Superior Vena Cava Syndrome Associated With Lymphoma Alon Yellin, MD; Matilda Mandel, MD; Gideon Rechavi, MD, PhD; Yoram Neuman, MD; Bracha Ramot, MD; Yair Lieberman, MD
Objective.\p=m-\Toevaluate the workup and treatment of lymphoma and superior vena cava syndrome. Design.\p=m-\Aretrospective survey. Setting.\p=m-\Statehospital serving as a secondary and tertiary referral center for pediatric oncology and pediatric \s=b\
children with
cardiac surgery.
Participants.\p=m-\Elevenchildren aged 11 months to 12 diagnosed as having lymphoma or T-cell acute lymphoblastic leukemia who presented with superior vena cava syndrome during an 11-year period. Interventions.\p=m-\Lymphnode biopsy (two patients), thoracenthesis (five patients), bone marrow aspiration (two patients), and thoracenthesis in addition to bone marrow aspiration (two patients). All aspirates were evaluated with immunohistochemical studies. Chemotherapy was the only years
management intervention.
Results.\p=m-\T-celllymphoma or leukemia accounted for nine cases and Hodgkin's disease for two cases. Respiratory symptoms occurred in 10 patients, including tracheal compression in six patients (compression was life-threatening in one patient). Diagnosis of superior vena cava syndrome was achieved in eight patients using surface-marker analysis of aspirates. The syndrome disappeared within 2 to 10 days. Seven of nine children whose conditions were diagnosed more than 1 year before this writing were alive and free of disease after mean follow-up of 37 months. Conclusions.\p=m-\(1)A specific diagnosis can be achieved in most children with superior vena cava syndrome and lymphoma; (2) Thoracic computed tomographic scans are essential, identifying minute pleural effusions that can aid diagnosis; (3) Anesthetic hazard is related only to severe tracheal compression; (4) Chemotherapy achieves excellent symptomatic relief; and (5) Long-term survival, without disease, is achievable.
(AJDC. 1992;146:1060-1063) vena cava syndrome (SVCS), the clinical pre¬ sentation of superior vena cava obstruction or severe reduction in venous return from the head, neck, and up¬ per extremities, is an uncommon condition in adults. In the pediatrie population, it is extremely rare. Issa et al1 and Ja¬ nin and coworkers2 reviewed the worldwide literature
Superior
Accepted for publication January 27, From the
1992.
Departments of Thoracic Surgery (Drs Yellin and Lieberman)
and Hematology (Drs Mandel, Rechavi, Neuman, and Ramot), The Chaim Sheba Medical Center, Tel Hashomer, Israel, and the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel (Drs Yellin, Mandel, Rechavi, Neuman, Ramot, and Lieberman). Reprint requests to Department of Thoracic Surgery, The Chaim Sheba Medical Center, Tel Hashomer 52621, Israel (Dr Yellin).
available until 1979 and found only 175 children and ad¬ olescents with SVCS. Most of the cases were associated with cardiovascular surgery or the introduction of ventriculoatrial shunts or catheterization of the superior vena cava. As in adults,3 the condition is regarded as an onco¬ logie emergency in the pediatrie literature, including the most recent textbooks,4 and radiotherapy is often initiated without an attempt to establish a diagnosis.2 Recently, we showed that these stigmata related to SVCS in adults5 are often based on misconceptions. From that study, we also learned that in Israel the main cause of SVCS in children is lymphoma. Bearing these points in mind, we reviewed our experience with SVCS in children during the last 11 years, focusing on lymphoma, in an attempt to draw conclusions pertinent to the diagnosis and management of this condition. PATIENTS AND METHODS A computerized search of the archives of the Chaim Sheba Medical Center, Tel Hashomer, Israel, was conducted to identify all pediatrie patients with SVCS from 1980 through October 1990. In addition, all outpatient records of pediatrie cases of lymphoma were manually screened for evidence of SVCS. Included in this search were 66 children with lymphoma, equal numbers of whom had Hodgkin's lymphoma (HL) and non-Hodgkin's lymphoma (NHL). We also included 22 patients with T-cell acute lymphoblastic leukemia, each of whom presented with a large mediasti¬ nal mass. The following data were then studied: duration and type of symptoms, age, sex, findings on physical examination, serum level of lactic dehydrogenase, type and duration of treat¬ ment, and results of complete follow-up. In addition, the original chest roentgenograms, computed tomographic scans, and histo¬ logie slides (when present) were reviewed. Patients coded in the archives for SVCS, but whose records showed no clear evidence of the syndrome, were not included. Also excluded were patients whose venous congestion could be attributed to pericardial or cardiac causes.
