Noninvasive evaluation of the swollen extremity: Experiences with 190 lymphoscintigraphic examinations.

Transcription

1 Noninvasive evaluation of the swollen extremity: Experiences with 190 lymphoscintigraphic examinations. Peter Gloviczki, M_D, David Calcagno, MD, Alexander Schirger, MD, Peter C. Pairolero, MD, Kenneth J. Cherry, MD, John W. HaUett, MD, and Heinz W. Wahner, MD, Rochester, Minn. Lymphoscintigraphy (LS), performed with technetium 99m-labeled antimony trisulfide colloid (Cadema Medical Products, Inc., Middletown, N.Y.), was used as a noninvasive diagnostic examination to evaluate the lymphatic c'trcadation in 190 extremities of 115 patients. Forty-six patients had primary lymphedema, 48 had secondary lymphedema, and 21 patients had other causes of limb swelling. To determine the value of LS in surgical decision making, preoperative and postoperative LS of 16 patients who underwent surgical repair of the lymphatic abnormality were studied separately. Semiquantitative evaluation of the lymphatic drainage and visual interpretation of the image patterns were reliable to differentiate lymphedema from edemas of other origin (sensitivity: 92%, specificity: 100%). Although certain image patterns were characteristic of either primary or secondary lymphedema, LS could not consistently differentiate between the two types. Episodes of cellulitis in lymphedema clearly delayed lymph transport. LS was helpful in patient selection and follow-up after lymphatic surgery, but it did not prove patency of lymphovenous anastomoses. It was diagnostic in the evaluation of lymphangiectasia and was used to document successful surgical treatment of reflux of chyle. LS is safe and reliable and has no side effects. It should replace contrast lymphangiography in the routine evaluation of the swollen extremity. (J Vase SURG 1989;9: ) Once a neglected field in medicine, chronic edema of the extremities has received considerable attention in recent years. This can be attributed to improvement in diagnosis, 1 to success of surgical treatment of chronic insufficiency of the deep veins, 2 and to the development of microvascular surgery that has made lymphovenous anastomoses 3-6 and lymph vessel transplantations 7 possible. Although the long-term cffcctivcness of these procedures still remains to be seen, there is a need for a noninvasive, inexpensive, and repeatable diagnostic examination to confirm the involvement of the lymphatic system as a factor in the development of chronic edema of the extremities. Contrast lymphangiography accurately delineates From the Section of Vascular Surgery, the Division of Cardiovascular Disease and Internal Medicine, and the Division of Diagnostic Nuclear Medicine, Mayo Clinic and Mayo Foundation. Supported in part by the 1987 Edwin J. Wylie Traveling Fellowship Award of the Society for Vascular Surgery. Presented at the Twelfth Annual Meeting of the Midwestern Vascular Surgical Society, Rochester, Minn., Sept , Reprint requests: Peter Gloviczki, MD, Section of Vascular Surgery, Mayo Clinic, 200 First St. S.W., Rochester, MN the anatomy of lymph vessels and nodes, but the procedure is invasive, is difficult to repeat, has side effects, s and gives little information on the ftmction of the lymphatic system. Lymphoscintigraphy (LS), first performed with radioactive gold in the 1950s, 8 has improved significantly because of advances in imaging techniques, the introduction of a better radioactive label (technetium 99m), and selection of the optimal size of the labeled particles Our preliminary experience with LS, performed with technetium 99m-labeled antimony trisulfide colloid, has been satisfactory, is The purpose of the present study was to determine the usefulness of LS in the diagnosis of chronic limb swelling, and secondarily its usefulness in preoperative selection and follow-up of patients undergoing surgical treatment for chronic edema of the extremities. PATIENTS Lymphoscintigraphy in diagnosis Between October 1, 1983, and June 30, 1988, 190 extremities in 115 patients were studied with 683

