Journal of Plastic, Reconstructive & Aesthetic Surgery (2013) 66, 201e208

p-Shaped lymphaticovenular anastomosis for headand neck lymphoedema: A preliminary study

Benoit Ayestaray a,b,*, Farid Bekara a, Jean-Baptiste Andreoletti b

aDepartment of Plastic and Reconstructive Surgery, Nimes University Hospital, pl Pr Robert Debre, 30000 Nimes, FrancebDepartment of Plastic and Reconstructive Surgery, Breast Institute, 15, av Jean Jaures, 90000 Belfort, France

Received 1 June 2012; accepted 28 August 2012

KEYWORDSLymphoedema;Head;Neck;Lymphadenectomy;End-to-side;Lymphaticovenularanastomosis;Supermicrosurgery

* Corresponding author. DepartmentFrance.

E-mail address: bayestaray@yahoo

1748-6815/$ - see front matter Crown CAesthetic Surgeons. All rights reserved.http://dx.doi.org/10.1016/j.bjps.2012.0

Summary Background: Head and neck lymphoedema secondary to jugular lymphadenect-omy is a severe issue, without efficient solution. Successful treatment of lymphoedema ofthe upper and lower limbs has become possible with supermicrosurgical lymphaticovenularanastomosis. The technique based on two end-to-side anastomosis is named p-shaped lympha-ticovenular anastomosis. We have evaluated this method for chronic head and neck lymphoe-dema.Methods: From November 2010 to April 2011, four patients with a chronic head and neck lym-phoedema were treated by p-shaped lymphaticovenular anastomosis. Three patients hada unilateral lymphoedema, and one patient had a bilateral lymphoedema. The mean age ofthe patients was 63.2 years (range, 46e77 years). The mean duration of the lymphoedemawas 2.6 years (range, 1e5). Every patient was operated under local anaesthesia througha face-lift skin incision. One p-shaped lymphaticovenular anastomosis was performed at eachoperative site.Results: The average operative time to perform one p-shaped lymphaticovenular anastomosiswas 1.9 h (range, 1.8e2.5). The calibre of lymphatic vessels used for lymphaticovenular anas-tomosis ranged from 0.3 to 0.7 mm (average, 0.5). A venous back-flow was found in seven lym-phaticovenular anastomosis (70%). Three patients (75%) had a qualitative improvement of skintissue and a significant circumferential reduction after surgery. The average circumferentialdifferential reduction rate was 3.7% (range, 0.6e7.8) (p Z 0.006). The average cross-sectional area differential reduction rate was 7.2% (range, 1.2e15.1) (pZ 0.007). The averagevolume differential reduction rate was 6.9% (range, 2e14.8) (p Z 0.05).Conclusions: The authors present a new option to treat head and neck lymphoedema. p-Shaped lymphaticovenular anastomosis is an effective method to reduce the severity of skintissue fibrosis and lymphoedema volume. Further studies with larger groups of patients arerequired to confirm the outcome of this preliminary study.

of Plastic and Reconstructive Surgery, Nimes University Hospital, pl Pr Robert Debre, 30000 Nimes,

.fr (B. Ayestaray).

opyright ª 2012 Published by Elsevier Ltd on behalf of British Association of Plastic, Reconstructive and

8.049

202 B. Ayestaray et al.

EBM Level Z level 4.Crown Copyright ª 2012 Published by Elsevier Ltd on behalf of British Association of Plastic,Reconstructive and Aesthetic Surgeons. All rights reserved.

Head and neck lymphoedema secondary to jugular lympha-denectomy is a debilitating condition with a severe psycho-logical impact. Few medical treatments can be applied:lymph drainage by physiotherapists is the most useful;facial compressive therapy can be used at night but is notsuitable in the daily life; elastic bandage is impossible forthe face.

Lymphaticovenular anastomosis using supermicrosurgi-cal techniques have proved their efficacy in chroniclymphoedema of the limbs.1e6 A successful case of facelymphoedema relief has also been reported by Miharaet al. with end-to-end lymphaticovenular anastomosis.7

p-Shaped lymphaticovenular anastomosis is a double end-to-side lymphaticovenular anastomosis. This method doesnot sacrifice the venous network and is a venous flow-sparingtechnique. We have evaluated the efficacy of p-shapedlymphaticovenular anastomosis in four patients havinga head and neck lymphoedema secondary to a jugularlymphadenectomy associated with an intraoral cancerresection.