Diagnosis was based either on histologie appearance (HL) or on results of immunohistochemical studies (NHL). In the latter cases, mononuclear cells from samples obtained by biopsy, bone mar¬ row aspiration, or thoracentesis were separated using Ficoll dis¬ continuous centrifugation. The cells were immunophenotyped using a panel of antibodies defining B, T, and myeloid antigens. Cases were classified according to Ray's modification of Lukes and Butler's nomenclature6 for HL and according to the National Cancer Institute's formulation for NHL.7,8 Patients with HL were treated with hybrid protocol9; patients with T-cell leukemia were treated according to 1983 Berlin-Frankfurt-Munster protocol.1" As a rule, radiotherapy was not part of the initial therapy. RESULTS Eleven patients younger than age 18 years with lym¬ phoma and SVCS were identified (Table). In addition, a 17-year-old girl with B-cell lymphoma was coded for
Downloaded From: http://archpedi.jamanetwork.com/ by a Oakland University User on 06/14/2015
Clinical Data of Children With
Lymphoma and SVCS* Time to
Roentgeno¬ Sex
Diagnosis T-ALL
Year of
(Ethnicity) Age Diagnosis M (A)
5 y
1982
Symptoms and
Signs
Dyspnea
Duration of Symptoms, d
LDH,
60
980
U/L
Pleural Effusion
Source of
Diagnosis Cervical biopsy
graphic
Time to Clinical
Evidence of Res¬
Results
Improve-
olution,
(Time After
ment, d
d
14
material (no
markers),
Diagnosis) Alive and well (3 mo)
bone marrow
T-ALL
F (A)
11
mo
1986
Hoarseness,
60
1234
respiratory distress,
T-cell
M (A)
lymphoma
3 y
1986
stridor Stridor
Died of
Effusion, bone
sepsis
marrow
(21 d) 30
1132
Effusion
16
Alive and well (4 y 3 mo)
12
Alive and well (3 y
(markers),
cervical fine-needle
aspirate T-cell
M(J)
lymphoma T-cell
M (A)
lymphoma,
10 y
1986
8 y
1986
12 y
1989
M (J)
Exertional
Effusion
11 mo)
Effusion
21
Unclear
Unclear
Died of
sepsis
dyspnea
stage III
Lymphoma, most probably
700
Dyspnea, flank pain
1000
Respiratory
Negative
(3 mo) Alive and well (1 y 6 mo)
2
results of bone
distress
T-cell
marrow
studies, clinical
HL, NS, stage IIA
M(A)
HL, NS, stage
M (J)
T-ALL
M(J)
11 y
1989
7 y
1989
7 y
1989
Hoarseness,
diagnosis Supraclavicular biopsy sample
14
exertional
dyspnea
IIIB
Fever,
Results of mediasti-
21
sweating
Dyspnea, cyanosis
Unclear
Unclear
10
45
2
5
2
8
notomy 21
1093
Negative
results of
two
lymph
Alive and well (1 y 5 mo) Alive and well (1 y 3 mo) Alive and well (1 y)
node
biopsies, bone
marrow
T-ALL
M(J)
5 y
1990
T-cell
F(J)
6 y
1990
lymphoma
Dyspnea
14
358
Effusion
Dyspnea, cyanosis,
70
523
Effusion
tachycardia
2
*SVCS indicates superior vena cava syndrome; LDH, lactate dehydrogenase; T-ALL, T-cell acute lymphoblastic leukemia; HL, NS, nodular sclerosing; A, Arab; J, Jewish; ellipses, data not available; minus sign, absent; and plus sign, present.
SVCS, but review of her records revealed no solid data confirming this diagnosis. Therefore, she was excluded from the study. The youngest adult with SVCS was a 22-
We also found five children with SVCS who did not have lymphoma. Three had undergone Mustard operation; one had undergone Glenn operation; and one suffered from seminoma. Our study included nine boys and two girls, with a mean age of 6.8 years. Five were Arab children from the Gaza
year-old man with HL.
10
Alive and well (2 mo) Alive and well (3 mo
wk)
Hodgkin's lymphoma;
Strip. Two patients had HL (with nodular sclerosing) and nine, T-cell lymphoblastic lymphoma or T-cell acute lymphoblastic leukemia. The prevalence of SVCS in patients with NHL (including T-cell acute lymphoblastic leukemia with large mediastinal masses) was 16% compared with 6% in patients with HL. Symptoms, mainly respiratory, were present for a mean of 26.5 days before diagnosis. Levels of lactic dehydrogenase were elevated in all eight patients with NHL in whom levels were determined. Lev-
Downloaded From: http://archpedi.jamanetwork.com/ by a Oakland University User on 06/14/2015
The two patients with HL were alive and in complete remission 15 and 17 months after diagnosis. Two children with T-cell lymphoma died of sepsis, one within 3 weeks of diagnosis, and the other, 3 months after diagnosis. At the time of this report, five patients were alive, three of whom were alive for more than 3 years after diagnosis. Two ad¬ ditional children were alive and well at the time of this re¬ port, but were followed up for very short periods.
overexposed chest roentgenogram shows severe lowertracheal, carinal, and bronchial narrowing. The resulting respiratory symptoms were unrelated to the superior vena cava syndrome per se. Case 7. A plain
els returned to normal, with disappearance of the medias¬ tinal mass. Pleural effusion was seen on roentgenograms of seven patients with NHL and of one patient with HL. The amount was usually small, and in three cases pleural effusion was detected with computed tomographic scans
only.
Diagnosis of all cases of NHL was based on results of immunohistochemical staining, and diagnosis of HL was based on imprints from biopsy material and histologie ap¬ pearance. Before admission to the Sheba Medical Center, Tel Hashomer, Israel, three patients underwent invasive procedures (cervical lymph node biopsy [two patients] and fine-needle aspiration [one patient]). Immunohistochemi¬ cal studies were not performed on the specimens obtained, and no definite diagnosis was determined at that time. Di¬ agnosis at the Sheba Medical Center was established after studying cells obtained by thoracenthesis (five patients), bone marrow aspiration (two patients), or both (one patient). In the sixth patient in our series, the bone marrow study was not diagnostic. Because the parents refused to permit further invasive procedures and the patient had severe trachéal compression, therapy was initiated. The presumed diagnosis of T-cell lymphoma was sustained by his clinical response to therapy. In the two patients with HL, material for diagnosis was procured by anterior mediastinotomy and supraclavicular biopsy. Therapy was usually initiated within 24 to 48 hours. Clinical improvement was prompt, occurring within 2 to 10 days (Table), with complete clearing of SVCS and res¬ piratory symptoms. Complete roentgenographic resolu¬ tion was relatively slower, requiring a mean of 9.7 days in patients with NHL. In one patient with HL, it took 45 days.
