feasibility of sentinel node biopsy in breast cancer surgery · node biopsy (snb) in lymph node...

FEASIBILITY OF SENTINEL NODE BIOPSY

IN BREAST CANCER SURGERY

Riitta Rönkä

Breast Surgery UnitHelsinki University Central Hospital

University of HelsinkiHelsinki, Finland

Academic dissertation

To be presented, with the permission of the Medical Faculty of the University of Helsinki,

for public examination in the Auditorium of Faltin at the Surgical Hospital,

Kasarminkatu 11–13, Helsinki University Hospital,

on the 22nd of October at 12 noon

Helsinki 2004

2

This thesis was supervised byDocent Karl von SmittenBreast Surgery UnitHelsinki University Hospital

and

Marjut Leidenius, MD, PhDBreast Surgery UnitHelsinki University Hospital

Reviewed byProfessor Kaija HolliPalliative MedicineMedical School of theUniversity of Tampere

and

Docent Markku HärmäDepartment of Plastic SurgeryKuopio University Hospital

Opponent for the DissertationProfessor Inkeri ElomaaDepartment of OncologyHelsinki University Hospital

ISBN 952-91-7754-2 (nid.)ISBN 952-10-2075-X (PDF)

Yliopistopaino, Helsinki.2004

3

To Juuli, Niila and Ari

4

List of Original Articles ............................................................................ 5

List of Abbreviations ................................................................................ 6

Abstract ................................................................................................... 7

Introduction ............................................................................................ 9

Review of the Literature ............................................................................ 111. Breast Cancer Incidence and Survival ................................................. 112. Lymph Node Metastases ................................................................... 113. Staging of Breast Cancer ................................................................... 134. Axillary Lymph Node Dissection ...................................................... 145. Axillary Lymph Node Sampling ....................................................... 156. Lymphatic Mapping and Sentinel Node Biopsy ................................ 15

6.1 History ...................................................................................... 156.2 Methods..................................................................................... 166.3 Accuracy..................................................................................... 18

7. Sentinel Node Biopsy and Cost Impacts ............................................ 208. Sentinel Node Biopsy and Surgery of Nonpalpable Breast Cancer ...... 209. Sentinel Node Biopsy and Morbidity After Breast Cancer Surgery ...... 22

9.1 Lymphedema ............................................................................. 229.2 Upper Arm Lymphedema .......................................................... 239.3 Breast Lymphedema ................................................................... 239.4 Functional Disorders of the Upper Extremity ............................. 249.5 Sensory Disorders of the Upper Extremity .................................. 24

Hypothesis of the Study............................................................................ 25

Patients and Methods ............................................................................... 26

Results ..................................................................................................... 33

Discussion ................................................................................................ 44

Conclusion ............................................................................................... 53

Acknowledgements .................................................................................. 54

References ................................................................................................ 56

Table of Contents

5LIST OF ORIGINAL ARTICLES

I Rönkä R, Krogerus L, Leppänen E, von Smitten K, Leidenius M. Sentinelnodes outside level I–II of the axilla and staging in breast cancer. AnticancerRes 22(5): 3109–3112, 2002

II Rönkä R, Krogerus L, Leppänen E, von Smitten K, Leidenius M. Radiogu-ided occult lesion localization (ROLL) in patients undergoing breast con-serving surgery and sentinel node biopsy Am J Surg 178 (4): 491–6, 2004

III Rönkä R, von Smitten K, Sintonen H, Kotomäki T, Krogerus L, LeppänenE, Leidenius M. The impact of sentinel node biopsy and axillary stagingstrategy on hospital costs. Ann Oncol 15 (1): 88–94, 2004

IV Rönkä R, Pamilo M, von Smitten K, Leidenius M. Breast lymphedema afterbreast conserving treatment and sentinel node biopsy. Acta Oncol 6(43):551–7, 2004

V Rönkä R, Tasmuth T, von Smitten K, Leidenius M. One-year morbidity ofsentinel node biopsy and breast surgery, in press

List of Original Articles

6 LIST OF ABBREVIATIONS

List of Abbreviations

AC axillary clearanceALND axillary lymph node dissectionBCS breast conserving surgeryBCT breast conserving therapyIMN internal mammary nodesLS lymphoscintigraphyRT radiotherapyROLL radioguided occult lesion localizationSN sentinel lymph nodeSNB sentinel node biopsyWGR wire-guided resectionUS ultrasound

7ABSTRACT

Abstract

Aim: The purpose of the present study was to evaluate the feasibility of sentinelnode biopsy (SNB) in lymph node staging in breast cancer with a special empha-sis on radioguided occult lesion localization (ROLL), hospital costs, and post-operative morbidity.

Patients and methods: The study population comprised of clinically stage T1-2and node-negative breast cancer patients, who underwent breast surgery and SNBbetween 31 May 2000 and 28 February 2002 at the Breast Surgery Unit, Helsin-ki University Hospital. Preoperative lymphoscintigraphy (LS), intraoperative gam-ma probe, and blue dye were used for identification of sentinel nodes (SNs).Axillary clearance (AC) was performed when the SNs in the axilla were involvedor were not identified. Any SN outside level I-II of axilla, when clearly visible inLS, was also excised. Single intratumoral injection of the radioactive tracer inultrasound (US) guidance for nonpalpable tumors visible in US, was the onlytumor localization method (ROLL).

The sequence of the treatment procedures of 237 consecutive patients wasrecorded and evaluated by use of three hypothetical scenarios: Diagnostic axillarylymph node dissection (ALND), SNB without frozen section diagnosis, and SNBas day-case surgery prior to the breast operation. The total hospital costs werecalculated in all protocols.

One-year breast morbidity was evaluated by clinical examination and breastultrasound (US) in 102 patients. To evaluate postoperative arm morbidity, aphysiotherapist examined 83 patients preoperatively and 3,6, and 12 monthspostoperatively; the range of shoulder motion on both sides and the sense oftouch in the axillary and breast areas was measured. Self-experienced morbiditywas estimated by a questionnaire concerning postoperative pain, range of shoul-der motion, numbness, muscle weakness, lymphedema, and sensory disorders.

Results: LS showed SN in the axilla in 88% of the 170 patients, 17% with SNoutside level I–II of the axilla. Axillary metastases were found in 40% ofthe patients with SNs outside the axilla; SN outside the axilla were involved in 3%of all cases. Two patients with extra-axillary node metastases had no axillary me-tastases.

Excision of the tumors was complete in 93% of all the 215 cases after theprimary operation and in 89% with radioguided occult lesion localization (ROLL).In addition, at least 10-mm free margins were achieved with ROLL in a primaryoperation in 69% of the patients.

The hospital costs per patient were 3750 € by frozen-section diagnosis ofSNs. Hospital costs per patient would have been 3020 € with the ALND for

8 ABSTRACT

axillary staging, 4087 € without the frozen section, and 4573 € with the “SNBas day-case surgery” model. Costs with or without intraoperative frozen-sectiondiagnosis would have been equal with a false-negative rate of 35%.

One year after surgery, the clinical examination revealed breast edema in 23%of 57 patients after SNB. US revealed subcutaneous edema and interstitial fluid inthe breast significantly more often after AC. Approximately a quarter of the 43patients experienced at least mild arm morbidity one year after SNB alone, where-as half the patients had similar symptoms after AC. None of the patients afterSNB alone received lymph therapy.

Conclusion: SNB is a feasible method in lymph node staging of breast cancer.Intratumoral tracer injection makes excision of nonpalpalpable tumors possiblewithout other localization methods. SNB seems to be associated with higherhospital costs than is diagnostic ALND, but frozen section diagnosis in SNB isworthwhile as long as the false-negative rate is less than 35%. The risk for severemorbidity after SNB seems to be low.

9INTRODUCTION

Introduction

The status of the lymph nodes, in the axilla and in the internal mammary chain,is the most significant prognostic factor for survival in breast cancer (Morrow etal 1981, Veronesi et al 1985, Noguchi et al 1991). Ten-year survival of 80% to100% has been reported, when both the axillary and the internal mammary nodes(IMN) were uninvolved, compared to 24% to 30% survival of patients with me-tastases in both nodal basins (Veronesi et al 1985, Noguchi et al 1991). Theintermediate 10-year survival rate was approximately 55%, when only one of thenodal stations was separately affected (Veronesi et al 1985). Furthermore, thenumber of involved lymph nodes correlates with survival. When over three nodesare involved, the 10-year survival can fall from 90% to 27% (Noguchi et al 1991).Metastases in the axillary lymph nodes are found in about 40% of patients and in20% to 25% in the internal mammary chain, and in even more than 30% when theaxillary nodes are also involved (Donegan 1977, Morrow et al 1981, Lacour et al1983, Veronesi et al, 1985).

Axillary lymph node dissection (ALND) has been the gold standard in axil-lary staging, providing information for decision-making about adjuvant therapy(Giuliano et al 1994, Albertini et al 1996, O’Hea et al 1998, Veronesi et al2003). ALND is considered also as a treatment leading to locoregional control,when the axillary lymph nodes are involved (Bland et al 1999, Axelsson et al2000). Recently, lymphatic mapping with sentinel node biopsy (SNB) has beenproposed to provide even more accurate lymphatic staging with less morbiditythan with diagnostic axillary clearance (AC), (Giuliano et al 1994, Albertini et al1996, Krag and Harlow 2003). Therefore, SNB serves as an indicator for axillarytreatment in an increasing number of breast surgery centers.

Since the beginning of the 1990’s the SNB technique has been a major focusof breast cancer research, although the universal standardized techniques for SNBare still lacking.

Numerous studies have reported arm morbidity after breast cancer treatment,with a focus especially on the extent of axillary surgery. A general and widelyknown postoperative complication after ALND is morbidity of the upper ex-tremity (Petrek and Heelan 1998, Duff et al 2001, Ververs et al 2001, Ernst et al2002, Gosselink et al 2003) including lymphedema, as well as sensory and func-tional disorders like numbness and motion restrictions of the shoulder joint.These symptoms manifest among 20% to 80% of women who have undergoneAC and received radiotherapy (RT), (Tasmuth et al 1996, Ververs et al 2001, Yapet al 2003, Gosselink et al 2003). The risk for breast and upper arm lymphedemaseems to increase after more extensive axillary surgery (Clarke et al 1982, McCor-mick et al 1989, Senofsky et al 1991, Temple et al 2002) and postoperative RT

10 INTRODUCTION

(Clarke et al 1982, McCormick et al 1989, Senofsky 1991, Isaksson and Feuk2000). Most of these symptoms manifest immediately or up to one year aftersurgery, but symptoms can appear until 2 to 5 years after ALND (Warmuth et al1998, Ernst et al 2002).

A decreased risk for early, late, and long-term morbidity has been expectedafter the introduction of SNB without further axillary treatment. The first studieshave reported significantly less arm morbidity after SNB than after AC (Schrenket al 2000, Miguel et al 2001, Sener et al 2001, Burak et al 2002, Ernst et al2002, Haid et al 2002, Temple et al 2002, Blanchard et al 2003, Gosselink et al2003, Leidenius et al 2003a, Peintinger 2003, Schijven et al 2003). SNB with-out further axillary treatment is, however, also associated with postoperativemorbidity, but the clinical significance of the SNB-related morbidity is still ob-scure.

The purpose of the present study was to evaluate the feasibility of sentinelnode biopsy in lymph node staging in breast cancer with a special emphasis onhospital costs and postoperative morbidity. A further purpose was to study whethera nonpalpable tumor can be localized by the same intratumoral injection of radi-oactive tracer used for lymphatic mapping.

11REVIEW OF THE LITERATURE

Review of the Literature

1. Breast cancer incidence and survival

Breast cancer is worldwide the most common cancer in women: up to one wom-an in eight will develop it during her lifetime. The annual incidence in Finland in2001 was 3646 new cases, and the age-adjusted incidence was 82.7/100 000(Finnish cancer registry, 2002). The predicted cancer-specific 5-year relative sur-vival rate for early stage breast cancer is 85% (Finnish Cancer Registry, 2002).

2. Lymph node metastases

Already in the end of the 1800’s, Halsted described the spread of breast cancer fromthe primary tumor to the regional lymph nodes and systemic circulation (Bland &Copeland 2004). The prevalence of axillary lymph node metastases has been 30%to 40% in all breast cancer patients (Rosen et al 1983, Veronesi et al 1997, Nieweget al 2001) and 10% to 18% in T1 tumors (Port et al 1998, Axelsson et al 2000,Rivadeneira et al 2000, Bland & Copeland 2004). The presence of internal mam-mary node metastases is strongly related to axillary node involvement. A sole me-tastasis directly to the internal mammary nodes is rare (Morrow et al 1981, Noguchiet al 1991). The prevalence of solitary internal mammary metastases varies from 3%to 20% (Morrow et al 1981, Noguchi et al 1991, Valdés Olmos et al 2001). Al-though metastasis is considered to proceed stepwise, skip metastases in the axillacan be found, but in less than 2% of the cases (Rosen et al, 1983). Table 1

Table 1: Prevalence of axillary and internal mammary lymph node metastases inbreast cancer, 1977–2003

Only internal Axillary andPublications Axillary mammary internal mammary

metastases metastases i metastases(%) (%) (%)

Donegan et al 1977 58.0 2.6 19.0Veronesi et al 1981 53.0 4.3 16.0Noguchi et al 1991 41.0 4.8 35.0Cody et al 1995 38.0 11.0 14.0Galimberti et al 2003 37.0 2.5 6.3

12 REVIEW OF THE LITERATURE

Knowledge of lymphatic drainage patterns and tumor cell dissemination is oneof the basic tools in cancer surgery. Introduction of the SN concept clarified howlittle is known about the mechanism of lymphatic metastasis (Quan et al 2002,Bland & Copeland 2004). It is considered to be an orderly progression of tumorcells within the lymphatic system (Borgstein et al 1997).

The mammary gland is embryologically derived from the ectoderm and devel-opes within a skin envelope. Lymphatic drainage from the skin of the breast, thenipple areolar complex, and some of the central portion of the gland occurs viathe superficial subareolar lymphatic plexus. This plexus is in connection with thedeep lymphatic plexus by interconnecting plexuses (Mortimer 1998, Nathansonet al 2001, Bland & Copeland 2004). Both the dermal and parenchymal or deep-er lymphatics of the breast drain to the axillary lymph nodes (Nathanson et al2001). Lymph drainage can be occluded indirectly by tumor growth or infec-tion (Mortimer 1998, Bland & Copeland 2004).

Pieter Camber (b. 1722) and Paolo Mascagni (b. 1752) were the first to de-scribe the lymphatic drainage from the breast to the internal mammary nodesand pectoral nodes (Bland & Copeland 2004). At the beginning of the twentiethcentury the theory of tumor dissemination by lymphatics and bloodstream, wasaccepted widely. This formed also the basis for radical mastectomy endorsed byHalsted (Bland & Copeland 2004). Sappey (1874) has described the superficialand deep lymphatic draining systems of the breast. Later anatomic studies havedemonstrated that approximately 75% of lymph drainage is directed to the axillaand 25% to the internal mammary chain (Bundred et al 1994, Uren et al 2001).Turner-Warwick (1959) produced autoradiographs and detected that axillary andinternal mammary nodes could receive lymph drainage from any quadrant of thebreast Bland & Copeland 2004).

The size of the primary tumor is the most significant predictive factor foraxillary lymph node metastases (Axelsson et al 2000, Rivadeneira et al 2000)(Table 2). Furthermore, younger age of the patient and poor histologic grade ofthe primary tumor increase risk for nodal involvement (Rivadeneira et al 2000).Axillary involvement is observed mostly at Berg level I of the axilla (see page 16),substantially elevating risk for the involvement of higher-level lymph nodes (Daniellet al 1988).


Table 2: Relationship of tumor size and axillary lymph node metastases in breastcancer

Tumor Prevalence of axillarysize metastases (%)

<0.5 20.60.5–0.9 20.61.0–1.9 33.22.0–2.9 44.93.0–3.9 52.14.0–4.9 60.0

>5.0 70.1

Data from Bland & Copeland 2004

3. Staging of breast cancer

The classification of breast tumors is based on the TNM system, which includeshistological tumor size (pT), nodal metastases (pN), and distant metastases (pM)(Table 3 and 4) (Sobin and Wittekind 2002). Originally, the TNM system wascreated for grouping patients into different survival and prognostic groups. After-wards, it has been a pragmatic and evidence-based model for targeting and deter-mining treatment, providing a uniform approach between different centers (Bland& Copeland 2004). The last edition of the TNM classification in particular isconsidered a reference for lymph nodes status, especially for micrometastases (Sobinand Wittekind 2002). This new TNM classification in addition includes quantita-tive criteria for classification of micrometastasis, an issue emerging after introduc-tion of the SNB. Revisions of the previous staging system are also related to thenumber of axillary lymph nodes involved and to the classification of metastases inlevel III axillary lymph nodes and lymph nodes in the internal mammary chain.

