Redefinition of Cutaneous Lymphatic Drainage With the Use of Lymphoscintigraphy for Malignant Melanoma James Norman, MD, C. Wayne Cruse, Robert Clark, MD, Hussain Saba,
MD, Carmen Espinosa, MD, MD, PhD, Karen Wells, MD,
Lymphoscintigraphy was performed on 82 patients with melanoma registered at the University Melanoma Clinic. From these data, precise lymphatic drainage basins could he drawn for the head, neck, shoulder, and trunk. These drawings differed significantly from the classic anatomic studies, providing a functional look at the cutaneous lymphatic drainage. This new method correlates much better with clinical experiences and demonstrates much larger areas of ambiguous drainage than previously reported. Data from the lymphoscintigrams also emphasize the individuality of cutaneous lymphatic flow. The implications of these data in planning elective node dissections for intermediate thickness melanomas are obvious, since it is estimated that up to 59% of the dissections for trunk and head and neck primary melanomas may he misdirected if based on classic anatomic studies. The data indicate that all patients with head, neck, and shoulder lesions should undergo lymphoscintigraphy to define possible drainage basins at risk for metastatic disease. Similarly, truncal lesions require scintigrams except when they are within four well-defined areas with an extremely low probability of ambiguous drainage. Lesions in these areas show very reliable and predictable drainage to a single nodal group.
From the Departments of Surgery (JN, CWC, CC, KW, DR), Patholo gy (CE), Radiology (CB, RC), and Medicine (HS), H. Lee Moflitt Cancer Center, University of South Florida, Tampa, Florida. Requests for reprints should be addressed to Douglas Reintgen, MD, H. Lee Moftitt Cancer Center, University of South Florida, P.O. Box 280179, Tampa, Florida 33682-0179. Manuscript submitted July 12, 1990, and accepted in revised form November 12, 1990.
432
THE AMERICAN JOURNAL OF SURGERY
Charles Cox, MD, Claudia Berman, Douglas Reintgen, MD, Tampa, Florida
MD,
he understanding of cutaneous lymphatic drainT age remains grounded in the work performed by Sappey in 1874. By injecting mercury into the skin of [I]
cadavers, he was able to show lines of demarcation running along the midline and from just above the umbilicus curving slowly backward to the second lumbar vertebra. Lesions above Sappey’s line were thought to drain to the ipsilateral axilla and those below to the ipsilateral groin. Minor revisions were later made by Haagensen et al [2] and Sugarbaker and McBride [3], who showed that this general pattern held true except for areas 2.5 cm on either side of the midline and on either side of Sappey’s line. Melanomas in these areas showed ambiguous lymphatic drainage that was unpredictable. These guidelines represented our view of dermal lymphatic anatomy until recently. Lymphoscintigraphy was introduced in 1953 by Sherman and Ter-Pogossian [4] to study cutaneous lymphatic flow. The technique has been extensively investigated in many clinical situations by numerous authors [5-211 and has since been shown to be highly reliable and reproducible. The common denominator in each study to date has been the unsuspected drainage frequently encountered that contradicts Sappey’s anatomic drawings. In addition, there is a known predilection for cutaneous tumors such as melanoma to metastasize to unexpected nodal groups, suggesting that new guidelines for cutaneous lymphatic drainage are needed. Although it is still controversial whether elective node dissection should be performed as part of the primary treatment for malignant melanoma, if it is performed, then the functional drainage for a particular primary skin site should direct the dissection. This report describes the use of lymphoscintigraphy in an attempt to map the lymphatic drainage of 82 consecutive patients with melanoma who were candidates for elective node dissection [22]. The data were combined with the drainage patterns of primary melanomas previously reported in the literature so that drawings of the ambiguous lymphatic drainage areas of the body could be prepared. These could then be compared with classic anatomic teachings. Likewise, areas of straightforward lymphatic drainage are defined so that elective lymph node dissections may proceed without the use of a preop erative lymphatic map. PATIENTS
AND METHODS
Lymphoscintigraphy was performed on 82 consecutive patients with malignant melanoma at the University of South Florida. All patients had clinical stage I or II
VOLUME 162 NOVEMBER 1991
TABLE I Lymphosclntigraphy for Mallgnant Melanoma
Reference
Patients Reported
institution
[51
0
11
A.B
New Mexico Michigan NYU Medical Center Miami
21 50 23 18 20 71 4 24 25 56
4 0 5 0 21 3 2 63 0 0 0 24
Denmark South Florida Mayo Clinic South Florida
53 17 22 65
5 2 0 1
A,B B
Alabama UCLA Duke Frankfurt, FRG Duke Virginia
[t81 [181 [201 [211 A = unclear
Reason
32
161 I71 181 [81 [lOI [ItI [I21 1131 [I41 v51 [t61 v71
lymphatic
Patients Rejected
documentation
flow demonstrated;
of primary C = extremfty
10
location;
B = technical
melanomas
excluded;
c D B B A.B
D
B failure.
No
D = 5.0cm-
wide local excision.
tumors of the head, neck, shoulder, and trunk and underwent the scan prior to elective lymph node dissection. Lymphoscintigraphy was performed using technetiumantimony trisullide as previously described [19]. Briefly, a mean dose of 0.94 mCi (range: 0.3 to 2.0 mCi) was injected intradermally in four equal parts circumferentially around the scar of the biopsy site. If the patient had a previous wide local excision and primary closure, the location of the primary melanoma was estimated and four injections were placed around this site. Imaging was performed using a large field-of-view gamma camera set at a 20% window and fitted with a low-energy, generalpurpose, parallel hole collimator. Anatomic landmarks were marked with a “hot” marker, and images of the head and neck, chest, abdomen, and pelvis were obtained. A minimum of 100,000 counts were obtained for each view and recorded on film. Scans of the primary site were made immediately upon injection, with all other views being performed at l/2, 2, and 4 hours after injection. Previous reports of the use of lymphoscintigraphy as a determinant of cutaneous lymphatic flow were retrospectively reviewed. Table I lists the various reports in which sufficient detail was provided such that the exact primary site of the melanoma and all the lymphatic basins positive by scan could be determined. These data were used to generate the new drainage guidelines for cutaneous lymphatic flow. The literature contains reference to 532 patients with primary melanoma who had undergone lymphoscintigraphy. A total of 123 patients were rejected due to inaccurate or unclear documentation of the primary tumor’s location, technical failure due to no drainage from the injection site, or the inclusion of extremity melanomas that have straightforward lymphatic drainage. Those who underwent scanning after wide local excision that
Fipe~ 1. Prlrnarymelanoma sitesthat show left axllkry &&age asallorpartofthekefferentlymphatlcflow.Thisincludsslesions that cross the midline as well as melanomas that are below Sappeyb line.
