Journal of Pediatric Surgery (2013) 48, 17271732

www.elsevier.com/locate/jpedsurg

Clinical outcomes in children with adrenal neuroblastoma

undergoing open versus laparoscopic adrenalectomy
Cassandra M. Kelleher a,, Lauren Smithson b , Louis L. Nguyen c , Giovanni Casadiego d ,
Ahmed Nasr d , Meredith S. Irwin e , Justin T. Gerstle d
a

MassGeneral Hospital for Children, Boston, MA

Buffalo General Hospital, Buffalo, NY
c
Brigham & Womens Hospital, Boston, MA
d
Division of General and Thoracic Surgery, Hospital for Sick Children, University of Toronto, Toronto, ON
e
Department of Pediatrics, Hospital for Sick Children, University of Toronto, Toronto, ON
b

Received 16 July 2012; revised 19 March 2013; accepted 20 March 2013

Key words:
Neuroblastoma;
Adrenal tumor;
Laparoscopy

Abstract
Background: Laparoscopic resection of adrenal neuroblastoma has become a common alternative to open
surgery. Prior reports have largely focused on short-term operative complications. This study compares longterm oncologic outcomes in children undergoing laparoscopic or open adrenalectomy for neuroblastoma.
Methods: Seventy-nine patients at a single center met inclusion criteria for having adrenal neuroblastoma
and undergoing operative resection. Patients were assigned to high or low/intermediate (L/I) risk groups
based upon Childrens Oncology Group (COG) trial enrollment. Criteria for laparoscopic resection were
absence of vascular encasement and size 5 cm in greatest dimension. Comparison between open versus
laparoscopic groups was performed by Wilcoxon ranked-sum and Fishers exact test. Multivariate Cox
proportional hazard models analyzed the primary outcomes of mortality and recurrence.
Results: In the L/I risk category (N = 30) there was one non-neuroblastoma related death in the open cohort.
Six of 7 patients in the High risk Group who underwent laparoscopic resection had favorable outcomes. Only
higher tumor stage (Hazard Ratio 8.455, P = 0.01) and earlier tumor recurrence were associated with
increased mortality (Hazards Ratio 0.932, P = 0.0002). Among patients who met selection criteria for
laparoscopic surgery there was no difference in mortality or recurrence rates between High risk and L/I risk.
Conclusions: Laparoscopic resection of adrenal neuroblastoma is feasible and can be performed with
equivalent recurrence and mortality rates in L/I risk patients and selected High risk patients. These data
suggest that laparoscopic resection of adrenal neuroblastoma should be considered in patients who meet
selection criteria, irrespective of risk group categorization.
2013 Elsevier Inc. All rights reserved.

Corresponding author. Mass General Hospital for Children, Boston,

MA 02114. Tel.: + 1 617 724 1602; fax: + 1 617 726 2167.
E-mail address: ckelleher3@partners.org (C.M. Kelleher).
0022-3468/$ see front matter 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jpedsurg.2013.03.056

The goal of surgical resection for malignant disease is to

confer the best possible oncologic outcome including
progression-free and overall survival. Application of a new
surgical method to resection for malignant disease requires

1728
that equivalent or superior oncologic outcomes be demonstrated when compared to the gold standard operative
approach. Although no such evidence exists for laparoscopic
resection of adrenal neuroblastomas, this procedure has
increased in popularity over the last decade. With this
study we sought to compare the oncologic outcomes in
patients undergoing laparoscopic resection of adrenal neuroblastoma with those undergoing the historical standard
treatment, open adrenalectomy.
Over the previous decade, a number of reports have been
published documenting the operative feasibility and safety of
laparoscopic adrenalectomy in children [110]. Many of
these studies included patients who had laparoscopic resections of adrenal neuroblastomas and other malignant
lesions. In these reports outcomes focused on operative
parameters such as operative time, length of stay and
complications and suggested that a laparoscopic approach
to adrenal neuroblastomas is feasible and safe in highly
selected patients. Although the literature supports the safety
and feasibility of the laparoscopic approach, few reports
comment on recurrence or overall survival [4,8,10]. Thus, it is
unclear whether the laparoscopic approach confers an
equivalent oncologic outcome to the open, gold standard
approach. Leclair and colleagues reported a predicted
disease-free survival of 84% at 5 years for 45 patients with
abdominal neuroblastoma undergoing laparoscopic resection
[11]. However, there are no studies that compare laparoscopic
to open adrenalectomy with respect to oncologic outcomes of
recurrence and mortality. Furthermore, there are no studies
that compare outcomes in open and laparoscopic surgery in
high versus L/I risk patient groups. The goal of this study was
to compare the clinical outcomes in patients undergoing
laparoscopic versus open adrenalectomy for neuroblastoma.