COMMENT Although most cases of SVCS in children and adoles¬ cents described in the literature are iatrogenic, our cases of SVCS were not. We encountered no cases of SVCS associ¬ ated with catheters, and detected only four cases of SVCS after cardiovascular surgery. Eleven of the 12 cases asso¬ ciated with malignant conditions were due to lymphomas, and these 11 cases formed the basis for this study. A literature review in 19831 revealed 150 cases of SVCS in childhood. Only 24 (16%) were neoplastic in origin, with 16 occurring with NHL and one occurring with HL. The more frequent association of SVCS with NHL rather than HL in adults was noted by us5" and others12 and is valid in children, too,1·2 as confirmed in the present study. Superior vena cava syndrome was described in 11 (6%) of 188 children with primary mediastinal tumors.13 Pokorny and Sherman14 described 109 children with medias¬ tinal masses, including 24 with lymphoma. Superior vena cava syndrome occurred in three (20%) of 15 children with NHL and in none with HL. The prevalence in our study was slightly lower. Nevertheless, both reports show that SVCS is not uncommon in children with NHL. Uncom¬ monly, SVCS in children may be caused by histoplasmo¬ sis,15 which is a rare disease in the Middle East. The correct diagnosis can usually be established based on results of serologie tests, with more invasive diagnostic procedures
rarely being required.16 D'Angio and coworkers17 used superior mediastinal syn¬
drome to describe
symptoms of trachéal and
venous com¬
pression occurring in children with superior mediastinal masses. They reported nine cases among 607 children with malignant tumors. Eight involved children with lymphoproliferative disorders. From the details given, it seems that only one child had clear-cut SVCS. Seven additional
of SVCS associated with NHL were reported re¬ cently1* among 25 children with mediastinal nonlymphoblastic lymphoma. These lymphomas do not commonly spread systemically, and their diagnosis usually requires lymph node biopsy, mediastinoscopy, or even thoracotomy. Most cases in our series and the series reported by Issa et al1 were cases of T-cell lymphoma, which tends to disseminate early. Mitchell and coworkers'1' described 72 children with malignant T-cell neoplasms, 19 of whom were considered to have lymphoma. Diagnosis, similar to our experience, was based on surface-marker analysis (immunohistochemical stains) of pleural fluid or bone marrow in most cases. The need for surgically obtained cases
specimens was very limited, in contrast to earlier studies.20 It may be of some significance that the two largest studies examining SVCS and lymphoma (the present study and that by Issa et al1) originated in the Middle East. Many authors1,2'417 have stressed the life-threatening
conditions associated with SVCS and advocate early treat¬ ment without definitive diagnosis. Although the danger of anesthesia in children with large mediastinal masses is well established20 22 and most cases of anesthetic catastro-
Downloaded From: http://archpedi.jamanetwork.com/ by a Oakland University User on 06/14/2015
phes were associated with lymphoma, SVCS per se did not cause the life-threatening conditions. Trachéal compres¬ sion is not uncommon in children with lymphoma21 and was observed in 55% of all the pediatrie cases of HL in an¬ other study.23 Trachéal compression was also common in our patients (Figure), occurring in six patients, and was the main cause of the high rate of severe respiratory symp¬ toms. Pleural effusions were detected in eight patients, but were moderate or large in only three. Before 1980, we en¬ countered two children with large mediastinal masses causing severe trachéal compression. One sustained respi¬ ratory arrest during an attempted mediastinoscopy and was thereafter treated empirically. The second patient un¬ derwent successful supraclavicular lymph node biopsy after induction with local anesthesia. In the present study, one child with severe respiratory distress, nondiagnostic bone marrow aspirate, and markedly elevated levels of lactic dehydrogenase was treated empirically. In all these cases, trachéal compression, rather than SVCS, imposed the anesthetic hazard. Because only two children under¬ went procedures after induction with general anesthesia in our medical center, we do not believe that this experience allows us to make conclusions regarding the hazards of anesthesia for patients with large mediastinal masses. We
conclude that children, like adults with SVCS but not trachéal compression, can undergo diagnostic surgery af¬ ter induction with general anesthesia without increased risk of complications and death.516 Therefore, it is crucial that every patient with a large mediastinal mass be eval¬ uated, preferably with computed tomography, for evi¬ dence of trachéal compression. Although many cases of SVCS associated with ante¬ rior and middle mediastinal masses in children prove to be caused by T-cell lymphoma, this is certainly not the rule. Nonlymphoblastic lymphoma,16 HL, neuroblas¬ toma, germ cell tumors, and even benign diseases can also cause superior vena cava obstruction. The manage¬ ment of these conditions is quite different from that of T-cell lymphoma. Therefore, identification of a specific diagnosis is advisable in any case of a mediastinal mass, regardless of the presence of SVCS. Only when severe trachéal compression imposes an immediate respiratory hazard does diagnosis become less important. Whereas diagnosis of all lymphomas in the past almost always can
required a surgical procedure,13 currently, cytochemical staining and surface-marker analysis of needle-obtained specimens can provide information for an accurate di¬ agnosis. In more localized lymphomas, (ie, nonlympho¬ blastic18 or Hodgkin's lymphoma), surgically obtained specimens may be needed. The two children with HL in our series underwent anterior mediastinotomy and su¬ praclavicular biopsy without difficulty. In evaluating children with mediastinal lymphomas, with or without SVCS, the outstanding improvement in
ries, all children were treated according to modern proto¬
cols. The fact that five of seven patients whose conditions were diagnosed at least 1 year before this report was com¬ pleted were alive and free of disease from 12 months to more than 8 years after diagnosis (mean, 46 months) sug¬ gests that survival could be expected to be similar to that of children with T-cell lymphoma-leukemia in general. We recommend that every effort be made to obtain an accurate diagnosis in children with mediastinal masses and SVCS. When extreme trachéal compression precludes use of general anesthesia and bone marrow or pleural as¬ pirate fails to provide adequate information for diagnosis, management should be directed to the disease most likely to affect these children; in the Middle East, this would be T-cell lymphoma. References 1. Issa PY, Brihi
ER, Janin Y, Slim MS. Superior
childhood: report of ten 71:337-341. 2. Janin Y, Becker
vena cava
ture and
syndrome
cases
vena cava
syndrome
in
and review of the literature. Pediatrics. 1983;
L, Wise K, Schneider D, Schwartz D, So H. Superior in childhood and adolescence: a review of the litera-
report of three
cases.
