(Chest. 2001;120:289-294.)
© 2001
American College of Chest Physicians
Thoracoscopy or CT-Guided Biopsy for Residual Intrathoracic Masses After Treatment of Lymphoma*
Dominique Gossot, MD;
Philippe Girard, MD, FCCP;
Eric de Kerviler, MD;
Pauline Brice, MD;
Jean-Didier Rain, MD;
Thierry Leblanc, MD and
Dominique Grunenwald, MD
*
From the Thoracic Department (Drs. Gossot, Girard, Grunenwald), Institut Mutualiste Montsouris; and Departments of Radiology (Dr. de Kerviler), Haematology (Drs. Brice and Leblanc), and Nuclear Medicine (Dr. Rain), Hôpital Saint-Louis, Paris, France.
Correspondence to: Dominique Gossot, MD, Thoracic Department, Institut Mutualiste Montsouris, 42 Bd Jourdan, F-75014 Paris, France; e-mail: dominique.gossot{at}imm.fr
 |
Abstract
|
|---|
Background: An intrathoracic mass persists after
completion of treatment in 20% of the patients treated for Hodgkin’s
disease (HD) or non-Hodgkin’s lymphoma (NHL). Gallium scan and
positron emission tomography allow for diagnosis in most cases.
However, in some patients, a pathologic examination of the residual
mass (RM) is required. The aim of this study was to evaluate the
results of a thoracoscopic approach for intrathoracic RM, as compared
with image-guided biopsies.
Patients and methods: From
1996 to 1998, 29 consecutive patients treated for NLH (n = 11) or HD
(n = 18) were referred either to radiology (group R; n = 8) or to
surgery (group S; n = 21) for biopsy of an intrathoracic RM. There
were 13 male and 16 female patients ranging in age from 15 to 56 years
(mean, 32 years). The reason for a biopsy was the inability to
determine the nature of the RM by means of radiologic examination or
scintigraphy. Biopsy was defined as successful when (1) residual
lymphoma was found in the specimen, or (2) benign tissue was found and
the patient remained disease-free after a minimal follow-up period of
12 months. A biopsy was defined as a failure when a local recurrence
occurred in a patient with a diagnosis of benign lesion.
Results: No significant procedure-related complications
occurred in either group. The mean follow-up was 26 months (range, 13
to 72 months). In group R, residual lymphoma was found in only one
patient. In group S, residual lymphoma was found in seven patients
(p = 0.5). In the seven patients of group R with a diagnosis of
benign mediastinal lesion, two patients had a local recurrence and one
had a recurrence within the abdomen. In the 15 patients of group S in
whom no residual disease was found, 1 patient had an intrathoracic
recurrence (p = 0.5) while 2 patients had recurrence in a remote
site.
Conclusion: Despite the limited number of
patients in this series, results suggest that a thoracoscopic approach
yields better data than image-guided biopsies.
Key Words: Hodgkin’s disease • gallium scan • lymphoma • mediastinum • residual mass • thoracoscopy
|
Introduction
|
|---|
Persistence
of an intrathoracic mass after completion of treatment of a Hodgkin’s
disease (HD) or a non-Hodgkin’s lymphoma (NHL) is a commonplace
situation. Residual thoracic masses are present in about one third of
patients after treatment for NHL or HD.1
Patients
refractory to treatment or presenting with poor response must be
detected early and changed to invasive therapeutic programs such
as high-dose chemotherapy and bone marrow transplantation. It is
therefore of utmost importance to obtain an accurate diagnosis of the
nature of an intrathoracic residual mass (RM). In these patients who
receive long-term therapy and follow-up, physicians are usually
reluctant to use an invasive method to achieve diagnosis. Among
noninvasive methods are CT, MRI, gallium citrate Ga67 scan
(GS),1
2
3
thallium scan,4
and positron
emission tomography (PET) with fluoro-2-deoxy-D-glucose.5
Until now, GS remained the preferred examination to stage NHL or HD at
completion of therapy. However, its sensitivity might not be as high as
usually asserted,6
and many factors may be responsible for
gallium citrate Ga67 uptake. PET scan is highly sensitive and specific
for carcinomas but is still being evaluated for staging of lymphomas.
Thus, some patients are still referred to the radiologist7
or the surgeon for biopsy.8
The aim of this work was to
assess the role of surgical endoscopic and image-guided techniques in
the diagnosis of intrathoracic RM.
|
Materials and Methods
|
|---|
From 1996 to 1998, 29 patients treated for NLH (n = 11) or HD
(n = 18) have been referred to radiology (group R) or surgery (group
S) for biopsy of an intrathoracic RM (Table 1
). There were 13 male and 16 female patients ranging in age from 13 to
56 years (mean, 32 years). Both groups were comparable. The choice of
radiology or surgery was based on localization criteria; when the mass
was easily reachable under CT scan, radiology was preferred.
