Applications of intraoperative angiography in head and neck reconstruction.

OBJECTIVE: Laser-assisted angiography with indocyanine green (LAIG) allows objective intraoperative evaluation of tissue vascularity. We endeavored to describe our experience with this technique in the head and neck region.
METHODS: A retrospective review from February 2016 till October 2018 was conducted. We included patients who underwent head and neck procedures in which LAIG was employed. The main outcome was postoperative wound complications. We analysed the influence of LAIG results in intraoperative decision-making process.
RESULTS: Nineteen patients were included, and follow-up was for at least 6 months. LAIG was employed in 11 local flaps, 9 free flaps and 6 cases of pharyngeal closure during total laryngectomies. Wound complications occurred in two cases with distal tip flap necrosis. LAIG findings resulted in changes in decision making intraoperatively in 84% of procedures, which consisted in trimming poorly perfused tissues. There were no pharyngocutaneous fistulas.
CONCLUSIONS: This represents a descriptive report on the use of LAIG on diverse head and neck reconstruction cases, with important impact on the decision-making process. A low number of postoperative wound complications were observed.

Introduction

Without appropriate tissue perfusion even the most elegant reconstruction can be doomed for failure. In head and neck surgery, employing vitally compromised tissues for reconstruction can have substantial consequences both functionally and aesthetically. Traditionally, tissue viability has been assessed using a combination of clinical observation and surgical expertise. This method lacks inherent objectivity, reliability and reproducibility [1].

Laser-assisted angiography with indocyanine green (ICG) dye utilises a near-infrared (NIR) imaging system that provides an intraoperative visual assessment of blood flow within tissues. The ICG molecule possess high affinity to plasma proteins, remaining entirely bound within the intravascular space, which makes it suitable to evaluate perfusion. It is metabolised by the liver and excreted in the kidneys, without any hepatic or renal toxic effects, with a short half-life of 3 to 5 minutes. The risk of anaphylactic reaction is low and is associated with iodine allergy. When the ICG molecules are exposed to light and excited in the NIR spectrum, tissue perfusion can be evaluated by identification and quantification of the fluorescence intensity of the particles: well-perfused tissue appear fluorescent and ischemic tissue appears dark. Post-imaging annotation software present in the NIR devices can analyse results qualitatively and quantitively, decreasing subjectivity and increasing interobserver reliability. This technology has been widely implemented in breast surgery, where authors highlighted the decrease in the occurrence of post-mastectomy necrosis and subsequent implant loss [2-6], as it allowed to identify and remove poorly vascularised tissue.

Nowadays, this technique is not as widely used in head and neck reconstructive surgery, even though there have been prior publications highlighting its benefits [7-10]. The objective of this study was to communicate our initial experience using laser-assisted angiography with ICG (LAIG) in major head and neck reconstructive procedures, describing the range of applications and report our rates of postoperative wound complications.

Materials and methods

A retrospective review from February 2016 till October 2018 at Sunnybrook Health Science Center was conducted. We only included patients who underwent a major head and neck procedure in which LAIG was employed. The decision to employ this technology in certain cases was purely made based on the availability of the equipment as it is mainly used by plastic surgery at our centre. Patient demographics, along with tumour histology, location, type of defect, previous treatment and reconstruction modality were accounted for. In addition, we also calculated the extra-costs and added time to the procedures. Costs were calculated based on total coded and non-coded billing information and contacting the pharmacy department, and time was calculated using a timer feature of the software after the ICG injection. Main outcome was postoperative wound complications (defined as any wound healing complication within 30 days after the surgery). We also analysed if LAIG results affected the intraoperative decision-making process. Research ethics board approval was obtained from Sunnybrook Health Sciences Centre (REB# 21-709.12).

LAICG Dye Angiography Protocol

Before using ICG, all patients were evaluated for an iodine or shellfish allergy. The SPY Elite machine (Novadaq Technologies, Mississauga, Ontario, Canada) is sterilely draped and the area of interest if visualised. Three ml of the reconstituted ICG is injected peripherally and followed by a 10 ml saline flush. Recording of the tissue of interest then commences. In the case of pedicled flaps, recording is done after flap elevation, to confirm adequate vascularisation of the tissue especially at the distal edges. In the case of free flaps, LAIG was used in addition to the hand held doppler to select the most suitable perforator (in Anterolateral Thigh -ALT- flaps), and before completing the flap elevation, to ensure that the selected perforator/s provided adequate vascularisation. It was also used in cases where free flaps were trimmed after their elevation (ALT, Temporoparietal fascia free flap and latissimus dorsi). In post-pharyngectomy cases, the edges of native pharyngeal mucosa, which are high risk areas for fistula formation are evaluated. Areas of concern were considered poorly perfused if, at 45-60 seconds after injection and flush of the IG, the intensity of the site by LAIG was 33% or less of a well perfused area set as a control on the SPY software. This threshold value was determined following previous reports of the literature [11,12]. The poorly perfused area was marked and trimmed back to confirmed well perfused tissue.

