A systematic review of chyle leaks and their management following axillary surgery

A systematic review of chyle leaks and their management following axillary surgery

Journal Pre-proof A Systematic Review Of Chyle Leaks And Their Management Following Axillary Surgery N. Farkas, J. Wong, S. Monib, S. Thomson PII: S0...

1MB Sizes 0 Downloads 10 Views

Journal Pre-proof A Systematic Review Of Chyle Leaks And Their Management Following Axillary Surgery N. Farkas, J. Wong, S. Monib, S. Thomson PII:

S0748-7983(20)30064-0

DOI:

https://doi.org/10.1016/j.ejso.2020.01.029

Reference:

YEJSO 5622

To appear in:

European Journal of Surgical Oncology

Received Date: 9 December 2019 Revised Date:

8 January 2020

Accepted Date: 24 January 2020

Please cite this article as: Farkas N, Wong J, Monib S, Thomson S, A Systematic Review Of Chyle Leaks And Their Management Following Axillary Surgery, European Journal of Surgical Oncology, https://doi.org/10.1016/j.ejso.2020.01.029. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Crown Copyright © 2020 Published by Elsevier Ltd. All rights reserved.

1

A Systematic Review Of Chyle Leaks And Their Management Following Axillary Surgery

2

N Farkas*, J Wong, S Monib, S Thomson

3

West Hertfordshire Breast Care Unit, West Hertfordshire Hospitals NHS Trust

4 5 6

* corresponding author

7

Abstract

8

Introduction:

9 10 11 12

Chyle leaks following surgery to the axilla are seldom encountered with an incidence <0.7%. Management varies with no consensus in the literature. Injury to branching tributaries of the thoracic duct may require lengthy management at significant cost to patient and clinical team. This paper aims to provide an up-to-date review to support clinical management.

13 14

Methods:

15 16 17

The term 'chyle' was combined with 'breast' or 'axilla.' EMBASE, Medline and PubMed database searches were conducted. All papers published in English were included with no exclusion date limits.

18 19

Results:

20 21 22 23 24 25 26 27 28

51 cases from 31 papers. All were female (mean age=53.3yrs). 47/51 leaks were left-sided. 5/51 underwent sentinel node biopsy, 19/51 level II axillary node clearance (ANC), 23/51 level III ANC, 5/51 not specified. 59% (30/51) of leaks were identified within 2 postoperative days (mean=3.3days). 96% initially managed conservatively: Drain=38/51; low-fat diet=34/51; compression bandaging=20/51; Aspiration=6/51. 40/51 (78%) were successfully managed conservatively, 11 patients returned to theater for secondary management. 7/11 recorded volumes >500mls/24hrs before secondary surgery. Mean resolution time from initial surgery was 17.3days (range=4-64days). No statistically significant difference (p=0.72) in time to resolution between conservatively and surgically managed patients.

29 30

Conclusions:

31 32 33 34

Chyle leaks are rarely seen following axillary surgery. Aberrant thoracic duct anatomy represents the likeliest etiology. We advocate early recognition and tailored individual management. Conservative management with non-suction drainage, low-fat diet and axillary compression bandaging appear effective where output <500ml/24hrs. Secondary surgical management should be considered in

35 36

high chylous output (<500mls/24hrs) patients unresponsive to conservative measures. We propose a management algorithm to aide clinicians.

37 38

Introduction:

39 40 41 42 43 44 45

Chyle leaks following surgery to the axilla are seldom encountered with an incidence of <0.7%. More commonly seen are chylous leaks following head or neck surgery with a reported incidence of between 0.5-1.4% of thyroidectomies and 2-8% of neck dissections(1). Management of chyle leaks following axillary surgery varies with no consensus in the literature. Options range from conservative measures to further explorative surgery. The purpose of this paper is to provide an upto-date review and to determine whether there are any trends from the documented cases to assist clinicians in future management.

46 47 48 49 50

Chyle is a milky fluid made up of emulsified fats (long chain triglycerides) and lymph derived from the absorption of fats from the small intestine into the lymphatic system. Chyle is transported via the lymphatic system to the thoracic duct that empties at the level of the left brachiocephalic vein into the bloodstream(2). Although transported in the same lymphatic system as lymph its composition differs due to the high fat content absorbed from the gut.

51 52 53

Injury to branching tributaries of the thoracic duct may require lengthy management with significant cost to both patient and clinical team. Early recognition and commencement of appropriate management will help reduce morbidity and improve patient outcomes.

54 55

Methodology:

56 57 58 59 60 61

The search term ‘chyle’ was combined with ‘breast’ or ‘axilla.’ Multiple database searches of EMBASE, Medline and PubMed were conducted. All papers published in English were included with no exclusion date limits. In addition to the database searches, hand searches were performed using Google Scholar with the same search terms. The first 50 hand search results were included for screening. The paper was registered with the International Prospective Register of Systematic Reviews, (PROSPERO) (registration=CRD42019155177).

62 63 64 65

Figure 1: PRISMA Diagram of Search Strategy

Identification

66

Electronic Database searches: Medline, Pubmed, Embase (n = 180) Hand searches (n = 50)

Included

Eligibility

Screening

Records after duplicates removed (n = 125)

Titles/ Abstracts screened (n = 125)

Records excluded (n = 89)

Full-text articles assessed for eligibility (n = 36)

Full-text articles excluded, no axillary procedure done (n = 5)

Studies included in quantitative synthesis (meta-analysis) (n = 31 (including 4 from hand searches))

67 68 69

In order to minimise bias, two independent reviewers analysed the searches and selected those papers deemed appropriate for the study based on the inclusion criteria, paper abstracts and content.

