Superficial temporal artery-to-middle cerebral artery side-to-side microvascular anastomosis using the in-situ intraluminal suturing technique

Abstract

Background

Superficial temporal artery (STA)-middle cerebral artery (MCA) side-to-side microvascular anastomosis can achieve the same clinical effects as traditional STA-MCA end-to-side anastomosis in extracranial-intracranial revascularization surgery, furthermore, STA-MCA side-to-side anastomosis has the lower risk of postoperative cerebral hyperperfusion syndrome (CHS) and the potential to recruit all scalp arteries as the donor sources via self-regulation. Therefore, STA-MCA side-to-side microvascular anastomosis seems to be a revascularization strategy superior to traditional STA-MCA end-to-side anastomosis. In this study, we presented seven cases in which a STA-MCA side-to-side microvascular anastomosis was performed with a 4–5 mm long arteriotomy using the in-situ intraluminal suturing technique.

Methods

Superficial temporal artery (STA)-middle cerebral artery (MCA) side-to-side anastomosis was performed in seven patients using the in-situ intraluminal suturing technique.

Results

The diameters of the recipient MCA and the donor STA were approximately 0.94 mm (range 0.8–1.4 mm) and 1.65 mm (range 1.4–2.0 mm), respectively, and the length of the arteriotomy was approximately 4.71 mm (range 4–5 mm). The MCA was temporarily occluded in approximately 25.00 min (range 20–29 min). 100% patency rates of the STA-MCA microvascular anastomosis were achieved in all patients. No obvious CHS was recorded. Intraoperative Indocyanine green videoangiography (ICG-VA) and postoperative digital subtraction angiography (DSA) demonstrated three different blood flow distribution patterns after the STA-MCA side-to-side anastomosis, the donor MCA received not only antegrade blood flow from the proximal preanastomotic STA but also retrograde blood flow from the distal postanastomotic STA in one case; the donor MCA received all the antegrade blood from the proximal STA without retrograde blood flow from the distal STA in two case; whereas, the recipient MCA territories received only partial antegrade blood flow from the proximal preanastomotic STA.

Conclusions

STA-MCA side-to-side microvascular anastomosis with a 4–5 mm long arteriotomy using the in situ intraluminal suturing technique is a safe and effective revascularization surgery, and the anastomosis can serve as a shunt for blood flow self-regulation.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00701-025-06421-x.

Introduction

The superficial temporal artery (STA) is the most commonly used donor artery for extracranial-to-intracranial (EC-IC) bypass surgery, and end-to-side anastomosis of the STA to the middle cerebral artery (MCA) is the traditional basic low-flow revascularization to the anterior circulation of the brain, which is mainly used for the treatment of steno-occlusive disease, moyamoya disease, etc [3]. Other revascularization strategies involving the STA include frontal and parietal STA‒MCA double-barrel bypass [5], STA‒to‒MCA one donor two recipient artery bypass [1, 2, 6], and STA‒MCA side‒to-side microvascular anastomosis [4, 7, 19]. However, the optimal revascularization strategy remains controversial [3]. Side-to-side microvascular anastomosis involves bypass between two separated arteries without compromising their individual inflow and outflow [17]. Recently, several authors reported that STA-MCA side-to-side anastomosis can achieve the same clinical effects as traditional STA-MCA end-to-side anastomosis [4, 7, 19]. Interestingly, patients with an STA-MCA side-to-side anastomosis had a relatively lower risk of postoperative cerebral hyperperfusion syndrome (CHS) than did those who underwent traditional end-to-side STA-MCA bypass; furthermore, the symptoms of CHS are mild, and the anastomosis has the potential to arouse all scalp arteries as donor sources for revascularization through the intact postanastomotic STA via self-regulation [19]. Therefore, STA-MCA side-to-side microvascular anastomosis seems to be a revascularization strategy superior to traditional STA-MCA end-to-side anastomosis. In this study, we presented seven cases in which an STA-MCA side-to-side microvascular anastomosis was performed with a 4–5 mm long arteriotomy using the in situ intraluminal suturing technique.

Methods and materials

In this study, seven patients who underwent STA-MCA side‒to-side microvascular anastomosis via the in-situ intraluminal suturing technique at our hospital between February 2024 and August 2024 were studied, including 2 patients with symptomatic M1 MCA occlusive disease and 5 patients with moyamoya disease. All side-to-side microvascular STA‒MCA anastomoses were performed by the same neurosurgeon (Z.X.).

The study protocol was approved by the institutional review board at our hospital and was performed in accordance with the Declaration of Helsinki revised in 1983. Written informed consent was obtained from all patients.

Surgical procedures

The patient was placed supine on the operating table, a roll was placed under the ipsilateral shoulder, and the head was rotated to the contralateral side. Doppler sonography was used to identify the course of the STA.