Results

Intraoperative angiography was employed in 19 patients (Tab. I). The average patient age was 65 years (range 35 to 84) and 14 were male. All but two cases were operated on for malignant disease; one trauma and one osteoradionecrosis case. LAIG was employed in eleven cases of local flaps, nine cases of free flaps and six cases of pharyngeal closure (three of these were salvage laryngectomies). Of note, LAIG was utilised for assessment in more than one type of reconstruction. The added cost to the procedure by the use of this technology was 30 USD as the ICG is nonproprietary and the pharmacy department facilitated it for us. Regarding the extra time consumed, the total operation was prolonged by an average of 13 minutes.

Table I.

Patients in which LAIG was employed.

Pt	Age	Gender	Pathology	Defect Location	Previous treatment	Reconstructive strategy	LAIG modified approach	Wound complications
1	69	M	Cutaneous SCC	Scalp		Lat	+	-
2	64	F	Metastatic SCC parotid	Preauricular skin		SCAIF	+	-
3	75	M	BCC	Cervical skin		ALT	+	-
4	35	M	Trauma	T-E fistula		TPFF	+	-
5	69	M	Supraglottic SCC (T4aN2c)	Partial Pharyngectomy		ALT/1ary pharyngeal closure	+	Distal tip necrosis of ALT skin paddle
6	60	F	Supraglottic SCC (T3N1)	Partial Pharyngectomy	+	1ary Pharyngeal closure	+	-
7	80	M	Melanoma	Preauricular skin		SCAIF	+	-
8	60	M	Supraglottic SCC (T4aN2a)	Partial Pharyngectomy		TPFF/1ary pharyngeal closure	+	-
9	58	M	BCC	Preauricular skin		Fascial RFFF	-	-
10	51	M	Cutaneous SCC	Cervical skin		SCAIF	+	-
11	52	M	Oral cavity SCC	Through-through buccal mucosa		ALT/Cervicofacial	+	Distal tip necrosis of cervicofacial
12	84	F	BCC	Forehead		Pericranium	-	-
13	52	F	Cutaneous SCC	Preauricular skin		Cervicofacial	+	-
14	67	M	Osteoradionecrosis	Scalp/Calvarial bone		Lat/Cervicofacial	+	-
15	53	M	Hypopharyngeal SCC (T3N2b)	Partial Pharyngectomy	+	ALT/1ary pharyngeal closure	+	-
16	92	M	Mucoepidermoid parotid	Preauricular skin		SCAIF	+	-
17	71	M	Hypopharyngeal SCC (T4N1)	Partial Pharyngectomy		Pharyngeal closure	+	-
18	79	F	Supraglottic SCC (T3N1)	Partial Pharyngectomy	+	SCAIF/1ary pharyngeal closure	+	-
19	61	M	Cutaneous SCC	Preauricular skin		SCAIF/Cervicofacial	-	-

T-E: tracheoesophageal fistula; Lat: latissimus dorsi free flap; ALT: anterolateral thigh free flap; SCAIF: supraclavicular artery island flap; TPFF: temporalis fascia free flap; SCC: squamous cells carcinoma; BCC: basal cells carcinoma.

LAIG findings resulted in changes in decision-making intraoperatively in 84% of the cases: tissues were trimmed when poor distal perfusion was observed in cases of local, free tissue transfer and native pharyngeal mucosa. Minor wound complications occurred in two cases: distal tip flap necrosis in an anterolateral thigh free flap and in a cervicofacial advancement flap. These were managed conservatively. There were no pharyngocutaneous fistulas or total free flap loss.

Example 1

61-year-old male with a neglected preauricular cutaneous squamous cell carcinoma. The patient underwent wide excision of the skin, superficial parotidectomy and neck dissection. To reconstruct the defect a supraclavicular artery island flap (SCAIF) and a cervicofacial advancement flap to replace the skin defect were employed. Both flaps were evaluated using the SPY system confirming good vascularisation. No wound complications were noted at follow up. In this case the distal tip of the cervicofacial advancement flap appeared dusky and had venous flow from the subdermal plexus, but had robust amplified ICG perfusion with values over the threshold so it was left intact.

Example 2

35-year-old quadriplegic male initially injured in a snowboarding accident in 2009. Shortly after injury he required anterior and posterior fusion to stabilise his c-spine. Unfortunately, the anterior plate eroded through his oesphagus creating a trachea-oesophageal fistula which healed as a large laryngeal cleft preventing oral intake. A TPFF was utilised to reconstruct the defect. The flap was divided into leaflets and LAIG confirmed adequate blood supply. This allowed reliable customisation of TPFF in order to re-establish the tridimensional pharyngeal framework, with a fascial flap typical difficult to assess vascularisation. After the microvascular anastomosis, the flap had continued perfusion and the patient had an uncomplicated postoperative course (Fig. 1).