70 71 72 73 74

The selected papers were analysed for multiple outcomes relating to side, sex, age, type of surgery, level of axillary clearance (if applicable), drain output, subsequent management, return to theatre (days), length of stay and mortality. Statistical analysis with the Mann Witney U test was undertaken where sufficient data existed, namely in the comparison of conservative and surgical treatment with respect to time to resolution.

75

Results:

76 77 78 79 80

Of the 125 papers derived from the searches, 89 papers were excluded once duplicates had been removed. 36 full texts were reviewed, of which a further 5 were subsequently excluded. Two of these papers reported chylous leaks following cosmetic breast surgery, but no axillary procedure had been undertaken. Therefore, a total of 31 papers were included. Figure 1 highlights the results of our search strategy.

81 82 83 84

A total of 51 cases from the 31 papers were described, as shown in Table 1. All 51 cases were female, with the mean age being 53.3 years. 47/51 (92%) of chyle leaks occurred after left sided axillary surgery. No intra-operative complications were reported, with 1 patient returning to theatre for haematoma evacuation in the immediate post-operative phase.

85

Table 1: Demographics, Procedure, Staging, Nodal Involvement and Diagnostic Tests

Author

Age

Sex

Side

Initial Surgery

Axillary Procedure

Tumour Staging

Nodes

Diagnosis

Abdelrazeq et al(3)

48

F

L

WLE and ANC

II

Grade II IDC

0 of 19

Al-Ishaq et al.(4)

41

F

L

Mastectomy and SLNB

SNB

Grade II IDC

0 of 4

Lymphoscintigraphy and biochemical Biochemical

Baek et al.(5)

38

F

L

Modified radical mastectomy

not stated

Grade III IDC

1 of 15

Biochemical

Caluwe et al.(6)

53

F

L

Bilateral lumpectomy and ANC

II

T1N0M0 IDC

0 of 19

Biochemical

Chan et al.(7)

53

F

L

Mastectomy and ANC

II

IDC

6 of 19

Clinical only

Chow et al.(8)

47

F

L

II

Not stated

Not stated

Biochemical

Cong et al.(9)

44

F

L

Skin-reducing mastectomy and ANC and immediate DIEP reconstruction Mastectomy and ANC

III

T2N0M0

Not stated

Biochemical

38

F

L

Mastectomy and ANC

III

T2N0M0

Not stated

Biochemical

39

F

L

Mastectomy and ANC

III

T1N1M0

Not stated

Biochemical

42

F

R

Mastectomy and ANC

III

T2N0M0

Not stated

Biochemical

48

F

L

Mastectomy and ANC

III

T2N1M0

Not stated

Biochemical

65

F

R

Mastectomy and ANC

III

T1N1M0

Not stated

Biochemical

Daggett et al.(10)

41

F

R

II

Not stated

2 of 4

Biochemical

Donkervoort et al.(11)

53

F

L

III

Grade III IDC

9 of 19

Biochemical

Flores et al.(12)

55

F

L

II

T3N2M1 IDC

6 of 12

Biochemical

Griffiths et al.(13)

81

F

L

Bilateral mastectomies with right SNB and immediate reconstruction Lumpectomy and axillary dissection Mastectomy and axillary node dissection WLE and ANC

III

Grade II IDC

9 of 24

Biochemical

Lagarde et al.(14)

56

F

R

SNB

SNB

Grade III IDC

0 of 2

Biochemical

Malik et al.(15)

76

F

L

Mastectomy with SNB

SNB

T1N0M0 IDC

Not stated

Biochemical

Nakajima et al.(16)

74

F

L

Mastectomy and ANC

III

IDC

Level I

Biochemical

30

F

L

Partial resection and ANC

II

IDC

Not stated

Biochemical

57

F

L

Partial resection and ANC

II

IDC

Level II

Biochemical

48

F

L

Partial resection and ANC

II

IDC

Not stated

Clinical only

Oba et al.(17)

69

F

L

Mastectomy and ANC

II

IDC Stage IIA

Biochemical

Pointer et al.(18)

75

F

L

Wire guided WLE with SNB

SNB

Not stated

Purkayastha et al.(19)

56

F

L

III

IDC

Lymphoscintigraphy and biochemical Biochemical

Rice et al.(20)

47

F

L

Modified radical mastectomy with ANC Modified radical mastectomy with ANC

2 metastatic nodes 3 sentinel nodes removed Not stated

III

Grade IV IDC

1 of 10

Biochemical

46

F

L

Modified radical mastectomy with ANC Modified radical mastectomy with ANC Modified radical mastectomy with ANC Modified radical mastectomy with ANC

III

T1N0 + DCIS

0 of 12

Biochemical

74

F

L

III

T1N0 + DCIS

0 of 11

Biochemical

43

F

L

III

T2N0

0 of 22

Biochemical

44

F

L

III

1 of 21

Biochemical

III

11 of 15

Biochemical

II

T2N0M0

Not stated

Biochemical

L

Modified radical mastectomy with ANC Modified radical mastectomy with ANC Axillary clearance

Multifocal lobular carcinoma T2N2

47

F

L

Sakman et al.(22)