Nineteen patients underwent GS before biopsy; 6 patients were in group
R and 13 patients were in group S. Results of GS are listed in Table 2
. The reason for a biopsy was the inability to determine the nature of
the RM by means of radiologic examination or scintigraphy.
Group R
Eight patients underwent a CT-guided biopsy of a mediastinal RM
(group R). There were neither pulmonary nor pleural lesions in this
group. A coagulation screen was performed in all patients, and informed
consent was obtained prior to the procedure. In teenagers, informed
consent was obtained from parents. The biopsy procedure was performed
on an outpatient basis in all cases.
Biopsies were performed by two experienced radiologists from our staff.
CT scanning was used for biopsy guidance. After disinfecting and
sterilely draping the patient’s skin, a local anesthesia was performed
with a 1% lidocaine solution. The puncture was performed by using
standard coaxial percutaneous biopsy technique in all cases using a
semiautomated biopsy gun (Quick-Core biopsy needle; Cook; Bloomington,
IN) that allows one to obtain a 17-mm-long core of tissue. In the
largest masses, we used a sextant technique that consists of tilting
the needle in several directions in order to obtain tissue samples in
different areas. Limited scanning was performed to localize the lesion
and to document needle progression in depth to the target. A 16-gauge
biopsy needle was used in six patients, and an 18-gauge needle was used
in two patients, making two to six passes (mean, 3.6 passes). Biopsies
were fixed in a solution of acetic acid, ethyl alcohol, and formol.
When malignant lymphoma was diagnosed on biopsy specimen, a panel of
antibodies was used to determine cell lineage and histologic subtype.
Group S
Twenty-one patients underwent a thoracoscopic approach under
general anesthesia according to the technique that has been previously
described9
(group S). A split ventilation was used in all
patients, and three to four ports were needed for a rigid 10-mm
thoracoscope and endoscopic instruments. Two patients had a history of
previous conventional surgical biopsy (one sternotomy and one
thoracotomy) that led to intraoperative difficulties.
In the case of a small pulmonary lung nodule, a preoperative
localization technique was used in order to facilitate and shorten the
thoracoscopic resection.10
The placement of a hook wire
combined with methylene blue labeling under CT-scan guidance was the
preferred method for localizing the nodule.11
When a large
tumor was present, a large partial biopsy was performed using scissors
and endoscopic suturing. When the nodule was small, it was fully
resected using a regular wedge resection with an endoscopic stapler. In
the case of a mediastinal tumor, multiple biopsies were performed at
different levels to minimize the hazard of overlooking residual cells.
A chest tube was placed and removed within the second postoperative
day. Patients were discharged from the surgical department between the
first and third postoperative day. The mean follow-up duration was 26
months (13 to 72 months). Two patients died of disease progression
during follow-up. Data were compared using a 
2
test with Yates correction.
Since none of the patients underwent a total removal of the RM, it was
difficult to define success and/or a failure criteria due to the lack
of complete pathologic proof. We chose to assert a biopsy as successful
in the two following instances: (1) residual lymphoma was found in the
specimen, or (2) benign tissue was found and the patient remained
disease free after a minimal follow-up period of 12 months. Similar
criteria are usually chosen when dealing with this
issue.12
A biopsy was defined as a failure when a local
recurrence occurred in a patient with a diagnosis of benign lesion.
|
Results
|
|---|
In group R (n = 8), residual lymphoma was found in only one
patient. Fibrosis was found in three patients; thymic hyperplasia was
found in two patients. Other diagnoses are listed in Table 2
. The only
complication in this group was one minor pneumothorax that did nor
require chest drainage and healed spontaneously.
In group S (n = 21), residual lymphoma was found in seven patients,
fibrosis in six patients, and thymic hyperplasia in two patients
(p = 0.5). Unexpected diseases (tuberculosis [n = 2] and
sarcoidosis [n = 1]) were found in three patients. Other rare
diagnoses are listed in Table 2
. In group S, two patients underwent
biopsy of both a pulmonary nodule and a mediastinal mass. In one of
these patients, residual HD was found in both RMs. In the second
patient, fibrosis was found in the mediastinum and tuberculosis was
found in the lung nodule. Another patient had a loculated mediastinal
lesion (Fig 1
). Fibrosis was found in the upper part of the lesion and residual HD in
the lower part. There was no morbidity in this group. The mean stay in
surgery was 2.2 days (1 to 4 days).
