At the time of diagnosis, 20 30% of patients present with
disseminated disease and consequent malignant carcinoid
syndrome defined by cutaneous flushing (90%), gastroin-
testinal hypermotility (70%), heart involvement (30%),
and bronchospasm (15%).1 The term carcinoid heart
disease is used to describe the severe fibrotic endocardial
plaquing resulting form elevated blood concentrations of
vasoactive substances such as 5-hydroxytrypatmine (sero-
tonin), histamine, tachykinins, prostaglandins, and possibly
other growth factors released from the tumour.2 Structural
changes in cardiac valve leaflet architecture leading to
pathological valve function occur in more than 50% of
patients with secondary hepatic metastases.3 This fibrous
reaction particularly involves the right-sided valves,

# The Board of Management and Trustees of the British Journal of Anaesthesia 2008. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
 Anaesthetic management of carcinoid heart disease

Downloaded from http://bja.oxfordjournals.org/ at departemen anestesi on December 2, 2014

Fig 1 Intraoperative analysis. Surgical view of the tricuspid (A) and PVs (B) revealing heavily retracted, thickened, and immobile white-pearly leaflets.
Exploration of the mitral valve shows thickened and retracted leaflets (C) with severe chordae tendineae involvement (D).

extending towards the subvalvular apparatus including the Methods

chordae tendineae and papillary muscles (Fig. 1).
Left-sided involvement is rare and mostly observed in the
presence of an interatrial shunt, endobronchial tumour
localization, and high tumour activity.4 In patients with
carcinoid heart disease, advanced lesions, and severe valv-
ular dysfunction, surgical therapy seems to be the only
definitive treatment option available to improve both the
quality of life and survival.5 6
Carcinoid heart disease poses two distinct challenges
for the anaesthetist during the perioperative period: carci-
noid crisis and low cardiac output syndrome secondary to
right ventricular (RV) failure. Carcinoid crisis, character-
ized by flushing, hypotension, and bronchospasm,7 may be
precipitated by the administration of catecholamines and
histamine-releasing drugs used routinely in the anaesthetic
management of patients presenting for cardiac surgery.8
10
Although the use of the somatostatin analogue octreo-
tide has been shown to stabilize haemodynamic variables
during the perioperative period, it remains clinically chal-
lenging to differentiate between hypotension due to a car-
cinoid crisis as opposed to low cardiac output syndrome
secondary to myocardial dysfunction.11
We report our operative experience and outcome in the
management of 11 patients with carcinoid heart disease
who underwent multivalvular surgery. We also analyse the
impact of vasoactive drugs in triggering a carcinoid crisis
leading to haemodynamic instability and a need for
additional intraoperative octreotide administration.
We retrospectively analysed a series of 11 patients with
severe symptomatic carcinoid heart disease who underwent
multivalvular surgery at our institution between January
2001 and December 2007. The protocol was approved by
our local institutional review board (IRB). The approval
included a waiver of informed consent.
Data collection
Data collection was performed prospectively. Clinical vari-
ables were entered into the New York State Department of
Health (NYSDH, State Cardiac Advisory Committee) data
registry. The NYSDH data registry is a mandatory verified
peer-reviewed data collection system, which includes all
cardiac surgery procedures in New York state. Anaesthetic
data were extracted from an information management
system used intraoperatively (CompuRecord, Philips
Medical, Andover, MA, USA) which includes variables
such as anaesthetic technique, haemodynamic course,
timing, and dosage of all administered medications,
including antifibrinolytics and vasoactive drugs and also
blood products transfused (red blood cells, cryoprecipitate,
fresh frozen plasma, platelets, plasmalyte, and albumin).
Additionally, a thorough medical chart review was per-
formed to obtain all tumour-related variables. Follow-up
survival information was obtained by postoperative visits
and by cross-matching the patients social security number
with a web-based death index (http://ssdi.rootsweb.com/).