required skin grafting for closure were also excluded since previously published work suggests that interruption of subdermal lymphatics at the time of such a wide local excision may give unreliable or no drainage from the primary site. Any lesion flowing to several nodal basins is represented on all appropriate diagrams; thus, a single lesion may be depicted up to four times if lymphoscintigraphy showed it to drain to four distinct regional basins. Rather than assume symmetry, all primary sites and the subsequent lymphatic flow were plotted for both sides of the body as further assurance of accuracy in the final definition of the drainage patterns. RESULTS Figure 1 shows all the primary sites that flow to the left axillary lymphatic basin. All primary melanomas from the anterior and posterior trunk that have the left axilla as all or part of their skin lymphatic drainage have been plotted. The shaded area extends 11 cm to the right of the midline and well below Sappey’s line both anteriorly and posteriorly. It also includes primary lesions in the left supraclavicular area, the left posterior cervical triangle, and the shoulder. The shaded areas in Figure 2 depict the distribution of primary melanomas that show drainage to the right axilla. Again, the lesions may be up to 11 cm to the left of the midline, as well as below Sappey’s line, and continue to drain to the right axilla. It is also apparent that some melanomas over the anterior and posterior cervical triangle have the right axilla as part of their cutaneous lymphatic drainage.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 162 NOVEMBER 1991
433
NORMANETAL
Figure 2.Cutaneous primariesthat chainto the rightaxiiia. Of note is the lymphaticflow from skin sites up to 11 cm on the opposite side of the midline.
Figure 3. Shadedarea depictsthe primarymelanomasthat drain to the left groin. This cutaneous area includes lesions above Sappey’sline as well as lesionsfrom well acrossthe midline.The vertical lined area illustratesrneianomas that have cutaneous flow to the left anteriorcervical basin.
Figure 3 demonstrates the cutaneous lesions that flowed to the left inguinal nodes and the left anterior cervical chain. Melanomas up to 11 cm to the right of the midline anteriorly and 5 cm to the right of the midline posteriorly drained to the left groin nodal basin. Similarly, drainage to the anterior cervical chain came from lesions well onto the shoulder, face, and even the posterior neck. The shaded areas in Figure 4 illustrate drainage to the right inguinal nodes. Again, contralateral inguinal drainage is seen up to 11 cm from the midline anteriorly and almost 6 cm posteriorly. Additionally, those lesions draining to the right posterior cervical chain are depicted by the vertical lines, and those draining to the left supraclavicular basin are shown by horizontal lines. The posterior cervical chain receives lymphatic flow from the skin well down onto the back, as well as the top of the shoulder and across the midline of the face. The supraclavicular nodes receive flow from the face, neck, shoulder, and superior chest. A comparison made of Sappey’s anatomic drawings (Figure 5, top) and the functional cutaneous lymphatic flow as defined by the lymphoscintigrams (Figure 5, bottom) depicts areas of ambiguous lymphatic drainage patterns for cutaneous lesions derived from superimposing Figures 1 through 4. Of note is that the area of unpredictable lymphatic drainage is much more extensive than originally thought. Instead of finding straightforward lymphatic drainage patterns from primary melanoma sites 2.5 cm off the midline or Sappey’s line, one does not
encounter predictable drainage until the lesion is 11 cm on either side of the midline or Sappey’s line. It is also apparent that the shoulder skin sites and the entire head and neck areas have totally unpredictable drainage patterns and are included in the shaded areas. The symmetry apparent in these drawings suggests that the functional scintigrams are valid, since symmetry was not assumed, yet the left and right drainage basins have nearly identical boundaries. It is also apparent from Figure 5 that there are four well-defined areas of the trunk in which lymphatic flow appears to be 100% predictable and lymphoscintigraphy does not add to the preoperative work-up of the melanoma patients in whom node dissections are planned (unshaded areas). An analysis was performed between the preoperative clinical prediction of lymphatic flow of the attending surgeon and the cutaneous drainage demonstrated with lymphoscintigraphy. In 48 of the 82 primary sites (59%) dissections based on accepted anatomic guidelines would not have removed all the lymphatic basins at risk for harboring occult metastatic disease or a lymphatic basin removed would not have shown functional drainage with the scintigram and presumably would not have been at risk for micrometastatic disease. Inevitably, the clinical prediction underestimated the possible paths of lymphatic flow from the primary site. The 82 patients had a mean follow-up of 2.1 years, a time period in which 75% of the recurrences are expected to recur. Only one patient has developed metastatic dis-
434
162
THE AMERICAN
JOURNAL
OF SURGERY
VOLUME
NOVEMBER
1991
Flglm4.shadedareail~possibleprimerysltesthatcan havedrainage~otherlghtinnodes.Verticellinedareashows possible prhwks that cm have tymphatic flow to the right
and neck @nary sites have tobdy unpdictable lymphaticdrainage*
ease in a lymphatic basin that was not positive by scan. This patient had a primary melanoma excised from her midline upper chest 2 years prior to presentation to the Melanoma Clinic. Recently, she developed a large intransit lesion just to the right of her primary site, as well as right axillary adenopathy. A lymphoscintigram around the primary midline site revealed drainage only to the left axilla. It was hypothesized that the large intransit metastasis just to the right of the primary site blocked lymphatic flow in this direction and the right axilla did not take up the tracer. Therapeutic node dissection of the right axilla confirmed metastatic disease and the only false-negative scan. COMMENTS By injecting mercury into the skin of cadavers, Sappey [I] defined lymphatic anatomy for the trunk and demonstrated a line of demarcation that now bears his name. Above this line, which circles the mid-abdomen from just above the umbilicus to the second lumbar vertebra posteriorly, cutaneous lesions drain to the ipsilateral axilla. Below, they drain to the ipsilateral inguinal nodes. Subsequently, other authors [2,3] have shown that areas of ambiguous drainage exist 2.5 cm on either side of the midline and Sappey’s line. By injecting radiocolloid gold intradermally, Sherman and Ter-Pogossian [4] introduced the technique of lymphoscintigraphy in 1953. Using gamma cameras, they were able to show migration of the radioactive partiTHE AMERICAN
F@re 5. Top, !hppey’s classic anatomic description of the lymphaticflow as c-iymptlaticfkiw.~mappedbylym@[email protected]+ b@ws lymphatic flow coqmred with the anatomic drawings andtheifwclusionoftheentheheadandneckareaandshoulderin unpredictableprimarysites.
