Abstract

OBJECTIVES

The pedicled intercostal muscle flap (IMF) is a high quality vascularized tissue commonly used to buttress the bronchial stump after pneumonectomy or bronchial anastomosis after sleeve lobectomy in order to prevent bronchopleural fistula formation. The evaluation of the viability of the muscle flap is difficult. The aim of this study was the assessment of the application of indicyanine green fluorescence for the evaluation of IMF perfusion.

METHODS

The study included 27 patients (10 males and 17 females), mean age 62.6 years (47–77 years). Indocyanine green fluorescence (ICG) was used for objective assessment of the IMF quality by a near-infrared camera system (Photodynamic Eye®, Hamamatsu Photonics, Japan). The following factors that may have an impact on the quality of the IMF were assessed: age, gender, body mass index, comorbidities, IMF length and thickness and timing of the harvesting during the procedure.

RESULTS

The following surgical pulmonary resections with IMF harvesting were performed: 12 pneumonectomies, 2 sleeve lobectomies and 13 lobectomies. Intercostal muscle flap (IMF) was harvested before rib spreader insertion in 23 patients (85%) and at the end of the surgery in 4 patients (15%). The mean length and thickness of the harvested intercostal muscle were 19.9 ± 2.9 cm (range 13–24 cm) and 2.4 cm ± 0.7 cm (range 1.0–3.5 cm), respectively. Indocyanine green angiography showed ischaemia in the distal part of the muscle in all cases, despite the lack of obvious macroscopic signs. Median length of the ischaemic part was 4 cm (range 0.5–20 cm). The IMF length and thickness had a significant impact on the length of the ischaemic segment. In 24 patients, the ischaemic part of the muscle flap was severed. In 3 patients with the longest ischaemic segment (11, 13 and 20 cm), an alternative tissue was used to cover the bronchial stump. No major complications occurred.

CONCLUSIONS

Our preliminary results confirmed the simplicity and high efficacy of ICG in the assessment of intercostal muscle blood perfusion. ICG was superior to macroscopic evaluation and influenced surgical proceeding.

INTRODUCTION

Bronchopleural fistula (BPF) is one of the most morbid complications of lung resections. Its occurrence substantially diminishes the quality of life after surgery, exposes the patient to additional surgical interventions and greatly increases the risk of death particularly after right pneumonectomy [1, 2]. One of the methods used for BPF prevention is the improvement of the bronchial stump healing by covering the suture line with various well-perfused tissues. The most frequently used tissues are: intercostal muscle flap (IMF), thymic or mediastinal fat pad, pericardium and parietal pleura. There are no clear data in the literature as to which tissue type is the most efficient in preventing bronchial fistula. Although in our institutional experience, pedicled IMF has been used for this purpose, the rate of BPF remained high and was 7%. IMF quality has been traditionally assessed only by a subjective visual examination due to the lack of available objective methods.

Indocyanine green fluorescence (ICG) has been used for the evaluation of tissue perfusion in many surgical procedures [3]. The aim of this study was the assessment of application of indicyanine green fluorescence for evaluation of IMF perfusion.

MATERIALS AND METHODS

The study was conducted at the Department of Thoracic Surgery of Poznań University of Medical Sciences from September 2011 to May 2012 after approval by the local Internal Review Board. The study included 27 patients (10 males and 17 females), with a mean age of 62.6 years (range 47–77 years), diagnosed with non-small-cell lung cancer (26 patients) and tuberculosis (1 patient) (Table 1). As a part of the preoperative workup the entire patient population underwent routine blood tests, CT scan of the chest, abdominal ultrasound, fiberoptic bronchoscopy and pulmonary function tests. All the surgical procedures were performed by three attending surgeons. The standard muscle-sparing antero-lateral thoracotomy approach was used. In the majority of patients (23 patients, 85%) scheduled for sleeve lobectomy or pneumonectomy, the IMF was harvested at the beginning of the procedure before the rib spreader was applied. In the remaining 4 patients (15%), the IMF was obtained later in the course of the procedure. The technique of IMF harvesting was as follows. The skin incision, subcutaneous tissue and serratus anterior muscle division were performed in the usual manner for antero-lateral muscle-sparing thoracotomy. The intercostal space was opened at the upper border of the sixth rib with cautery. The IMF harvesting was initiated at the level of anterior axillary line. The periosteum of the fifth rib was cut longitudinally and dissected from the rib down to its lower border with a periosteal elevator. The intercostal muscle with adjacent parietal pleura was meticulously detached from the fifth rib using cautery up to the point located ∼2–3 cm from the dorsal end of the thoracotomy incision. Particular attention was paid to the protection of the intercostal vascular bundle and any unnecessary cauterization of minor bleeding sites was avoided if possible. After harvesting, the length and thickness of the IMF were measured, and signs of muscle ischaemia were macroscopically assessed (colour, temperature, bleeding from the distal end of IMF).