65

F

L

Sales et al.(23)

21

F

Singh et al.(24)

56

F

III

Not stated

0 of 17

Biochemical

L

Axillary clearance

III

T3N1M0

Not stated

Biochemical

47

F

L

Axillary clearance

III

T1N0M0

Not stated

Biochemical

33

F

L

Axillary clearance

III

T1N0M0

Not stated

Biochemical

59

F

L

Axillary clearance

III

T4N1M0

Not stated

Biochemical

56

F

L

Axillary clearance

III

T2N2M0

Not stated

Biochemical

50

F

L

Axillary clearance

III

T2N0M0

Not stated

Clinical only

Tan et al.(25)

48

F

L

SNB

IDC

2 of 3

Biochemical

Taylor et al.(26)

82

F

L

Nipple sparing mastectomy and SNB Mastectomy and ANC

ANC

Not stated

Biochemical

Thang et al.(27)

78

F

L

Mastectomy and ANC

ANC

Multicentric lobular carcinoma Not stated

Not stated

Lymphoscintigraphy

Wong et al.(28)

51

F

L

II

IDC

Not stated

Biochemical

Zhou et al.(29)

70

F

L

Mastectomy, ANC and latissimus dorsi flap reconstruction Mastectomy and ANC

II

IDC

3 of 13

Biochemical

44

F

L

Mastectomy and ANC

II

IDC

0 of 15

Biochemical

42

F

L

Breast conserving and ANC

II

IDC

1 of 26

Biochemical

37

F

L

Mastectomy and ANC

II

IDC

0 of 33

Biochemical

Haraguchi et al.(30)

71

F

L

II

T2N0M0

0 of 18

Biochemical

di Summa et al.(31)

F

L

Not stated

IDC

Not stated

Biochemical

Gonzalez et al.(32)

Not stated 72

F

L

Modified radical mastectomy and axillary dissection Mastectomy, ANC and immediate DIEP reconstruction Modified radical mastectomy

ANC

T2N2M0

Not stated

Biochemical

Fothiadaki et al.(33)

55

F

L

II

Grade II IDC

7 of 22

Biochemical

II (19/51) III (23/51) SNB (5/51) ANC (3/51) Not stated (2/51)

Stated (47/51) Not stated (4/51)

Stated (28/51) Not stated (23/51)

Biochemical (46/51) Lymphoscintigraphy (1/51) Clinical only (2/51) Lymphoscintigraphy and biochemical (2/51)

Rijken et al.(21)

Avg. (n=50 )53.3

F L (51/51) (47/51) R (4/51)

Central quadrantectomy and axillary dissection

WLE = Wide local excision, ANC = Axillary node clearance, SNB = Sentinel node biopsy, DIEP = deep inferior epigastric perforators

86 87 88 89

The majority of patients (82%) underwent either a level II (19/51) or III (23/51) axillary clearance. 5 patients underwent a sentinel node biopsy (SNB). Under half of reports specified the status of the nodes (25/51). Of those detailed, 56% had metastatic nodes (14/25), whilst 44% (11/25) were clear.

90 91 92

59% (30/51) of chyle leaks were identified within 2 days of initial surgery as demonstrated by Figure 2. The mean time for identifying a chyle leak was 3.3 days post initial surgery. However, leaks were identified up to 14 days following surgery.

93 94

Chyle leak diagnosis was most commonly made using biochemical analysis 42/51 (82%), whilst lymphoscintigraphy was utilised in 4 cases, and clinically alone in 8.

95 96

Figure 2: Day Chyle Leak Identified

97 98 99 100 101

Table 2 demonstrates that the majority of patients (96%) initial management was conservative. Only 2 cases were immediately managed with surgery as the leaks were identified intraoperatively. The most popular conservative management modalities were the use of a drain 38/51 (75%), low fat diet 34/51 (67%) and compression bandaging 20/51 (39%). Aspiration was reported in 6/51 (12%).

102 103

Conservative management was successful in 40/51 (78%), whilst 11 patients returned to theatre (22%) for secondary management.

104 105 106 107

The chylous volumes of all patients following initial surgery are documented in Table 2. 11 of 51 returned to theatre, with most authors recording chylous volumes greater than 500mls/24hrs. 2 of 40 cases that were successfully managed conservatively reported a chylous volume of >500mls/24hrs.

108 109

Table 3 outlines the secondary management utilised for the 11 patients who returned to theatre. The respective time courses from initial surgery and the date of chyle leak identified are also stated.

Table 2: Management Variable

Day Chyle leak identified (post operative day)

Drain output

Abdelrazeq

7

Al-Ishaq et al.

2

Baek et al.

4

400-700mls/ 48hrs for 2 weeks 200mls/ 48hrs next 4 weeks 180mls/ 24hrs decreasing to 100mls/ 24hrs for 4 weeks 100mls/ 24hrs decreasing to 50mls/ 24hrs over subsequent 2 weeks Up to 700mls/ 24 hrs for first 3 days

Caluwe et al.

1

Chan et al.

4

Chow et al.

5

Cong et al.

