**3. Indications for neuromodulation**

Primary headaches like tension headache, migraine (CM) or trigeminal autonomic cephalagias (TACs) show the highest incidence. TACs are particularly incapacitating [5]. Most cases can be controlled with medication and physiotherapy. Abuse of medication is common with these patients, via the dose, the drugs or both [6]. When the pharmacological and conservative treatments fail, surgery may be considered. In the past, ablative surgical techniques have been applied. These techniques have been replaced by neuromodulation techniques. In them, the anatomical structures are not lesioned, but instead, the electric impulses block the nervous structures in a reversible fashion [7]. These techniques can be subdivided into two broad categories: noninvasive and invasive [1]. The noninvasive options include transcranial stimulation either electric [8–10] or magnetic [11–13] and transdermal stimulation of occipital [14, 15], supraorbital [9, 14–19] or vagus [20–23] nerves. Invasive procedures include stimulation of occipital [5, 17, 24, 25–35], supraorbital [19, 31, 36], infraorbital [31] or greater auricular [37] nerves as well as sphenopalatine ganglion [38–44], cervical spinal cord [45–48] or hypothalamus [4, 25, 27, 49–52]. Noninvasive neuromodulation techniques are user friendly and have low costs and few and minor side effects [8–15, 19–23]. Unfortunately, their effectivity is lower than their invasive counterparts. Invasive neuromodulation is reserved for the most refractory cases, as they are associated with increased aggressiveness, more severe adverse events (AE) and higher costs [7]. All costs have to be taken into consideration. The full cost of neuromodulation would include

the disability grants, as well as further possible treatments for AEs [33, 34, 53].

ing time and resources.

124 Transcranial Magnetic Stimulation in Neuropsychiatry

analysed, describing the possible future trends.

**2. Historical aspects of neuromodulation**

Taking the net expenditure into consideration, some have reported that the reductions in cost are evident at 5-year postimplantation [53]. In any case, invasive neuromodulation must only be used in the most refractory cases and only after all other medical and noninvasive treatments have failed [46]. This is particularly important due to the high incidence of AEs and the possibility of new surgical procedures to solve them [29, 33]. A trial of temporary stimulation is required to evaluate the possible response of a definitive implant [45, 46, 54] to avoid wast-

These techniques present promising new treatment strategies. The available evidence will be

Electricity to treat chronic headaches was first used in ancient Rome [55], but it was not until the 1950s that neuromodulation was used in the treatment of chronic pain disorders [55, 56]. Thalamic stimulation to treat chronic headaches was introduced in 1976 [57] and percutaneous peripheral nerve stimulation a year later [58]. At the beginning of the 1990s, hypothalamic deep brain stimulation (hDBS) was applied to the treatment of some chronic headache syndromes and particularly in TACs [59]. The first report of occipital nerve stimulation (ONS) to treat occipital neuralgia was in 1999 [60]. In the year 2000, the hypothalamic stimulation was applied in the treatment of drug-resistant cluster headache (CH) [61]. The first two cases of supraorbital nerve stimulation (SONS) were reported in 2002 [62]. The first cases of hDBS in the treatment of CH were reported in 2003 [61]. In 2004, the ONS was applied in the treatment The first step is to diagnose the patient and select an appropriate treatment by an experienced team that is familiarised with all available treatments. Neuromodulation techniques are indicated in cases that have failed all other medical treatments available for this specific headache type. It is also recommended that patients receive a psychological assessment.

The next step is to attempt noninvasive neuromodulation techniques particularly useful in this type of chronic headache. Should all fail, a period of temporary trial stimulation is suitable [33]. Patients showing no response are not implanted and are redirected to other forms of treatment. This temporary stimulation also helps to predict the results to be expected if the definitive implant is attempted [32, 46].