Patients and methods

From November 2010 to April 2011, four patients with a headand neck lymphoedema were treated by p-shaped lympha-ticovenular anastomosis. All patients had a chronic lymphoe-dema secondary to a jugular lymphadenectomy associatedwith an intraoral cancer resection. This lymphoedemaevolved since at least 1 year and was not effectivelyresponding to conservative treatment, such as compressiontherapy with a nightly elastic stocking and lymph drainageduring at least 6 months (Table 1). The mean age of thepatients was 62 years (range, 46e77). The mean durationof the lymphoedema was 2.5 years (range, 2e3).

Two criteria were analysed, before and every 2 monthsafter the surgery, during a follow-up of 12 months, to assessthe outcomes of the surgery:

- a qualitative analysis of the soft-tissue characteristics:softness, thickness (pinch test) and sensibility

- a quantitative analysis, by measuring the circumfer-ence (Figure 1) of the head and neck at three differentlevels (C1: head circumference, C2: superior part of theneck above the thyroid cartilage and C3: inferior part ofthe neck under the cricoid cartilage) and by calculatingthe cross-sectional area (CSA Z p.r2 Z C2/4p) at thesame levels and the mean volume of the lymphoedema(V Z p.h (C12 þ C32 þ C1.C3)/12).

Statistical analysis

Quantitative data were analysed by an independent biostat-istician from the department of biostatistics of MontpellierUniversity Hospital. The t-testwas used to compare themean

of postoperative measures to the preoperative value, overalland for each patient, with a follow-up of 12months. Values ofp < 0.02 were considered statistically significant.

Surgical procedure

All patients were operated under local anaesthesia with 1%lidocaine with 1/100 000 epinephrine. The mean volume oflocal anaesthetic used was 12.3 ml (range, 8e15). Througha tragal face-lift skin incision, the superficialis temporalisvein was dissected between the tragus and the parotidfascia. The adjacent superficial lymphatic vessels (pre-auricular lymphatic vessels)8,9 were dissected subdermally,along the superficialis temporalis vein. After dissection ofthe subdermal lymphatic vessels, their calibres weremeasured with a professional scaled ruler (Shinwa SokuteiCo., Ltd., Sapporo Eigyosho, Japan). p-Shaped lymphatico-venular anastomoses were performed with EMI SuperMicro-surgery instruments (EMI Factory Co., Ltd., Nagano, Japan)and a 11/0 nylon monofilament on a 50-mm needle (Ethilon,Ethicon, Johnson and Johnson Co., Cornelia, GA, USA). Thelarger superficial lymphatic vessel was transected withmicroscissors. Then, two end-to-side lymphaticovenularanastomosis were performed on the lateral side of thesuperficialis temporalis vein wall. This lymphaticovenularanastomosis configuration has the shape of a p (Figure 2).One p-shaped lymphaticovenular anastomosis was per-formed for three patients (cases 1, 2 and 3) having a unilat-eral lymphoedema. Two p-shaped lymphaticovenularanastomosis (one on the right side, one on the left side)were performed for one patient (case 4) having a bilaterallymphoedema.

Results

The average operative time to perform one p-shapedlymphaticovenular anastomosis was 1.9 h (range,1.8e2.5). The calibre of the lymphatic vessels used forthe lymphaticovenular anastomosis ranged from 0.3 to0.7 mm (average, 0.5). A venous back-flow10 was found inseven lymphaticovenular anastomoses (70%).

No patient (0%) experienced a postoperative worseningof his/her lymphoedema during the year of follow-up. Nopatient (0%) experienced a postoperative haematoma orinfection (Table 1). One patient (25%) developed a partialskin necrosis (5 mm) along the tragal scar (case 4). Allpatients (100%) had a qualitative improvement of skintissue after surgery. The soft tissues were softer, the pinchtest was thicker (average reduction of 6 mm) and thesubjective skin sensibility was improved. Three patients(75%) had a significant circumferential reduction(p < 0.02) after surgery (Table 2). The average circumfer-ential differential reduction rate was 3.7% (range,0.6e7.8) (p Z 0.006). The average cross-sectional area

Table

1Su

mmary

ofpatients

treatedbyp-shapedlymphatico

venularanastomosis.