619
 Castillo et al.
Preoperative workup
All patients underwent preoperative evaluation includ-
ing measurements of urine 5-hydroxyindoleacetic acid
(5-HIAA), blood serotonin, chromogranin A, and other
chemical markers. Valve morphology and function were
assessed first with transthoracic (TTE) and if necessary by
subsequent transoesophageal (TOE) echocardiography.
The severity of valvular regurgitation, as determined by
Doppler echocardiography, was graded on a scale from 1
to 4 (1, mild; 2, moderate; 3, moderately severe;
and 4, severe). Mean and peak transvalvular gradients
were also measured to grade the degree of valvular steno-
sis. RV and left ventricular functions were also assessed.
Cardiologists experienced in echocardiography interpreted
the preoperative TTE/TOE investigations. Cardiac cathe-
terization was performed in all patients.
Anaesthetic management
A balanced anaesthetic technique was used in all cases.
Before securing the airway with a standard tracheal tube,
general anaesthesia was induced with etomidate and main-
tained with isoflurane. Fentanyl or sufentanil, midazolam,
and vecuronium or rocuronium were used. The doses
administered were at the discretion of the consultant
anaesthetist. Patient monitoring included standard ASA
monitors, an indwelling radial arterial catheter, and a pul-
monary artery catheter inserted through the right internal
jugular vein and TOE for intraoperative assessment of
valve and ventricular function.
A precise anaesthetic care protocol was applied to avoid
carcinoid crisis. A pre-emptive infusion of octreotide at
50 100 mg h21 was started in the holding area before
insertion of the arterial catheter and then continued
throughout the case. During induction, an additional bolus
of 50 100 mg was given. Additionally, octreotide was
administered intermittently throughout the procedure as
bolus injections of 20 100 mg to patients with carcinoid
symptoms or unexplained decline in venous return during
cardiopulmonary bypass (CPB). The infusion of octreotide
was increased to a maximum dose of 300 mg h21 if
required.
Vasoactive medications were used, when necessary, to
wean patients off CPB providing a carcinoid crisis was not
the cause of systemic vasoplegia. Medications included
phenylephrine in incremental doses (40 100 mg) and
calcium chloride as a single bolus dose (500 1500 mg)
with the aim to normalize the concentration of ionized
calcium. When additional inotropic support was needed,
catecholamine infusions were given in patients with severe
ventricular dysfunction and as rescue therapy for those
patients who presented with refractory hypotension.
Antifibrinolytic therapy consisted of the administration
of aprotinin (Trasylolw) to all patients except to those
patients with severe renal dysfunction defined as a
creatinine .221 mmol litre21. These patients received
epsilon-aminocaproic acid (EACA, Amicarw). Throughout
the study period from 2000 to 2006, aprotinin was admi-
nistered according to the Hammersmith protocol. We gave
a loading dose of 2 million kallikrein inhibitory units
(KIU) i.v. over a 30 min time period. After completion of
the loading dose, a maintenance dose of 500 000 KIU h21
was started and continued until the surgical procedure was
finished. In addition, 2 million KIU were added to the
CPB circuit prime. EACA 150 mg kg21 was administered
as a bolus over 30 min, followed by a continuous infusion
of 15 mg kg21 min21. No EACA was added to the CPB
circuit prime. The infusion was continued until the end of
surgery.
The heparin dosage required for adequate anticoagula-
tion for CPB was calculated by the heparin dose response
method to maintain the patients activated clotting time
Downloaded from http://bja.oxfordjournals.org/ at departemen anestesi on December 2, 2014
480 s with a heparin level at or greater than 200 U kg21,
or more specifically 2.7 U ml21 circulating volume.
Additionally, 10 000 U of heparin was added to the pump
prime. After the completion of CPB, heparin was reversed
with protamine sulphate. The protamine dose was calcu-
lated based on the heparin dose and was 0.015 mg U21
of heparin. After the administration of protamine, blood
products ( packed red blood cells, fresh frozen plasma,
platelets, and cryoprecipitate) were given as necessary to
correct coagulopathy.
Our standard postoperative pain management protocol
included the administration of parenteral opioids such as
fentanyl or morphine and non-steroidal anti-inflammatory
drugs such as ketorolac tromethamine. Patients who con-
tinued complaining of pain received a patient-controlled
analgesia pump with one of the aforementioned opioids.
In our centre, it is not a departmental policy for cardiac
surgical patients to receive thoracic epidural catheters after
operation.
Statement of responsibility
The authors had full access to the data and take full
responsibility for their integrity. All authors have read and
agreed to the manuscript as written.
Results
Patient characteristics
Characteristics of 11 patients are summarized in Table 1.
The median (range) age of the patients was 60 (42 73) yr
and five patients were female. Congestive heart failure
with New York Heart Association (NYHA) functional
class III or IV was present in all patients. Symptoms of
severe right-sided heart failure were observed in six
patients. Additional preoperative comorbidity included
hypertension (n5), hepatic failure (n3), renal failure
(n2), and peripheral vascular disease (n1).
620
 Anaesthetic management of carcinoid heart disease

Table 1 Patient characteristics. 5-HIAA, 5-hydroxyindoleacetic acid (reference range 2 10 mg 24 h21); CgA, chromogranin A; Dx, diagnosis; F, female; HF,
hepatic failure; HTN, systemic hypertension; M, male; NYHA, New York Heart Association; PVD, peripheral vascular disease; RF, renal failure

Patient Age/gender Age at Dx Primary tumour Preoperative 5-HIAA CgA Hb NYHA ASA Preoperative
location octreotide (mg 24 h21) (nmol litre21) (gm dl21) class status comorbidities