Table 1:

Patient characteristics

Patients (n27 
Gender 
 Male (n, %) 17 (63%) 
 Female (n, %) 10 (37%) 
Age (range, mean) 47–77; 62.6 years 
Body mass index 24.7 (±4.2) 
Smoking history 26 (96%) 
Pack-years 50 (range 0–90) 
Induction chemotherapy 
Comorbidities 
 Chronic obstructive pulmonary disease 
 Coronary heart disease 10 
 Diabetes mellitus 
Patients (n27 
Gender 
 Male (n, %) 17 (63%) 
 Female (n, %) 10 (37%) 
Age (range, mean) 47–77; 62.6 years 
Body mass index 24.7 (±4.2) 
Smoking history 26 (96%) 
Pack-years 50 (range 0–90) 
Induction chemotherapy 
Comorbidities 
 Chronic obstructive pulmonary disease 
 Coronary heart disease 10 
 Diabetes mellitus 

Subsequently, IMF quality assessment by ICG fluorescence was performed. The IMF was surrounded with clean surgical towels to avoid any background signal noise from other vascular tissues. 10 ml of aqueous solution of 2.5 mg of ICG dye was injected intravenously. One minute after the injection, operating-room lights were dimmed for better visualization. ICG fluorescence imaging was assessed with a near-infrared camera system (Photodynamic Eye®, Hamamatsu Photonics, Japan). Well-fluorescing tissues were regarded as properly perfused, while poorly fluorescing as inadequately perfused (Fig. 1). The ischaemic part was measured and severed ∼30–40 min after harvesting. IMF was laid down freely in the pleural cavity and an appropriate anatomical resection was completed followed by a mediastinal lymphadenectomy. After the resection, the pedunculated muscle flap was attached to the stump with interrupted absorbable sutures. ICG fluorescence of the IMF sutured to the bronchus was performed again at the end of the operation to look for the signs of unexpected ischaemia (Fig. 2). The following factors were analysed in terms of the influence on the length of the IMF ischaemic part: gender, age and BMI and underlying conditions (COPD, DM and cardio-vascular disease). Additionally, muscle features such as: flap length, thickness and timing of harvesting were included in the analysis.

Figure 1:

ICG showing well-perfused proximal part and ischaemic distal part of harvested intercostal muscle. IMFperf: intercostal muscle flap with good perfusion; IMFisch: ischaemic IMF.

Figure 1:

ICG showing well-perfused proximal part and ischaemic distal part of harvested intercostal muscle. IMFperf: intercostal muscle flap with good perfusion; IMFisch: ischaemic IMF.

Figure 2:

ICG after bronchial stump buttressing showing good perfusion of IMF. Asc: ascending aorta; Arch: aortic arch; Desc: descending aorta; AP: aortopulmonary window after lymphadenectomy; IMF: intercostal muscle flap.

Figure 2:

ICG after bronchial stump buttressing showing good perfusion of IMF. Asc: ascending aorta; Arch: aortic arch; Desc: descending aorta; AP: aortopulmonary window after lymphadenectomy; IMF: intercostal muscle flap.