Patient

Age

(years)

Sex

Durationof

lymphoedema

(years)

Seve

rity

Side

Cause

of

lymphadenectomy

Operative

time

Number

ofLV

Calibre

ofLV

LVA

configu

ration

Complica

tion

168

F2

þþL

CheekSC

C2

20.3e

0.5

p0

00

246

M2

þR

CheekSC

C1.8

10.6

p0

00

377

F3

þþþ

LCheekSC

C1.6

20.4e

0.7

p0

00

457

M3

þþþ

Rþ

LTongu

eSC

C2.5

R:1

R:0.5

p0

0þ

L:2

L:0.5e

0.7

p0

00

M:male,F:female.

Seve

rity:moderate

(þ),

seve

re(þ

þ),fibrosis(þ

þþ).

R:righ

t,L:

left.

SCC:squamousce

llca

rcinoma.

LV:lymphaticve

ssel.

LVA:lymphatico

venularanastomosis.

Figure 1 Anatomical model for the follow-up of the circum-ferential and cross-sectional area evolution after surgery.Three different levels are chosen to measure the circumfer-ence of the head and neck (C1: head, C2: superior part ofthe neck above the thyroid cartilage, and C3: inferior part ofthe neck under the cricoid cartilage). The cross-sectionalarea is calculated at the same levels. The average volume oflymphoedema is calculated between C1 and C3.

Figure 2 Intraoperative view of one p-shaped lymphaticove-nular anastomosis. Two windows are performed through thewall of the superficialis temporalis vein, (star) as large as thecalibre of the adjacent lymphatic vessel. The lymphaticchannel is cut with microscissors. Each stump (arrow) isconnected to the lateral side of the vein, in an end-to-sidefashion. This configuration does not interrupt the venous flowin the recipient vein.

p-Shaped lymphaticovenular anastomosis 203

Table 2 Summary of cases: clinical outcomes.

Patient mC(cm)

mCreduction(cm)

Average mCreductionrate (%)

mCSA(cm2)

mCSAreduction(cm2)

Average mCSAreductionrate (%)

mV(cm3)

mVreduction(cm3)

Average mVreductionrate (%)

p

1 49.6 1.6 3.2 195.8 12.4 6.3 64,877.9 3751.5 5.8 0.022 48.6 1.6 3.3 188 12.2 6.5 62,362.7 3242.8 5.2 0.013 50.6 0.3 0.6 203.8 2.4 1.2 68,509.3 1299.9 2 0.04*4 51 4 7.8 207 31.2 15.1 68,276.2 10,110 14.8 0.004

mC: mean circumference.mC Z (C1 þ C2 þ C3)/3.mCSA: mean cross-sectional area.mV: Mean volume.*p > 0.02: Non significant.

204 B. Ayestaray et al.

differential reduction rate was 7.2% (range, 1.2e15.1)(p Z 0.007). The average volume differential reductionrate was 6.9% (range, 2e14.8) (p Z 0.05). Three patients(75%) stopped lymph drainage and wearing their nightlyelastic stocking definitively 10 months after surgery. Threepatients (75%) concluded to a better quality of life 1 yearafter surgery.

Case report

Case 4

A 57-year-old male patient presented in our clinic fora bilateral head and neck lymphoedema, secondary toa bilateral jugular lymphadenectomy, associated with theexcision of a squamous cell carcinoma of the tongue, 3years ago. He was also treated by radiochemotherapy atthe same time. Lymph drainage and compression therapydid not lead to a sufficient oedema relief. He alsoexperienced recurrent skin infections at the lower part ofhis lymphoedema (Figure 3(a) and (b)). One p-shaped lym-phaticovenular anastomosis was performed at each superfi-cial temporalis vein with one adjacent superficial lymphaticvessel, under local anaesthesia. After a follow-up of 12months, the average circumferential reduction rate was7.8%, the average cross-sectional area reduction rate was15.1% and the average volume reduction rate was 14.8%(Figure 3(c) and (d)). He definitively stopped wearing hisnightly compressive stocking and lymph drainage 9 monthspostoperatively (Figures 4e6).