Dissection of the parietal branch of the STA

When the parietal branch of the STA was chosen as the donor artery, a linear skin incision along the course of the STA was made from the level of the zygoma anterior to the tragus to approximately 5 cm above the superior temporal line (Fig. 1A). The STA was dissected under a microscope as previously described [13, 8], the skin was cut just directly over the course of the STA, and then blunt dissection using mosquito forceps was performed under the subcutaneous tissue to identify the STA. The STA was then meticulously dissected along its course under a microscope. The surrounding tissue of the STA was carefully dissected via needle point monopolar cautery or microscissors, and these small branches arising from the STA were carefully coagulated via bipolar forceps and then cut with microscissors. Generally, the STA has a serpentine morphology, which can be evaluated via preoperative digital subtraction angiography (DSA), computed tomography angiography (CTA) or magnetic resonance angiography (MRA). After the serpentine course of the STA is dissected long enough, the STA is tension free and can be easily mobilized to the appropriate location of the potential recipient MCA later. Great care was taken to ensure that the STA was not injured and that the distal STA remained patent without any injury (Fig. 1BC). A piece of cotton with diluted papaverine (15 mg/100 mL of 0.9% saline) was placed on the STA to prevent vasospasm [8]. The temporalis muscle was then incised in a T-shaped fashion via monopolar cautery and then retracted bilaterally.

Fig. 1.

Dissection of the parietal branch of the STA. A linear skin incision along the course of the STA was marked from the level of zygoma anterior to the tragus to 5 cm above the superior temporal line (A), the STA was fully dissected, and the distal STA remained patent without any injury (BC). Dist. = distal; STA = superficial temporal artery; Prox. = proximal

Dissection of the frontal branch of the STA

When the frontal branch of the STA was used, a frontotemporal curved skin incision with the frontal STA inside the skin flap was made (Fig. 2A). After the frontotemporal flap was reflected, the STA was meticulously dissected from the skin flap while the distal STA remained patent (Fig. 2B). The temporalis muscle was subsequently cut along the posterior skin incision and 5 mm below the superior temporal line; thus, a cuff of the temporalis muscle was left on the superior temporal line for later reapproximation, after which the temporalis muscle was retracted anteriorly and inferiorly.

Fig. 2.

Dissection of the frontal branch of the STA. A frontotemporal curved skin incision with the frontal STA inside the skin flap was used (A). After the frontotemporal flap was reflected, the STA was meticulously dissected from the skin flap while the distal STA remained patent (B). Dist. = distal; STA = superficial temporal artery; Prox. = proximal

Craniotomy

Two burr holes were made under the course of the dissected STA, one proximal (temporal) and one distal (either parietal or frontal), and then a 4*5 cm craniotomy around the sylvian fissure was performed after the dura was carefully detached from the bone; moreover, measures were implemented to ensure that the dissected STA and the middle meningeal artery were not damaged during the craniotomy.

STA-MCA side-to-side anastomosis using the in-situ intraluminal continuous suturing technique

The dura was cut in a cruciate fashion, and the largest M4 cortical artery almost parallel to the course of the STA without vascular tension was chosen as the recipient artery. Both the donor and recipient vessels were placed together in a parallel position, and approximately 3 times the vessel diameters of the planned arteriotomies were marked with ink. First, the marked segment of the STA was temporally occluded with two temporary aneurysm clips, and the two clips were placed far from the anastomotic area so that these clips would not interfere with the subsequent microvascular procedure later. A beveled 25-gauge needle was then used to pierce the vessel (Fig. 3A), and the lower blade of the microscissors was subsequently inserted into the puncture hole (Fig. 3B). An arteriotomy on the anterior wall of the STA was subsequently performed along the marked line (Fig. 3C). Heparinized saline (100 U/mL) was used to wash the lumen of the vessel, and the adventitia around the STA arteriotomy was carefully removed. A 10–0 nonabsorbable monofilament polypropylene suture (W2790, 13 cm in length, 3.8 mm, 3/8 circle taper point, BV 75–3, Ethicon) was subsequently placed at one corner of the arteriotomy of the STA. Then, the STA and MCA were placed parallelly again to check the location of the MCA arteriotomy, and the MCA was prepared (Fig. 4A). Small branches from the recipient segment of the MCA were coagulated and cut, and larger branches from the recipient MCA were temporarily clipped during the procedure. A piece of rubber dam was placed under the MCA, and the MCA was temporarily clipped. The distance between the corner of the planned arteriotomy and the clip should be at least 1 mm, as we previously described for side-to-side anastomosis in an animal training model [17]. Then, an arteriotomy of the same length as the STA arteriotomy was created on the anterior wall of the MCA using the same technique described above. The arachnoid around the MCA arteriotomy was carefully removed. Side-to-side anastomosis was performed using the in-situ intraluminal continuous suturing technique under high magnification. The first suture placed at one corner of the STA arteriotomy was connected with one corner of the MCA arteriotomy as one staying suture. Then, the thread was introduced into the lumen of one of the arteries under the first staying suture (Fig. 4B), the posterior wall was closed intraluminally using the continuous suturing technique, and these sutures were placed loosely (Fig. 4C). The depth of the sutures was one to two times the thickness of the STA wall, and the spacing was approximately 3–5 sutures per millimeter. After all the intraluminal continuous suture loops were loosely placed in the posterior wall, these loose loops were tightened sequentially in the appropriate place with appropriate tension [15–17]. Next, the suture thread exited the vessel when the suture reached the other corner of the arteriotomy, and a surgical knot was made at this corner of the arteriotomy using the same thread. The anterior wall of the anastomosis was subsequently closed with the same thread or a new stitch using the extraluminal continuous suturing technique (Fig. 4D). Finally, the temporary clips on the MCA were released to restore blood flow. Some minor bleeding from the anastomotic site was controlled with gentle compression using a SURGICEL FIBRILLAR (Johnson & Johnson Medical Devices). If there was some obvious bleeding or leakage, additional interrupted sutures were placed either with or without temporary clips. The temporary clips on the STA were subsequently removed to restore the blood flow of the STA-MCA side-to-side microvascular anastomosis (Fig. 4E). Indocyanine green videoangiography (ICG-VA) was used to confirm the immediate patency of the anastomosis (Fig. 4F, Video 1).