Data on interval measurement scale were presented as means and standard deviations or medians and interquartile ranges in case the data did not follow a normal distribution. Categorical data were presented as percentages. The assumption that data follow a normal distribution was checked by Shapiro–Wilk's test. For comparison of three unpaired groups, one-way analysis of variance (ANOVA) for the data that followed normal distribution and homogeneity of variances was performed. Homogeneity of variances was tested with Levene's test. In case data were not normally distributed nonparametric tests were used. Mann–Whitney U-test for comparing, two groups and Kruskal–Wallis for the comparison of more than two groups simultaneously were used. If statistically significant difference existed the Dunn's multiple comparisons post hoc test was applied. All results were considered significant at P < 0.05. Statistical analysis was performed using statistical packages: STATISTICA 10.0 PL (StatSoft., Inc.) or StatXact 8.0 (CytelStudio).

RESULTS

The following surgical pulmonary resections with IMF harvesting were performed: 12 pneumonectomies (4 right and 8 left), 2 sleeve lobectomies and 13 lobectomies. Mean length and thickness of the harvested intercostal muscle were 19.9 ± 2.9 cm (range 13–24 cm) and 2.4 ± 0.7 cm (range 1.0–3.5 cm), respectively (Table 2). The macroscopic symptoms of the muscle ischaemia were never observed in the harvested IMFs. The ICG fluorescence was easily detected by the camera at 1 min after injection and in all cases at least a small portion of the distal part of the muscle showed a complete ischaemia. There was a significant variability of the length of the ischaemic part ranging from 0.5 up to 20 cm (median 4 cm). Before covering of the bronchial stump the ischaemic part of the muscle flap was severed. In 3 patients with the longest ischaemic segments (11, 13 and 20 cm) an alternative tissue had to be used to cover the bronchial stump.

Table 2:

IMF characteristics

Patient number IMF—length Ischaemic part—length IMF thickness 
19.5 1.0 3.5 
24.0 4.0 3.0 
20.0 8.0 1.5 
23.0 20.0 3.0 
19.0 4.0 1.5 
23.0 3.0 3.0 
22.0 5.0 2.0 
21.5 11.0 1.0 
18.0 4.0 3.0 
10 22.0 13.0 2.5 
11 17.0 2.0 2.5 
12 21.0 5.0 3.0 
13 23.0 6.0 3.0 
14 24.0 8.0 1.5 
15 19.0 7.0 1.5 
16 20.0 1.0 2.0 
17 20.0 1.0 2.0 
18 18.0 6.0 2.0 
19 18.0 7.0 2.5 
20 18.5 8.5 1.0 
21 19.0 1.5 2.5 
22 21.0 2.0 3.0 
23 13.0 0.5 3.0 
24 20.0 4.0 2.5 
25 15.0 0.0 3.0 
26 14.0 1.0 2.5 
27 24.0 9.0 3.0 
Patient number IMF—length Ischaemic part—length IMF thickness 
19.5 1.0 3.5 
24.0 4.0 3.0 
20.0 8.0 1.5 
23.0 20.0 3.0 
19.0 4.0 1.5 
23.0 3.0 3.0 
22.0 5.0 2.0 
21.5 11.0 1.0 
18.0 4.0 3.0 
10 22.0 13.0 2.5 
11 17.0 2.0 2.5 
12 21.0 5.0 3.0 
13 23.0 6.0 3.0 
14 24.0 8.0 1.5 
15 19.0 7.0 1.5 
16 20.0 1.0 2.0 
17 20.0 1.0 2.0 
18 18.0 6.0 2.0 
19 18.0 7.0 2.5 
20 18.5 8.5 1.0 
21 19.0 1.5 2.5 
22 21.0 2.0 3.0 
23 13.0 0.5 3.0 
24 20.0 4.0 2.5 
25 15.0 0.0 3.0 
26 14.0 1.0 2.5 
27 24.0 9.0 3.0 

IMF: intercostal muscle flap.

The statistical analysis showed that length and thickness of the harvested muscle flap had a significant influence on the extent of the ischaemia (Table 3). There was no correlation between the timing of the harvesting (before or after rib-spreader insertion) and length of the ischaemic part of the muscle, even though it was longer when harvested later in the course of the procedure. The patient factors including comorbidities, BMI, age and gender had no significant impact on the length of the ischaemic muscle.