Figure 1.

(A) Temporoparietal fascial free flap harvested to reconstruct the defect; (B) TPFFF divided into leaflets for reconstructive purposes; (C) LAIG analysis on TPFFF confirming adequate blood supply in all its regions.

Example 3

60-year-old female diagnosed with a T3N1 supraglottic squamous cell carcinoma who previously received chemoradiation as part of an organ preservation strategy. Unfortunately, the tumour recurred mandating a salvage surgical approach. LAIG was used to evaluate the pharyngeal strip after tumour extirpation. The SPY system suggested an area of poor perfusion and the mucosa was trimmed to healthy tissue (Fig. 2). The patient had an uneventful post-operative course.

Figure 2.

(A) Intraoperative picture showing a poorly perfused left pharyngeal area after tumour resection; (B) LAIG analysis showed an area of poor perfusion (black dotted line). P: pharyngeal remnant; SMG: submandibular gland; CA: carotid artery; IJV: internal jugular vein; SCMM: sternocleidomastoid muscle.

Discussion

In the head and neck surgical oncology setting, wound complications after reconstructive procedures may lead to catastrophic consequences. Even minor cases of flap necrosis that can be managed conservatively may lead to suboptimal aesthetic results and affect quality of life. In cases of pharyngeal and oral cavity reconstructions, salivary fistulas cause delayed oral intake, and can result in life-threatening conditions such as carotid blow-out syndrome. Reconstruction failures can also indirectly impact on patient prognosis as they may delay the initiation of adjuvant treatment [13,14]. Advocators for the use of LAIG highlight its ability to provide real-time intraoperative and objective assessment of areas of tissue perfusion that are not identified by the surgeon, serving as a superior clinical indicator compared with clinical intraoperative judgement alone [15]. There is extensive breast reconstructive literature reporting the advantages of this technology [3,5], where a reduction in post-mastectomy necrosis rates was found and therefore in implant loss [3,6]. In addition, it has been reported that the use of LAIG had an impact in decision-making intraoperatively in reconstructive surgery [10]. However, in head and neck surgery this technique is not routinely used. We endeavored to report the use of LAIG in a variety of clinical scenarios in head and neck reconstruction.

Necrosis of the distal tip of the flap is the most common complication when using large locoregional fasciocutaneous flaps that extend beyond their vascular pedicle primary and adjacent angiosomes [16]. Cervicofacial flaps and SCAIF have a 9% [17] and 18% [18] risk of partial necrosis respectively. A recent survey capturing 221 head and neck surgeons revealed that the rate of complications with SCAIF was 29.8% [19]. In our series when using SCAIF we experienced only one distal tip necrosis (16,6%), and no complications were reported when using cervicofacial flaps.

In free-flaps with high perforator variability such as the anterolateral thigh (ALT) free flap [20], intraoperative angiography was described as an adjunct to select the most reliable skin perforator, as well as allowing reliable trimming prior to harvesting with good results [21]. Although we did not utilise LAIG for any fibula flap, Beckler et al. [13] also had very promising results with the intraoperative angiography approach observing a reduction in partial skin flap necrosis of fibula flaps in a cohort of 73 patients that underwent oromandibular reconstructions with through and through defects.

The reported post-laryngectomy pharyngocutaneous fistula rates range between 16% to 65% [22]. Two prospective studies including 37 and 41 patients respectively, compared mucosal perfusion detected by this technology in salvage laryngectomies and found that patients who developed postoperative fistulas had lower mucosal perfusion values when compared with patients who did not develop fistulas [12,23], suggesting that there might be a role for LAIG in assessing vascularisation of the pharyngeal mucosa remnant. In the same line as these two last reports, we did not experience any pharyngocutaneous fistulas after pharyngectomies.

Limitations of this study are a small sample size, the retrospective nature of the study, lack of a control group, and the absence of a formal statistical analysis. Nevertheless, it represents a valuable descriptive series highlighting the versatility of the use of LAIG in head and neck reconstruction. It is also important to consider the cost of this technology. In many tertiary-care centres the system is already available as it is largely employed in breast surgery, so the cost centres around the indocyanine green. If the ICG or the equipment are not readily available, the cost can escalate, being 650 USD according to Kanuri et al. [24]. Regarding added time spent in the operating room, it is not a time-consuming procedure, and the average in our series was 13 minutes since the time of the ICG injection.

Conclusions

We employed LAIG in a diverse range of head and neck oncologic and reconstruction applications with encouraging results. Postoperative wound complications were limited (10%) and were all managed successfully with conservative strategies. This technology determined a change in decision making of the surgeon in a high percentage of the procedures. Subsequent studies evaluating its application in head and neck surgery are required to validate our findings.