Fig. 3.

Performance of the arteriotomy of the STA. A beveled 25-gauge needle was used to pierce the vessel after the STA was temporarily clipped (A), the lower blade of the microscissors was subsequently inserted into the puncture hole (B), and an arteriotomy on the anterior wall of the STA was subsequently created along the marked line (C). STA = superficial temporal artery

Fig. 4.

STA-MCA side-to-side anastomosis using the in-situ intraluminal continuous suturing technique. The STA and MCA were placed parallelly (A). After the first staying suture was placed at one corner of the arteriotomy, the thread was introduced into the lumen of one of the arteries under the first staying suture (B), the posterior wall was closed intraluminally using the continuous suturing technique (C), and the anterior wall was closed extraluminally to finish the anastomosis (D), the temporary clips were subsequently removed to restore the blood flow of the STA-MCA side-to-side microvascular anastomosis (E), and patency of the anastomosis was confirmed by intraoperative Indocyanine green videoangiography (F). (Scale bar = 1 mm). Dist. = distal; MCA = middle cerebral artery; STA = superficial temporal artery; Prox. = proximal

Closure

The leaflets of the dura were reversed to ensure contact with the brain surface as an indirect revascularization source, and a piece of artificial dural substitute was used to reconstruct the dura. The superior and inferior edges of the bone flap were trimmed so that the proximal and distal donor STAs would not be compressed; then, the bone flap was reapproximated with titanium plates. The temporalis muscle and the skin were reapproximated anatomically.

Follow-up angiographic examination

Postoperative DSA and CTA were used to evaluate the patency of the STA-MCA side-to-side anastomosis, postoperative DSA was performed at one week after bypass, and CTA or DSA was performed one month after surgery.

Statistical analysis

Descriptive summary statistics are presented as the means ± standard deviations (SDs). Analyses were conducted using SPSS 19.0 (SPSS, Inc., Chicago, IL, USA).

Results

In this study, seven adult patients (mean [range] age 51.7 [40–66] years) were included. The diameters of the recipient MCA and the donor STA were approximately 0.94 mm (range 0.8–1.4 mm) and 1.65 mm (range 1.4–2.0 mm), respectively, and the length of the arteriotomy was approximately 4.71 mm (range 4–5 mm). Three frontal STA‒MCA and four parietal STA‒MCA side‒to-side microvascular anastomoses were performed in this study. The size and diameter of the donor STA and the recipient MCA were nearly the same in Patient 1, and thick-walled, large-diameter, strong STAs were sutured to a very thin-walled, small-diameter fragile MCA in the other six patients. The MCA was temporarily occluded for approximately 25.00 min (range 20–29 min). 100% of the STA-MCA side-to-side microvascular anastomoses were confirmed to be immediately patent via intraoperative ICG-VA, and all seven STA-MCA bypasses were confirmed to be patent via computed tomography angiography (CTA) or digital subtraction angiography (DSA) one month later. In four patients, both the proximal and distal MCA and STA were bright through antegrade blood flow of the proximal STA almost at the same time intraoperatively via ICG-VA. Postoperative DSA also confirmed that both the distal and proximal MCA and the distal STA were filled by the antegrade blood flow of the proximal STA on ipsilateral external carotid artery DSA, and no blood flow was observed from the recipient MCA to the donor STA on ipsilateral internal carotid artery DSA. In patient 6, the four limbs of the anastomosis were turned to be bright at the same time as the other four patients via intraoperative ICG-VA; however, postoperative DSA showed that only a small segment of the distal postanastomotic STA was visible while the anastomosis, the proximal STA and recipient MCA were patent. Interestingly, in two patients (patient 1 and 7), although intraoperative ICG-VA showed four limbs of the STA-MCA side-to-side anastomosis were patent, the distal post-anastomotic STA was visible several seconds after the proximal preanastomotic STA and recipient MCA were bright. Furthermore, postoperative DSA demonstrated that the distal post-anastomotic STA was hardly visible, while the other three limbs of the anastomosis was patent. Additionally, retrograde blood flow was observed from the distal STA to the recipient MCA on the contralateral external carotid artery DSA in patient 1.