cles through the subdermal lymphatics to draining lymph nodes. Fee and associates [5] reported their use of the technique in predicting the metastatic possibilities of malignant melanoma in 1978. Since that time, numerous studies have appeared in the literature confirming the reliability and accuracy of lymphoscintigraphy in predicting lymphatic drainage in individual patients [6-211. Several institutions, including our own, are now directing surgical therapy based on these scans. The advantage of lymphoscintigraphy over classic anatomic guidelines is the scan’s reflection of functional and not just structural anatomy. Sensitivity is greatly enhanced over gross or even microscopic anatomic examination because of its in uiuo use and the small size of the particles that allow early migration into the interstitial
JOURNAL
OF SURGERY
VOLUME
162
NOVEMBER
1991
435
space and lymphatics. False negative results, defined as metastatic lesions appearing in basins not visualized by scan, are extremely rare. In our series of 82 patients, there has been only one false-negative result. This low rate permits lymphoscintigraphy to be used to reliably predict the possible lymphatic routes of metastatic spread. The one false-negative scan, presumably caused by the disruption of normal cutaneous flow by a large in-transit lesion, emphasizes the need for lymphoscintigraphy as early as possible during the treatment of melanoma. If the test is performed prior to the wide excision of the primary, there will be little disruption of draining lymphatics. Prior to this report, the largest published series consisted of 7 1 patients having lymphoscintigrams of lesions located on the head, neck, trunk, and extremities, with most studies having fewer than 25 patients. It has been difficult, therefore, to draw conclusions about new guidelines for lymphatic drainage even though it was quite evident that many patients had scans that did not follow accepted anatomic standards. With the addition of our 82 patients to the 450 patients who have already been described in the literature, the population size is sufficiently large such that accurate anatomic drawings can be made. The overlap of basins draining the head, neck, and shoulders is so extensive that no prediction of which nodal group drains a particular lesion can be made by anatomic location alone. The typical head and neck lesion drains to three separate nodal groups (range: 1 to 4) and is unpredictable more than 60% of the time. Similarly, more than 50% of shoulder lesions have unexpected drainage patterns and most commonly drain to two or three distinct nodal basins [19]. Primary melanomas located on the trunk show an interesting pattern of ambiguous drainage that is greatly expanded over Sappey’s watershed areas. These drawings do resemble Sappey’s work by having a line of ambiguous drainage down the midline and around the waist. However, a primary skin site does not drain in a straightforward fashion until it is more than 11.Ocm off the midline or Sappey’s line. It is important to note that the discrete areas centered about each axilla and lateral hip that do not show tendencies toward ambiguous or bilateral drainage are very well defined. All lesions located in these areas drain to a single nodal basin, that being the adjacent axillary or inguinal nodes, and they do not require lymphoscintigraphy prior to elective lymph node dissection. There were no lesions located within these areas that drained anywhere except the single adjacent nodal basin. It is interesting to hypothesize that these are lesions that lend themselves to an incontinuity node dissection at the time of local excision. As expected, the closer a lesion is to the midline or Sappey’s line, the higher the probability for drainage to more than one nodal basin. This was apparent for lesions on the midline that drained bilaterally 89% of the time. There were many lesions with bilateral flow that were located more than 2.5 cm from the midline but none were further than 11 cm. Two areas were identified: one anterior near the sternal notch and one posterior at the base of the neck. In these two areas, drainage is possible to six separate nodal 436
THE AMERICAN JOURNAL OF SURGERY
basins: bilateral anterior and posterior cervical basins, and bilateral axillary basins. No lesions were identified, however, that drained to more than four discrete basins. Whether elective lymph node dissection should be performed as part of the primary treatment for malignant melanoma is still controversial. Two prospective randomized trials [23,24] have shown no increase in disease-free or ultimate survival in patients who had elective lymph node dissection with wide local excision of the primary melanoma. However, there are three retrospective studies, each based on large databases [22,25,26], that imply an increase in actuarial survival in patients with intermediate thickness melanoma (tumor thickness more than 0.76 mm and less than 4.0 mm) who have had elective lymph node dissection with the wide excision. Two prospective randomized trials to evaluate the efficacy of elective lymph node dissection are in progress. The World Health Organization trial (Veronesi l-J, unpublished data) is currently evaluating the role of elective lymph node dissection in patients with trunk primary melanomas, since the organization’s original study [23] investigated only distal extremity lesions. Another intergroup trial (Balch C, unpublished data) is investigating the role of elective lymph node dissection in patients with head and neck melanomas whose tumors are more than 1.0 mm thick. This particular trial requires the preoperative use of lymphoscintigraphy in areas of ambiguous lymphatic drainage to plan the node dissection so that the basins at risk for micrometastases are removed. The results of both of these trials should settle the question of the effectiveness of elective lymph node dissection for clinical stage I melanoma. The individuality of cutaneous lymphatic drainage is emphasized in the current study since the majority of lesions will fall into these newly defined areas of unpredictable drainage. It is apparent that many elective lymph node dissections performed on the head, neck, shoulders, and trunk may be misdirected since classic anatomic guidelines cannot predict their lymphatic drainage and therefore potential metastatic spread. To eliminate the possibility of misdirected lymph node dissection, lymphoscintigraphy is suggested for these patients unless the primary tumor is located on an extremity or within one of the four discrete areas of isolated single basin drainage. The authors provide a succinct review of the current state-of-the-art of lymphoscintigraphy. Their study of 82 consecutive patients with stage I melanoma demonstrates ambiguous patterns of lymphatic drainage, especially for primary melanomas of the trunk and head and neck. REFERENCES 1. Sappey MPC. Anatomie, physiologie, pathologie, des vaisseaux lymphatiques consideres chez I’homme et les vertebres. Paris 1874, A. DeLahaye and E. Lecrosnier. 2. Haagensen CD, Feind CR, Hester FP, et al. Lymphatics of the trunk. In: Haagensen CD, editor. The lymphatics in cancer. Philadelphia: WB Saunders, 1972: 441-51.