1. Methods
A retrospective chart review of patients operated on at
The Hospital for Sick Children, Toronto was conducted over
a ten-year period, 19982008, with Ethics Review Board
approval. Data were collected from electronic and paper
medical records. 128 patients were identified as having
neuroblastoma. The seventy-nine patients who had primary
adrenal neuroblastoma were included in this study. Tumor
size was reported as the largest dimension measured by the
radiologist on pre-operative imaging (CT, MRI or ultrasound). As per institutional guidelines, patients were deemed
candidates for laparoscopic resection if the pre-operative
imaging showed no vascular encasement and a tumor with
the greatest dimension 5 cm. For this review, vascular
encasement and size were evaluated on pre-operative
imaging and not prior to neoadjuvant therapy nor at the
time of operative intervention. The majority of the
laparoscopic procedures (83%) were performed in the last
5 years of the study period. Operative times listed were for
the entire operative procedure performed and in certain cases

C.M. Kelleher et al.

included additional procedures such as central line placement, bone marrow biopsy and aspirates, nephrectomy, liver
or lung wedge resection, and/or lymphadenectomy in the
majority of cases. Recurrence was defined as new metastatic
disease or new disease identified in the operative vicinity on
any imaging study that was performed after surgery and until
the time of data collection.
Patients were classified into high, intermediate or low risk
groups based upon the Childrens Oncology Group (COG),
Pediatric Oncology Group (POG) or Childrens Cancer Group
(CCG) protocols into which they were enrolled. Treatment
protocols were the active studies open at our center at the time
of each patients diagnosis. For high risk patients these
included COG A3973, COG ANBL0532, POG9341 and CCG
3891; the majority of these high risk patients received
induction chemotherapy for 46 cycles prior to surgical
resection and autologous stem cell transplant following
surgery. Intermediate risk patients were enrolled on COG
A3961 and ANBL0531 with low risk patients on COG
ANBL00P2, POG9641 or had surgery alone without adjuvant
chemotherapy. There were only 6 patients with intermediate
stage disease, 1 who underwent laparoscopic resection and 5
who underwent open resection. The small number of patients
in a separate intermediate risk group would have precluded
valid statistical analysis. Therefore, low and intermediate
groups were combined into one group for analysis (L/I risk
group) as intermediate risk patients have been shown to have
outcomes similar to low risk patients [12,13].
Laparoscopic resections were carried out using three
5 mm trocars and one 12 mm trocar. Adrenal vessels were
clipped and all adrenal glands were removed intact from the
umbilical port within a bag. No lymphadenectomies were
performed during laparoscopic procedures, as these smaller
tumors were not associated with enlarged nodes on imaging.
Open resections were performed through subcostal, chevron
or thoracoabdominal incisions. Tumors encasing or adherent
to blood vessels were removed along adventitial plains and
were removed piecemeal, as is standard for advanced
neuroblastoma cases with vascular encasement, in an attempt
to decrease gross residual disease. Nephrectomy was
performed in five patients, four of whom had direct tumor
invasion into the renal parenchyma or hilum and one of
whom underwent preoperative tumor embolization complicated by renal artery thrombosis.
Statistical analysis was performed using Students t-test,
Wilcoxon ranked-sum, Fishers exact test, multivariate Cox
Proportional Hazard regression analysis, and KaplanMeier
survival plots (SAS 9.2 (Cary, NC)). An alpha level of
0.05, corresponding to a P-value of 0.05 and 95% confidence
intervals, was considered statistically significant.

2. Results
Of the seventy- nine patients eligible for inclusion, 49 were
in the high risk group and thirty in the L/I risk group. In the

Open versus laparoscopic adrenalectomy for neuroblastoma

Fig. 1

Included patients and risk group stratifications.