Primary management

Secondary management (Yes/ No)

Time to resolution from day chyle leak identified (days)

Time to resolution from initial surgery (days)

No

37

44

No

40

42

Yes

12

16

210ml/ 24hrs initially, 235mls (210-260mls)/ 24hrs Week 1 23mls (15-60mls)/ 24 hrs Week 2 125mls (90-225mls)/ 24 hrs Week 3 6.5mls (0-15mls)/ 24 hrs Week 4 Maximum 300mls/ 24hrs

No

26

27

No

6

10

No

30

35

2

250mls/ 24 hrs from POD 5 until 2 weeks when drain fell out 170mls (120-250mls)/ 24hrs from POD 2

No

2

170mls (120-250mls)/ 24hrs from POD 2

No

2

170mls (120-250mls)/ 24hrs from POD 2

No

2

170mls (120-250mls)/ 24hrs from POD 2

No

2

170mls (120-250mls)/ 24hrs from POD 2

No

2

170mls (120-250mls)/ 24hrs from POD 2

No

Daggett et al.

1

No

Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) 11

Donkervoort

1

No

6

7

Flores et al.

5

80mls/ 8hrs in right drain 30mls/ 8hrs in left drain Subsequently low output 170-210mls/ 24hrs for POD 1-6 100mls/ 24hrs from POD 7 230mls/ 24hrs initially,

Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) Median of 5 days (range: 3-7 days) 10

No

15

20

Conservative

Compression

Drain

Aspiration

Low fat diet

Theatre

90mls/ 24hrs after fat-free diet Griffiths et al.

2

Lagarde et al.

6

Malik et al.

9

Nakajima et al.

3 3

2

2

Oba et al. Pointer et al.

14 Not stated

Purkayastha et al. Rice et al.

1

Rijken et al.

1

1

4 1 2 1

400mls/ 24hrs on POD 2 400-500mls/ 24hrs on POD 3-7 Milky fluid aspirated POD 6

No

5

7

Yes

58

64

2L aspirated POD 9 10L aspirated over next 3 weeks 60mls/ 24hrs

Yes

36

45

No

1

4

80mls/ 24hrs POD 3 100mls/ 24hrs POD 4 90mls/ 24hrs POD 5 60mls/ 24hrs POD 6 60mls/ 24hrs POD 2 40mls/ 24hrs POD 3 30mls/ 24hrs POD 4 80mls/ 24hrs POD 2 60mls/ 24hrs POD 3 20mls/ 24hrs POD 4 60mls aspirated POD 7, 20mls aspirated POD 8, 250mls aspirated POD 15, 180mls aspirated POD 18 15mls aspirated on POD 14, 400mls aspirated on POD 21 800mls/ 24hrs initially, decreased to 200mls/ 24hrs after 4 days. Reduced further to 60mls/ 24hrs by day 6. Increased to 200mls after restarting normal diet 1000mls/ 24hrs

No

4

7

No

3

5

No

16

18

No

10

31

Yes

22

23

Yes

17

18

No

5

6

No

8

9

No

11

15

No

8

9

No

12

14

No

13

14

275mls/ 24hrs on POD 2, Output reduced over subsequent 5 days Max 200mls/ 24hrs, total drain production 960mls Max 360mls/ 24hrs, total drain production 2970mls Max 410mls/ 24hrs, total drain production 1400mls Max 260mls/ 24hrs, total drain production unknown Max 240mls/ 24hrs, total drain production 1230mls

Sakman et al.

1

No

2

350mls/ 24hrs POD 1, decreasing drainage over subsequent 4 days Mean of 341mls/ 24 hrs from POD 2-8,

Sales et al. Singh et al.

6

7

1

1000mls/ 24hrs for 14 days

No

9

10

Yes

17

18

2

Median drain output of 350mls/ 24hrs

No

10

12

2

Median drain output of 400mls/ 24hrs

No

12

14

2

Median drain output of 200mls/ 24hrs

No

13

15

Intraoperative

Median drain output of 125mls/24hrs

No

12

12

Intraoperative

0

No

0

Tan et al.

1

50mls/ 24hrs

Yes

Resolved following surgery Not stated

Not stated

Taylor et al.

11

120mls/ 24hrs POD 11

No

1

12

Thang et al.

5

Not stated

Yes

Not stated

Not stated

Wong et al.

10

1000mls/ 24hrs

No

6

16

Zhou et al.

4

<500mls/ 24hrs

No

Between 7-34 days

Between 7-34 days

8

<500mls/ 24hrs

No

Between 7-34 days

Between 7-34 days

3

<500mls/ 24hrs

No

Between 7-34 days

Between 7-34 days

Not stated

1

>700mls/ 24hrs for 7 days

Yes

17

18

Haraguchi et al.

5

Yes

34

39

di Summa et al. Gonzalez et al.

3 10

318mls/ 24hrs 200-780mls/ 24hrs POD 6-29 Not stated Not stated

Yes No

Not stated 12

Fothiadaki et al.

1

50mls/ 24 hrs initially, 95mls (30-150mls)/ 24hrs POD 2-14

No

13

POD = Post-operative day, N/A = Not applicable, hrs = hours

Not stated 22 14

Table 3: Secondary Management Author

Secondary Management

Time secondary management implemented from date chyle leak identified (days)

Time to resolution following secondary management (days)

Baek et al.

Ligation of single duct inferior to lateral pectoral bundle branch

7

9

Lagarde et al.

Return to theatre POD 19 - two lymphatic branches clipped

19

45

Malik et al.

Return to theatre POD 37 - Suture ligation of infero-lateral tissue on chest wall

28

8

Pointer et al.

Return to theatre POD 7 - ligation of lymphactics along superolateral aspect of cavity

5

17

Purkayastha et al.