VOLUME 162 NOVEMBER 1991
3. Sugarbaker EV, McBride CM. Melanoma of the trunk: the results of surgical excision and anatomic guidelines for predicting nodal metastasis. Surgery 1976; 80: 22-30. 4. Sherman A, Ter-Pogosaian M. Lymph node concentration of radioactive colloidal gold following interstitial injection. Cancer 1953; 6: 123840. 5. Fee HJ, Robinson DS, Sample WF, Graham LS, Holmes EC, Morton DL. The determination of lymph shed by colloidal gold scanning in patients with malignant melanoma: a preliminary study. Surgery 1978; 84: 626-32. 6. Meyer CM, Lechletner ML, Logic JR, Balch CE, Bessey PQ, Tauxe WM. Technetium-99 sulfur-colloid cutaneous lymphoscintigraphy in the management of truncal melanoma. Radiology 1979; 131: 205-9. 7. Rees WV, Robinson DS, Holmes EC, et al. Altered lymphatic drainage following lymphadenectomy. Cancer 1980; 45: 3045-9. 8. Sullivan DC, Croker BP Jr, Harris CC, er al. Lymphoscintigraphy in malignant melanoma: 99-m Tc antimony sulfur colloid. AJR Am J Roentgen01 1981; 137: 847-51. 9. Munz DC, Altmeyer P, Sessler MJ, et 01. Axillary lymph node groups-the center in lymphatic drainage from the tnmkal skin in man. Lymphology 1982; 15: 143-7. 10. Reintgen DS, Sullivan D, Coleman E, er al. Lymphoscintigraphy for malignant melanoma-surgical considerations. Am Surg 1983; 49: 672-8. 11. Wanebo HJ, Harpole D, Testes CD. Radionuclide lymphoscintigraphy with technetium 99m antimony sulfide colloid to identify lymphatic drainage of cutaneous melanoma at ambiguous sites in the head and neck and trunk. Cancer 1985; 55: 1403-13. 12. Bergvist L, Strand S, Hafstrom L, et al. Lymphoscintigraphy in patients with malignant melanoma: a quantitative and qualitative evaluation of its usefulness. Eur J Nucl Med 1984; 9: 129-35. 13. Logic JR, Balch CM. Defining lymphatic drainage patterns with cutaneous lymphoscintigraphy. In: Balch CM, Milton WG, editors. Cutaneous melanoma. Philadelphia: Lippincott, 1985: 159-70. 14. Doss LL, Padilla RS, Hladik WB. Technetium antimony sulfur colloid scintiscan in selective lymph node dissection for malignant
melanoma. J Dermatol Surg Oncol 1986; 12: 1280-4. 15. Eberbach MA, Wahl RL, Argenta LC, et al. Utility of Iymphoscintigraphy in directing surgical therapy for melanomas of the head, neck, and upper thorax. Surgery 1987; 102: 433-9. 16. Kramer EL, Sanger JJ, Golomb F, et al. The impact of intradermal lymphoscintigraphy on surgical management of clinical stage I truncal malignant melanoma. J Dermatol Surg Gncol 1987; 13: 508-15. 17. Eberbach MA, Wahl RL. Lymphatic anatomy: functional nodal basins. Ann Plast Surg 1989; 22: 25-31. 18. Lock-Andersen J, Rossing N, Drzewiecki KT. Preoperative cutaneous lymphoscintigraphy in malignant melanoma. Cancer 1989; 63: 77-82. 19. Norman JG, Cruse CW, Ruas E, et al. The expanding role of lymphtintigraphy in the management of cutaneous melanoma. Am Surg 1989; 12: 689-94. 20. Woods JE, Freedman AM, Brown ML. Lymphoscintigraphy as a guide to treatment in malignant melanoma. Ann Plast Surg 1989; 22: 150-5. 21. Berman CG, Norman J, Cruse CW, Reintgen DS, Clark R. Lymphoscintigraphy in malignant melanoma. Ann Plast Surg. In press. 22. Reintgen DS, Cox EB, McCarthy KS, Vollmer R, Seigler HF. Efficacy of elective lymph node dissection in patients with intermediate thickness melanoma. Ann Surg 1983; 198: 379-85. 23. Veronesi U, Adamus J, et al. Inefficacy of immediate node d§ion in stage I melanoma of the limbs. N Engl J Med 1977; 297: 627-30. 24. Sim FH, Taylor WF, Ivins JC, et al. A prospective randomized study of the efficacy of routine elective lymphadenectomy in the management of malignant melanoma. Cancer 1978; 41: 448-56. 25. Urist MM, Balch CM, Soong SJ, et 01. Head and neck melanoma in 536 clinical stage I patients: a prognostic factor analysis and results of surgical treatment. Ann Surg 1984; 200: 769-75. 26. Milton GW, Shaw HM, McCarthy WH, et al. Prophylactic lymph node dissection in dinicai stage I cutaneous malignant melanoma: results of surgical treatment in 1,319 patients. Br J Surg 1982; 69: 108-l 1.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 162 NOVEMBER 1991
437
MD, Carmen Espinosa, MD, MD, PhD, Karen Wells, MD,
Lymphoscintigraphy was performed on 82 patients with melanoma registered at the University Melanoma Clinic. From these data, precise lymphatic drainage basins could he drawn for the head, neck, shoulder, and trunk. These drawings differed significantly from the classic anatomic studies, providing a functional look at the cutaneous lymphatic drainage. This new method correlates much better with clinical experiences and demonstrates much larger areas of ambiguous drainage than previously reported. Data from the lymphoscintigrams also emphasize the individuality of cutaneous lymphatic flow. The implications of these data in planning elective node dissections for intermediate thickness melanomas are obvious, since it is estimated that up to 59% of the dissections for trunk and head and neck primary melanomas may he misdirected if based on classic anatomic studies. The data indicate that all patients with head, neck, and shoulder lesions should undergo lymphoscintigraphy to define possible drainage basins at risk for metastatic disease. Similarly, truncal lesions require scintigrams except when they are within four well-defined areas with an extremely low probability of ambiguous drainage. Lesions in these areas show very reliable and predictable drainage to a single nodal group.