1 56/F 53 Unknown 1000 mg day21 256 567 12.6 III IV

2 55/M 53 Ileum 30 mg month21 254 830 13.1 IV IV HTN, HF
3 64/F 60 Caecum 30 mg month21 239 850 12.8 IV IV
4 66/M 62 Testis 30 mg month21 264 782 10.9 III IV RF
5 68/M 66 Jejunum 2000 mg day21 229 400 11.7 IV IV HTN, RF
6 42/M 38 Ileum 60 mg month21 258 830 15.6 III IV
7 53/F 51 Caecum 60 mg month21 241 1423 11.3 III IV
8 73/M 68 Ileum 30 mg month21 217 1121 10.5 III IV HF
9 60/F 57 Unknown 30 mg month21 267 668 12.1 III IV HTN, HF, PVD
10 49/M 46 Ileum 60 mg month21 223 553 12.8 III IV HTN
11 67/F 65 Ileum 30 mg month21 315 1255 11.6 III IV HTN

Downloaded from http://bja.oxfordjournals.org/ at departemen anestesi on December 2, 2014

Echocardiography findings
Two-dimensional and Doppler echocardiography revealed
restricted leaflet motion during diastole (Carpentiers func-
tional classification type IIIa) of the tricuspid valve (TV)
in all patients. Tricuspid leaflets were heavily thickened
and retracted resulting in severe valve regurgitation. Four
patients had associated moderate or severe tricuspid steno-
sis with a median (range) peak gradient of 11 (10 13)
mm Hg. Pulmonary valve (PV) lesions were also present
in 10 patients with significantly thickened and retracted
leaflets leading to combined pulmonary regurgitation and
stenosis. The median (range) PV peak gradient was 51
(40 75) mm Hg.
Left-sided valvular disease was noted in three patients
with type IIIa mitral valve (MV) dysfunction due to leaflet
thickening/retraction, and also chordal fusion and shorten-
ing. Mitral regurgitation was graded as moderate to severe
in all patients. No patient presented with mitral stenosis.
One patient presented with moderate aortic valve regurgi-
tation. A patent foramen ovale (PFO) was noted in three
patients ( patients number 2, 6, and 8 in Table 2). Visual
qualitative RV function assessed by TOE revealed mild,
moderate, and severe RV dysfunction before surgery in
one, four, and six patients, respectively.
Anaesthesia
As noted above, anaesthesia was induced with etomidate
and maintained with isoflurane. Other anaesthetic drugs
were administered as follows: fentanyl median (range)
dose 2000 (500 2500) mg or sufentanil 250 (250 500)
mg, midazolam 10 (4 20) mg, and neuromuscular block-
ing agents to facilitate tracheal intubation including vecur-
onium 15 (3 26) mg or rocuronium 65 (48 80) mg.
or 60 mg (n3) of long-acting octreotide monthly,
whereas two patients were taking 1000 and 2000 mg of
s.c. octreotide daily, respectively. Intraoperatively, octreo-
tide was administered to all patients with a median (range)
dose of 650 mg (300 1050). Those patients who were on
a higher octreotide regime (60 mg monthly or 2000 mg
daily) required additional intraoperative octreotide admin-
istration [320 (300 850) vs 750 (650 1050) mg]. Further-
more, patients in whom peak glucose levels were 8.3
mmol litre21 required greater doses of octreotide [310
(300 600) vs 750 (650 1050) mg]. Three patients
required an intraoperative insulin infusion.
Ninety per cent of patients (n10) were given vaso-
active medications. Calcium was administered to eight
patients, phenylephrine to six, epinephrine to five, and
dopamine to two patients. Ten patients required two or
more vasoactive medications. Administration of catechol-
amines was different between patients with congestive
heart failure in NYHA functional class IV (n3) as
opposed to patients in NYHA functional class III (n4).
Ten patients received aprotinin intraoperatively whereas
only one patient was given EACA. The four patients who
were given increased amounts of aprotinin (.5 million
KIU) mostly due to longer surgical procedures also
required higher dosages of octreotide [475 (300 700) vs
750 (320 1050) mg].
Ten patients required blood product administration.
Packed red blood cells were transfused to seven patients
(range 1 4 units), platelets to six patients (range 4 6
units), and fresh frozen plasma to three patients (3 6
units). Total transfusion requirements were higher in five
patients with longer (.200 min) CPB times [3 (1 5) vs 5
(2 14) units], but no differences were found when itemiz-
ing transfusion requirements (RBC vs non-RBC).