Table 3: pT classification and histological size of the tumor in breast cancer

Classification of primaryTumor diameter

tumor

pT1 ≤ 2cmpT1a >0.1 – < 0.5 cmpT1b > 0.5 – < 1.0 cmpT1c > 1.0 – < 2.0cmpT2 > 2.0 cm – < 5.0cmpT3 > 5.0cmpT4 Tumor of any size with direct extension

to chest wall or skin

Data from Sobin and Wittekind, 2002


Table 4: pN classification of breast cancer based on axillary lymph node dissec-tion with or without sentinel node biopsy

Classification of regional Status of regional lymph nodes

lymph nodes

pNx Cannot assessPN0 / PN0(i-) No metastasis / no metastasis histologically, negative IHC

PN0(i+) No metastasis / no metastasis histologically, positive IHC,no IHC clusters > 0.2 mm

pN0(mol-) No metastasis / no metastasis histologically or molecularlypN0(mol+) No metastasis / no metastasis histologically, positive

molecular findingspN1a / pN1b Metastasis in 1–3 axillary lymph nodes and / or IMN

pN1c Metastasis in 1–3 axillary lymph nodes and IMNpN1mic Micrometastasis >0.2 mm – < 2.0 mm

PN2a / PN2b Metastasis in 4–9 axillary lymph nodes or in IMNPN3a Metastasis in ≥10 axillary lymph nodes or in infraclavicular

lymph nodesPN3b Metastasis in ipsilateral IMN and ≥1 positive axillary

lymph nodes or >3 positive axillary lymph nodes andmicroscopically defined IMN metastases

PN3c Metastasis in ipsilateral supraclavicular lymph nodes

IHC= immunohistochemistryIMN= internal mammary nodes

Data from Sobin and Wittekind, 2002

4. Axillary lymph node dissection

ALND has been the gold standard for decades in axillary staging of breast cancer,providing excellent loco-regional control of the disease and probably also a sur-vival advantage (Blichert-Toft 2000, Axelsson et al 2000). No imaging tech-nique or molecular biological marker can replace surgically defined nodal statusthus far (Blichert-Toft 2000).

One of the main basic questions is how many lymph nodes have to be exam-ined to determine reliably the status of nodes in the nodal basin (Recht andHoulihan 1995). The extent of axillary surgery is based on the Berg levels of theaxilla: level I (proximal), tissue inferior and lateral to the pectoral minor muscle;level II (middle/ central), nodes posterior to the pectoral minor muscle andbelow the axillary vein; level III (apical), nodes medial to the pectoral minormuscle against the chest wall (Bundred et al 1994). These levels, taken together,contain approximately 20 lymph nodes, which are mainly on level I. In general,level I–II axillary lymph node dissection is performed for staging purposes.


At least 10 axillary nodes should be examined to reach reliable staging andloco-regional control, when the axillary nodes are involved. For low- risk tumors(<10 mm), recurrence rate and recurrence-free survival was significantly betterwhen at least ten axillary nodes were removed (Axelsson et al 2000). The incidenceof “skip-metastases” has been 0.2% to 3% at level III, without concomitant metas-tasis to level I and II nodes (McMasters et al 1998, Rosen et al 1983, Veronesi et al1987). The main site for “skip-metastasis” is level II (Blichert-Toft, 2000).

When ALND is used as a staging procedure, extra-axillary nodal involvementremains unknown. Routine excision of IMN is not considered worthwhile, be-cause of the low incidence of involved nodes and increased risk for morbidity(Krag 2003).

5. Axillary lymph node sampling

The axillary lymph node sampling procedure includes random or palpation-di-rected biopsy of the lower axillary lymph nodes (Forrest et al 1995). The theoryis based on the impression that every intraoperatively palpable tumor containsmetastases (Forrest et al 1995). Loco-regional control has been equal to bothfour-node sampling and ALND (Forrest et al 1995, Chetty et al 2002). Intraop-erative axillary node sampling in combination with SNB may reduce the false-negative rate from 14% to 4% in tumors under 3 cm in diameter (Hoar andStonelake 2003). When only SN is involved, however, combining these proce-dures is not recommended (Hoar and Stonelake 2003).

The position of this procedure in diagnosis of lymph node status is controver-sial, because several authors consider the sampling procedure as a method equalto clinical examination and thus inadequate in breast cancer staging (Steele et al1985, Kissin et al 1982, Davies et al 1980). According to the Kissin and Daviesreports, axillary sampling failed to yield positive nodes in up to 24% of the pa-tients (Kissin et al 1982, Davies et al 1980). However, sampling like SNB isassociated with decreased morbidity such as lymph edema (Rampaul et al 2003,Forrest et al 1995).

6. Lymphatic mapping and sentinel node biopsy

6.1 History

Urologist Ramon Cabañas originally described the SN concept and usage of lym-phangiograms in 1977 in surgery for penile cancer. He confirmed that the pro-gression of the tumor cells into the lymphatic system is an orderly process, andthe first node to contain the metastasis shall be called an SN (Cabañas 1977). Themethod of Cabañas did not achieve wide acceptance, due to the arbitrary locali-zation method for SNs. Morton et al (1992) reported on cutaneous LS in iden-


tifying lymph nodes in melanoma, hypothesizing that identification of lymphdrainage pathways from the primary tumor might indicate the site of the metas-tases. Lymphatic mapping was developed for localization of these nodes, withvital blue dye to stain the lymphatics. Krag et al introduced in 1993 radiolabelledcolloids and a gamma probe for identification of SNs in breast cancer. They wereable to identify SNs in 82% of the patients (Krag et al 1993). Giuliano et al(1994) described first the intraoperative vital blue dye mapping in breast cancerthat achieved a 65% success rate. Albertini et al (1996) combined both tech-niques with a 92% success rate (Figure 1).

The sentinel node biopsy technique was developed as a diagnostic test todetermine the status of regional lymph nodes. It is considered a minimally inva-sive procedure, providing the same information as ALND with less morbidity. Astandardized technique for SNB is, however, thus far lacking.

As yet no results exist from long-term randomized trials comparing outcomesafter SNB and ALND (Krag et al 2003), but SNB has become a widely acceptedtechnique for axillary staging in breast cancer patients. SNB is usually applied inbreast cancer patients with tumors that are clinically node-negative, unicentric,and at stage T1 to T2. In addition to SNs in the axilla, LS visualises SNs outsidethe axilla in up to 34% of patients (Nieweg and Bartelink 2004).

Figure 1: Progression of lymph flow from a primary tumor to sentinel node and afterwardsto the second- and third-echelon nodes or two different sentinel nodes (SN)(Nieweg 2001 c, permission of Lippincott Williams & Wilkins)

6.2 Methods

Methods for SNB vary in regards to performing the LS and harvesting the lymphnodes (Borgstein et al 1998, Paganelli et al 1998, Glass et al 1999, Nieweg et al1999, Uren et al 2001, Valdés-Olmos et al 2000, McMasters et al 2001a, Krag etal 2003). The most variable parts of the method are the dosage, volume, andparticle size of the radioactive tracer, the site of the injection, and the use ofscintigraphy with or without blue dye.


Preoperative LS is performed in order to identify the draining LN basins andthe number of SNs in the basin (Jansen et al 1998, Nieweg et al 1999). The flowof radioactive colloid simulates the pathway of lymphatic spread and allows anaccurate prediction of nodes possibly involved (Kapteijn et al 1998).

Nodal uptake and the retention of the tracer depend on particle size and thevolume of the injection (Nieweg et al 1999, Valdés Olmos et al 2001, Leideniuset al 2004). Accumulation of the tracer in a node depends also on the conditionof the lymphatic channels entering from the primary tumor to an SN and the rateof the lymph flow (Nieweg et al 2001b).

The ideal tracer is drained swiftly to an SN and retained in the first drainingnode without overflow to the secondary, second-echelon, nodes. A small parti-cle-size agent accumulates fast and drains very quickly via the SN to the second-echelon nodes and can easily lead to visualization and harvesting of the second-echelon node (Nieweg et al 1999, Glass et al 1999). A large particle-size tracerremains at the injection site and may diminish the visualization of the non-SNs(Paganelli et al 1998, DeCicco et al 1998, Noguchi 2002).

Different radiocolloids are used for lymphatic mapping, but we still lack theideal one (Leppänen et al 2002, Leidenius et al 2004). The most widely usedradiopharmaceuticals are 99mTc- and isosulfran technetium and 99mTc-human col-loidal albumin (Krag et al 1993, Giuliano et al 1994, Veronesi et al 1997, Glasset al 1999). Administration of blue dye helps in the identification of blue-stainedlymph channels and their origin from the primary tumor (Miltenburg et al 1999).

The tracer may be injected intraparenchymally – intra- or peritumorally – orsuperficially – sub- or intradermally into the overlying skin at the tumor site orinto the subareolar lymphatic plexus. The idea of the superficial technique is thatthe overlying skin has the same embryologic origin as the breast parenchyma andshould share the same lymphatic drainage patterns (Borgstein et al 1997). Borg-stein proposed that both peritumoral tracer injection and intradermal injectiondrain to the same lymph node in the axilla (Borgstein et al 1997, 2000). Theadvantage of the superficial injection technique is the excellent visualization rateof SNs in LS, because of richer lymphatic drainage of the skin than of the breastparenchyma (McMasters et al 2001a, Mateos et al 2001).

On the other hand, it has been proposed that the tracer injection should beadjacent to the tumor; otherwise the risk is that a watershed of lymphatics iscrossed, and the visualized node will be draining another area of the breast andnot the tumor site (Canavese et al 2000, Valdés Olmos et al 2000). Drainagepatterns from the skin may be different from those of the underlying parenchy-ma, because the SN in the internal mammary chain is seldom visualized follow-ing an intradermal injection of the tracer (Roumen et al 1997). Deep injectioninto or around the tumor is suggested to lead to absorption of the tracer into thesame lymph channel as the one that the cancer cells enter (Nieweg et al 1999,Nathanson et al 2001) (Table 5).


Table 5: Visualization and identification rates for sentinel nodes (SN) by differ-ent lymphatic mapping techniques

Visualization Identification in LS rate of axillary

No. Injection (%) SNAuthor of technique and

Pts dose of tracer Axilla Outside (%)axilla

Krag 1998 443 PT (37 MBq) – 8 93van der Ent 1999 70 PT(370MBq) 97 34 100Rahusen 2000 115 PT (40–80MBq) 91 17 92Jansen 2000 113 IT (40–60MBq) – 19 88Valdés Olmos 2000 100 IT (42–88MBq) 96 19 90Valdés Olmos 2000 50 IT (68–124MBq) 96 17 98Doting 2000 141 IT (40–60MBq) 87 18 93Van der Ent 2001 256 PT (370MBq) 95 25 97Uren 2001 159 PT (5–10MBq) 93 8 96Galimberti 2002 182 PT/SF 92 8 99

IT= intratumoral, PT= peritumoral, SF=superficial

6.3 Accuracy

Before any technique can be accepted as a reliable, clinically useful, and validdiagnostic test, it should be sufficiently sensitive and specific. Overall accuracy isestimated by false-negative and false-positive rates, in addition to positive andnegative predictive values. The crucial factor concerning the validity of the meth-od is the false-negative rate, because any false-negative result could lead to under-staging and undertreatment. Furthermore, the false-negative and -positive ratesof the procedure should be acceptable before the method can become the goldstandard (McMasters et al 1998).

An SN is defined as a lymph node receiving direct drainage from the primarytumor (Morton et al 1992). It is considered to be the closest node to a primarytumor or to be the first node visible in lymphoscintigraphy images. This, howev-er, is true only if it actually is the first to receive direct drainage from the injectionsite (Nieweg et al 2001c). The first visible node is therefore not always an SN,because of preferential flow to the other nodes when the SN is occluded. It isnoteworthy that more than one SN can exist in the LS, when two or more lym-phatic channels originate from the primary tumor.

The accurate performance of SNB requires a pilot study phase for the surgeon(Cox et al 2001, Nieweg et al 2001b, McMasters et al 2001, Tanis et al 2002). Inthe initial studies, most surgeons were able to identify the SN in over 90% of the


patients. In these studies, a confirmative ALND was performed. The false-nega-tive rate of SNB, that is, the proportion of patients with a negative SN withsubsequently proven axillary metastases has ranged from 0 to 29% of all node-positive patients (Alex and Krag 1996, Albertini et al 1996, Giuliano et al 1997,Borgstein et al 1997, Krag et al 1998, Miltenburg et al 1999, Fentiman andMansel 2002). The false-negative rate is the most important parameter in thelearning phase of SNB, not forgetting the identification rate. Reliable perform-ance and quality of SNB requires in the learning phase a sufficient number ofinvolved SNs, which is, according Tanis et al, 150 procedures with 60 involvednodes (Tanis et al 2002). Identification rate for SNs correlates with a decreasingfalse-negative rate and the experience of the surgeon in performing the SNB(Giuliano et al 1994, Giuliano et al 1997). Profound knowledge of the nuancesof the procedure facilitates the performance (Nieweg et al 1999), even more thandoes the frequency of performances (McMasters et al 1998, Nieweg et al 1999,Tanis et al 2002) (Table 6).

In addition, a successful and accurate performance of SNB requires close co-operation between surgeon, nuclear medicine physician, and pathologist. Theaccuracy of the SN procedure is based on lymphoscintigraphy and on biopsy ofthe real SN and its meticulous histological examination (Canavese et al 2000,Tanis et al 2002).

Table 6: Success, accuracy, and false-negative rate by percentage for different lym-phoscintigraphy techniques for performing sentinel biopsy

No of Technique of Success Accuracy FalseAuthor Year patients injection (%) (%) negative

(%)

Giuliano 1994 174 VD 66 96 11Giuliano 1997 107 VD 94 100 0Veronesi 1997 163 SD 98 98 5Alex 1996 70 PT 71 100 0Albertini 1996 62 PT 92 100 0Krag 1998 443 PT 91 96 11Rahusen 2000 115 PT 92 87? 37Borgstein 1997 25 ID 100 100 0Borgstein 1998 130 PT 94 98 2

ID= intradermal, PT= peritumoral, SD= subdermal

VD= vital blue dye

Accuracy= true positive + true negative/total number of patientsFalse negative: false negative/false negative+ true positive


7. Sentinel node biopsy and cost impacts

SNB has currently been introduced into many breast surgery centers as an alterna-tive to ALND for axillary staging in breast cancer. The method is less invasive andprovides a benefit for postoperative recovery and also probably for long-timemorbidity (Schrenk et al 2000, Burak et al 2002, Sener et al 2001, Leidenius et al2003a, Haid et al 2002, Miguel et al 2001). SNB is expected to be a cost-effec-tive procedure, although only a few studies cover this topic (Gemignani et al2000, Chirikos et al 2001, Fenaroli et al 2004).

Estimation of total costs of a procedure is far from simple. Charges for aprocedure include both indirect and direct costs of illness (Leivo 2001). Hospitalcharges usually include hospital-related costs like medication, laboratory tests,lymphoscintigraphy, costs in the operating room including anesthesia, hospitalstay, and pathological examination of the primary tumor and the lymph nodes(Gemignani et al 2000). As regards hospital charges, SNB has one weak link,namely the intraoperative diagnosis, especially the false-negative rate involved(Gemignani et al 2000).

As a consequence of the time-consuming extra workload in a pathology labo-ratory, the intraoperative analysis of SNs increases charges. On the other hand,overall charges may decline, because SNB enables breast surgery, axillary staging,and treatment during the same procedure for the majority of patients (Gemigna-ni et al 2000, Weiser et al 2000, Leidenius et al 2004). SNB does not significantlyraise hospital-related charges in stage ™ breast cancer, while only a few patients atthat stage need ALND as a second operation (Gemignani et al 2000). Further-more, from the economic point of view, for T1 tumors, the frozen section anal-ysis of SNs may not be worthwhile, because lymph node involvement in thisgroup is rare (Weiser et al 2000, Fenaroli et al 2004).

8. Sentinel node biopsy and surgery of nonpalpablebreast cancer

The proportion of nonpalpable breast lesions has increased during recent yearsdue to widespread use of mammographic screening and early diagnosis of breastcancer (McLelland 1991, Luini et al 1999, Harlow et al 1999, Gray et al 2001,Paganelli et al 2002). The most important factor in surgery for non-palpablelesion is exact localization and the excision of the tumor with adequate margins(Feggi et al 2001).

Localization of nonpalpable tumors is most commonly done by a stereotacti-cally placed guide wire (Gray et al 2001). This method has many disadvantages;the insertion of the wire is uncomfortable for the patient, displacement of a wireoccurs easily, and transsection of a wire is possible. Nonpalpable tumors are alsopossible to localize by intraoperative ultrasound (Rissanen et al 1994, Rahusen etal 1999, Harlow et al 1999)


The nonpalpable tumors can be localized by injection of a radioactive seedinto a tumor with ultrasonic (US) or mammography guidance (Gray et al, 2001).Accordingly, Gray et al (2001) found this method to be as feasible in localizationand retrieval of a nonpalpable lesion as the standard wire-guided method. Fur-thermore, with their seed localization method, they observed reduced incidenceof histologically involved margins.

A method called radioguided occult lesion localization (ROLL) has been usedfor the localization of non-palpable tumors since 1999 (Zurrida et al 1998, Feg-gi et al 2001). Different techniques and combinations of radiopharmaceuticalscompounded with a colloid have been used in intratumoral and often also super-ficial tracer administration (Zurrida et al 1998, Luini et al 1999, Gennari et al2000, Feggi et al 2001, Tanis et al 2001, Barros et al 2002). After a singleintratumoral tracer injection, performance of simultaneous LS, SNB, and ROLLis possible (Feggi et al 2001) (Table 7).

Table 7: Success rates in localization of nonpalpable lesions, visualization of sen-tinel nodes in lymphoscintigraphy, and radicality of excision by radioguided oc-cult lesion localization

Author Patients/ Localized Surgery Visualized VisualizedInjection lesions sufficiently sentinel extra-

radical nodes axillarysentinelnodes

(%) (%) (%) (%)

Luini et al,1999 331 /IT 100 99.1 – –

Gennari et al,2000 647 /IT 99 99.5 – –

Feggi et al,2001 73/IT + SF 100 94.5 97.3 15.0

Tanis et al,2001 56/ IT 100 92.3 93.0 43.0

Gallegos Hernandezet al, 2004 65/ IT 100 83.0 – –

IT=intratumorally, SF = superficially


9. Sentinel node biopsy and morbidity after breastcancer surgery

The breakthrough of breast conserving therapy (BCT) with improved survivalhas shifted the focus to postoperative morbidity after breast cancer surgery. Eventhough these symptoms are not life threatening, they occur over a prolongedperiod postoperatively and reduce significantly the quality of life in some breastcancer patients (Voogd et al 2003, Bland & Copeland 2004).