Table 3:

Analysis of IMF ischaemia risk factors

Variables n Spearman's rank correlation coefficient r P-value 
Length of ischaemic part of IMF and length of IMF 27 0.51 0.006 
Length of ischaemic part of IMF and IMF thickness 27 −0.4 0.04 
Length of ischaemic part of IMF and BMI 27 ns 0.31 
Length of ischaemic part of IMF and age 27 ns 0.38 
Variables n Spearman's rank correlation coefficient r P-value 
Length of ischaemic part of IMF and length of IMF 27 0.51 0.006 
Length of ischaemic part of IMF and IMF thickness 27 −0.4 0.04 
Length of ischaemic part of IMF and BMI 27 ns 0.31 
Length of ischaemic part of IMF and age 27 ns 0.38 

IMF: intercostal muscle flap; BMI: body mass index.

The IMF perfusion rechecked after bronchial stump buttressing was optimal in all cases. Mean operation time and blood loss were 153 ± 41 min (range 90–260 min) and 248 ± 93 ml (50–400 ml), respectively. Chest drainage median time was 3 days (range 3–12 days) and postoperative hospital stay was 9 days (range 6–35 days). No major postoperative complications occurred and no other minor morbidities that could be attributed to intercostal muscle harvesting were observed. No blood transfusion was needed.

In 1 patient, after right pneumonectomy a bronchial stump dehiscence without bronchopleural fistula was observed on postoperative day 13. During bronchoscopy, a viable IMF was visualized completely covering the stump and separating it from the post-pneumonectomy space. A successful healing of the stump by secondary intention within the IMF coverage was achieved 3 months after surgery, and it could clearly be attributed to the use of the fully viable IMF.

DISCUSSION

BPF, despite refinements in surgical technique, still remains one of the most feared complications of lung resections. Its occurrence after pneumonectomy varies substantially between patient series, but is generally reported to be between 0 and 12% [4]. One of the methods to decrease the risk of BPF is improving the blood supply of the bronchial stump area by transposing well-vascularized tissues [5, 6]. In our experience, the use of IMF did not significantly decrease the rate of BPF after pneumonectomy. Therefore, we concluded that the suboptimal quality of the muscle flap could be one of the reasons.

In general, the essential desirable properties of transposed flaps are good perfusion and viability. There are many tests that have been developed for monitoring these features of tissue transfers: clinical tests, chemical techniques (fluorescein and indocyanine), radioactive isotopes (technetium-99 m, xenon and sodium), instrumental methods (Doppler flowmetry, photoplethysmography, microdialysis and MRI). Unfortunately, most of these methods are not suitable for IMF appraisal due to their complexity, cost, potential side-effects, need of waste utilization or patient irradiation [7].

The ideal method of evaluation of tissue perfusion and viability should be inexpensive, simple to use, rapid, reliable, repeatable, recordable, accurate, objective, with a short learning-time and no side-effects [8]. From this point of view, the properties of indocyanine green make it a promising option for assessing the blood flow in the IMF.

First, ICG has the ability to fluorescence. This is defined as, after absorbing the light of particular wavelength (maximum absorption occurs at 805 nm), ICG emits the light of a different wavelength (835 nm) that can be recorded by infrared camera. ICG is the only dye that has been approved by the US Food and Drug Administration for use in humans [9]. ICG is water-soluble, which allows its easy intravenous administration [10].

Secondly, ICG has been used for more than half a century and has proved to have a very high safety profile [11]. In this study, the total dose was 5 mg per patient, which equalled ∼0.05–0.1 mg/kg.

Due to its physicochemical properties and safety profile, ICG has been implemented in many fields of medicine. Primarily used for ophthalmic angiography [12] and liver function assessment [13], ICG proved to be useful for plasma volume measurement [14] and cardiac output evaluation [15]. It was also implemented in sentinel lymph node navigation in breast [16], gastrointestinal [17], skin [18] and lung cancer [19] as a much safer and easier agent to use than radionuclides. Intraoperative angiography based on ICG fluorescence has been applied with success for the immediate assessment of the patency of vascular anastomoses in the surgery of congenital cardio-vascular anomalies [20], coronary artery bypass grafting [21], microsurgical free flap transfers [22] and for gastric conduit perfusion during esophagectomy [23]. ICG has been recently used in thoracic surgery for the assessment of intersegmental planes [24].