J Pediatr Surg. 1982;17:290-295.
3. Carabell SC, Goodman RL. Oncologic emergencies: superior vena cava syndrome. In: Devita VT, Hellman S, Rosenberg SA, eds. Cancer. 2nd ed. Philadelphia, Pa: JB Lippincott; 1985:1985-1960. 4. Lange B, D'Angio G, Ross AJ III, O'Neill JA Jr, Packer RJ. Oncologic emergencies: superior vena cava syndrome and superior mediastinal syndrome. In: Pizzo PA, Poplack DG, eds. Pediatric Oncology. Philadelphia, Pa: JB Lippincott; 1989:799-802. 5. Yellin A, Rosen A, Reichert N, Lieberman Y. Superior vena cava syndrome: the myth\p=m-\thefact. Am Rev Respir Dis. 1989;141:1114-1118. 6. Lukes RJ, Butler JJ. The pathology and nomenclature of Hodgkin's disease. Cancer Res. 1966;26:1063-1081. 7. Magrath IT. Lymphocyte differentiation pathways: an essential basis for the comprehension of lymphoid neoplasia. J Natl Cancer Inst. 1981;67:501\x=req-\ 504. 8. Magrath IT. Malignant lymphomas. In: Levine AS, ed. Cancer in the Young. New York, NY: Masson Publishing USA Inc; 1982:473-574. 9. Connors JM, Klimo P. MOPP/ABV hybrid chemotherapy for advanced Hodgkin's disease. Semin Hematol. 1987;24(suppl 1):35-40. 10. Riehm H, Feickert HJ, Schrappe M, Henze G, Schellong G. results in five ALL-BFF studies since 1970: implications of risk factors for
Therapy
prognosis. Haematol Blood Transfusion. 1987;30:139-146. 11. Yellin A, Pak HY, Burke JS, Benfield JR. Surgical management of lymphomas involving the chest. Ann Thorac Surg. 1987;44:363-369. 12. Miller JB, Variakojis D, Bitran JD, et al. Diffuse histiocytic lymphoma with sclerosis: a clinicopathologic entity frequently causing superior vena caval obstruction. Cancer. 1981;47:748-756. 13. King RM, Telander RL, Smithson WA, Banks PM, Han MT. Primary mediastinal tumors in children. J Pediatr Surg. 1982;17:512-520. 14. Pokorny WJ, Sherman JO. Mediastinal masses in infants and children. J Thorac Cardiovasc Surg. 1974;68:869-975. 15. Pate JW, Hamman J. Superior vena cava syndrome due to histoplasmosis in children. Ann Surg. 1965;161:778-787. 16. Gaebler JW, Kleiman MB, Cohen M, etal. Differentiation of lymphoma from histoplasmosis in children with mediastinal masses. J Pediatr. 1984;
survival in the last decade should further prompt clinicians
104:706-709. 17. D'Angio GJ, Mitus A, Evans AE. The superior mediastinal syndrome in children with cancer. AJR Am J Roentgenol. 1965;93:537-544. 18. Bunin NJ, Hvizdala E, Link M, et al: Mediastinal nonlymphoblastic lymphoma in children: a clinicopathologic study. J Clin Oncol. 1986;4:154\x=req-\ 159. 19. Mitchell CD, Gordon I, Chessels JM. Clinical, haematological, and radiological features in T-cell lymphoblastic malignancy in childhood. Clin
with NHL who survived for extended periods usually had only limited disease,13 today, survival without disease of children with T-cell lymphoma19 and nonlymphoblastic lymphoma18 can reach 51% and 74%, respectively. Whether survival in the presence of SVCS was as good is not clear, as this was not reported. Moreover, in the more specific studies, either the follow-up period was too short2 or the management was not based on current protocols.1 The lat¬ ter study could serve only as an historic control. In our se-
1170.
to obtain accurate diagnoses. Whereas before 1980 children
Radiol. 1986;37:257-261. 20. Northrip DR, Bohman BK, Tsueda K. Total airway occlusion and superior vena cava syndrome in a child with an anterior mediastinal tumor. Anesth Analg. 1986;65:1079-1082. 21. Neuman GG, Weingarten AE, Abramowitz RM, Kushins LG, Abramson AL, Ladner W. The anesthetic management of the patient with an anterior mediastinal mass. Anesthesiology. 1984;60:144-147. 22. Halpern S, Chatten J, Meadows AT, Byrd R, Lange B. Anterior mediastinal masses: anesthesia hazards and other problems. J Pediatr. 1983;102: 407-410. 23. Mandell GA, Lentieri R, Goodman LR. Tracheobronchial compression in Hodgkin lymphoma in childhood. AJR Am J Roentgenol. 1982;139:1167\x=req-\
Downloaded From: http://archpedi.jamanetwork.com/ by a Oakland University User on 06/14/2015
Objective.\p=m-\Toevaluate the workup and treatment of lymphoma and superior vena cava syndrome. Design.\p=m-\Aretrospective survey. Setting.\p=m-\Statehospital serving as a secondary and tertiary referral center for pediatric oncology and pediatric \s=b\
children with
cardiac surgery.