Postoperative computed tomography (CT) and magnetic resonance imaging (MRI) revealed that there was no bypass-related ischemia or hematoma. The symptoms of four patients were gradually improved, especially the function of the extremities of two patients significantly improved after bypass (Patients 2 and 4), and the neurological status of the other three patients remained unchanged. No obvious CHS was observed in the seven patients (Table 1). These patients were discharged home without any neurological deficits, the skin incision healed well, and no symptoms of steal phenomena were observed.

Table 1.

Characteristics of the Patients and measurements of the STA-MCA side to side microvascular anastomosis

Patient	Age, yr	Sex	Side	Indication	Parietal or Frontal STA	STA diameter (mm)	MCA diameter (mm)	The length of arteriotomy (mm)	MCA temporary occlusion time (minute)	Intraoperative immediate bypass patency	Postoperative bypass patency	complications	Postoperative symptoms
1	55	F	R	Moyamoya disease	Parietal	1.5	1.4	5	29	Patent	Patent	N	No change
2	56	F	R	Moyamoya disease	Parietal	1.4	0.8	4	22	Patent	Patent	N	Improved
3	62	M	L	MCA Occlusion	Frontal	1.4	1	4.5	21	Patent	Patent	N	improved
4	66	M	R	MCA Occlusion	Frontal	1.5	1	5	25	Patent	Patent	N	Improved
5	41	F	L	Moyamoya disease	Frontal	2	0.8	5	29	Patent	Patent	N	Improved
6	42	F	L	Moyamoya disease	Parietal	1.8	0.8	4.5	20	Patent	Patent	N	No change
7	40	M	R	Moyamoya disease	Parietal	2	0.8	5	29	Patent	Patent	N	No change

STA = superficial temproal artery, MCA = middle cerebral artery

Illustrative case 1 (parietal branch of the STA-MCA side-to-side anastomosis)

A 56-year-old female (patient 2) was admitted to our department due to hemorrhagic moyamoya disease. The patient underwent a left-parietal craniotomy to remove the parietal hematoma seven months prior. She experienced right-sided hemiparesis immediately after the onset of the parietal hematoma, and she exhibited Grade 4 strength in the upper and lower extremities at admission. Preoperative CTA and DSA revealed bilateral ICA stenosis and bilateral M1 occlusions, which confirmed the diagnosis of moyamoya disease (Fig. 5A). A left-parietal branch of the STA-MCA side-to-side microvascular anastomosis was successfully performed, with an arteriotomy measuring 5 mm in length (Fig. 5B). Intraoperative ICG-VA and postoperative DSA on postoperative day 6 confirmed the patency of the anastomosis (Fig. 5CDE), and no blood flow was observed from the MCA to the STA (Fig. 5F).

Fig. 5.

Illustrative case 1 (Parietal branch of the STA-MCA side-to-side anastomosis). Preoperative CTA and DSA revealed bilateral ICA stenosis and bilateral M1 occlusions, which confirmed the diagnosis of moyamoya disease (A). A left-parietal branch of the STA-MCA side-to-side microvascular anastomosis was successfully performed with an arteriotomy measuring 5 mm in length (B). Immediate patency of the anastomosis was confirmed by intraoperative Indocyanine green videoangiography (C). And postoperative patency of the anastomosis (Arrow) was confirmed by postoperative DSA on postoperative day 6 (D). 3D reconstruction of the STA-MCA side-to-side anastomosis was illustrated (E). No blood flow was observed from the MCA to the STA (F). (Scale bar = 1 mm). Dist. = distal; MCA = middle cerebral artery; STA = superficial temporal artery; Prox. = proximal

Illustrative case 2 (frontal branch of the STA-MCA side-to-side anastomosis)

A 66-year-old male (patient 4) was diagnosed with occlusion of the right M1 segment of the MCA and was admitted to our department due to left-sided mild hemiparesis ten years prior (Fig. 6A). Preoperative magnetic resonance imaging (MRI) revealed ischemic changes in the right basal ganglion (Fig. 6B). Computed tomography perfusion (CTP) revealed a prolonged mean transit time (MTT) predominantly in the right MCA territory (Fig. 6C). A right frontal branch of the STA-MCA side-to-side microvascular anastomosis was successfully performed, with an arteriotomy measuring 5 mm in length. Intraoperative ICG-VA and postoperative DSA confirmed the patency of the STA-MCA bypass (Fig. 6D), and postoperative DSA also revealed that there was no blood flow from the donor MCA to the recipient STA (Fig. 6E). Postoperative CTP revealed significant improvement in the perfusion of the right MCA territory after surgery (Fig. 6F).