high risk group, 7 of 49 patients underwent laparoscopic

adrenalectomy. Eleven of 30 patients in the L/I intermediate
risk group underwent laparoscopic adrenalectomy (Fig. 1).
Demographic and operative data are included in Table 1. The
age of the patients in the L/I risk groups tended to be lower than
that of the patients in the high risk groups. Median stage for
high risk patients was 4 (range 24,4S) and for L/I risk patients
median stage was 1 to 2 (range 14 with 4S in the open group).
Tumor dimension was larger for the patients undergoing open
resections than those selected for laparoscopic resection,
6.8 cm versus 2.9 cm for the high risk groups and 6.1 cm
versus 2.4 cm for the L/I risk groups, respectively. Of the L/I
risk patients, 15 underwent surgery only without additional
therapy, with 9 receiving laparoscopic and 6 open adrenalectomies. There was one conversion to an open procedure in each
of the laparoscopic groups with patients maintained in
Table 1

1729

intention-to-treat groups for data analysis. Conversion to

open was elected in both cases secondary to adherence of the
tumor to surrounding structures and inability to safely proceed
with laparoscopic resection.
Operative time was longer for patients undergoing open
adrenalectomy when compared to laparoscopic adrenalectomy in both groups. Operative time was also longer in the
high risk open group when compared to the L/I risk open
group. Additional procedures were performed during both
laparoscopic and open adrenalectomies. In the L/I risk
groups, the additional procedures often included bone
marrow aspiration and biopsies. In the high risk group
significant additional surgical procedures were more often
performed in the patients undergoing open resection and
included nephrectomy, retroperitoneal or para-aortic lymphadenectomy and/or liver or lung wedge resections. These

Demographic and operative data.

Risk Group

High (n = 49)

Operative approach
Mean age (mo)
Surgery alone
Median stage [range]
Largest tumor dimension (cm) [range]
OR time (min)
Converted to Open (%)
Transfused (%)
LOS (days)
Average F/U (mo)

Open (n = 42)
41.3 #
n/a
4 [24]
6.8 [1.113.9]
332 #
n/a
18/48 (37%)
11.31
44

Low/ Intermediate (n = 30)

Lap (n = 7)
48.3 #
n/a
4 [4, 4S]
2.9 [0.94.3]
180
1/7 (14%)
0
4.43 #
19

Open (n = 19)
22.5 #
4
2 [14,4S]
6.1 [1.414.8]
252 #
n/a
2/13 (26%)
9.47
69

n/a = not applicable.

= p 0.05 lap v open approach within risk group.
#
= p 0.05 between High and Low/Intermediate Risk groups undergoing open or laparoscopic adrenalectomy.

Lap (n = 11)
17.0 #
9
1 [12]
2.4 [1.72.9]
157
1/11 (9%)
0
2.5 #
42

1730
Table 2

C.M. Kelleher et al.

Recurrence and mortality.

Risk Group

High (n = 49)

Operative approach
Recurrence
Time to recur [range]
Mean time to mortality [range]^
Mortality

Open (n = 42)
17/36 (47%)*#
20 mos [464]
28 mos [6d-104 mos]
21/48 (50%)#

Low/ Intermediate (n = 30)

Lap (n = 7)
0*
n/a
12 days
1/7 (14%)

Open (n = 19)
0#
n/a
6 days
1/19 (5%)#

Lap (n = 11)
0
n/a
n/a
0

* = p 0.05 lap v open approach within risk group , # = p 0.05 between High and Low/Intermediate Risk groups undergoing open or laparoscopic
adrenalectomy, ^ = only for patients with mortality, n/a=not applicable. Denominator n when data was unavailable for some patients.

procedures were included in the recorded operative times

making the direct comparison of operative times difficult and
conclusions from such comparisons unreliable. There was a
need for transfusion of blood only in patients who received
open operations. Length of stay was significantly longer for
patients receiving open adrenalectomy. There was also a
longer length of stay (LOS) in the patients undergoing
laparoscopic adrenalectomy in the high risk group versus the
L/I risk group. There was a trend towards increased LOS for
the high risk patients undergoing open adrenalectomy when
compared to the high risk laparoscopic cohort (P = 0.07).
Median follow times were 19 and 44 months and 42 and
69 months for the high laparoscopic and open and L/I
laparoscopic and open risk groups, respectively.
Recurrence and mortality data were collected to determine
if laparoscopic resection resulted in acceptable oncologic
outcomes (Table 2). Recurrence, new disease identified in
the operative vicinity or new metastatic lesions on imaging
studies obtained for follow-up, were seen only in the high
risk open group and occurred in 47% of patients (data not
available for 6 patients).
There were 21 deaths in the high risk group undergoing
open surgery with a progression-free survival of 51.4% and
overall survival of 52.4% at 5 years. In the L/I risk group
undergoing laparoscopic surgery, the progression-free sur-

