Study

First author, year

Appropriate and clearly focused question?¹

Yes/no/unclear

Comprehensive and systematic literature search?²

Yes/no/unclear

Description of included and excluded studies?³

Yes/no/unclear

Description of relevant characteristics of included studies?⁴

Yes/no/unclear

Appropriate adjustment for potential confounders in observational studies?⁵

Yes/no/unclear/notapplicable

Assessment of scientific quality of included studies?⁶

Yes/no/unclear

Enough similarities between studies to make combining them reasonable?⁷

Yes/no/unclear

Potential risk of publication bias taken into account?⁸

Yes/no/unclear

Potential conflicts of interest reported?⁹

Yes/no/unclear

Man et al., 2009

Yes

 Yes

No

Note: relevant studies excluded at final selection are not referenced with reasons

Yes

 Unclear

No

Yes

Yes

 Unclear

Note: not reported for each of included studies

Study reference

(first author, year of publication)

Bias due to a non-representative or ill-defined sample of patients?¹

(unlikely/likely/unclear)

Bias due to insufficiently long, or incomplete follow-up, or differences in follow-up between treatment groups?²

(unlikely/likely/unclear)

Bias due to ill-defined or inadequately measured outcome ?³

(unlikely/likely/unclear)

Bias due to inadequate adjustment for all important prognostic factors?⁴

(unlikely/likely/unclear)

 Vrije TRAM versus DIEP

Nelson et al., 2010

Likely

Note: choice of group was made based on anatomic intraoperative findings.

Likely

Note: differences in follow –up between groups and loss –to-follow up is not reported

Unclear

Note: outcome measures were not defined

Likely

Note: small groups in combination with lots of prognostic factors

Takeishi et al., 2008

Unclear

Note: selection criteria not reported

Unclear

Note: follow –up and loss-to-follow –up not reported

Unclear

Note: outcome measures were not defined

Likely

Note: failure to adequately adjust for overweight, other prognostic factors not reported

Gesteelde TRAM versus vrije TRAM

Gherardini et al.,

1994

Likely

Note: patients were assigned to the groups according to the condition of the donor site (scars), the recipient site (irradiation) and the size of the residual breast.

Likely

Note: follow-up varied between 1 and 57 month.

Unclear

Note: outcome measures were not defined

Likely

Note: small groups in combination with lots of prognostic factors (smoking, BMI, primary/secondary reconstruction, use of mesh, chemotherapy, ect.)

Baldewin et al.

1993

Likely

Note: choice of group was made by surgeon according to immediate reconstruction, smoking history, obesity and abdominal scarring

Unclear:

Note: loss-to-follow–up is not reported

Unlikely

Note: outcome measures were defined

Likely

Note: small groups in combination with lots of prognostic factors (timing of reconstruction, smoking, obesity, chemotherapy etc)

Moran et al.,

2000

Unclear

Note: patients were divided into BMI subgroups according to previously reports of pedicled TRAM reconstruction

Likely

Note: follow- up varied between 6 and 72 months

Unclear

Note: outcome measures were not defined

Unclear

Note: adjusted for age, BMI category and length of surgery

Gesteelde TRAM versus DIEP

Chun et al., 2010

Unclear

Note: selection criteria not reported

Likely

Note: loss-to-follow- up is not reported

Unlikely

Unclear

SIEA versus vrije TRAM

Selber et al., 2007

Unclear

Note: choice of group was made based on intraoperative findings. Performing TRAM from 1992 – 2003 and SIEA from 2003

Unclear:

Note: length of follow –up and lost-to-follow-up not reported

Unclear

Note: outcome measures were not defined

Likely

Note: failure to adequately adjust for prognostic factors (smokers, overweight and bilateral operation)

Chevray, 2003

Unclear

Note: in each case, reconstruction with an SIEA flap was attempted first. If the criteria for an SIEA flap were not met,

reconstruction with a DIEP flap was attempted. If the criteria for a DIEP flap were not met, reconstruction was performed with a muscle sparing free TRAM flap

Likely

Note: follow-up varied between 190 and 318 days.

Unclear

Note: outcome measures were not defined

Note: exclusion of study, see comment evidence table

Study reference

Study characteristics

Patient characteristics

Intervention (I)

Comparison / control (C)

Follow-up

Outcome measures and effect size

Comments

Man et al., 2009

[individual study characteristics deduced from Man et al., 2009]

SR and meta-analysis of comparative retrospective case series

Literature search up to April 2007

A: Bajaj et al., 2006

B: Blondeel et al., 1997

C: Bonde et al., 2006

D: Kroll et al., 2000

E: Nahabedian et al., 2005

F: Scheer et al., 2006

Setting and Country:

University

School of Medicine, USA

Source of funding:

Not reported

Inclusion criteria:

-studies comparing DIEP and TRAM flap surgery for breast reconstruction that report complication rates for at least one of the following complications: fat necrosis, partial flap loss, total flap loss, abdominal wall bulge, laxity or weakness, abdominal wall hernia

- clearly describe study design, number of patients and number of flaps

Exclusion criteria:

-abstract only, case reports, letters, comments, reviews

-case series with non-consecutive recruitment of patients

-languages other than English, novel or modifications of surgical techniques, animal or cadaveric studies

-physiologic or anatomical studies

6 studies included

Important patient characteristics at baseline:

No. of patients:

A: I: 124 (150 flaps) C: 35 (43 flaps)

B: I: 20, C: 18

C: I: 233, C: 44

D: total of 241 patients (I: 31 flaps, C: 279 flaps)

E: I: 89 (113 flaps), C: 88 (110 flaps)

F: I: 40 (46 flaps), C: 68 (84 flaps)

Groups comparable at baseline?

Difficult to assess, prognostic factors not always available.

Free TRAM

(in 3 studies a muscle –sparing free TRAM flap technique was used)

DIEP

Mean end-point of follow-up (months):

A: I: 10.2, C: 9.8

B: I: 32.1, C: 17.8

C: NR

D: NR

E: I: 23, C: 23

F: NR

For how many participants were no complete outcome data available?

(intervention/control)

Not reported

Any flap loss

Defined as number of total and partial flap losses (combined because flap loss was rare)

Effect measure: RR (95% CI):

A: 10.47 (2.17 – 19.04)

B: NR

C: 2.12 (0.42 – 10.57)

D: 6.43 (2.17 – 19.04)

E: 1.54 (0.26 – 9.04)

F: 0.66 (0.21 – 2.04)

Pooled effect:

RR 2.05 (95% CI 1.16 to 3.61) favoring free TRAM

P =0.04 for heterogeneity

Fat necrosis

Defined as number of flaps with fat necrosis

Effect measure: RR (95% CI):

A: 1.25 (0.48 – 3.26)

B: NR

C: 0.27 (0.02 – 4.62)

D: 2.25 (1.20 – 4.22)

E: 0.90 (0.34 – 2.39)

F: 4.93 (1.87 – 12.98)

Pooled effect:

RR 1.94 (95% CI 1.28 to 2.93) favoring free TRAM

P = 0.07 for heterogeneity

Abdominal bulge or hernia

Defined as patients with abdominal wall bulge or hernia (combined because not all studies distinguished between bulge and hernia)

Effect measure: RR (95% CI):

A: 0.94 (0.33 – 2.67)

B: 0.16 (0.01 – 2.86)

C: 0.41 (0.05 – 3.03)

D: NR

E: 0.25 (0.06 – 1.16)

F: 0.52 (0.22 – 1.25)

Pooled effect:

RR 0.49 (95% CI 0.28 to 0.86) favoring DIEP

P=0.57 for heterogeneity

Aesthetic result
Not reported

Facultative:

Conclusion authors: this analysis suggests that the DIEP flap reduces abdominal morbidity but increases flap-related complications compared with the free TRAM

flap in breast reconstruction.

Study reference

Study characteristics

Patient characteristics

Intervention (I)

Comparison / control (C)

Follow-up

Outcome measures and effect size

Comments

Vrije TRAM versus DIEP

Nelson et al., 2010

Type of study:

Retrospective cohort

Setting:

University hospital

Country:

USA

Source of funding:

In part by the Doris Duke Clinical Research Fellowship

Inclusion criteria:

-Patients with mastectomy receiving breast reconstruction between July 2006 and June 2008

-patients reconstructed with unilateral and bilateral free TRAM flaps (I) or DIEP flaps (C ) and patients undergoing bilateral reconstructions utilizing one of each flap (I +C)

Exclusion criteria:

Not reported

N total at baseline:

Intervention: 91 patients (123 breasts)

Control: 53 patients (71 breasts)

Bilateral I +C: 31 patients

Total flaps:

I: 154

C: 102

Important characteristics:

Previous smokers:

I: 34 (37%)

C: 18 (34%)

I+C: 10 (32%)

Current smokers:

I: 5 (5%)

C: 3 (6%)

I+C: 3 (10%)

Groups comparable at baseline?

Difficult to assess.

Muscle -sparing free TRAM

DIEP

Length of follow-up (mean ±SD?):

I: 369.3 (190) days

C: 366.8 (218) days

I+C: 424.5 (169) days

Loss-to-follow-up:

Not reported

Total flap loss

I: 0/154

C: 1/102

Partial flap loss

I: 0/154

C: 1/102

Fat necrosis

I: 7/154

C: 6/102

Abdominal bulge

I: 4/154

C: 1/102

Hernia

I: 4/154

C: 0/102

Aesthetic result

Not reported

Retrospective;

No study protocol (follow-up etc.); imprecision

Takeishi et al., 2008

Type of study:

Retrospective cohort

Setting:

Jikei University School of Medicine

Country:

Japan

Source of funding:

Not reported

Inclusion criteria:

Patients with breast cancer, giant mammary fibroadenoma or silicone implant receiving breast reconstruction between October 1999 and April 2006

Exclusion criteria:

Not reported

N total at baseline:

Intervention: 79 (82 breasts)

Control: 30

Important characteristics:

Obese (BMI ≥25 kg/m²):

I: 14/79

C: 4/30

Smokers:

Not reported

Groups comparable at baseline?