Fig. 6.

Illustrative case 2 (Frontal branch of the STA-MCA side-to-side anastomosis). Preoperative CT angiograph confirmed the occlusion of the right M1 segment of the MCA (White arrow A). Preoperative magnetic resonance imaging revealed ischemic changes in the right basal ganglion (B), preoperative CTP revealed a prolonged MTT predominantly in the right MCA territory (C). Postoperative DSA confirmed the patency of the STA-MCA bypass (Arrow D), and postoperative DSA revealed that there was no blood flow from the donor MCA to the recipient STA (E). Postoperative CTP showed significant improvement in the perfusion of the right MCA territory after surgery (F). CTP = computed tomography perfusion; Dist. = distal; MTT = mean transit time; MCA = middle cerebral artery; STA = superficial temporal artery; Prox. = proximal

Illustrative case 3 (parietal branch of the STA-MCA side-to-side anastomosis)

A 55-year-old female (patient 1) was diagnosed with moyamoya disease (Fig. 7A) and admitted to our department; no significant neurological deficits were detected. The diameters of the STA and MCA were 1.5 mm and 1.4 mm, respectively, and the sizes and thicknesses of the donor and recipient arteries were almost perfectly matched (Fig. 7B). A parietal branch of the STA-MCA side-to-side anastomosis was successfully performed (Fig. 7C), and the patency of the anastomosis was confirmed via intraoperative ICG-VA and postoperative DSA. Interestingly, at the beginning of the intraoperative ICG-VA, the preanastomotic STA and the M4 recipient arteries were easily visible at the same time, whereas the distal postanastomotic STA remained dark (Fig. 7D), and only a short segment of the postanastomotic STA was intermittently brightened by the antegrade blood flow of the preanastomotic STA (Fig. 7E). The postanastomotic STA subsequently turned bright through retrograde blood flow from the postanastomotic distal STA approximately 4 s later, and the distal STA and the recipient MCA remained bright even after the ICG signal on the proximal preanastomotic STA obviously decreased (Fig. 7F, Video 2). These interesting blood flow results were also observed on postoperative DSA images taken on postoperative day 5. Robust antegrade blood flow was observed through the anastomosis from the proximal preanastomotic STA to the MCA on ipsilateral external carotid artery angiography images; only a very small segment of the postanastomotic distal STA was observed, as shown on intraoperative ICG-VA images, and the remaining distal STA was hardly visible (Fig. 7G). On left external carotid artery angiography images, retrograde blood flow into the right STA-MCA side-to-side microvascular anastomosis was observed via the postanastomotic distal STA, and the retrograde blood flow was derived mainly from the interconnection of the right distal postanastomotic STA with the left parietal branch of the STA and occipital arteries (Fig. 7H). No blood flow was observed from the recipient MCA to the donor STA on ipsilateral internal carotid artery DSA images (Fig. 7I).

Fig. 7.

Illustrative case 3 (Parietal branch of the STA-MCA side-to-side anastomosis). Preoperative DSA confirmed the diagnosis of moyamoya disease (A). Intraoperatively, the donor STA and the recipient MCA were almost perfectly matched (B). A parietal branch of the STA-MCA side-to-side anastomosis was successfully performed (C). At the very beginning of the intraoperative ICG-VA, the preanastomotic STA and the M4 recipient arteries were easily visible at the same time, whereas the distal post-anastomotic STA remained dark (D); and only a short segment of the postanastomotic STA was intermittently brightened (Red arrow) by the antegrade blood flow of the pre-anastomotic STA (E). Then the post-anastomotic STA subsequently turned bright, and the distal STA and the recipient MCA remained bright even after the ICG signal on the proximal pre-anastomotic STA obviously decreased (F). Robust antegrade blood flow was observed through the anastomosis (Arrow) from the proximal pre-anastomotic STA to the MCA on ipsilateral external carotid artery angiography images, but the distal post-anastomotic STA was hardly visible (G). On left external carotid artery angiography images, retrograde blood flow into the right STA-MCA side-to-side microvascular anastomosis was observed via the post-anastomotic distal STA, and the retrograde blood flow was mainly derived from the interconnection of the right distal post-anastomotic STA with the left parietal branch of the STA and occipital arteries (H). No blood flow was observed from the recipient MCA to the donor STA on ipsilateral internal carotid artery DSA images (I). Dist. = distal; DSA = digital subtraction angiography; ICG-VA = Indocyanine green videoangiography; MCA = middle cerebral artery; STA = superficial temporal artery; Par = Parietal; Prox = proximal; OA = Occipital artery