We hypothesized that ICG fluorescence could be an objective method for intercostal muscle blood flow assessment. Our results proved that clinical features were not sufficient for the reliable evaluation of IMF perfusion and might be inaccurate in determining the extent of muscle ischaemia. Indocyanine green angiography clearly showed that distal parts of all of the harvested muscles were ischaemic. The length of the ischaemic part was very variable and it was >3 cm in 70% of patients. Potentially, without the objective assessment, the whole bronchial suture line would be buttressed with a blood-supply deficient tissue that would certainly not fulfill the task of improving bronchial stump viability. Moreover, in case of ischaemic IMF necrosis, a chronic inflammatory state might develop in the bronchial stump. This could contribute to increased susceptibility to infection and BPF formation. This hypothesis could explain the higher percentage of BPF after bronchial stump reinforcement observed by some authors and discrepancies in the rates of BPF occurrence between different publications [4].

The first factor related to the IMF quality that significantly influenced the length of the ischaemic part of the IMF was the thickness of the harvested muscle flap. It seems to be particularly important in thin patients with poorly developed intercostal muscles. In these patients, we should expect longer ischaemia and pay particular attention to the technique of harvesting.

Secondly, our results showed that the length of the ischaemic part was directly correlated to the entire length of the harvested IMF. It may indicate that perfusion of the IMF gradually diminishes with the increasing distance from the origin of the intercostal artery from the aorta. Based on these results we suggest avoiding harvesting of exceedingly long intercostal muscles if possible.

One of the most important issues is the timing of the IMF harvesting [25]. We strongly believe that any additional injury of the intercostal bundle by compression related to the use of the rib spreader may result in the local damage of the vascularization of the muscle. This factor was reflected in our results in the longer distal ischaemic part of the IMF (median 7.0 vs 4.0 cm) when the harvesting was performed at the end of surgery after significant period of compression by the rib spreader. It seems to be crucial that when anticipating IMF use, it should always be harvested at the initial stage of thoracotomy. Although this difference was not statistically significant (P < 0.3715), it is likely that with bigger patient groups the difference would reach statistical significance. In each case of unanticipated need of IMF use, after lung resection, one has to be very cautious and expect potentially longer ‘local ischaemia’.

The quality of the IMF could be impaired by blood flow disturbances resulting from arteriosclerosis. Potential arteriosclerosis risk factors that often occur in lung cancer patients, i.e. age, gender and BMI, had no significant impact on the length of the ischaemic part of IMF. Similarly, commorbidities, i.e. COPD, cardio-vascular disease did not influence the quality of the IMF.

This method could be also used as a tool for evaluating the process of teaching as well as supervising of harvesting of IMFs and other pedicled flaps used in surgery. Although we have been using IMFs for many years and felt that our technique of muscle harvesting was adequate, the novel ICG use made us realize how fragile and prone to injury intercostal vessel are. This has influenced us to put even more effort into improving our technique of muscle harvesting.

By using ICG and defining the range of IMF ischaemia, we were able to avoid improper bronchial stump buttressing by either cutting the ischaemic part (in 24 of 27 patients, 79%) or abandoning its use in favour of the other tissues (in 3 of 27 patients, 11%). ICG fluorescence substantially influenced our surgical procedure.

As a pilot study, this obviously has some limitations: limited number of patients, varied extent of pulmonary resections, lack of randomization and short time of the follow-up. The ICG fluorescence assessment of blood supply with the currently available infrared camera remains subjective, as there are still no quantative measures of IMF perfusion except tissue brightness level.

Our results indicate that ICG can become a valuable tool for the assessment of IMF viability but further evaluation of the method is needed.

Conflict of interest: none declared.