2 684 Gloviczki et al. Journal of VASCULAR SURGERY Table I. Clinical data of 115 patients with swollen extremity Type of edema No. of patients No. of extremities Upper Lower Total Primary lymphedema* Secondary lymphedema Malignant tumor Lymphadenectomy Lymphadenectomy with irradiation Advanced cancer Surgery for benign disease Femoropopliteal graft Trauma Infection Insect bite Other causes of swelling Lipedema Chronic venous insufficiency Idiopathic edema Benign cyclic edema Myxedema Reflex sympathetic dystrophy Chylothorax, chylous ascites Arteriovenous malformation Miscellaneous 3 -- ~ 5 Total * Includes five patients with congenital lymphangiectasia. LS at the Mayo Clinic. There were 86 female and 29 male patients. Mean age was 43.2 years and ranged from 6 to 77 years. Twenty patients had upper extremity limb swelling, and 95 patients had lower extremity limb swelling. Sixty-nine patients had unilateral limb involvement, and 43 patients had bilateral limb involvement. LS was performed on a total of 155 swollen limbs and 35 clinically normal limbs. However, our "normal control group" consisted of 17 patients, only those who had contralateral secondary lymphedema. To establish the role of LS in the diagnosis of the swollen extremity, the 115 patients were divided into three distinct groups: (1) primary lymphedema, (2) secondary lymphedema, and (3) other causes of limb swelling (Table I). Forty-six patients had primary lymphedema. Criteria for inclusion into this group were (1) a lack of definite cause of limb swelling and (2) if the disease process was presumed to be the result of primary abnormality of the lymph conducting elements of the lymph vessels and lymph nodes. 16 Five of the 46 patients in this group had evidence of congenital lymphangiectasia; one of these patients had reflux of chyle through incompetent varicose lymph vessels into the pelvis and the lower extremity. In 48 other patients a definite or highly probable cause of the interruption of the lymphatic pathways could be determined, and these patients were con- sidered to have secondary lymphedema. The largest subgroup was made up of 31 patients with malignant rumor. However, limb swelling in the patients with cancer was usually not the result of the tumor itself but rather the consequence of therapeutic interventions, such as lymphadenectomy, frequently followed by irradiation. Adenocarcinoma of the breast was the most frequent tumor in patients with upper extremity lymphedema (16 of 18 patients), and cervical carcinoma, lymphoma, and adenocarcinoma of the prostate were the most frequent tumors in patients with lower extremity involvement (four patients each). Eight patients had a history of trauma, although direct correlation between the swelling and the injury could not always be proved. The only patient who had infection as the cause of limb swelling had documented filariasis. The third group of 21 patients had limb swelling from causes other than lymphatic disease. The most frequent cause was lipedema followed by chronic venous stasis. Overall, 21 of the 115 patients had evidence of either deep venous incompetence or obstruction or had removal of the greater saphenous vein for varicosity or arterial bypass. Seventeen of these 21 were included in the group of patients with primary or secondary lymphedema, since the main cause of limb swelling was attributed to lymphatic obstruction and not to venous disease. To define the role of LS in making operative

3 Volume 9 Number 5 May 1989 Lymphoscintigraphy, lymphedema, lymphangiectasia 685 decisions, we separately analyzed preoperative and postoperative lymphoscintigraphic studies in 16 of the 115 patients who underwent surgical treatment for the lymphatic abnormality. METHODS Lymphoscinfigraphy was performed as previously described from our laboratory. Four hundred to 500 ~zci of99mtc Sb2S3 (technetium 99m-labeled antimony trisulfide colloid, Cadema Medical Products, Inc., Middletown, N.Y.) was injected in the second interdigital space subcutaneously by means of a tuberculin syringe and 24-gange needle. Total volume injected varied between 0.3 and 0.6 ml. In addition to our previously described procedure, is we now exercise the injected limbs immediately after the injection by asking the patient to repeatedly squeeze a rubber ball or to perform toe stands for 3 minutes. This maneuver has significantly shortened the arrival time of the isotope to the regional lymph nodes and has resulted in more reproducible studies. After injection, the patient was placed in a supine position under the