A common and well-known postoperative complication after ALND is mor-bidity of the upper extremity (Petrek et al 1990, Pezner et al 1986) includinglymphedema of the breast and upper extremity, as well as sensory and functionaldisorders like numbness and motion restriction of the upper arm. These symp-toms manifest among 20% to 80% women who have undergone AC and radio-therapy (RT), (Tasmuth et al 1996, Recht et al 2001, Ververs et al 2001, Gos-selink et al 2003, Yap et al 2003). Risk for breast and upper arm lymphedemaseems to be higher after extensive axillary surgery (Clarke et al 1982, McCormicket al 1989, Senofsky 1991, Temple et al 2002) and postoperative RT (Clarke et al1982, McCormick et al 1989, Senofsky 1991, Isaksson and Feuk 2000). Most ofthese symptoms manifest immediately or up to one year after surgery, but symp-toms can appear until 2 to 5 years after ALND (Warmuth et al 1998).

A decreased risk was expected for early and long-time morbidity after the in-troduction of SNB without further axillary treatment. The first studies have re-ported significantly less arm morbidity after SNB than after AC (Schrenk et al2000, Miguel et al 2001, Sener et al 2001, Burak et al 2002, Ernst et al 2002,Haid et al 2002, Temple et al 2002, Blanchard et al 2003, Gosselink et al 2003,Leidenius et al 2003 a, Peintinger et al 2003, Schijven et al 2003). Furthermore,immediate postoperative recovery is faster after sole SNB than after AC (Buraket al 2002, Leidenius et al 2003, Peintinger et al 2003). SNB without furtheraxillary treatment has, however, also been associated with postoperative morbid-ity in these studies, but the clinical significance of any SNB-related morbidity isstill obscure.

9.1 Lymphedema

Lymphedema is an accumulative process, due to an imbalance between capillaryfiltration and lymph drainage (Mortimer et al 1998, Zippel et al 2003). In breastcancer, the disturbance of this balance occurs by interruption of the axillary lym-phatic system by surgery or RT, leading to dysfunction, increased pressure, valvedysfunction in lymphatic vessels, and dermal back flow, as well as accumulation ofprotein-rich fluid in subcutaneous tissue (Mortimer 1998, Erickson et al 2001).Edema may develop immediately after surgery or in a later postoperative phase(Recht et al 2001).


9.2 Upper arm lymphedema

Postoperative lymphedema in the upper extremity is the most well-known andfeared complication after breast cancer surgery. Incidence of cancer treatment-related lymphedema in the upper arm may vary from 27% to 45% during the firstpostoperative year, depending on operative technique – mastectomy or resection(Tasmuth et al 1996). Tasmuth et al (1996) also revealed that upper arm edemais more often detected than self-reported. The risk for upper arm edema is mini-mal if the axilla is neither dissected nor irradiated, whereas the risk may increaseup to 27%, when one or the other treatment modality is added (Senofsky et al1991, Meek 1998). Immediate or late postoperative arm edema occurs in 10 to56% of patients who have undergone ALND (Gerber et al 1998, Schrenk et al2000, Duff et al 2001, Sener et al 2001,Golshan et al 2003, Voogd et al 2003).Combination of axillary RT and ALND increases the risk and incidence of upperarm edema (Erickson et al 2001, Yap et al 2003). The other contributory factorsfor lymphedema with surgery and irradiation are advanced stage at diagnosis,obesity, and older age (Pezner et al 1986, Meek et al 1998, Deutsch and Flick-inger 2003).

Objective measurement of lymphedema is difficult. A water displacement tech-nique was used in initial measurements of arm lymphedema (Engler and Sweat1962, Stranden et al 1981), and an optioelectronic volumetry and dielectricmethod (Nuutinen et al 1998, Duff et al 2001). The most common method ismeasurement of the arm circumference and comparison of results between theoperated and non-operated sites.

9.3 Breast lymphedema

Breast lymphedema is considered a minor problem among postoperative compli-cations; perhaps one reason that it has not been studied extensively (Clarke et al1982, McCormick et al 1989, Senofsky et al 1991). Mild symptoms may lead toa most feared complication called delayed breast cellulitis, with an estimated inci-dence of 3% to 5% (Zippel et al 2003). This may be difficult to distinguish froman infection or underlying neoplastic process (Zippel et al 2003). Diagnosis ofbreast lymphedema is complicated, due to lack of standardized measuring andreporting criteria (Petrek et al 1996). Diagnosis is often a combination of clinicalfindings and patients’ complaints (Clarke et al 1982, Rockson et al 1998).

The incidence of breast edema is associated with the extent of axillary surgeryand postoperative RT (Clarke et al 1982, Senofsky et al 1991). It ranges accord-ing to different publications from 5% to 24% with limited axillary dissection upto 15% to 80% when a full ALND is performed (Clarke et al 1982, Senofsky et al1991, Meek 1998). A minimal risk (6%) for breast edema is reported in patientswith breast resection without RT (Senofsky et al 1991, Recht and Houlihan1995).


9.4 Functional disorders of the upper extremity

Postoperative functional disorders in breast cancer patients are associated withextent of axillary treatment (Erickson et al 2001, Ernst et al 2002) and axillaryRT (Sugden et al 1998, Gosselink et al 2003). Furthermore, the type of breastoperation – mastectomy or resection – can influence shoulder motion. Upperarm morbidity and motion restriction manifests more often after mastectomythan after BCT (Sugden et al 1998, Ernst et al 2002, Gosselink et al 2003),usually immediately after surgery and during the following 6 postoperative months(Duff et al 2001). Motion restriction is associated with ALND and postoperativeRT, appearing in 73% of patients with RT and in 35% of patients without (Sug-den et al 1998, Gosselink et al 2003).

9.5 Sensory disorders of the upper extremity

Results of most studies of sensory disorders are based on the patient’s self-experi-ence, using either a validated or non-validated questionnaire as the measuringinstrument (Tasmuth et al 1996, Warmuth et al 1998, Schrenk et al 2001, Buraket al 2002, Temple et al 2002). In most studies, sensory disorders are describedby the terms numbness, tenderness, or pain (Schrenk et al 2001, Burak et al2002, Haid et al 2002).

Sensory disorders like numbness most often manifest immediately after sur-gery and improve significantly during the first 3 months after surgery (Temple etal 2002). Prolonged, mild sensory changes have been evident even 5 years afterALND in 55% to 72% of patients (Warmuth et al 1998, Haid et al 2002).

Postmastectomy pain syndrome is also related to breast cancer treatment, af-fecting 30% to 60% of the patients (Tasmuth et al 2003). The etiology of thissyndrome is supposed to be injury of the intercostobrachial nerve or brachialplexus, or both (Freeman et al 2003). It is disabling, affecting daily- life functionof the upper arm (Tasmuth et al 2003). According to Miguel and co-workers’study (2001), referrals to their pain clinic decreased between 1991 and 1998,suggesting that the increasing number of SNBs have reduced the incidence ofsurgery-related pain syndrome.

Pain often presents in combination with other sensory disorders like numb-ness. Preservation of the intercostobrachial nerve may prevent sensory deficit andreduce long-term symptoms after AC (Freeman et al 2003).

25HYPOTHESIS OF THE STUDY

Hypothesis of the Study

1. SNB facilitates nodal staging in breast cancer.2. Postoperative morbidity after SNB is minimal.3. Simultaneous lymphatic mapping and SNB can be combined into a simple

localization method for nonpalpable breast tumors.4. In breast cancer staging, SNB is thus a cost-effective method.

26 PATIENTS AND METHODS

Patients and methods

The study was carried out at the Breast Surgery Unit of Helsinki University Hospi-tal between 30 May 2000 and 28 February 2002. The study included 426 consec-utive breast cancer patients with clinical T1-T2 stage axillary node-negative cancerundergoing SNB during the study period. In addition, 22 breast cancer patientsundergoing diagnostic ALND were included (Study IV) (Table 1).

Table 1: Number of study patients and exclusion criteria

No Name of Number of Number in Number Exclusionstudy patients groups excluded criteria

I Sentinelnodesoutside 170 – – None

level I–II ofaxilla

II Radioguided ROLL 64 Prior excisionaloccult lesion 231 PALP 137 16 biopsylocalization WGR 14

III Impact ofSNB onhospital

237 – – None

costs

IV Breast SNB 57lymphedema 160 AC– 57 – Noneafter BCT AC+ 46

V One year SNB 43 Preoperativemorbidity 109 AC 40 26 examinationafter SNB impossible

scheduling

AC– = patients in axillary clearance group, uninvolved nodesAC+ = patients in axillary clearance group, involved nodesAC = patients with axillary clearancePALP =patients with palpable tumorROLL = patients with radioguided occult lesion localizationSNB = sentinel node biopsyWGR = patients with wire-guided resection

27PATIENTS AND METHODS

Lymphoscintigraphy

Lymphoscintigraphy was performed the day before surgery, 4 hours after an in-tratumoral injection of 80 to 100 MBq 99m Tc-labeled human albumin colloidNanocoll® (particle size <80 nm: Nanocoll®, Sorin Biomedica diagnostics, Salug-gia, Italy) or Albu-Res® (particle size 200-3000 nm, ALBU-RES® 99m Tc albu-min microcolloid, Nycomed Amersham Sorin s.r.l.) in a volume of 0.2 ml.

The radioactive tracer was injected intratumorally under palpation control bya nuclear medicine physician when the tumor was palpable. In cases with nonpal-pable tumor, the tracer was injected under US control or by use of a stereotacti-cally placed guide wire. In patients with nonpalpable tumors (Study II) the sin-gle intratumoral injection of the radioactive tracer for LS was the only tumor-localization method and was used also for localization of the SN.

Anterior and lateral planar views were obtained with a gamma camera. Esti-mated locations of sentinel nodes were marked on the skin with a permanent penfor both projections.

Sentinel node biopsy

At least 5 minutes prior to incision, 1 ml of Patent Blue dye was injected intratu-morally (Bleu Patenté V; Laboratoire Geuerbet, Aulnay-sous-Bois, France). TheSNs were harvested with a hand-held gamma probe (Neoprobe 2000, Johnson &Johnson Medical, Hamburg, Germany or Navigator GPS, Auto Suture EuropeanServices Center, S.A., France) and by searching for blue-stained lymphatic vesselsand nodes. SNs in the axilla were sought on all occasions, and when clearly visiblein LS, attempts were made to retrieve them. AC was performed when the SNs inthe axilla were involved, were not identified, were blue only – without any radio-activity, or if the tumor proved to be multifocal.

Breast surgery

In all patients with breast conserving surgery (BCT), a wide local excision of thetumor was performed aiming at 1- to 2- cm free margins and including theunderlying pectoral fascia and most often a slice of overlying skin (Aspegren et al1988). In Study II, the nonpalpable tumors visible in US (ROLL group) wereexcised guided by a gamma detector searching for the area of maximal radioactiv-ity, which showed the center of the tumor, or excised with the guidance of astereotactically placed guide wire (WGR group). In a case of a nonpalpable tu-mor, a specimen X-ray ensured radiological free margins.

During primary surgery, if the tumor proved not to be suitable for BCT becauseof multifocality or larger size than evaluated preoperatively, the operation was con-verted to mastectomy. Tumor multifocality was confirmed by intraoperative frozensection diagnosis, and permission for conversion had been obtained preoperatively.


When the margins were involved or close (< 3 mm), in agreement with the patient,a second operation was performed, either mastectomy or re-resection.

Adjuvant therapy

Patients with BCS received postoperative RT to the breast area, in addition to theaxillary, supraclavicular, and internal mammary fields, if axillary nodes were in-volved. The patients received adjuvant chemotherapy and hormonal therapy ac-cording to the guidelines of the Department of Oncology, Helsinki UniversityCentral Hospital (Tables 2 and 3).

Table 2: Adjuvant treatment of 160 breast cancer patients with sole SNB or withAC with uninvolved or with involved nodes (Study IV)

Adjuvant SNB Patients with Patients withtreatment AC and tumor- AC and tumor-

negative nodes positive nodesN=57 N=57 N=46

Radiotherapy 55 (96%) 56 (98%) 46 (100%)Chemotherapy 8 (14%) 7 (12%) 20 (57%)

AC= axillary clearanceSNB= sentinel node biopsy

Table 3: Axillary nodal status, breast surgery, and adjuvant therapy for the 83breast cancer patients with sole SNB, and patients with axillary clearance (Study V)

Event SNB ACN=43 N=40

RadiotherapyNone 7(16%) 1 (3%)Breast only 35 (82%) 5 (14%)Breast, axillary, and supraclavicular fields 1 (2%) 34 (83%)

Adjuvant systemic therapyNone 22 (51%) 3 (7%)Cytotoxic treatment 3 (7%) 2 (5%)Tamoxifen 15 (35%) 11 (28%)Cytotoxic treatment and Tamoxifen 3 (7%) 24 (60%)

Breast operationResection 35 (81%) 28 (70%)Mastectomy 8 (19%) 12 (30%)

Axillary nodal statusN1 1 (2%) 33 (82%)


Histological methods

Sentinel lymph nodes (Studies I–IV)

SNs were sent to the pathology laboratory as separate samples labeled with theirsites of origin. The fresh specimens were cleaned of all extra capsular fat tissue,measured, sliced into 1 to 1.5 mm thick sections, and mounted on iced OCT®.Touch preparations and frozen sections from two levels were made from theseslices; these were then stained with toluidine blue and viewed. All remaining tissuewas fixed in phosphate-buffered 10% formalin and embedded in paraffin, and sec-tions were stained with H&E and with Cam 5.2 immunostain. When no frozensection diagnosis was required (sentinel nodes outside the axilla), the nodes werefixed directly in phosphate-buffered 10% formalin. After fixation, the nodes werecleaned of fat, cut, and totally embedded. H&E sections were made from two levelsof each lymph node, and cytokeratin immunostaining was done from level 1.

Breast specimens (Study II)

All resected specimens were sent to the pathology laboratory, pinned on a styroxboard with a sketch of the appropriate breast. In cases of nonpalpable tumors,the radiographic edges of the tumors were marked with needles placed in connec-tion with a specimen X-ray. The specimen was painted with commercial vital dyes(WAK-CHEMIE MEDICAL GMBH, Bad Soden, Germany), always in the samefashion, yellow on the anterior surface of the palpable tumor and between themarking needles in the specimens with nonpalpable tumors. The medial marginof the specimen was marked red, the lateral green, the cranial black, and the fasciablue. The specimen was detached from the styrox board, weighed, and cut fron-tally. If no tumor was visible in either of the two sections, they were cut frontallyonce more, and so on, yielding less than 1-cm-thick slices. These slices were kepton bits of paper and numbered. The greatest diameter and the distance to thelateral margins were measured with a ruler for all tumors observed in the speci-men. After 1- to 2- days’ fixation in 10% formalin, representative areas were se-lected for histological examination. These included one center section contain-ing the tumor at its greatest diameter in gross examination, and at least theclosest lateral margin (most often all the lateral margins), as well as crosswisesections from the bottom and the surface containing the anterior and posteriormargins. During the microscopic examination, the tumor margins were ink-markedon the slides; the vital dye marks were often reinforced on the glass with ink. Allmargins were re-measured from the histological sections.

When, during the primary operation or afterwards, the resected specimen,proved to be multifocal, the closest free margin of the resected specimen wasdetermined based on tumor closest to the edge of the specimen, irrespective ofwhether it was the index tumor diagnosed preoperatively or an additional focusdetected during the primary operation or in the definite histology.


In order to estimate whether the tumor had been excised without excessivemass of healthy tissue, we related the size of the specimen to the histologicaltumor size; the tumor-specimen ratio was evaluated by subtracting the histolog-ical tumor size (the greatest dimension) from the maximum length and width ofthe specimen. This tumor-specimen ratio was calculated by use of the histologi-cal size of the index tumor diagnosed preoperatively also in cases with a multifo-cal carcinoma.

Hospital costs (Study III)The sequence of the staging and treatment procedures was registered for 237consecutive patients (Study III). Axillary staging for the 237 patients was per-formed by SNB with intraoperative frozen section diagnosis of SN metastases. Inaddition, three hypothetical alternative treatment protocols were developed foreach of the 237 patients: diagnostic ALND, SNB as day-case surgery prior to thebreast operation, or SNB without frozen-section diagnosis. Thereafter, the hypo-thetical sequence of the staging and treatment procedures for each patient wasestimated, first, application of diagnostic ALND for axillary staging, second, ifSNB as day-case surgery prior to the breast operation had been applied, andthird, application of SNB without frozen section diagnosis. Hospital costs werecalculated based on the real or hypothetical sequences of surgical staging andtreatment procedures for each axillary staging alternative.

For calculation of costs, we used unit costs based on the detailed internalaccounting system of the Helsinki University Hospital. These were estimatedaccording to the production costs of the various services and procedures in 2001.The costing included hospital inpatient care, outpatient visits, LS, surgery, andpathological analysis. The cost of hospital inpatient care comprises room andboard, medication, blood products, and laboratory costs, as well as pre- andpostoperative nursing care. The cost of an outpatient visit comprises pre- andpostoperative check-ups. The cost of pathology analysis includes intra-operativefrozen section diagnosis of SNs, and postoperative histological examinations ofSNs and breast specimens, as well as the ALND specimens in patients with AC.The costs of surgery comprise all costs arising from the operation including an-esthesia and the recovery room.