vival and overall survival were 100% at 5 years while in the

open group progression-free survival and overall survival
were 94.7% and at 5 years (Fig. 2). One patient in the L/I
group undergoing open resection died in the early postoperative period (post-operative day 6) from sepsis. One
death from progressive pulmonary complications from
metastatic disease occurred in the high risk laparoscopic
group within the first month after which there were no
additional events. In the high risk laparoscopic group small
sample size and shorter follow-up time precluded calculation
of progression-free or overall survival at 5 years. However,
in the five patients in the high risk laparoscopic group who
were followed for greater than 12 months, the one-year
progression-free survival and overall survival were both
100%. There was no statistically significant difference in
recurrence or survival between the patients undergoing
laparoscopic surgery regardless of risk categorization. The
rates of recurrence and mortality in the high risk group
undergoing laparoscopic adrenalectomy were equivalent to
those patients with L/I risk tumors. Multivariate regression
analysis for mortality included the risk factors of age, tumor
size, operative time, need for blood transfusion, approach to
operation (laparoscopic or open), operative time, risk group
(High, Intermediate or Low) and tumor stage. Higher tumor
stage (Hazard Ratio 8.455, P = 0.01) and earlier recurrence

Fig. 2 KaplanMeier Survival Curve. KaplanMeier survival curves for each group: High Open, High laparoscopic (High Lap), L/I Open
and L/I laparoscopic (L/I Lap). Circles represent censored patients.

Open versus laparoscopic adrenalectomy for neuroblastoma

were associated with increased mortality (Hazard Ratio
0.932, P = 0.0002). The need for blood transfusion (Hazard
Ratio 2.749, P = .038), longer operative time (Hazard Ratio
1.003, P = .045) and assignment to the high risk group
(Hazard Ratio 0.0085, P = .009) correlated with earlier
recurrence. There was a trend towards earlier recurrence time
with higher stage tumors (Hazard Ratio 0.36, P = .088).
Interestingly, the type of surgical approach was not
associated with recurrence or mortality.

3. Discussion
Over the past decade, laparoscopic resection of malignant
tumors has become more commonly performed. The
literature suggests that laparoscopic resection of adrenal
tumors, including some malignancies, in children can be
performed with similar early operative outcomes compared
to open resection. Some studies suggest that earlier time to
post-operative feeding and lengths of stay can be accomplished with a laparoscopic approach to adrenalectomy [6,7].
The outcomes pertaining to oncologic events including
neuroblastoma risk groups, recurrence and mortality rates
have been rarely reported in children undergoing laparoscopic adrenalectomy for neuroblastoma.
In this retrospective review, recurrence and mortality rates
in patients undergoing laparoscopic resection, regardless of
COG neuroblastoma risk category, were equivalent to L/I
Risk patients undergoing open resection. We do not believe
that the surgical approach improved the outcome for these
patients, but rather that there may be a subset of high risk
patients who have favorable clinical and biological features
that influence the size of the tumor, response to chemotherapy, and lack of vascular encasement at the time of resection.
Retrospective analyses of surgical data from the International
Neuroblastoma Risk Group (INRG), International Society of
Paediatric Oncology European Neuroblastoma (SIOPEN),
and the German Society for Paediatric Oncology and
Haematology (GPOH) suggest that patients without vascular
encasement on preoperative imaging may have an improved
event-free survival [1417]. Although we did not directly
correlate vascular encasement with events in this review, the
need for blood transfusion and longer operative times are
indicators of a difficult tumor resection that is often
secondary to vascular encasement. Our data similarly show
that these indicators, the need for blood transfusion and
longer operative times, correlated with an increased risk of
tumor recurrence but not with decreased survival. Additionally, we did not collect data on the size of the tumors prior to
neoadjuvant chemotherapy in the high risk laparoscopic
group and therefore cannot determine whether these patients
initially had smaller or more chemotherapy sensitive tumors
when compared to the high risk open group.
It is inherently difficult to compare outcomes between
these treatment groups because of variations in tumor size
and demographics between the cohorts undergoing open