From the Departments of Surgery (JN, CWC, CC, KW, DR), Patholo gy (CE), Radiology (CB, RC), and Medicine (HS), H. Lee Moflitt Cancer Center, University of South Florida, Tampa, Florida. Requests for reprints should be addressed to Douglas Reintgen, MD, H. Lee Moftitt Cancer Center, University of South Florida, P.O. Box 280179, Tampa, Florida 33682-0179. Manuscript submitted July 12, 1990, and accepted in revised form November 12, 1990.
432
THE AMERICAN JOURNAL OF SURGERY
Charles Cox, MD, Claudia Berman, Douglas Reintgen, MD, Tampa, Florida
MD,
he understanding of cutaneous lymphatic drainT age remains grounded in the work performed by Sappey in 1874. By injecting mercury into the skin of [I]
cadavers, he was able to show lines of demarcation running along the midline and from just above the umbilicus curving slowly backward to the second lumbar vertebra. Lesions above Sappey’s line were thought to drain to the ipsilateral axilla and those below to the ipsilateral groin. Minor revisions were later made by Haagensen et al [2] and Sugarbaker and McBride [3], who showed that this general pattern held true except for areas 2.5 cm on either side of the midline and on either side of Sappey’s line. Melanomas in these areas showed ambiguous lymphatic drainage that was unpredictable. These guidelines represented our view of dermal lymphatic anatomy until recently. Lymphoscintigraphy was introduced in 1953 by Sherman and Ter-Pogossian [4] to study cutaneous lymphatic flow. The technique has been extensively investigated in many clinical situations by numerous authors [5-211 and has since been shown to be highly reliable and reproducible. The common denominator in each study to date has been the unsuspected drainage frequently encountered that contradicts Sappey’s anatomic drawings. In addition, there is a known predilection for cutaneous tumors such as melanoma to metastasize to unexpected nodal groups, suggesting that new guidelines for cutaneous lymphatic drainage are needed. Although it is still controversial whether elective node dissection should be performed as part of the primary treatment for malignant melanoma, if it is performed, then the functional drainage for a particular primary skin site should direct the dissection. This report describes the use of lymphoscintigraphy in an attempt to map the lymphatic drainage of 82 consecutive patients with melanoma who were candidates for elective node dissection [22]. The data were combined with the drainage patterns of primary melanomas previously reported in the literature so that drawings of the ambiguous lymphatic drainage areas of the body could be prepared. These could then be compared with classic anatomic teachings. Likewise, areas of straightforward lymphatic drainage are defined so that elective lymph node dissections may proceed without the use of a preop erative lymphatic map. PATIENTS
AND METHODS
Lymphoscintigraphy was performed on 82 consecutive patients with malignant melanoma at the University of South Florida. All patients had clinical stage I or II
VOLUME 162 NOVEMBER 1991
TABLE I Lymphosclntigraphy for Mallgnant Melanoma
Reference
Patients Reported
institution
[51
0
11
A.B
New Mexico Michigan NYU Medical Center Miami
21 50 23 18 20 71 4 24 25 56
4 0 5 0 21 3 2 63 0 0 0 24
Denmark South Florida Mayo Clinic South Florida
53 17 22 65
5 2 0 1
A,B B
Alabama UCLA Duke Frankfurt, FRG Duke Virginia
[t81 [181 [201 [211 A = unclear
Reason
32
161 I71 181 [81 [lOI [ItI [I21 1131 [I41 v51 [t61 v71
lymphatic
Patients Rejected
documentation
flow demonstrated;
of primary C = extremfty
10
location;
B = technical
melanomas
excluded;
c D B B A.B
D
B failure.
No
D = 5.0cm-
wide local excision.
tumors of the head, neck, shoulder, and trunk and underwent the scan prior to elective lymph node dissection. Lymphoscintigraphy was performed using technetiumantimony trisullide as previously described [19]. Briefly, a mean dose of 0.94 mCi (range: 0.3 to 2.0 mCi) was injected intradermally in four equal parts circumferentially around the scar of the biopsy site. If the patient had a previous wide local excision and primary closure, the location of the primary melanoma was estimated and four injections were placed around this site. Imaging was performed using a large field-of-view gamma camera set at a 20% window and fitted with a low-energy, generalpurpose, parallel hole collimator. Anatomic landmarks were marked with a “hot” marker, and images of the head and neck, chest, abdomen, and pelvis were obtained. A minimum of 100,000 counts were obtained for each view and recorded on film. Scans of the primary site were made immediately upon injection, with all other views being performed at l/2, 2, and 4 hours after injection. Previous reports of the use of lymphoscintigraphy as a determinant of cutaneous lymphatic flow were retrospectively reviewed. Table I lists the various reports in which sufficient detail was provided such that the exact primary site of the melanoma and all the lymphatic basins positive by scan could be determined. These data were used to generate the new drainage guidelines for cutaneous lymphatic flow. The literature contains reference to 532 patients with primary melanoma who had undergone lymphoscintigraphy. A total of 123 patients were rejected due to inaccurate or unclear documentation of the primary tumor’s location, technical failure due to no drainage from the injection site, or the inclusion of extremity melanomas that have straightforward lymphatic drainage. Those who underwent scanning after wide local excision that
Fipe~ 1. Prlrnarymelanoma sitesthat show left axllkry &&age asallorpartofthekefferentlymphatlcflow.Thisincludsslesions that cross the midline as well as melanomas that are below Sappeyb line.