REFERENCES

1
Algar
FJ
Alvarez
A
Aranda
JL
Salvatierra
A
Baamonde
C
López-Pujol
FJ
Prediction of early bronchopleural fistula after pneumonectomy: a multivariate analysis
Ann Thorac Surg
 , 
2001
, vol. 
72
 (pg. 
1662
-
7
)
2
Darling
GE
Abdurahman
A
Yi
QL
Johnston
M
Waddel
TK
Pierre
A
, et al.  . 
Risk of right pneumonectomy: role of bronchopleural fistula
Ann Thorac Surg
 , 
2005
, vol. 
79
 (pg. 
433
-
7
)
3
Połom
K
Murawa
D
Rho
Y
Nowaczyk
P
Hünerbein
M
Murawa
P
Current trends and emerging future of indocyanine green usage in surgery and oncology: a literature review
Cancer
 , 
2011
, vol. 
117
 (pg. 
4812
-
22
)
4
Deschamps
C
Bernard
A
Nichols
FC
Allen
MS
Miller
DL
Trastek
VF
, et al.  . 
Empyema and bronchoplerural fistula after pneumonectomy
Ann Thorac Surg
 , 
2001
, vol. 
72
 (pg. 
243
-
8
)
5
Anderson
TM
Miller
JI
Use of pleura, azygos vein, pericardium and muscle flaps in tracheobronchial surgery
Ann Thorac Surg
 , 
1995
, vol. 
60
 (pg. 
729
-
33
)
6
Rendina
EA
Venuta
F
Ricci
P
Fadda
GF
Bognolo
DA
Ricci
C
, et al.  . 
Protection and revascularization of bronchial anastomoses by the intercostal pedicle flap
J Thorac Cardiovasc Surg
 , 
1994
, vol. 
107
 (pg. 
1251
-
4
)
7
Whitaker
IS
Karoo
ROS
Oliver
DW
Ganchi
PA
Gulati
V
Malata
CM
Current techniques in the post-operative monitoring of microvascular free-tissue transfers
Eur J Plast Surg
 , 
2005
, vol. 
27
 (pg. 
315
-
21
)
8
Creech
B
Miller
S
Myers
Grabb WC
Evaluation of circulation in skin flaps
Skin flaps
 , 
1975
Boston
Little, Brown
9
te Velde
EA
Veerman
T
Subramaniam
V
Ruers
T
The use of fluorescent dyes and probes in surgical oncology
Eur J Surg Oncol
 , 
2009
, vol. 
36
 (pg. 
6
-
15
)
10
Alander
JT
Kaartinen
I
Laakso
A
Pätilä
T
Spillmann
T
Tuchin
VV
, et al.  . 
A review of indocyanine green fluorescent imaging in surgery
Int J Biomed Imaging
 , 
2012
, vol. 
2012
 pg. 
940585
 