Intraoperative variables
Surgical procedures and perioperative variables are
reported in Table 2. All patients were treated with octreo-
tide before operation. Nine patients were on an octreotide
protocol consisting of the administration of 30 mg (n6)
Surgical outcome
Surgical outcome is detailed in Table 3. Intraoperatively,
one patient had a mild carcinoid crisis ( patient number 1)
during the fourth hour of anaesthesia, presenting with

621
 Castillo et al.

Table 2 Intraoperative variables. A, arterial; AV, aortic valve; Ca, calcium; CPB, cardiopulmonary bypass time; CV, central venous; EACA, epsilon-aminocaproic acid; MV, mitral valve; P, repair; PA, pulmonary
facial oedema and flushing. The crisis was overcome with

X-clamp
the administration of an octreotide bolus (100 mg).

(min)

153

102

102
288
146

161
182
43
78

Two patients (18%) in this series died. One patient pre-
sented with multiple postoperative complications (respiratory
(min)
CPB

182

147

188
338

246
350
178

211
209
146
failure and renal failure requiring dialysis) leading to

67
multiorgan system failure. This patient eventually died on
Octreotide

postoperative day 17. The second patient presented with

severe biventricular failure and postoperative hypotension

1050
(mg)

300

310

300
320

650
700
800
850
600

700
due to vasoplegia, and developed hepatorenal syndrome.
Additionally, her postoperative course was complicated by
Antifibrynolitic

heparin-induced thrombocytopenia and thrombosis leading

to ischaemia of the lower extremities. This patient eventually
Aprotinin

Aprotinin

Aprotinin
Aprotinin

Aprotinin
Aprotinin
Aprotinin
Aprotinin
Aprotinin
Aprotinin
therapy

developed multiorgan system failure and died on postopera-

EACA
tive day 10. Finally, one patient developed a haemorrhagic
rectal ulcer, which responded to conservative treatment.
()

()

()
()
()
()
()
()
()
No major postoperative complications, particularly no
Ca

Downloaded from http://bja.oxfordjournals.org/ at departemen anestesi on December 2, 2014

re-exploration for bleeding, occurred in the remaining eight
Phenylephrine

patients. The median (range) length of hospital stay was 11

(626) days.
Vasoactive medications

()

()

()

()
()
()

Follow-up
artery; PFOC, patent foramen ovale closure; PV, pulmonic valve; R, replacement; TV, tricuspid valve; X-clamp, aortic cross-clamp time

Catecholamines

Follow-up was complete for all patients. All nine dis-

charged patients were alive at a median (range) follow-up
of 21 (4 75) months. Five patients were classified as
NYHA class I and four patients as NYHA class II.
()

()

()
()

()

()
()

Patients medication included beta-blockers (n5), diure-

tics (n6), angiotensin-converting enzyme inhibitors
blocking agents
Neuromuscular

Cisatracurium
Cisatracurium
Vecuronium

Vecuronium

Rocuronium

(n2), and calcium channel blockers (n1). All patients

pancuronium
Vecuronium

Vecuronium
Vecuronium
Vecuronium
Vecuronium
Vecuronium
Vecuronium
rocuronium

were on octreotide (30 60 mg monthly, n7; 1000 mg

daily, n2). Echocardiographic findings at follow-up are
reported in Table 3. Although there were no cases of pros-
thetic ring dehiscence, endocarditis, or thromboembolic
Short-acting

events, one of the patients developed early bioprosthetic

Sufentanil

Sufentanil

Sufentanil
Sufentanil

Fentanyl
Fentanyl
Fentanyl
Fentanyl
Fentanyl
Fentanyl
Fentanyl
opioids

structural valve deterioration due to secondary carcinoid

plaquing within 26 months from the first procedure.
Subsequently, this patient underwent re-operative mechan-
Etomidate

Etomidate

Etomidate
Etomidate

Etomidate
Etomidate
Etomidate
Etomidate
Etomidate
Etomidate
Etomidate
induction
Primary

ical pulmonary and TV replacement.

()

()

()
()

()

()

()
PA

Discussion

Carcinoid tumours secrete large amounts of biogenic pep-

()

()

()
()

()
()
()
()
()
()
CV
Catheters

tides and other vasoactive substances, of which serotonin

is the most prominent. Serotonin produced by the carci-
()

()

()
()

()
()
()
()
()
()
()
A

noid tumour is transported to the liver through the portal

PFOC

system and metabolized to 5-HIAA.12 In patients with

()

()