Patients were admitted to the hospital on the day before surgery. On average,they were discharged on the first day after breast surgery and SNB and on thesecond day after AC. These patients left with a drainage tube, which was removedas an outpatient procedure on the fifth day after surgery.

Clinical breast examination (Study IV)Clinical examination of the operated and the contralateral breast was performed amedian of 12.6 (11.3–18.8) months after the operation by specialist surgeons or


surgical residents. Data were collected from case report forms (CRFs). The size,edema, tenderness, and pigmentation of the breast and the condition of the skinin the scar area and in the whole breast were recorded.

Radiological methods (Study IV)

In order to evaluate the breast edema, an experienced breast radiologist performedUS on the breasts after the clinical follow-up visit blind to the exact surgicalprocedure performed. Patients from remote residential areas were excluded fromthe extra US examination. The US examinations were performed with a real-timeUS unit having a linear probe of 5 to 13 MHz focused upon the area of interest.The total skin thickness of the four quadrants of the breasts was recorded bothfor the operated breast and the contralateral breast, as was the presence of anyinterstitial edema and fluid collection.

The thickness of the breast skin varies between 1 and 2 mm, with a meanthickness of 1.7 mm (Nuutinen et al 1998). The skin complex is comprised oftwo thin echogenic lines with a hypoechoic dermis between them (Mendelson1992).

Thickening of the skin of over 2 mm with increased echogenicity, distur-bance, or poor visibility of the deeper echogenic line and interstitial fluid accu-mulations were considered edema. Both interstitial fluid and fluid accumulationis considered as varying degrees of lymphedema, and in this study both wereregistered as lymphedema.

Clinical arm examination (Study V)

The range of shoulder motion, circumference of the upper extremity at six levels,and skin sensation in the innervation area of the intercostobrachial nerve weremeasured preoperatively, as well as at 2 weeks and 3, 6, and 12 months postoper-atively.

Flexion, abduction, and external and internal rotation were measured by agoniometer, with reasons for limited movement registered. A reduction of 10°or more in the range of shoulder motion compared with the preoperative meas-urement was considered as motion restriction.

The circumference of the upper extremity was measured with a tape measurearound the ring finger, the wrist, and 10 cm and 20 cm below the lateral epi-condyle, as well as 10 cm and 20 cm above the lateral epicondyle.

For evaluating lymphedema, in each patient, the differences between the pre-and postoperative circumference were compared between the ipsi- and contralat-eral upper extremities at each measurement point and expressed as percentages ofthe preoperative circumference of the ipsilateral side by the formula:

(PoI-PreI) – (PoC-PreC) × 100%PreI


where PoI is the postoperative and PreI the preoperative circumference of theipsilateral upper extremity, and PoC and PreC are the post- and preoperativecircumference of the contralateral arm, correspondingly. Increase in arm circum-ference (LE%) was defined as the mean difference at the six measurement points.

Skin sensation was measured on the medial aspect of the upper arm, the axilla,and the breast area with von Frey filament stimulation. Difference in skin sensa-tion more than the thickness of one filament was considered impaired sensation.

Self-reported morbidity (Study V)

Self-experienced morbidity one year after surgery in regards to lymphedema andsensory and functional disorders and their interference with daily life was evaluat-ed by a questionnaire. Prevalence and intensity of symptoms such as pain, lymph-edema, strange sensations, numbness, muscle weakness, frozen-shoulder-like symp-toms, and shoulder stiffness were measured on a visual analogy scale (VAS), (10cm scale, 0= no symptoms, 10 cm = worst possible). The VAS scale result in allthese symptoms was considered as follows: <20 mm mild, >20–40 mm moder-ate, and >40 mm severe (Huskisson 1974, McQuay and moore 1998).

Statistical analysis

The significance of difference between the study groups was tested by the two-tailed Fisher’s exact test (Studies I, II, IV, V). The medians were compared usingthe non-parametric test, Mann-Whitney-U. Cross tabulations were used in com-paring the groups, and differences were tested by the Chi Square- and Fisher’sexact tests. P-values less than 0.05 in two-sided tests were considered statisticallysignificant.

The statistical analysis was performed by a computer system (SPSS, Inc., Chi-cago IL, USA), (Studies IV–V). Calculations in the study concerning hospitalcost (Study III) were carried out by a computer-based decision tree model builton DATA 3.5 (Tree Age Software Inc.).

33RESULTS

Results

Prevalence of SLN and SLN metastases outside level I–IIof the axillaLS showed SNs in the axilla in 150 (88%) of the 170 cases and outside level I toII of the axilla in 30 (17%) cases. SNs in the internal mammary basin were visual-ized in 22 (13%) patients, and drainage solely outside the axilla was visible in two(1%) patients. Lymphatic mapping failed to visualize SNs in 20 (12%) cases.

In 30 patients, a total of 55 SNs were visualized outside the axilla in the LS.These included 38 internal mammary, 9 subclavicular, and 7 intramammary nodes,and one interpectoral node.

Patients with SNs outside the axilla were somewhat younger, with a medianage of 51 years (range 26–81), than those were without such nodes, with theirmedian age of 57 (range 28–86) (p=0.02). No differences existed in BMI, later-ality of the breast cancer, tumor size, grade, histology, and particle size of thetracer, or in the rate of axillary metastases between cases with or without SNsoutside the axilla. Lymphatic drainage to SNs in the IMN occurred from bothmediocentral and lateral tumors (Table 4).

Table 4: Tumor location and incidence of sentinel nodes in internal mammarybasin, axillary metastases, and metastases in internal mammary sentinel nodes in172 breast cancer cases (Study I)

Event Upper Lower Upper Lower CentralLateral Lateral Medial Medial

(N=95) (N=26) (N=33) (N=7) (N=11)

Sentinel nodes ininternal mammary 10(11%) 7(2%) 2(9%) 2(29%) 1(5%)basinAxillary metastases 31(33%) 13(50%) 12(36%) 0 (0%) 3 (27%)Metastases in internalmammary basin 0 (0%) 2 (8%) 2 (7%) 0 (0%) 0 (0%)

The median number of harvested SNs in the axilla was 2 (1–15). Nodes identifiedin the axilla were both blue and radioactive in 100 (58%) cases, only blue in 5(3%) and only radioactive in 46 (27%). Neither blue nor radioactive nodes ap-peared in the axilla in 21 (12%) patients.

SNs outside the axilla were successfully harvested in 27 of 30 (90%) of thecases, with 43 (78%) retrieved of the 55 SNs visualized outside the axilla in the

34 RESULTS

LS. The success rate in harvesting the SNs in the internal mammary chain was86% (33 of 38).

An additional skin incision was necessary in 15 patients (79%) of those withBCS and in none of those who underwent mastectomy during harvesting of theSNs outside the axilla. A minimal perforation of the parietal pleura occurred inthree (10%) patients, who all recovered uneventfully without pleural drainage.

Axillary metastases were found in 59 (34%) of all 170 cases and in 12 (40%) ofthe 30 cases with SNs outside the axilla. In the axilla, SNs were the only metastat-ic nodes in 65% of the cases. Five patients (3% of all cases, 17% of those with SNsoutside the axilla) had metastases in the SNs outside the axilla, and two of thesefive patients had no axillary metastases.

Radioguided surgery of nonpalpable tumor (Study II)Localization of the tumor was successful in all the 215 cases, and excision of thetumor was complete in 200 of 215 (93%) after the primary operation. Free mar-gins of at least 10 mm were achieved in 166 of 215 (77%) in the primary opera-tion without statistically significant differences between patient groups. The medianlength of the closest margin was 15 mm (range 0–40 mm), without differencesbetween the ROLL, WGR, or palpation groups. The proportion of cases withinvolved margins was highest in the WGR group (29%) as was excessively widemargins (closest margin over 20 mm), 29% (Table 5).

Table 5: Median histological tumor size and free tissue margins after resectionwith radioguided occult lesion location (ROLL), wire-guided resection (WGR),or palpable tumors resection

Closest margin ROLL WGR Palpable tumorsN=64 N=14 N=137

Length (median, mm range,) 12 (0–35) 10 (0–30) 15 (0–40)10 mm or more 44 (69%) 8 (57%) 114 (83%)Involved * 3 (5%) 4 (29%) 8 (6%)Close or involved (0–3 mm) # 5 (8%) 4 (29%) 9 (7%)Excessively wide (>20 mm) 9 (14%) 4 (29%) 32 (23%)

* p= 0,03 between ROLL and WGR and between WGR and palpable tumors# p<0,05 between WGR and palpable tumors

Median size of the resected specimen was 8 (SD±2.9) × 7 (SD±2.2) cm, with thesize largest in the WGR group, although median histological tumor size wassmallest. However, the operation technique (resection with ROLL, WGR, or inpalpation did not influence significantly the tumor-specimen ratio (maximumlength of the specimen – histological tumor size or maximum width of the spec-imen – histological tumor size, or both) (Table 6).

35RESULTS

Table 6: Median size of the resected specimen in centimeters, length and widthof the resected specimen subtracted from the tumor and the tumor-specimenratio after radioguided occult lesion location (ROLL), wire-guided resection(WGR), or palpable tumors resection

ROLL Palpable tumors WGRSize n=64 n=137 n=14

Length of specimen * 8.0 (4.5–15.0) 8.0 (1.8–25.0) 9.75 (7.5–25.0)Length – tumor size 6.9 (1.9–13.9) 6.35 (0–22.6) 7.55 (2.6–23.0)Width of specimen # 6.0 (3.3–14.0) 7.0 (1.5–13.0) 7.5 (6.8–20.0)Width – tumor size 5.3 (0.9–12.9) 4.95 (0–10.6) 6.4 (0–18.0)

Median (range)*p<0.01between ROLL and WGR and between palpable tumors and WGR

#p< 0.05 between ROLL and WGR and between palpable tumors and WGR

An additional excision to ensure free tissue margins was performed on 52 of the215 (24%) during the primary operation, based on palpation of the specimen(palpable tumors) or on findings in the specimen X-ray (nonpalpable tumors).An additional excision was performed equally often in all patient groups. Thehistological finding in the resected additional tissue was benign in 70%. None ofthese 52 patients needed re-operations due to insufficient margins.

The operation was converted to mastectomy during the primary operation in13 (6%) of all patients, a conversion was equally common in all patient groups. Asecond operation (mastectomy or re-excision) was considered necessary due tomultifocal carcinoma or close or involved margins in 17 (8%). The second oper-ation was required most often in the WGR group. The overall mastectomy rateduring the primary or as a second operation was 11% (24 of 215), with no differ-ence between patient groups (Table 7).

Table 7: Additional procedures during primary surgery or second operation in pa-tients with palpable and nonpalpable tumors undergoing breast conserving surgery

Event Radioguided occult Wire-guided Palpablelesion location (ROLL) resection (WGR) tumors (P)

N=64(%) N=14(%) N=137(%)

Extra excision during 14 (22) 4 (23) 34 (25)primary operationRe-operation * 7 (11) 3 (21) 5 (4)

Mastectomy during 5 (8) 1 (7) 7 (5)primary operationMastectomy, total 10 (16) 3 (21) 11 (8)

*p = 0.05 between WGR and P

36 RESULTS

The median histological tumor size was largest, 15 mm, for palpable tumors,compared to 12 mm in the ROLL group or 9.5 mm in the WGR group (p<0.001 between palpable tumors and WGR, p < 0.005 between ROLL and pal-pable tumors). Ductal carcinoma in situ (DCIS) was encountered most often(21%, 3 of 14) in the WGR group, compared to the ROLL group (4%, 4 of 64)or in patients with palpable tumors (1%, 1 of 137, p<0.005 between WGR andpalpable tumors).

The histology of the resected specimen revealed multifocal carcinoma in 12%(17 of 137) of patients with palpable and in 6% (5 of 78) with nonpalpabletumors (NS). Multifocality seemed to be less common in the ROLL group (5%,3of 64) than in the WGR group (14%, 2 of 14), but without a significant differ-ence. The prevalence of an extensive intraductal component was highest (36%, 5of 14) in the WGR group compared to the ROLL group’s 14% (9 of 64) orcompared to palpable tumors (7%, 9 of 137, p<0.01 between WGR and palpabletumors).

SNB and impact on hospital costsThe hospital costs per patient with SNB and intraoperative frozen section diag-nosis were 3750 €. Costs per patient would have been 3020 € in the ALNDmodel, 4087 € had the frozen section not been applied, and 4573 € in the “SNBas day-case surgery” model. The most expensive axillary staging was thus one anda half time times as expensive as the most economical.

The false-negative rate with the intraoperative frozen section diagnosis was13% (9 of 68) of patients with tumor-positive SN. Based on the sensitivity analy-sis, all economic advantage of intraoperative diagnosis would have been lost witha false-negative rate of 35% or more. In other words, the costs with or withoutfrozen section diagnosis would be equal at a threshold false-negative rate of 35%.

Among these 237 patients, two operations were necessary for the treatmentand axillary staging of breast cancer in 19 (8%). More than one operation wouldalways have been necessary in the “ SNB day-case surgery” model (Table 8).

37RESULTS

Table 8: Treatment procedures in 237 breast cancer patients with sentinel nodebiopsy (SNB) and intraoperative frozen-section diagnosis of sentinel node metas-tases

Treatment PatientsN %

Breast surgeryBreast conserving surgery 185 78Mastectomy 50 21Lumpectomy prior to SNB 2 1

Second operationMastectomy 8 4Axillary clearance as second operation: false-negativefinding in frozen section 9 4Axillary clearance as second operation: multifocalcarcinoma 5 2

Axillary surgerySNB only during primary operation: negative finding infrozen section 144 61AC during primary operation: true-positive finding infrozen section 59 25AC during primary operation: false-positive finding infrozen section 1 <1AC during primary operation: SN unidentified 33 14

Postoperative breast lymphedema in clinical examination(Study IV)In the clinical examination the operated breast was more tumid than the contral-ateral one in 34% (55 of 160) of the patients. Clinical breast edema was mostcommon, in 48% (22 of 46) of the patients with AC and involved axillary nodes,and least common in 23% (13 of 57) with SNB only (p<0.05).

After RT, of the 160 patients, 37% had edematous or thickened skin (orangepeel), erythema, significant pigmentation, or other skin damage. In general, pig-mentation of the skin in the operated breast was a common clinical finding, at aprevalence of 59% (94 of 160), including also the slightest pigmentation. Ab-normal tenderness during breast palpation was experienced by 46% (73 of 160).The prevalence of these signs was not influenced by the extent of axillary surgery(Table 9).

38 RESULTS

Table 9: Findings in the clinical examination one year after breast-conservingtreatment and sole sentinel node biopsy (SNB) or axillary clearance (AC)

Breast finding SNB Patients with Patients withAC and tumor- AC and tumor-negative nodes positive nodes

N=57 N=57 N=46

Edema * 13 (23%) 20 (35%) 22 (48%)

Breast sizesEqual 23 (40%) 21(37%) 23 (50%)Smaller 26 (46%) 28 (49%) 20 (44%)Larger 7 (12%) 8 (14%) 3 (6%)Not applicable 1 (2%)

PigmentationNone 28 (49%) 21 (37%) 16 (35%)Slight 24 (42%) 24 (42%) 18 (39%)Clear 5 (9%) 12 (21%) 11 (24%)Not evaluated 1 (2%)“Orange peel” skin 0 (0%) 4 (7%) 2 (5%)Erythema 1 (2%) 2 (4%) 5 (12%)

Breast firm/compact 7 (12%) 10 (18%) 8 (17%)

Tenderness 21 (37%) 30 (53%) 22 (48%)

Scar area normal 47 (82%) 39 (68%) 38 (83%)

*p= 0.029

Postoperative breast lymphedema in ultrasoundexamination (Study IV)Thickness of the skin on the non-operated breast was constant in every segmentof the breast. On the operated side, thicker skin was evident more often in pa-tients after AC than in patients with SNB only (p=0.01–0.001) (Table 10).

39RESULTS

Table 10: Median skin thickness on the breast measured by ultrasonography oneyear after breast conserving treatment and sole sentinel node biopsy (SNB) oraxillary clearance (AC)

Localization of SNB Patients with Patients withmeasurement AC and tumor- AC and tumor-

negative nodes positive nodes

N=36 N=37 N=29

Non-operated breastBreast SegmentUpper lateral 1.1 (0.7–2.5) 1.1 (0.6–1.6) 1.0 (0.6–1.6)Upper medial 1.2 (0.9–2.9) 1.2 (0.7–1.6) 1.3 (0.9–1.7)Lower lateral 1.3 (0.7–2.1) 1.3 (0.8–1.9) 1.3 (0.8–1.9)Lower medial 1.1 (0.7–2.2) 1.1 (0.7–1.7) 1.1 (0.7–1.6)

Operated breastBreast SegmentUpper lateral * 1.5 (1.0–4.4) 1.9 (0.7–5.5) 1.8 (1.1–6.6)Upper medial # 1.7 (0.9–4.6) 2.5 (1.2–6.5) 2.7 (0.9–6.1)Lower lateral # 1.9 (0.9–5.5) 2.9 (1.4–64) 3.4 (1.1–7.5)Lower medial # 2.1 (1.1–4.6) 3.1 (1.4–6.4) 3.6 (1.1–7.5)

mm (range) *p<0.01 #p<0.001

Subcutaneous edema in the operated breast was encountered after AC in 70% of thepatients and in 28% after SNB alone. Fluid collection in the operated breast wasmore common in the AC group – in 56% – than after SNB only, 17% (Table 11).