11
Benya
R
Quantana
J
Brundage
B
Adverse reactions to indocyanine green: a case report and a review of the literature
Cathet Cardiovasc Diagn
 , 
1989
, vol. 
17
 (pg. 
231
-
3
)
12
Flower
RW
Evolution of indocyanine green dye choroidal angiography
Optical Engineering
 , 
1995
, vol. 
34
 (pg. 
727
-
36
)
13
Paumgartner
G
The handling of indocyanine green by the liver
Schweiz Med Wochenschr
 , 
1975
, vol. 
105
 (pg. 
1
-
30
)
14
Haller
M
Akbulut
C
Brechtelsbauer
H
Fett
W
Briegel
J
Finsterer
U
, et al.  . 
Determination of plasma volume with indocyanine green in man
Life Sci
 , 
1993
, vol. 
53
 (pg. 
1597
-
604
)
15
Lund-Johansen
P
The dye dilution method for measurement of cardiac output
Eur Heart J
 , 
1990
, vol. 
11
 (pg. 
6
-
12
)
16
Murawa
D
Hirche
C
Dresel
S
Hunerbein
M
Sentinel lymph node biopsy in breast cancer guided by indocyanine green fluorescence
Br J Surg
 , 
2009
, vol. 
96
 (pg. 
1289
-
94
)
17
Hirche
C
Dresel
S
Krempien
R
Hunerbein
M
Sentinel lymph node biopsy by indocyanine green retention fluorescence detection for inguinal lymph node staging of anal cancer: preliminary experience
Ann Surg Oncol
 , 
2010
, vol. 
17
 (pg. 
2357
-
62
)
18
Tanaka
R
Nakashima
K
Fujimoto
W
Sentinel lymph node detection in skin cancer using fluorescence navigation with indocyanine green
J Dermatol
 , 
2009
, vol. 
36
 (pg. 
478
-
70
)
19
Yamashita
SI
Tokuishi
K
Anami
K
Miyawaki
M
Moroga
T
Kamei
M
, et al.  . 
Video-assisted thoracoscopic indocyanine green fluorescence imaging system shows sentinel lymph nodes in non-small-cell lung cancer.
J Thorac Cardiovasc Surg
 , 
2011
, vol. 
141
 (pg. 
141
-
4
)
20
Kogon
B
Fernandez
J
Kanter
K
Kirshbom
P
Vincent
B
Maher
K
, et al.  . 
The role of intraoperative indocyanine green fluorescence angiography in pediatric cardiac surgery
Ann Thorac Surg
 , 
2009
, vol. 
88
 (pg. 
632
-
6
)
21
Desai
ND
Miwa
S
Kodama
D
Koyama
T
Cohen
G
Pelletier
MP
, et al.  . 
A randomized comparison of intraoperative indocyanine green angiography and transient-time flow measurement to detect technical errors in coronary artery bypass grafts
J Thorac Cardiovasc Surg
 , 
2006
, vol. 
132
 (pg. 
585
-
94
)
22
Giunta
RE
Holzbach
T
Taskov
C
Holm
PS
Brill
T
Busch
R
, et al.  . 
Prediction of flap necrosis with laser induced indocyanine green fluorescence in a rat model
Br J Plast Surg
 , 
2005
, vol. 
58
 (pg. 
695
-
701
)
23
Murawa
D
Hünerbein
M
Spychała
A
Nowaczyk
P
Połom
K
Murawa
P
Indocyanine green angiography for evaluation of gastric conduit perfusion during esophagectomy—first experience
Acta Chir Belg
 , 
2012
, vol. 
112
 (pg. 
275
-
80
)
24
Misaki
N
Chang
S
Igai
H
Tarumi
S
Gotoh
M
Yokomise
H
New clinically applicable method for visualizing adjacent lung segments using an infrared thoracoscopy system
J Thorac Cardiovasc Surg
 , 
2010
, vol. 
149
 (pg. 
752
-
6
)
25
Venuta
F
Anile
M
Rendina
EA
Advantages of the segmental nondivided intercostal muscle flap
J Thorac Cardiovasc Surg
 , 
2010
, vol. 
140
 pg. 
485
 

APPENDIX. CONFERENCE DISCUSSION

Dr M. Zielinski(Zakopane, Poland): This was really something new in thoracic surgery. I think your study might have great value for all of us in the future. I have two questions. Probably you removed the rib during preparation of the flap: was it one or two or, indeed, any? That is one question. And the other is, you said that you started your procedure from preparation of the flap, but were there any cases in which you did the dissection afterwards? Have you any experience of what the flap looks like if we start from thoracotomy and then we get the idea to use the flap?

Dr Piwkowski: First we do not resect the ribs. We just, as you could see from the picture, dissect the muscle flap posteriorly, the length of the thoracotomy, in fact, and just go down. So we don't remove the rib. Regarding the second question, which I think is very important, that is why I asked in the beginning when do you harvest this muscle? And just to try to avoid any kind of risk of additional injury to the intercostal vessels, we designed the study so that in every case we do, if we suspect that a pneumonectomy may have to be done, we harvest the muscle in the beginning, and that is why so many lobectomies are there.

But when you look at one of these patients with a 13 cm ischaemic portion, the muscle was harvested later during the procedure when, in the beginning, lobectomy was planned, and during surgery we decided that, unfortunately, pneumonectomy had to be done. So we just harvest it later. And this is one of the patients in the group with the unexpectedly long ischaemic segments. Of course, it is difficult to draw any general conclusion on this small group of patients, but definitely the time of harvesting may have an important impact on the quality.

Dr E. Black(Oxford, UK): If the flap doesn't look so good, do you have any recommendation for what to use, because presumably if you use fluorescein, a bit of pleural or pericardial fat pad is not going to be much.

Dr Piwkowski: In this situation, we prefer intercostal muscle flap because we think that it is the most worthwhile tissue, let's say. But, of course, in cases where there is no possibility of using it, we just use fatty tissue on the right side from the mediastinum with azygos vein or sometimes pericardial flap. But in every case, if we can, we prefer to use intercostal muscle flap.

Author notes

Presented at the 20th European Conference on General Thoracic Surgery, Essen, Germany, 10–13 June 2012.