()

liver metastases, large amounts of serotonin enter the sys-

temic venous circulation resulting in carcinoid syndrome,
AV

a constellation of symptoms including intermittent flush-

ing, secretory diarrhoea, bronchospasm, hypotension, and
MV

cardiac involvement.13
P

P
PV

R
R

R
R
R

R
R
R
Surgery

General anaesthetic considerations

TV

R
R

R
R
R
R
R
R
R

The anaesthetic goal for managing patients with carcinoid

10
11

syndrome is to avoid drugs or situations that may directly

1

3
4

5
6
7
8
9

622
 Anaesthetic management of carcinoid heart disease

Table 3 Surgical outcome and follow-up. AR, aortic regurgitation; DIC, disseminated intravascular coagulation; EF, ejection fraction; HITT, heparin-induced
thrombocytopenia; LOS, length of stay; MR, mitral regurgitation; PR, pulmonary regurgitation; RSF, respiratory failure; TR, tricuspid regurgitation

Patient Major complications LOS (days) Discharge status Patient follow-up Follow-up echocardiography

Status (months) EF (%) TR PR MR AR

1 7 Alive Alive (75) 68 Mild Mild

2 6 Alive Alive (60) 55 None Trace
3 11 Alive Alive (50) 55 Mild Mod
4 DIC, MSOF 17 Dead
5 RSF 26 Alive Alive (34) 55 Mild Mild
6 11 Alive Alive (21) 50 Mod Mod Mild Mild
7 7 Alive Alive (23) 55 Mild Trace
8 Rectorragia 15 Alive Alive (19) 65 Mod
9 HITT, MSOF 10 Dead
10 9 Alive Alive (6) 53 None Trace Mild
11 9 Alive Alive (1) 57 None Trace

Downloaded from http://bja.oxfordjournals.org/ at departemen anestesi on December 2, 2014

or indirectly trigger a carcinoid crisis.14 Several factors
have been shown to mediate the release of peptides from
carcinoid tumours, such as emotional stress, hypercapnia,
hypothermia, hypotension (which all have the potential to
release catecholamines), histamine-releasing medications,
and hypertension, which releases bradykinin.8 10 15 Rec-
ommendations for the use of certain anaesthetic drugs are
based mostly on anecdotal but logical experiences reported
in the literature.16 17 Drugs that are considered capable of
triggering the release of mediators include long-acting
opioids, specifically meperidine and morphine, histamine-
releasing neuromuscular relaxants, and also catecholamines.
Although histamine release is most likely to occur in the
presence of a gastric carcinoid tumour,1 drugs with potential
to release histamine such as thiopental, atracurium, succi-
nylcholine, meperidine, or morphine are best avoided. The
anxiolytic properties of benzodiazepines make this class of
drug very useful since emotional stress is an important
factor in the development of carcinoid crises.
In our experience, anaesthetic induction can be accom-
plished safely with etomidate and muscle relaxation
achieved with rocuronium or vecuronium. Short-acting syn-
thetic opioids such as fentanyl or sufentanil are also safe.
Intraoperative role of octreotide
Octreotide represents the pillar of treatment for patients
with carcinoid syndrome and has replaced nearly all pre-
viously used drugs such as ketanserin, methysergide, and
cyproheptadine as the drug of choice for any carcinoid
event. It has been shown to be the most effective treatment
for the deleterious cardiovascular and pulmonary effects
of serotonin and bradykinin, preventing their release from
carcinoid cells and subsequently reducing symptoms in
more than 70% of patients.18 19 Octreotide may be admin-
istered i.v., i.m., or by s.c. depot, and daily octreotide
therapy usually ranges from 100 to 600 mg in two to four
separated doses. Intraoperatively, it is usually given as an
infusion (50 100 mg h21), using i.v. boluses, should a
carcinoid crisis occur. However, treatment of patients with
octreotide raises several issues, particularly with respect to
glucose metabolism. Octreotide is a somatostatin analogue
that is known to suppress several hormones, including
insulin.20 Therefore, its use in combination with steroids
in obese or non-insulin-dependent diabetic patients man-
dates close monitoring of glucose concentrations
throughout the surgical procedure. In line with these con-
siderations, a notable finding in our study was the associ-
ation of significantly elevated glucose levels with greater
doses of octreotide.
Administration of antifibrinolytics
Aprotinin has been shown to reduce platelet activation on
bypass, to decrease the activation threshold of the clotting
factor cascade, and to prevent fibrinolysis. Although the
efficacy of aprotinin has been reported with variable results
in the setting of carcinoid syndrome, it has been tradition-
ally used to inhibit kallikrein peptide released by the
tumour, thus reducing the risk of hypotension and intra-
operative bleeding.21 EACA has been used as an inhibitor
of carcinoid hormones when aprotinin was not available.16
Previous studies have documented the beneficial effects of
EACA on facial oedema during cardiac surgery, but aproti-
nin is still considered a more potent and effective inhibitor
of kallikrein.14 In our series, those patients who were given
higher amounts of aprotinin mostly due to longer operation
times also required higher dosages of octreotide. Our
finding is consistent with previously reported results from
the Mayo Clinic. Weingarten and colleagues,22 in the only
contemporary series on anaesthetic management of patients
undergoing cardiac surgery for carcinoid heart disease,
showed that the use of aprotinin correlated with an
increased intraoperative requirement of octreotide. This
observation, and the fact that aprotinin failed to prevent
carcinoid symptoms after catecholamine administration as
documented in classic studies, led us to suppose that there
might be no need to administer further kallikrein inhibitors
to block carcinoid burden.23 24 Furthermore, recent publi-
cations have corroborated that in addition to previously

