Introduction

In recent years, surgical therapy has taken on an increasingly decisive role in the therapeutic armamentarium of migraine. Since Guyuron’s initial popularization more than two decades ago [1], an increasing number of scientific publications have described and corroborated the surgical therapy [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17]. The current approach is based on the neurolysis of certain extracranial nerves, called trigger points. The most common, in our experience, are at the occipital, temporal, and frontal levels. The aim of this paper is to describe the techniques we use to treat these three specific trigger points, based on clinical experience gained over the past twelve years [18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36].

Patients and methods

A single-center, retrospective study was conducted on all patients who underwent surgical treatment of migraine between 2011 and 2022. Patients were asked to fill out a headache diary and complete a migraine questionnaire assessing parameters before surgery, after 3 months, and 1 year after surgery.

Surgical technique

Occipital trigger site

In this region, the nerves that may be affected are the greater (medially) and/or the lesser (laterally) occipital, mono- or bilaterally. The third occipital nerve, the most medial of all, is never affected. The patient usually indicates with great precision the point (or points) where the attacks originate, corresponding to the location of the inflamed nerve(s). The ultimate goal of the procedure is to carry out a complete neurolysis (Fig. 1) of the involved nerve(s). In most cases it is possible to notice a conflict between the nerve and branches, usually dilated or ectatic, of the occipital artery (Fig. 2). As we have recently described [37, 38], these vessels have several and specific ultrastructural anomalies: above all, tunica intima hyperplasia and internal elastic lamina fragmentation, as well as phenotypic conversion from a contractile-quiescent to “synthetic”-active state of smooth muscle cells. Surgical procedures are performed as one-day surgery under local anesthesia with sedation. There is no need to cut patients’ hair. According to the patient’s symptoms, a cutaneous incision (of about 5) cm is performed, mono- or bilaterally, horizontal above the superior nuchal line and near the trigger point(s) reported. The blade must be oriented parallel to the adjacent hair shafts, so as not to dissect the underlying bulbs and generate a more visible scar. Subcutaneous and muscular tissues are then undermined in order to identify and visualize the occipital nerves involved. The greater occipital nerve is usually about 3 cm laterally from the midline, while the lesser occipital nerve is about 5 cm more laterally. The vascular-nervous bundle is usually incorporated into abundant fibrous tissue (Fig. 3a), which is the result of chronic tissue inflammation, and which must be removed. Once the fibrous tissue has been removed, the nerve branches and the nearby branches of the occipital artery are visualized, usually ectatic and in close connection with the nerve (Fig. 3b). Complete neurolysis must therefore include coagulation of the facing, dilated vessels (Fig. 3c).

Fig. 1
figure 1

Complete neurolysis of the right greater occipital nerve, highlighting the main trunk (white circle) and peripheral branches (yellow stars)

Fig. 2
figure 2

Right occipital trigger point: strict correlation between the greater occipital nerve (white arrow) and the right occipital artery (blue arrow)

Fig. 3
figure 3

Main stages of occipital surgery: highlighting of the fibrous tissue that incorporates the neurovascular bundle (a); complete neurolysis of the nerve (white arrow), close to an ectatic occipital artery (black arrow) (b); coagulation of all the dilated vascular branches, with complete release of the nerves (white stars) (c)

Temporal trigger site

In this site, the involved nerve is the auriculotemporal one, usually compressed by aneurysmatic branches of the superficial temporal artery. In fact, this site is characterized by a typical symptomatology: a pulsating pain that, from the temporal region, radiates to the frontal, occipital, or vertex areas. The compression point is classically localized at the intersection of two perpendicular lines running parallel to the anterior and superior borders of the ipsilateral auricle. The ectatic branch of the superficial temporal artery can easily be localized by means of a handheld Doppler (Fig. 4). In many patients, this vessel is also clearly visible on the skin surface. The surgical procedure includes complete neurolysis of the nerve and coagulation of dilated vessels that are always present at this trigger point (Figs. 5 and 6).

Fig. 4
figure 4

Identification of the right superficial temporal artery by means of a handheld Doppler

Fig. 5
figure 5

Visualization of the right auriculotemporal nerve (yellow arrow), together with dilated branches of the superficial temporal artery (blue arrow) and vein (black arrow). Note the cutaneous incision parallel to the hair follicles

Fig. 6
figure 6

An edematous, bifurcated right auriculotemporal nerve (white arrow) enveloping an ectatic superficial temporal artery (blue arrow)

Frontal trigger site

In this site, the involved nerves are the supraorbital and the supratrochlear ones. The emergence of the supraorbital nerve from the supraorbital foramen is classically localizable along a straight hemipupillary line. The supratrochlear nerve is located about 2 cm more medially. The nerves are mechanically compressed by the mimic muscles of the upper third of the face that they must cross to reach the skin surface: the depressor supercilii (sharper, thinner and lighter, vertically angled), the corrugator supercilii (deeper, thicker, darker and crumbly, horizontally angled), and the procerus. In addition, it is not uncommon to still visualize ectatic vessels in the vicinity of nerves (Fig. 7), which can contribute to compression. Also, in this case, the procedure involves a complete neurolysis of both nerves (which must be completely freed from adjacent muscle and fibrous tissues) and coagulation of the facing dilated vessels (Figs. 8 and 9).