Discussion

Postoperative lymphoedema remains a problem after lym-phadenectomy associated with a cancer resection. Conser-vative treatments are most of the time not efficient, andthe social impact is important. Thus, the concept oflymphatic vessel connections to veins was developed, inview to create new ways of lymphatic drainage.1e3 Upperand lower limb lymphoedema have been first treatedsuccessfully by lymphaticovenular bypasses. Secondaryhead and neck lymphoedema has not developed the same

interest, until supermicrosurgical techniques weredescribed by Koshima et al.3e6 A successful case of facelymphoedema reduction was reported by Mihara et al. in2011.7 The authors used one end-to-end lymphaticovenularanastomosis in the submandibular region and one end-to-side lymphaticovenular anastomosis in the pre-auricularregion, between superficial lymphatic vessels andsubdermal venules.

End-to-end lymphaticovenular anastomosis is the mostcommonly used configuration of lymphaticovenular anasto-mosis. The main issue of this technique is to sacrifice thedistal venous network. On the other hand, end-to-sidelymphaticovenular anastomosis does not interrupt thevenous flow or sacrifice the distal venous stump. In p-shaped lymphaticovenular anastomosis, which is a doubleend-to-side lymphaticovenular anastomosis, the distalendothelial cells and the venous flow are respected. Thevenous system is important in lymphatic fluid drainage fortwo reasons. First, it is responsible for the reabsorption of90% of the interstitial fluid.11 Then, venous endothelial cellsare involved in lymphangiogenesis by centrifugal sprout-ing.12,13 One endothelial receptor, vascular endothelialgrowth factor receptor 3 (VEGFR-3), has been recognisedas the most specific for lymphangiogenic growth factors,VEGF-C and VEGF-D.14,15 The venous network, by itsdrainage potential and its part in lymphangiogenesis, seemsto be a major element to preserve in lymphoedemapatients.

We developed a p-shaped lymphaticovenular anasto-mosis to treat four patients, having a head and necklymphoedema, addressed in our department betweenNovember 2010 and April 2011. This configuration of lym-phaticovenular anastomosis was experienced by Narushimaet al.16 and Yamamoto et al.10 in lymphoedema of the upperand lower limbs. The operation was performed under localanaesthesia. The incision site was a face-lift tragal incision,for the scar to be as undetectable as possible postopera-tively. The outcomes were satisfying with a significant qual-itative and quantitative improvement of lymphoedema(p < 0.02) in three patients. The skin became softer, thepinch test was thicker and the sensibility was improved aftersurgery. The quantitative improvement was moderate witha clinically significant reduction in three patients. Theaverage circumferential reduction rate was 3.7%, cross-

Figure 3 Case 4: Bilateral head and neck lymphoedema, secondary to a bilateral jugular lymphadenectomy associated with theexcision of a tongue carcinoma. Preoperative views (a,b) and postoperative views (c,d) 12months after performing onep-shaped lym-phaticovenular anastomosis, between each superficial temporalis vein and one superficial lymphatic vessel. The average circumfer-ential reduction rate was 7.8%, the cross-sectional area reduction rate was 15.1%, and the volume reduction rate was 14.8%.

Figure 4 Variations of the head and neck average circumfer-ence in patients treated by p-shaped lymphaticovenularanastomosis.

Figure 5 Variations of the head and neck average cross-sectional area in patients treated by p-shaped lymphaticove-nular anastomosis.

p-Shaped lymphaticovenular anastomosis 205

Figure 6 Variations of the head and neck average volume inpatients treated by p-shaped lymphaticovenular anastomosis.