623
 Castillo et al.
reported renal and vascular damage, aprotinin adminis-
tration is linked to a significant increase in hospital and
long-term mortality among cardiac surgical patients when
compared with recipients of EACA, particularly in those
undergoing coronary artery bypass grafting.25 26 These
findings support the use of alternative and safer generic
medications such as EACA or tranexamic acid in this
patient population. Consequently, we have restricted the
routine use of aprotinin during cardiac surgery.
Intraoperative hypotension and vasoactive
medication
Haemodynamic instability in patients with carcinoid disease
may be directly related to the tumour activity, to severe
ventricular failure, and to functional changes after CPB.
If hypotension in the setting of a carcinoid crisis is not
responsive to the standard management of optimal volume
replacement, correction of electrolyte abnormalities and
octreotide administration, more effective vasoactive agents
may be considered. In this scenario, calcium and catechol-
amines such as dopamine provide additional inotropic
action to achieve an immediate reaction in response to myo-
cardial depression. Historically, it has been taught that cat-
echolamines trigger release of kallikrein from the tumour,
which in turn leads to bradykinin formation causing vasodi-
lation, increased capillary permeability, and bronchial
constriction.10 Therefore, its use has been avoided due to a
paradoxical hypotensive effect. In our series, all patients
received vasoactive medications. Seven patients required
the administration of catecholamines. Their use was not
associated with higher requirements of octreotide, poten-
tially indicating stable carcinoid tumour activity and
secretion. Previous publications have also reported the
safety of vasopressors and inotrope administration in
patients with carcinoid heart disease,16 22 24 and we there-
fore believe that in conjunction with octreotide, the admin-
istration of catecholamines can be done safely, particularly
in the presence of primary myocardial depression. With the
availability of octreotide, the historical recommendation to
avoid these agents is no longer valid.
RV dysfunction in patients with carcinoid
heart disease
Elevated blood concentrations of vasoactive substances
such as serotonin lead to a severe fibrotic endocardial pla-
quing mostly involving the surface of the right-sided
chambers and valves. As a consequence of these structural
changes, there is a characteristic restrictive filling pattern
due to reduced myocardial compliance and subsequent
diminished ventricular diastolic volume with near-normal
systolic function. Additionally, the tricuspid and PVs
appear echocardiographically thickened, retracted, and
locked in a semi-open position during both phases of the
cardiac cycle (Carpentiers type IIIa dysfunction) which
624
results in progressive RV volume overload and further RV
diastolic pressure elevation.
Intuitively, carcinoid heart disease should be clas-
sified as a restrictive cardiomyopathy resulting in diastolic
dysfunction.27 28 Unfortunately, a review of the most
recent literature reveals no study specifically examining
the incidence or degree of diastolic dysfunction in this
patient population. Without doubt, right-sided heart failure
can be documented clinically and represents the most
common indication for surgery among patients with carci-
noid heart disease.5 However, although classic Doppler-
derived parameters to assess LV diastolic dysfunction
(transmitral flow velocities and pulmonary vein flow vel-
ocities) have been widely developed and validated, modern
emerging concepts of characterizing RV diastolic function
(e.g. strain and strain rate) are still under review for accu-
Downloaded from http://bja.oxfordjournals.org/ at departemen anestesi on December 2, 2014
racy and validation and have yet to be utilized in this par-
ticular patient population.29 Consequently, we were not able
to evaluate our patient population for diastolic dysfunction.
Systolic function of the right ventricle appears initially
normal, yet may be misleading. The afterload reduced
state seen in patients with long-standing insufficient atrio-
ventricular valves could lead to an overestimation of true
systolic function. In addition, the echocardiographer is
confronted with the difficulty of quantifying RV function.
Mortality and morbidity
During the last decades, advances in medical therapy with
somatostatin analogues and more effective oncological
therapies for tumour metastases have resulted in better