gamma camera with the groin or axilla positioned in the field of vision. Twelve-minute images were then obtained for the first hour. At 1 and 3 hours additional 5-minute images were obtained of the trunk and the affected extremity. If desired for better anatomic detail, other images in posterior or lateral positions and delayed images at 6 to 24 hours were recorded. Informed consent was obtained from all the patients. The radiopharmaceutical, 99mTcSb2S3, is all investigational drug and is restricted by the Food and Drug Administration for investigational use. Evaluation of lymphoscintigrams Semiquantitative evaluation of the lymphatic drainage and description of thc image patterns were performed from the timed images done in each test. Computer-assisted quantitation of radioactivity above the injection site, above major lymph nodes, and in the liver were performed in our previously reported study, is but results were inconsistent because of significant variability of the counts. For this reason, direct quantitation was subsequently discontinued. With semiquantitative LS, normal and delayed lymph transport cotfld be determined. Appearance time of the isotope in major lymph nodes (axillary and inguinal) combined with the degree of visualization of the lymph nodes (clearly visualized, faintly visualized, not seen) provided information on the degree of delay in lymph transport. In all patients Fig. 1. Normal lymphoscintigram. with normal transport, lymph nodes were clearly visualized at 1 hour after injection; the transport was mildly delayed if lymph nodes were only faintly seen at 1 hour and moderately delayed if lymph nodes were not visualized until 3 or 6 hours after injection. The transport was severely delayed if lymph nodes were not visualized at all. Visual interpretation of the image patterns revealed a distinct normal pattern and several abnormal patterns. The normal lymphoscintigraphic pattern of the superficial lymphatic system of the lower extremities (Fig. 1) shows lymph channels along the medial aspect of the leg and lymph nodes in the groin, pelvis, and paraaortic region at 1 hour. The liver is also generally well outlined at I hour. At 3 hours, activity in the pelvic and abdominal lymph nodes and in the liver is more pronounced. There is usually no activity seen in the chest. In patients with lymphedema, seven abnormal patterns of the lymph drainage could be seen (Figs. 2,A through C, 3, and 4,A): (1) no removal of the isotope from the injection site; (2) dermal backflow pattern consistent with interstitial dermal lymph transport and found if main lymph channels were obstructed or nonfunctioning; (3) large collateral lymph channels; (4) mixed pattern (2 and 3); (5) crossover filling of major nodes caused by proximal obstruction and collateral circulation; (6) extravasation of the isotope into lymphatic fistula or lymphocele; and (7) lymphangiectasia.

4 686 Gloviczki et al. Journal of VASCULAR SURGERY Fig. 2. Abnormal patterns of lymphoscintigram in lymphedema. A, No removal of the isotope from the injection site; B, large lymphatic collaterals at the calf; C, dermal pattern. RESULTS Lymphoscintigraphy in diagnosis All 17 extremities from the normal control group had a similar normal pattern and had no delay in lymphatic transport (specificity: 100%). One patient with contralateral primary lymphedema had moderate delay in lymph transport; the possibility of latent edema could not be excluded. The degree in delay in lymph transport, observed in 54 extremities with primary lymphedema and in 52 extremities with secondary lymphedema, is summarized on Table II. There was no significant difference in lymph transport between primary and secondary lymphedema; however, certain image patterns were characteristic of one or the other type: Congenital lymphangiectasia could clearly be demonstrated by LS, which showed wide lymph channels in areas where lymph nodes are not available. Reflux of the labeled particles into the edematous extremity could also be visualized (Fig. 4, A). On the other hand, lymphatic fistula or lymphocele is characteristic of extravasation of lymph caused by surgical trauma to the lymph vessel (Fig. 3) that could be shown in both patients with edema after femoropopliteal bypass grafting. History of previous cellulitis (n = 25) significandy (p < 0.05) delayed lymph transport in the af- Fig. 3. Extravasation of the colloid at the right knee 3 hours after injection confirms the lymphocele in a patient with edema after femoropoplitcal bypass graft, fected extremity when compared to those who did not have previous episodes of infection (n = 79). The LS was normal in patients with lipedema, myxedema, reflex sympathetic dystrophy, or idiopathic or "benign cyclic" edema. The four patients with chronic venous insufficiency had a normal lymphoscintigraphic pattern, although one of the patients had early intensive uptake in the inguinal lymph nodes, which suggested faster lymph transport. Ten other patients who had evidence of chronic venous incompetence or obstruction also had abnormal lymphatic patterns and delayed transport, and were diagnosed as having edema of mixed venous and lymphatic cause. Four of these patients had in- " guinal or axillary lymph node dissection, and in addition to secondary lymphedema, they also had occlusion of the axiuary or iliofemoral veins. The other six patients had primary lymphedema and venous stasis. The thoracic duct could bc visualized in only one patient in the 190 LS studies. This patient had congenital lymphangiectasia (Fig. 4, B). All studies were performed with good patient tolerance and without any complications. The value of LS in operative decisions LS provided useful information in those 12 patients who underwent lymphovenous anastomosis because of chronic lymphedema, and results in 10

5 Volume 9 Number 5 May 1989 Lymphoscintigraphy, lymphedema, lymphangiectasia 687 Table II. Lymphatic transport in 54 extremities with primary lymphedema and 52 extremities with secondary lymphedema Type and location of lymphedema Normal Lymphatic transport Mild delay Moderate delay Severe delay Total no. of limbs Primary lymphedema Upper extremity (4%) 2 (4%) Lower extremity 7(13%) 19(35%) 13(24%) 13(24%) 52(96%) Total 7(13%) 19(35%) 13(24%) 15(28%) 54(100%) Secondary lymphedema Upper extremity 2(4%) 5(10%) 7(13%) 4(8%) 18(35%) Lower extremity -- 17(33%) 11(21%.~ 6(11%) 34(65%) Total 2(4%) 22(42%) 18(35%) 10(19%) 52(100%) patients have been previously published. 6 In several patients lymph channels that were visualized on LS were marked on the skin preoperatively, and were found during the operation and used for lymphovenous anastomoses. In four patients who had no lymph channels visualized preoperatively, patent channels at exploration could be found in only one. However, LS in one patient failed to visualize two major lymph channels that were present on contrast lymphangiogram. At subsequent exploration the lymph vessels still contained the contrast material that was injected into the leg lymphatics 4 months previously. Although lymphovenous anastomoses with these lymphatics were performed, this patient did not benefit from the procedure. As previously reported, 6 postoperative LS gives only indirect evidence oflymphovenous anastomosis patency (visualization of lymph channels). In one patient who underwent suprapubic lymph vessel crossover transposition, postoperative LS failed to show patency of the lymphatic grafts. In the two patients who underwent anastomosis of thoracic duct to azygos vein because of acquired chylothorax, contrast lymphangiography was necessary preoperatively to delineate exact anatomy. However, in another patient with chylothorax and congenital lymphangiectasia, preoperative studies showed mediastinal uptake to be higher than in the liver, and a pleural sample showed counts four times higher than in serum. A 16-year-old female patient with congenital lymphangiectasia had left lower extremity edema and drainage of chyle from vesicles of the skin and of the vulva. Reflux of chyle through incompetent dilated lymph vessels from the mesentery to the pelvis and to the left lower extremity was confirmed by preoperative LS. The test was performed by injecting the isotope into the clinically normal right foot (Fig. 4, A). Studies at 3 and 4 hours showed increased activity over the pelvis and reflux of the isotope into the left thigh through the incompetent lymph vessels. These findings were subsequently confirmed by contrast lymphangiography. This patient underwent resection of the large varicose pelvic lymph channels and had two lymphovenous anastomoses