1731
versus laparoscopic resections. However, the goal of this
study is to highlight the oncologic outcomes in patients with
neuroblastoma who are carefully selected to undergo
laparoscopic adrenalectomy. Currently there is a lack of
evidence to support equivalent or improved oncologic
outcomes for laparoscopic adrenalectomy for neuroblastoma, a procedure that many pediatric surgeons now perform
routinely. This retrospective review provides an evaluation
of survival and recurrence after laparoscopic resection for
adrenal neuroblastoma that includes high risk patients.
Limitations to this study include the small sample size in
the laparoscopic groups and shorter follow-up in the high
risk laparoscopic group. The shorter follow-up time in the
high risk group make predictions of long-term survival less
accurate. However, in the high risk open group, 52% of
deaths and 75% of recurrences occurred within 2 years of
follow up and the median time to relapse in neuroblastoma is
approximately 13 months [18]. Therefore, we would have
expected to observe recurrences and deaths in the high risk
laparoscopic group during a similar time period if laparoscopic resection was associated with an inferior outcome.
Although the small sample sizes limit our analyses, the
results are statistically significant. Further studies with larger
patient numbers are warranted to corroborate and extend the
results of this analysis.

4. Conclusion
Results from this study indicate that laparoscopic
adrenalectomy for neuroblastoma in patients who meet
surgical selection criteria, tumor size smaller than 5 cm and
absence of vascular encasement, is feasible and offers
equivalent progression-free survival and overall survival
regardless of other indicators of risk including age and stage.
These data suggest that laparoscopic resection of adrenal
neuroblastoma should be considered in patients who meet
selection criteria, irrespective of risk group categorization.

References
[1] Romano P, et al. Adrenal masses in children: the role of minimally
invasive surgery. Surg Laparosc Endosc Percutan Tech 2007;17(6):
504-7.
[2] Stanford A, et al. Surgical management of open versus laparoscopic
adrenalectomy: outcome analysis. J Pediatr Surg 2002;37(7):1027-9.
[3] Laje P, Mattei PA. Laparoscopic adrenalectomy for adrenal tumors in
children: a case series. J Laparoendosc Adv Surg Tech A 2009;
19(Suppl 1):S27-9.
[4] Al-Shanafey S, Habib Z. Feasibility and safety of laparoscopic
adrenalectomy in children: special emphasis on neoplastic lesions.
J Laparoendosc Adv Surg Tech A 2008;18(2):306-9.
[5] Castilho LN, et al. Laparoscopic adrenal surgery in children. J Urol
2002;168(1):221-4.
[6] Iwanaka T, et al. Surgical treatment for abdominal neuroblastoma in
the laparoscopic era. Surg Endosc 2001;15(7):751-4.

1732
[7] Iwanaka T, Kawashima H, Uchida H. The laparoscopic approach of
neuroblastoma. Semin Pediatr Surg 2007;16(4):259-65.
[8] de Lagausie P, et al. Laparoscopic adrenal surgery for neuroblastomas
in children. J Urol 2003;170(3):932-5.
[9] Saad DF, et al. Laparoscopic adrenalectomy for neuroblastoma in
children: a report of 6 cases. J Pediatr Surg 2005;40(12):1948-50.
[10] Kadamba P, Habib Z, Rossi L. Experience with laparoscopic
adrenalectomy in children. J Pediatr Surg 2004;39(5):764-7.
[11] Leclair MD, et al. Laparoscopic resection of abdominal neuroblastoma. Ann Surg Oncol 2008;15(1):117-24.
[12] Baker DL, et al. Outcome after reduced chemotherapy for intermediate-risk neuroblastoma. N Engl J Med 2010;363(14):1313-23.
[13] Strother DR, et al. Outcome after surgery alone or with restricted use of
chemotherapy for patients with low-risk neuroblastoma: results of
Children's Oncology Group study P9641. J Clin Oncol 2012;30(15):
1842-8.

C.M. Kelleher et al.

[14] Monclair T, et al. The International Neuroblastoma Risk Group
(INRG) staging system: an INRG Task Force report. J Clin Oncol
2009;27(2):298-303.
[15] Simon T, et al. Review of image defined risk factors in localized
neuroblastoma patients: results of the GPOH NB97 trial. Pediatr Blood
Cancer 2008;50(5):965-9.
[16] Cecchetto G, et al. Surgical risk factors in primary surgery for localized
neuroblastoma: the LNESG1 study of the European International
Society of Pediatric Oncology Neuroblastoma Group. J Clin Oncol
2005;23(33):8483-9.
[17] Warmann SW, et al. Vascular encasement as element of risk
stratification in abdominal neuroblastoma. Surg Oncol 2011;20(4):
231-5.
[18] London WB, et al. Clinical and biologic features predictive of survival
after relapse of neuroblastoma: a report from the International
Neuroblastoma Risk Group project. J Clin Oncol 2011;29(24):3286-92.