View larger version (71K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1.. Thymic rebound with negative GS result in a
19-year-old woman with HD. Top, a: GS.
Bottom left, b: CT scan at
the end of chemoradiotherapy. Bottom right,
c: CT scan 7 months later demonstrating thymic
enlargement.
|
|
Outcome
In the seven patients of group R with a diagnosis of benign
mediastinal lesion, two patients had a local recurrence and one had
recurrence within the abdomen. In the 14 patients of group S in whom no
residual disease was found, there were two recurrences in a remote site
and one intrathoracic recurrence (p = 0.5; Table 3
). In this patient, the surgical examination had been difficult and
remained incomplete because of tight adhesions related to a previous
sternotomy.
Correlation With GS
In group R, four of the six GS results were positive. Residual
disease was found only in one of the four positive GS findings. Two of
the three patients with a positive GS and negative CT-guided biopsy
results had recurrence. The third patient did not have recurrence but
was treated on the basis of the positive GS result.
In group S, 13 patients underwent a preoperative GS. Six of these
results were positive. In only one patient, fibrosis was found and he
recurred early. In the five other patients with positive GS results,
residual lymphoma was found. On the other hand, in one of the six
patients with negative GS results, persistent HD was found. One patient
with nonconclusive GS result had only necrotic tissue (Table 4
). The correlation with GS was better in group S than in group R.
|
Discussion
|
|---|
An intrathoracic RM is present in > 20% of patients after
treatment of an HD or a NHL.1
Eighteen percent of these
patients experience a relapse at the level of the RM.13
Theoretically, only a complete surgical resection of the RM could
determine that the patient is disease free. A second-look surgery has
been advocated as the method of choice by some
authors.14
15
This attitude is questionable for the
following reasons: (1) complete removal of tumoral and fibrotic tissues
requires large incisions in patients who may have undergone previous
surgery, (2) the morbidity is high, and (3) < 20% of mediastinal RMs
lead to relapse.16
Thus, the use of noninvasive methods such as CT, GS, or PET scan seems
preferable. However, there is no consensus about the method of choice.
CT alone does not allow one to discriminate between residual tumor and
fibrosis or necrosis. According to most recent studies, CT has a poor
sensitivity with respect to the diagnosis of residual disease. In a
retrospective review of HD and NHL, Stumpe et al12
have
found the specificity of CT to be only 41% for HD and 67% for NHL.
MRI has been shown17
to be slightly more reliable with
respect to the diagnosis of fibrosis.
GS is much more reliable than CT and MRI since gallium uptake is
proportional to the amount of residual cells.2
In a recent
study,18
some of us demonstrated that GS has a high
specificity (91%) and a positive predictive value of 81% in 53
patients with RM after treatment of HD. Other authors3
have found similar results for aggressive diffuse NHL. However, for
follicular NHL, results appear to be less reliable.6
The
specificity of GS is not optimal because of numerous causes of
false-positive findings, such as tuberculosis, sarcoidosis, bronchitis,
or thymic hyperplasia.19
The latter is a commonplace cause
of mediastinal enlargement after chemotherapy. It is known as
"rebound" thymic hyperplasia.20
It may occur up to 14
months after completion of chemotherapy and is found in > 11% of the
patients. Although thymic enlargement usually appears as an homogeneous
mass of the anterior mediastinum (Fig 1
, 2
), doubt may remain in some patients. Furthermore, most thymic rebounds
are associated with gallium citrate Ga67 uptake.3
In
addition, GS must not be performed close to the end of chemotherapy to
avoid false-negative findings. Partially necrotic or small-sized tumors
may also lead to false-negative findings in GS.
View larger version (63K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2.. Thymic rebound with positive GS result in a
20-year-old woman with large-cell NHL located in the pelvis.
Left, a: GS. Center,
b: CT scan before chemotherapy. Top
right, c: CT scan at the end of chemotherapy.
Bottom right, d: CT scan 4 months
later.
|
|
PET seems to reach higher specificity. In a series of 44 RMs with
positive CT results, Zinzani et al21
found 100% relapse
among the 13 patients with positive PET findings and only one relapse
(4%) among the 24 patients with negative PET findings. Stumpe et
al12
have found CT and PET to have similar sensitivities
but PET is significantly more specific than CT. Its specificity for RM
is 96% in patients with HD and 100% in patients with NHL. However,
although the predictive positive value of PET is 100%,22
it seems that it mainly predicts early progression but cannot exclude
the presence of minimal residual disease.22
Indeed, in the
series of Jerusalem et al,22
clinical relapses were
observed more frequently in patients with than without RM. These data
and the limited follow-up do not allow to draw definite conclusion
about the accuracy of PET. To our knowledge, no published data are
available concerning the ability of PET to distinguish mediastinal
recurrence from benign thymic hyperplasia.