required skin grafting for closure were also excluded since previously published work suggests that interruption of subdermal lymphatics at the time of such a wide local excision may give unreliable or no drainage from the primary site. Any lesion flowing to several nodal basins is represented on all appropriate diagrams; thus, a single lesion may be depicted up to four times if lymphoscintigraphy showed it to drain to four distinct regional basins. Rather than assume symmetry, all primary sites and the subsequent lymphatic flow were plotted for both sides of the body as further assurance of accuracy in the final definition of the drainage patterns. RESULTS Figure 1 shows all the primary sites that flow to the left axillary lymphatic basin. All primary melanomas from the anterior and posterior trunk that have the left axilla as all or part of their skin lymphatic drainage have been plotted. The shaded area extends 11 cm to the right of the midline and well below Sappey’s line both anteriorly and posteriorly. It also includes primary lesions in the left supraclavicular area, the left posterior cervical triangle, and the shoulder. The shaded areas in Figure 2 depict the distribution of primary melanomas that show drainage to the right axilla. Again, the lesions may be up to 11 cm to the left of the midline, as well as below Sappey’s line, and continue to drain to the right axilla. It is also apparent that some melanomas over the anterior and posterior cervical triangle have the right axilla as part of their cutaneous lymphatic drainage.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 162 NOVEMBER 1991
433
NORMANETAL
Figure 2.Cutaneous primariesthat chainto the rightaxiiia. Of note is the lymphaticflow from skin sites up to 11 cm on the opposite side of the midline.
Figure 3. Shadedarea depictsthe primarymelanomasthat drain to the left groin. This cutaneous area includes lesions above Sappey’sline as well as lesionsfrom well acrossthe midline.The vertical lined area illustratesrneianomas that have cutaneous flow to the left anteriorcervical basin.
Figure 3 demonstrates the cutaneous lesions that flowed to the left inguinal nodes and the left anterior cervical chain. Melanomas up to 11 cm to the right of the midline anteriorly and 5 cm to the right of the midline posteriorly drained to the left groin nodal basin. Similarly, drainage to the anterior cervical chain came from lesions well onto the shoulder, face, and even the posterior neck. The shaded areas in Figure 4 illustrate drainage to the right inguinal nodes. Again, contralateral inguinal drainage is seen up to 11 cm from the midline anteriorly and almost 6 cm posteriorly. Additionally, those lesions draining to the right posterior cervical chain are depicted by the vertical lines, and those draining to the left supraclavicular basin are shown by horizontal lines. The posterior cervical chain receives lymphatic flow from the skin well down onto the back, as well as the top of the shoulder and across the midline of the face. The supraclavicular nodes receive flow from the face, neck, shoulder, and superior chest. A comparison made of Sappey’s anatomic drawings (Figure 5, top) and the functional cutaneous lymphatic flow as defined by the lymphoscintigrams (Figure 5, bottom) depicts areas of ambiguous lymphatic drainage patterns for cutaneous lesions derived from superimposing Figures 1 through 4. Of note is that the area of unpredictable lymphatic drainage is much more extensive than originally thought. Instead of finding straightforward lymphatic drainage patterns from primary melanoma sites 2.5 cm off the midline or Sappey’s line, one does not
encounter predictable drainage until the lesion is 11 cm on either side of the midline or Sappey’s line. It is also apparent that the shoulder skin sites and the entire head and neck areas have totally unpredictable drainage patterns and are included in the shaded areas. The symmetry apparent in these drawings suggests that the functional scintigrams are valid, since symmetry was not assumed, yet the left and right drainage basins have nearly identical boundaries. It is also apparent from Figure 5 that there are four well-defined areas of the trunk in which lymphatic flow appears to be 100% predictable and lymphoscintigraphy does not add to the preoperative work-up of the melanoma patients in whom node dissections are planned (unshaded areas). An analysis was performed between the preoperative clinical prediction of lymphatic flow of the attending surgeon and the cutaneous drainage demonstrated with lymphoscintigraphy. In 48 of the 82 primary sites (59%) dissections based on accepted anatomic guidelines would not have removed all the lymphatic basins at risk for harboring occult metastatic disease or a lymphatic basin removed would not have shown functional drainage with the scintigram and presumably would not have been at risk for micrometastatic disease. Inevitably, the clinical prediction underestimated the possible paths of lymphatic flow from the primary site. The 82 patients had a mean follow-up of 2.1 years, a time period in which 75% of the recurrences are expected to recur. Only one patient has developed metastatic dis-
434
162
THE AMERICAN
JOURNAL
OF SURGERY
VOLUME
NOVEMBER
1991
Flglm4.shadedareail~possibleprimerysltesthatcan havedrainage~otherlghtinnodes.Verticellinedareashows possible prhwks that cm have tymphatic flow to the right
and neck @nary sites have tobdy unpdictable lymphaticdrainage*
ease in a lymphatic basin that was not positive by scan. This patient had a primary melanoma excised from her midline upper chest 2 years prior to presentation to the Melanoma Clinic. Recently, she developed a large intransit lesion just to the right of her primary site, as well as right axillary adenopathy. A lymphoscintigram around the primary midline site revealed drainage only to the left axilla. It was hypothesized that the large intransit metastasis just to the right of the primary site blocked lymphatic flow in this direction and the right axilla did not take up the tracer. Therapeutic node dissection of the right axilla confirmed metastatic disease and the only false-negative scan. COMMENTS By injecting mercury into the skin of cadavers, Sappey [I] defined lymphatic anatomy for the trunk and demonstrated a line of demarcation that now bears his name. Above this line, which circles the mid-abdomen from just above the umbilicus to the second lumbar vertebra posteriorly, cutaneous lesions drain to the ipsilateral axilla. Below, they drain to the ipsilateral inguinal nodes. Subsequently, other authors [2,3] have shown that areas of ambiguous drainage exist 2.5 cm on either side of the midline and Sappey’s line. By injecting radiocolloid gold intradermally, Sherman and Ter-Pogossian [4] introduced the technique of lymphoscintigraphy in 1953. Using gamma cameras, they were able to show migration of the radioactive partiTHE AMERICAN
F@re 5. Top, !hppey’s classic anatomic description of the lymphaticflow as c-iymptlaticfkiw.~mappedbylym@[email protected]+ b@ws lymphatic flow coqmred with the anatomic drawings andtheifwclusionoftheentheheadandneckareaandshoulderin unpredictableprimarysites.