Table 11: Prevalence of subcutaneous edema and interstitial fluid collections di-agnosed by ultrasonography one year after breast conserving treatment and solesentinel node biopsy (SNB) or axillary clearance (AC)

Localization of SNB Patients with Patients withmeasurement AC and tumor- AC and tumor-

negative nodes positive nodes

N=36 N=37 N=29

Subcutaneous edema * 10 (28%) 26 (70%) 20 (69%)

Interstitial fluid # 6 (17%) 20 (54%) 17 (59%)*p<0.001 between SNB and AC groups, p<0.001 between SNB and AC with tumor-negative nodes andp<0.001 between SNB and AC positive nodes #p<0.001 between SNB and AC groups, p=0.001 between SNB and AC with tumor-negative, p<0.001between SNB and AC with tumor-positive lymph nodes

40 RESULTS

Postoperative functional disorders (Study V)In the preoperative measurement, the range of at least one of the ipsilateral shouldermotions was restricted at least 10° from its anatomical range in 26 of the 83(31%) patients. Of the 83 patients, 27 (33%) complained of strain in the pectoralmuscle during the preoperative evaluation. Such strain was also the most com-mon reason for postoperative restricted shoulder motion, irrespective of axillarytreatment.

One year after surgery, four (10%) of the patients in the SNB group and nine(22%) in the AC group had impaired flexion of the ipsilateral shoulder (p=ns). Asimilar, non-significant difference in shoulder abduction occurred. Extent of ax-illary treatment had no influence on internal and external rotation of the ipsilat-eral shoulder. The range of shoulder motion of the contralateral shoulder re-mained unchanged during follow-up (Table 12).

Table 12: Incidence of motion restriction in the ipsi- and contralateral shoulders2 weeks and 3, 6, and 12 months after sole sentinel node biopsy (SNB) in 43patients or after axillary clearance (AC) in 40 patients

Upper extremity function Follow-up

2 weeks 3 months 6 months 12 months

FlexionIpsilateral * SNB 16 (37%) 3 (7%) 4 (9%) 4 (10%)

AC 32 (78%) 1 (3%) 1 (3%) 9 (22%)Contralateral SNB 1 (2%) 3 (7%) 3 (7%) 1 (2%)

AC 2 (5%) 0 (0%) 0 (0%) 0 (0%)

AbductionIpsilateral # SNB 19 (44%) 4 (10%) 3 (7%) 3 (7%)


AC 3 (7%) 0 (0%) 0 (0%) 1 (0%)

External rotationIpsilateral SNB 8 (19%) 2 (5%) 4 (9%) 6 (16%)


AC 1 (3%) 0 (0%) 0 (0%) 3 (8%)

Internal rotationIpsilateral SNB 1 (2%) 0 (0%) 1 (2%) 0 (0%)

AC 0 (0%) 0 (0%) 1 (2%) 0 (0%)

Contralateral SNB 0 (0%) 0 (0%) 0 (0%) 0 (0%)AC 0 (0%) 0 (0%) 0 (0%) 0 (0%)

*p<0.001 between SNB and AC 2 weeks after surgery #p<0.01 between SNB and AC 2 weeks after surgery

41RESULTS

Of the 43 SNB-group patients, 8 (19%) of complained of mild or moderate mus-cular weakness in the ipsilateral arm one year after surgery, while patients in the ACgroup more often reported muscle weakness (17 of 40, 42%, p<0.05). Shoulderstiffness and tension neck-like symptoms seemed to bother patients slightly moreoften in the AC group, but differences between the groups were non-significant.

Postoperative breast and upper extremity sensorydisorders (Study V)Impaired skin sensation in the medial aspect of the ipsilateral upper arm, axilla,and breast was more common after AC up to 6 months after surgery (p<0.001).The impaired skin sensation improved in the AC group, and all differences be-tween the groups disappeared during the last 6 follow-up months.

Self-reported numbness in the ipsilateral arm (p<0.001) and breast area(p<0.01) were clearly more common after AC than after SNB alone. After moreextensive axillary treatment 3 SNB group patients reported only mild arm numb-ness, whereas 14 patients complained of mild, three of moderate, and six of severearm numbness (p<0.001). Strange arm sensations were also more common, in25 (62%) AC group patients compared to 8 (19%) SNB group patients (p=0.001).

Altogether 62 (75%) patients complained of at least some pain in the surgicalarea or in the ipsilateral arm. Pain in the arm was more common in the AC group,in 21 (52%) patients compared to 12 (28%) in the SNB group (p<0.05). Eight(22%) AC-group patients complained of daily pain in the arm and the breast area,while in the SNB group, four patients suffered daily pain in the breast (p=ns), andtwo in the arm (p<0.01). Medication to relieve the pain was necessary in 12 to 20%of the patients, but the difference between groups was non-significant (Table 13).

Table 13: Incidence of impaired sensation in 83 breast cancer patients 2 weeksand 3, 6, and 12 months postoperatively after sentinel node biopsy (SNB) oraxillary clearance (AC)

Sensory area Time2 weeks 3 months 6 months 12 months

Medial aspect of upper armSNB 1 (2%) 0 (0%) 0 (0%) 0 (0%)AC 8 (24%) * 5 (15%)# 3 (10%) 0 (0%)

Axilla SNB 3 (7%) 0 (0%) 0 (0%) 1 (2%)AC 16 (48%)# 14 (42%)# 10 (30%)∆ 1 (3%)

Breast SNB 2 (5%) 0 (0%) 0 (0%) 0 (0%)AC 13 (39%)# 10 (32%)¤ 8 (24%)• 3 (9%)

*p<0.01 between SNB and AC groups 2 weeks after surgery#p<0.001 between SNB and AC groups 2 weeks after surgery¤p<0.001 between SNB and AC groups 3 months after surgery∆p<0.001 between SNB and AC groups 6 months after surgery•p<0.01 between SNB and AC groups 6 months after surgery

42 RESULTS

Postoperative lymphedema (Study V)None of the patients after SNB alone was referred to lymph therapy, whereas 15(37%) patients after AC underwent manual lymph drainage, and 3 of them alsowore a compression sleeve (p<0.001). A similar increase in arm circumference(LE%) was observed in both patient groups one year after surgery.

Patients in the AC group more often reported edema in the breast area (23 vs.10, p<0.001) and in the upper extremity (30 vs. 5, p<0.001) than did those inthe SNB group (Table 14).

Table 14: Need for lymph therapy, increase in arm circumference (LE%), and self-reported lymph edema one year after sentinel node biopsy (SNB) alone or axillaryclearance (AC)

Event SNB ACN=43 (%) N= 40 (%)

Lymph therapy *

None 43 (100%) 25 (63%)Manual lymph drainage 0 (0%) 12 (30%)Manual lymph drainageplus compression sleeve 0 (0%) 3 (7%)

Increase in arm circumference (LE%)0–5% 37 (86%) 32 (80%)5–10% 4 (10%) 5 (13%)> 10% 1 (2%) 3 (7%)Missing data 1 (2%) 0 (0%)

Self-reported breast oedema #

None 32 (75%) 17 (42%)Mild 7 (16%) 16 (41%)Moderate 3 (7%) 6 (15%)Severe 0 (0%) 1 (2%)Missing data 1 (2%) 0 (0%)

Self-reported arm oedema *

None 36 (84%) 9 (23%)Mild 4 (9%) 21 (53%)Moderate 1 (3,5%) 7 (1%)Severe 0 (0%) 2 (5%)Missing data 1 (3,5%) 1 (2%)

*p<0.001, #p<0.01

43RESULTS

Interference of postoperative morbidity with daily life(Study V)Arm morbidity affected work (p<0.01), hobbies (p<0.01), and daily life in gen-eral (p<0.01) more often after AC than after SNB. The interference reported wasalways mild after SNB alone. Patients had no difficulties in eating, combing theirhair, brushing teeth, dressing, or fastening a seat belt with the ipsilateral arm,regardless of axillary treatment. Carrying bags on the operated side was difficultfor 11 (28%) patients in the AC group and for one single patient in the SNBgroup (p<0.01). In the AC group, nine patients (23%) and in the SNB groupthree (7%) could not sleep on the ipsilateral side (p=ns).

Work aggravated pain in the ipsilateral upper extremity more often in the ACgroup, in 15 patients (37%), than after SNB, in six (15%) (p<0.05). Work, nee-dlework, carrying items, and pressure from clothing were the most commonpain-inducers. Of the 83 patients, 7 (9%) experienced sleep disturbance due topain (Table 15).

Table 15: Interference of arm morbidity with work, hobbies, and daily life ingeneral one year after sentinel node biopsy (SNB) alone or axillary clearance (AC)

Interference with SNB ACN=43(%) N=40(%)

Daily life in general *


Sleeping on the ipsilateral sideNo 3 (7%) 9 (23%)

Working #


Hobbies #


*p<0.01#p<0.01

44 DISCUSSION

Discussion

Extra-axillary lymph nodesThe anatomical studies of Urban (1978), based on extended radical mastectomyseries, demonstrated that the breast lymphatics drain mostly, in 75% of the cases,directly to the axilla and in 25% to the IMN. After the introduction of LS, IMNdrainage has been seen in up to 28% to 87% of all patients with breast carcinoma(Uren et al 1995, Veronesi et al, Jansen et al 2000, Krag et al 1998, van der Ent1999, Rahusen 2000, Valdés-Olmos et al 2001, Estourgie et al 2003, Estourgieet al 2004) The visualization rate of SN has varied between 75% to 99% in theaxilla and from 8% to 87% outside the axilla, when intra- or peritumoral injectiontechniques were used (Borgstein et al 1998, Krag et al 1998, O’Hea et al 1998,Haigh et al 2000, Jansen et al 2000, van der Ent et al 2001, Uren et al 2001,Valdés-Olmos et al 2001, Estourgie et al 2003, Estourgie et al 2004). In recentas in earlier studies, the SNs have been located solely outside the axilla in only afew cases (Krag et al 1998, Jansen et al 2000, van der Ent et al 2001).

The particle sizes of radionuclides and doses and volumes of the tracers have amarked effect on the visualization of SNs and especially regarding SNs outside theaxilla (Valdés-Olmos et al 2001, Leppänen et al 2002). Furthermore, lymph flow isinfluenced by individual factors like physical exercise, medication, and the status ofhydration of the patients; these may influence the visualization of SNs in the LS(Borgstein et al 1998, Krag et al 1998 Jansen et al 2000, Nieweg et al 2001b).

The reported prevalence of IMN metastases in breast cancer patients has been17% to 25% (Jansen et al 2000, Galimberti et al 2002, Estourgie et al 2004).Metastases in the internal mammary chain associate strongly with axillary metas-tasis and large tumor size, and IMN metastases alone are present in 9% to 14% ofthe patients (Morrow 1981, Veronesi et al 1985, Lacour et al 1983, Noguchi etal 1991, Galimberti et al 2002, Estourgie et al 2004). The high prevalence ofIMN metastases in reports from the late 70´s and early 80´s (Veronesi et al 1985,Lacour et al 1983, Donegan et al1977) probably does not represent the situationnowadays, when patients undergoing lymphatic mapping and SNB include ahigh proportion at stage T1 without axillary involvement. Even the introduc-tion of and increasingly routine lymphatic mapping and SNB have broadenedprocedures towards IMN harvesting and better nodal staging (Galimberti et al2002, Estourgie et al 2003, Estourgie et al 2004).

Before the introduction of SNB, the investigators considered the extra axillarynodes to be present in the first two intercostal spaces (Donegan et al 1977, Noguchiet al 1991) and thus biopsied nodes only from the first two intercostal spaces. Ina multicentre non-sentinel study by Lacour and co-workers, the number of IMNexamined was relatively small, the median number being only 1.9 to 4.1 (Lacouret al 1983).

45DISCUSSION

Our study, like the study of van der Ent (2001,) confirmed that the mostcommon site of lymphatic drainage in the internal mammary chain was the thirdintercostal space.

Findings on the location of IMN metastases by Noguchi and co-workers (1991)and on lymphatic drainage patterns to the internal mammary chain in both thestudy of van der Ent and co-workers (2001) and the present study are contradic-tory. This, as well as the limited number of IMN examined earlier (Donegan1977, Lacour et al 1983) opens the speculation the prevalence and the locationof internal mammary metastases in breast cancer patients.

Reappraisal of IMN metastases as a prognostic factor and the findings of re-cent publications of trials evaluating the value of postmastectomy radiation in-cluding internal mammary field radiation have initiated discussion concerningadoption of routine elective IMN irradiation. A small survival benefit has beengained by including the IMN in the RT field in patients with mediocentrallylocated tumors and axillary metastases (Overgaard et al 1997, Ragaz et al 1997,Freedman et al 2000). It should be remembered, based on ours and previousstudies, that lymphatic drainage or lymph node metastases can occur irrespectiveof tumor location; the location of the tumor thus cannot be the main indicationin decision-making as to IMN-field RT (Donegan et al 1977, Lacour et al 1983,Veronesi et al 1985, Borgstein et al 1998, Uren et al 1995, van der Ent et al2001, Estourgie et al 2004).

Furthermore, the nonpalpable tumors, regardless of quadrant, more frequentlyhave their drainage towards the IMNs (Estourgie et al 2004). If all patients withaxillary metastases, regardless of tumor size or location, receive RT to the axilla,the subclavicular, and the internal mammary fields, the proportion of patientswith IMN metastases not receiving RT to the IMN-field is probably negligible(Recht et al 2001). On the other hand, by this treatment strategy, a substantialproportion of patients will receive needless extensive radiation and become pre-disposed to the risks from irradiation such as cardiac toxicity (Freedman et al2000). The introduction of LS and SNB has provided better tools for the diag-nosis of regional lymph node status and maybe also tools for decision-makingregarding adjuvant treatments. Methods for more accurately targeting the IMN-field RT are therefore urgently needed.

Harvesting of the SN outside the axilla is more demanding, with a success rateof 78% in the present study and 63% to 73% in previous studies (Jansen et al2000, van der Ent et al 2001), compared to SNB in the axilla with a practically100% retrieval rate, when the nodes are visualized in LS. Insufficient uptake ofthe tracer and/or difficult anatomic circumstances cause most of the harvestingfailures (Jansen et al 2000). Severe complications, however, seem to be rare (Jansenet al 2000, Galimberti et al 2002).

Improved nodal staging includes also the status of SNs outside the axilla.Surgeons, radiotherapists, and medical oncologists should consider the implica-tions of the presence or absence of metastatic disease in the nodes. Knowing thestatus of these extra-axillary SNs will lead to better selection of those who will

46 DISCUSSION

benefit from adjuvant systemic treatment and maybe also postoperative RT to theIMN chain.

Radioguided occult lesion localizationIn breast conserving surgery, estimation of the free tissue margins may be diffi-cult even when the tumor is palpable. Excision of a nonpalpable tumor withadequate margins by guidance of a wire is much more challenging, because offrequent displacement of the wire. The use of a gamma probe makes easier tocontrol of the tumor edges, because the radioactivity clearly decreases outside thetumor (Zurrida et al 1998). Furthermore, residual radioactivity is easy to checkfrom the wound to ensure that the tumor is resected radically (Zurrida et al1998).

In the present study, close or involved margins were observed in less than 10%of our ROLL specimens. The closest free margin was at least 10 mm in almost70% of the cases. Gray and co-workers (2001) achieved initial margins of over 1mm for 75% of patients, while Feggi and co-workers (2001) obtained free mar-gins of over 3 mm in 95% of in their quadrantectomy specimens, and GallegosHernandez and co-workers (2004) obtained adequate margins for 83% after thecompletion of the primary operation.

ROLL has been compared with wire localization (Gray et al 2001, Luini et al1999, Gallegos Hernandez et al 2004), and all these studies concluded that ROLLwas quite a quick, easy, and accurate method for a surgeon to find an occultbreast lesion. Furthermore, ROLL allows for a reduced excision volume and bet-ter lesion centering within the specimen compared to those of wire localization(Luini et al 1999, Gallegos Hernandez et al 2004).

In the present study, involved or close margins, and on the other hand, alsoexcessively wide margins were encountered more often with the use of WGR thanwith ROLL or surgery for palpable lesions. However, a reliable comparison be-tween results obtained by ROLL and wire localization is unfortunately not pos-sible on the basis of our findings. The subgroup of patients operated on with useof a guide wire was small and selected. The proportion of cases with pure ductalinvasive in situ or invasive carcinoma with an extensive invasive component washigh compared with other patient groups. Gallegos Hernandez and co-workersfound decreasing pathological diameter, increasing weight of the specimen, andabsence of microcalcifications or carcinoma in situ to be predictive factors forclear margins. Thus, the differences observed between ROLL and wire localiza-tion may reflect the different nature of tumors visible or invisible in breast US,not only the differences between two localization methods.

Only a couple of previous studies have addressed the feasibility of ROLL inconnection with LS and SNB (Tanis et al 2001, Feggi et al 2001). We, as authorsof previous studies, have found the simultaneous performance of ROLL and SNBin the management of early breast cancer to be feasible (Tanis et al 2001, Feggi etal 2001).

47DISCUSSION

Hospital costsIt is not surprising that the new method, SNB, turned out to be the most expen-sive alternative for axillary staging. At the beginning of its introduction, newmethods and procedures are associated with elevated costs due to the specialequipment and examinations necessary for successful performance of the meth-od. Gemignani et al (2000) found SNB to be less costly than ALND because ofthe shorter hospital stay after SNB. Only patients with a low probability ofaxillary metastases, those with T1 tumors, were included in their study. Theirlower false-negative rate (4–9%) in frozen section diagnosis also resulted in lowercosts for SNB in relation to diagnostic ALND; a large majority (73%) of SNBpatients with tumor-negative findings in frozen section were discharged the sameday, and their postoperative hospital stays were shorter in general (Gemignani etal 2000). We discharge our own axillary node-negative SNB patients the follow-ing day, because of postoperative nursing care, counseling and physiotherapy.