performed between the lumbar veins and two large paraaortic lymph vessels. Postoperative LS showed no evidence of residual reflux (Fig. 4, B). Six months after the operation the patient has no significant chylous drainage and no evidence of progression of her edema. DISCUSSION Evaluation and correct management of the swollen extremity remains a challenge. Clinical examination is the first approach and is most valuable to exclude systemic causes such as cardiac, renal, hepatic, nutritional, or endocrine causes of edemas. Taking patient history and physical examination are frequently successful to diagnose he two most frequent causes of regional edema, chronic venous insufficiency, and lymphedcma. Secondary lymphedema is more frequent than primary lymphedema, and its correct evaluation is of utmost importance, since in this study either malignancy or the surgical or radiologic treatment of a malignant minor was directly responsible for 94% of upper cxtremity secondary lymphedemas and 52% of lower extremity secondary lymphedemas. Whereas computed tomography scanning is routinely used in our practice to cxctude underlying malignancy, LS has become the standard test to confirm the diagnosis of lymphedema. 17 Visual interpretation of the image pattern together with estimation of appearance time of the colloid in the regional lymph nodes provides reliable information as to whether the lymphatic system is normal or abnormal (specificity, 100%; sensitivity,

6 688 Gloviczki et al. Journal of VASCULAR SURGERY Fig. 4. A, Lymphoscintigraphy of a 16-year-old girl with lyrnphangiectasia. Injection of the isotope into the right foot reveals reflux into the pelvis at 3 hours and into the left lower extremity at 4 hours. B, Postoperative result with no evidence of reflux at 4 hours after injection. 92%). These results correspond with those reported by others) 2'~* In our experience quantitation of the lymph clearance gave us inconsistent results. 15 However, other investigators reported reliable quantitative delay in lymph transport Uptake of the isotope depends on several factors independent of the lymphatic system of the extremity. These factors include the depth of injection of the labeled colloid, the inadvertent injection of the colloid into a small vein, and muscular exercise after injection, which is most important. As previously repotted by Weissleder et al.,19 we found that muscular exercise increases measured values of radioactivity as much as 10 times. Consequently, we believe that unless muscular exercise is strictly standardized, ~9 quantitative studies are of limited value. Based on estimation of delay in lymph drainage and visual interpretation of drainage patterns, we could not differentiate primary lymphedema from secondary lymphedema. However, in some patients drainage patterns were characteristic of primary lymphedema (lymphangiectasia) or secondary postsurgical lymphedema (lymphocele). In patients with primary lympedema there was a trend toward an increase in delay of lymph transport as compared to that in patients with longer duration oflymphedema. The three patients with congenital primary lymphedema had no uptake of the isotope from the injection site, and all had severe lymphedema. A similar correlation between the severity of lymphedema and the delay in lymph clearance has also been shown by other authors) 2 Our data also confirmed significant delay of lymph transport in those patients who had previous episodes of cellulitis. Patients with edema as a result of Venous insufficiency had normal or, in one case, an enhanced lymph transport, which is similar to the findings reported by Stewart et al. 12 We also identified a definite group of patients who came for treatment with mixed, lymphatic, and venous edema. LS was able to differentiate lymphedema from lipedema, reflex sympathetic dystrophy, or other idiopathic edemas. LS is helpful for preoperative selection of patients who may benefit from lymphovenous anastomoses; however, patency of the anastomoses cannot be proved postoperatively with this method. 15 LS is also essential to show normal pattern before removal of lymph vessels for cross-femoral transposition. Unfortunately, in contrast to Baumeister et al.,7 we could not show patency of the transposed lymphatic grafts in our patient postoperatively.