Discussion

STA-MCA end-to-side bypass is the widely accepted direct extracranial‒intracranial revascularization surgery for symptomatic patients with steno-occlusive disease, moyamoya disease, etc [3]. Side-to-side microvascular anastomosis involves bypass between two separated arteries without compromising their individual inflow and outflow [17]. Recently, several authors reported that STA-MCA side-to-side anastomosis can achieve the same clinical effects as traditional STA-MCA end-to-side anastomosis [4, 7, 19]. Interestingly, patients who underwent STA-MCA side-to-side anastomosis had a lower incidence of postoperative CHS than did those who underwent STA-MCA end-to-side anastomosis; furthermore, the symptoms of CHS are milder, the duration of CHS is shorter, and the anastomosis has the potential ability to arouse all scalp arteries as donor sources for revascularization through the intact postanastomotic distal STA via self-regulation [19]. These advantages suggest that STA-MCA side-to-side microvascular anastomosis may be a potential revascularization method superior to traditional end-to-side STA-MCA anastomosis.

In 2020, Lang [7] reported the first true side-to-side STA‒MCA microvascular anastomosis while preserving the patency of the distal STA in a patient with moyamoya disease. Later, Daggubati [4] reported another side-to-side STA-MCA bypass in a patient with symptomatic common carotid artery occlusion. In 2022, Morcos [6] used sequential side-to-side temporal and end-to-side frontal bypasses from the same donor parietal STA to revascularize ischemic brain tissue, which is also called the one-donor, two-recipient (1D2R) bypass technique. However, the distal STA must be transected during the revascularization procedure, as in the traditional procedure of STA-MCA end-to-side anastomosis; therefore, 1D2R bypass is not a true side-to-side STA-MCA microvascular anastomosis. Subsequently, Jianjian Zhang [19] analyzed a series of 35 cases of STA‒MCA side‒to-side microvascular anastomosis for moyamoya disease.

Side-to-side microvascular anastomosis is considered the most difficult type of anastomosis because it is commonly performed in-situ in a deep and narrow surgical corridor using the difficult in-situ intraluminal continuous suturing technique, and the risk of vascular endothelium injury is relatively greater than that of the extraluminal suturing technique, which requires a high-level microvascular technique to avoid endovascular thrombosis due to endothelium injury. However, the intraluminal suturing technique can be used to safely and efficiently resolve many difficult microvascular anastomotic situations when the traditional interrupted extraluminal suturing technique is impossible or difficult to perform [16]. Currently, the side-to-side in-situ intraluminal suturing technique has been successfully and sufficiently used in anterior cerebral artery (ACA)-ACA bypass in the interhemispheric fissure, posterior cerebral artery (PCA)-superior cerebellar artery (SCA) bypass in the crural or ambient cistern, and middle cerebral artery (MCA)-MCA bypass in the deepest sylvian fissure, and these side-to-side microvascular anastomoses could remain patent in the long term [9–12, 14]. Although the intraluminal suturing technique is difficult to perform, it can be mastered through deliberate practice in microsurgical laboratories prior to its use in real human operations. Deliberate practice in microsurgical laboratories can achieve proficiency in microsurgical techniques, and “so-called” difficult anastomoses can be performed safely and efficiently in a limited vascular occlusion time. The use of living animal training models is optimal for simulating real anastomoses; for example, arterial side‒to‒side microvascular anastomoses between two perfectly matched arteries, such as the common carotid artery (CCA)-CCA and common iliac artery (CIA)-CIA anastomoses, have been successfully established in rat training models [16, 17]. However, mismatched vessels are frequently and inevitably encountered in real surgery, regardless of precise preoperative surgical plan [18, 17], especially in revascularization surgery for those with moyamoya disease, in which a thick-walled, large-diameter, strong STA is frequently anastomosed to a very thin-walled, small-diameter fragile MCA. In this study, vascular mismatch was encountered in 6 patients, and only one donor STA was almost matched with the recipient MCA. Therefore, it is very important to practice anastomoses between vessels with differences in diameter, thickness, texture and consistency. Arteriovenous anastomoses using rat vessels, such as CCA-external jugular vein (EJV) side-to-side anastomosis, are used as models to mimic this type of anastomotic situation [17].