SNB as an initial outpatient procedure under local anesthesia would have beenthe most tempting alternative for SNB, at least from a theoretical standpoint. Afterthe initial procedure, the histopathological report of SNs would be final, meaningthat the patient would come to the subsequent breast operation with axillary stag-ing already performed. However, according to our calculations, this would be themost expensive alternative, probably due to the numerous operations. Moreover,based on our clinical experience, AC as a second operation seems to be more de-manding and laborious because of scar formation after SNB and because sparingthe intercostobrachial nerves in the second procedure may be difficult.

From the economic point of view, frozen section analysis of SNs is probablythe most crucial part of the whole procedure. Frozen section diagnostics is ex-pensive due to the substantial workload in the pathology laboratory, but it al-lows AC for most axillary node-positive patients during the primary operation.A false-negative result leads to re-operation and elevated costs of surgical treat-ment. Recovery, including sick leave, will be longer and the risk for post–opera-tive morbidity may be higher after two procedures. The sensitivity analysis re-vealed that frozen section diagnosis seems to be economically worthwhile as longas the false-negative rate does not exceed 35%, which is far higher than the rateobserved in our clinic.

The lower costs of a shorter hospital stay may not fully compensate for thecosts of LS and the histological evaluation of SN; economic justification of themethod has to be sought from the wider social aspects such as decreased postop-erative morbidity and recovery time. SNB is associated with less early postopera-tive morbidity than is AC, and thus the return to work should be quicker. Wetherefore tried to evaluate the differences in sick leaves, but the length of the sickleave appeared to be dependent more on psychosocial factors than simply onphysical recovery from surgery. Postoperative sick leave often continued until theend of adjuvant chemo- and/or radiotherapy because of factors like adverse ef-fects of treatment, the patient’s psychological status, and distance between theresidence and the Department of Oncology.

48 DISCUSSION

For these reasons, the most significant advantage of SNB will probably bedeclining long-term morbidity. As regards chronic post-mastectomy pain andlymphedema, the results from the present and previous studies are encouraging(Schrenk et al 2000, Miguel et al 2001, Sener et al 2001, Burak et al 2002,Haidet al 2002). However, larger studies with longer follow-up periods are neededuntil the real benefits of SNB can be determined. Patients have to be followed upfor several years, with detailed data recorded regarding need for physiotherapydue to arm morbidity, periods of sick leave, and change in work place or in workresponsibilities, or both. However, these work-related events are irrelevant for asubstantial percentage of breast cancer patients because they are already retireddue to old age or disability caused by other conditions. Anyway, physiotherapyand especially the treatment of severe chronic lymphedema are expensive and con-cern all breast cancer patients, irrespective of age and working status.

Comparison of the costs of different procedures is demanding, and the resultsmay not be generalizeable, for instance due to differences in treatment practicesand unit costs of resources and services (Gemignani et al 2000, Palit et al 2000).The applicability of our results to other units thus depends on factors like thecosts of the LS and inpatient care, as well as salaries of the pathologists, surgeons,and nurses. In addition, the length of hospital stay after SNB or ALND, thesuccess rate in identification of SN, and the false-negative rate in frozen-sectiondiagnosis vary not only between different units but may also vary over time in thesame hospital. For example, our clinic’s SN identification rate has improved fromthe reported 86% and is nowadays practically 100%.

The reliability of our results may be disputed because our patients were notrandomized to the different axillary staging alternatives (ALND, SNB as day-casesurgery, SNB with and without intraoperative diagnosis). The costs of SNB withfrozen section reflect actual costs, but the costs for the hypothetical strategieswere only estimated and may therefore differ from the actual ones at least in somerespects. In addition, we assumed all patients to be eligible for SNB as day-casesurgery under local anesthesia. Because the costs of LS as well as those of theoperations and histopathological assessments of breast and axillary lymph nodespecimens are averages based on a large number of cases, only the length of hos-pital stay and possible complications after various axillary staging modalities leaveroom for speculation. Because the shortest possible postoperative hospital staywas applied when estimating the costs of the hypothetical strategies, actual costsmight be somewhat higher if several patients needed a longer stay after AC. It is,however, unlikely that the hospital costs of diagnostic ALND exceed those ofSNB, at least when all patients eligible for SNB are considered. SNB may also beless expensive than ALND in patients with a very low probability of axillary me-tastases (Gemignani et al 2000).

49DISCUSSION

Postoperative lymphedemaThe symptoms and signs of breast lymphedema are well known, but its etiologyis still obscure (Clarke et al 1982, Meek et al 1998). Postoperative edema can bedivided into early and late onset. Edema is considered as early, when it developswithin the first 2 months, whereas late breast edema occurs about 20 monthsafter surgery and RT, or after either one alone. (Clarke et al 1982). Breast surgeryas well as RT to the breast or to the axilla, or both, can disturb the lymphaticcirculation of the breast. (Clarke et al 1982). Risk for postoperative breast lymph-edema seems to be higher in obese patients with surgery for an upper outerquadrant tumor (Schrenk et al 2000).

According to the findings of the present and the previous studies, approxi-mately one patient in three suffers from breast edema one year after surgery (Clarkeet al 1982, McCormick et al 1989). Breast edema seems undoubtedly to be relat-ed to the extent of axillary surgery. After axillary sampling, Clarke and co-work-ers (1982) observed breast lymph edema in 25% of their patients, a finding inclose agreement with our 23% prevalence of breast edema after SNB withoutfurther axillary treatment. Breast edema was observed clearly more often, in 35%to 79% of the patients, after AC in the present study and than that observed byClarke and co workers (Clarke et al 1982), who found risk for breast edema with-out any axillary surgery to be about 6%. (Clarke et al 1982)

Adjuvant breast RT is another risk factor for breast lymphedema. Senofskyand co-workers (1991) observed a 21% incidence of breast edema after breastresection and RT, but incidence of breast edema was only 6% among non-irradi-ated patients. The influence of breast RT on risk for breast edema was not as-sessed in the present study, because practically all our patients received adjuvantbreast RT.

It is insufficiently clear whether addition of RT to the axillary and supraclavic-ular fields after AC further elevates risk for breast edema. In the present study,patients with tumor-positive axillary nodes, all of whom received RT also to theaxillary and the supraclavicular fields, manifested more breast edema. Whetherthe reason for increased breast morbidity in this patient group was the moreextensive RT or the axillary node involvement itself, or both, could not be con-cluded on the basis of our findings. Another possible explanation for the moreprevalent breast edema in patients with AC and involved axillary nodes is the 3-month shorter period between completion of RT and clinical and US examina-tion. The timing of RT in relation to the US examination may also have influ-enced our findings concerning skin thickness of the operated breast.

The location of the tumor in the breast may also influence the prevalence ofbreast lymphedema (Temple et al 2002, Schrenk et al 2000). Especially delayedbreast cellulitis, the most severe form of breast lymphedema, seems to be com-mon in patients with tumor in the upper lateral breast segment and breast con-serving treatment (Temple et al 2002). However, in the present study, location ofthe tumor in the breast had no influence on the prevalence of breast lymphede-

50 DISCUSSION

ma. None of the our patients had delayed breast cellulitis, but only milder formsof breast lymphedema. In addition, all of our patients underwent axillary sur-gery, SNB, or AC. Although SNB is less extensive than AC, it is also an invasiveprocedure disrupting lymphatics draining from the breast to the axilla. Preciseevaluation of the role of the location of the resection in the breast requires apatient group with breast surgery without any axillary procedure.

In addition to the varying extent and timing of RT, our results may be biasedbecause of the small study population and the non-randomized study setting asregards AC and SNB. Other confounding factors like systemic adjuvant therapy,age, and body mass index of the patients have also possibly influenced our results.Despite the limitations of our study, our results as well as those of the Clarke andco-workers (1982) indicate, however, that the extent of the axillary procedureelevates the risk for breast lymphedema.

Few simple objective methods exist for evaluation of post-treatment symp-toms in the upper extremity and especially those in the breast. In addition to theclinical examination, we evaluated the breasts with US and compared the operat-ed breast to the non-operated one. The US findings were concordant with thoseof the clinical examination. Furthermore, the thickness of the skin and the amountof interstitial fluid was similar on the non-operated side in all three treatmentgroups. For these reasons breast US seems clearly to be a feasible method forevaluation of breast edema.

Postoperative upper extremity functional disordersRestricted shoulder motion was more common 2 weeks after AC, as reported inour study examining early postoperative morbidity in these same patients (Lei-denius et al 2003). Peintinger (2003) also shows less motion restriction in favorof SNB.

A slight but not significant difference in shoulder abduction and flexion, aswell as self-reported shoulder stiffness was evident after SNB and AC also at theend of the study period. Muscle weakness in the ipsilateral arm was reported morefrequently after more extensive axillary treatment. It is difficult to distinguishwhether such symptoms are due to breast or axillary surgery, adjuvant radio- orchemotherapy, or musculosceletal disorders unrelated to breast cancer treatment.It is noteworthy that almost one patient in every three already had restrictedshoulder motion before surgery. Strain in the pectoral muscle was a commoncomplaint after but also before the operation. In addition, two SNB-group pa-tients fell and hurt the ipsilateral shoulder just before the one-year follow-up. Forthese reasons, the real incidence of axillary surgery-related functional disorders isdifficult to estimate. Knowledge of preoperative status as to range of motion iscrucial in evaluating the influence of axillary treatment.

51DISCUSSION

Postoperative upper extremity sensory disordersThe cause of upper arm sensory disorders is suggested to be transsection of orinjury to the intercostobrachial nerve (Freeman et al 2003). This causes impair-ment of skin sensation in the area of the axilla and upper arm and can also lead toan edematous sensation in the upper arm. This kind of nerve injury is moreprobable after AC than after SNB. The present and the other studies (Peinttingeret al, Temple et al 2002, Burak et al 2002) show that sensory disorders are morecommon after AC than after SNB.

During the first 6 postoperative months, skin sensation, tested by von Freyfilament stimulation, was significantly altered after AC compared to that afterSNB. Improvement in sensation back to the preoperative level was recorded dur-ing the one-year follow-up in almost all patients regardless of the extent of axil-lary treatment, probably due to the recovery of spared sensory nerves. However,at the end of follow-up, pain and self- reported sensory disorders like numbnessand strange sensations in the ipsilateral arm were still common, especially afterextensive axillary treatment. One-fourth of the patients in the SNB group report-ed mild or moderate pain or sensory disorders in the arm, but none of themsuffered severe pain, numbness, or strange sensations. On the other hand, somepatients also in the SNB group experienced pain, with daily need for medication.Unfortunately, the data did not differentiate among the locations of bothersomepain, in the breast operation area, arm, or both.

Postoperative lymphedemaMeasurement of lymphedema is challenging, because it lacks simple, objective,reliable methods (Gerber et al 1998, Rampaul et al 2003). Distinguishing be-tween the lymph edema-causative or -associative factors – extent of breast andaxillary surgery, RT, BMI, or infection – is not easy.

Most of the studies evaluate lymphedema by measuring the circumference ofthe arm at several points (Schrenk et al 2001, Golshan et al 2003) and definelymphedema as an increase in circumference of at least 2 cm compared with pre-operative status or the contralateral arm (Sener et al 2001, Schrenk et al 2001).This could be misleading, if factors such as BMI of the patient, change in theweight and location of the lymphedema in the arm, muscle wasting due to de-creased use of the ipsilateral arm, strenuous exercise using the affected arm, recentmanual lymphatic draining, and the use of a compression sleeve are not takeninto account. Nor are measurement points using bony protuberances as land-marks exact. In addition, a 2-cm change in arm circumference is certainly signif-icant in a very slim patient but hardly noticeable in an obese one. Probably forthese reasons we found no differences in arm circumferences between groups.This finding is in close agreement with the observations of Temple and co-work-ers (2002). Not surprisingly, increase in arm circumference correlated poorly withneed for lymph therapy or self-experienced lymphedema, even in patients withsevere edema and a compression sleeve.

52 DISCUSSION

Subjective estimation of lymphedema in questionnaires may lead to over-esti-mation and increased figures for prevalence of lymphedema. The feeling of numb-ness and thickness due to injury of the intercostobrachial nerve may be misinter-preted as lymphedema, leading to over-reporting. Accordingly, self-experiencedarm edema, most often mild, was a common finding in the present study, espe-cially after extensive axillary treatment. Nevertheless, less than 15% of patients inthe SNB group complained of arm edema, and none of them experienced it assevere. Furthermore, none of the patients with SNB alone wore a compressionsleeve or received lymph therapy. This is in agreement with recent findings ofnone or only minimal risk for clinically significant lymphedema after axillary nodesampling (Rampaul et al 2003) or SNB (Blanchard et al 2003) without furtheraxillary treatment. For these reasons, the risk for disabling lymphedema afterSNB seems only minimal.

Interference of postoperative morbidity with daily lifeInterference of arm morbidity with work, hobbies, or daily life in general was lesscommon and less severe after SNB alone than after more extensive axillary treat-ment. Arm morbidity affected the daily life of about 20% of patients after SNBalone, but the interference was mild in this patient group.

Our results are in agreement with previous reports showing faster recoveryand less postoperative morbidity after SNB alone than after more extensive axil-lary treatment (Miguel et al 2001, Roumen et al 2001, Schrenk et al 2001, Seneret al 2001, Burak et al 2002, Ernst et al 2002, Haid et al 2002, Temple et al2002, Blanchard et al 2003, Gosselink et al 2003, Leidenius et al 2003, Peintingeret al 2003, Schijven et al 2003) Regrettably, we could not estimate reliably theinfluence of extent of axillary surgery on one-year postoperative morbidity, be-cause most patients in our AC group were axillary node-positive and received RTto the axillary and supraclavicular fields. They also received adjuvant chemother-apy more frequently than did the mainly node-negative patients in the SNB group.RT increases especially the risk for lymphedema after AC (Bentzen et al 2000,Hojris et al 2000,Ververs et al 2001), but is also associated with sensory andfunctional disorders (Bentzen and Dische 2000, Hojris et al 2000, Ververs et al2001) The frequent use of adjuvant axillary RT is the most probable explanationfor the high proportion of our AC-group patients’ receiving lymph therapy.

53CONCLUSION

Conclusion

SNB is a feasible method for nodal staging of breast cancer, providing informa-tion on nodal metastases, both in the axilla and outside the axilla. Simultaneousresection of a nonpalpable lesion in combination with SNB is possible after asingle intratumoral radioactive tracer injection without other localization meth-ods. The results of this technique are comparable with results achieved by surgeryon palpable lesions.

SNB is a more expensive method for nodal staging than is diagnostic ALND.However, the intra-operative diagnosis of SN metastases seems to be worthwhileas long as the false-negative rate is relatively low. Risk for severe arm morbidityand long-time morbidity is minimal after SNB; in particular the risk for disablinglymph edema is negligible.

54 ACKNOWLEDGEMENTS

Acknowledgements

This study was carried out at the Breast Surgery Unit, Maria Hospital, of theHelsinki University Hospital.

I wish to express my deepest gratitude to a number of people who made thiswork possible:

Professor Krister Höckerstedt, you provided me excellent facilities for this re-search.

Docent Karl von Smitten, my supervisor, provided the possibility to performthis study. I always admire you as an experienced and skillful colleague. Mysecond supervisor, Dr. Marjut Leidenius, you pushed me energetically to the goalon such a tight schedule. I am grateful to both of you for the teamwork weenjoyed during this study. I really appreciate Marjut’s energy during this workand the enthusiasm that infected me, as well.

My respectful thanks to Professor Kaija Holli and Docent Markku Härmä,official reviewers of my thesis. I thank you for attending to this study carefullyand helping me with constructive and valuable criticism to complete the finalmanuscript.

My co-workers Drs. Leena Krogerus, Esa Leppänen, Martti Pamilo, and TiinaTasmuth who took part in this study besides their own work. I am grateful foryour excellent collaboration, constructive criticism, and helpful advice.

Professor Harri Sintonen and Teija Kotomäki, I thank you sincerely for col-laboration in my best publication, even though the study construction and re-sults didn’t correspond with your own view.

Karin von Smitten-Dubb and Johanna Möller, and Minna Ronkainen, I thankfor you examining all the study patients and saving the data careful. Withoutyour contribution, I could never have finished this study on schedule.

My colleagues in the Breast Cancer Unit, I also express my deepest thanks toyou and your work and contributions during the examination of all these patients.

Nurses in the outpatient department of surgery in Maria Hospital, I admireyour sense of responsibility in carrying out the study project besides your ownwork and pressures. Marcus Nilsson in the clinic of anesthesia in Maria Hospital,I thank you for helping me in collecting the charges for all procedures concern-ing my study.

Elina Nurmi my sincere thanks for designing the layout of my posters, andalso Ari Karhu, you both edited the posters with originality.

Special thanks to my Dutch colleague, Omgo Nieweg, who has guided me tothe world of the sentinel node via his valuable publications and permitting me touse figures of his publications in my thesis.

55ACKNOWLEDGEMENTS

Juhani Lassander has been worth gold during this study, answering the mostpeculiar e-mails and requests in minimal time. Thanks and hugs.

Vesa Perhoniemi, I am grateful for your continuous encouragement and tak-ing care of the arrangement during my surgical training.

Special thanks to Carol Norris. She came to my house on one of the rainy daysthis summer and helped me in editing the language of the thesis.

Sincere thanks to all my colleagues in Maria Hospital; thanks for your interestin my study and your encouragement, especially, Professor Mirja Ruutu, DocentMartti Ala-opas, Eero Wuokko, and my dear tutor Ilkka Perttilä and all othercolleagues in the department of urology, you all have lived right beside me dur-ing this work and made it possible for me to finish it on the planned schedule.I am sorry about my absence and my inadequately attending to urology.