Participants.\p=m-\Elevenchildren aged 11 months to 12 diagnosed as having lymphoma or T-cell acute lymphoblastic leukemia who presented with superior vena cava syndrome during an 11-year period. Interventions.\p=m-\Lymphnode biopsy (two patients), thoracenthesis (five patients), bone marrow aspiration (two patients), and thoracenthesis in addition to bone marrow aspiration (two patients). All aspirates were evaluated with immunohistochemical studies. Chemotherapy was the only years
management intervention.
Results.\p=m-\T-celllymphoma or leukemia accounted for nine cases and Hodgkin's disease for two cases. Respiratory symptoms occurred in 10 patients, including tracheal compression in six patients (compression was life-threatening in one patient). Diagnosis of superior vena cava syndrome was achieved in eight patients using surface-marker analysis of aspirates. The syndrome disappeared within 2 to 10 days. Seven of nine children whose conditions were diagnosed more than 1 year before this writing were alive and free of disease after mean follow-up of 37 months. Conclusions.\p=m-\(1)A specific diagnosis can be achieved in most children with superior vena cava syndrome and lymphoma; (2) Thoracic computed tomographic scans are essential, identifying minute pleural effusions that can aid diagnosis; (3) Anesthetic hazard is related only to severe tracheal compression; (4) Chemotherapy achieves excellent symptomatic relief; and (5) Long-term survival, without disease, is achievable.
(AJDC. 1992;146:1060-1063) vena cava syndrome (SVCS), the clinical pre¬ sentation of superior vena cava obstruction or severe reduction in venous return from the head, neck, and up¬ per extremities, is an uncommon condition in adults. In the pediatrie population, it is extremely rare. Issa et al1 and Ja¬ nin and coworkers2 reviewed the worldwide literature
Superior
Accepted for publication January 27, From the
1992.
Departments of Thoracic Surgery (Drs Yellin and Lieberman)
and Hematology (Drs Mandel, Rechavi, Neuman, and Ramot), The Chaim Sheba Medical Center, Tel Hashomer, Israel, and the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel (Drs Yellin, Mandel, Rechavi, Neuman, Ramot, and Lieberman). Reprint requests to Department of Thoracic Surgery, The Chaim Sheba Medical Center, Tel Hashomer 52621, Israel (Dr Yellin).
available until 1979 and found only 175 children and ad¬ olescents with SVCS. Most of the cases were associated with cardiovascular surgery or the introduction of ventriculoatrial shunts or catheterization of the superior vena cava. As in adults,3 the condition is regarded as an onco¬ logie emergency in the pediatrie literature, including the most recent textbooks,4 and radiotherapy is often initiated without an attempt to establish a diagnosis.2 Recently, we showed that these stigmata related to SVCS in adults5 are often based on misconceptions. From that study, we also learned that in Israel the main cause of SVCS in children is lymphoma. Bearing these points in mind, we reviewed our experience with SVCS in children during the last 11 years, focusing on lymphoma, in an attempt to draw conclusions pertinent to the diagnosis and management of this condition. PATIENTS AND METHODS A computerized search of the archives of the Chaim Sheba Medical Center, Tel Hashomer, Israel, was conducted to identify all pediatrie patients with SVCS from 1980 through October 1990. In addition, all outpatient records of pediatrie cases of lymphoma were manually screened for evidence of SVCS. Included in this search were 66 children with lymphoma, equal numbers of whom had Hodgkin's lymphoma (HL) and non-Hodgkin's lymphoma (NHL). We also included 22 patients with T-cell acute lymphoblastic leukemia, each of whom presented with a large mediasti¬ nal mass. The following data were then studied: duration and type of symptoms, age, sex, findings on physical examination, serum level of lactic dehydrogenase, type and duration of treat¬ ment, and results of complete follow-up. In addition, the original chest roentgenograms, computed tomographic scans, and histo¬ logie slides (when present) were reviewed. Patients coded in the archives for SVCS, but whose records showed no clear evidence of the syndrome, were not included. Also excluded were patients whose venous congestion could be attributed to pericardial or cardiac causes.