Return to theatre, mass ligation using 2.0 silk of leaking duct posterior to axillary vein and rotation of pectoralis major muscle

14

4

Singh et al.

Return to theatre, ligation of single duct inferior to axillary vein and rotation of pectoralis major muscle flap

14

4

Tan et al.

Return to theatre for axillary clearance as nodes positive and on going chyle leak - ligation of duct

Not stated

Resolved following surgery

Thang et al.

Re-exploration, 800ml of chyle drained. Tissue glue to cavity inferior to biceps tendon, surgical closure and compression dressing.

Not stated

Not stated

Zhou et al.

Re-exploration and ligation of leaking lymph ducts in lower skin flap

7

11

Haraguchi et al.

Return to theatre POD 29 for suture ligation of lymphatic vessel lateral to pectoralis major and fibrin glue

24

10

di Summa et al.

Re-exploration and washout

POD = Post-operative day

3

Not stated

Average (n=9) 13.1

Average (n=8) 13.5

Table 4 outlines the comparison of data between those managed conservatively and those requiring secondary surgical management. The overall mean time to resolution from initial surgery was 17.3 days with a range of 4-64 days. The mean time to resolution from chyle leak identification was 14.2 days (range 1-58 days). Those patients requiring secondary management on average returned to theatre 14.4 days (range 3-19 days) following the identification of a chyle leak. On average chylous leakage resolved 13.5 days after secondary surgical management, whereas the average time to resolution of those managed conservatively was 11.5 days. Table 4: Conservative vs. Surgical Management of Chyle Leaks Following Axillary Surgery

Conservative Management

Time To Resolution From Initial Surgery (Days)

Time To Resolution From Date Chyle Leak Identified (Days)

14.5

11.5

Time Secondary Management Implemented From Date Chyle Leak Identified (Days)

Time To Resolution Following Secondary Surgical Management (Days)

14.4

13.5

Secondary Surgical Management

36.9

28.3

Overall

17.4

14.3

Range

4-64

1-58

3-19

4-45

43 stated, 8 not stated

43 stated, 8 not stated

8 stated, 3 not stated

9 stated, 2 not stated

Statistical analysis between the two groups with the Mann Whitney U test was undertaken. The time to resolution from date leak identified in those managed conservatively was compared with time to resolution following secondary surgical management. No significant difference was demonstrated (p<0.72).

Discussion: Reports of chyle leakage following axillary surgery are rare. Zhou et al quote a chyle leak incidence of 0.36% (4 of 1096 patients)(29). Other authors have comparable figures; Nakajima 0.47% (4/851)(16), Cong 0.68% (4/882)(9). We acknowledge that the majority of the data relates to case reports and is therefore somewhat subjective and anecdotal. Likewise, the clarity and depth of reporting has significant discrepancies between authors meaning some data is missing. Previous literature reviews have been conducted(10)(33), however, none as recently as our review nor with the number of cases included. All of the cases in our review related to female patients, with a mean age of 53.3yrs. This is not unexpected given that male breast cancers make up only 1% of new cases(34), whilst post menopausal women represent the majority of women with breast cancer(35). Chylous leaks may occur following traumatic injury or surgery. Iatrogenic injuries to the thoracic duct during thoracic or head and neck surgery most commonly account for this rare sequela. Axillary surgery is rarely associated with chylous leaks owing to the anatomically more remote position of the thoracic duct. However, as evidenced by the results, chylous leaks are not confined purely to the left side. Anatomical variants of the thoracic duct are well documented in the literature(36). This is

not unsurprising given that the typical anatomy is present in only 50% of people. In 2-3% the duct may empty on the right and bilateral emptying occurs in 1.5%(37). The type of axillary procedure may play a causative role in determining which patients might sustain a chyle leak. As evidenced by the results, only 5 patients had a chylous leak following sentinel node biopsy compared to the vast majority undergoing axillary clearance. One may postulate that a more invasive procedure such as a level III axillary clearance may carry a higher risk of injury. This may be due to the anatomical proximity of relevant structures to possible variant lymphatics arising from the thoracic duct, when compared to a sentinel lymph node biopsy. Nodal involvement does not appear to be linked to a higher likelihood of a leak occurring. Both patients with negative and positive nodes suffered chyle leaks with similar numbers reported in each group. Likewise, the stage of cancer does not appear to be a causative factor. More patients with T1 disease incurred leaks compared to those with T4 disease. These factors therefore point towards other aetiology associated with chylous leaks e.g. aberrant anatomy of the thoracic duct and its branching tributaries. The diagnosis of a chyle leak is primarily a clinical one. The continued presence of milky fluid in a postoperative drain following axillary surgery should raise the possibility. Nevertheless, clinicians may want confirmation. Two diagnostic modalities are available to determine the cause of such fluid: biochemical analysis and lymphoscintigraphy. Biochemical analysis is non-invasive, quick and cheap whereas lymphoscintigraphy carries the risks of radiation, contrast dye and a time consuming investigation(3). Neither offers therapeutic management. Authors highlight the diagnostic accuracy of utilising SPECT in conjunction with lymphoscintigraphy to determine more precisely the location of a leak(27). However, the benefit of utilising lymphoscintigraphy as a diagnostic tool is unclear, as management may not alter significantly following its use. The majority of chyle leaks were diagnosed within the first 48 hours of axillary surgery. This would suggest iatrogenic disruption or injury to chylous channels at the time of surgery. Nonetheless, delayed presentations up to 14 days post surgery are evidenced by our data. It is not unsurprising that some leaks are missed at the time of initial surgery given that patients will have been starved for a number of hours. Consequently chyle reabsorption is likely to be low. Given this factor, we advocate a systematic check of the axillary cavity before closing. No major intraoperative complications were reported in any of the papers. This highlights that chylous leaks can be insidious and subtle, with no clear and obvious source necessarily located intraoperatively. One postoperative haematoma following surgery required a return to theatre. Chow et al report a left mastectomy and level II axillary clearance followed by immediate DIEP flap reconstruction resulting in a 600ml axillary haematoma 6 hours post surgery(8). The authors point to a trainee surgeon carrying out the axillary clearance as a possible factor for a subsequent chyle leak, which was managed conservatively. In a report excluded prior to analysis, Tam et al describe a 34-year-old woman who underwent removal and replacement of breast implants but returned 7 days after surgery with a left sided haematoma. Interestingly, this patient subsequently returned to theatre again and was found to have a right-sided chylous leak at the medial edge of the capsule diagnosed 28 days post initial