The area targeted for manipulation for STA-MCA side-to-side anastomosis is superficially located on an open and shallow working space over the surface of the brain tissue, and shallow STA-MCA side-to-side anastomosis is relatively easier to perform than deep and narrow surgical corridors. Lang [7] and Daggubati [4] successfully performed STA-MCA side-to-side microvascular anastomosis using the intraluminal suturing technique. Lang [7] successfully performed a patent STA-MCA side-to-side bypass in a moyamoya patient. In their procedure, they created an arteriotomy measuring approximately 3 times the vessel diameter and then created an STA-MCA side-to-side anastomosis using the intraluminal technique; however, they did not mention the size of the STA or MCA or the total time needed to clip the MCA. In Daggubati’s case [4], the patient underwent ligation of the CCA due to a ruptured pseudoaneurysm at the carotid bulb because of neck cancer, and the blood flow of the ipsilateral distal STA was derived mainly from the contralateral external carotid artery branches over the scalp vertex. The author subsequently successfully created a STA‒MCA side‒to-side bypass to revascularize the MCA territory through retrograde blood flow through the distal STA. In Daggubati’s case [4], the diameters of the STA and MCA were approximately 2 mm and 1.1 mm, respectively, but they did not mention the length of the arteriotomy. The STA‒MCA side‒to-side anastomosis was also performed using the intraluminal technique, the total clipping time was 65 min, and the bypass was confirmed to be patent via intraoperative ICG-VA and DSA two months after the surgery. Instead of using the difficult intraluminal suturing technique, as Lang [7] and Daggubati [4] reported, Zhang Jianjian [19] used the traditional extraluminal suturing technique to finish their STA-MCA side-to-side anastomosis; their STA-MCA side-to-side arteriotomy was approximately 1.2–1.5 mm, they generally placed 3–4 interrupted stitches for each wall to finish the anastomosis, they achieved 100% immediate intraoperative bypass patency rates, and only one case of graft occlusion was recorded in the follow-up. The average temporary occlusion time of the MCA is approximately 23.34 ± 4.38 min. The long-term advantages of using the interrupted suturing technique instead of the continuous suturing technique are that the arteriotomy may be expanded and that the time needed to suture a small arteriotomy to decrease the MCA occlusion time may be shorter. However, the interrupted suturing technique requires a surgical knot after every stitch, making it more time-consuming than the continuous suturing technique [16]. In this study, we created a long arteriotomy approximately 4–5 mm between the STA and MCA, and there was no need for the arteriotomy to be expanded after surgery. We completed the anastomosis in approximately 20–29 min using the in-situ intraluminal suturing technique, and a 100% patency rate was achieved.

In the present study, the STA remained patent during surgery. If there is no appropriate recipient artery for the anastomosis, the dissected STA with patent distal blood flow may be placed over the surface of the brain tissue as an indirect revascularization procedure. Moreover, if an STA-MCA side-to-side anastomosis was successfully created, the other segment of the STA in contact with the brain surface could also serve as an additional indirect revascularization source. In the procedure, we always prepared the STA prior to the MCA and placed a stich at one corner of the STA arteriotomy; thus, the MCA occlusion time could be minimized.

In traditional end-to-side, double barrel or 1D2R bypass, the blood supply of the skin may be compromised by transection of the distal STA. Moreover, all the blood from the donor STA flows into the recipient MCA, and CHS may occur when the recipient cerebral territories do not require much blood flow. The advantages of 1D2R bypass are that the blood flow of the two cerebral territories is augmented by one single parietal donor STA, and the possibility of CHS in 1D2R bypass may be lower than that of traditional STA end-to-side bypass because the blood flow from the donor STA is distributed into two cerebral vascular territories, and blood flow between the two vascular territories may be regulated by itself depending on the vascular requirements. Although the two anastomotic areas are located in separate areas, blood flow from the two anastomotic STAs may accumulate in specific brain tissue through the blood flow network, and if the two separated vascular territories cannot withstand blood flow from the same donor STA, CHS may occur.