Thus, there are still circumstances where doubt remains and where a
pathologic proof is required. Due to the minimal invasiveness,
CT-guided biopsies are often considered as the method of choice when no
peripheral palpable lymph node can be biopsied. The accuracy of
CT-guided biopsies for the diagnosis of lymphoma ranges from 80 to 90%
in many series,23
24
25
especially when several samples can
be obtained using a coaxial technique. In the series of Pappa et
al,23
the size of the samples were sufficient for
phenotyping in all cases. The CT-guided approach has become our method
of choice for initial diagnosis of NHL and HD. However, our data
underline its limitations when dealing with RM. In RM, performing
multiple biopsies is essential since residual disease usually coexists
with fibrotic and/or necrotic tissue or thymic hyperplasia (Fig 3
).26
The fact than two of the seven CT-guided biopsy
patients with negative results did have recurrences within the chest
indicates that residual cells were most likely missed despite the use
of large cutting needles and a sextant sampling technique. On the other
hand, in the surgical group, only 1 of 14 patients with negative biopsy
results had a local relapse. Furthermore, in this patient, a previous
history of sternotomy made the thoracoscopic examination difficult and
incomplete.
View larger version (100K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3.. Left, a: loculated
mediastinal RM after completion of chemotherapy in a 13-year-old
patient with HD. Thoracoscopic biopsy revealed fibrosis in the anterior
aspect of the lesion (top right, b) and
residual HD in the posterior aspect (bottom
right, c). A CT-guided procedure would
have more likely sampled the anterior aspect of the mass.
|
|
In conclusion, despite the increasing use of GS and PET scan, there are
still cases where certainty is wanted, ie, where a biopsy is
required. Although our results are not significant because of the
limited number of patients, our data show that CT-guided biopsies do
not determine definitely the nature of a RM after treatment of HD or
NHL. Only surgery allows for multiple biopsies in different sites (as
demonstrated in this series, benign and malignant lesions may coexist),
thus avoiding the risk of missing remaining malignant tissue. In these
patients submitted to invasive therapeutic programs, open conventional
surgery should be avoided for the following reasons: (1) satisfactory
specimens can be obtained by endoscopic surgical techniques, and (2)
open surgery makes an eventual secondary thoracoscopy more difficult
and less profitable.
|
Footnotes
|
|---|
Abbreviations:
GS = gallium citrate Ga67 scan; HD = Hodgkin’s disease;
NHL = non-Hodgkin’s lymphoma; PET = positron emission tomography;
RM = residual mass
Received for publication November 22, 2000.
Accepted for publication January 25, 2001.
|
References
|
|---|
-
Brice, P, Rain, JD, Frija, J, et al (1993) Residual mediastinal mass in malignant lymphoma: value of magnetic resonance imaging and gallium scan. Nouv Rev Fr Hematol 35,457-461
-
Front, D, Ben-Haim, S, Israel, O, et al (1992) Lymphoma: predictive value of Ga-67 scintigraphy after treatment. Radiology 182,359-363[Abstract/Free Full Text]
-
Peylan-Ramu, N, Haddy, TB, Jones, E, et al (1989) High frequency of benign mediastinal uptake of gallium-67 after completion of chemotherapy in children with high-grade non-Hodgkin’s lymphoma. J Clin Oncol 7,1800-1806[Abstract]
-
Fletcher, BD, Xiong, X, Kauffman, WM, et al (1998) Hodgkin disease: use of T1–201 to monitor mediastinal involvement after treatment. Radiology 209,471-475[Abstract/Free Full Text]
-
Hoh, CK, Glaspy, J, Rosen, P, et al (1997) Whole-body FDG-PET imaging for staging of Hodgkin’s disease and lymphoma. J Nucl Med 38,343-348[Abstract/Free Full Text]
-
Gallamini, A, Biggi, A, Fruttero, A, et al (1997) Revisiting the prognostic role of gallium scintigraphy in low-grade non-Hodgkin’s lymphoma. Eur J Nucl Med 24,1499-1506[CrossRef][ISI][Medline]