surgery. The authors postulate that this leak was as a result of aberrant anatomy associated with a lymphatic channel accompanying an intercostal vessel(38). The majority of patients were initially managed conservatively with a range of modalities utilised. A drain left in situ following axillary surgery appears to be standard practice as documented by various authors. Therefore, utilisation of this conduit by most is not unexpected. Non-suction drains appear most favoured. The prevention of fluid accumulation without increasing the pressure gradient between body cavity and exterior averts increasing chylous output. Many authors document the introduction of a low fat diet in conjunction with a drain or other conservative measures in order to slow chylous output. A diet containing medium chain triglycerides (MCT) is advocated, as these are thought to be absorbed directly across the mucosa without the need for lymphatic transport(39). However, in some instances total parental nutrition (TPN) may be required when a patient is unable to tolerate such a diet(40). A review by McCray suggests dietary modification plays a role in treating some patients. However, the authors do highlight a number of drawbacks including cost, palatability, compliance and unclear endpoints(39). In our review aspiration was utilised relatively infrequently when compared to other conservative management options, not usually being necessary with a drain in situ. Needle aspiration carries an infection risk each time the procedure is undertaken, and is also potentially painful. Such management may be appropriate in those patients who cannot tolerate a drain e.g. patients with delirium or dementia. Some authors advocate the use of somatostatin analogues such as octreotide to help reduce gastric motility and secretions in order to decrease chylous output(1). Jain et al report successful outcomes using octreotide in the context of neck chylous fistula after other conservative measures had failed(41). In the axillary setting this appears as an appropriate adjunct with other conservative measures(32). In just 2 of the cases successfully managed conservatively were chylous volumes greater than 500mls over 24hrs reported(3)(28). In comparison, of those that returned to theatre for secondary management, 7 of the 9 reports recorded a volume greater than 500mls over 24 hours. Evidently chyle volumes do not necessarily determine the success or failure of conservative management but do appear to play a role in clinical decision-making. Clinicians appear eager to avoid a return to theatre, as evidenced by a 2-week period (on average) where conservative management is attempted. Understandably, surgery and associated anaesthesia carry risk in co-morbid patients on low fat diets. Of the 11 patients that returned to theatre all leaks were identified and managed intraoperatively. However, as documented in other surgeries, methylene blue dye represents a cheap adjunct without contraindication that could be utilised to help identify a leak not seen with the naked eye. Subsequent ligation can then be undertaken at the site of blue pigmentation seen within the cavity. Kapila et al highlight its effectiveness in helping to distinguish between anastomotic and chylous leaks following oesophagectomy(42). Other dyes including patent blue and Sudan black B could also be used(43).

Conservatively managed patients’ symptoms resolved on average 11.5 days after chyle leak identification, whereas those requiring secondary management in theatre resolved on average 13.5 days after identification. These findings suggest that conservative management appears effective in those patients with lower output leaks (<500mls/24hrs), offering faster recovery times. Nevertheless, caution should be taken when interpreting these results, as without a return to theatre some patients’ recoveries are likely to have been much longer. Equally, a return to theatre does not guarantee a quick resolution of all chyle leaks. Recovery times appear lengthy in both subgroups with some patients taking months to settle. No significance in time to resolution was demonstrated between those managed conservatively and those returning for secondary surgery. The inconvenience of regular monitoring of a drain, aspirations, compression bandaging and dietary modification should not be discounted. Such patients may require specialist input on a regular basis from nutritionists, nurse specialists and surgeons. No mortalities were reported following axillary surgery. However, it is not inconceivable given the patient profile undergoing such procedures that such an outcome could occur if managed inappropriately. We suggest a management algorithm (Figure 3) to help clinicians who encounter such a presentation. Although it is difficult to draw a definitive consensus when managing such patients given the small numbers, it may offer a guide around which to tailor management.

Figure 3: Suggested Management Algorithm

Conclusion: Chyle leaks are rarely seen following axillary surgery and represent an infrequent clinical presentation encountered by breast surgeons. Our study has considered many factors associated with chyle leaks. Left sided and more invasive surgeries may be implicated in chylous leaks. However, aberrant anatomy presents the likeliest aetiology in the development of such a leak. We advocate early recognition of a leak where possible and tailoring management on an individual basis.