The true STA‒MCA side‒to-side microvascular anastomosis connects the blood flow between the STA and MCA while preserving the patency of the postanastomotic distal STA without compromising the inflow and outflow of the STA and MCA. Therefore, the distal STA continues to supply the scalp; thus, the blood supply of the scalp could be protected to increase skin wound healing in STA-MCA side-to-side anastomosis compared with transection of the distal STA in end-to-side anastomosis. There are three different cerebral blood flow distribution patterns after STA-MCA side-to-side bypass. If the ischemic hemisphere requires massive blood flow, side-to-side anastomosis may recruit not only antegrade blood flow from the preanastomotic proximal STA but also retrograde blood flow from the postanastomotic distal STA from the vascular network of the entire scalp with the distal STA, such as the ipsilateral other branch of the STA, contralateral STA, ipsilateral and contralateral occipital arteries and postauricular arteries [19]. In this study, patient 1 exhibited this type of blood flow distribution pattern. In addition to robust antegrade blood flow from the proximal STA, retrograde blood flow from the distal STA with the interconnection of scalp arteries was also identified, especially from the contralateral parietal branch of the STA and occipital arteries. These blood flow distribution patterns were very similar to those reported by Lang [7]. In Lang’s case [7], both antegrade blood flow from the proximal STA and retrograde blood from the distal STA with the connection of the contralateral occipital artery were also observed [7]. Zhang [19] also presented an interesting result in a moyamoya patient three months after successful parietal STA-MCA side-to-side bypass; the proximal preanastomotic segment of the parietal STA became much thinner but was still patent and provided blood flow into the brain, and the distal frontal STA, which was not observed on preoperative DSA images, was observed and provided very strong blood flow to the brain through the side-to-side anastomosis from the connection with the distal parietal STA [19]. These surgical results demonstrated that the ischemic hemisphere of the recipient MCA may receive blood from either the proximal or the distal STA through side‒to-side anastomosis if the hemisphere requires a large amount of blood; however, how much blood flow the hemisphere requires and how the blood flow is redistributed in the long term are unknown and need to be studied further. The second blood flow distribution pattern is that if the recipient MCA territory requires almost the blood flow as the STA, all the antegrade blood from the proximal STA will flow into the recipient MCA without retrograde blood flow from the distal STA, and the distal STA will not receive any blood from the proximal STA. Therefore, the side-to-side STA-MCA anastomosis serves as an end-to-side STA-MCA anastomosis. This type of blood flow distribution pattern was observed in the postoperative DSA for patients 6 and 7. However, the amount of blood flow that the recipient cerebral territory requires could not be precisely predicted, as it changed according to the level of blood pressure, activities and metabolism of the brain issues, etc. Therefore, the blood flow distribution pattern dynamically transforms to other patterns through the self-regulation function of side-to-side anastomosis. The third blood flow distribution pattern after the STA-MCA side-to-side bypass indicates that if the extent of ischemia in the hemisphere is not severe or if the ischemic hemisphere does not require much blood flow at that time, even if the hemisphere is already under severe conditions of ischemia, the excessive blood flow may be shunted to the postanastomotic distal STA while the intracranial blood flow may not be aggressively changed [19]. Then, the ischemic hemisphere may gradually obtain appropriate blood flow, as it requires via self-regulated side-to-side STA-MCA bypass. In the other four patients in our study, ICG-VA or DSA confirmed that the distal postanastomotic donor STA was observed through the ipsilateral STA antegrade blood flow, which demonstrated that the recipient ischemic hemisphere did not need much blood flow at that time, and the brain tissue received only partial antegrade blood flow from the proximal preanastomotic STA; however, interestingly, mildly paralyzed extremities in one patient with this blood flow distribution pattern gradually improved one week after bypass, so we thought that these ischemic situations gradually improved in a safe and slow way while minimizing the aggressive blood flow changes in the ischemic area. Therefore, the blood supply of the scalp is better protected in STA-MCA side-to-side anastomosis; moreover, the hemodynamic changes in the recipient MCA cerebral territories after STA-MCA side-to-side bypass are safer via the self-regulation shunting function of the bypass, and the blood flow to the MCA territories through the donor STA may be maximized through either antegrade or retrograde blood flow.

However, one of the potential problems associated with STA-MCA side-to-side anastomosis is that blood may flow from the MCA to the STA; thus, ischemic problems may worsen after surgery. Fortunately, no blood flow was observed from the MCA to the STA with internal carotid artery injection on DSA in this study, and no patient experienced symptoms of the steal phenomenon; however, further investigation is needed.

There were several limitations in this study. First, we presented only seven cases in which the in-situ intraluminal continuous suturing technique was used by the same neurosurgeon, and many more cases need to be studied further. Second, we followed these patients for only 1–6 months, and the long-term patency rate of the STA‒MCA side‒to-side anastomosis and how blood flow is redistributed after surgery should be further investigated.

Therefore, STA-MCA side-to-side anastomosis with a long arteriotomy using the in situ intraluminal suturing technique is a safe and effective revascularization strategy, and the anastomosis can serve as a shunt for blood flow self-regulation. When the ischemic hemisphere requires massive blood flow, side-to-side anastomosis may recruit not only antegrade blood flow from the preanastomotic proximal STA but also retrograde blood flow from the postanastomotic distal STA. When the ischemic hemisphere does not require much blood flow from the proximal preanastomotic STA, excessive blood is shunted away through the anastomosis to the distal postanastomotic STA.

Conclusions

STA-MCA side-to-side microvascular anastomosis with a 4–5 mm long arteriotomy using the in situ intraluminal suturing technique is a safe and effective revascularization surgery, and the anastomosis can serve as a shunt for blood flow self-regulation.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conception and design: Zongyu Xiao. Analysis and interpretation of data: all authors. Drafting the article: Zongyu Xiao. Critically revising the article: Zongyu Xiao. Reviewed submitted version of manuscript: all authors.

Funding

This study was supported by project of the National Natural Science Foundation of China (82303851), Guangdong Basic and Applied Basic Research Foundation (2022A1515111065), Gusu Talent Program.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethical approval

The study protocol was approved by the institutional review board at our hospital and was performed in accordance with the Declaration of Helsinki revised in 1983. Written informed consent was obtained from all patients.

Disclosure

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.

Competing interests

The authors declare no competing interests.