Diagnosis was based either on histologie appearance (HL) or on results of immunohistochemical studies (NHL). In the latter cases, mononuclear cells from samples obtained by biopsy, bone mar¬ row aspiration, or thoracentesis were separated using Ficoll dis¬ continuous centrifugation. The cells were immunophenotyped using a panel of antibodies defining B, T, and myeloid antigens. Cases were classified according to Ray's modification of Lukes and Butler's nomenclature6 for HL and according to the National Cancer Institute's formulation for NHL.7,8 Patients with HL were treated with hybrid protocol9; patients with T-cell leukemia were treated according to 1983 Berlin-Frankfurt-Munster protocol.1" As a rule, radiotherapy was not part of the initial therapy. RESULTS Eleven patients younger than age 18 years with lym¬ phoma and SVCS were identified (Table). In addition, a 17-year-old girl with B-cell lymphoma was coded for
Downloaded From: http://archpedi.jamanetwork.com/ by a Oakland University User on 06/14/2015
Clinical Data of Children With
Lymphoma and SVCS* Time to
Roentgeno¬ Sex
Diagnosis T-ALL
Year of
(Ethnicity) Age Diagnosis M (A)
5 y
1982
Symptoms and
Signs
Dyspnea
Duration of Symptoms, d
LDH,
60
980
U/L
Pleural Effusion
Source of
Diagnosis Cervical biopsy
graphic
Time to Clinical
Evidence of Res¬
Results
Improve-
olution,
(Time After
ment, d
d
14
material (no
markers),
Diagnosis) Alive and well (3 mo)
bone marrow
T-ALL
F (A)
11
mo
1986
Hoarseness,
60
1234
respiratory distress,
T-cell
M (A)
lymphoma
3 y
1986
stridor Stridor
Died of
Effusion, bone
sepsis
marrow
(21 d) 30
1132
Effusion
16
Alive and well (4 y 3 mo)
12
Alive and well (3 y
(markers),
cervical fine-needle
aspirate T-cell
M(J)
lymphoma T-cell
M (A)
lymphoma,
10 y
1986
8 y
1986
12 y
1989
M (J)
Exertional
Effusion
11 mo)
Effusion
21
Unclear
Unclear
Died of
sepsis
dyspnea
stage III
Lymphoma, most probably
700
Dyspnea, flank pain
1000
Respiratory
Negative
(3 mo) Alive and well (1 y 6 mo)
2
results of bone
distress
T-cell
marrow
studies, clinical
HL, NS, stage IIA
M(A)
HL, NS, stage
M (J)
T-ALL
M(J)
11 y
1989
7 y
1989
7 y
1989
Hoarseness,
diagnosis Supraclavicular biopsy sample
14
exertional
dyspnea
IIIB
Fever,
Results of mediasti-
21
sweating
Dyspnea, cyanosis
Unclear
Unclear
10
45
2
5
2
8
notomy 21
1093
Negative
results of
two
lymph
Alive and well (1 y 5 mo) Alive and well (1 y 3 mo) Alive and well (1 y)
node
biopsies, bone
marrow
T-ALL
M(J)
5 y
1990
T-cell
F(J)
6 y
1990
lymphoma
Dyspnea
14
358
Effusion
Dyspnea, cyanosis,
70
523
Effusion
tachycardia
2
*SVCS indicates superior vena cava syndrome; LDH, lactate dehydrogenase; T-ALL, T-cell acute lymphoblastic leukemia; HL, NS, nodular sclerosing; A, Arab; J, Jewish; ellipses, data not available; minus sign, absent; and plus sign, present.
SVCS, but review of her records revealed no solid data confirming this diagnosis. Therefore, she was excluded from the study. The youngest adult with SVCS was a 22-
We also found five children with SVCS who did not have lymphoma. Three had undergone Mustard operation; one had undergone Glenn operation; and one suffered from seminoma. Our study included nine boys and two girls, with a mean age of 6.8 years. Five were Arab children from the Gaza
year-old man with HL.
10
Alive and well (2 mo) Alive and well (3 mo
wk)
Hodgkin's lymphoma;
Strip. Two patients had HL (with nodular sclerosing) and nine, T-cell lymphoblastic lymphoma or T-cell acute lymphoblastic leukemia. The prevalence of SVCS in patients with NHL (including T-cell acute lymphoblastic leukemia with large mediastinal masses) was 16% compared with 6% in patients with HL. Symptoms, mainly respiratory, were present for a mean of 26.5 days before diagnosis. Levels of lactic dehydrogenase were elevated in all eight patients with NHL in whom levels were determined. Lev-
Downloaded From: http://archpedi.jamanetwork.com/ by a Oakland University User on 06/14/2015
The two patients with HL were alive and in complete remission 15 and 17 months after diagnosis. Two children with T-cell lymphoma died of sepsis, one within 3 weeks of diagnosis, and the other, 3 months after diagnosis. At the time of this report, five patients were alive, three of whom were alive for more than 3 years after diagnosis. Two ad¬ ditional children were alive and well at the time of this re¬ port, but were followed up for very short periods.
overexposed chest roentgenogram shows severe lowertracheal, carinal, and bronchial narrowing. The resulting respiratory symptoms were unrelated to the superior vena cava syndrome per se. Case 7. A plain
els returned to normal, with disappearance of the medias¬ tinal mass. Pleural effusion was seen on roentgenograms of seven patients with NHL and of one patient with HL. The amount was usually small, and in three cases pleural effusion was detected with computed tomographic scans
only.
Diagnosis of all cases of NHL was based on results of immunohistochemical staining, and diagnosis of HL was based on imprints from biopsy material and histologie ap¬ pearance. Before admission to the Sheba Medical Center, Tel Hashomer, Israel, three patients underwent invasive procedures (cervical lymph node biopsy [two patients] and fine-needle aspiration [one patient]). Immunohistochemi¬ cal studies were not performed on the specimens obtained, and no definite diagnosis was determined at that time. Di¬ agnosis at the Sheba Medical Center was established after studying cells obtained by thoracenthesis (five patients), bone marrow aspiration (two patients), or both (one patient). In the sixth patient in our series, the bone marrow study was not diagnostic. Because the parents refused to permit further invasive procedures and the patient had severe trachéal compression, therapy was initiated. The presumed diagnosis of T-cell lymphoma was sustained by his clinical response to therapy. In the two patients with HL, material for diagnosis was procured by anterior mediastinotomy and supraclavicular biopsy. Therapy was usually initiated within 24 to 48 hours. Clinical improvement was prompt, occurring within 2 to 10 days (Table), with complete clearing of SVCS and res¬ piratory symptoms. Complete roentgenographic resolu¬ tion was relatively slower, requiring a mean of 9.7 days in patients with NHL. In one patient with HL, it took 45 days.