cles through the subdermal lymphatics to draining lymph nodes. Fee and associates [5] reported their use of the technique in predicting the metastatic possibilities of malignant melanoma in 1978. Since that time, numerous studies have appeared in the literature confirming the reliability and accuracy of lymphoscintigraphy in predicting lymphatic drainage in individual patients [6-211. Several institutions, including our own, are now directing surgical therapy based on these scans. The advantage of lymphoscintigraphy over classic anatomic guidelines is the scan’s reflection of functional and not just structural anatomy. Sensitivity is greatly enhanced over gross or even microscopic anatomic examination because of its in uiuo use and the small size of the particles that allow early migration into the interstitial
JOURNAL
OF SURGERY
VOLUME
162
NOVEMBER
1991
435
space and lymphatics. False negative results, defined as metastatic lesions appearing in basins not visualized by scan, are extremely rare. In our series of 82 patients, there has been only one false-negative result. This low rate permits lymphoscintigraphy to be used to reliably predict the possible lymphatic routes of metastatic spread. The one false-negative scan, presumably caused by the disruption of normal cutaneous flow by a large in-transit lesion, emphasizes the need for lymphoscintigraphy as early as possible during the treatment of melanoma. If the test is performed prior to the wide excision of the primary, there will be little disruption of draining lymphatics. Prior to this report, the largest published series consisted of 7 1 patients having lymphoscintigrams of lesions located on the head, neck, trunk, and extremities, with most studies having fewer than 25 patients. It has been difficult, therefore, to draw conclusions about new guidelines for lymphatic drainage even though it was quite evident that many patients had scans that did not follow accepted anatomic standards. With the addition of our 82 patients to the 450 patients who have already been described in the literature, the population size is sufficiently large such that accurate anatomic drawings can be made. The overlap of basins draining the head, neck, and shoulders is so extensive that no prediction of which nodal group drains a particular lesion can be made by anatomic location alone. The typical head and neck lesion drains to three separate nodal groups (range: 1 to 4) and is unpredictable more than 60% of the time. Similarly, more than 50% of shoulder lesions have unexpected drainage patterns and most commonly drain to two or three distinct nodal basins [19]. Primary melanomas located on the trunk show an interesting pattern of ambiguous drainage that is greatly expanded over Sappey’s watershed areas. These drawings do resemble Sappey’s work by having a line of ambiguous drainage down the midline and around the waist. However, a primary skin site does not drain in a straightforward fashion until it is more than 11.Ocm off the midline or Sappey’s line. It is important to note that the discrete areas centered about each axilla and lateral hip that do not show tendencies toward ambiguous or bilateral drainage are very well defined. All lesions located in these areas drain to a single nodal basin, that being the adjacent axillary or inguinal nodes, and they do not require lymphoscintigraphy prior to elective lymph node dissection. There were no lesions located within these areas that drained anywhere except the single adjacent nodal basin. It is interesting to hypothesize that these are lesions that lend themselves to an incontinuity node dissection at the time of local excision. As expected, the closer a lesion is to the midline or Sappey’s line, the higher the probability for drainage to more than one nodal basin. This was apparent for lesions on the midline that drained bilaterally 89% of the time. There were many lesions with bilateral flow that were located more than 2.5 cm from the midline but none were further than 11 cm. Two areas were identified: one anterior near the sternal notch and one posterior at the base of the neck. In these two areas, drainage is possible to six separate nodal 436
THE AMERICAN JOURNAL OF SURGERY
basins: bilateral anterior and posterior cervical basins, and bilateral axillary basins. No lesions were identified, however, that drained to more than four discrete basins. Whether elective lymph node dissection should be performed as part of the primary treatment for malignant melanoma is still controversial. Two prospective randomized trials [23,24] have shown no increase in disease-free or ultimate survival in patients who had elective lymph node dissection with wide local excision of the primary melanoma. However, there are three retrospective studies, each based on large databases [22,25,26], that imply an increase in actuarial survival in patients with intermediate thickness melanoma (tumor thickness more than 0.76 mm and less than 4.0 mm) who have had elective lymph node dissection with the wide excision. Two prospective randomized trials to evaluate the efficacy of elective lymph node dissection are in progress. The World Health Organization trial (Veronesi l-J, unpublished data) is currently evaluating the role of elective lymph node dissection in patients with trunk primary melanomas, since the organization’s original study [23] investigated only distal extremity lesions. Another intergroup trial (Balch C, unpublished data) is investigating the role of elective lymph node dissection in patients with head and neck melanomas whose tumors are more than 1.0 mm thick. This particular trial requires the preoperative use of lymphoscintigraphy in areas of ambiguous lymphatic drainage to plan the node dissection so that the basins at risk for micrometastases are removed. The results of both of these trials should settle the question of the effectiveness of elective lymph node dissection for clinical stage I melanoma. The individuality of cutaneous lymphatic drainage is emphasized in the current study since the majority of lesions will fall into these newly defined areas of unpredictable drainage. It is apparent that many elective lymph node dissections performed on the head, neck, shoulders, and trunk may be misdirected since classic anatomic guidelines cannot predict their lymphatic drainage and therefore potential metastatic spread. To eliminate the possibility of misdirected lymph node dissection, lymphoscintigraphy is suggested for these patients unless the primary tumor is located on an extremity or within one of the four discrete areas of isolated single basin drainage. The authors provide a succinct review of the current state-of-the-art of lymphoscintigraphy. Their study of 82 consecutive patients with stage I melanoma demonstrates ambiguous patterns of lymphatic drainage, especially for primary melanomas of the trunk and head and neck. REFERENCES 1. Sappey MPC. Anatomie, physiologie, pathologie, des vaisseaux lymphatiques consideres chez I’homme et les vertebres. Paris 1874, A. DeLahaye and E. Lecrosnier. 2. Haagensen CD, Feind CR, Hester FP, et al. Lymphatics of the trunk. In: Haagensen CD, editor. The lymphatics in cancer. Philadelphia: WB Saunders, 1972: 441-51.