Fig. 7
figure 7

Dilated vessels (yellow arrow) in close proximity to the right frontal trigger site

Fig. 8
figure 8

The left supratrochlear (top) and supraorbital (bottom) nerves at the end of the procedures

Fig. 9
figure 9

Complete neurolysis of the left supratrochlear (blue spot) and supraorbital (yellow spot) nerves

Results

In the last 12 years, we have operated on 612 patients (71% females, 29% males; age range: 19–83 ys): n.266 occipital, n.248 temporal, n.98 frontal. All the patients had a diagnosis of chronic primary headache or chronic migraine confirmed by a board-certified neurologist, after failure of several attempts of different medical therapies. Patients were asked to fill out a headache diary and complete a migraine questionnaire assessing parameters before surgery, after 3 months, and 1 year after surgery. Data regarding age, sex, age at onset, migraines per month (in days), associated symptoms, severity (on a scale from 1 to 10), inability to work per month (in days), health status, history of neck trauma, and family history were collected. The Migraine Disability Assessment Scale (MIDAS) score was also used to evaluate the degree of disability for each patient [39].

After a 3-month post-operative period, the surgical procedure elicited a positive response in:

  1. a)

    Occipital surgery: remarkable improvement in 95% of patients (86% complete recovery);

  2. b)

    Frontal surgery: remarkable improvement in 87% of patients (32% complete recovery);

  3. c)

    Temporal surgery: remarkable improvement in 88% of patients (66% complete recovery).

Results were substantially confirmed at the 12-month follow-up.

Complications were mild and temporary: mainly, hypoesthesia and numbness of the undermined areas, lasting from 3 weeks to 6 months. We observed the onset of secondary trigger sites in 257 (42%) patients.

Discussion

After Guyuron’s initial, monumental pioneering work [40,41,42], the surgical treatment of migraine has experienced an exponential diffusion both in the USA and in Europe. Actually, several migraine therapy surgeries had been described in the past. Most of these procedures were based on the section or removal of extracranial vessels [9]. In more recent years, our studies have highlighted numerous abnormalities in the vessels facing the trigger points of migraine patients [37, 38]. Interestingly, according to what has been hypothesized and described by several authors in the past [9], an extracranial vascular involvement in the etiopathogenesis of migraine therefore seems to be supported by morphological evidence. In normal arteries, the tunica intima (endothelial layer) consists of a single layer of endothelial cells surrounded by a connective tissue basement membrane with elastic fibers. The middle layer, the tunica media, is primarily composed of smooth muscle and is usually the thickest layer. During migraine surgery, we routinely tunica intima hyperplasia and internal elastic lamina fragmentation, as well as cellular alterations in the tunica media. In particular, a consistent fraction of vascular smooth muscle cells shifted from contractile versus synthetic phenotype [37, 38]. Vascular smooth muscle cells are highly specialized cells that regulate vascular tone and participate in vessel remodeling in physiological and pathological conditions. Phenotypic conversion from a contractile-quiescent to “synthetic”-active state contributes to vascular pathologies. Further research is needed to highlight the therapeutic and surgical implications of these observations. In addition, over the years, numerous trigger points have been described, including the nasal (rhinogenic) and zygomaticotemporal [43,44,45]. In our experience and case studies, the three most represented trigger points were certainly those described: the occipital, the temporal, and the frontal. Regarding the treatment of the occipital trigger point, the most common approach involves (in addition to neurolysis of the occipital nerves) the exeresis of a portion of semispinalis capitis muscle and the transposition of a subcutaneous flap, while the commonly adopted surgical approach for the frontal trigger point (in addition to neurolysis of the supraorbital and the supratrochlear nerves) includes glabellar muscle group avulsion and the use of an adipose flap to cover and protect the nerves [46,47,48]. Our approach, which can be defined as somewhat minimally invasive, allowed us to obtain success rates similar to the previous ones, reducing, in small part, the complexity of the surgical procedure.

The effort to identify decisive and increasingly less invasive therapeutic approaches, such as surgery, is even more important in times of COVID. During the pandemic and lockdown, migraine patients are facing an enormous problem in getting optimum care because of difficulty in access, forced social isolation, and encountering a health system that is getting rapidly overwhelmed [49]. In this perspective, given the excellent therapeutic results expected, in our opinion, the surgical approach is one of the main therapeutic resources for optimal management of this complex disease.

Conclusions

The described procedure is minimally invasive and highly reproducible and allows to obtain a high degree of success with a negligible rate of complications. Nevertheless, given the high percentage of occurrence of secondary trigger points, it is particularly important to warn the patient of this possibility in pre-operative visits.