Difficult to assess.

Muscle -sparing free TRAM

(58 primary; 20 secondary)

DIEP

Length of follow-up:

Not reported

Loss-to-follow-up:

Not reported

Total flap loss

I: 0/82

C: 1/30

Partial flap loss

Not reported

Fat necrosis

I: 8/82

C: 4/30

Abdominal bulge

I: 5/82

C: 2/30

Hernia

Not reported

Aesthetic result

Not reported

Retrospective;

No study protocol (follow-up etc.); imprecision

Gesteelde TRAM versus vrije TRAM

Gherardini et al.,

1994

Type of study:

Retrospective cohort

Setting:

Karolinska Hospital, Stockholm

Country:

Sweden

Source of funding:

Supported by the National Council of Researches of Italy, Comett Forum, Stiftelsen Blanceflor Boncompagni Foedd Bildt Foundation, Wenner-Gren Foundation and the Rotary Foundation.

Inclusion criteria:

Patients with modified radical or partial mastectomy receiving breast reconstruction

Exclusion criteria:

Not reported

N total at baseline:

Intervention: 27 (28 breasts)

Control: 11

Important characteristics:

Heavy smokers (>30 cigarettes /day):

I: 4/27

C: 4/11

Groups comparable at baseline?

Not comparable for heavy smokers; unclear for other prognostic variables.

Pedicled TRAM

(3 primary; 24 secondary)

Free TRAM

(1 primary; 10 secondary)

Length of follow-up:

Median: nine month (range 1-57)

Loss-to-follow-up:

Not reported

Total flap loss

I: 1/28

C: 0/11

Partial flap loss

I: 9/28

C: 1/11

Fat necrosis

Not reported

Abdominal bulge

Not reported

Hernia

Not reported

Aesthetic result

Not reported

Retrospective;

No study protocol (follow-up etc.);

Imprecision;

1994: indirectness?;

Baldwin et al.,

1993

Type of study:

Retrospective cohort

Setting:

M. D. Anderson Cancer Center

Country:

Texas, USA

Source of funding:

Not reported

Inclusion criteria:

Patients with bilateral mastectomies (including elective contralateral mastectomy) receiving bilateral breast reconstruction

Exclusion criteria:

Not reported

N total at baseline:

Intervention: 28 (56 breasts)

Control: 18 (36 breasts)

Important characteristics:

Smokers:

I: 13%

C: 33%

Groups comparable at baseline?

Not comparable for smokers; unclear for other prognostic variables

Pedicled TRAM

( 21 immediate, 5 delayed at second mastectomy and 2 after both mastectomies)

Free TRAM

(17 immediate, 1 delayed at second reconstruction)

Length of follow-up:

Minimum of 1 year

Loss-to-follow-up:

Not reported

Total flap loss

I: 0/56

C: 3/36

P= not significant

Partial flap loss

Defined as loss of part of the skin paddle and/or adjactent subcutaneous tissue within the first 30 days

I: 7/56

C: 0/36

P=0.03

Fat necrosis

Defined as nonvascularity of subcutaneous flap tissue presenting a firm mass after 30 days

I: 4/56

C: 0/36

P = not significant

Abdominal bulge

Not reported

Hernia

Not reported

Aesthetic result

Not reported

Authors: free TRAM-procedure has a steep learning curve.

Moran et al.,

2000

Type of study:

Retrospective cohort

Setting:

University hospital

Country:

USA

Source of funding:

Not reported

Inclusion criteria:

Obese patients (BMI >25.8 kg/m²) receiving breast reconstruction

Exclusion criteria:

Not reported

N total at baseline:

Intervention: 36 (8% bilateral)

Control: 78 (4% bilateral)

Important characteristics:

Smokers:

I: 17%

C: 26%

Prior radiation treatment:

I: 11%

C: 9%

Groups comparable at baseline?

Difficult to assess.

Pedicled TRAM

Free TRAM

Length of follow-up:

Average: 24 months (range 6 to 72 months)

Loss-to-follow-up:

Not reported

Total flap loss

I: 1/36

C: 0/78

Partial flap loss

I: 4/36

C: 0/78

Fat necrosis

I: 3/36

C: 8/78

Abdominal bulge or hernia

I: 1/36

C: 2/78

Aesthetic result

Not reported

An increasing BMI was found to have a significant effect on free TRAM flap complications.