7 Volume 9 Number 5 May 1989 Lymphoscintigraphy, lymphedema, lymphangiectasia 689 During operation for lymphedema, we found lymph vessels that contained oily contrast material injected into the leg 4 months previously for lymphangiography. Although we acknowledge the value and necessity of contrast lymphangiography in selected cases (lymphangiectasia, visualization of the thoracic duct), this latter finding supports the opinion of those who state that in lymphedema contrast lymphangiography may destroy functioning lymphatics. 2 Our young patient with congenital lymphangiectasia proves the value of LS in diagnosis. Lymphangiectasia detected with LS has been previously reported, n but according to our knowledge, this is the first published case where reflux into the pelvis and left lower extremity was documented with LS and where the early result of successful surgical treatment was shown on postoperative LS. In conclusion, our data support the usefulness of LS in the evaluation of the chronically swollen extremity. The examination has no known contraindications, and the diagnosis of lymphedema can be made with an overall accuracy of approximatcly 93 %. In addition, LS allows differentiation of purely venous edemas from mixed or lymphatic edemas, and it can detcct lymphatic abnormalities such as lymphocele or lymphangiectasia with reflux. Quantitative studies are possible but appear to be of limited usefulness. The test is helpful in paticnt sclection and in follow-up after repair, but it does not prove patency of lyrnphovenous anastomoses. Finally, whereas contrast lymphangiography is still needed for selective evaluations, we recommend that LS replace lymphangiography in the routine evaluation of patients with chronic edema of the extremity. The authors thank Alex E. Schabauer for his help in chart reviews and Denise Colwell for manuscript preparation. REFERENCES 1. Kismer RL. Diagnosis of chronic venous insufficiency. 1 VASC SURG 1986;3: Bergan JJ, Yao JST, Flinn WR, McCarthy WJ. Surgical treatment of venous obstruction and insufficiency. J VASC SURG 1986;3: O'Brien BM, Shafiroff BB. Microlymphaticovenous and resectional surgery in obstructive lymphedema. World J Surg 1979;3: Huang G-K, Hu R-Q, Liu Z-Z, Shen Y-L, Lan T-D, Pan G-P. Microlymphaticovenous anastomosis in the treatment of lower limb obstructive lymphedema: analysis of 91 cases. Plast Reconstr Surg 1985;76: Gloviczki P, Hollier LH, Nora FE, Kaye MP. The natural history of microsurgical lymphovenous anastomoses: an experimental study. J VAsc SURG 1986;4: Gloviczki P, Fisher J, HoUier LH, Pairolero PC, Schirger A, Wahner HW. Microsurgical lymphovenous anastomosis for treatment of lymphedema: a critical review. J VASC SVR6 1988;7: Baumeister RG, Siuda S, Bohmert H, Moser E. A microsurgical method for reconstruction of interrupted lymphatic pathways: autologous lymph-vessel transplantation for treatment of lymphedemas. Scand J Plast Reconstr Surg 1986; 20: Sage HH, Gozum BV. Lymphatic scintigrams: a method for studying the fimctional pattern of lymphatics and lymph nodes. Cancer 1958;11; Jackson FI, Bowen P, Lentle BC. Scintilymphangiography with 99mTc-antimony sulfide colloid in hereditary lymphedema (Nonne-Milroy disease). Clin Nud Med 1978;3: Ege GN, Warbick A. Lymphoscintigraphy: a comparison of ~rfcm antimony sulphide colloid and ~fcm stannous phytate. Br J Radiol i979;52: Sty JR, Starshak RJ. Atlas of pediatric radionuclide lymphography. Clin Nucl Med 1982;7: Stewart G, Gaunt JI, Croft DN, Browse NL. Isotope lymphography: a new method of investigating the role of the lymphatics in chronic limb oedema. Br J Surg 1985;72: Nawaz K, Hamad MM, Sadek S, Awdeh M, EklofB, Abdel- Dayem HM. Dynamic lymph flow imaging in lymphedema. Normal and abnormal patterns. Clin Nud Med 1986;11: Ohtake E, Matsui K. Lymphoscintigraphy in patients with lymphedema. A new approach using intradermal injections of technetium-99m human serum albumin. Clin Nucl Med 1986;11: Vaqueiro M, Gloviczki P, Fisher J. Hollier LH, Schirger A, Wahner HW. Lymphoscintigraphy in lymphedema: an aid to nficrosurgery. J Nucl Med 1986;27: Browse NL, Stewart G. Lymphoedema: pathophysiology and classification. J Cardiovasc Surg (Torino) 1985;26: Gloviczki P, Schirger A. Lymphedema. In: Spittell JA Jr, ed. Clinical medicine. Philadelphia: Harper & Row, 1985:1-10. i8. Kleinhans E, Baumeister RGH, Hahn D, Siuda S, Bull U, Moser E. Evaluation of transport kinetics in lymphoscintigraphy: follow-up study in patients with transplanted lymphatic vessels. Eur J Nucl Med 1985;10: Weissleder H, Weissleder R. Lymphedema: evaluation of qualitative and quantitative lymphoscintigraphy in 238 patients. Radiology 1988;167: Casley-Smith JR, Foldi M, Ryan TJ, et al. Lymphedema. Summary of the 10th International Congress of Lymphology working group discussions and recommendations. Adelaide, Australia, August 10-17, Lymphology 1985;18:

8 690 Gloviczki et al. Journal of VASCULAR SURGERY DISCUSSION Dr. Walter McCarthy (Chicago, Ill.). This interesting presentation has brought to our attention an important diagnostic modality that may replace conventional lymphangiograms. As a word of follow-up, the girl treated for the lymph fistulas was seen at Northwestern. She recently returned for follow-up and is very happy with the surgical result she obtained. I have two questions. Is this technique quantitative enough to follow the natural history of this condition over time? What sort of special equipment is required to establish this type of imaging at our hospitals? Dr. Gloviczki. Lymphoscintigraphy is a semiquantitative study, and our specificity and sensitivity results are based on the visual interpretation of the image patterns and on the time it takes for the isotope to appear in major nodes, such as those in the axilla. We did not find that lymph clearance measurements were accurate; neither were they necessary, The isotope itself presently is under restriction by the Food and Drug Administration, but it can be obtained by request. A single kit costs $30, and the entire study, about $300. You also need a gamma camera and an enthusiastic nuclear physicist who believes in the test. Dr. Donald Silver (Columbia, Mo.). Although this very nice study suggests that lymphoscintigraphy may help with the diagnosis and follow-up of patients with lymphedema, I have several questions. Before we accept any new test, it must be compared with established ones. Since your presentation did not include any comparisons between lymphoscintigraphy and standard lymphangiography, perhaps you could clarify this issue for us. Second, does this approach distinguish between hypoplasia and hyperplasia as the source for primary lymphedema? And third, your follow-up period of 4 months is a little short. As you know, almost anything we do for lymphedema provides shortterm relief, but several years are required to evaluate the procedure. Dr. Jeffrey R. Rubin (Cleveland, Ohio). We have done similar work using technetium-labeled serum albumin and an assimilation counter with a pencil probe to quantify flow after lymphaticovenous anastomosis. Why did you choose your type of component tube to perform lymphoscintigraphy? With regard to quantification, it is imperative to measure and quantify flow after performing these procedures. We were not able to document flow very well with the visual demonstrations on our pictures, and that is why we used the quantification techniques. Why have you not tried this simple assimilation counter with a pencil probe? Dr. Gloviczki. Mr. Norman Browse, who works in London at the same hospital in which lymphangiography was developed, has published a study comparing lymphoscintigraphy with lymphangiography and found good correlation between the two. Only 11 patients in our series underwent contrast lymphangiography and lymphoscintigraphy. We performed lymphoscintigraphy 4 months after lymphangiography in one patient and failed to show the lymph channels that were seen on the original contrast study. We operated on this patient 4 months after contrast medium had been injected and found that the lymphatics still were filled with dye. So there may be some truth in the axiom, "if you see a fianctional lymphangiogram, it may be for the last time." In general, we found that the correlation was quite good between the results of lymphangiography and lymphoscintigraphy in our 11 patients. LymphoscintigraphL ~ was especially helpful in excluding other sources of edema, such as in patients who are under consideration for deep venous reconstruction. We cannot predict hypoplasia or hyperplasia in primary lymphedema, but I do not know whether it is really necessary to do so. The use of human serum albumin is a faster technique than ours, but we do not have any experience with it. Dr. Krishna Jain (Kalamazoo, Mich.). Please tell us the indications for this test. For example, should it be recommended to the patient who has a swollen leg in conjunction with negative venous study results? We usually have managed this problem with compressive stockings and have not obtained lymphangiography because lymphovenous reconstruction is so uncommon to begin with. What should we be doing in community hospitals? Dr. Gloviezki (closing). In many cases, the distinction between venous edema and lymphedema may be made on clinical grounds. However, others require objective documentation. If results of a lymphoscintigraph are positive, then that patient has lymphedema. If results are positive and the noninvasive venous study results also are abnormal, that patient may have mixed edema. If results are negative, the possibility of true lymphedema may be omitted entirely. In that situation another diagnosis must be considered, such as lipedema or benign cyclic edema. In the adult population, a computed tomographic scan may be necessary to exclude the possibility of pelvic malignancy.