control of carcinoid symptoms and potentially improved
survival.20 30 Consequently, right-sided valvular disease
has become a major source of morbidity and mortality.
Owing to the rarity of the disease, limited information is
available regarding the outcome of multivalvular surgery
in patients with carcinoid heart disease. Knot-Craig and
colleagues31 reported one of the first case series with 10
patients who mostly underwent right-sided valve surgery
with an operative mortality of 10%. Later on in 1995,
Robiolio and colleagues2 published a series of nine patients
undergoing right-sided valve replacement with a high
operative mortality of 63%. In the same year, Connolly and
colleagues5 reported nine operative deaths (35%) in a
surgical series of 26 patients. More recently, Moller and
colleagues32 updated the Mayo Clinics experience and
showed that despite high postoperative mortality (16%), a
trend towards improved surgical outcome was achieved.
Similar to these studies, we report two operative deaths
among 11 patients (18%), both of which highlight the chal-
lenges (vasoplegia and right heart failure) associated with
cardiac surgery in this patient population. No major post-
operative complications occurred in the remaining nine
patients. Furthermore, with the meticulous application of
our perioperative protocol of octreotide administration, we
only observed one episode of carcinoid crisis (9%).
 Anaesthetic management of carcinoid heart disease
Others have reported perioperative coagulopathy as a
major source of mortality and morbidity among carcinoid
heart disease patients undergoing cardiac surgery.5 31 33
This complication was mostly observed in elderly patients,
particularly in those with an abnormal liver profile. In a
more recent study, Connolly and colleagues6 reported a
lower incidence of this complication that was comparable
with that observed in our series. The trend towards the
reduction of postoperative bleeding in these patients is
probably related to the advances in perioperative manage-
ment and surgical techniques. As mentioned previously,
all patients except for one receiving aprotinin intraopera-
tively. We believe that the systematic use of this drug,
which is a potent antifibrinolytic agent, may have contrib-
uted in significantly reducing the incidence of bleeding.
Limitations
The retrospective observational nature of the study may
lead to conclusions necessarily limited in their application.
Additionally, the small sample size of the study population
prevents us from evaluating any independent casual
relationship between medications or risk factors and oper-
ative complications.
Conclusions
Carcinoid crisis and low cardiac output syndrome secondary
to RV failure still remain the primary challenges that the
anaesthetist is confronted with while managing patients
with carcinoid heart disease. Our study findings further
support the administration of supportive catecholamines to
wean patients off CPB, particularly in the setting of hypo-
tension due to myocardial dysfunction rather than carcinoid
crisis. Those patients on a greater dose of octreotide before
operation may require greater intraoperative octreotide
dosages, which may require close monitoring of glucose
levels. Finally, despite high operative mortality, a trend
towards improved surgical outcome in patients with carci-
noid heart disease has been achieved potentially due to
earlier patient referral and better perioperative management.
References
1 Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carci-
noid tumors. Cancer 2003; 97: 934 59
2 Robiolio PA, Rigolin VH, Wilson JS, et al. Carcinoid heart disease.
Correlation of high serotonin levels with valvular abnormalities
detected by cardiac catheterization and echocardiography.
Circulation 1995; 92: 790 5
3 Lundin L, Norheim I, Landelius J, Oberg K, Theodorsson-
Norheim E. Carcinoid heart disease: relationship of circulating
vasoactive substances to ultrasound-detectable cardiac abnor-
malities. Circulation 1988; 77: 264 9
625
4 Pellikka PA, Tajik AJ, Khandheria BK, et al. Carcinoid heart
disease. Clinical and echocardiographic spectrum in 74 patients.
Circulation 1993; 87: 1188 96
5 Connolly HM, Nishimura RA, Smith HC, Pellikka PA, Mullany CJ,
Kvols LK. Outcome of cardiac surgery for carcinoid heart
disease. J Am Coll Cardiol 1995; 25: 410 6
6 Connolly HM, Schaff HV, Mullany CJ, Rubin J, Abel MD, Pellikka
PA. Surgical management of left-sided carcinoid heart disease.
Circulation 2001; 104: I36 40
7 Kahil ME, Brown H, Fred HL. The carcinoid crisis. Arch Intern
Med 1964; 114: 26 8