-
Wittich, G, Nowels, K, Korn, R, et al (1992) Coaxial transthoracic fine-needle biopsy in patients with a history of malignant lymphoma. Radiology 183,175-178[Abstract/Free Full Text]
-
Gossot, D, de Kerviler, E, Brice, P, et al (1998) Surgical endoscopic techniques in the diagnosis and follow-up of patients with lymphoma. Br J Surg 85,1107-1110[CrossRef][ISI][Medline]
-
Gossot, D, Toledo, L, Celerier, M (1996) The thoracoscope as diagnostic tool for solid mediastinal masses. Surg Endosc 10,504-507[CrossRef][ISI][Medline]
-
Gossot, D, Miaux, Y, Guermazi, A, et al (1994) The hook-wire technique for localization of pulmonary nodules during thoracoscopic resection. Chest 105,1467-1469[Abstract/Free Full Text]
-
de Kerviler, E, Gossot, D, Frija, J (1996) Localization techniques for the thoracoscopic resection of pulmonary nodules. Int Surg 81,241-244[ISI][Medline]
-
Stumpe, K, Urbinelli, M, Steinert, H, et al (1998) Whole-body positron emission tomography using fluorodeoxyglucose for staging of lymphoma: effectiveness and comparison with computed tomography. Eur J Nucl Med 25,721-728[CrossRef][ISI][Medline]
-
Canellos, GP (1988) Residual mass in lymphoma may not be residual disease [editorial]. J Clin Oncol 6,931-933[Free Full Text]
-
Goodman, G, Jones, S, Villar, H (1982) Surgical restaging of Hodgkin’s disease. Cancer Treat Rep 66,751-757[ISI][Medline]
-
Sutcliffe, S, Wrigley, P, Timothy, A (1982) Post-treatment laparotomy as a guide to management in patients with Hodgkin’s disease. Cancer Treat Rep 66,759-765[ISI][Medline]
-
Radford, J, Cowan, R, Flanagan, M (1988) The significance of residual mediastinal abnormality on the chest radiograph following treatment for Hodgkin’s disease. J Clin Oncol 6,940-946[Abstract/Free Full Text]
-
Devizzi, L, Maffioli, L, Bonfante, V, et al (1997) Comparison of gallium scan, computed tomography, and magnetic resonance in patients with mediastinal Hodgkin’s disease. Ann Oncol 8,53-56
-
Ionescu, I, Brice, P, Simon, D, et al (2000) Restaging with gallium scan identifies chemosensitive patients and predicts survival of poor-prognosis mediastinal Hodgkin’s disease patients. Med Oncol 17,127-134[ISI][Medline]
-
Barthold, S, Donohoe, K, Fletcher, J (1997) Procedures guideline for gallium scintigraphy in the evaluation of malignant disease. J Nucl Med 38,990-994[Free Full Text]
-
Kissin, CM, Husband, JE, Nicholas, D, et al (1987) Bening thymic enlargement in adults after chemotherapy: CT demonstration. Radiology 163,67-70[Abstract/Free Full Text]
-
Zinzani, P, Magagnoli, M, Chierichetti, F, et al (1999) The role of positron emission tomography (PET) in the management of lymphoma patients. Ann Oncol 10,1181-1184[Abstract/Free Full Text]
-
Jerusalem, G, Beguin, Y, Najjar, F, et al (1999) Whole-body positron emission tomography using fluorodeoxyglucose for posttreatment evaluation in Hodgkin’s disease and non-Hodgkin’s lymphoma has higher diagnostic and prognostic value than classical computed tomography scan imaging. Blood 94,429-433[Abstract/Free Full Text]
-
Pappa, V, Hussain, H, Reznezk, R, et al (1996) Role of image-guided core-needle biopsy in the management of patients with lymphoma. J Clin Oncol 14,2427-2430[Abstract]
-
Ben-Yehuda, D, Pollicak, A, Okon, E, et al (1996) Image-guided core-needle biopsy in malignant lymphoma: experience with 100 patients that suggests the technique is reliable. J Clin Oncol 14,2431-2434[Abstract]
-
de Kerviler, E, Guermazi, A, Zagdanski, A, et al (2000) Image-guided core-needle biopsy in patients with suspected or recurrent lymphomas. Cancer 89,647-652[CrossRef][ISI][Medline]
-
Gossot, D, Fritsch, S, Toledo, L, et al (1996) Mediastinoscopy vs thoracoscopy for isolated mediastinal lymph nodes: results of a prospective non-randomized trial. Chest 110,1328-1331[Abstract/Free Full Text]
TOP
Abstract
Introduction