COMMENT Although most cases of SVCS in children and adoles¬ cents described in the literature are iatrogenic, our cases of SVCS were not. We encountered no cases of SVCS associ¬ ated with catheters, and detected only four cases of SVCS after cardiovascular surgery. Eleven of the 12 cases asso¬ ciated with malignant conditions were due to lymphomas, and these 11 cases formed the basis for this study. A literature review in 19831 revealed 150 cases of SVCS in childhood. Only 24 (16%) were neoplastic in origin, with 16 occurring with NHL and one occurring with HL. The more frequent association of SVCS with NHL rather than HL in adults was noted by us5" and others12 and is valid in children, too,1·2 as confirmed in the present study. Superior vena cava syndrome was described in 11 (6%) of 188 children with primary mediastinal tumors.13 Pokorny and Sherman14 described 109 children with medias¬ tinal masses, including 24 with lymphoma. Superior vena cava syndrome occurred in three (20%) of 15 children with NHL and in none with HL. The prevalence in our study was slightly lower. Nevertheless, both reports show that SVCS is not uncommon in children with NHL. Uncom¬ monly, SVCS in children may be caused by histoplasmo¬ sis,15 which is a rare disease in the Middle East. The correct diagnosis can usually be established based on results of serologie tests, with more invasive diagnostic procedures
rarely being required.16 D'Angio and coworkers17 used superior mediastinal syn¬
drome to describe
symptoms of trachéal and
venous com¬
pression occurring in children with superior mediastinal masses. They reported nine cases among 607 children with malignant tumors. Eight involved children with lymphoproliferative disorders. From the details given, it seems that only one child had clear-cut SVCS. Seven additional
of SVCS associated with NHL were reported re¬ cently1* among 25 children with mediastinal nonlymphoblastic lymphoma. These lymphomas do not commonly spread systemically, and their diagnosis usually requires lymph node biopsy, mediastinoscopy, or even thoracotomy. Most cases in our series and the series reported by Issa et al1 were cases of T-cell lymphoma, which tends to disseminate early. Mitchell and coworkers'1' described 72 children with malignant T-cell neoplasms, 19 of whom were considered to have lymphoma. Diagnosis, similar to our experience, was based on surface-marker analysis (immunohistochemical stains) of pleural fluid or bone marrow in most cases. The need for surgically obtained cases
specimens was very limited, in contrast to earlier studies.20 It may be of some significance that the two largest studies examining SVCS and lymphoma (the present study and that by Issa et al1) originated in the Middle East. Many authors1,2'417 have stressed the life-threatening
conditions associated with SVCS and advocate early treat¬ ment without definitive diagnosis. Although the danger of anesthesia in children with large mediastinal masses is well established20 22 and most cases of anesthetic catastro-
Downloaded From: http://archpedi.jamanetwork.com/ by a Oakland University User on 06/14/2015
phes were associated with lymphoma, SVCS per se did not cause the life-threatening conditions. Trachéal compres¬ sion is not uncommon in children with lymphoma21 and was observed in 55% of all the pediatrie cases of HL in an¬ other study.23 Trachéal compression was also common in our patients (Figure), occurring in six patients, and was the main cause of the high rate of severe respiratory symp¬ toms. Pleural effusions were detected in eight patients, but were moderate or large in only three. Before 1980, we en¬ countered two children with large mediastinal masses causing severe trachéal compression. One sustained respi¬ ratory arrest during an attempted mediastinoscopy and was thereafter treated empirically. The second patient un¬ derwent successful supraclavicular lymph node biopsy after induction with local anesthesia. In the present study, one child with severe respiratory distress, nondiagnostic bone marrow aspirate, and markedly elevated levels of lactic dehydrogenase was treated empirically. In all these cases, trachéal compression, rather than SVCS, imposed the anesthetic hazard. Because only two children under¬ went procedures after induction with general anesthesia in our medical center, we do not believe that this experience allows us to make conclusions regarding the hazards of anesthesia for patients with large mediastinal masses. We
conclude that children, like adults with SVCS but not trachéal compression, can undergo diagnostic surgery af¬ ter induction with general anesthesia without increased risk of complications and death.516 Therefore, it is crucial that every patient with a large mediastinal mass be eval¬ uated, preferably with computed tomography, for evi¬ dence of trachéal compression. Although many cases of SVCS associated with ante¬ rior and middle mediastinal masses in children prove to be caused by T-cell lymphoma, this is certainly not the rule. Nonlymphoblastic lymphoma,16 HL, neuroblas¬ toma, germ cell tumors, and even benign diseases can also cause superior vena cava obstruction. The manage¬ ment of these conditions is quite different from that of T-cell lymphoma. Therefore, identification of a specific diagnosis is advisable in any case of a mediastinal mass, regardless of the presence of SVCS. Only when severe trachéal compression imposes an immediate respiratory hazard does diagnosis become less important. Whereas diagnosis of all lymphomas in the past almost always can
required a surgical procedure,13 currently, cytochemical staining and surface-marker analysis of needle-obtained specimens can provide information for an accurate di¬ agnosis. In more localized lymphomas, (ie, nonlympho¬ blastic18 or Hodgkin's lymphoma), surgically obtained specimens may be needed. The two children with HL in our series underwent anterior mediastinotomy and su¬ praclavicular biopsy without difficulty. In evaluating children with mediastinal lymphomas, with or without SVCS, the outstanding improvement in
ries, all children were treated according to modern proto¬
cols. The fact that five of seven patients whose conditions were diagnosed at least 1 year before this report was com¬ pleted were alive and free of disease from 12 months to more than 8 years after diagnosis (mean, 46 months) sug¬ gests that survival could be expected to be similar to that of children with T-cell lymphoma-leukemia in general. We recommend that every effort be made to obtain an accurate diagnosis in children with mediastinal masses and SVCS. When extreme trachéal compression precludes use of general anesthesia and bone marrow or pleural as¬ pirate fails to provide adequate information for diagnosis, management should be directed to the disease most likely to affect these children; in the Middle East, this would be T-cell lymphoma. References 1. Issa PY, Brihi
ER, Janin Y, Slim MS. Superior
childhood: report of ten 71:337-341. 2. Janin Y, Becker
vena cava
ture and
syndrome
cases
vena cava
syndrome
in
and review of the literature. Pediatrics. 1983;
L, Wise K, Schneider D, Schwartz D, So H. Superior in childhood and adolescence: a review of the litera-
report of three
cases.