206 B. Ayestaray et al.

sectional area reduction rate was 7.2% and volume reductionrate was 6.9% after a follow-up of 12 months. Surgery did notlead to a complete resorption of the lymphoedema butimproved the aesthetic and functional outcomes. Thequality of life was also felt to be better 1 year after surgery.These results are very encouraging. In this way, p-shapedlymphaticovenular anastomosis is a minimal invasivemethod, which seems to improve the severity of chronichead and neck lymphoedema, resisting the classical conser-vative treatment. Its low rate of morbidity and benefits forpatients make it an interesting technique for head andneck lymphoedema surgery.

The severity of head and neck chronic lymphoedemacan be reduced by p-shaped lymphaticovenular anasto-mosis. The aesthetic and functional outcomes areimproved by this method, which is minimally invasiveand can be performed under local anaesthesia. Furtherstudies with larger groups should be done to confirm thispreliminary study.

Conflict of interest

None.

Acknowledgements

The authors thank very much Gregoire Mercier, M.D., M.S.from the department of biostatistics of Montpellier Univer-sity Hospital for the statistical analysis of the data.

References

1. Yamada Y. Studies on lymphatic venous anastomosis in lymphe-dema. Nagoya J Med Sci 1969;32:1e21.

2. O’Brien BM, Sykes P, Threlfall GN, Browning FS. Microlymphati-covenous anastomoses for obstructive lymphedema. PlastReconstr Surg 1977;60:197e211.

3. Koshima I, Kawada S, Moriguchi T, Kajiwara Y. Ultrastructuralobservations of lymphatic vessels in lymphedema in humanextremities. Plast Reconstr Surg 1996;97:397e405. discussion406-7.

4. Koshima I, Inagawa K, Urushibara K, et al. Supermicrosurgicallymphaticovenular anastomosis for the treatment of lymphe-dema in the upper extremities. J Reconstr Microsurg 2000;16:432e7.

5. Koshima I, Nanba Y, Tsutsui T, et al. Long-term follow-up afterlymphaticovenular anastomosis for lymphedema in the leg. JReconstr Microsurg 2003;19:209e15.

6. Nagase T, Gonda K, Inoue K, et al. Treatment of lymphedemawith lymphaticovenular anastomoses. Int J Clin Oncol 2005;10:304e10.

7. Mihara M, Uchida G, Hara H, et al. Lymphaticovenous anasto-mosis for facial lymphoedema after multiple courses oftherapy for head-and-neck cancer. J Plast Reconstr AesthetSurg 2011;64:1221e5.

8. Millard FP, Walmsley AG. Applied anatomy of the lymphaticsof the head and neck. In: Millard FP, editor. Appliedanatomy of the lymphatics. New York: Kessinger PublishingCo.; 1922. p. 31e52.

9. Zuther JE. Lymphatic drainage of the scalp, face and neck. In:Zuther JE, editor. Lymphedema management. 2nd ed. Stutt-gart: Thieme Verlag; 2009. p. 18e9.

10. Yamamoto T, Narushima M, Kikuchi K, et al. Lambda-shapedanastomosis with intravascular stenting method for safe andeffective lymphaticovenular anastomosis. Plast Reconstr Surg2011;127:1987e92.

11. Starling EH. On the absorption of fluids from the connectivetissue spaces. J Physiol (Lond) 1895;19:312e26.

12. Sabin FR. On the origin of the lymphatic system from the veinsand the development of the lymph hearts and thoracic duct inthe pig. Am J Anat 1902;1:367e91.

13. Wigle JT, Harvey N, Detmar M, et al. An essential role for Prox1in the induction of the lymphatic endothelial cell phenotype.EMBO J 2002;21:1505e13.

14. Joukov V, Pajusola K, Kaipainen A, et al. A novel vascular endo-thelial growth factor, VEGF-C, is a ligand for the Flt4 (VEGFR-3)and KDR (VEGFR-2) receptor tyrosine kinases. EMBO J 1996;15:290e8.

15. Jussila L, Alitalo K. Vascular growth factors and lymphangio-genesis. Physiol Rev 2002;82:673e700.

16. Narushima M, Mihara M, Yamamoto Y, et al. The intravascularstenting method for treatment of extremity lymphedemawith multiconfiguration lymphaticovenous anastomoses. PlastReconstr Surg 2010;125(3):935e43.