VOLUME 162 NOVEMBER 1991
3. Sugarbaker EV, McBride CM. Melanoma of the trunk: the results of surgical excision and anatomic guidelines for predicting nodal metastasis. Surgery 1976; 80: 22-30. 4. Sherman A, Ter-Pogosaian M. Lymph node concentration of radioactive colloidal gold following interstitial injection. Cancer 1953; 6: 123840. 5. Fee HJ, Robinson DS, Sample WF, Graham LS, Holmes EC, Morton DL. The determination of lymph shed by colloidal gold scanning in patients with malignant melanoma: a preliminary study. Surgery 1978; 84: 626-32. 6. Meyer CM, Lechletner ML, Logic JR, Balch CE, Bessey PQ, Tauxe WM. Technetium-99 sulfur-colloid cutaneous lymphoscintigraphy in the management of truncal melanoma. Radiology 1979; 131: 205-9. 7. Rees WV, Robinson DS, Holmes EC, et al. Altered lymphatic drainage following lymphadenectomy. Cancer 1980; 45: 3045-9. 8. Sullivan DC, Croker BP Jr, Harris CC, er al. Lymphoscintigraphy in malignant melanoma: 99-m Tc antimony sulfur colloid. AJR Am J Roentgen01 1981; 137: 847-51. 9. Munz DC, Altmeyer P, Sessler MJ, et 01. Axillary lymph node groups-the center in lymphatic drainage from the tnmkal skin in man. Lymphology 1982; 15: 143-7. 10. Reintgen DS, Sullivan D, Coleman E, er al. Lymphoscintigraphy for malignant melanoma-surgical considerations. Am Surg 1983; 49: 672-8. 11. Wanebo HJ, Harpole D, Testes CD. Radionuclide lymphoscintigraphy with technetium 99m antimony sulfide colloid to identify lymphatic drainage of cutaneous melanoma at ambiguous sites in the head and neck and trunk. Cancer 1985; 55: 1403-13. 12. Bergvist L, Strand S, Hafstrom L, et al. Lymphoscintigraphy in patients with malignant melanoma: a quantitative and qualitative evaluation of its usefulness. Eur J Nucl Med 1984; 9: 129-35. 13. Logic JR, Balch CM. Defining lymphatic drainage patterns with cutaneous lymphoscintigraphy. In: Balch CM, Milton WG, editors. Cutaneous melanoma. Philadelphia: Lippincott, 1985: 159-70. 14. Doss LL, Padilla RS, Hladik WB. Technetium antimony sulfur colloid scintiscan in selective lymph node dissection for malignant
melanoma. J Dermatol Surg Oncol 1986; 12: 1280-4. 15. Eberbach MA, Wahl RL, Argenta LC, et al. Utility of Iymphoscintigraphy in directing surgical therapy for melanomas of the head, neck, and upper thorax. Surgery 1987; 102: 433-9. 16. Kramer EL, Sanger JJ, Golomb F, et al. The impact of intradermal lymphoscintigraphy on surgical management of clinical stage I truncal malignant melanoma. J Dermatol Surg Gncol 1987; 13: 508-15. 17. Eberbach MA, Wahl RL. Lymphatic anatomy: functional nodal basins. Ann Plast Surg 1989; 22: 25-31. 18. Lock-Andersen J, Rossing N, Drzewiecki KT. Preoperative cutaneous lymphoscintigraphy in malignant melanoma. Cancer 1989; 63: 77-82. 19. Norman JG, Cruse CW, Ruas E, et al. The expanding role of lymphtintigraphy in the management of cutaneous melanoma. Am Surg 1989; 12: 689-94. 20. Woods JE, Freedman AM, Brown ML. Lymphoscintigraphy as a guide to treatment in malignant melanoma. Ann Plast Surg 1989; 22: 150-5. 21. Berman CG, Norman J, Cruse CW, Reintgen DS, Clark R. Lymphoscintigraphy in malignant melanoma. Ann Plast Surg. In press. 22. Reintgen DS, Cox EB, McCarthy KS, Vollmer R, Seigler HF. Efficacy of elective lymph node dissection in patients with intermediate thickness melanoma. Ann Surg 1983; 198: 379-85. 23. Veronesi U, Adamus J, et al. Inefficacy of immediate node d§ion in stage I melanoma of the limbs. N Engl J Med 1977; 297: 627-30. 24. Sim FH, Taylor WF, Ivins JC, et al. A prospective randomized study of the efficacy of routine elective lymphadenectomy in the management of malignant melanoma. Cancer 1978; 41: 448-56. 25. Urist MM, Balch CM, Soong SJ, et 01. Head and neck melanoma in 536 clinical stage I patients: a prognostic factor analysis and results of surgical treatment. Ann Surg 1984; 200: 769-75. 26. Milton GW, Shaw HM, McCarthy WH, et al. Prophylactic lymph node dissection in dinicai stage I cutaneous malignant melanoma: results of surgical treatment in 1,319 patients. Br J Surg 1982; 69: 108-l 1.
THE AMERICAN JOURNAL OF SURGERY
VOLUME 162 NOVEMBER 1991
437