8 Adamson AR, Grahame-Smith DG, Peart WS, Starr M.
Pharmacological blockade of carcinoid flushing provoked by cat-
echolamines and alcohol. Lancet 1969; 2: 293 7
9 Levine RJ, Sjoerdsma A. Pressor amines and the carcinoid flush.
Ann Intern Med 1963; 58: 818 28
10 Peart WS, Robertson JI, Andrews TM. Facial flushing produced in
patients with carcinoid syndrome by intravenous adrenaline and
Downloaded from http://bja.oxfordjournals.org/ at departemen anestesi on December 2, 2014
noradrenaline. Lancet 1959; 2: 715 6
11 Kinney MA, Warner ME, Nagorney DM, et al. Perianaesthetic
risks and outcomes of abdominal surgery for metastatic carcinoid
tumours. Br J Anaesth 2001; 87: 447 52
12 Kema IP, de Vries EG, Muskiet FA. Clinical chemistry of serotonin
and metabolites. J Chromatogr B Biomed Sci Appl 2000; 747: 3348
13 Thorson A, Biorck G, Bjorkman G, Waldenstrom J. Malignant
carcinoid of the small intestine with metastases to the liver, valvu-
lar disease of the right side of the heart ( pulmonary stenosis and
tricuspid regurgitation without septal defects), peripheral vaso-
motor symptoms, bronchoconstriction, and an unusual type of
cyanosis; a clinical and pathologic syndrome. Am Heart J 1954;
47: 795 817
14 Mason RA, Steane PA. Carcinoid syndrome: its relevance to the
anaesthetist. Anaesthesia 1976; 31: 228 42
15 Frolich JC, Bloomgarden ZT, Oates JA, McGuigan JE, Rabinowitz D.
The carcinoid flush. Provocation by pentagastrin and inhibition by
somatostatin. N Engl J Med 1978; 299: 1055 7
16 Neustein SM, Cohen E. Anesthesia for aortic and mitral valve
replacement in a patient with carcinoid heart disease. Anesthes-
iology 1995; 82: 1067 70
17 Propst JW, Siegel LC, Stover EP. Anesthetic considerations for
valve replacement surgery in a patient with carcinoid syndrome.
J Cardiothorac Vasc Anesth 1994; 8: 209 12
18 Kvols LK, Martin JK, Marsh HM, Moertel CG. Rapid reversal of
carcinoid crisis with a somatostatin analogue. N Engl J Med 1985;
313: 1229 30
19 Marsh HM, Martin JK Jr, Kvols LK, et al. Carcinoid crisis during
anesthesia: successful treatment with a somatostatin analogue.
Anesthesiology 1987; 66: 89 91
20 Kvols LK, Moertel CG, OConnell MJ, Schutt AJ, Rubin J,
Hahn RG. Treatment of the malignant carcinoid syndrome.
Evaluation of a long-acting somatostatin analogue. N Engl J Med
1986; 315: 663 6
21 Oates JA, Melmon K, Sjoerdsma A, Gillespie L, Mason DT. Release
of a kinin peptide in the carcinoid syndrome. Lancet 1964; 1: 5147
22 Weingarten TN, Abel MD, Connolly HM, Schroeder DR, Schaff
HV. Intraoperative management of patients with carcinoid heart
disease having valvular surgery: a review of one hundred consecu-
tive cases. Anesth Analg 2007; 105: 11929
23 Gardner B, Dollinger M, Silen W, Back N, OReilly S. Studies of
the carcinoid syndrome: its relationship to serotonin, bradykinin,
and histamine. Surgery 1967; 61: 846 52
24 Hamid SK, Harris DN. Hypotension following valve replacement
surgery in carcinoid heart disease. Anaesthesia 1992; 47: 490 2
 Castillo et al.
25 Mangano DT, Miao Y, Vuylsteke A, et al. Mortality associated with
aprotinin during 5 years following coronary artery bypass graft
surgery. J Am Med Assoc 2007; 297: 471 9 30
26 Mangano DT, Tudor IC, Dietzel C. The risk associated
with aprotinin in cardiac surgery. N Engl J Med 2006; 354:
353 65 31
27 Child JS, Perloff JK. The restrictive cardiomyopathies. Cardiol Clin
1988; 6: 289 316
28 Stollberger C, Finsterer J. Extracardiac medical and neuromuscu- 32
lar implications in restrictive cardiomyopathy. Clin Cardiol 2007;
30: 375 80
29 Teske AJ, De Boeck BW, Olimulder M, Prakken NH, Doevendans 33
PA, Cramer MJ. Echocardiographic assessment of regional
right ventricular function: a head-to-head comparison between
626
2-dimensional and tissue Doppler-derived strain analysis. J Am Soc
Echocardiogr 2008; 21: 275 83
Eriksson BK, Larsson EG, Skogseid BM, Lofberg AM, Lorelius LE,
Oberg KE. Liver embolizations of patients with malignant neuro-
endocrine gastrointestinal tumors. Cancer 1998; 83: 2293 301
Knott-Craig CJ, Schaff HV, Mullany CJ, et al. Carcinoid disease of
the heart. Surgical management of ten patients. J Thorac Cardiovasc
Surg 1992; 104: 47581
Moller JE, Pellikka PA, Bernheim AM, Schaff HV, Rubin J,
Connolly HM. Prognosis of carcinoid heart disease: analysis of
200 cases over two decades. Circulation 2005; 112: 3320 7
Robiolio PA, Rigolin VH, Harrison JK, et al. Predictors of
outcome of tricuspid valve replacement in carcinoid heart
disease. Am J Cardiol 1995; 75: 485 8
Downloaded from http://bja.oxfordjournals.org/ at departemen anestesi on December 2, 2014