J Pediatr Surg. 1982;17:290-295.
3. Carabell SC, Goodman RL. Oncologic emergencies: superior vena cava syndrome. In: Devita VT, Hellman S, Rosenberg SA, eds. Cancer. 2nd ed. Philadelphia, Pa: JB Lippincott; 1985:1985-1960. 4. Lange B, D'Angio G, Ross AJ III, O'Neill JA Jr, Packer RJ. Oncologic emergencies: superior vena cava syndrome and superior mediastinal syndrome. In: Pizzo PA, Poplack DG, eds. Pediatric Oncology. Philadelphia, Pa: JB Lippincott; 1989:799-802. 5. Yellin A, Rosen A, Reichert N, Lieberman Y. Superior vena cava syndrome: the myth\p=m-\thefact. Am Rev Respir Dis. 1989;141:1114-1118. 6. Lukes RJ, Butler JJ. The pathology and nomenclature of Hodgkin's disease. Cancer Res. 1966;26:1063-1081. 7. Magrath IT. Lymphocyte differentiation pathways: an essential basis for the comprehension of lymphoid neoplasia. J Natl Cancer Inst. 1981;67:501\x=req-\ 504. 8. Magrath IT. Malignant lymphomas. In: Levine AS, ed. Cancer in the Young. New York, NY: Masson Publishing USA Inc; 1982:473-574. 9. Connors JM, Klimo P. MOPP/ABV hybrid chemotherapy for advanced Hodgkin's disease. Semin Hematol. 1987;24(suppl 1):35-40. 10. Riehm H, Feickert HJ, Schrappe M, Henze G, Schellong G. results in five ALL-BFF studies since 1970: implications of risk factors for
Therapy
prognosis. Haematol Blood Transfusion. 1987;30:139-146. 11. Yellin A, Pak HY, Burke JS, Benfield JR. Surgical management of lymphomas involving the chest. Ann Thorac Surg. 1987;44:363-369. 12. Miller JB, Variakojis D, Bitran JD, et al. Diffuse histiocytic lymphoma with sclerosis: a clinicopathologic entity frequently causing superior vena caval obstruction. Cancer. 1981;47:748-756. 13. King RM, Telander RL, Smithson WA, Banks PM, Han MT. Primary mediastinal tumors in children. J Pediatr Surg. 1982;17:512-520. 14. Pokorny WJ, Sherman JO. Mediastinal masses in infants and children. J Thorac Cardiovasc Surg. 1974;68:869-975. 15. Pate JW, Hamman J. Superior vena cava syndrome due to histoplasmosis in children. Ann Surg. 1965;161:778-787. 16. Gaebler JW, Kleiman MB, Cohen M, etal. Differentiation of lymphoma from histoplasmosis in children with mediastinal masses. J Pediatr. 1984;
survival in the last decade should further prompt clinicians
104:706-709. 17. D'Angio GJ, Mitus A, Evans AE. The superior mediastinal syndrome in children with cancer. AJR Am J Roentgenol. 1965;93:537-544. 18. Bunin NJ, Hvizdala E, Link M, et al: Mediastinal nonlymphoblastic lymphoma in children: a clinicopathologic study. J Clin Oncol. 1986;4:154\x=req-\ 159. 19. Mitchell CD, Gordon I, Chessels JM. Clinical, haematological, and radiological features in T-cell lymphoblastic malignancy in childhood. Clin
with NHL who survived for extended periods usually had only limited disease,13 today, survival without disease of children with T-cell lymphoma19 and nonlymphoblastic lymphoma18 can reach 51% and 74%, respectively. Whether survival in the presence of SVCS was as good is not clear, as this was not reported. Moreover, in the more specific studies, either the follow-up period was too short2 or the management was not based on current protocols.1 The lat¬ ter study could serve only as an historic control. In our se-
1170.
to obtain accurate diagnoses. Whereas before 1980 children
Radiol. 1986;37:257-261. 20. Northrip DR, Bohman BK, Tsueda K. Total airway occlusion and superior vena cava syndrome in a child with an anterior mediastinal tumor. Anesth Analg. 1986;65:1079-1082. 21. Neuman GG, Weingarten AE, Abramowitz RM, Kushins LG, Abramson AL, Ladner W. The anesthetic management of the patient with an anterior mediastinal mass. Anesthesiology. 1984;60:144-147. 22. Halpern S, Chatten J, Meadows AT, Byrd R, Lange B. Anterior mediastinal masses: anesthesia hazards and other problems. J Pediatr. 1983;102: 407-410. 23. Mandell GA, Lentieri R, Goodman LR. Tracheobronchial compression in Hodgkin lymphoma in childhood. AJR Am J Roentgenol. 1982;139:1167\x=req-\
Downloaded From: http://archpedi.jamanetwork.com/ by a Oakland University User on 06/14/2015