Conservative management appears effective in controlling chylous leaks where daily volumes are less than 500ml every 24 hours. Given the potential risks of axillary surgery and associated anaesthesia we support the use of conservative measures with non-suction drainage, low fat diet and compression bandaging to the axilla as first line management. Secondary surgical management should be considered in patients who fail to respond to conservative treatment. Given the complexities of such patients, we advocate a multidisciplinary team approach with regular input from surgeons, nurse specialists and nutritionists. We propose a management algorithm to guide other clinicians when encountering this challenging and unusual presentation.

Figures Figure 1: PRISMA Diagram of Search Strategy ........................................................................................ 3 Figure 2: Day Chyle Leak Identified......................................................................................................... 6 Figure 3: Suggested Management Algorithm ....................................................................................... 15

Tables Table 1: Demographics, Procedure, Staging, Nodal Involvement and Diagnostic Tests ....................... 4 Table 2: Management ............................................................................................................................ 7 Table 3: Secondary Management ......................................................................................................... 10 Table 4: Conservative vs. Surgical Management of Chyle Leaks Following Axillary Surgery ................ 11

References: 1.

Delaney SW, Shi H, Shokrani A, Sinha UK. Management of chyle leak after head and neck surgery: review of current treatment strategies. Int J Otolaryngol. Hindawi; 2017;2017.

2.

Jacob S. Chapter 3 - Thorax. In: Jacob SBT-HA, editor. Churchill Livingstone; 2008. p. 51–70. Available from: http://www.sciencedirect.com/science/article/pii/B9780443103735500063

3.

Abdelrazeq AS. Lymphoscintigraphic demonstration of chylous leak after axillary lymph node dissection. Clin Nucl Med. LWW; 2005;30(5):299–301.

4.

Al-Ishaq Z, Gupta S, Collins MA, Sircar T. Chyle leak following an axillary sentinel lymph node biopsy for breast cancer in a patient with superior vena caval thrombosis–a case report and review of the literature. Ann R Coll Surg Engl. Royal College of Surgeons; 2018;100(6):e147–9.

5.

Baek JM, Lee JA, Nam YH, Sung GY, Lee DS, Won JM. Chylous leakage: a rare complication after axillary lymph node dissection in breast cancer and surgical management. J Breast Cancer. 2012;15(1):133–4.

6.

Caluwe GLF, Christiaens M-R. Chylous leak: a rare complication after axillary lymph node dissection. Acta Chir Belg. Taylor & Francis; 2003;103(2):217–8.

7.

Chan AC, Sarojah A. Chylous leakage post mastectomy and axillary clearance: clinical aspect, causes and review of literature. Med J Malaysia. Citeseer; 2013;68(3):262–3.

8.

Chow WTH, Rozen WM, Patel NG, Ramakrishnan V V. Chyle leak after axillary lymph node dissection. J Plast Reconstr Aesthetic Surg. Elsevier; 2015;68(5):e105–6.

9.

Cong M, Liu Q, Zhou W, Zhu J, Song C, Tian X. Six cases of chylous leakage after axillary lymph node dissection. Oncol Res Treat. Karger Publishers; 2008;31(6):321–4.

10.

Daggett JD, Watt AW, Smith PD. Chyle leak following right axillary lymph node dissection: A case report and review of current literature. Int J Surg Case Rep [Internet]. 2016/01/07. Elsevier; 2016;20:68–73. Available from: https://www.ncbi.nlm.nih.gov/pubmed/26826928

11.

Donkervoort SC, Roos D, Borgstein PJ. A case of chylous fistula after axillary dissection in breast-conserving treatment for breast cancer. Clin Breast Cancer. Elsevier; 2006;7(2):171–2.

12.

Flores E V, de Castro G, Casal E, Sobrino C. Chylous fistula following axillary lymphadenectomy. Cir Esp. 2014;92(1):55.

13.

Griffiths O, Adjogatse J. Chyle leak: A rare complication of Axillary Lymph Node Clearance– Case report. Int J Surg. Elsevier; 2017;47:S20.

14.

Lagarde SM, Tanis PJ, van der Meij S, Rutgers EJ. Chylous fistula after internal mammary chain sentinel node biopsy. Am Surg. Southeastern Surgical Congress; 2010;76(11):1309.

15.

Malik M, Ruiz DE, Annamalai A, Lee S, Karsif K. Chyle fistula after simple mastectomy and sentinel lymph node biopsy: a rare occurrence. Breast J. Blackwell Publishing Ltd Oxford, UK; 2012;18(5):488–90.

16.

Nakajima E, Iwata H, Iwase T, Murai H, Mizutani M, Miura S, et al. Four cases of chylous fistula after breast cancer resection. Breast Cancer Res Treat. Springer; 2004;83(1):11–4.

17.

Oba T, Ono M, Iesato A, Hanamura T, Watanabe T, Ito T, et al. Chylous leakage after axillary lymph node dissection in a patient with breast cancer. Breast J. Wiley Online Library;

2018;24(3):438–40. 18.

Pointer Jr DT, Durgan DM, Kis B, Khakpour N, Kiluk J V. High-output chyle leak after breastconserving surgery and sentinel lymph node biopsy. Breast J. Wiley Online Library; 2019;

19.

Purkayastha J, Hazarika S, Deo SVS, Kar M, Shukla NK. Post-mastectomy chylous fistula: Anatomical and clinical implications. Clin Anat Off J Am Assoc Clin Anat Br Assoc Clin Anat. Wiley Online Library; 2004;17(5):413–5.