Gesteelde TRAM versus DIEP

Chun et al., 2010

Type of study:

Retrospective cohort

Setting:

Brigham and Women’s Hospital (TRAM) and Beth Israel Deaconess Medical Center (DIEP)

Country:

USA

Source of funding:

Funded by Brigham and Women’s Hospital Department of Surgery Junior Fellowship in honor of Robert T. Osteen and Doris Duke Charitable Foundation

Inclusion criteria:

Patients with mastectomies receiving bilateral breast reconstruction

Exclusion criteria:

Not reported

N total at baseline:

Intervention: 105 (210 breasts)

Control: 58 (116 breasts)

Important characteristics:

Smokers:

I: 11.4%

C: 6.9%

Prior radiation treatment:

I: 38.5%

C: 39.6%

BMI >30:

I: 20%

C: 14.5%

Groups comparable at baseline?

Difficult to assess

Pedicled TRAM

(reconstruction: 86 immediate, 19 delayed)

DIEP

(reconstruction: 41 immediate, 13 delayed and 4 immediate of one breast and delayed of contralateral breast)

Length of follow-up (mean):

I: 6.2 years ( 11months -15 years);

C: 2.3 years (11 months – 4 years)

Loss-to-follow-up:

Not reported

Total flap loss

I: 0/105

C: 1/58

Partial flap loss

Not reported

Fat necrosis (per flap)

Defined as a palpable discrete firmness greater than 2 cm identified by physical examination during postoperative evaluation or by intraoperative finding for any

subsequent procedure.

I: 24/210

C: 25/116

Abdominal bulge

Defined as abdominal wall bulging without a fascial defect as assessed by physical examination and/or radiographic study

I: 3/105

C: 4/58

Hernia

Defined as a fascial

defect identified by physical examination and radiographic

study and/or confirmed at the time of operative repair.

I: 3/105

C: 0/58

Aesthetic result

Not reported

Satisfaction: there were no significant differences between groups

SIEA versus vrije TRAM

Selber et al., 2007

Type of study:

Retrospective cohort (performing TAMP from 1992 – 2003 and SIEA from 2003-2007)

Setting:

Teaching hospital of University of Rochester (TRAM and SIEA) and University of Pennsylvania (only SIEA)

Country:

USA

Source of funding:

No

Inclusion criteria:

Patients with mastectomies receiving breast reconstruction

Exclusion criteria:

Not reported

N total at baseline:

Intervention: 63 (69 flaps)

Control: 500 (569 flaps)

Important characteristics:

Smokers:

I: 24.64%

C: 9.8%

Overweight (BMI ≥ 26):

I: 39.13%

C: 14%

Groups comparable at baseline?

Not comparable for smokers, overweight and bilateral operation

SIEA

(53 immediate)

Muscle-sparing free TRAM

(390 immediate)

Length of follow-up:

Not reported

Loss-to-follow-up:

Not reported

Total flap loss

I: 2/69

C: 1/569

P=0.032

Partial flap loss

I: 1/69

C: 9/569

P=0.629

Fat necrosis

I: 5/69

C: 19/569

P=0.167

Abdominal bulge or hernia

I: 0/69

C: 11/569

P=0.338

Aesthetic result

Not reported

The SIEA-flap had a higher rate of thrombotic complications.

Chevray, 2003

Type of study:

Prospective cohort

Setting:

University

Country:

Texas

Source of funding:

Inclusion criteria:

Breast reconstruction

using autologous tissue from the

lower abdomen during the 17-month period

from August of 2001 through December of

2002 were included.

Exclusion criteria:

Not reported

N total at baseline:

SIEA: 12 (14 flaps)

DIEP: 7 (8 flaps)

TRAM: 21 (25 flaps)

Important characteristics:

Groups comparable at baseline:

SIEA

(8 immediate; 6 delayed)

DIEP
(6 immediate; 2 delayed)

free TRAM

(15 immediate; 10 delayed)

Length of follow-up (days):

SIEA: 271

DIEP: 318

TRAM: 190

Loss-to-follow-up:

None (however, not explicitely stated)

Total flap loss

SIEA: 1/14

DIEP: 0/8

TRAM: 0/25

Partial flap loss

SIEA: 0/14

DIEP: 1/8

TRAM: 0/25

Fat necrosis

SIEA: 2/14

DIEP: 1/8

TRAM: 3/25

Abdominal bulge or hernia

SIEA: 0/14

DIEP: 1/8

TRAM: 1/25

Aesthetic result

Results SIEA were similar to those obtained by TRAM or DIEP

Statistical comparisons of patient characteristics

and incidence of complications were not valid for the entire group because many patients

underwent bilateral breast reconstruction with various combinations of the three

flap types à exclusion