I extend my sincere thanks to my friends: my dearest friend Kristiina Aalto-nen, I thank you pushing me forward in life and also for offering all possiblesupport during these years. Erno Polus, my personal trainer, you taught me manyuseful things about sports and helped me keep in condition, even during strenu-ous days when I was finishing the first version of the manuscript. Thanks tocookbook authors Antti Vahtera and Visa Nurmi for inspiring me in cookingand helping me keep touch with normal life during this study.

Finally, I wish to express my deepest gratitude to my whole family, even mydeceased dad, Kauko. My dear mother Elsa, I can ever thank you too much. Youhave humbly traveled to Espoo and taken care my dear children Juuli and Niilaand have been patient even though I have been tired and surly. Thank you, mydear husband Ari and our Juuli and Niila, the most important persons in my life,you have always been present and reminded me: “carpe diem”: I love you.

This study was supported by the Finnish Breast Cancer Group, the FinnishMedicine Foundation, and Helsinki University Hospital.

Espoo, August 2004

56 REFERENCES

References

Albertini JJ, Lyman GH, Cox C, Yeatman T, Balducci L, Ku N, Shivers S, Ber-man C, Wells K, Rapaport D, Shons A, Horton J, Greenberg H, Nicosia S,Clark R, Cantor A, Reintgen DS. Lymphatic mapping and sentinel node biopsyin the patient with breast cancer. JAMA 276(22):1818–22, 1996.

Alex JC, Krag DN. The gamma-probe-guided resection of radiolabeled primary lymphnodes. Surg Oncol Clin N Am 5(1):33–41, 1996.

Aspegren K, Holmberg L, Adami O. Standardization of the surgical technique inbreast-conserving surgery of mammary cancer. Br J Surg 75: 807–10, 1988.

Axelsson CK, Rank F, Blichert-Toft M, Mouridsen HT, Jensen MB. Impact ofaxillary dissection on staging and regional control in breast tumors < or = 10 mm –the DBCG experience. The Danish Breast Cancer Cooperative Group (DBCG),Rigshisoutalet, Copenhagen, Denmark. Acta Oncol 39(3):283–9, 2000.

Barros A, Cardoso MA, Sheng PY, Costa PA, Pelizon C. Radioguided localisationof non-palpable breast lesions and simultaneous sentinel lymph node mapping. EurJ Nucl Med Mol Imaging 29(12):1561–5, 2002.

Bentzen SM, Dische S. Morbidity related to axillary irradiation in the treatment ofbreast cancer. Acta Oncol 39(3):337–47, 2000.

Blanchard DK, Donohue JH, Reynolds C, Grant CS. Relapse and morbidity inpatients undergoing sentinel node biopsy alone or with axillary dissection for breastcancer. Arch Surg 138(5): 482–7; 2003.

Bland KI, Scott-Conner CE, Menck H, Winchester DP. Axillary dissection in breast-conserving surgery for stage I and II breast cancer: a National Cancer Data Basestudy of patterns of omission and implications for survival. J Am Coll Surg 188(6):586–95; 1999.

Bland & Copeland. The Breast. Comprehensive management of benign and malig-nant disorders, (ed), Saunders, 2004. Pp. 8–18, 1019–1129

Blichert-Toft M. Axillary surgery in breast cancer management – background, inci-dence and extent of nodal spread, extent of surgery and accurate axillary staging,surgical procedures. Acta Oncol 39(3):269–75, 2000.

Borgstein PJ, Meijer S, Pijpers R. Intradermal blue dye to identify sentinel lymph-node in breast cancer. Lancet 7; 349:1668–9, 1997.

Borgstein PJ, Meijer S, Pijpers R. Sentinel lymph node biopsy in breast cancer: guide-lines and pitfalls of lymphoscintigraphy and gamma probe detection. J Am Coll Surg186(3):275–83, 1998.

Borgstein PJ, Meijer S, Pijpers RJ, van Diest PJ. Functional lymphatic anatomy forsentinel node biopsy in breast cancer: echoes from the past and the periareolar bluemethod. Ann Surg; 232(1):81–9, 2000.

57REFERENCES

Bundred NJ, Morgan DA, Dixon JM. ABC of breast diseases. Management of re-gional nodes in breast cancer. BMJ 309(6963): 1222–5, 1994.

Burak WE, Hollenbeck ST, Zervos EE, Hock KL, Kemp LC, Young DC. Sentinellymph node biopsy results in less postoperative morbidity compared with axillarylymph node dissection for breast cancer. Am J Surg 183(1): 23–7, 2002.

Cabañas RM. An approach for the treatment of penile carcinoma. Cancer 39(2): 456–66, 1977.

Canavese G, Gipponi M, Catturich A, Di Somma C, Vecchio C, Rosato F, TomeiD, Nicolo G, Carli F, Villa G, Agnese G, Bianchi P, Buffoni F, Mariani G,Badellino F. Pattern of lymphatic drainage to the sentinel lymph node in breastcancer patients. J Surg Oncol 74(1):69–74, 2000

Chetty U, Jack W, Prescott RJ, Tyler C, Rodger A. Management of the axilla inpatients with operable breast cancer treated by breast conservation: a randomisedtrial. Br J Surg 87 (2): 163–169, 2000.

Chirikos TN, Berman CG, Luther SL, Clark RA. Cost consequences of sentinellymph node biopsy in the treatment of breast cancer. A preliminary analysis. Int JTechnol Assess Health Care 17(4):626–31, 2001.

Clarke D, Martinez A, Cox RS, Goffinet DR. Breast edema following staging axil-lary node dissection in patient with breast carcinoma treated by radical radiotherapy.Cancer 49:2295–99, 1982.

Cox CE, Salud CJ, Cantor A, Bass SS, Peltz ES, Ebert MD, Nguyen K, ReintgenDS. Learning curves for breast cancer sentinel lymph node mapping based onsurgical volume analysis. J Am Coll Surg 193(6):593–600, 2001.

Daniell HW. Increased lymph node metastases at mastectomy for breast cancer associ-ated with host obesity, cigarette smoking, age, and large tumor size. Cancer62(2):429–35, 1988.

Davies GC, Millis RR, Hayward JL. Assessment of axillary lymph node status. AnnSurg 192(2):148–51, 1980.

De Cicco C, Cremonesi M, Luini A, Bartolomei M, Grana C, Prisco G, Galimber-ti V, Calza P, Viale G, Veronesi U, Paganelli G. Lymphoscintigraphy and radi-oguided biopsy of the sentinel axillary node in breast cancer. J Nucl Med 39(12):2080–4, 1998.

Deutsch M, Flickinger JC. Arm edema after lumpectomy and breast irradiation. AmJ Clin Oncol 26(3): 229–31, 2003.

Donegan WL. The influence of untreated internal mammary metastases upon the courseof mammary cancer. Cancer 39(2):533–8, 1977.

Doting MH, Jansen L, Nieweg OE, Piers DA, Tiebosch AT, Koops HS, RutgersEJ, Kroon BB, Peterse JL, Olmos RA, de Vries J. Lymphatic mapping withintralesional tracer administration in breast carcinoma patients. Cancer1;88(11):2546–52, 2000.

58 REFERENCES

Duff M, Hill AD, McGreal G, Walsh S, McDermott EW, O’Higgins NJ. Prospec-tive evaluation of the morbidity of axillary clearance for breast cancer. Br J Surg.88(1):114–7, 2001.

Engler HS, Sweat RD. Volumetric arm measurements: technique and results. AmSurg 28:465–8, 1962.

Erickson VS, Pearson ML, Ganz PA, Adams J, Kahn KL. Arm edema in breastcancer patients. J Natl Cancer Inst 93(2):96–111, 2001.

Ernst MF, Voogd AC, Balder W, Klinkenbijl JH, Roukema JA. Early and latemorbidity associated with axillary levels I-III dissection in breast cancer. J SurgOncol 79(3): 151-5, 2002.

Estourgie SH, Tanis PJ, Nieweg OE, Valdes Olmos RA, Rutgers EJ, Kroon BB.Should the hunt for internal mammary chain sentinel nodes begin? An evaluation of150 breast cancer patients. Ann Surg Oncol 10(8):935–41, 2003.

Estourgie SH, Nieweg OE, Olmos RA, Rutgers EJ, Kroon BB. Lymphatic drain-age patterns from the breast. Ann Surg 239(2): 232–7, 2004.

Feggi L, Basaglia E, Corcione S, Querzoli P, Soliani G, Ascanelli S, Prandini N,Bergossi L, Carcoforo P. An original approach in the diagnosis of early breastcancer: use of the same radiopharmaceutical for both non-palpable lesions and sen-tinel node localisation. Eur J Nucl Med 28(11):1589–96, 2001.

Fenaroli P, Merson M, Giuliano L, Bonasegale A, Virotta G, Pericotti S, Valenti-ni M, Poletti P, Labianca R, Personeni A, Tondini C. Population-based sentinellymph node biopsy in early invasive breast cancer. Eur J Surg Oncol 30(6):618–23,2004.

Fentiman IS, Mansel RE. The sentinel node negative axilla: is it a no-go zone? CancerTreat Rev 28(1):1–3, 2002.

Finnish Cancer Registry. Cancer Statistics for Finland. http://www.cancerregistry.fi/eng/statistics.htm (updated November 2002).

Freedman GM, Fowble BL, Nicolaou N, Sigurdson ER, Torosian MH, BoraasMC, Hoffman JP. Should internal mammary lymph nodes in breast cancer be atarget for the radiation oncologist? Int J Radiat Oncol Biol Phys 1;46(4):805–14,2000.

Freeman SR, Washington SJ, Pritchard T, Barr L, Baildam AD, Bundred NJ.Long term results of a randomised prospective study of preservation of the intercos-tobrachial nerve. Eur J Surg Oncol 29(3):213–5, 2003.

Forrest APM, Everington D, McDonald CC, Steele RJC. Chetty U, Stewart HJ.The Edinburgh randomised trial of axillary sampling or clearance after mastectomy.Br J Surg 82(11):1504–1508, 1995.

Galimberti V, Veronesi P, Arnone P, De Cicco C, Renne G, Intra M, Zurrida S,Sacchini V, Gennari R, Vento A, Luini A, Veronesi U. Stage migration afterbiopsy of internal mammary chain lymph nodes in breast cancer patients. Ann SurgOncol 9(9):924–8, 2002.

59REFERENCES

Gallegos Hernandez JF, Tanis PJ, Deurloo EE, Nieweg OE, Th Rutgers EJ, KroonBB, Valdes Olmos RA. Radio-guided surgery improves outcome of therapeuticexcision in non-palpable invasive breast cancer. Nucl Med Commun 25(3):227–32,2004.

Gemignani ML, Cody HS 3rd, Fey JV, Tran KN, Venkatraman E, Borgen PI.Impact of sentinel lymph node mapping on relative charges in patients with early-stage breast cancer. Ann Surg Oncol 7(8):575–80, 2000.

Gennari R, Galimberti V, De Cicco C, Zurrida S, Zerwes F, Pigatto F, Luini A,Paganelli G, Veronesi U. Use of technetium-99m-labeled colloid albumin for pre-operative and intraoperative localization of nonpalpable breast lesions. J Am CollSurg 190(6):692–8, 2000.

Gerber LH. A review of measures of lymphoedema. Cancer 83(12): 2803–4, 1998;.

Giuliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic mapping andsentinel lymphadenectomy for breast cancer. Ann Surg. 220(3):391–8; 1994.

Giuliano AE, Jones RC, Brennan M, Statman R. Sentinel lymphadenectomy in breastcancer. J Clin Oncol 15(6):2345–50, 1997.

Glass EC, Essner R, Giuliano AE. Sentinel node localization in breast cancer. SeminNucl Med 29(1):57–68, 1999.

Golshan M, Martin WJ, Dowlatshahi K. Sentinel lymph node biopsy lowers the rateof lymphedema when compared with standard axillary lymph node dissection. AmSurg 69(3):209–11, 2003.

Gosselink R, Rouffaer L, Vanhelden P, Piot W, Troosters T, Christiaens MR.Recovery of upper limb function after axillary dissection. J Surg Oncol; 83(4): 204–11, 2003.

Gray RJ, Giuliano R, Dauway EL, Cox CE, Reintgen DS. Radioguidance for non-palpable primary lesions and sentinel lymph node(s). Am J Surg 182(4):404–6,2001.

Haid A, Koberle-Wuhrer R, Knauer M, Burtscher J, Fritzsche H, Peschina W,Jasarevic Z, Ammann M, Hergan K, Sturn H, Zimmermann G. Morbidity ofbreast cancer patients following complete axillary dissection or sentinel node biopsyonly: a comparative evaluation. Breast Cancer Res Treat 73(1): 31–6, 2002.

Haid A, Kuehn T, Konstantiniuk P, Koberle-Wuhrer R, Knauer M, Kreienberg R,Zimmermann G. Shoulder-arm morbidity following axillary dissection and sentinelnode only biopsy for breast cancer. Eur J Surg Oncol 28(7): 705–10, 2002.

Halsted WS. The results of operations for the cure of cancer of the breast performed atthe Johns Hopkins Hospital from June, 1889, to January 1894. Ann Surg 20:497,1894.

Harlow SP, Krag DN, Ames SE, Weaver DL. Intraoperative ultrasound localizationto guide surgical excision of nonpalpable breast carcinoma. J Am Coll Surg189(3):241–6, 1999.

60 REFERENCES

Hoar FJ, Stonelake PS. A prospective study of the value of axillary node sampling inaddition to sentinel lymph node biopsy in patients with breast cancer. Eur J SurgOncol. 29(6):526–31, 2003.

Hojris I, Andersen J, Overgaard M, Overgaard J. Late treatment-related morbidityin breast cancer patients randomized to postmastectomy radiotherapy and systemictreatment versus systemic treatment alone. Acta Oncol. 39(3): 355–72, 2000.

Huskisson EC. Measurement of pain. Lancet; ii: 1127–31, 1974.

Isaksson G, Feuk B. Morbidity from axillary treatment in breast cancer – a follow-upstudy in a district hospital. Acta Oncol 39(3):335–6, 2000.

Jansen L, Nieweg OE, Valdes Olmos RA, Rutgers EJ, Peterse JL, de Vries J,Doting MH, Kroon BB. Improved staging of breast cancer through lymphaticmapping and sentinel node biopsy. Eur J Surg Oncol 24(5):445–6, 1998.

Jansen L, Doting MHE, Rutgers EJ, de Vries J, Valdés Olmos RA, Nieweg OE.Clinical relevance of sentinel lymph nodes outside the axilla in patients with breastcancer. Br J Surg 87(7):920–925, 2000.

Kapteijn BA, Nieweg OE, Petersen JL, Rutgers EJ, Hart AA, van Dongen JA,Kroon BB. Identification and biopsy of the sentinel lymph node in breast cancer.Eur J Surg Oncol 24(5):427–30, 1998.

Kissin MW, Thompson EM, Price AB, Slavin G, Kark AE. The inadequacy ofaxillary sampling in breast cancer. Lancet 29;1(8283):1210–2, 1982.

Krag DN, Weaver DL, Alex JC, Fairbank JT. Surgical resection and radiolocalizationof the sentinel lymph node in breast cancer using a gamma probe. Surg Oncol. 2(6):335–9; 1993.

Krag D, Weaver D, Ashikaga T, Moffat F, Klimberg VS, Shriver C, Feldman S,Kusminsky R, Gadd M, Kuhn J, Harlow S, Beitsch P. The sentinel node inbreast cancer – a multicenter validation study. N Engl J Med 339(14): 941–6, 1998.

Krag DN, Harlow S. Current status of sentinel node surgery in breast cancer. Oncology17(12):1663–6; 2003.

Lacour J, Le M, Caceres E, Koszarowski T, Veronesi U, Hill C. Radical mastectomyversus radical mastectomy plus internal mammary dissection. Cancer 51(10):1941–43, 1983.

Leidenius M, Leppänen E, Krogerus L, von Smitten K. Motion restriction andaxillary web syndrome after sentinel node biopsy and axillary clearance in breastcancer. Am J Surg 185(2): 127–30, 2003a.

Leidenius MH, Krogerus LA, Toivonen TS, Von Smitten KJ. The feasibility ofintraoperative diagnosis of sentinel lymph node metastases in breast cancer. J SurgOncol 84(2): 68–73, 2003.

Leidenius MH, Leppänen EA, Krogerus LA, Smitten KA. The impact of radiop-harmaceutical particle size on the visualization and identification of sentinel nodes inbreast cancer. Nucl Med Commun 25(3):233–8, 2004.

61REFERENCES

Leivo Tiina. Economic evaluation of selected Finnish screening programmes. Dep PublHealth, academic dissertation, 2001

Leppänen E, Leidenius M, Krogerus L, von Smitten K. The effect of patient andtumour characteristics on visualization of sentinel nodes after a single intratumouralinjection of Tc 99m labelled human albumin colloid in breast cancer. Eur J SurgOncol 28(8):821–6, 2002.

Luini A, Zurrida S, Paganelli G, Galimberti V, Sacchini V, Monti S, Veronesi P,Viale G, Veronesi U. Comparison of radioguided excision with wire localization ofoccult breast lesions. Br J Surg 86(4):522–5, 1999.

Mateos JJ, Vidal-Sicart S, Zanon G, Pahisa J, Fuster D, Martin F, Ortega M,Fernandez P, Pons F. Sentinel lymph node biopsy in breast cancer patients: sub-dermal versus peritumoural radiocolloid injection. Nucl Med Commun 22(1):17–24, 2001.