20.

Rice DC, Emory JRE, McIlrath DC, Meland NB. Chylous fistula: an unusual occurrence after mastectomy with immediate breast reconstruction. Plast Reconstr Surg. 1994;93(2):399–401.

21.

Rijken A, Chaplin BJ, Rutgers EJT. Chyle in the drain after modified radical mastectomy: an easy manageable problem. The Breast. Elsevier; 1997;6(5):299–300.

22.

Sakman G, Parsak CK, Demircan O. A rare complication in breast cancer surgery: chylous fistula and its treatment. Acta Chir Belg. Taylor & Francis; 2007;107(3):317–9.

23.

Sales F, Trepo E, Brondello S, Lemaitre P, Bourgeois P. Chylorrhea after axillary lymph node dissection. Eur J Surg Oncol. Elsevier; 2007;33(8):1042–3.

24.

Singh M, Deo SVS, Shukla NK, Pandit A. Chylous fistula after axillary lymph node dissection: incidence, management, and possible cause. Clin Breast Cancer. Elsevier; 2011;11(5):320–4.

25.

Tan B, Bloom R. Chylous fistula after sentinel lymph biopsy. J Plast Reconstr Aesthetic Surg. 2016;6(69):873–4.

26.

Taylor J, Jayasinghe S, Barthelmes L, Chare M. Chyle Leak Following Axillary Lymph Node Clearance–a Benign Complication: Review of the Literature. Breast Care. Karger Publishers; 2011;6(2):130–2.

27.

Thang SP, Tong AKT, Ng DCE. Postmastectomy/Axillary Node Dissection Chyloma: The Additional Value of SPECT/CT Lymphoscintigraphy. J Breast Cancer. 2014;17(3):291–4.

28.

Wong K, Elfaki A, Irwin M. Chyle Leak After Breast Surgery. Br J Surg. 2015;102.

29.

Zhou W, Liu Y, Zha X, Zheng W, Chen L, Xie X, et al. Management of chylous leakage after breast surgery: report of four cases. Surg Today. Japan; 2011 Dec;41(12):1639–43.

30.

Haraguchi M, Kuroki T, Tsuneoka N, Furui J, Kanematsu T. Management of chylous leakage after axillary lymph node dissection in a patient undergoing breast surgery. The Breast. Elsevier; 2006;15(5):677–9.

31.

di Summa PG, Roos E, Tay SK. Chylous leak: an unexpected complication after microsurgical breast reconstruction. Gland Surg. AME Publications; 2018;7(5):496.

32.

González-Sánchez-Migallón E, Aguilar-Jiménez J, García-Marín JA, Aguayo-Albasini JL. Chylous fistula following axillary lymphadenectomy: benefit of octreotide treatment. Case Rep Surg. Hindawi; 2016;2016.

33.

Fothiadaki IR, Stamatiou D, Askoxylakis J, Zoras O. Chylous leakage after breast-conserving surgery and axillary lymph node dissection for breast cancer. Hell Surg Oncol. 2014;5(3):134.

34.

Weiss JR, Moysich KB, Swede H. Epidemiology of male breast cancer. Cancer Epidemiol Prev Biomarkers. AACR; 2005;14(1):20–6.

35.

Key TJ, Verkasalo PK, Banks E. Epidemiology of breast cancer. Lancet Oncol. Elsevier; 2001;2(3):133–40.

36.

Smith ME, Riffat F, Jani P. The surgical anatomy and clinical relevance of the neglected right lymphatic duct. J Laryngol Otol. Cambridge University Press; 2013;127(2):128–33.

37.

Ilahi M, Ilahi TB. Anatomy, Thorax, Thoracic Duct. In: StatPearls [Internet]. StatPearls Publishing; 2018.

38.

Tam D, Scamp T, Tan J. Chyle leak after breast augmentation. Aesthetic Surg J. Blackwell Publishing Ltd Oxford, UK; 2009;29(2):113–5.

39.

McCray S, Parrish CR. When chyle leaks: nutrition management options. Pract Gastroenterol. SHUGAR PUBLISHING INC; 2004;60–77.

40.

Smoke A, DeLegge MH. Chyle leaks: consensus on management? Nutr Clin Pract. Wiley Online Library; 2008;23(5):529–32.

41.

Jain A, Singh SN, Singhal P, Sharma MP, Grover M. A prospective study on the role of octreotide in management of chyle fistula neck. Laryngoscope. Wiley Online Library; 2015;125(7):1624–7.

42.

Kapila S, Rozen WM, Huang T, Wu T, Fairbank S. Determining between chyle leak and anastomotic leak after esophageal reconstruction: The utility of methylene blue dye. Laryngoscope. Wiley Online Library; 2012;122(4):779–80.

43.

Kuroiwa M, Toki F, Suzuki M, Suzuki N. Successful laparoscopic ligation of the lymphatic trunk for refractory chylous ascites. J Pediatr Surg. Elsevier; 2007;42(5):e15–8.

Conflict of Interest Statement

We the authors have no affiliation with any organization with a direct or indirect financial interest in the subject matter discussed in the manuscript. This manuscript has not been submitted to, nor is under review at, another journal or other publishing venue. Kind Regards, Mr Nicholas Farkas Breast Surgical Registrar West Hertfordshire Hospitals NHS Trust