McCormick B, Yahalom J, Cos L, Shank B, Massie MJ. The patient’s perception ofher breast following radiation and limited surgery. Int J Radiat Oncol Biol Phys17(6):1299–1302, 1989.

McLelland R. Screening mammography. Cancer 67(4):1129–31, 1991.

McMasters K. M., Giuliano A. E., Ross M. I., Reintgen D. S., Hunt K. K., ByrdD. R., Klimberg V. S., Whitworth P. W., Tafra L. C., Edwards M. J. Sentinel-lymph-node biopsy for breast cancer – not yet the standard of care. N Engl J Med.339(14):990–5, 1998.

McMasters KM, Wong SL, Martin RC 2nd, Chao C, Tuttle TM, Noyes RD,Carlson DJ, Laidley AL, McGlothin TQ, Ley PB, Brown CM, Glaser RL,Pennington RE, Turk PS, Simpson D, Cerrito PB, Edwards MJ; University ofLouisville Breast Cancer Study Group. Dermal injection of radioactive colloid issuperior to peritumoral injection for breast cancer sentinel lymph node biopsy: re-sults of a multiinstitutional study. Ann Surg 233(5):676–87, 2001a.

McMasters KM, Wong SL, Chao C, Woo C, Tuttle TM, Noyes RD, Carlson DJ,Laidley AL, McGlothin TQ, Ley PB, Brown CM, Glaser RL, PenningtonRE, Turk PS, Simpson D, Edwards MJ; University of Louisville Breast Can-cer Study Group. Defining the optimal surgeon experience for breast cancer sen-tinel lymph node biopsy: a model for implementation of new surgical techniques.Ann Surg 234(3): 292–9, 2001.

McQuay HJ, Moore RA. An evidence-based resource for pain relief, Oxford, 1998,pp 15–18.

Meek AG. Breast radiotherapy and lymphedema. Cancer 83(12):2788–97, 1998.

Mendelson EB. Evaluation of the postoperative breast. Radiol Clin North Am 30:107–138, 1992.

Miguel R, Kuhn AM, Shons AR, Dyches P, Ebert MD, Peltz ES, Nguyen K, CoxCE. The effect of sentinel node selective axillary lymphadenectomy on the incidenceof postmastectomy pain syndrome. Cancer Control 8(5): 427–30, 2001.

62 REFERENCES

Miltenburg DM, Miller C, Karamlou TB, Brunicardi FC. Meta-analysis of sentinellymph node biopsy in breast cancer. J Surg Res 84(2):138–42, 1999.

Morrow M, Foster RS. Staging of breast cancer. A new rationale for internal mamma-ry node biopsy. Arch Surg 166:748–751, 1981.

Morrow M, Rademaker AW, Bethke KP, Talamonti MS, Dawes LG, Clauson J,Hansen N. Learning sentinel node biopsy: results of a prospective randomized trialof two techniques. Surgery 126(4): 714–20, 1999.

Mortimer PS. The pathophysiology of lymphedema. Cancer 15(83):2798–802, 1998.

Morton DL, Wen DR, Wong JH, Economou JS, Cagle LA, Storm FK, FoshagLJ, Cochran AJ. Technical details of intraoperative lymphatic mapping for earlystage melanoma. Arch Surg 127(4):392–9, 1992.

Nathanson SD, Wachna DL, Gilman D, Karvelis K, Havstad S, Ferrara J. Path-ways of lymphatic drainage from the breast. Ann Surg Oncol 8(10):837–43, 2001.

Nieweg OE, Jansen L, Valdes Olmos RA, Rutgers EJ, Peterse JL, Hoefnagel KA,Kroon BB. Lymphatic mapping and sentinel lymph node biopsy in breast cancer.Eur J Nucl Med 26(4):S11–6, 1999.

Nieweg OE. Lymphatics of the breast and the rationale for different injection tech-niques. Ann Surg Oncol 8(9 ): 71S–3S, 2001a.

Nieweg OE, Rutgers EJ, Jansen L, Valdes Olmos RA, Peterse JL, Hoefnagel KA,Kroon BB. Is lymphatic mapping in breast cancer adequate and safe? World J Surg25(6): 780–8, 2001b.

Nieweg OE, Tanis PJ, Kroon BB. The definition of a sentinel node. Ann Surg OncolJul;8(6):538–41, 2001c.

Nieweg OE, Bartelink H. Implications of lymphatic mapping for staging and adjuvanttreatment of patients with breast cancer. Eur J Cancer 40(2):179–81, 2004.

Noguchi M, Ohta N, Koyasaki N, Taniya T, Miyazaki I, Mizukami Y. Reappraisalof internal mammary node metastases as a prognostic factor on patients with breastcancer. Cancer 68:1918–1925, 1991.

Noguchi M. Sentinel lymph node biopsy and breast cancer. Br J Surg 89(1):21–34,2002.

Nuutinen J, Lahtinen T, Turunen M, Alanen E, Tenhunen M, Usenius T, KolleR.A dielectric method for measuring early and late reactions in irradiated humanskin. Radiother Oncol 47:249–254, 1998.

O’Hea BJ, Hill AD, El-Shirbiny AM, Yeh SD, Rosen PP, Coit DG, Borgen PI,Cody HS 3rd. Sentinel lymph node biopsy in breast cancer: initial experience atMemorial Sloan-Kettering Cancer Center. J Am Coll Surg. 186(4):423–7, 1998.

Overgaard M, Hansen PS, Overgaard J, Rose C, Andersson M, Bach F, Kjaer M,Gadeberg CC, Mouridsen HT, Jensen MB, Zedeler K. Postoperative radiother-apy in high-risk premenopausal women with breast cancer who receive adjuvantchemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med337(14):949–55,1997.

63REFERENCES

Paganelli G, De Cicco C, Cremonesi M, Prisco G, Calza P, Luini A, Zucali P,Veronesi U. Optimized sentinel node scintigraphy in breast cancer. Q J Nucl Med42(1):49–53, 1998.

Paganelli G, Luini A, Veronesi U. Radioguided occult lesion localization (ROLL) inbreast cancer: maximizing efficacy, minimizing mutilation. Ann Oncol 13(12):1839–40, 2002.

Palit TK, Miltenburg DM, Brunicardi FC. Cost analysis of breast conservation sur-gery compared with modified radical mastectomy with and without reconstruction.Am J Surg 179(6): 441–445, 2000.

Peintinger F, Reitsamer R, Stranzl H, Ralph G. Comprasion of quality of life andarm complaints after axillary lymph node dissection vs. sentinel lymph node biopsyin breast cancer patients. Br J Cancer 89(4): 648–52, 2003.

Petrek JA, Heelan MC. Incidence of breast carcinoma-related lymphedema. Cancer83(12):2776–81,1998.

Petrek JA, Peters MM, Nori S, Knauer C, Kinne DW, Rogatko A. Axillary lym-phadenectomy. A prospective, randomized trial of 13 factors influencing drainage,including early or delayed arm mobilization. Arch Surg 125(3):378–82, 1990.

Pezner RD, Patterson MP, Hill LR, Lipsett JA, Desai KR, Vora N, Wong JY, LukKH. Arm lymphedema in patients treated conservatively for breast cancer: relation-ship to patient age and axillary node dissection technique. Int J Radiat Oncol BiolPhys 12(12):2079–83, 1986.

Port ER, Tan LK, Borgen PI, Van Zee KJ. Incidence of axillary lymph node metas-tases in T1a and T1b breast carcinoma. Ann Surg Oncol 5(1):23–7, 1998.

Quan ML, McCready D, Temple WJ, McKinnon JG. Biology of lymphatic metas-tases in breast cancer: lessons learned from sentinel node biopsy. Ann Surg Oncol9(5):467–71, 2002.

Ragaz J, Jackson SM, Le N, Plenderleith IH, Spinelli JJ, Basco VE, Wilson KS,Knowling MA, Coppin CM, Paradis M, Coldman AJ, Olivotto IA. Adjuvantradiotherapy and chemotherapy in node-positive premenopausal women with breastcancer. N Engl J Med 337(14):956–62,1997.

Rahusen FD, Taets van Amerongen AH, van Diest PJ, Borgstein PJ, BleichrodtRP, Meijer S. Ultrasound-guided lumpectomy of nonpalpable breast cancers: Afeasibility study looking at the accuracy of obtained margins. J Surg Oncol 72(2):72–6, 1999.

Rampaul RS, Mullinger K, Macmillan RD, Cid J, Holmes S, Morgan DA, BlameyRW. Incidence of clinically significant lymphoedema as a complication followingsurgery for primary operable breast cancer. Eur J Cancer 39(15):2165–7, 2003.

Recht A, Houlihan MJ. Axillary lymph nodes and breast cancer: a review. Cancer76(9):1491–512, 1995.

Recht A, Edge SB, Solin LJ, Robinson DS, Estabrook A, Fine RE, Fleming GF,Formenti S, Hudis C, Kirshner JJ, Krause DA, Kuske RR, Langer AS, SledgeGW Jr, Whelan TJ, Pfister DG; American Society of Clinical Oncology. Post-

64 REFERENCES

mastectomy radiotherapy: clinical practice guidelines of the American Society ofClinical Oncology. J Clin Oncol 19(5): 1539–69, 2001.

Rissanen TJ, Mäkäräinen HP, Kiviniemi HO, Suramo II. Ultrasonographically guidedwire localization of nonpalpable breast lesions. J Ultrasound Med 13(3):183–8, 1994.

Rivadeneira DE, Simmons RM, Christos PJ, Hanna K, Daly JM, Osborne MP.Predictive factors associated with axillary lymph node metastases in T1a and T1bbreast carcinomas: analysis in more than 900 patients. J Am Coll Surg 191(1): 1–6,2000.

Rockson SG, Miller LT, Senie R, Brennan MJ, Casley-Smith JR, Foldi E, FoldiM, Gamble GL, Kasseroller RG, Leduc A, Lerner R, Mortimer PS, NormanSA, Plotkin CL, Rinehart-Ayres ME, Walder AL. American Cancer SocietyLymphedema Workshop. Workgroup III: Diagnosis and management of lymphe-dema. Cancer 83(12): 2882–5, 1998.

Rosen PP, Lesser ML, Kinne DW, Beattie EJ. Discontinuous or “skip” metastases inbreast carcinoma: analysis of 1228 axillary dissections. Ann Surg 197(3): 276–83,1983.

Roumen RM, Valkenburg JG, Geuskens LM. Lymphoscintigraphy and feasibility ofsentinel node biopsy in 83 patients with primary breast cancer. Eur J Surg Oncol23(6): 495–502, 1997.

Sappey MPC. Anatomie, physiologie, pathologie des vaisseaux lymphatiques consid-eres chez l’homme at les vertebras A. DeLahaye and E. Lecrosnier, Paris, 1874.

Schijven MP, Vingerhoets AJ, Rutten HJ, Nieuwenhuijzen GA, Roumen RM,van Bussel ME, Voogd AC. Comparison of morbidity between axillary lymphnode dissection and sentinel node biopsy. Eur J Surg Oncol 29(4): 341–50, 2003.

Schrenk P, Rieger R, Shamiyeh A, Wayand W. Morbidity following sentinel lympnode biopsy versus axillary lymph node dissection for patients with breast carcino-ma. Cancer 88 (3): 608–614, 2000.

Sener SF, Winchester DJ, Martz CH, Feldman JL, Cavanaugh JA, WinchesterDP, Weigel B, Bonnefoi K, Kirby K, Morehead C. Lymphedema after sentinellymphadenectomy for breast carcinoma. Cancer 92(4): 748–52, 2001.

Senofsky GM, Moffat FL, Davis K, Masri MM, Clark KC, Robinson DS, SabtesB, Ketcham AS. Total axillary lymphadectomy in the management of breast can-cer. Arch Surg 126(11):1336–42, 1991.

Sobin LH, Wittekind CH (eds). TNM classification of malignant tumours. U.I.U.A.Cancer. (ed). Wiley-Liss: New-York, 2002. Pp131–141.

Steele RJ, Forrest AP, Gibson T, Stewart HJ, Chetty U. The efficacy of loweraxillary sampling in obtaining lymph node status in breast cancer: a controlled ran-domized trial. Br J Surg 72(5):368–9, 1985.

Stranden E A comparison between surface measurements and water displacement vol-umetry for the quantification of leg edema. J Oslo City Hosp 31(12):153–5, 1981.

65REFERENCES

Sugden EM, Rezvani M, Harrison JM, Hughes LK. Shoulder movement after thetreatment of early stage breast cancer. Clin Oncol (R Coll Radiol) 10(3):173–81,1998.

Tanis PJ, Deurloo EE, Valdes Olmos RA, Rutgers EJ, Nieweg OE, Besnard AP,Kroon BB. Single intralesional tracer dose for radio-guided excision of clinicallyoccult breast cancer and sentinel node. Ann Surg Oncol 8(10):850–5, 2001.

Tanis PJ, Nieweg OE, Hart AA, Kroon BB. The illusion of the learning phase forlymphatic mapping. Ann Surg Oncol 9(2):142–7, 2002.

Tasmuth T, von Smitten K, Kalso E. Pain and other symptoms during the first yearafter radical and conservative surgery for breast cancer. Br J Cancer 74(12):2024–31, 1996.

Tasmuth T, Kalso E. The postmastectomy pain syndrome. Neuropathic pain 3, 2003.

Temple LKF, Baron R, Cody HS, Fey JV, Thaler HT, Borgen PI, Heerdt AS,Montgpmery LL, Petrek JA, Van Zee KJ. Sensory morbidity after sentinel lymphnode biopsy and axillary dissection: a prospective study of 233 women. Ann SurgOncol 9(7):654–62, 2002.

Turner-Warwick RT. The lymphatics of the breast. Br J Surg. 46:574–82, 1959.

Urban JA. Management of operable breast cancer: the surgeon’s view. Cancer42(4):2066–77, 1978.

Uren RF, Howman-Giles R, Renwick SB, Gillet D. Lymphatic mapping of thebreast: locating the sentinel lymph nodes. World J Surg 25(6):789–93, 2001.

Valdés-Olmos RA, Jansen L, Hoefnagel CA, Nieweg OE, Muller SH, Rutgers EJ,Kroon BB. Evaluation of mammary lymphoscintigraphy by a single intratumoralinjection for sentinel node identification. J Nucl Med 41(9):1500–6, 2000.

Valdés Olmos RA, Tanis PJ, Hoefnagel CA, Nieweg OE, Muller SH, Rutgers EJ,Kooi ML, Kroon BB. Improved sentinel node visualization in breast cancer byoptimizing the colloid particle concentration and tracer dosage. Nucl Med Commun22(5): 579–86, 2001.

van der Ent FW, Kengen RA, van der Pol HA, Povel JA, Stroeken HJ, HoofwijkAG. Halsted revisited: internal mammary sentinel lymph node biopsy in breastcancer. Ann Surg 234(1): 79–84, 2001.

Veronesi U, Cascinelli N, Greco M, Bufalino R, Morabito A, Galluzzo D, ContiR, De Lellis R, Delle Donne V, Piotti P, et al. Prognosis of breast cancer pa-tients after mastectomy and dissection of internal mammary nodes. Ann Surg. 202:702–7, 1985.

Veronesi U, Rilke F, Luini A, Sacchini V, Galimberti V, Campa T, Dei Bei E,Greco M, Magni A, Merson M, et al. Distribution of axillary node metastases bylevel of invasion. An analysis of 539 cases. Cancer 59(4): 682–7, 1987.

Veronesi U, Paganelli G, Galimberti V, Viale G, Zurrida S, Bedoni M, Costa A,de Cicco C, Geraghty JG, Luini A, Sacchini V, Veronesi P. Sentinel-node biop-sy to avoid axillary dissection in breast cancer with clinically negative lymph-nodes.Lancet 349(9069):1864–7, 1997.

66 REFERENCES

Ververs JM, Roumen RM, Vingerhoets AJ, Vreugdenhil G, Coebergh JW, Crom-melin MA, Luiten EJ, Repelaer van Driel OJ, Schijven M, Wissing JC, VoogdAC. Risk, severity and predictors of physical and psychological morbidity after ax-illary lymph node dissection for breast cancer. Eur J Cancer 37(8): 991–9, 2001.

Voogd AC, Ververs JM, Vingerhoets AJ, Roumen RM, Coebergh JW, Crommel-in MA. Lymphoedema and reduced shoulder function as indicators of quality of lifeafter axillary lymph node dissection for invasive breast cancer. Br J Surg 90(1):76–81, 2003.

Warmuth MA, Bowen G, Prosnitz LR, Chu L, Broadwater G, Peterson B, LeightG, Winer EP. Complications of axillary lymph node dissection for carcinoma of thebreast: a report based on a patient survey. Cancer 83(7):1362–8, 1998.

Weiser MR, Montgomery LL, Susnik B, Tan LK, Borgen PI, Cody HS. Is routineintraoperative frozen-section examination of sentinel lymph nodes in breast cancerworthwhile? Ann Surg Oncol 7(9):651–5, 2000.

Yap KP, McCready DR, Narod S, Manchul LA, Trudeau M, Fyles A. Factorsinfluencing arm and axillary symptoms after treatment for node negative breastcarcinoma. Cancer 97(6): 1369–75., 2003.

Zippel D, Siegelmann-Danieli N, Ayalon S, Kaufman B, Pfeffer R, Zvi Papa M.Delayed breast cellulitis following breast conserving operation. Eur J Surg Oncol29(4): 327–30, 2003.

Zurrida A, Galimberti V, Monti S, Luini A. Radioguided localization of occult breastlesions. Breast 7: 11–13, 1998.

feasibility of sentinel node biopsy in breast cancer surgery · node biopsy (snb) in lymph node...

Documents